Full text of "The future of arid lands : papers and recommendations from the International Arid Lands Meetings"

See other formats

THE FUTURE
OF ARID LANDS

ull ilu"
wal
|

Tied
| i

Nii

Sadia

@ TedEEOO TOEO O

WOM M00 00

IOHM/181N

THE FUTURE OF
ARID LANDS

Papers and Recommendations from
the International Arid Lands Meetings

Edited by GILBERT F. WHITE

DATA LIBRARY
HOLE OCEANOGRAPHIC INSTITUTIOp:

Publication Number 43 of the
AMERICAN ASSOCIATION FOR THE ADVANCEMENT OF SCIENCE
Washington, D.C. 1956

Library of Congress Catalog
Card Number 56-6107

Preface

This volume sets down the efforts of scientists from 17 coun-
tries and from as many disciplines to assess the state of man’s
struggle to make productive and stable use of the world’s arid
lands. Areas of meager and undependable rainfall and of sparse
vegetation commonly called ‘“‘arid” account for roughly one-third
of the land surface of the globe. Except where water is imported
for irrigation, as down the Nile and Colorado rivers, most of that
arid zone is sparsely settled and man lives in delicate adjustment
to uncertain moisture and shallow soils. Unlike the other great
land areas of low population density—the cold lands—the arid
lands have been greatly affected by man’s use and misuse, and the
margins of grazing, cropping and non-agricultural activity are
shifting, unstable frontiers of occupance.

The future of that occupance hinges in part upon success in
maintaining the present resources base at present levels of living:
range deterioration, water exhaustion, salt accumulation and
accelerated erosion are among the hazards to permanent use. The
future also hinges upon ingenuity in finding new and improved
ways of increasing the usefulness of these great physical expanses:
possibilities range from radical innovations in finding new water
sources to patient application of principles known centuries ago.

The whole range of thinking is recorded here without attempt-
ing to reconcile differences of view or to fill obvious gaps. No
single, clear answer emerges. Troublesome questions are identified
and new avenues of attack are plotted. The individual papers and
the group recommendations may be considered guideposts to
scientific development in at least three ways. They mark a promis-
ing method of collaboration across both national and disciplinary
boundaries. They point out specific areas of research in which
more vigorous activity isneeded. They suggest methods of think-
ing about the future that may play a significant role in shaping
that future.

ese
il

iv PREFACE

Several factors combined to bring together the scientists who
took part in the International Arid Lands Meetings in New
Mexico, April 26-May 4, 1955. A group within the Southwestern
and Rocky Mountain Division of the AMERICAN ASSOCIATION FOR
THE ADVANCEMENT OF SCIENCE under the chairmanship of Peter
Duisberg had felt that much could be learned from an interna-
tional exchange of thinking on arid lands research problems. The
national Association, interested in promoting the cooperation of
scientific groups and also in bringing basic research needs to the
attention of a wider public, lent its formal support and the able
guidance of John Behnke to the enterprise. Under its broad pro-
gram for fostering the development of arid lands the United
Nations Educational, Scientific and Cultural Organization gave
its strong backing and the weight of its experience in sponsoring
other international explorations of this character in Israel,
Turkey, France, and India. A program committee (listed else-
where in this volume) was appointed by the Board of the Associa-
tion to draw up plans for the meetings. It was decided to arrange
a public symposium around a few key problems and to invite
thinking without regard to traditional boundaries of academic
fields. The chairmen were to offer summary statements at the
conclusion of each session. At the same time it was felt important
to provide for informal discussion among participants so as to
make the most of possible exchange of ideas and experience among
those who were at the forefront of their respective fields. Accord-
ingly, a series of discussion groups were planned for the final day
of the symposium, a two-day field trip was scheduled, and a con-
ference was planned for a smaller group selected so as to represent
the various countries and disciplines.

With this program in hand, financial aid was sought and ob-
tained. The National Science Foundation, the Rockefeller Foun-
dation, and UNESCO granted funds which made it possible to
pay the expenses of some of the scientists who would not otherwise
be able to attend, thus assuring a highly representative group.
Substantial support was given by local business groups through a
finance committee of the Division. The University of New Mexico
generously served as host for the Symposium, and the New

PREFACE v

Mexico Institute of Mining and Technology graciously enter-
tained the Conference.

The program centered upon those areas of investigation where
prediction of the future currently must be based upon insufficient
understanding and data. The state of our knowledge and the need
for new research are described in the Symposium papers as well
as the Conference recommendations. They show the difficulty of
judging the capacity of arid lands resources in the present state of
knowledge but they also reveal certain ways of looking at the
problem which bear heavy significance for the future.

The whole concept of the water budget provides a framework
of thought in which much detailed analysis in climatology, hy-
drology, ecology, and geography begins to takeon new importance.

Simple as it may appear, the concept of selecting, breeding, and
improving plants and animals for arid conditions rather than
concentrating upon adjusting the arid environment for plants and
animals imported from humid areas opens out an enormous field
for exploration.

When many disciplines join in viewing an old problem the
traditional perspectives are challenged, and this is the case with
views of priorities of water use in arid areas. The relative efficiency
of various plant covers as water users, the proper allocations of
water among upstream and downstream uses, and the importance
of industrial in contrast to agricultural needs are among the
priorities called into question. Radical revisions in public views
of water priorities seem in prospect.

Two other trends in thinking about the future of arid lands
showed strongly in the conference discussions. One was the
emphasis placed upon integrated analysis of resources problems on
a regional basis. In many instances the advantages were recog-
nized of joining archaeological with hydrologic studies or of link-
ing botanical with climatological and geomorphic studies. The
trend is toward a cooperative approach to common problems.
This does not necessarily mean full integration of basic surveys in
the field but it does lead to a more nearly unified attack in such
areas as the Upper Rio Grande basin.

Second, the natural scientists repeatedly emphasized the im-

vi PREFACE

portance of translating scientific findings into action at the level
of operating farmers, herders, and land owners. This is a problem
wherever science and technology advance in agricultural societies,
but it has special relevancy to large-scale reduction in the zone’s
resources. Much of the deterioration of semi-arid lands now being
used at low efficiency is due to such conditions as tenure, property
rights, political control, social attitudes and taxes which impede
application of new knowledge and techniques. The investigator
of plant physiology or the geomorphologist sees his results gaining
usefulness only to the extent that there are peaceful means of
promoting social change toward accepted, wise ends. And this
leads him to encourage research which will reveal the processes of
decision in resource management or will point the way to public
education.

Preparation of the proceedings for the printer has been greatly
aided by Anne U. White. A separate report containing discussion
on the final day of the Albuquerque Conference of “Problems of
the Upper Rio Grande River” is to be published by local insti-
tutions cooperating through the Division’s Committee on Desert
and Arid Zone Research.

Although the papers from the Symposium and the recommenda-
tions from the Conference are printed in this volume, the more
important scientific outcomes are not recorded. They are showing
themselves and will continue to show themselves for years to
come in the work of individual scientists who took part or have
been influenced by the suggestions which emerged there. In this
sense the Symposium was a point of departure rather than a
summing up, and in this sense the future of the arid lands is in
some part shaped by those who sought to assess its prospect in

1955-

GILBERT F. WHITE

Department of Geography
University of Chicago
May, 1956

Contents

THE BROAD VIEW
History and Problems of Arid Lands Development

H. L. SHantz 3
The Role of Science in Man’s Struggle on Arid Lands
Cuar.es E. KELLocc 26

The Challenge of Arid Lands Research and Development for the
Benefit of Mankind
B. T. Dickson 47

VARIA BIEIDY AND PREDICTABILITY OF WATER SUPPLY
Climatology in Arid Zone Research

C. W. THORNTHWAITE 67
Water Resources in Arid Regions

J. TrxeRont 85
Data and Understanding

Luna B. LEopotp 114
Variability and Predictability of Water Supply

Frank DIxey 121
Fluctuations and Variability in Mexican Rainfall

C. C. WaLLEN 141
Beneficial Use of Water in Arid Lands

Joun H. Dorrou, Jr. 156
Geochronology as an Aid to Study of Arid Lands

Tera L. SMILEY 161
Summary Statement

ReeEp W. BalLey 72

BETIER USE OF PRESENT RESOURCES

GRAZING RESOURCES

R. O. WuyTe 179
Water Resources

L. N. McCLetian 18g
Geography’s Contribution to the Better Use of Resources

Hiicarpb O’REILLY STERNBERG 200

Agricultural Use of Water under Saline Conditions
L. A. RicHarps 221

Vill CONTENTS

Consequences of Using Arid Lands Beyond Their Capabilities
Cyrit LuKER 226

Possibilities of Increasing and Maintaining Production from
Grass and Forest Lands without Accelerating Erosion
RaymonpD PRICE 233

Land Reclamation and Soil Conservation in Indian America
PEDRO ARMILLAS 245

Better Use of Present Resources: Concluding Remarks
KANWAR SAIN 250

PROSPE@IS: FOR ADDIMIONAL, WATERS SOURCES

Demineralization of Saline Waters

SHEPPARD T. POWELL VET
Demineralization as an Additional Water Source for Arid Lands

Wirueimus F. J. M. Kru 272
The Salinity Factor in the Reuse of Waste Waters

H. E. Haywarp 279
Induced Precipitation

E. G. Bowen 291

Some Relationships of Experimental Meteorology to Arid Land
Water Sources
VINCENT J. SCHAEFER 300
Some Problems in Utilizing Water Resources of the Air
GLENN W. BRIER 314
The Economics of Water Sources
Louis KoEnic 320

BETTER ADAPTATION OF PLANTS AND ANIMALS
TO ARID CONDITIONS

Adaptation of Plants and Animals

Omar Draz RAT
Better Adaptation of Plants to Arid Conditions
R. Merron Love 343

Animals and Arid Conditions: Physiological Aspects of Productivity
and Management
Knut Scumipt-NIELSEN 368
The Locust and Grasshopper Problem in Relation to the Develop-
ment of Arid Lands
B. P. Uvarov 383
Desert Agriculture: Problems and Results in Israel
Micuaert Evenart AND Dov KOLLER 390

CONTENTS ix

Problems in the Development and Utilization of Arid Land Plants

IL, IMIS Jee, 414
Plants, Animals, and Humans in Arid Areas

EnrIQUE BELTRAN 419
Summary Statement

Oar S. AamMopr 424
RECOMMENDATIONS BY THE SOCORRO CONFERENCE 427
CoMMITTEES FOR THE MEETINGS 436
PARTICIPANTS IN THE SOCORRO CONFERENCE 441
INDEX 445

Enp Papers

The maps of arid homoclimates appearing in the end papers were
drawn from maps prepared by Peveril Meigs for the UNESCO Arid

Zone Program.

THE BROAD VIEW

What 1s the future of the arid lands?

History and Problems
of Arid Lands Development

H. L. SHANTZ

Santa Barbara, California

Definition of the Arid Zone

The arid zone has not been precisely defined. Probably in no
zone on the earth are there greater swings in precipitation, tem-
perature, and aridity than in this zone. It is customary to charac-
terize this area in terms of the minimum of precipitation and the
maximum of heat and aridity. Rainfall in this zone is rare, local-
ized, irregular, and often violent: precipitation in the more arid
portions of the zone averages less than an inch a year, but more
than the yearly average may fall in a single storm. The precipi-
tation in the semi-arid parts may be as high as 30 or 40 inches a
year, enough, if maintained, to move the area into the humid
zone. The relative humidity in the extremely arid parts may drop
to less than 5%. The highest air temperature on earth, 136° F,
was recorded 1n 1922 in this zone at Aziza in Libya.

To define such a zone properly would probably not be possible,
for if done the definition would be bound largely to a single factor
of the environment. A definition based on climatic data would not
be the same as one based on soils, on vegetation, on animal dis-
tribution, or on land use.

I would be inclined to include all the area from extremely arid
to semi-arid in this belt. In this whole range the only safe assump-
tion is that any year may be extremely arid. The more humid
years will take care of themselves, but the more arid years set the
pattern of use and must be anticipated and planned for if man is

4 THE FUTURE OF ARID LANDS

TABLE 1
Area of Arid Lands Based on Vegetation
Semi-arid
Sclerophyll brushland I, 180,000
Thorn forest 340,000
Short grass 1,200,000
Total semi-arid land — 2,720,000
Arid
Desert grass savanna 2,300,000
Desert grass-desert shrub 10, 600,000
Total arid land a 12,900,000
Extreme arid
Desert 2,430,000
Total extreme arid — 2,430,000
Total 18,050,000
Rercentas 35 (Land area 15,970,000)

safely to utilize this area. Moreover, the methods of use best
suited to arid lands are the safest means of utilizing the semi-
arid lands.

The area of the arid lands may be estimated on the basis of
vegetation, of climate, of soils, and by the area of interior drainage.

Estimates based on vegetation, expressed in square miles, are
given in Table 1.

The arid zone is designated sharply by vegetation. All plants
are adjusted to drought conditions, are either xerophytes or the
short-lived annuals. There are long rest periods during which
there is little or no growth. The open plant cover with much of the
soil surface bare is a characteristic of arid zone vegetation. How-
ever, drought rest periods are characteristic of many of the
grasslands and monsoon forests beyond the boundary of the arid
zone. Much of the tall grass or prairie and the tall grass and high
grass savannas in the tropics have drought rest periods late in
the season which stop growth and prepare these areas for the
characteristic fires which sweep over them. However, drought is
not the chief characteristic, and the plants are not xerophytes but
are mesophytes.

The extreme arid, the arid, and the semi-arid climates of the
earth have been mapped by Meigs (11), using the method de-
veloped by Thornthwaite (18). These estimates, expressed in
square miles, have been supplied by Dr. Meigs and are shown in

ablear

HISTORY AND PROBLEMS )

TABLE 2
Area of Arid Lands Based on Climate
Semi-arid 8,202,000
Arid 8,418,000
Extreme arid 2,244,000
Total 18,864,000
ercent 530 (Land area 51,970,000)

Drought due to lack of moisture is characteristic of the arid
zone. There are areas at high elevations and at high latitudes with
low precipitation where drought is not a characteristic. High
temperatures are generally characteristic of the arid zone but they
are not limited to this zone. The extremely high temperatures in
the arid zone are largely due to the absence of water for evapora-
tion or transpiration to cool the air. In the arid zone there is often
a marked inverse ratio between evaporation rate as measured and
the actual loss of water from the region. In an area with no
water to transpire or evaporate the measurements of loss from a
water surface or a wet bulb or any other measurement of dew
point will be very high although no water may be lost from the
land. Absence of precipitation has no damaging effect on vegeta-
tion if the soil has a supply of available moisture. This is evident
in any irrigated region where drought can occur only when the
source of irrigation water is cut off.

The vegetation cover is a measure of aridity, but here also there
are difficulties in application. An example will serve to illustrate
this difficulty and to emphasize the danger of assuming that
measurements based on a single species can be applied to all
plants, as is so generally done. The open sand blowouts in eastern
Colorado were being invaded and partly held down in 1915 by
Redfieldia flexuosa, Psoralea lanceolata, and the Andropogons
gerardi, hallit, and scoparius, and an occasional plant of Muhlen-
bergia pungens. During the dry hot years of the thirties these
plants were killed out, with the exception of the Muhlenbergia. One
would have expected the open sand areas to be greatly extended,
but just the opposite occurred for the sand areas were rapidly
covered by Muhlenbergia pungens, a grass more at home on the
hot arid sands of Arizona and New Mexico and before the hot dry
years only occasionally present in eastern Colorado. Had the
measurement of the years been made on the basis of the amount of

6 THE FUTURE OF ARID LANDS

sand covered by vegetation, the conclusion might well have been
reached that 1937 was a less arid year than 1915.

On the basis of soi/s about 43% of the land area of the earth is
estimated to be of pedocals. These soils have dry subsoils. The
moisture received at the surface hangs like a blanket above the
dry subsoil. Moisture will pass down into the dry soil only if the
water content of the surface soil has been raised above the field
carrying capacity. This soil moisture is lost to the air by two
methods. (1) Soil moisture in the first few inches can be evaporated
into the air, but moisture does not move up by capillarity to take
the place of the water lost, and the dry soil, be it of dust or of
hard soil, protects the moisture below. (2) If there are any growing
plants on this soil they absorb the soil moisture in contact with the
roots and pass it out into the air by transpiration.

The soluble carbonates are carried down by the percolating soil
moisture but they are not returned to the surface and gradually
accumulate at the bottom of the moist layer. This carbonate
layer marks the depth to which the soil is generally moistened by
the precipitation. In much of the more arid region the carbonates
are present in the surface soil, but the layer on the more humid
side of the chernozems may be 3 feet below the surface. Three
feet of soil would hold enough soil moisture to produce a luxuriant
grass cover and such an area would not be called arid or semi-arid.
These pedocals are not leached and are usually productive. The
plant growth is generally terminated by drought and the dry
vegetation is often burned off by fires started by man or by
lightning. Lightning fires have repeatedly been observed in the
short grass in eastern Colorado. Fire is also a characteristic of the
sclerophyll brushland of the Mediterranean type where the hot
and the dry periods come together and rainfall is confined to the
cool period of the year.

A world map of interior drainage has been drawn by de Mar-
tonne and Larfrére (g) which outlines very clearly the arid zone
as discussed above if we eliminate such areas as the upper Volga
and the upper Shari which although they do not drain into the
ocean come from relatively humid areas. The pedocal soils also
extend beyond the arid zone.

HISTORY AND PROBLEMS .

The estimated area of arid land is 43% if based on pedocals,
36% if based on climate, and 35% if based on the natural vegeta-
tion, and about the same area is marked by interior drainage.
Here then is an area of about one third of the land area of the
earth in which moisture is the chief limiting factor in the pro-
duction of plant growth and the dependent animal and human
populations.

Plant Adjustments to Arid Conditions

Since both plant life and animal life originated in an aquatic
_ environment great adjustments were necessary to enable them to
survive in the air and still greater adjustments to enable them to
live in the arid zone.

Water

Soil
moisture per cent

Figure 1. To illustrate relation of hydrophytes, mesophytes, and
xerophytes to moisture supply in a loam soil: hydrophytes @ to c; meso-
phytes and xerophytes ¢ to d. Mesophytes die below d and xerophytes
become dormant. Soil under desert drought sinks to about 5% or e.
Rainfall at f raises water content to about field carrying capacity, 19.6 %
at g.

8 THE FUTURE OF ARID LANDS

The adaptive features by which plants were able to move from
a hydrophytic environment to a mesophytic and then to a xero-
phytic environment have been studied for many years (10). Still
there is much misunderstanding with respect to these adjustments.
Plants are separated on a very broad basis into hydrophytes,
water plants, mesophytes, moist soil plants, and xerophytes
plants characteristic of drought regions and which many people
still think can grow in dry soil. These groups are not sharply
separated and merge into each other, but they serve as most con-
venient groups in discussing the many adjustments. These groups
are found under all climates, and their presence is determined by
the condition of the substratum.

To make this clear a simple diagram is used (Figure 1). A loam
soil is chosen to illustrate the relationships of the moisture retain-
ing power of the soil. If the soil chosen were sand the values would
still have the same definite relation, but the values would be
lower and if a heavier soil were chosen the values would be higher.
In order to simplify the graph only a few of the laboratory and
field determinations of the moisture holding capacity are entered
here.

In Figure 1:
Dry soil is at o% (based on dry weight)
Desert dry soil under desert vegetation 5% (16)
Hygroscopic coefficient 6.8% (5)
Wilting coefficient 10% (2)
Field carrying capacity 19.6% (3)
Moisture holding capacity 50% (6)
Saturated soil 50%
Water table at 50%

In free water from a to 4 on the graph we would find floating,
submerged, and amphibious hydrophytes. In soil conditions be-
tween 4 and c only hydrophytes would grow. Most of these hydro-
phytes would continue alive and grow it the soil moisture were
lowered from c to d but would die below this point. At water con-
tents between c and d mesophytes and xerophytes would grow,
but below d the mesophytes would die and the xerophytes become
semi-dormant or dormant. The soil moisture below d would be
slowly absorbed and lost to the air and reduced slowly to about

HISTORY AND PROBLEMS 9

one-half the moisture content at the wilting coefficient, or to
about 5%, somewhat less than Heinrich’s hygroscopic coefficient.
During this period growth is suspended and the plant is in a semi-
active or inactive condition often called estivation. Xerophytes
have no more ability to grow in dry soil than mesophytes or
hydrophytes with finely divided root systems such as rice.

When rain falls, the soil moisture rises from 5%, or even less
at the soil surface, to 19.6%, or the field carrying capacity. No
higher percentage can be added and no lower percentage if the soil
is not confined and there is free contact with dryer soil below.
Two inches of absorbed rainfall as compared with 1 inch moistens
twice as much soil but does not increase the moisture 1n percentage
of the dry weight of the soil. On the diagram growth begins again
at g.

Hydrophytes cannot grow in the mesophytic and xerophytic
habitat because of lack of abundant water, and mesophytes and
xerophytes cannot grow in a hydrophytic habitat because of a lack
of oxygen since they are adjusted to live only in the presence of
air in the soil and around their stems and leaves. A liter of water
at 20° C can contain the same amount of CO, as a liter of air at
the same temperature but only 1/65 as much oxygen. In hydro-
phytes the major problem is to secure oxygen, and mesophytes
and xerophytes cannot do this in a hydrophytic environment.
Xerophytes have the principal problem of not succumbing to
desiccation and are especially adjusted to prevent harmful des-
iccation. Mesophytes are not adjusted to hydrophytic or to
xerophytic conditions.

In measurements of the water economy of mesophytes in con-
tainers, one of the difficulties is to maintain the water content
in the container at the proper percentage range, that is, never
below the wilting point and never above the field carrying ca-
pacity. When water is added to a container the water is raised
to field carrying capacity, and if the proper amount of water is
added, practically all the soil mass is moistened to this point. If
half enough water is added half the soil remains dry. If more
water is added swamp conditions are produced and none will drain
out until complete saturation is reached. The plants will probably

10 THE FUTURE OF ARID LANDS

suffer or die because of lack of oxygen. If they are watered from
below, saturated soil or a water table is established at the bottom
of the container. Therefore all experimental containers a few feet
deep in which the experimenter expects the excess of rainfall or
irrigation water to drain out will find only hydrophytic or swamp
conditions which do not even approximate field conditions.

Let us consider some of the many adjustments xerophytes
have made to enable them to survive and to grow in the arid
lands. They must be able to grow under the following conditions.
The climate varies from semi-arid to extremely arid. The relative
humidity is often as low as § to 35%. The evaporation rate is high.
The soils are pedocals moistened only at the surface and with a
permanently dry sub-soil.

The roots of these plants are extensive but limited to the soil
moistened by the precipitation. The root surface is large as com-
pared with the aerial plant and root hairs are abundant.

The stems are small, often with deciduous leaves or none.
Cladophylls or bracts or stems often function as leaves. The
branches are often crowded together or are underground for
protection.

The /eaves are generally thick, small, firm, and leathery (sclero-
phyllous); they do not wilt but endure drought and recover when
water 1s again available. Leaves are often deciduous.

The internal structures also show many adjustments.

The roots of xerophytes often have cork or sand sheaths near
the soil surface.

The stems have cork or thick cuticular coverings. The stems
often function as leaves, especially when leaves are absent, and
have several layers of palisade tissue. The stomata are sunken
or protected.

The /eaves, if present, have the leaf lamina reduced or absent
and the cells and stomata are small. They are sclerophyllous with
much supporting tissue, do not wilt, but often roll or fold to pro-
tect the stomatal areas. The outer epidermal walls have thick
cuticles with waxy or resinous coating. The stomata are sunken
and protected. The epidermis often has woolly, scaley, or stellate
hairs filled with air to reflect light. The leaves have several layers

HISTORY AND PROBLEMS 11

of palisade tissue and small intercellular spaces. The spongy
tissue is reduced.

The drought-resistant succulents show the same structures for
preventing water loss but also have impounded water in roots,
stem, or leaf, or all of these, and are able to continue growth when
no water is available in the soil. They rapidly absorb water from
summer showers, have an extensive superficial root system, are
often globular or columnar, and have a large volume as compared
with the surface. They resist drought by impounding water in the
plant body.

The physiological adjustments are equally striking. Unlike
the mesophytes, the leaf cells of xerophytes do not wilt when they
lose water. The leaves are either dropped or, if they remain on
the stem, they may lose as much as 40 to 80% of their water and
still recover when water is again available. The reason for this
recovery is that, unlike the mesophytes, the dehydration does
not cause coagulation of the protoplasm or rupture of the protoplast
to produce permanent and irreversible harm. A possible explana-
tion of this condition is found in the high osmotic pressure of from
20 to 7§ atmospheres and possibly the effect of hydrophylic col-
loids and sugars on the protoplast.

The ecological means by which plants meet drought conditions
and grow in regions where droughts occur are of four types (8).

1. Drought escaping plants grow only where or when there is no
drought. Here belong the summer and winter annuals of our South-
west such as the six-week grasses and the ‘‘ashab”’ of the Sahara.
They usually grow under mesophytic conditions and except in the
seed ripening and seed condition, do not encounter drought. They
live through the hot, dry, drought period in the seed stage, a
dormant stage similar to estivation.

2. Drought evading plants are plants economical in the use of
the limited soil moisture supply. This is accomplished by wide
spacing, by keeping the plants small with a small leaf surface and
a small amount of annual growth. The root systems are propor-
tionally very large. These plants are often very efficient in the use
of water. Under like conditions plants may require as much as from
300 pounds to 2200 pounds of water to produce 1 pound of dry

12 THE FUTURE OF ARID LANDS

matter—a variation of more than seven times (17). This range for
different plants is about the same as the difference measured be-
tween the same plant grown in a semi-arid and an extremely arid
climate. The grain crops grown most successfully in the arid
zone are of this group. They have a short season, rapid growth,
large grain yield in proportion to the straw, and a low water
requirement.

3. The term drought resistance can be reserved for the suc-
culents, which impound water within their plant body and are
able to continue growth when soil moisture is entirely unavailable.
They are admirably adjusted to absorb soil moisture rapidly as
soon as it is available in light showers and to store it in their
bodies. They are not characteristic of the extreme deserts.

4. Drought enduring plants estivate when drought occurs and
they can resume growth as soon as water is again available. In
other words, they become drought dormant. They also possess
most of the characteristics of the drought evading plants by which
they are enabled to prolong the growth period. The most im-
portant and prominent desert plants belong to this group. When
soil moisture is no longer available, these drought enduring plants
become drought dormant or estivate in the vegetative stage,
while the drought escaping plants pass through the drought only
in the seed stage. Many bacteria, fungi, algae, lichens, mosses,
and ferns are able to endure drought.

Great adjustments are made by plants to enable them to grow
in the arid lands. Great as are the physiological Fences
they are small compared with the variation in size shown by the
drought escaping plants which will still ripen seeds when less than
one thousandth of normal size.

Animal Adjustments to Arid Conditions

Animals show as great adjustments as plants do to enable
them to survive in the arid lands (15). They too developed in
water and they had to develop lungs or tracheal systems in order
to secure oxygen from the air. They must also have a constant
supply of water for their active growth. One of the important
reasons why animals can live in the arid zone is found in their

HISTORY AND PROBLEMS 13

ability to move about. Birds, insects, and large and fleet mammals
can occupy arid lands if there is a water supply within their range.
Smaller mammals and reptiles are more dependent on local con-
ditions and have shown great adjustments to arid conditions. We
may consider these adjustments under the same classification
used in discussing the plants.

1. Drought escaping animals are those that avoid drought areas
and are not found permanently in or dependent on the arid lands.
One may include here animals that enter the arid region only when
an adequate supply of moisture is available either as water or lush
vegetation, or that can move to the water supply. By far the
greatest number of drought escaping animals found in arid areas
belong to the insects and the lower groups of the invertebrates.
Many of these animals are able to complete their life history or
the active part of it during the short period when conditions are
favorable. Here belong most of the insects that abound on the
summer and winter annuals. These forms pass the drought period
as drought enduring eggs or larvae and resume their active life
only at the end of the drought period. One might include all
animals of the humid zone and in water habitat as drought es-
caping for they are not found in drought areas.

2. Drought evading animals show many adjustments even more
striking than those shown by plants. The ability of animals to
move has enabled them to burrow into the soil and thus live in an
environment much more favorable during the day and confine their
activity to the night and early morning when the conditions are
not extremely arid. Partly as a result of this ability some of
the rodents do not need to provide water for temperature control
except in extreme cases. K. and B. Schmidt-Nielsen in their excel-
lent summary have listed about a dozen genera of rodents from
widely separated deserts which do not expend water for heat regu-
lation (15). This is a significant example of the development of the
same adjustment in widely separated areas by animals not closely
related phylogenetically. They show certain gross morphological
similarities. They all are leaping animals with elongated hind legs
and small front legs, with reduced number of toes, and they have
cheek or gular pouches.

14 THE FUTURE OF ARID LANDS

3. Drought resistance in animals is different from drought
resistance in plants in which large quantities of water are im-
pounded within the plant body and growth is continued when no
other water is available. Although animals resist drought differ-
ently the term drought resistance can still be used since normal
activities are maintained. Dr. Schmidt-Nielsen’s studies of the
camel and the donkey in Algeria will do much to explain the
physiological processes by which these animals are able to resist
desert conditions. (See pp. 370-380). He has found no reason to
agree with the common idea that the camel carries in pouches in
his stomach a reserve of water to be used when no drinking water
is available. Drought resistance in animals rests upon the physio-
logical processes by which animals are able to concentrate the
urine, stop perspiration, lose little water in the feces, endure
dehydration, and still remain active.

The methods by which the kangaroo rat, jaroba, and some other
animals avoid loss of water is by voiding nearly dry feces and
concentrating the urea in the urine to as high as 23%. Some
birds and reptiles can void urea and nitrates in solid form. Ex-
tensive studies of the kidney functions of these animals have
been made and are under way to try to determine the exact
means by which such concentration is accomplished. This ability
might well be a deciding factor in enabling some animals to escape
death by drought in an arid desert.

Drought resistance in animals has been accomplished in some
groups not by impounding water but by the physiological process
of developing metabolic water (11). Dry fat when oxidized pro-
duces 1.1 grams of water from each gram of fat. A gram of starch
or sugar produces 0.6 gram, and a gram of protein 0.3 gram of
water. The water formed by the oxidation of the hydrogen in dry
food can be used to carry on the life functions. Protein has a
great disadvantage as a source of water for it requires the loss
of much water to remove the urea that is formed.

A great amount of work has been done on the kangaroo rat to
determine the water balance under desert conditions (7 and 14).
These animals can be kept on dry pearled barley almost in-
definitely without any other food or water if the relative humidity

HISTORY AND PROBLEMS 15

is not lower than 10%. They do not store water when abundantly
supplied and do not lose weight on dry feed. They do not lose
water to control temperature under normal conditions. When fed
only on dry protein, dehydration resulted in a reduction to
67.2% of normal weight, or the animals lost nearly a third of their
weight and died shortly afterward. But they are not forced to
endure desiccation when feeding on a desert dry food supply of
grasses and seeds. The kangaroo rat, pocket mice, and jarobas
can maintain balance on a dry diet and thus resist dehydration.
Babcock in his important work (1) on metabolic water in I912
discussed its role in insects that feed on dry plant and animal
material, for many are able to oxidize the hydrogen of their foods
to obtain enough water for their normal life processes.

4. In the field of drought enduring we have the least physiological
details, and yet we know that many rodents that do not de-
velop metabolic water pass into an inactive state called estivation
in summer from which they can recover when the hot dry period
ends. There are many invertebrates that can be desiccated
and recover quickly when water is again available. Many cases
could be cited. Among the mammals the round-tailed ground
squirrels and pocket gophers estivate during the hot dry periods,
but apparently they do not suffer a very marked desiccation. They
reduce the rate of respiration and heart beat and assume the
temperature of the air around them and are inactive. This summer
sleep or estivation still offers a most promising field of physio-
logical investigation.

Primitive Man

Primitive man made remarkable progress in the use of the arid
lands. In the field of animal husbandry he domesticated all of our
animals with the possible exception of the turkey (13). His use
of arid lands was made possible partly by grazing these lands with
cattle, camels, goats, sheep, asses, and horses. Sauer traced the
origins and dispersals of not only the domesticated animals but
also many of the cultivated crops. Excepting the grazing animals
and seed crops, most of the crops and domestic animals have de-
veloped in the humid rather than the arid zone. However, the

16 THE FUTURE OF ARID LANDS

great nutritive value of the semi-arid and arid vegetation has
attracted stockmen to these areas and probably from the begin-
ning of man’s use has had a part in pushing back the vegetation
which had by tremendous adjustments established itself in this
arid area.

In the field of crop production most of the plants used by primi-
tive man were dependent on well-watered land. He brought into
cultivation all the principal crops now used. At the edge of the
arid zone he showed marked ingenuity in the profitable use of
arid land for the production of seed crops. In the Sahara and other
Asian and African desert areas he made great use of the ‘‘ashab”’
or temporary growth of lush weedy drought escaping plants. He
also seeded short-lived barley, a very well-adapted plant, to pro-
duce a catch crop following a temporary summer rain. Every
advantage was taken of flood water. Sorghum was planted close
to the receding water edge and the planting was continued until
the water had disappeared. This resulted in a field in which the
plants were younger toward the center of the area.

In the Southwest of the United States the Pimas relied chiefly
upon flood water and irrigation to produce their crops. They also
made great use of wild plants. These include about 22 varieties
of plants of which they used stems, leaves, or flowers, 4 that fur-
nish bulbs, 24 seeds or nuts, and 15 fruits or berries. This may
indicate why primitive man so far excelled modern man in the
domestication of plants and animals. Mesquite beans were collected
and the fruit of the saguaro was made into dried sweetmeats,
jams, and jellies. All suitable fruits of cacti were used. The
fields were prepared and planted to cotton, maize, squash, water-
melons, beans, and devils claw. The Papagoes practiced dry
farming with or without flood water. Maize, wheat, barley, beans,
and cotton were the principal crops.

The Hopi are outstanding as successful crop producers under
arid conditions. With a very low rainfall, they developed irrigation
and dry farming to a degree hardly reached by modern agriculture
even with the aid of such powerful weapons as soil physics, soil
chemistry, plant physiology, and plant genetics. Hopi maize
grown in fields of varying sizes and shapes presents an interesting

HISTORY AND PROBLEMS 17

study. Planted usually on land that has received some flood water,
the soil is cleared for planting. Holes are dug with a stick down to
moist soil and 1§ to 20 grains of maize are dropped in the hole.
Although ordinary varieties of maize cannot reach the surface if
planted over 3 inches deep, Hopi maize because of a greatly
elongated hypocotyl can be planted as much as 14 inches under
ground. Whereas ordinary maize develops three roots from the
seed, Hopi maize develops only one, which can be pushed deep
into the soil in search of moisture. These two adjustments make
Hopi maize superior to other varieties in an arid environment. A
hill of dense stems protected at first by a rock and later by many
leaves is not easily damaged by drifting sand. Planted 6 to 8
feet apart the roots can slowly elongate into a large soil mass
available to them. Beans are next in importance, and squash and
melons are grown without irrigation. Peaches and a few apricots
and apples are grown on the sand hills. The Hopi also irrigate
land when and where water is available. Here onions, chili, wheat,
sorghum, tomatoes, potatoes, grapes, garlic, cucumbers, tobacco,
and false saffron are grown. The tepary beans were domesticated
in prehistorical time and are now important crops of this region.

To primitive man one great advantage of the arid zone was the
ease of curing and preserving food in the dry air. Seeds can be
stored and such plants as squash and melons cut into strips and
dried to be kept for future use.

The use of all features of the natural environment by primitive
man is well illustrated in South Africa where the Bushmen and
Hottentots depended so largely on the fruits of the mtsama melon,
the wild watermelon, an annual plant which gathers moisture
from the desert soils during the growing season and stores it in
the small thick rind melons which hold it for months. The melons
could be cached in the desert sands and used largely as a water
supply when needed. Travel in the Kalahari was possible only
when this source of water was available. Again in the extremely
dry Southwestern African desert the Acanthosicyos horrida, a
shrublike spiny cucurbit, furnishes in its large fruits both water
and valuable oil seeds for food and drink when no other water is
available. The barrel cactus serves as a like source of water in

18 THE FUTURE OF ARID LANDS

cases of emergency. Those who die of thirst in the desert are
usually lacking in resourcefulness.

Ancient Man

It is difficult to determine how far ancient man pushed his agri-
culture into the arid land. Here is not included the oasis type of
development, in which great advances were made. Originally
the Nile Valley was watered naturally by flood waters. The oasis
type was either supplied with water by small streams or water was
raised from wells by a dillus operated by a camel or a donkey or
by a hand-operated chaduf. These methods are ancient as is also
the use of the old aqueducts or foggaras. Ancient man also used
flood plains and constructed large rock terraces to hold back flood
waters. All desert forage within travel distance of drinking water
was used by domestic animals. I find little to indicate the use of
any land but that well-watered by either precipitation or irriga-
tion from springs and running water in the ancient accounts avail-
able in the Bible, Koran, Book of Mormon, Talmud, Cato, Virgil,
Theophrastus, Pliny, or many others. The many references to
and the discussions of agriculture deal with lands well watered.

The development of olive culture by wide spacing and clean
cultivation is a fine example of ancient man’s use of dry land.
Grain and legumes were also grown extensively. It is probable
that a more thorough search of ancient literature, possibly in the
writings of Mago, the Carthaginian, which are not available to
me, will throw light on the use of semi-arid lands by the
Carthaginians. There is evidence that grain culture was an im-
portant industry on their arid lands.

Modern Man: Our Attempts at Dry Land Agriculture

The use of the so-called dry land could be traced in many parts
of the world on the better or semi-arid lands. I am most familiar
with the development in what was generally known as the Great
American Desert. J. W. Powell’s Report of the Lands of the Arid
Regions in 1878 (12) was exceptionally comprehensive and sound.
He recommended not less than 4 square miles as the size of a home-
stead and stated that ‘‘In those localities, and, so far as I am

HISTORY AND PROBLEMS 19

aware, in all others where dry land has been successfully farmed,
the soil is sandy, and this appears to be an essential condition.”’

Sand or sandy soil offers an especially favorable environment,
for sand or sand dunes act as sponges and absorb all water that
falls, and except for a few inches at the surface do not pass it back
to the air by evaporation. One walks for miles through desert
with no sign of animal life to come upon a small sand dune marked
by mammal, bird, reptile, and insect tracks. Here the penetrability
of the soil to water results in an increased animal and plant popu-
lation. The meat eaters are naturally found where the plant eaters
abound. |

The Homestead Law in 1862 made possible the acquisition of
160 acres of land by a residence of five years. The Timber Culture
Act in 1874 enabled settlers to acquire title to 160 acres of land
on condition of growing a certain amount of timber. In 1916 the
640 acre Homestead Law was passed and was confined to land
suitable only for grazing and the production of forage. In the
meantime the grasslands of the Great Plains had attracted the
cattle men. Migratory cattle became abundant on the range about
1866 and increased to millions in fifteen years, but in 1886 the
cattle boom, which had reached its height in 1882, came crashing
down. At that time there was a great influx of homesteaders in
Eastern Colorado. A few favorable years resulted in an agricultural
boom based chiefly on maize and potatoe production. A few cattle
men remained in this region, mostly in the sand hills. Then came
the drought of 1893-94, so severe that all but a few settlers left
the region. By 1908 about 13% of the land had been plowed, and
about half of this plowed land was in crop and about half was
abandoned and returning to short grass, the original plant cover.
In 1949, 96% of this land was plowed, and not an acre was going
back to grass. It was largely in wheat and summer fallow.

The beginning of dry farming was greatly influenced by false
conceptions and by propaganda. There was a feeling, supported
by propaganda, that, if the loss of water from the surface could
be stopped or retarded, capillarity would raise water from the
water table to produce the crop. It was easily demonstrated
that a great depth of dry soil lay between the moist surface soil

20 THE FUTURE OF ARID LANDS

and the capillary fringe, and no water could be secured from the
water table by capillarity.

It was also thought that a dust mulch maintained by repeated
cultivation would prevent water loss and therefore provide a
solution of the water problem. Again it was easily proved that a
dust mulch provided no more protection to the moisture below
than the same amount of hard soil (3), that the only advantage of
the repeated cultivation was in keeping down the weeds which
would otherwise pull the water out of the soil. This same effect
was accomplished in Australia by overpasturing the land with
sheep. But the dust mulch had two very serious, harmful results.
In a heavy rain, and many of the rains are of this character, the
soil sealed over, prevented water penetration, and caused a heavy
runoff. The dust mulch was also subject to wind erosion and was
a predisposing cause of dust storms. A cloddy surface was much
more desirable for preventing blowing and for increasing water
penetration.

Another bad practice was introduced by the advocates of loosen-
ing up the subsoil by blasting, deep plowing, and by the subse-
quent use of a sub-soil packer. Closely allied to these harmful
practices was the deep planting of fruit trees. The roots of such
trees came promptly to the surface where and where only could
they find moist soil. The use of manure and nitrate fertilizer had
to be abandoned for they caused the loss of the crop by drought.
In the short grass the nitrates released in fungus fairy rings killed
out the short grasses.

The conservation of moisture transcends all other arid land
problems. All precipitation should be absorbed by the soil and
held there until needed by the crop. Weeds should never be
allowed to compete with the crop, even before the land is seeded.
The moisture should be used as soon as practicable by short season
drought evading crops which are thinly seeded or widely spaced
to allow a large amount of moist soil to each plant.

In eastern Colorado maize was the crop used by the earlier dry
farmers. Now wheat has crowded out most of the other crops and
as a result of the sustained price about 70% of the land is devoted
to this crop.

HISTORY AND PROBLEMS 21

Human Extension of Desert Lands

Man, prehistoric, ancient, and modern, has been responsible
for greatly increasing the desert lands. His herds have removed
much of the scanty cover of nutritious grasses and shrubs, and
thereby favored the non-palatable ones. His path has been marked
by ruins of human settlements. It is doubtful if desiccation has
been the cause. The cause seems to have been over-use. Within
the life span of many of us beautiful areas of desert grassland have
been reduced to bare soil and useless weeds by over-grazing, the
short grass of our high plains have been replaced by wheat and
summer fallow, followed by the dust bowl, and the brushlands of
the Mediterranean type have been reduced to non-palatable brush
by fire followed by grazing of the young sprouts, a practice gen-
erally employed and one very detrimental to palatable plants. It
is almost impossible to reverse these destructive trends under
increasing population pressure, but it must be done if future
generations are to find the resource in as good condition as we
found it.

Great areas of our most productive soils are being occupied by
cities, highways, landing fields, reservoirs, recreational areas, and
factories, in which the crop production capacity of the land is
lost. It is possible that some of this unproductive occupancy can
be halted by shifting some of these uses to more arid land. The
arid zone is a delightful place in which to live during at least a
part of the year, and it affords a retreat to many who can choose
their place of abode and to some who find health only on the
arid lands.

Future Lines of Development

In general the ancients used flood waters by using naturally
flooded areas and by building deflectors and large stone terraces
to retain flood water in selected places. It was, in fact, a type of
irrigation agriculture. The Indians of our Southwest did the same,
but they took every advantage of the natural drainage areas,
allowing nature to concentrate water in channels, fans, and
temporary pools. Dry farming is based on a very different prin-
ciple: to catch the water where it falls, hold it in the soil, and

22 THE FUTURE OF ARID LANDS

utilize it as soon as possible by a crop adjusted in its demands to
the supply of moisture in hand.

We have by no means fully utilized the highly adjusted plants
of the arid lands so rich in fibers and valuable chemical com-
ponents. Duisberg (4) has pointed to many of them, but the
field is still largely unexplored. Esparto and guayule are examples
of useful products developed by nature in this zone. In this field
M. C. Caldwell of the University of Arizona has found most
promising antibacterial compounds which indicate that these
desert plants have produced many substances which may fit into
the important field of antibiotics. The genetic work of Gordon
Whaley of the University of Texas points to a great resource of
genetic material in the grasses of this arid belt. These are only a
few examples of what seems a promising field of research. There
are probably many chemical components developed in these
drought enduring plants not present in plants grown and de-
veloped under a less exacting environment.

We know far too little from the standpoint of physiology and
water economy to attempt a reasonable management of range
plants. The plants that are the largest and produce the most
forage are favored in the management with no questions asked
as to the economic use of the valuable water supply. To choose
the largest may be very misleading. If A, B, and C are grown to-
gether in the same soil mass we may choose A with a water re-
quirement of goo as a better plant than B with a water require-
ment of 600 or C with a water requirement of 300 pounds of water
to one pound of dry forage. In this case by choosing the plant
on the basis of production, we are choosing the least efficient
plant of the three. Not until we have a physiological balance sheet
of the principal components of the range can we separate the in-
eficient from the efficient.

We also know far too little of the water balance of animals. In
attempting to improve the food supply for the Navajos the prairie
dog was poisoned and destroyed in the area. But to the Navajos
the prairie dog is a delicious morsel. We do not know that the cow
or sheep can produce as much food with a ton of grass as can the
prairie dog.

Primitive man used the rodents as an important food supply.

HISTORY AND PROBLEMS 23

With the exception of squirrels and rabbits, modern man has
almost neglected this group as a food source. Rodents probably
are the greatest consumers of plant material and on these 20,000
species man and other carnivores could largely depend for food.
They among mammals seem to have made the greatest adjust-
ments to life in an arid environment. Yet modern man has re-
garded them largely as pests and competitors of his larger domesti-
cated mammals. It would seem possible to explore this resource
as a food source for man, as have the more primitive peoples of
the earth.

Nature set almost impossible tasks for the plant and the animal
world. Water, the substance about which all life was developed,
would seem to be all essential. Here in the drought deserts nature
has provided a minimum of supply. At the same time the tem-
perature and other atmospheric conditions demand the greatest
amount of water. In the plant world the challenge has been met
wonderfully and almost all of the earth surface has been occupied.
The same is true of the animal world.

Modern man has been given mighty instruments to extend this
use. So far he has relied largely on engineering skill to lead waters
into this desert region from regions of ample supply. He has not
accomplished much in economical use for he still grows crops
during the hot dry part of the year at a very high rate of water
consumption. Neither has he attempted under irrigation to im-
prove the efficiency of these irrigated crops in the use of water.
Ground water has been too generally regarded as a renewable
resource but experience has often proved it to be largely non-
renewable.

He hopes to demineralize sea and alkali waters when the demand
justifies the expense, thus to aid in a small degree nature’s great
distillery which pours over the land an average of about 30 inches
of water of the highest quality each year. He likewise hopes to
control to a degree the distribution of this rainfall and to direct
it to more arid areas.

One thing he can hope to accomplish and that is to stop the
enormous loss of water to the ocean, when it is so badly needed on
land adjacent to the streams that carry it away.

Nature has produced a magnificent controlled area occupying

24 THE FUTURE OF ARID LANDS

about one-third of the land surface. Here is an experiment already
planned and carried out. All we have to do to advance our knowl-
edge in the field of the more complete utilization of these arid
lands is to study and interpret properly what nature has already
accomplished. Here is a great field for the physiologists. Here also
is a rich resource for genetic exploration and the geneticist can be
well guided by the physiologist as to the means by which plants
and animals can be better adjusted to this rather extreme environ-
ment. Can the geneticist breed an animal that can live on meta-
bolic water or can concentrate the urea in the urine, or can he
produce a maize plant with only one seed root and with an elon-

gated hypocotyl? Nature has pointed the way.

REFERENCES

1. Babcock, S. M. 1912. Metabolic water: Its production and role in
vital phenomena. Univ. Wis. Agr. Expt. Sta., Research Bull. 22,
87-181.

2. Briggs, L. J., and H. L. Shantz. 1912. The wilting coefficient for
different plants and its indirect determination. U. S. Dept. Agr., Bur.
Plant Industry Bull. 230, 1-83.

3. Burr, W. W. 1914. The storage and use of soil moisture. Neb. Agr.
Expt. Sta., Research Bull. 5, 1-88.

4. Duisberg, PC. 1953. Chemical components of useful or potentially
useful desert plants of North America and the industries derived
from them. Desert Research. Proceedings of the International Sym-
posium, Jerusalem, May 7-14, 1952.

5. Heinrich, R. 1894. Die Absorptionsfahigkeit der Bodenarten ftir
Wasserdampf und deren Bedeutung ftir die Pflanzen. Zweiter Be-
richt tiber die Verhaltnisse und Wirksamkeit der Landwirthschaftlichen
Versuchs-Station zu Rostock, pp. 19-85.

6. Hilgard, E. W. 1906. Soils: Their Formation, Properties, Compo-
sition and Relations to Climate and Plant Growth in the Humid and
Arid Regions, pp. 1-593.

7. Howell, A. B., and I. Gersh. 1935. Conservation of water by the
rodent Dipodomys. 7. Mammal. 16, 1-9.

8. Kearney, T. H., and H. L. Shantz. 1911. The water economy of dry-
land crops. U. S. Dept. Agr., Yearbook Agriculture, TQIT, 351-362.

g. Martonne, Em. de, and L. Larfrére. 1927. Map of Interior Basin
Drainage. Geog. Reo. 17 (@), Jel @

10. Maximov, N. A. 1929. The Plant in Relation to Water. > igindlarn
11. Meigs, Peveril. 1952. Distribution of arid homoclimates. Maps Nos.
392 and 393, United Nations.

HISTORY AND PROBLEMS ALS)

. Powell, J. W. 1878. Report of the lands of the arid regions of the

United States with a more detailed account of the lands of Utah.
45th Cong. 2d Session. House Ex. Doc. 73, Washington, D. C.

. Sauer, C. O. 1952. Agricultural origins and dispersals. Am. Geog.

Soc., New York.

. Schmidt-Nielsen, K., and B. Schmidt-Nielsen. 1951. A complete

account of the water metabolism in kangaroo rats and an experi-
mental verification. 7. Cellular and Comp. Physiol. 38, 165-181.

. Schmidt-Nielsen, K., and B. Schmidt-Nielsen. 1952. Water metabo-

lism of desert mammals. Physiol. Revs. 32 (2), 135-167.

. Shantz, H. L. and R. L. Piemeisel. 1924. Indicator significance of

the natural vegetation of the southwestern desert region. F. 4gri.
Research 28 (8), 721-802.

. Shantz, H. L., and Lydia N. Piemeisel. 1927. The water requirement

of plants at Akron, Colorado. 7. Agr. Research 34 (12), 1093-1190.

. Thornthwaite, C. W. 1948. An approach toward a rational classifi-

cation of climate. Geog. Rev. 38 (1), 55-94.

The Role of Science
in Man’s Struggle on Arid Lands

CHARLES E. KELLOGG

Soil Conservation Service, United States Depart-
ment of Agriculture, Washington, D. C.

History tells us a great deal about the relationship of man to
his land in arid regions. Broadly it tells us that man can live on
such lands, but to live well and to prosper he must make full use
of science and technology for the combined use of all the resources.

First, ‘‘arid lands” as used in this paper must be defined. In-
cluded specifically are those regions in which the normal soils,
although perhaps productive of grass and browse, are usually too
low in moisture for the dependable production of cultivated
plants without irrigation. Many of the characteristics of arid
lands thus defined are found in slightly less dry regions where the
soils will support crops of cereal grains in alternate years without
irrigation if special practices are followed to conserve moisture.

The boundary between the sown and the unsown is wide and
shifting. Under extreme pressure for food, cereal growing com-
monly pushes out onto arid soils too dry for sustained production.

The Balance We Seek

The central problem of land use and human living in arid re-
gions, as elsewhere, is to maintain a reasonable fit between society
and resources.

Among very primitive people, comparatively isolated and hav-
ing few tools, perhaps one may talk about a “natural balance”’
of some sort. But such a concept has limited relevance today.

SCIENCE IN MAN’S STRUGGLE ON ARID LANDS 27

True, there are limits beyond which the use of renewable resources
of soil, water, and vegetation cannot be intensified without
trouble. Once we go beyond such limits it is always difficult and
usually painful to restore their optimum use. Overgrazing of
ranges, overcutting of forests, or excessive tillage on farm lands
may initiate new cycles of erosion hard to control. Restoration of
the soil may call for very difficult physical and social adjustments.
The continued overdraft on the local ground water, using it faster
than it recharges, can mean only reduced water some day and
severe social penalties on someone.

Such limits on resource use are only partly fixed in nature. They
depend as well upon man’s skill, patience, and industry. Using
the principles of science, we invent new tools of enormous power—
in engineering, in biology, in soil and water management, and in
administration; and we adapt them to local conditions. Even
more important than the tools yet invented are the basic skills of
science, the skills of basic research by which we get the principles
for inventing still more.

An attempt “‘to return to nature,’”’ to a way of life without tech-
nology, would condemn the majority of the world’s population
to starvation and death. This majority has learned in recent years
that starvation is unnecessary. In fact, the present unrest in the
world about food and opportunity is not due to material shortages
in terms of the past. Never was there greater abundance. The
unrest is due to a new realization of the enormous shortages we
have in terms of what is possible with the full use of modern
science.

We are seeking a cultural balance or, more accurately, a cultural
dynamic of relationship between resources and people for efficient
sustained production. For this we use all the skills of science and
engineering. The relationship we seek is hardly a balance, except
perhaps momentarily along the way from one point of efficiency
and abundance to a higher one.

Where Do We Stand?2

First, I shall review some of the basic aspects of arid land
resources.

28 THE FUTURE OF ARID LANDS

Climate

Arid regions have little rain; and the little they do have is often
exceedingly irregular. It may seem a bit ironical to many that
desert landscapes, such as those of Death Valley in the United
States, are dominated by erosion features. With too little moisture
for a protective plant cover, the infrequent but severe storms cause
severe erosion.

Rainfall and temperature have been recorded for large parts of
the world for some time now; but these alone do not give a precise
definition of environmental conditions. The water in the soils,
for example, is partly a matter of total rainfall, as modified by
slope, plant cover, and soil permeability, and partly a matter of
evaporation, which depends upon humidity, cloudiness, and wind,
as well as upon the temperature.

Nor do the total or average values for these factors tell enough.
The patterns of each factor, in relation to the others, through the
seasons, must be defined. Many arid lands near the Mediter-
ranean, for example, have the most rain during the coolest months
and the least during the hottest months. Others, such as those
near us here in Albuquerque, have the reverse pattern; both rain-
fall and temperature are highest together. Even with the same
totals and annual averages, these are quite different environments
for soils and plants.

Renewed efforts are being made to define climatic types and to
classify them. Description is difficult because a climatic type has
no morphology in the sense that a rock, a plant, or a soil has a
morphology. The “morphological” definition of a climatic type
depends upon what the climatologist decides to be relevant
seasonal patterns of factors that he can measure and of the al-
lowable limits of variation over a long period of years. He cannot
“see”? what he fails to define. His notions of the relevance of
factors and of variations within them depend, in turn, upon
studies of the effects of the climatic environments upon soils,
water, plants, animals, and people.

The results of these combined studies are leading to better
definitions and classifications of climatic types. The closer the
network of observation stations of long record with complete data,

SCIENCE IN MAN’S STRUGGLE ON ARID LANDS 29

the more accurately these climatic types can be defined and shown
on maps (5, 9). The interpretation of such maps in terms of plant
growth and available water, even where highly accurate, depends
upon a knowledge of the interactions between the local climatic
type and the local land form and kind of soil.

Within one climatic type, for example, a pervious soil of medium
texture can take in water, even of sharp showers, and hold it for
plant growth; whereas impervious soils refuse water, shallow ones
cannot hold much, and loose, sandy soils allow the water to pass
beyond the reach of plant roots. Yet water falling on thin soils
over cracked limestone or basalt, finds its way into the deep
substratum from which it appears again in springs or artesian
wells, or may be tapped by pumping.

Such basic relationships between soil and climate are sig-
nificantly modified by variations in plant cover and soil use.

Detailed studies of the interactions among climate, ground
water, land form, soils, and plants are increasing in several parts of
the world and are furnishing a continually better basis for assess-
ing the potentialities that we have and the critical problems of use.

Especially do these results give an essential part of the firm
basis we need for transferring plant materials, techniques, and the
results of research and experience from one place to another. But
climate data alone are not enough. In fact, unless the other re-
sources and the existing social patterns are well known also, the
use of climatic analogues alone may give a sense of scientific se-
curity in faulty conclusions less reliable than the direct impressions
of an observant traveler.

A challenging aspect of climate is the variation of its features
from year to year and from century to century. There is abundant
evidence in the rocks, in the soils, and in the cultural remains that
climates of various parts of the world have changed drastically
many times. Studies in some regions show that they have gone
through several cycles of change. The recent studies of radioactive
carbon in buried wood and soils have greatly foreshortened our
geological time scale. Drastic differences must have existed in
many areas less than 9,000 years ago; and there have been highly
significant cycles within this recent period.

30 THE FUTURE OF ARID LANDS

Drought prediction for a few years ahead has great immediate
interest. Although one can arrange the data from previous years
by cycles, up to now none of these can be projected into the
future as a sound basis for prediction. Yet the problem is so im-
portant that active research is continuing, and the chances for
such long-time predictions steadily improve with studies of the
mass movements of the upper air and the factors which influence
them.

Recently some people have tried ‘‘to do something” about
climate, or rather about the weather. We have no firm basis for
guessing how important cloud seeding may turn out to be, nor
will we until the scientists in this field have spoken more clearly
than they have so far.

Geomorphology

The land forms of arid regions and the processes by which they
are modified differ importantly from those of well-watered regions.
Only in recent years have there been the detailed studies neces-
sary to forecast the evolution of land forms well enough to plan
enduring engineering measures and land-use schemes with reason-
able certainty. In fact, world history is crowded with examples of
such failures to predict landscape changes that ruined the works
of man. Some of these failures were inevitable. Irrigation works
and cities were located in places where it was inevitable that they
would be destroyed by the natural processes of landscape change.
Other failures were stimulated by overuse of the soil or by failures
to build or to maintain simple protective works that would have
been practicable for controlling the movement of water and of
soil by wind or water.

The geomorphologist has the task of sorting out the relatively
stable from the relatively unstable landscapes and of predicting
the changes that will be induced by changes in water courses and
in the use of the land.

Because of the low rainfall, arid soils generally have a sparse
plant cover; thus normally the soil is exposed directly to wind and
to running water. The fine particles of dry soil exposed to severe
wind blow away, often to great distances. The sand moves more

SCIENCE IN MAN’S STRUGGLE ON ARID LANDS 31

slowly with the wind and gets piled into a variety of deposits. An
exposed soil surface continues to lose fine soil until bedrock or
hardpan has been uncovered; or until enough stones have been
exposed to form a protective desert or erosion pavement.

Fine soil material exposed to the forces of running water, in-
cluding the infrequent but torrent-producing storms of the desert,
is carried down to the quiet reaches of the streams. Catastrophic
erosion with cutting and down-grading of whole regions can be
initiated by changes in the grade of streams or by some weakening
of the plant cover. Examples include earth movements that raise
the land above the normal grade line of the streams. The grade
of a stream may also be changed by cutting through one layer of
rock into another softer one or by earth changes that shorten its
course. The straightening of well-graded winding streams, for
example, may so increase the new grade of the stream that down-
cutting is stimulated through the whole of a drainage basin. Great
changes in climate toward desert conditions have brought about
devastating natural erosion. Even showers of hot volcanic ash
have killed the vegetation and initiated a new erosion cycle; so
has fire and overuse by grazing animals.

I should not want to minimize the effects of man in causing in-
stability of soils. But the tendency has often been to overlook
these natural changes and potential changes and to assume that
all active erosion has been caused by overgrazing or other wrong
land use. Yet where the erosion is in fact due to other causes,
grazing control or other simple soil management practices cannot
stop it.

Many of the predictions about the stability of landscapes have
come from historical research. Such research can be most helpful,
but only if done thoroughly. One can be led into serious error by
leaving out some of the critical factors. For example, some old
cultural ruins around the Mediterranean and in the Near East,
now covered with sand or erosion debris, were not in agricultural
regions. The original towns or forts were established primarily as
military posts and to serve transport routes. Although some agri-
culture, or rather gardening, may have been attempted on a small
scale, the processes of sedimentation that covered some of those

32 THE FUTURE OF ARID LANDS

ruins were not really influenced much by land use. Other cities
were located in places where their destruction was inevitable.

Scientists in this field are only now getting into good position to
sort out the unlike situations and the reasons back of them.
[See, as a good example, Leopold and Miller (4).] Obviously, re-
search in geomorphology is most productive when associated
with parallel researches in climatology, soil science, hydrology,
botany, and archeology.

We must avoid building dams and irrigation canals, for ex-
ample, in places where they will soon become choked with sedi-
ment or be undercut, regardless of our soil management practices.
Other less critical areas can be used successfully only if precau-
tions are taken to insure no weakening of the surface that would
initiate severe erosion. Still other areas are comparatively safe
and no costly precautions are necessary.

I should like to emphasize that such situations cannot be evalu-
ated in general or by superficial examinations. They are not
directly related to any one of the factors taken by itself. The only
reliable course is detailed investigation of all the features of each
drainage basin, one by one.

As a detail in this connection I should like to add this further
caution: We cannot always assume that all the soil erosion we see
is more or less equally responsible for the silting of our streams
and reservoirs; commonly it is the sediment from a few critical
stream banks or gullies that is doing most of the damage. In
such areas control measures need to be pinpointed, not generalized.
In areas of greatest erosion potential grazing may need to be
avoided altogether, although commonly light grazing of grasslands
gives a better cover than none at all.

Ground Water

Advances are being made in our knowledge of ground water
storage and recharge. Although a good vegetative cover is essential
to soil stability in many landscapes, it does not follow, as many
formerly assumed, that maximum vegetation gives maximum
water yield, either in the ground water or in surface storage.
Especially where the main source of mountain water comes as

SCIENCE IN MAN'S STRUGGLE ON ARID LANDS 33

snow, trees that hold this snow in the air, away from the ground,
actually reduce the water intake of the soil. Yet one cannot easily
generalize. For example, on a thin sloping soil over cracked lime-
stone, pine plantings may slow down runoff and give added water
to the substratum, thus restoring springs or wells in the adjacent
lowland. On a sloping deep soil of little rain, vigorously growing
eucalyptus trees can use so much water that springs and wells
in the lowland dry up. Nor does undergrazed grassland necessarily
give the best situation. Moderate grazing commonly leads to more
spreading of the plants and less chance for runoff between the
bunches of grass. Thus on many soils moderate grazing gives both
optimum production of livestock and optimum water yield.

Much interest is developing among engineers in the possibility
of control of underground water storage in contrast to storage by
dams. In hot, dry areas evaporation losses from artificial ponds
are enormous. If the water can be stored underground, these
losses can be avoided. But such storage is useful only where leaking
through deep cracks or contamination by salt can be avoided and
where pumping costs are not excessive.

Tritium may be a splendid new tool to help in the study of
water shortage.

Radioactive hydrogen with an atomic weight of three, or
tritium, is produced in the upper atmosphere by cosmic ray bom-
bardment and is brought to the earth as a component of rain
water. The longer the time the water vapor remains in the air,
the higher the content of tritium. Hence the normal amounts for
various regions differ, but they are fairly constant for any one
place, barring unusual disturbances. Tritium formation ceases
when the rain reaches the earth and that present in the water
decays at a known rate: its half-life is about 12 years. Thus by
comparing the tritium content of a water sample with that of the
rainfall in the same region, one can estimate the time since the
water fell as rain.

Libby and his co-workers (1) have made extensive studies of
water from wells, springs, rivers, and other sources with interesting
results. Since the life of tritium is short and its content in rain is
small, dating by this means is probably limited to periods under

34 THE FUTURE OF ARID LANDS

100 years. Rather elaborate techniques and apparatus are needed
and several factors affect the interpretation of the results; yet the
method holds promise for studying the fate of the rainfall in criti-
cal areas and for calculating the contribution of recent rainfall to
ground water, streams, and plant growth.

Despite some fair progress, we need more detailed data on
supplies of ground water, rates of recharge, and conditions aftect-
ing recharge. That is, we need more application to specific areas
of current methods for hydrological definition.

Still adequate use is not being made in many places of the data
already in hand. It has been definitely shown that ground water
is being used in some areas far faster than the recharge. This has
been explained to the people in these areas. In some areas the
people have developed local voluntary associations or legal
schemes to protect the water supplies. Good progress has been
reported from North Africa for example, where herdsmen have
worked out ways to protect their water supplies for livestock
and for feed reserves against the dry years.

In other areas, legal and administrative devices to deal with the
problem are lacking. In these, social research and invention are
lagging behind physical research and invention.

Soil

Ibn-al-Awan, the Moorish agriculturist of the twelfth century,
began his great book on agriculture with the sentence: ‘The
first principle of agriculture is an understanding of soils and of
how to distinguish those of good quality from those of poor
quality.”’ But despite such admonitions, many people have wasted
their lives trying to irrigate unresponsive soils.

It was not until about the end of the nineteenth century that
examinations of soil were made as a part of a more or less routine
evaluation of their potential use and capabilities. At first, texture,
slope, salt content, and wetness or depth to the water table were
the features considered. Even such limited studies were largely
confined to the young alluvial valleys where it was convenient to
use irrigation water. It is only in the last few years that studies
have been made of the old and stable upland soils of the desert

SCIENCE IN MAN’S STRUGGLE ON ARID LANDS 35

where the full effects of climate and living matter are recorded in
the basic morphology of the soil.

Since World War I researches on the soils of arid regions have
begun to throw light on their genesis, and considerable work has
been done recently in parts of Australia, North Africa, the western
part of the United States, and elsewhere. Important studies had
started before that time in what is now the Soviet Union.

Because of the unique environment, the characteristics of soils
in arid regions differ importantly from those of other regions. At
the recent Desert Symposium in Israel I summarized these dif-
ferences and their implications (3). I should emphasize, however,
that in the detailed appraisal of the potentialities of arid soils
we must continually recall that individual properties, such as
texture and structure, have a different significance from those of
soils of humid regions where much of the scientific research about
soils has been conducted.

Detailed examinations of soil characteristics in field and labora-
tory are especially critical to predictions about the use of arid
soils for irrigated crops because the soils are being studied in an
obviously different environment from the one in which they will
be used. The bringing of large quantities of irrigation water onto
the soil amounts to giving it artificially a humid climate. Soil
characteristics of little or no significance to native plants are very
important to the productivity and stability of soils under irriga-
tion. These include the porosity, texture, mineralogy, salinity, and
reaction of deep layers.

The combinations of soil characteristics that are most nearly
ideal for irrigation are not necessarily those best for continued
production of native forage without irrigation. For example,
nearly level, deeply pervious, salt-free loams of good structure are
nearly ideal for irrigation. A sloping soil consisting of about equal
amounts of fine soil material and basaltic rock fragments to several
feet would be useless for crops under irrigation; yet with the con-
centration of moisture around the roots it would probably give
double the yield of uncultivated forage plants.

Despite the great gaps in our knowledge we have learned a
good deal about arid soils in recent years. We now have reasonably

36 THE FUTURE OF ARID LANDS

good soil survey methods that have been tried out successfully
in several countries (7, 8). We have learned to sample soils for
laboratory work in relation to genetic soil horizons and geological
layers and with regard to the marked influence upon soils only a
few inches apart of both the kind of plants above them and of
small differences in water relations. These improved techniques
have made it possible to make the predictions from soil surveys
more quantitative and to extend greatly the use of data from
laboratories and small field plots.

Reliable methods are now available for classifying soils accord-
ing to defined kinds of soil that can be evaluated quantitatively
as to productivity, both without irrigation and with different
systems of management under irrigation.

Laboratory methods are now available for characterizing the
chemical properties and moisture relations of soil samples with
an accuracy that permits us to predict changes in the physical
and chemical properties of soils with irrigation (10). These new
technical tools, developed mainly during the last fifteen years,
make possible far better decisions between arable and nonarable
soils and between different systems of soil-water management
for specific crops.

The results of modern researches make it possible to determine
with reasonable precision the amount of moisture required at
different depths for the best growth of important crops. We now
have practical devices for measuring this moisture in the soil.
The tolerance of various kinds of plants for different degrees of
saltiness and alkalinity, under different sets of soil conditions, is
becoming known with some precision. We have even, perhaps, the
possibility of using these techniques for identifying more efficient
plants to use as breeding stock.

Suitable instruments for measuring soil moisture and for the
control of irrigation water now make possible well-controlled field
experiments for studying, on different kinds of soil, the interac-
tions among moisture levels, fertility levels of the several plant
nutrients, and plant spacing in order to give the data required
for calculating the most economical combinations of inputs with
variable costs for the individual items.

SCIENCE IN MAN’S STRUGGLE ON ARID LANDS Sy/

All these new results together make us realize that a large per-
centage of the water now being delivered to irrigated lands is
wasted, not counting losses through evaporation from reservoir
pools and from poorly engineered canals. In other words, farmers
could get a great deal more out of the water they are using, cer-
tainly twice as much; and I think they will do better after the
educational work catches up more nearly with the research results.

As a concluding remark about irrigated soils, I should like to
emphasize the great importance of maintaining high levels of plant
nutrients. Very few irrigated soils give good yields of crops with-
out abundant fertilization. Most of them need additional nitrogen
since arid soils are low in organic matter; many need additional
phosphorus; some need potassium fertilizer, especially after long
cropping; and many need one or more of the minor nutrients.

Since early times the horticulturist on irrigated land has been
plagued with chlorosis, the loss of normal green color in the leaves
of plants. We know that much of this trouble is due to deficiencies
of zinc, iron, and other minor nutrients. Only within the last three
or four years have the new chelates furnished an effective iron
fertilizer that may go a long way toward correcting this trouble-
some deficiency.

Since the addition and control of water are expensive, econom1-
cal production depends upon good husbandry in all other phases
of the farming enterprise.

Plant Distribution

Plants may be assumed to be in themselves indicators of the
combined effects of the growth factors in the spots where they are
grown. Even on uncultivated soils, however, the influence of the
type and condition of the vegetation as modified by use must be
evaluated along with the basic soil and climate.

Most plant scientists had formerly assumed soil (or the “edaphic
factor”) to be one of the fixed features of the landscape along with
climate and relief. Actually the plants themselves have a great
deal to do with the soil under them. In arid lands, for example,
some plants are salt collectors much more than others (6). Large
differences exist in the physical and chemical properties of soils

38 THE FUTURE OF ARID LANDS

only a few inches apart, depending upon the type of shrub above
them. The relationship between plant and soil in the natural
landscape is so intimate that rarely can we say that one is due to
the other: they evolve together.

Actually, useful plants (and useless ones) are distributed un-
evenly over the world. One need think only of the great value of
subterranean clover to Australia, and of the great harm done in
Australia by the prickly pear cactus from the United States, now
being brought under biological control by introduced pests.
Research is now going on in order to find out where other useful
transfers may be made.

Plant scientists are probably making their greatest contribu-
tions to more efficient grazing just now through improved plant
selection and better management practices. Perennials with
drought resistance, or at least drought tolerance under heavy use,
are being identified. Such plants include palatable shrubs as well
as grasses and legumes.

Research has found ways to increase these good species through
management and to spread some of them through seeding. For
others, practical methods of collecting seed are not yet available.
Although we must be on the lookout for good exotics, greatest
progress is being made with local varieties that have persisted for
a long time.

Along with selection and reseeding, where possible, is the
usually overriding matter of management, of controlled use.
Almost phenomenal effects of temporary rest or ‘‘guarding”’ of
the range to bring it into a high state of production have been
reported from many parts of the world, especially where adequate
measures for controlling water by contouring, terraces, spreading,
and the like can be installed. Such guarding for even two years
has raised the level of productivity of old ranges several fold; and
the new high levels can be maintained through controlled use.

Besides the normal cultural practices, the new methods for weed
control are proving to be important tools for the ranchers and
farmers in arid lands. Already brush is being eradicated from arid
lands for as little as three dollars per acre. Many differential
plant-killing hormones and other chemicals are in practical use,
and testing is going forward with thousands of new kinds. We can

SCIENCE IN MAN'S STRUGGLE ON ARID LANDS 39

confidently expect continued improvement in arid land manage-
ment through our ability to remove the useless plants that waste
water so that we may conserve it for the good ones.

Despite this progress, I feel that plant breeding will make as
great or even greater long-time improvement. Most of our out-
standing results from breeding new varieties, in contrast to selec-
tion alone, have been of plants to be grown under favorable condi-
tions of soil and water. Crops grown under irrigation in
arid regions, such as cotton, alfalfa, corn, and potatoes, have been
greatly improved through breeding. In unirrigated arid regions,
however, we shall be breeding plants for growth under severe limi-
tations of moisture.

We know that most of our plants are highly inefficient in their
use of water and light. As plants proceed with their basic function
—the manufacture of plant food by photosynthesis in sunlight—
they are required to take in carbon dioxide. In most of them, as
they take this in, water can escape. But the extent of this water
loss varies widely. Basic researches on these processes can give our
plant breeders new plant materials. Some plants can even take in
large amounts of water from humid air, such as that over the
plants during a cool night. Some plants can fix nitrogen, or at
least sustain other organisms that can. It is know known that this
ability is not limited to legumes like clover and alfalfa.

If characteristics of plants such as drought resistance, drought
tolerance, and low water requirement can be related to specific
genetic patterns, I see no reason why the plant breeder cannot
combine them with other desirable characteristics as he has
already done with disease resistance.

Thus besides selection and testing of plants collected from arid
lands, I feel that the basis for breeding plants for arid conditions
has improved. Such breeding on a significant scale, except for
lucky accidents, will be preceded by more fundamental research
of the genetic basis for the primary growth factors in promising
species and varieties. The newly emerging discipline, sometimes
called “physiological genetics,” holds great promise for giving us
far greater potentialities than can be had from plant selection
alone.

The long-time potentialities of such plant breeding, based on

40 THE FUTURE OF ARID LANDS

fundamental studies of physiological genetics, applies also to crops
grown under near-arid conditions. Here I am thinking of cereals,
including crops like grain sorghum, to be used for feed as well as
for food.

Other Research

So far I have spoken mainly about lines of research that are
directly related to arid lands. Before we come back to some of the
specific problems, we should remind ourselves of the many other
researches that may have important special implications in the
arid region.

Perhaps one of the first of these that comes to mind is the possi-
bility of low-cost production of potable water, or even irrigation
water, from the sea or other brackish sources. Active research is
going on in this field; but from what I can discover the industrial
chemist has a task to bring down the costs.

This problem is not unrelated to the future developments of
low-cost power. Wind power has already proved to be practicable
in many parts of northwestern Europe. It may have a place in
those arid regions with nearly continuous wind. Or perhaps the
current research on ways of converting sun power to electric
power may lead to practical inventions.

Then too, modern medical research is telling us a lot of new
things about human health in hot countries. These results in the
hands of the engineering specialist in air conditioning may greatly
increase the efficiency of labor and the ease of living in places
where most Europeans have not formerly been able to adjust
themselves.

Soil-Use Problems

Despite this reasonably optimistic picture of accurate methods
for predicting the harvest and the effects of management under
grazing, dry farming, and irrigation, our soils are not yet used
under anything like optimum sustained production. In other
words, the potentialities of the arid lands of the world are very
much higher indeed than our present realization.

Although good methods for research are available, they have

SCIENCE IN MAN’S STRUGGLE ON ARID LANDS Al

not been generally applied. We have either inadequate soil sur-
veys, or none at all, for a large part of the arid lands. Many
important kinds of soil are still very poorly understood, to say
nothing of their relations to vegetation and climate.

Lest what I have said so far lead to overoptimism, let us sum-
marize the soil-use problems under grazing, dry farming, and
irrigation.

Grazing

The productivity of arid range lands varies within wide limits
from essentially zero to something like 200 pounds of beef per
acre (or say 30 pounds of mutton and to pounds of wool), de-
pending upon the local soil conditions and effective rainfall.
Thus operational units of comparable size in terms of livestock
production vary enormously in terms of acres. For this reason
alone the units on poor soil tend to be too small for the individual
family or village, and this leads to overgrazing. Since soil moisture
is variable, there is a tendency to overstock in moist years and
for such overstocking to continue into average or drier-than-
average years with resulting overgrazing and damage to the
basic potentialities of the vegetation and, finally, even of the
soil itself.

Once serious overgrazing has taken place, the carrying capacity
falls far below normal and its recovery depends upon a tem-
porary sharp reduction in stocking together with reseeding, con-
touring, water spreading, and other measures to control the water
and keep it in the soil.

Then, too, the effective use of grazing land usually depends
upon sure supplies of feed for critical hot, dry, or cold periods.
This means that grazing land needs to be associated with irri-
gated land or with arable land in the mountains or in favorable
spots where hay or feed crops can be grown. Improved techniques
for wells and for surface water storage lead to more watering
places and thus spread the livestock evenly over the range. This
helps to control the use of the range.

In predominantly grazing areas generous amounts of water
should be allocated to use as livestock water and to irrigate

A2 THE FUTURE OF ARID LANDS

emergency feed crops. Effective local schemes are necessary to
give firm protection of water supplies for such use. In fact, water
and favorable soil for feed crops on a small area can stabilize the
grazing economy of a region several hundred times its size.

The economic, social, and administrative problems for range
improvement are really more difficult now than the purely tech-
nical ones. The areas most needing reduced grazing and careful
management are often the very ones where local residents most
strongly feel that they cannot reduce their livestock, especially
if marketing facilities are inadequate.

In large parts of the world, arid grazing lands could support
several times the present numbers of livestock if well-known ~
management practices could be installed during a temporary
period of reduced use. Even where experiments have demon-
strated greater production of livestock products from half or
less the numbers of animals, it is not easy to convince local herds-
men to reduce numbers.

Although we have made much progress in the basic and applied
research among the natural sciences related to range management,
much less has been done in the social sciences aside from produc-
tion economics. Associations of farmers and ranchers in grazing
districts and soil conservation districts have helped. Still these
devices fall short of meeting the problem in many places.

Dry Land Cereal Growing

During periods of unusual moisture and with strong economic
pressure, or local demand for food, many arid soils are broken
out for cereal grains. This happened during both world wars in
parts of the United States as well as in many other countries.
In parts of the Near East, cereal growing presses very hard
against the desert all the time.

Near the margin between arid soils and other soils, higher
yields on a more nearly sustained basis can be expected on deep
soils of good structure than on thin soils with low water-holding
capacity or on structureless sandy soils easily subject to serious
blowing. But once these thin or sandy soils are in cultivated
fields, it is hard to get them back into grass. Here again the most

SCIENCE IN MAN’S STRUGGLE ON ARID LANDS 43

difficult phases of the problem are economic. Let us suppose that
a farmer starts a cereal grain farm on a holding of some 600 acres
on soil unsuitable for crop use on a sustained basis. A great many
have done just that. In a few years when it is clearly established
that the soil cannot be maintained in production except under
grass, how is he to manage to enlarge his unit? A unit economical
for grazing would be much larger. During the transition period of
reseeding and low-carrying capacity, how is the farmer to live
and pay expenses? Large numbers of cultivators on the edge of
the desert in the Near East, for example, were born into such a
situation.

Despite advances in natural science, clear answers to the tough
problems of adjustment are lacking. More nearly appropriate
institutional techniques are needed.

Irrigation Farming

The improved methods for irrigation farming have already
been explained. It has been estimated that we have many more
millions of acres of arid soil in the world that could be developed
(3). Yet irrigation can be overdone. Where arable soil exists only
in small areas, intermingled with land suitable only for grazing,
the cutting out of these small irregular areas for industrial crops,
such as cotton and sugar beets, harms the use of the region as a
whole. It is by conserving such soil areas for hay and feed crops,
rather than using them for other crops, that efficient use can be
made of the range land.

Even at best irrigation is an expensive undertaking. If it is to
be done, it should be well done so that water is properly controlled
and so that other limiting factors, such as low fertility, salinity,
hazards of soil blowing, and waterlogging are avoided.

We have the scientific techniques to make these determinations.
We have fairly good methods for the economic analyses of the
results over short-run periods; but good methods for appraising
the long-run economic benefits and costs of irrigation are lacking.

Thus, given accurate soil surveys, current research results for
guiding the physical management of arid lands are more nearly
adequate to our present problems than those for guiding their
social management.

44 THE FUTURE OF ARID LANDS

Combined Resource Use

We know that the use of rural land depends also upon the use
of other resources. The more modern science and technology are
used, the more delicate becomes the balance among separate
lines of production and among individual resources (2). Even
similar kinds of land produce very differently in unlike economic
environments. An efficient agriculture at high levels of labor
income requires industry, transport, education, and medical
facilities in the same region.

Perhaps no example of the principle of combined use is more
striking than the growing competition for water in arid lands.
We must think not only of the competition within agriculture,
which I have already mentioned, but also the competition be-
tween agriculture, industry, and urban use. Where highly valuable
mineral deposits can be exploited only with methods requiring
large amounts of water, this use competes directly with irriga-
tion. We cannot assume offhand that it is wrong to use water for
industry and thus deny it to agriculture. Rather, we need to
develop through research, applicable criteria for appraising the
long-run benefits to society of alternative combinations of uses.

Only rarely can any single segment of modern economy succeed
by itself in any community or region. To do so requires that it
have an enormous relative advantage. For example, where an
area 1s so wholly agricultural that all public services must come
out of agricultural income alone, the soil must be excellent indeed.
Actually, some of the best agricultural areas of the world are
located on soils that were of only mediocre quality to begin with.
This is because agriculture generally thrives best where industry
and other resource uses contribute to the general economy and
share in the cost and benefits of the social services.

Thus there is more to the appraisal of arid lands than examina-
tions of the soils, vegetation, and water. We must look at the
other resources and the relation of their use to agriculture and
to social services. Obviously, we must include fuel and power for
industry and mineral resources tor mines and factories. Where
roads and social facilities serve two or more enterprises, eficiency
is far greater than where they serve a single one. And perhaps

SCIENCE IN MAN'S STRUGGLE ON ARID LANDS 45

above all is the question of how we shall divide the water without
its overuse.

At such an international conference it may not be out of place
to emphasize again the need for international cooperation and
joint planning. Many excellent opportunities can be realized only
by the pooling of resources, including water, that are arbitrarily
divided by country boundaries. For example, United States and
Mexico and the United States and Canada have worked together
for their mutual benefit. Going back again to the Near East, the
agriculturist can only hope for full peace and cooperation so that
the planning of combined resource use can become effective.

Summary

I have tried to create the impression that science has found out
a lot about arid lands. Progress has been substantial during the
last twenty years. But this progress has been mainly within
individual disciplines. It is not balanced; some fields have pro-
gressed more than others. Some lines of research have been weak
on the fundamental side. I should say that this is true in biology.
Here, for example, great potentialities for advances in plant
breeding depend upon more fundamental researches in genetics as
related to plant physiology. My experience leads me to the view
that the economic and legal researches, basic to improved institu-
tional devices, are less advanced than research in most fields of
natural science.

Great opportunities for science to serve the development of
arid lands lie in more emphasis upon basic research, in the wider
application of accurate methods for basic resource surveys, and
in the integration of research results from the several disciplines
in ways that make possible good predictions for combinations of
resource use in communities and regions. Perhaps we need more
team research. At least I think we do. And special emphasis should
be given to joint research with both natural scientists and social
scientists on the same team.

REFERENCES

1. Buttlar, von Haro, and W. F. Libby. 1955. Natural distribution of
cosmic ray produced tritium II. ¥. Inorg. and Nuclear Chem. London.

Io.

THE FUTURE OF ARID LANDS

. Kellogg, Charles E. 1950. Food, soil, and people. UNESCO Food and

People Series No. 6. Manhattan Pub. Co., New York.

. Kellogg, Charles E., 1953. Potentialities and problems of arid soils.

Desert Research. Proc., Jerusalem.

. Leopold, L. B., and J. P. Miller. 1954. A postglacial chronology for

some alluvial valleys in Wyoming. U. S. Geological Survey Water-
supply paper 1261, Washington, D. C.

. Meigs, Peveril. 1953. Design and use of homoclimatic maps: dry

climates of Israel as example. Desert Research. Proc., Jerusalem.

. Roberts, R. C. 1950. Chemical effects of salt-tolerant shrubs on soils.

Fourth International Congress Soil Sci. Trans. 1, 404, Amsterdam.

. Soil Survey Staff, 1951. Soil Survey Manual. U. S. Dept. of Agricul-

ture Handbook No. 18, Washington, D. C.

. Stephens, C. G. (Editor). 1953. Soil surveys for land development.

FAO Agri. Studies No. 20, Rome.

. Thornthwaite, C. W. 1948. Approach toward a rational classifica-

tion of climate. Geog. Rev. 38, 55.

U. S. Salinity Laboratory Staff. 1954. Diagnosis and improvement
of saline and alkaline soils. U. S. Dept. of Agriculture Handbook No.
60, Washington, D. C.

The Challenge of Arid Lands
Research and Development
for the Benefit of Mankind

B. T. DICKSON

Member, UNESCO Advisory Committee on Arid
Zone Research, Canberra, Australia

It is a pleasure on this occasion to remind the people of this
part of the United States that we in Australia owe much to the
wonderful work done by George and William Benjamin Chaffey,
those Canadian brothers who established Etiwanda as an irriga-
tion unit on the Santa Fé trail and later Ontario in California.
In 1886-87 they went to Australia to begin the irrigation programs
on the Murray. Where before had been arid land supporting
rabbits and a scattering of sheep, Renmark and Mildura de-
veloped the thriving centers of production.

Let us examine the title ‘““The Challenge of Arid Land Research
and Development for the Benefit of Mankind.” First, it is a
challenge, and a challenge indicates a contest or struggle, so that
sooner or later one ought to take a side. If we look at the end of
the title, which reads ‘“‘for the Benefit of Mankind,” there ought
not to be much doubt about which is the side to take.

Arid lands, wherever they may be located, especially if they
are hot, are characterised by intense blue skies by day, immense
distances in a shimmering atmosphere, sparse vegetation and still
sparser animal and human population, with a limited and erratic
rainfall of less than 10 inches in a good year. Yet we recall that
great civilizations were developed in arid areas crossed by great
rivers, the Tigris-Euphrates, the Indus, and the Nile.

48 THE FUTURE OF ARID LANDS

How big is the challenge so far as area goes? It is difficult to
get precise figures, but it is generally believed that the total
land surface of the earth is of the order of twenty-five thousand
million acres and of this, at present, about two thousand five
hundred million, or 10%, are under some form of cultivation.
But it is also estimated that about six thousand four hundred
million acres are arid, in other words, about one-quarter of the
total land surface of the earth. That is to say the arid area is Just
over two and a half times as large as the presently cultivated area.

Potentialities of Arid Lands

The end of World War I saw the end of large scale settlement
in, and development of, new lands by pioneering individuals in
whom the spirit of adventure or the desire for a freer life was
strong. Came World War II with a startling realization of the
precarious way in which most nations lived, a way fraught with
all the explosive possibilities of further bitter struggles between
peoples.

Today we know that Malthus was just ahead of his time and
we have to ask ourselves whether adequate food requirements of
the people can be provided from present sources with all the
technological experience available to us. We suggest that two
possibilities arise, one a reduction in the rate of growth of the
population of the world and the other an increase in production
of foodstuffs not only just to feed people on a minimal diet but
on a diet nearer to the optimal. With the control of the rate of
population growth we arid zone people have little or nothing to
do. We do say, however, with regard to the second, that it is
possible to increase world food supplies by increasing production
from present areas and by bringing into production additional
land at present uneconomically used.

By applying technological knowledge to the fullest extent in
what are at present underproducing areas, a great increase in
food production by individual growers is possible. India, for
example, possesses irrigation works on a great scale,.and nowhere
else in the world is so large a population dependent on irrigation
for food supplies. Nevertheless it is estimated that less than 10%,
of the runoff of her rivers is utilized.

RESEARCH AND DEVELOPMENT 49

When we come to consider the possibilities in arid lands, it is
recognised that the efforts of individual growers can play but a
small part in the development of new areas of production or the
re-establishment of production in great areas which used to be
fertile in ancient times and are now out of use. I refer to the mil-
lions of acres in the valleys of the Tigris-Euphrates, the Indus-
Chenab, and the Nile, in the dry north of Ceylon where still are
to be seen the remains of ancient irrigation channels and reser-
voirs. It is estimated that in Latin America over twelve million
acres are susceptible of development, and an area bigger than
Egypt’s productive land is said to be available in the middle
Niger. The re-establishment of these ancient areas and the de-
velopment of new areas require the joint efforts of research
institutions, of governments, and the instrumentalities of the
United Nations, such as FAO, WHO, and WMO, not forgetting
the international financial authorities.

Let us now look at some of the problems with which those
charged with arid land development have to deal. I propose to
list them under the following heads of water, soils, plants, animals,
and man, to give you what must obviously be a very sketchy
outline of some of the salient features of each, and then to come
to a conclusion.

Water

Water is placed first for obvious reasons, and adequate informa-
tion about the water resources of a region is essential to a safe
usage of that water for full development in food production,
domestic use, and industry. In the desert one conserves every
drop of water for the maintenance of life, in great urban centers
we turn a tap and water flows, but sometimes restrictions are
placed on turning those taps. In other lands the fresh water from
great mountain ranges flows several thousand miles to the sea
and constitutes a fresh water delta of tremendous size in the
ocean.

Rainfall

Whence comes the supply, great or small? The answer is rain-

fall, and this is the beginning of what is called the hydrologic

50 THE FUTURE OF ARID LANDS

cycle and the controlling factor in the hydrologic cycle. Some of
the rain soaks into the ground, some runs off in streams, some is
evaporated, and some is transpired by vegetation. That which is
evaporated or transpired by plants, like stream waters, may
travel great distances, but ultimately it again falls somewhere as
rain, snow, hail, fog, or dew. In tropical areas rainfall may actually
be beyond the capacity of man to use, as with the Amazon, but
in arid and semi-arid areas, rainfall is low, and every effort must
be made to conserve usage to the best ends. This applies even
when an arid area is using imported precipitation as in Egypt,
where the Nile valley is dependent in the main for its supply of
water on the rainfall in the Abyssinian highlands.

One of the first problems in the hydrologic cycle of arid areas
is that of accurately measuring precipitation. Because rainfall
is generally low, it can be very local and extremely sporadic, and
this erratic distribution both geographically and in time makes
for considerable discrepancies in accurate measurement.

While on the subject of rainfall, thought naturally turns to
the fascinating possibility of rain-making, and a preliminary
report by the World Meteorological Organization entitled,
“Artificial inducement of precipitation with special reference to
the arid and semi-arid regions of the world,” prepared by the
Technical Division of WMO Secretariat from reports received

(4) should be noted:

From the consideration of the regional reports quoted in this paper,
the following conclusions might be drawn:

1. Operations which have so far been carried out have produced re-
sults that could be termed, at best, inconclusive; neither the complete
failure of the methods employed nor the cértainty of getting substantial
increases of rainfall have been demonstrated.

2. The most favourable meteorological conditions for the artificial
inducement of precipitation are to be sought in regions and during sea-
sons where natural precipitation is most likely.

3. Present day techniques, either “cold” or “warm” cloud seeding,
have very little value, if any, in augmenting the precipitation in areas
of very low rainfall or during dry periods in regions of normally medium
rainfall.

These tentative conclusions should not be taken as the expression of
a negative attitude towards studies and experiments on the artificial

RESEARCH AND DEVELOPMENT -51

modification of clouds and precipitation. On the contrary, they indicate
that further effort is necessary.

It is therefore recommended that:

(a) new scientifically designed and rigorously checked experiments be
undertaken in all regions where there is a possibility of success;

(2) precise methods of evaluating the amount of precipitation result-
ing from such experiments be developed;

(c) in all research experiments and applied operations, the collabora-
tion of the meteorological authorities be sought, in order to ensure the
greatest reliability in conducting the experiments and assessing the
results;

(d) information on all projects already carried out or those in opera-
tion now be released and made available to all scientific workers, putting
development of science before commercial and other interests.

It is well known that advanced civilizations formerly occupied
what are today arid regions. Excavations show the remains of
reservoirs, canals, fortifications, and human habitations, pre-
suming an intensive and extensive cultivation by a large popula-
tion. Such is the case in India, Pakistan, North Africa, the Near
East, and even in the desolate region of Lob Nor in Central Asia
there are remains of oasis towns and irrigation works. The dis-
appearance of these civilizations may have been brought about
by severe climatic changes or by the doings of man not for the
benefit of mankind.

Tixeront, our colleague from Tunisia, in asking whether it is
possible to forecast weather over long periods, points out that
we have not yet sufficient, reliable records, but he states that in
Tunisia they use one of nature’s records in the growth rings of
trees and one of man’s in archaeological studies. The meteorological
service of Tunisia studied the climate of Ain Draham from 1736
to 1955 by examining the growth rings of an oak tree. The infer-
ence drawn was that from 1736 to 1790 there was significantly
more rain than later, and this seems to be confirmed from histor1-
cal documents which refer to abundant crops in the eighteenth
century. (See pp. 91-92.)

Tixeront considers that a study of the ruins of Arab and Roman
irrigation works, of historical texts, of the continuity of cultiva-
tion methods, and of the cultivated plant species indicates that
the climate is stable and has not become decidedly drier. Simi-

o2 THE FUTURE OF ARID LANDS

larly it was reported at the conference in Jerusalem in 1953
that all the historic, botanical and archaeological evidence pointed
to little change in climate during the last eighty years or so in
Israel and India.

Tixeront points out that droughts do occur and we ought to
become aware of their statistical probability, and he makes a
plea that we study all the climatic factors which may lead to the
development of drought periods, as conversely to flood times.

In practice in Tunisia information from historical and archaeo-
logical sources has been used to good effect. The planting of olive
trees at the same spacing as used by the Romans in dry farming
is successful today. So, too, wells, cisterns, and irrigation channels
of Roman origin give a guide to modern siting and use.

In semi-arid and arid areas transpiration and evaporation, or
to put them together as Thornthwaite does under the term
evapotranspiration, constitute the major factor in the recircula-
tion of rainfall into the air, but it is notably difficult to measure
satisfactorily because of differences in plant cover. It is becoming
clearly evident that in some instances an increase in crops or
pastures or tree cover may severely tax the capacity of ground
water supplies, and in general a balance has to be struck between
the needs of crops and the water supply. Lysimeter experiments
in South Africa indicate that only about 3% of rainfall goes lower
than the root zone of veldt grasses. Phreatophytes, plants which
grow their roots down to the water table like alfalfa, are notable
for their efficient transpiration. Indeed, I have on occasion recom-
mended the use of alfalfa to drain waterlogged orchards in irri-
gated country. Worthless phreatophytes may waste millions of
acre-feet of water in some arid country.

In order to assess the water requirements of an area, wherever
it may be, some way of measuring evaporation and transpiration
must be used, yet strange though it may seem, we have not yet
achieved exactitude even in reading evaporation pans. Thorn-
thwaite, who has devoted his scientific life to the study of climate,
has proposed a method of estimating the water need of a region
so that it is possible, if the rainfall and the Wee
are known, to determine how much additional water, if any, 1
needed by way of irrigation. He defines the total water need as

RESEARCH AND DEVELOPMENT Ss)

the amount of water which will return to the atmosphere from a
surface completely covered with vegetation when there is in the
soil sufficient moisture for the full use of the vegetation at all
times, and this he calls the “potential evapotranspiration.”
Geen 73:)

One way of determining evapotranspiration is by the “vapor
transfer’ method based on the rate at which air near the ground
is mixing with air above it at a given height and by measuring
the difference in water vapor content at the two levels.

It is possible also to measure rainfall, the inflow of irrigation
water, and the outflow water, regarding the amount which does
not run off as evapotranspired.

Latterly specially designed soil tanks 4 square meters in area
and 70 centimeters deep, in which plants can be grown under
field conditions, have been set up in a number of places, each
tank being surrounded by a large buffer area to ensure greater
accuracy in results, but not enough are yet in use to give the
range of variation from one area to another. Meantime Thorn-
thwaite and his colleagues have come to the conclusion that the
computation of potential evapotranspiration for any place can
be done from data on air temperature and latitude alone. With
these it is possible to determine the water needs of an area and,
as it were, keep accounts whereby the most economical use may
be made of irrigation water.

Underground Water

Having briefly referred to rainfall, using that as a general term,
we may naturally turn to what happens to the rainfall apart
from evapotranspiration. Some infiltrates the soil and other layers
and goes underground, where it may be stored or move slowly in
suitable layers gradually toward the sea. The most generally used
method of determining infiltration is to examine the data of the
use and fall of water in wells, although today radioactive isotopes
are available to make evident the movement of water through
permeable strata. There are in the United States about seven
thousand observation wells, and approximately 5% of these have
automatic recording equipment.

The examination of an area for underground water in the first

54 THE FUTURE OF ARID LANDS

place calls for geological knowledge and properly the use of
geologists to make the survey with or without the aid of geo-
physics. By this means exact records would be kept of all the
wells or bores put down, the strata through which they went
down, the quality and quantity of the water, and so on. Such a
survey would enable estimates to be made of the total volume of
the underground supply, of its area and depth in a confined source,
or of its flow if unconfined, and the region of flow in the aquifer.
It may be of interest to remind you that below the Nile there is
an underground river about 560 miles long reaching from about
80 miles south of Luxor to about 70 miles north of Cairo. Accord-
ing to Mohamed El Sayed Ayoub, one-time Inspector General
for Nile Control, the mean width of the stream is 10 kilometers,
the strata of sand and gravel in which it flows ranges from 100
meters to 300 meters in depth, with a water storage capacity of
nearly 500,000 million cubic meters, and the water takes nearly
100 years to arrive at the head of the delta. Each year 1,400 mil-
lion cubic meters are used for irrigation, another 1,000 million
cubic meters are planned to be used on 25,000 acres of a new irri-
gation project, about another 1,000 million cubic meters are used
by plants, and nearly 4,000 million cubic meters flow into the delta
unused.

This great aquifer under the Nile and the Nile itself receive
their water from distant sources, but were they to rely on local
rainfall for the infiltration and stream flow they would be dry
each year for six months.

The Thal Development Authority

I would like at this stage to tell you briefly the story of another
great arid area which is being reclaimed, the reclamation of which
illustrates regional organization of the order of the Tennessee
Valley Authority. I refer to the Thal desert area in western
Pakistan. It consists of a triangular area of nearly five million
acres with a base of 65 miles along the Salt Range to the north,
and a length of about 175 miles to the apex at the-south, and is
in the Punjab between the Indus, Jhelum, and the Chenab.

Tradition, supported by geological evidence, has it that the Indus

RESEARCH AND DEVELOPMENT 55

formerly flowed down the middle and deposited huge quantities
of sand and silt and later changed its course to the west. The
superficial sand dunes arise from fine material blown from coastal
and desert regions of Sind and Rajasthan.

The vegetation consists of low brush and scanty grass on which
camels browse. There are no indications of early occupation, such
as are found in other parts, prior to the fourteenth century, when
a few tanks of about an acre each were constructed by the Em-
peror Sher Shah Sun.

The question of developing the Thal area was first considered
in 1870. No action was taken until 1g0o1 when a Colonization Bill
authorized the construction of a canal to the Shamlat area, but
nothing was done until 1936 when the distribution of the waters
of the Indus and its great tributaries was considered. Work on a
Thal Project was begun in 1939 but was held up because of World
War II, and channels were filled up with sand when in 1947 the
flood of refugees from India moved into Pakistan. Of these 250,000
are being settled in the Thal.

In late August 1949 the Thal Development Authority was
established, to be responsible for the full development of an area
of 834,500 acres, with an area of 638,000 acres to be developed
by private enterprise with the assistance of the Authority.

It was believed that the agricultural development of the area
and the establishment of villages and small towns throughout the
area needed the balance of industrial development, and so today
there are sugar mills, cotton textile mills, a woollen mill, and a
cement factory in the area. Some 640 villages have been estab-
lished, each with forty or fifty houses on a total of 100 acres,
with a green belt all round each village and a timber area of
50 acres alongside.

Each settler is allowed 15 acres of land at not more than a mile
and a half from his village. These acres he must cultivate satis-
factorily.

The authority of the TDA originally covered the million and
half acres commandable by canal for irrigation, but in 1953 a
wider scheme was examined for parts of the three and a half
million acres not commanded by canals. In certain belts masonry

56 THE FUTURE OF ARID LANDS

wells have for years been used to supply water for small holdings.
The aquifer consists of sand layers with a water table at 40 to
60 feet in ample quantity, and so a tube well scheme has been
initiated. How successful this will be remains to be seen because
percolation is heavy and evaporation, with summer temperatures
up to 120°F. is high, but it is hoped that each well can irrigate
150 acres. Early in 1954 Australia supplied tube wells to the
TDA under the Colombo Plan.

It seems safe to prophesy that in perhaps a decade there will
be need to study a salt problem in parts of the Thal, and FAO
is already at work in Pakistan on this problem in the Indus valley.

In arid and semi-arid areas where the need for recharging the
underground water is acute there are often long periods of no
rain, interspersed with short bursts of storm rains with extremely
rapid runoff, carrying astonishing quantities of surface material
of sizes ranging from silt particles to boulders. These storm waters
are gone in a relatively short time, and the problem is how to
make good use of what are sometimes quite large supplies, by
spreading and slowing down the rush of waters, by the use of
dams and tanks, by selection of the site for percolation, and so on.

It is impossible in this composite paper to do more than indicate
the complexity of the problems concerning water supplies. The
U. S. Geological Survey has prepared a list of thirty ground water
problems needing research, and in arid areas the general problem
of the development of water resources to the fullest economic
capacity will always be a vital one.

Salinity Problems

Rainfall is relatively free from salts, and where rainfall is
adequate for agricultural production excess soluble salts in the
soil are leached away in the drainage water, but where rainfall is
low, leaching is reduced and salt accumulation can occur. All
irrigation waters contain salts dissolved from the rocks and soils
through which the water moves. Some years ago Scofield studied
irrigated areas in this part of the United States and described the
salt balance as the relation between the amount of salts being
delivered in irrigation water and the amount removed from the

RESEARCH AND DEVELOPMENT 5Y/

area in drainage waters (2). This concept backed by suitable
methods for its application may be valuable in preventing salting
and in remedying existing salt conditions.

Hayward, of the U. S. Salinity Laboratory at Riverside, has
prepared a most comprehensive review for UNESCO of research
on plant growth under saline conditions (3). In it he refers to the
classification of saline and alkali soils, the quality of waters for
irrigation, the physiological bases in plants for salt and alkali
tolerance, the effects on plant growth and on seed germination,
and then succinctly reviews the position in Australia, India and
Pakistan, South and Central America, and North America.

As human, animal, and plant bodies are so largely made up of
water it is little wonder that the ability of man to live is dependent
on having plenty of good water. Reference has been made to the
relationship between population and food supply. The same sort
of relationship obtains between population and water supply. It
is little wonder then that men think of those seemingly inexhaust-
ible supplies of water, the seas and oceans, and wish it were
economically feasible to desalt sea water in immense quantities.
On one occasion a sincere good wisher asked me whether it would
be possible to construct a canal from the Mediterranean through
the Negev desert to the Dead Sea using desalted sea water for
irrigating the desert and raising the level of the Dead Sea waters
with the drainage. The answer is that success in producing large
quantities of fresh water economically from salt water is not just
round the corner. There is no magic wand, but research is going
on in many parts, and there is little doubt that the day will come
when in some arid areas it will be possible to provide desalted
water at lower cost than, for example, water transmitted over
great distances.

Howe, of the University of California, prepared for UNESCO
an excellent summary of research on the utilization of saline
water and we have heard during this series of meetings from
Powell, a member of the Advisory Group to the U. S. Secretary
of the Interior on the Saline Water Conversion Program (see
p- 257.) This program was established by the U. S. Congress under
Public Law 448, and the research projects financed by grants in

58 THE FUTURE OF ARID LANDS

aid under this enactment are already highly productive, especially
in assessing the merits and costs of producing fresh water from
saline supplies. The Saline Water Program is due to end in July
1957, but I am sure we hope that it may be extended beyond that
date.

Soils

Soil, the base on which food production on a scale to satisfy
world needs rests, must now be considered. It is not proposed to
go into any detail about arid soils, because Kellogg can write
books about them. Suffice it to call attention to a few character-
istics such as their low content of organic matter and so of nitro-
gen, the fact that they are more usually alkaline than acid and
so may develop permeability problems with irrigation, and their
sometimes rather high content of soluble salts. Despite these
characteristics, with suitable fertilizer treatments crop yields
under irrigation can be remarkably high and large scale prosperous
communities may be established, as witness the position in the
United States and in Australia.

I cannot do better than quote Kellogg’s summary to his intro-
ductory paper at the Jerusalem Desert Research Symposium
(1) as follows:

Several important areas of needed soil research are the following:

1. Morphological study of the soils of relatively unexplored regions
and their classification as a basis for preliminary reconnaissance mapping
and appraisal. Previous soil experience has been highly concentrated on
alluvial soils and especially in areas easily reached by existing transport.
Soil scientists have had inadequate opportunities to make detailed
studies of arid soils remote from present population centres. Even in
places where water is scarce, principles of great fundamental value to
our knowledge of the formation and development of arid soils can be
learned—principles important generally and to the stabilization and use
of arid soils for grazing even though they cannot be irrigated.

2. Relation of soil permeability to drainage and salt removal. We
need to know more precisely the lower limit of permeability, especially
in subsoils and substrata, for satisfactory management with different
kinds and amounts of irrigation water, the factors that control the per-
meability, and how permeability may be modified by chemical treat-
ments, growing plants, and water management.

RESEARCH AND DEVELOPMENT 59

3. Development of structure and water-holding capacity in arid
sandy soils through use of better adapted green-manuring crops, organic
soil conditioners, or in other ways.

4. Better methods are needed for appraising the salt balance in whole
watersheds where irrigation water is taken from streams and the drain-
age water is returned into the streams to be used again for irrigation at
one or more lower levels.

5. More precise studies are needed of the soil properties that lead to
chlorosis in plants and of ways to modify them.

6. The reasonable alternative combinations of plant nutrients, water
supply, plant spacing, and cropping systems need to be tested in order
to find the most nearly optimum ones for each kind of soil in terms of
harvest, sets of practices, and the long-time effects on soil productivity.
Vine agulkss moe be eee onan ta de specific terms required for calculat-
ing costs and input-output ratios needed in farm budgeting.

Plants

In the section on climate reference was made to the fact that
there is little evidence of any marked climatic change for the
worse since man used or misused the land for living. The effective-
ness of the rainfall has however been seriously reduced by man’s
overuse of plough, the axe, and the grazing animal, particularly
the goat. Marginal areas have become man-made desert areas,
and it is in this sense that the desert is advancing. In any attempt
to re-establish marginal areas for better production and living
conditions it is necessary to survey and map the existing plant
cover, whether natural or cultivated, and also land use. These
are being done or are planned by FAO working with national
authorities.

The main task is to attempt the regeneration of a better plant
cover and to do this while the population is engaged in gaining a
living from the area. To study such natural regeneration as may
occur it 1s essential to have as guides in this work enclosed or
protected areas which are ungrazed. Sometimes the results may
be startling. One thing cannot be done, and that is go over from
one area in one part of the world to another area in another part
and at once begin to apply procedures in the expectation that they
will be successful. Much experimentation is needed to select suit-
able plants, to get them to seed, to germinate, and to grow in
these very old environments.

60 THE FUTURE OF ARID LANDS

It is dangerous to disturb the land surface more than neces-
sary, even with good intent, because of the possibility of soil
removal or seedling scarification by the sand-loaded winds which
blow in the hot season. I have seen this in the Thal desert of
Pakistan where we are making trials with a number of plants.
It seems to me that the provision of some shelter from these
searing winds is essential and this can begin around nursery
areas and in strips suitably placed across the prevailing winds.
The limitless horizons of the desert are very interesting to write
about but they are no good for proper land care.

I come from a land where it has been necessary to introduce
and establish every kind of plant food for man and his animals,
with the exception of the native grasses and top feed dry climate
trees like the mulga (an Acacia). We have successfully introduced
every kind of fruit and vegetable which can be grown anywhere
and we are still testing many grasses and legumes. So has the
United States, where I think trial introductions from all over the
world total more than 65,000. Outstanding in Australia has been
the success of the establishment of subterranean clover, which
has meant untold millions to the sheep men of southern Australia.
Phalaris tuberosa and rye grasses (Lolium spp.) in the south,
Cenchrus spp. in the west, and Rhodes grass in Queensland are
other illustrations. The United States and Australia use alfalfa,
called by Australians lucerne. (The American name is nearer the
Arabic which means the good plant.) I refer to this to illustrate
the possibilities of plant exploration for grasses and legumes,
particularly in old arid areas. Much must yet be done in this
field, and it is good to hear from Whyte about the work which is
in progress at the moment under his guidance in FAO (See p. 185).

In the detailed studies that are essential with respect to any of
these introductions there will be plenty of scope for selection and
genetical studies of those that show promise of establishment.
We have now, for example, a number of established strains of
subclover, and it may be that Trifolium hirtum, native of Turkey
and now established in California, will develop ecotypes.

It is obviously necessary to learn what are the physiological
factors which enable desert and near-desert plants to survive

RESEARCH AND DEVELOPMENT 61

long periods with limited water supply under conditions of ex-
cessive insolation, very high day temperatures and low night
temperatures.

Animals

Draz, Director of the Desert Range Development Project in
Egypt, also referred to the need for a thorough understanding of
the ecological, logical, genetic, and physiological bases which will
enable the selection of plants and animals most suited to arid
conditions, and drew attention to the importance in animals of
heat tolerance and heat dissipation, about which much is still
to be learned. He stressed the idea that it is short-sighted to look
down on local breeds which have become adapted to the condi-
tions of living in the environment (See p. 335).

Man

Whatever research work is done in any or all the fields we have
so far considered, the end result should be for the benefit of man.
It is appropriate, therefore, to think at this stage about man
himself, his well-being and living conditions. One striking feature
of man in the desert is his nomadism. While it may always be
that some movement of flocks must occur to and from grazing
areas, it does not follow that the shepherds remain nomads.
It is unnecessary to do more than remember how very different
the lives of men, women, and children are under nomadic condi-
tions from those enjoyed in the dry climate of New Mexico. Any
changes in modes of living must mean great social adjustments
for those people.

Ladell, Director of the Hot Physiological Research Unit in
Nigeria, dealing with the influence of environment in arid regions
(In a paper in press) points out the wide range in temperature
which man has to live under, as for example at Basra where the
mean monthly minimum varies from freezing to 83°F and the
mean monthly maximum from 67° to 1og°F. Under these condi-
tions there is also little cloud and scanty vegetative cover so that
the ground radiates heat, and dust-laden air radiates still more.
Add to this wind, and water loss from the body may proceed

62 THE FUTURE OF ARID LANDS

faster than physiologically desirable. He refers to heat acclimati-
zation, by which is meant the physiological changes resulting in
an improvement in work following exposure in a hot environment,
and he believes that man can live under conditions more severe
than occur in the hottest parts of the world. Protection from direct
solar radiation is desirable in the form of a light broad-brimmed
hat or a canopy on a tractor over the driving seat.

Good housing is essential—cool by day and of a kind to give
adequate protection at night. In this respect the thick-walled,
small-windowed, pisé houses of the Near East, Pakistan, and
India are types.

Water is the essence of life, and it seems a pity that because of
religious beliefs or tradition desert dwellers do not take to the
use of galvanized iron tanks to store roof water.

Desert dwellers may suffer from prickly heat and from mal-
nutrition and vitamin A deficiency which leads to slow healing
of wounds.

We who live in comfort find it hard to realize what life is like
without electricity or gas, without water in pipes, without re-
frigeration and radios, air conditioning in buildings, good roads
and fast cars, and so forth. But many thousands live where
hydroelectric power is not available nor where fossil fuels like
coal and oil can be used. So we turn to such sources of power as
wind and sun.

Successful wind-driven generators up to 70 kilowatts capacity
are operating in Denmark, and two prototypes of 10co kilowatts
are functioning in the United Kingdom. Much thought is being
given to automatic regulation in variable winds in order to make
the fullest use of wind power. It is most important to choose the
right site for a windmill.

The use of wind power to pump water either directly or through
the use of electricity should result in saving bullock power and
thereby acreage for human food.

Solar energy has already been put to work for cooking and
heating water and we saw demonstrations of equipment in action
at New Delhi, India, in November, 1954. The problem is to
reduce the cost to within the means of the average Indian. In

RESEARCH AND DEVELOPMENT 63

California hot water heaters have been installed on the roof with
insulated hot water storage tanks.

Considerable thought and experimentation is being devoted to
the possibility of developing a solar engine. Abbott of the Smith-
sonian Institution has long pioneered in this field.

Need for Scientific Teamwork

I have merely touched on a few points in a few of the many
fields of research in which we must achieve results.

In any scientific work it is first essential to survey the field. In
this case it is a survey of the fields. Therefore it is a highly complex
operation calling for teamwork of the highest order. I do not de-
cry individual effort. Indeed some of the greatest contributions to
knowledge have been made by the Newtons and Einsteins. But
in the title of this paper, ‘“The Challenge of Arid Land Research
and Development for the Benefit of Mankind,” it is impossible
to avoid the inference that we must have teamwork, and that
teamwork should be between individuals, between universities
and research institutions, and between peoples—in other words,
between United Nations organizations.

Particularly does it seem appropriate that we who belong to
those sections of mankind that enjoy the highest standards of
living should see a plain duty to help in every way possible to
benefit the less well-off sections of mankind. I often think how
wasteful it is that the billions of money are expended in defending
part of mankind against possible aggression by another part of
mankind when they could be spent in research work of this sort
for the benefit of mankind. Imagine our young folk being called
up for service and electing to serve for a period in one of the fields
we have been discussing during these meetings. But that is
Utopian.

It seemed to me that the title of this paper required me to be a
sort of missioner, and if I have succeeded in confirming in the
reader a determination to support in every way practicable the
efforts of scientists, the work of institutions and above all the
great work being done by UNESCO, FAO, WHO and others,
I am amply rewarded.

64 THE FUTURE OF ARID LANDS

REFERENCES

1. Desert Research. 1953. Special Publication No. 2, Research Council of
Israel, Jerusalem.
2. Scofield, C. S. 1940. Salt balance in irrigated areas, 7. dgr. Research
ls LIBRO.
. UNESCO. 1954. Reviews of Research on Problems of Utilization of
Saline Water. Paris.
4. World Meteorological Organization. 1954. Artificial inducement of
precipitation, Technical Note No. 1. Geneva.

VARIABILITY AND PREDICTABILITY
OF WATER SUPPLY

Questions

How predictable 1s precipitation in an arid region?

Are there distinct drought cycles?

What are the prospects for usable ground water occurrence in
arid areas?

What 1s the practicability of locating and estimating volume and
rate of natural recharge of underground water supplies?

Within a given watershed, to what degree can the water sources
and water yield be determined?

Climatology in Arid Zone Research

C. W. THORNTHWAITE

Laboratory of Climatology, Centerton, New
Jersey

The problems of the arid regions are climatic in origin and
stem from the imbalance between the water supply and the
water need. Two obvious proposals have been suggested in order
to relieve this disparity between water need and supply: (1) in-
crease the water supplies through artificial induction of rainfall;
(2) make better use of the existing water supplies by study of the
water requirements of crops and avoidance of overirrigation.
This paper will be directed toward the second—ways in which
existing water supplies may be utilized more effectively.

One cannot determine the amount by which precipitation fails
to supply the water needs of crops unless these water needs are
known. Thus, it is first necessary to determine the water need.
This most important climatic element is defined as the amount of
water which will return to the atmosphere from a surface com-
pletely covered with vegetation when there is sufficient moisture
in the soil for the use of the vegetation at all times. I have called
this the potential evapotranspiration (10, 17).

Measurement of Evapotranspiration

Precipitation is easily measured by means of rain gages and
has been recorded in most settled areas of the world. It is not
easy to measure evapotranspiration, however. In fact, no weather
service in the world yet determines this important element, and
the little known about its distribution has been pieced together
from various scattered determinations.

Scientists have tried various ways to determine the amount of

68 THE FUTURE OF ARID LANDS

water used by plants. Experiments which attempt to measure
the water loss from a leaf or a branch detached from the plant,
or from isolated plants in special pots, are highly artificial, and
generalizations from such studies have sometimes been greatly
in error. The only method that measures the evapotranspiration
from a field or any other natural surface without disturbing the
vegetation cover in any way is the so-called vapor transfer method
(14). Water vapor when it enters the atmosphere from the ground
or from plants is carried upward by the moving air in small
eddies or bodies of air that are replaced by drier eddies from
above. If we determine the rate at which the air near the ground
is mixing with that above it and at the same time measure the
difference in water vapor content at the two levels, we can deter-
mine both the rate and the amount of evapotranspiration.

This method is not easy to understand or to use. It requires
physical measurements more precise than are usually made.
However, the method can and should be perfected for it will
answer many important questions for climatology and biology.

There are other ways of determining both water use and water
need. In some irrigated areas rainfall, irrigation water, and water
outflow are all measured. The fraction of the applied water that
does not run off is the evapotranspiration. In a few isolated
places, mostly in western United States, irrigation engineers
have determined the evapotranspiration from plants growing in
sunken tanks filled to ground level with soil in which water
tables are maintained at different predetermined depths beneath
the soil surface (19).

Since 1946 increasing thought has been devoted to the problem
of measuring the water use of plants under optimum soil moisture
conditions, and an instrument has been developed and standard-
ized (4, 18). It consists of a large soil tank so constructed that
plants can be grown in it under essentially field conditions and
can be provided with water as they need it. The tanks are 4 square
meters in area and contain soil to a depth of approximately 70
centimeters. They have means for subirrigation from a supply
tank designed so that actual amounts of water used can be accu-
rately measured, or they can be irrigated by sprinkling from above.

CLIMATOLOGY IN ARID ZONE RESEARCH 69

The latter method proves to be much more satisfactory in prac-
tice. When it rains, any excess water drains through the soil and
is similarly measured. Thus, evapotranspiration can be deter-
mined as a difference since every other term in the hydrologic
equation is measured. A number of these evapotranspirometers
are now in operation in widely scattered areas of the world.
Many additional installations are needed if we are to understand
the variation of evapotranspiration from one area to another.
The analysis of the observations of evapotranspiration from
collaborators in various climatic regions of the earth has revealed
that it is easy to get erroneous answers, particularly in arid areas
where measurements give values that are likely to be too large
(5). The errors occur when the vegetation on the area surround-
ing the tanks is not the same as on the tanks or when the soil
moisture inside the tanks differs from that in the soil outside.
There are three possible sources of energy for evaporation or
evapotranspiration: solar radiation, heat that reaches the evap-
orating surface from the air, and heat that is stored in the
evaporating body. However, with no external source of energy,
the surface temperature of an evaporating body would quickly
drop to the dew point of the air and evaporation would cease.
Consequently, evaporation can occur as a continuing process
only while energy is being received from some outside source.
The sun is the original source of all energy that is involved in
the transformation from liquid to water vapor, but not all the
energy that is received from the sun is used in evaporating water.
Some of the incoming solar radiation is immediately reflected
from the surface back to the sky. For a vegetation-covered
surface about 25% of the incoming radiation is lost in this way.
Also a certain percentage of the incoming radiation 1s radiated
from the surface back to the sky, the amount depending upon the
temperature of the earth’s surface and on the sky above. This
amount is often between 10 and 15% of the incoming radiation.
After deducting the losses due to reflection and back radiation,
the remainder, which is known as the net radiation, must be
partitioned into three parts; that which heats the soil, that which
heats the air by convection through contact with the soil surface,

70 THE FUTURE OF ARID LANDS

and that which is utilized in evaporation. Recent measurements
have shown that when the soil is very moist more than 80% of
the net radiation is used in evaporation. As the soil becomes dry,
the evaporation rate declines and more of the net radiation is
devoted to heating the air and soil with little remaining for
evaporation.

The potential rate of evapotranspiration is realized only when
the area of the evaporating surface is large enough so that all the
energy for evaporation comes from radiation and none from
advection. Obviously, the area of a standard evapotranspirometer
(4 square meters) is too small and can give reliable values only
when it is surrounded by an extensive buffer area identical in
vegetation cover and soil moisture. If the area of the evaporat-
ing surface is large, the influence of the air passing over it becomes
small and solar radiation is the primary source of energy for
evaporation. Under these circumstances the atmospheric humid-
ity is unimportant. If the air is moist, the temperature of the
evaporating surface will rise to a point, above the dew point of
the air, at which the evaporation will just use the energy that is
available. Similarly, in dry air, rapid evaporation will lower the
temperature of the evaporating surface until the evaporation is in
balance with the available energy.

When a psychrometer is used to determine the humidity of the
air, a thermometer bulb is moistened and becomes an evaporating
surface. The water evaporating from the wet bulb thermometer
cools the bulb. The surface area of the bulb is small, and the
amount of water vaporized is very small. Heat flows into the
water film on the bulb from the warmer surrounding air, and the
evaporation process will reach equilibrium at a rate and at a wet
bulb temperature at which the energy appropriated from the air
is just sufficient to maintain the evaporation. Solar radiation
contributes almost no energy to this process. The water evapo-
rated from the wet bulb moistens the air, but the amount is so
minute that the effect on the moisture content of the air is
completely negligible.

The Piche evaporimeter is also small with an "evaporating
surface of approximately 13 square centimeters. The evapora-
tion from the surface of a Piche evaporimeter is likewise incom-

CLIMATOLOGY IN ARID ZONE RESEARCH 71

petent to raise the humidity in the air to any significant degree.
Here, most of the energy used in evaporation comes from the air.

The Weather Bureau Class A evaporation pan is 4 feet in
diameter, and on a summer day in a dry situation it may
evaporate 2 gallons of water. Solar radiation contributes an
important share of the energy for evaporation, the amount
depending on the turbidity of the water and on the albedo of the
pan, which varies greatly with type, age, and condition of the
material of which the pan 1s made. Additional energy for evapora-
tion is available from the air. The amount of water evaporated
from the pan will do little to modify the moisture content of the
air; but immediately over the water surface the humidity is
raised, the moisture gradient reduced, and the evaporation im-
peded. The extent of this influence depends on the rate at which
fresh air passes across the evaporating surface from outside.

An extreme example of the effect of evaporation on the humid-
ity of the air is taken from a series of observations made during
the period 1907-10 in connection with an investigation of the
evaporation from Salton Sea in the Colorado desert of California.
The sea had been formed accidentally in 1904 while the Colorado
River was out of control and pouring into the dry bed of Salton
sink. When the river was again confined to its channel in 1907,
Salton Sea consisted of a sheet of fresh water 45 miles long and
10 to 15 miles wide with 440 square miles of surface area. The sea
was expected to disappear in 10 or 12 years so 1t was recognized
as presenting an unexcelled opportunity to study evaporation on a
large scale in an arid climate (1).

Evaporation pans were exposed at various heights on towers
over the sea surface and over the land at several locations away
from the water. Unfortunately, the pans were not all the same
size and were not uniform as to height or exposure. Nevertheless,
the results for 1910, the only ones that I have been able to find,
are very interesting. Table 1 gives the annual total evaporation
for the calendar year. Four other stations located up to 40 miles
away from the sea gave results that are similar to those from
Salton Sea Tower 1. At Indio, the evaporation from a 2-foot
pan, 10 feet above the ground, was more than 200 inches.

The actual evaporation from Salton Sea was determined by

VP THE FUTURE OF ARID LANDS

TABLE 1
: F Height of Pan Diameter Evaporation
L :

Station ocation (ft) (ft) (in.)

Tower 1 1500 ft inland Ground 2 164.50
Tower I 1500 ft inland 40 2 193.44
Tower 2 500 ft at sea 2 4 108.65
Tower 2 500 ft at sea 45 4 137-75
Tower 4 7500 ft at sea 2 4 106.45
Tower 4 7500 ft at sea 45 4 140.02

measuring inflow, outflow, and change in water level. The results
for three years are as follows:

Period Evaporation (in.)
June 1, 1907—May 31, 1908 51
June 1, 1908-May 31, 1909 59
June 1, 1909—May 31, Ig10 69

The average annual evaporation for the 3-year period was about
60 inches.

The evaporation from the inland ground pan was more than
two and one-half times that from the sea and that from the two
pans mounted 2 feet above the sea surface, 80% greater than
from the sea. The similar evaporation from comparable pans on
towers 2 and 4 indicates that the rate of evaporation from the
sea is nearly uniform, beginning a short distance from the shore.
The much smaller evaporation from the 45-foot high pans over
the water than from the 40-foot pan over the land proves the
existence of a ‘“‘vapor blanket’ over the water and shows the
strong reciprocal relation between the moisture structure of the
air and the evaporation.

The evaporation from Lake Okeechobee in the humid climate
of Florida has been determined by careful measurement of pre-
cipitation, surface inflow, and outflow (3). Inseepage and out-
seepage were not measured but are believed to be small. The lake
is 600 to 800 square miles in area. At the same time, the Weather
Bureau has maintained a standard 4-foot evaporation pan at
Belle Glade, on the south shore of the lake (6). The comparative

mean monthly evaporation in inches is as follows:

J 1 AMIE NET] i BS O INP = ID Yieate
Dake 254) 2586" 4.20) 666, 5.98 5-360 16.045 263) 14-155 990 02 7 Ee ge
Pan ~ 322) 4-05 (15-79) 6264.77.25, (6.28 16.32. (6-36) 5533) 5-03) 6 3-77eg Oo lOR EOS

CLIMATOLOGY IN ARID ZONE RESEARCH 73

In this example, in a humid climate, where the lake evaporation
is 83% as great as the evaporation from the pan, it is seen that
size of evaporating surface is a much less important factor than
in an arid climate.

Thus, it is evident that measurements made with evapotrans-
pirometers are subject to serious error. To insure that the observa-
tions are reliable it 1s necessary to keep an area around the field
tanks under high soil moisture. The size of this buffer area varies
with the climate; it must be much larger in the arid climates
then in humid.

Conditions Affecting Evapotranspiration

Although the various methods of determining evapotranspira-
tion have many faults and the determinations are scattered and
few, we get from them an idea of how much water is transpired
and evaporated under different conditions. We find that the rate
of evapotranspiration depends on five things: climate, soil mois-
ture supply, plant cover, soil type and texture, and land manage-
ment. There is considerable evidence to show that, when the root
zone of the soil is well supplied with water, the amount used by
the vegetation will depend more on the amount of solar energy
received by the surface and the resultant temperature than on
the kind of vegetation growing in the area. Soil type and texture
and farming practices likewise have little effect on the rate of
evapotranspiration under high moisture conditions. The water
loss under optimum soil moisture conditions, the potential
evapotranspiration, thus appears to be determined principally by
climatic conditions.

Using the most reliable measurements of evaporation and
transpiration that are available a valid and practical relationship
between certain climatic parameters and potential evapotrans-
piration has been obtained. This relationship permits the com-
putation of potential evapotranspiration for any place from
information on air temperature and latitude alone. The relation-
ship is given and its use described elsewhere (11). Work is pro-
ceeding toward the development of a new formula that is based
on sound physical principles. In the meantime, the present em-

74 THE FUTURE OF ARID LANDS

pirical formula is being widely used in various water balance
studies.

Needed Inventory of Arid Climates

It has long been recognized that in any adequate program of
desert research a first step would be an inventory of the dry
climates of the world. In 1951, at the first session of the Advisory
Committee on Arid Zone Research in Algiers it was recommended
that maps of the arid climates be prepared. As a result, Meigs
drew up for UNESCO a series of homoclimatic maps of the arid
zones which delineated the basic arid and semi-arid regions by
means of the moisture indices of my 1948 classification. These
maps were published in a provisional edition in 1952. They
represent a starting point in such an arid zone inventory but they
fail to provide all the information possible from such a mapping
of arid and semi-arid regions, information which must be made
available for a complete understanding of arid zone problems (12).

Water Surplus and Deficiency

Since rainfall and evapotranspiration are due to different
things, through the year they are not often the same either in
amount or in distribution. In some places more rain falls month
after month than the vegetation can use. The surplus moves
through the ground and over it to form streams and rivers and
flows back to the sea. In others, the rainfall is deficient in one
season and excessive in another, so that a period of drought is
followed by one with runoff. In still other areas, month after
month, there is less water in the soil than the vegetation could
use if it were available. There is no excess of rainfall and no
runoff, except locally where the soil cannot absorb the rain water
as it falls. Consequently, there are no permanent rivers and there
is no drainage to the ocean.

From a comparison of the monthly march of precipitation
with potential evapotranspiration at different stations, it 1s pos-
sible to obtain a clearer picture of the periods of water surplus
and deficiency and to bring into perspective the nature ot the
water problems in an area. Figure 1 shows the march of precipita-

CLIMATOLOGY IN ARID ZONE RESEARCH 75

inches

Albuquerque, New Mexico

Barahona, Dominican Republic

Gaza, Egypt

Zeerust, Union of South Africa

“

O 9
J FM AMJJ AS ON D J

©---o Precipitation EEF] Soil Moisture Utilization
o—— Potential Evapotranspiration SSSN Soil Moisture Storage
X—-=X Actual Evapotranspiration Water Deficit

Figure 1. The water budget at four selected stations.

tion and potential evapotranspiration at four places in arid or
semi-arid regions of the world. Because of variations in precipita-
tion at these four places there is a considerable variation in the
periods of moisture surplus and deficiency which they display.
At Albuquerque, average precipitation never exceeds 1.§ inches

76 THE FUTURE OF ARID LANDS

a month, whereas the summertime need for water is as high as
6 inches a month. Only in the winter does the meager precipita-
tion equal or exceed the reduced evapotranspiration, and a slight
storage of moisture in the soil occurs. The moisture deficiency in
summer ts large, being over 21 inches. No form of moisture con-
servation would make agriculture possible here. Irrigation would
be an absolute necessity.

Precipitation is highly variable at Barahana, Dominican Re-
public. It is less than 2 inches per month in the winter and again
in July, and more than 6 inches per month in May and October.
As a result there is a considerable deficiency of moisture during
both the summer and winter, amounting to over 24 inches for
the year, while there is some soil moisture recharge in spring and
fall.

Gaza is the only Egyptian station which has an average water
surplus at any time of year. On account of its Mediterranean type
of climate, the precipitation at Gaza is concentrated in the winter
half of the year and exceeds the water need of that period by a
considerable amount. As a result there is storage of moisture in
the ground and a water surplus of 2.7 inches in January and
February. However, due to the lack of summer rainfall and the
high summer water need a moisture deficiency of over 27 inches
occurs in summer. Thus, in spite of its more than 15 inches of
precipitation a year, agriculture without irrigation is extremely
hazardous because of the poor distribution of rainfall through the
year. At Zeerust in the Union of South Africa, the march of pre-
cipitation closely follows that of water need. In no month does
precipitation equal need so that there is no soil moisture storage.
However, because of the fact that precipitation and need are
nearly parallel and fairly close, the area chronically suffers from
hidden drought. Agriculture may be possible without supple-
mental irrigation but would be far more successful with it.

The values of moisture surplus and deficit have been combined
to form a moisture index which is the basis for the division of land
areas into moisture provinces. The use of such an index in deline-
ating moisture regions is a well-accepted practice. The need of a
second index, the annual potential evapotranspiration itself, to

CLIMATOLOGY IN ARID ZONE RESEARCH Ul

define climatic provinces is not well understood. Recent studies
have indicated that crop growth is closely related to potential
evapotranspiration, or the water use of plants (15). Thus, the an-
nual potential evapotranspiration provides an index of the growth
potential of an area. A single parameter, annual potential evapo-
transpiration, because of its dependence on the energy balance,
can serve both as a moisture and a thermal index.

The indices are important in defining world climatic regions.
Their greatest value, however, lies in specifying at every point a
thermal growth souenclal and the degree of moistness or aridity
of a climate. In other words, the indices are continuously dis-
tributed about the earth and do not exist merely along inter-
provincial boundaries, as do Képpen’s limits.

As part of a task which the Laboratory of Climatology has
undertaken for the Office of Naval Research, maps are being pre-
pared of precipitation, potential evapotranspiration, water sur-
plus, water deficiency, and the moisture regions for all parts of
the world on a scale that is consistent with the density of the
climatic network. As an example, maps of Japan, Korea, and For-
mosa were made in scale 1:1,000,000. Maps of the entire con-
tinent of Africa have been completed in color, using the new Amer-
ican Geographical Society base map on a scale of 1:3,000,000. It
is anticipated that the mapping of the entire earth will be com-
pleted within a short time. These large-scale maps give a clear
picture of the distribution of the arid and semi-arid regions and
add considerably to an understanding of their water balance
problems. For instance, from such detailed maps it 1s possible to
work out exactly the overall irrigation needs and the water sup-
ply that is available from precipitation. Such information is basic
in any intelligent evaluation of the water resource potential.

Variability in Climatic Factors

The inventory of the arid and semi-arid regions of the world
should go farther than merely making available the average an-
nual distribution of the different moisture regions. It has long been
recognized that variability is one of the main characteristics of
dry climates. The reliability of precipitation becomes less as the

THE FUTURE OF ARID LANDS

amount decreases, so that in dry regions one might expect great
yearly changes in the extent of the arid and semi-arid zones. As
part of the inventory it is necessary to determine the year-to-year
shifts in the extent of the climatic regions and to make available

formation in graphical or cartographic form. Good examples

are the maps showing the year-to
over the United States which I prepared

of Agriculture in 1

the in

year variations in climatic zones

for the U. S. Department

941 (g). From these data it was possible to de-
f the different climatic zones

f years each o

termine the number o

2 shows the frequency of arid

arid climates over the United States based on the period

occurred over different areas. Figure
and semi

f this type, when coupled with the de-

formation o

tailed data on the moisture regions of the world
much more adequate basis for the interpretation of moisture prob-

lems than is presently available.

—39. Inf

Igol

would provide a

)

year fluctuations in the

f the broad-scale moisture regions it is

ar-

In addition to determining the ye

eographic distribution o
desirable to know the probabilities of occurrence o

oO
5

f values of the

other important factors which constitute the water balance. In

FREQUENCY OF CLIMATES, SEMIARID AND DRIER

ANNUAL

ORR)

PERCENT OF YEARS

ear] 100 SSE
[EES] 7-99 ZA \-25

= s5:-75 BSQ o

. AAA ANI yoy.

Frequency of semiarid and drier climates in the United

Figure 2.
States.

CLIMATOLOGY IN ARID ZONE RESEARCH 79

arid climates the moisture deficiency is the most significant value
than can be determined from the water balance bookkeeping pro-
cedure since it gives an indication of the severity of the water
problem and reveals how much water must be supplied in other
ways. It is thus necessary to carry the bookkeeping procedure one
step farther to determine the year-to-year fluctuations in the
moisture deficiency and to derive the probabilities of occurrence
of different levels of deficiency. Figure 3 gives the magnitude of
the water deficiency in the Missouri Valley area of the United
States for the period 1920-44. This region is one of critical impor-
tance in any research on arid and semi-arid zone problems as it
marks the transition zone between regions where agriculture is
feasible without irrigation and where it is not. Average annual
values have little meaning in this region. The limit of economi-
cally feasible agricultural production can vary so greatly over the
region from year to year that probabilities based on a number of
years of observations must be studied in any intelligent approach
to the problem.

Daily Water Balance Bookkeeping

It is possible to compare the precipitation with the evapotrans-
piration or water need on a daily as well as a monthly basis. If
this is done and a regular account is kept of the day-to-day addi-
tions and withdrawals from the soil moisture bank, it is possible
to determine when moisture is lacking and replace it through ir-
rigation. This insures that there is never any deficit of water in
the soil and also that there is no overirrigation and wasteful misuse
of the water resource. An irrigation schedule can be derived as a
natural result of the daily bookkeeping of precipitation and water
use.

Until recently the most common practice of farmers had been
to watch the crops for signs of moisture deficiency and to irrigate
an undetermined amount when such signs were recognized. This
practice was far from satisfactory, for when the crops showed
signs of distress, a considerable reduction in yield had already oc-
curred. The irrigation bookkeeping procedure eliminates this prob-
lem by telling when and how much to irrigate before any defi-
ciency exists in the soil.

S 5
Sie OWE, INOlrg BONE Wy ZO

ll, ==4
: —

— a;
7 |

60
—7]
——S,

\;
+

emi ama

itll |

SSS
U
°

——S

SS oe

{0} ——— ree
5

ee ——
—=>
== ee 9
—_—laS=
i.

it
~

|
|

oS (oo)
ce

Lee

Cee
O53
ee.
i, 284)
il tage 2
i Tm 225

of water defi

Tens] 3 F
Fate eas Pee N

TUL °

—————

Sr
'

4 poe
————SSSS
__——————SS>S>=
lil

4
60,
—<—$—— ———J
——
——SS—SS=S=
o...—
aapeegeceal

Figure 3. Magnitude

igu
ciency mor

defi
; lower left, deficie

CLIMATOLOGY IN ARID ZONE RESEARCH 81

The irrigation scheduling procedure which was developed for
use in the humid climate of New Jersey will also apply in arid
climates; but here the problem is even simpler than in the moister
climates because there 1s little or no need to be concerned over the
possibility of rainfall affecting the program (13, 16). Thus, it
would not be necessary to provide a “‘safety factor,’ and one
could schedule irrigation to bring soil moisture just back to field
capacity in the desired root zone each time.

Estimating Water Needs

In the arid and semi-arid regions of the world, agriculture 1s
not feasible without recourse to supplemental irrigation. It is
therefore of primary importance that reliable information on the
magnitude of the water need in these areas be made available.
Computations of the water need for the various climatological sta-
tions in the states of Utah, Colorado, Arizona, and New Mexico
are plotted in Figure 4. It varies from a little over 60 inches in the
hot, dry regions of southwestern Arizona to less than 15 inches in
the mountainous regions of Colorado. Irrigation engineers have
attempted to arrive at figures for water need from the few avail-
able measured observations of evaporation from pans. Meyer has
presented a map of the evaporation from shallow lakes and reser-
voirs which indicates values of over 100 inches a year in the south-
western part of Arizona and up to 80 inches a year in New Mexico
(7). Horton, analyzing records from Weather Bureau Class A
evaporation pans, found values of evaporation of over 120 and go
inches a year respectively in these same areas (2). Horton’s map
of evaporation from pans has been reprinted, but it is reprinted
as a map of potential evapotranspiration (8).

The high values of evaporation which have been found by
Meyer and Horton without due consideration of the effect of the
size of the evaporating area on the rate of evaporation have been
used to arrive at figures for water need. They have led to false
conclusions concerning the water requirements in these arid areas
and, in certain cases, have resulted in the excessive use of water
with damaging results to the soil. It is time for irrigation engi-
neers and others who are attempting to determine the water needs
and requirements of these dry regions to reevaluate the available

82 THE FUTURE OF ARID LANDS

2445 2465

26395025 20)

_ 2008 2039
Eke a +2830
2598 [2185679 243 hs 2602504 2465 \
2815 to) 2638 igso2 [= 2
6 * 2429 2567 :
21 21
856 2 (sla 24:09 2648
1 2284)" 2110 22896 3277 aha
] 25 jest W929 2189 /*2q25 2807
7 eet ge Fon 3028
a a e288 06
2923 2665° 0
2339 Za 2406 2072
ee 2237(fes 35 2829/4177 3173, 3158}
2595 of
ce 2740 274 0as0 {31 £6)
‘ 28 86 5x40] °2461 aes 30.00
‘ M 2835 aed, 3
sf 296 Sas ‘
25°28 ae 3\’30!
9 2 o
aioe 2567 3087

= sje8ss [3827

SS i

A Ga28 3209 3356 3496 }

92 6 =

|
| Average annual potential evapotranspiration in Utah,
| Colorado, Arizona and New Mexico, in inches.

Light figures, potential evapotranspiration computed by T-E method
Bold faced figures, consumptive use, from various sources

Figure 4. Distribution of average annual potential evapotranspira-
tion in Utah, Colorado, Arizona, and New Mexico.

information on the basis of these known limitations and to read-
just their estimates of irrigation requirements and of the water
resources of the arid areas.

One of the objects of this paper has been to show the usefulness
of the concept of potential evapotranspiration in arid and semi-

CLIMATOLOGY IN ARID ZONE RESEARCH 83

arid research. For instance, with this concept it is possible to de-
termine the water needs of an area and to evaluate the utility
of the water supply to meet those needs. Through its use in the
classification of climates, it becomes basic to any inventory of the
arid regions of the world. It 1s possible to use it to work out a com-
plete and accurate irrigation schedule that will permit the maxi-
mum returns with the water available. Thus, it becomes more
abundantly clear as studies of the concept of potential evapo-
transpiration progress that it is a tool of increasing usefulness in
the solution of problems not only in humid regions but also in the
potentially important arid and semi-arid regions of the world.

REFERENCES

1. Bigelow, Frank H. 1907. Studies on the phenomena of the evapora-
tion of water over lakes and reservoirs. Monthly Weather Rev. 35,
311-16.

2. Horton, R. E. 1944. Evaporation—Maps of the United States.
Am. Geophys. Union, Trans. of 1943. Pp. 743-753-

3. Langbein, W. B. 1951. Research on evaporation from lakes and
reservoirs. U.G.G.I. Assoc. Intern. a’Hydrologie Scientific. Assemblée
Générale de Bruxelles. Vol. 3, pp. 409-19.

4. Mather, J. R. 1950. Manual of Evapotranspiration. Mzcrometeor-
ology of the Surface Layer of the Atmosphere, Supplement to Interim
Report No. 10, The Johns Hopkins Univ. Lab. of Climatology,
Centerton, N. J.

5. Mather, J. R. (Editor). 1954. The measurement of potential evapo-
transpiration. Publications in Climatology, Vol. VII, No. 1. The
Johns Hopkins Univ. Lab. of Climatology, Centerton, N. J.

6. Mean monthly and annual evaporation from free water surfaces.
U. S. Weather Bureau, Tech. Paver No. 13, July 1950.

7. Meyer, A. F. 1942. Evaporation from Lakes and Reservoirs. Minn.
Research Committee. Minneapolis.

8. Reichelderfer, F. W. 1952. Physical basis of water supply for the
United States. In H, The Physical Basis of Water Supply and Its
Principal Uses. Pp. 11-24. Interior and Insular Affairs Committee,
U. S. Congress.

g. Thornthwaite, C. W. 1941. Atlas of Climatic Types in the United
Sines Wee 40), UW, S.dDeon oF Zhi, SiGSo, Mose. I-02. INO, 21.

10. Thornthwaite, C. W. 1946. The moisture factor in climate. Trans.
Am. Geophys. Union 21, No. 1, 41-48.

11. Thornthwaite, C. W. 1948. An approach toward a rational classifi-
cation of climate. Geogr. Rev. 38, No. 1, 55-94.

19).

16.

THE FUTURE OF ARID LANDS

Thornthwaite, C. W. 1953. The water balance in arid and semiarid
climates. Desert Research Proc., Inter. Symposium of Research

Council of Israel and UNESCO, Jerusalem, pp. 112-135.

. Thornthwaite, C. W. 1953. Climate and scientific irrigation in New

Jersey. Publications in Climatology, Vol. VI, No. 1. The Johns Hop-
kins Univ. Lab. of Climatology, Centerton, N. J.

. Thornthwaite, C. W., and Benjamin Holzman. 1942. Measurement

of Evaporation from Land and Water Surfaces. U. S. Dept. of Agr.
Tech. Bull. No. 817.

. Thornthwaite, C. W., and J. R. Mather. 1954. Climate in relation

to crops. In Recent Studies in Bioclimatology, Am. Met. Soc. Meteoro-
logical Monographs, 2, No. 8, 1-10.

Thornthwaite, C. W., and J. R. Mather. 1955. The Water Budget and
Its Use in Irrigation. U.S. Dept. of Agr., Yearbook of Agriculture,
346-58.

Tale C. W., with H. G. Wilm and others. 1945. Report of
the Committee on Evaporation and Transpiration, 1943-1944. 4m.
Geophys. Union Trans. of 1944, pp. 686-93.

. Thornthwaite, C. W., with H. G. Wilm and others. 1946. Report of

the Committee on Evaporation and Transpiration, 1945-1946.
Trans. Am. Geophys. Union 27, No. 5, 721-23.

. Young, A. A., and H. F. Blaney. 1942. Use of water by native vege-

tation. Bull. No. 50, Calif. Dept. of Public Works, Div. of Water Re-
search.

Water Resources in Arid Regions

J. TIXERONT

Chief Engineer of Public Works, Tunis

In general we consider as arid those regions where water is
the usual limiting factor of agricultural production. In North
Africa, these regions are approximately those that have an
annual rainfall of less than 500 m/m (6). They present extremely
varied characteristics. A Tunisian arid area such as the Sahel
has a population density of more than 70 inhabitants per square
kilometer, although the annual rainfall is well below 400 milli-
meters. In the south, on the other hand, where the rainfall is
less than 100 millimeters, the population density is insignificant,
except for the oases.

Arid regions often draw their water resources from humid
regions on the outside. This is so, for example, in the Nile Valley,
in Mesopotamia, and in the valley of the Indus. This study is
limited to the cases of typically arid areas and to the water re-
sources contained within their perimeters.

The hydrology of arid regions naturally obeys the same physical
laws as the hydrology of every other region, but also has particular
traits to which we shall limit ourselves. We realize that our experi-
ence is related especially to the arid areas of North Africa and,
to a lesser degree, to the rest of the Mediterranean basin. Ac-
cordingly, to that extent this study lacks generality. There will
be no discussion of the cold arid zones.

In the study of water resources, it will be impossible to con-
sider individually the agricultural and economic phases since
these are inseparable from the hydrology in arid regions.

86 THE FUTURE OF ARID LANDS

Effects of Variability of Climate on Economic Life of Arid Regions

Arid zones are characterized by lack of water and variability
of climate. This variability entails numerous economic, social,
and political consequences. The enumeration of some of these
will be useful for they may orient studies of hydrology and
climatology.

Nomadism

The rainfall must exceed certain amounts to be usable either
for agriculture or for cattle-raising. These amounts are not
exceeded everywhere at one time. The population must plant
where the rain has fallen and likewise transport their flocks there.

Nomadism is therefore obligatory for complete utilization of
very arid areas.

Storing

By storing is meant all the methods of storing resources or
products from one year or from one group of years to another.

Storing first has to do with the water supply, and in arid
regions it necessitates regulation continuing over several years
in the interest of the oaglereround aquifers which alone are
capable of resisting loss by evaporation.

Storing should also cover all agricultural products, the
principal ones in Tunisia being wheat, vegetables, oil, and fodder.
Stocking of these products requires the intervention of the state
or of powerful cooperatives, and seed supplies should be furnished
by the state in many areas.

In any case, material means for storing are not sufficient and
should be supplemented by a system of credit.

Exchanges

Despite all the methods of storing that an arid country can
utilize, its possibilities of demographic expansion remain very
limited. Beyond these possibilities, and without the importation
of water from surrounding humid lands, it is necessary to proceed
to exchanges of agricultural products with regions where the
climate is temperate or complementary. Such exchanges can be

WATER RESOURCES IN ARID REGIONS 87

within the network of the nation itself, of an association of states
permitting compensation, or on an international scale.

This raises inevitable political problems, and it can be stated
from the record that the prosperity of arid regions has been ex-
tremely sensitive to political conditions.

Economic Planning

In arid regions planning should be for long periods. The five-
year period, often used in temperate countries, is too short in the
arid zones. We give two examples.

Tunisia had established a plan for agricultural development
for the years 1949-53, based on statistics for preceding years.
The years 1942-48 had been very dry, but the years 1949-53
were extremely humid. Agricultural production increased con-
siderably, but it is very difficult to determine what part of the
increase was the result of the measures taken under the plan.

With the same thought (8) Mexico established a six-year plan
providing for a 50% increase in wheat production for the period
1953-59, 1n comparison with the period 1948-51. Argentina
aimed at a 27% increase in wheat production before 1958, taking
for base the production of 1947-51. Variability in climate can
cause as great variations, despite the fact that these countries
are larger than Tunisia. When these plans terminate, it will
perhaps be difficult to determine whether the objectives have
been attained; yet it is necessary to check their efficacy.

A rational procedure should include (a) planning for more
than 5 years, depending on the variability of the climate (20 or
30 years, for example), and (4) submitting the economic statistics
to climatic corrections resulting from a parallel study by statis-
ticlans, agronomists, and hydrologists.

Variability in Rainfall

Rainfall is the most variable element of climate and the one
that most conditions the economy of arid regions. Variability in
rainfall has repercussions on harvests, where dry culture is used,
on the volume of water reaching the water table, and on the
volume of runoff.

88 THE FUTURE OF ARID LANDS

One might ask whether, in desert areas, the mean annual
rainfall is of any interest. If we take a station such as Adrar,
where the average annual rainfall is 17.4 millimeters, there is
evidently little chance of that amount of rainfall in any one year,
the standard deviation being 16.7 millimeters. On the other
hand, there is much more chance that 174.0 millimeters will fall
in ten years. The annual average value has consequently little
bearing on the most probable value, but it gives an idea of the
amount of precipitation that is more realistic in less arid regions.
Such is the conclusion drawn by Dubief who has specially studied
the noalnancan@)):

The element which may most easily be considered is the annual
rainfall. The yield of crops depends not only on the total annual
rainfall, but also on the distribution of the rain by season and
even by days.

However, to simplify matters, the discussion in this section
will be limited to:

1. What is the probability of having, at a given point, a total
annual rainfall greater than a given amount?

2. What is the probability of having a given annual total rain-
fall over the entire country?

3. Can one forecast variations in the annual rainfall over a
period of time?

Rainfall at Given Point

Suppose that it is possible to make observations over a long
period of time, say more than 50 years. The variability of the
rainfall at a given point can then be represented during the
length of observation by the law of Galton Gibrat. If H is the
annual amount of rainfall in millimeters, HM the medium amount,
and HO the corrective amount, normal distribution will be the
logarithm of (H + HO). On a Gausso logarithmic diagram, we
obtain a straight line equation:

log (HW + HO) = ax + log (HM + HO)

Figure 1 gives the frequencies of the annual rainfall for Tunis,
and for the average of five stations scattered over all Tunisia

WATER RESOURCES IN ARID REGIONS 89

1000
mm.

500

Tunis
excluding
the desert

Tunis —

py
(o>)
(2)
Annual depth of
precipitation in mm.

150

100

O. Medjerda — ge

Runoff in mm.

O. Kebir

(3)

PH MNW RUD OO

C5 $0) 80 70 6) 50 40 30 20 10 5
Frequency in %

Figure 1. Top, frequencies of annual precipitations, Tunis and aver-
age of § stations, raised by 200 m/m. Bottom, frequencies of annual
runoff of two watercourses in Tunisia.

(the desert zone excepted). This figure also gives the frequencies
of the annual average runoff for two watercourses whose basins
receive an average of 500 millimeters of rain per year:

Oued Kebir, whose basin surface is 225 square kilometers, and
Oued Medjerda, whose basin surface is 16,000 square kilo-
meters.

Rainfall Over Entire Country

As indicated by the curve of Figure 1, relative to the average
of the five stations, a deficit of rainfall at one point can be com-
pensated by a surplus at another, and the variability of the total
rainfall on a territory is less than in any one locality.

90 THE FUTURE OF ARID LANDS

However, the variability of the annual rainfall is still very
large, even for a country of the size of Tunisia. It is easy to be-
come aware of this in comparing the distribution of rainfall during
two four-year periods, one very dry (1944-47), the other very
humid (1931-34) (Figure 2). If we assume that desert begins
where there is less than 200 millimeters annual rainfall, we see
that the limits of the desert moved some 200 kilometers in the
Sahel between these two periods. The four-year period was chosen
because it marks the influence of rainfall not only on annual
harvests but also on deep-rooted plants such as olive trees. In
fact, in the Sfax area, famous for its orchards, the olive trees in
1947 had lost all their leaves and if the drought had lasted longer,
great numbers would have died. Sand dunes were developing with
great rapidity all the way into the center of Tunisia. Harvests
of dry culture were of no account in the center and in the south.

The drought of 1944-1947, moreover, was felt outside Tunisia,
in Italy for instance (5). Hence the necessity, when studying
the variability of climate, to cover territories of different areas to
determine the extent of regions within which there can be an
economic compensation for the variability of the climate.

Forecasts on Variability of Rainfall

If the rainfall varies much from one year to another, it varies
also in the course of consecutive groups of years. Statistics give
some idea of this. For example, the Sfax station gives the data
in Table 1. The importance of the difference in the average five-
year precipitation illustrates what has been said earlier on the
subject of equipment plans. But, the statistics cannot be extra-
polated with certainty beyond the period of observations. Be-
yond that period one must have recourse to other methods of
forecasting, which leads one to interrogate witnesses of a more-
distant past.

The only methods employed in Tunisia are the examination of
tree growth and archaeological studies. The first method bore only
on a humid locality but, however, one neighboring to the arid
zone. The Meteorological Service of Tunisia has applied it to the
study of the climate of Ain Draham since 1736, in examining the
growth rings of an old oak knocked down in 1905, and in com-

WATER RESOURCES IN ARID REGIONS 91

Kairouan

a
200 ae

@89
Kebili

+
+
28 | +
@ Foum Tataouine ne
4931-1934 +

50
Scale in kilometers

Position of the Mean Annual

Isoheyet for the Periods
1931-1934 and 1944-1947

® 1564 Normal amount of precipitation,
1901-1950
Isohyet of 200 mm.
—— — Isohyet of 700 mm.
e+ Ghadames

Tunisia, Position of isohyetes in 1931-1934 and in 1944—-

Figure 2.
1947. Annual averages for 4 years.

92 THE FUTURE OF ARID LANDS

TABLE 1
Sian Minimum Maximum
Annual Average Rainfall " Depth, om Depth,
by period ear san an ea an/ian
I year 1946 62 1921 372
2 consecutive years 1945-46 67 1905-06 333
5 consecutive years 1943-47 II§ Ig18—22 263
IO consecutive years 1938-47 143 1905-14 229
20 consecutive years 1933-52 180 1903-22 224
50 consecutive years General average Igo1I-52, 197 m/m.

paring them with the rainfall during the period 1890-1905 (10).
The author of this study has inferred from it that the period
1736-go was clearly more rainy than the present period. Despite
all the reservations that one can have on the methods followed by
Ginestous, one 1s struck by the fact that the historical documents
actually make no mention of famines in the eighteenth century
(2), but do mention, on the contrary, abundant rains and harvests.
One can proceed no farther in Tunisia because there is a lack of
trees to observe, but the method may be developed in other
countries of the arid zone. Very old trees exist even in the center
of the Sahara.

The ensemble of historical and archaeological studies argues in
favor of a great stability of climate. For Tunisia, this conclusion
was drawn from numerous studies based on the examination of
ruins, of Roman and Arab water works, of historical texts, of the
constancy of methods of cultivation and types of plants raised,
and finally of the exceptional character of the two four-year
periods cited above. The same conclusion was arrived at in Israel
(a8) armel iva Ikachia, (a).

So the rainfall seems to have remained the same since antiquity,
with the exception of variations of the same kind as those that
we noticed at the present time. Can these variations be foreseen?

It must be noted in this respect that a dry year, even an ex-
ceptional one, does not result in famine nor does it constitute an
economic catastrophe, unless it is one of a group of consecutive
dry years extending over a vast territory. It is important to fore-

WATER RESOURCES IN ARID REGIONS 93

cast and to evaluate these periods of drought, as well as the periods
of abundance that present the same characteristics. Even if the
arrival of a given period cannot be predicted, it is useful to know
its statistical probability which has evident political and economic
consequences: a completely unorganized country will suffer
famines which could be avoided by an organization at the pro-
vincial level. More and more serious famines could be avoided
only by national, federal, or international organization.

There exist, in fact, periods of famine, of long duration, cover-
ing more or less extensive territories. They have meteorological
causes that should be studied in two phases: (a) the meteorological
situations which cause the periods of famine, and (4) the causes
of such meteorological situations.

The statements made regarding the historical permanence of
the climate lead to the supposition that variations in climate are
the result of causes which themselves fluctuate. Thus one pro-
ceeds to a study of the fluctuations in climate and to a study of
the fluctuations in all the geophysical or cosmic phenomena
capable of varying in a concomitant manner, choosing those
phenomena whose fluctuations can be forecast, such as cycles of
sunspots.

The simplest method has been to try to deduce periodicities
from the analysis of past observations and to suppose that they
will continue in the future. This analysis, to be convincing, should
be made by dividing the time into two periods for which one
possesses information. The first serves for analysis, the second for
verification. The conclusions are of value only if they cover a
group of stations for which there are available long-term observa-
tions.

To my knowledge these attempts have not yet yielded usable
conclusions. It is hoped that conclusions may be reached in the
future since documentation at hand is constantly increasing and
also the field of study of geophysical and cosmic phenomena is
widening.

In addition, the study of natural fluctuations in climate has
become more and more urgent in the past few years with the
appearance of possible causes of artificial man-made variations:

94 THE FUTURE OF ARID LANDS

artificial nucleation and atomic processes, whose effects can be
considered a part of the first plan. A new field of study of varia-
bility is opening. This will become more complicated because the
causes of natural variability will be less clear (14).

Since the artificial modification of rainfall is beyond the bounds
of this study, we make only one comment. The present studies of
the processes of nucleation aim above all at increasing rainfall
when the conditions favorable to natural precipitation have
already been met. These processes permit the making of one
general reservation. If they are effective, they risk, in fact,
increasing the variability of the climate, and their application
should consequently be considered under such conditions that
the processes will contribute not only to an increase of precipita-
tion but also to an increase in water reserves.

Water Resources

The water resources of a region, other than the rain absorbed
directly by the soil, are the portion ot the rainfall stored in under-
ground aquifers and the portion which runs off and which can be
stored in the soil by dispersion or in surface reservoirs. All these
fractions are interconnected. For convenience of discussion, how-
ever, we shall look at them successively. But first it is necessary
to call attention to several fundamental characteristics of the

hydrology of arid regions.

General Water Balance

Tunisia receives per year: 32.5 billion cubic meters of rain
water and o.§ billion brought in from neighboring regions, of
which 2 billions return to the sea, and less than 1 billion passes
by the underground aquifers to be used for irrigation or uselessly
evaporated—all this for a population of 3,500,000.

The fraction evaporated and recoverable for dry culture
amounts to some 30 billion cubic meters. It represents therefore a
much greater volume than that which is recoverable for irrigation.
But the importance of irrigation cannot be measured in cubic
meters of water. The fractions usable for irrigation are, in fact,
put to much better and efficient agricultural use, and, in countries
where the variability of the climate is the main obstacle to their

WATER RESOURCES IN ARID REGIONS 95

development, they alone may be regulated. The aims of the
development plans of Tunisia are: (a) increase of production by
seeking better utilization of rainfall by dry-land cultures;
(4) among the dry cultures, the development especially of the
tree cultures, since the root system of trees permits the exploita-
tion of deep soil layers and consequently maximum utilization of
soil water reserves; (c) independently of the economic means
enumerated earlier the exploitation of all reserves usable for
irrigation whethe: they be large or small, underground, or surface
reservoirs. The demographic expansion of the country being
what it is, none should be neglected.

Water Needs: Evapotransptration

Calculation of the evapotranspiration according to the method
of Thornthwaite (20, 21) has been done for Tunisia by Preciozi
(17). Because this method permits a more detailed analysis of the
climate than others, it is of interest, for we have noted the great
variety of conditions in arid regions.

One of its principal applications is, perhaps, the prediction of
water needs for irrigation. In Table 2 are the results of the
calculation of these needs, after Preciozi, for three Tunisian

TABLE 2
Stations
Tabarca Gabes Tozeur
Annual rainfall, m/m 1029 175 89
Climatic indices* of Thorn- B: B3 S: al EB; d a! EA! d bi
thwaite
Water insufficiency in July, m/m 150 166 214
Corresponding flow, liter/second/ 0.40 0.45 0.58
hectare
Water insufficiency through the 407 820 1082
year, m/m
Corresponding flow, liter/second/ 0.13 0.26 0.34
hectare

¢B,, humidclimate;E, arid climate; B3, mesothermal: evapotranspiration between
855 and 997 m/m; A!, megathermal: evapotranspiration more than 1140 m/m; Sz, _im-
portant humidity deficit in summer; d, very important humidity deficit; a’, summer
concentration of thermic efficiency, 48%; bi, summer concentration of thermic effi-

ciency, 51.9%.

96 THE FUTURE OF ARID LANDS

stations, one in the most humid region, the others in very dry
areas.

From our experience, it seems that the water needs thus calcu-
lated are too weak for Gabes and especially so for Tozeur. Cor-
rections would be necessary. One is evident: the water at Tozeur
is saltier than at Tabarca and the rate of irrigation must therefore
be increased. Other corrections would perhaps be necessary if
there were at our disposal more observations on the evapotranspi-
ration in arid regions.

Moreover, dry cultures are s pase tle in the arid zone by re-
ducing evaporation surfaces by plant spacing, or by concen-
trating on only a part of the land the water which falls on more
extensive impluvium.

Discontinuity of Procedures for Establishment of Water Resources

In the first part of this report, we stressed the variability of
rainfall. We limited our treatment of the topic to annual rainfall.
It is now necessary to push the analysis farther by showing the
discontinuity of the rainfall and its distribution over the year.

Not only is the rainfall discontinuous, but in general it falls on
non-saturated surfaces. The effects of rain on the runoff and on
the supply of underground aquifers are felt only after exhaustion
of a volume necessary to saturate the soils, or when the rate of
rainfall exceeds the rate of absorption by the soil.

Insufficiency of Climatological Information

In order to study the flow of a great river of the humid zone,
one can be content with several spaced-out measures, for example,
once a week, and thus obtain a sufficiently accurate evaluation
of runoff. The discontinuous nature of the flow in arid zones makes
much more necessary the continuity of observation. It is very
dificult to get an observer to maintain a continuous watch.
When he observes a watercourse for months without seeing any
flow, his attention is likely to relax, and he may absent himself
precisely during the several hours when there is something to
observe. This continuity of observation is even more necessary if
one wishes to study the silt flow of streams. Available observa-
tions are of insufficient duration precisely in those regions in

WATER RESOURCES IN ARID REGIONS 97

which observations of much longer duration are more necessary
than in humid regions.

As a consequence of the remoteness of the stations, the expenses
are very high and cannot easily be borne by the arid regions,
which are economically weak, without having recourse to aid by
more humid regions. Apparatus runs the risk of being destroyed
by nomadic peoples who, with no evil intent, consider a rain
gage an object worthy of attention in the desert, interesting to
dismount, and useful as a domestic apparatus. Some difficulties
would be resolved if cheap and particularly strong automatic
devices were used which did not require having observers in each
place. Be that as it may, so great are the difficulties that the
hydrology of arid regions, now in its infancy, will remain so as
long as the network of observations remains largely undeveloped,
especially in the mountains where the rains are most abundant.

In Tunisia we have tried to fill the gaps in the pluviometric
network by establishing an approximate map of the rainfall, on
the following bases.

The reports furnished by the stations of the network serve as
reference marks. Between these stations isopluviometric lines
have been interpolated by considering the variation in the rainfall
according to altitude and exposure, and by complementing these
means of evaluation by phytosociological information (g). The
advantage of the phytosociological method is that it integrates
the rainfall of a long period. It demands much care and time
because annual vegetation can have a different appearance ac-
cording to the year the survey is made. Therefore, perennial
plants are of peculiar interest. Anyway one must not neglect
any means of cross-checking. We consider, however, the result
obtained satisfactory. It completes, at modest cost, the quite
insufficient information on annual rainfall given by the clima-
tological stations.

The phytosociological method is not, after all, the only bio-
logical method to determine the amount of rainfall. Another
method consists of making, in the course of the seasons, evalua-
tions of leaf surfaces and of the evaporating power of the soil and
its vegetal cover (15).

98 THE FUTURE OF ARID LANDS

Archaeological and Historical Studies

It happens that most of the arid zones were, in the past, seats of
very advanced agricultural civilizations, remains of which are well
preserved. This is true in North Africa, in all the countries of the
Near East, in India, etc.

The disappearance of these civilizations receives two explana-
tions: the first attributes it to variations in the climate; the
second, to human factors. We believe that it has been shown that
the second is more probable. It is true for North Africa, and the
reclamation of Tunisian lands was very much facilitated by
using archaeologicalinformation. The planting of olive trees in the
Sfax region was decided upon because ruins of oil presses lay
scattered about the surface. As it has happened, the olive trees
have prospered in dry farming when they were planted with
the same spacing as that recommended by the Phoenician and
Roman agronomists.

At present, every year, many wells or water projects are carried
out by using as base the indications given by Roman ruins.
For example, in regions where no wells are found, but only ruins
of cisterns, cisterns must be constructed or water must be
sought at depths greater than those reached by Roman techniques,
which, however, took advantage of phreatic water levels by
means of wells more than 100 meters deep.

Irrigation, both by perennial or flood waters was practiced
also, with techniques that have survived in certain spots up to
our epoch and that are always useful of application. It therefore
pays to begin research for the development of an arid country
by establishing an archaeological survey directed toward water
resources. This survey 1s now made easier by using aerial photos
on which appear the limits of the fields, and traces of plantations
cultivated even many centuries ago.

Thus we know that in certain steppe regions the population in
antiquity was more numerous than it is now and therefore can
be increased.

Problems of Salinity

The complete utilization of water means the recovery and
reutilization of already used water. This part of the program is

WATER RESOURCES IN ARID REGIONS 99

limited by the increase in salinity which accompanies each
utilization. Its increase is especially great where evaporation
is stronger. Therefore in arid regions one must study the salt
cycle as well as the water cycle.

The recovery of water used in towns poses salinity problems.
Waters distributed for public use are charged with salt before
passing into the sewers by the following process: evaporation in
the course of urban utilization, consumption of salt by the
inhabitants, industrial waste products, and penetration into the
sewers of water coming down from saline water strata, if the
sewers are not watertight.

For Tunis, starting with drinking water having a saline content
of around 500 m/g per liter, we hope to arrive at sewage water
having a salinity of less than 1,500 m/g and usable for irrigation.

Surface Waters

Stream flow in arid regions is subject to evaporation which
excludes in many cases any possibility of utilization by surface
reservoirs when the annual rainfall is less than 300 m/m. More-
over, these reservoirs would have to operate on interannual regu-
lation over many years because of the irregularity of the runoff.
This irregularity is illustrated in Figure 1, where at the bottom is
given the distribution of the runoff of two Tunisian water courses
whose basins receive an annual rainfall of the order of 500 m/m.
and for which thirty years’ observations were plotted.

However, the study of runoff in the arid zone is fundamental:
(a) because the runoff waters can be stored up in the soil; (4)
because they contribute to the recharge of underground aquifers
for a more and more important part as one goes toward drier
regions.

Recovery of Runoff Waters

We shall return to point (2) when we study the underground
reserves. For the moment let us consider point (2). Runoff waters
can be recovered in the soil, either by practicing retentive means
where the rain falls, 1.e., at the origin of runoff, or by letting the

100 THE FUTURE OF ARID LANDS

water collect in streams and recovering it downstream on suitable
lands. The former method needs no special study of the runoff.
It presents only a limited interest because the quantity of water
reserved by evapotranspiration for dry farming will be increased
rather insignificantly, and also because in the arid zone, the places
where rainfall is most abundant are often mountains with poor
soil.

The second method is of much greater interest. It permits
utilization of soils of good quality, on which the complement of
water, as compared with the rainfall per surface unit watered, is
important enough to increase the frequency of harvest years.
It is the method of water spreading. To be applied scientifically,
it requires a study of the soils, in order to choose those which are
most likely to give the best agricultural eficiency and the best
retaining of the water. It is also obviously necessary to be familiar
with the conditions of runoff.

Relationship between Annual Runoff and Annual Rainfall

In humid regions, runoff is mostly dependent on soil saturation.
Since, during long periods the rainfall is greater than the potential
evapotranspiration, the runoff is characterized by great regularity.
It has been studied in numerous countries. Formulas have been
established for annual average runoff which in general give
satisfaction in the regions for which they were established.

Even more general formulas have been established (19) or the
documentation condensed into curves of very wide application

(13).
Let us consider a simple formula:
I == Aller”
where R = annual average runoff in meters,

H = annual average rainfall in meters,
a = a coefficient greater than 2.

To apply this formula we must have a map of the precipitation
of a given watershed in order to divide it into surfaces of equal
depth of precipitation and to arrive at the total of the partial
runofts thus computed.

WATER RESOURCES IN ARID REGIONS 101

The coefficients 4 and a naturally vary with the temperature
and the distribution of the rainfall in the course of the year.
By referring to the study of the basins of South Africa (24),
it can be verified that regions of summer rainfall yield less runoft
than regions of winter rainfall.

The available data on runoff are very scanty in arid regions
with rainfall less than 400 m/m. For lack of other elements, one
can thus try to use a formula of this type which represents rather
well the relationship of rainfall to runoff in Tunisia with a = 3
and 4 between 0.25 and 0.4; but this can in no way hide the at
present gross insufficiency of the basic data.

The high value of the coefficient a brings out the interest in the
mountainous parts of watersheds, which not only are more watered
but also have steeper slopes and a lower water retention than the
plains.

Runoff Conditions

The very notion of average runoff leaves room for the same
observations as the average rainfall. It would be of little use to
know an annual average rate of runoff without knowing its
distribution and variability. The runoff depends on many other
factors besides the annual rainfall.

In humid regions, the runoff is relatively continuous for long
periods. In arid regions, the flow presents continuity only if it
has been regularized upstream of the measuring point by under-
ground reservoirs. We reserve the study of this last case until the
section on underground water. In general, the phenomenon is
discontinuous.

If the rain falls on a saturated soil, there will be runoff, but
there can be runoff even if the soil is not saturated, if the intensity
of the rainfall is fairly great. Thus one can classify floods into
saturation floods and intensity floods.

Saturation of the soil by the rainfall becomes rarer in proportion
as the aridity of the region increases. However, even in the deserts
there can be times when the soil is saturated. This comes from the
fact that the rains, while rare, can be abundant, as indicated by
Table 3, which gives the annual hundred-year rainfall by extra-

102 THE FUTURE OF ARID LANDS

TABLE 3
Axncal Daily Rainfall
Stations R et Be Maximum 100-Year Length of Observation
ania’, ™M Observed, Rainfall,
(1) (2)mm (3) mm
Ain Draham 15 gy 171 230 (1) Normals: 1g01—-s50
Tunis 420 144 200 (2) Totality of the observa-
tion from 1890 to 1945
Motmatas 239 160 250 (3) Extrapolation observation
period, 1901-45
Gabes 175 103 150
Tozeur 89 63 58
Adrar 17 33 (4) (4) Period, 1926-50

polation according to Galton Gibrat’s method from the observa-
tions covering the period 1901-45.

On the other hand, there is not sufficient information on the
water-retaining capacity of arid soils. There is in Tunisia a great
difference in this regard between regions whose rainfall is greater
than 200 mm and regions where it is less. In the region of Oued
Kebir, the annual rainfall is around 500 mm, the soil is saturated
when the rainfall approximately attains 100 mm. In the Sfax
region, the annual rainfall reaches 200 mm. The soil depth 1s
sometimes very great, and evaporation can reach a great depth,
which can therefore bring an even more important deficit in
saturation. On the contrary, in the regions of rainfall of less than
200 mm and especially in the deserts, the retentive capacity can
be very weak because of the slowing down, by lack of water, of
the processes of soil formation. In fact, over vast areas the rocks
are bare. In any event the quantity of rainfall is so reduced that
it generally falls on non-saturated terrain and yet rarely do
streams remain several years without flow, even in the middle of
the desert. So it must be admitted that a flood takes place either
because the soil is saturated within limited regions of the basin
or because the chief factor in the flow is the intensity of precipita-
tion.

Dubief estimates that in central Sahara there is a flood when
the rate of the rainfall exceeds 5 mm with an intensity greater
than 0.5 mm. per minute.

WATER RESOURCES IN ARID REGIONS 103

It certainly would be too costly to maintain many hydro-
metric stations. A rational program of study should be based on
the following thoughts:

1. To establish a general pluviometric network much more
dense than the present one, chiefly in mountainous areas.

2. To equip for complete observations a small number of basins
of limited extent in order to make possible a detailed study.

3. To equip these basins with a sufficiently dense network of
self-recording rain gages.

4. To analyze for these basins the elements of the flow of each
individual flood.

Thus, for the basins studied, one could obtain increasingly
exact correlations between runoff and meteorological conditions.
The comparison of these basin prototypes would allow differentia-
tion of the effects of physiographical factors. One could then come
to know the rates of flow of other basins from the information
furnished by the posts of the general pluviometric network and
from the topographical, geological, and pedological elements.

In passing let us call attention to a difficulty peculiar to hydro-
metric measurements. Since the water courses are dry for long
periods, the devices for taking the water level (floaters or mano-
metric capsules) risk being blocked by sand. Equipment that is
well adapted to these conditions should be sought.

Erosion

We will not close the section on surface waters without giving
some consideration to erosion and the silt flow of streams in arid
regions.

The annual average erosion, calculated in tons of sediment
per square kilometer carried away by the waters, reaches im-
portant amounts in the arid Mediterranean regions that have the
most rainfall (say about 500 mm). It decreases in going toward
the desert areas (23), but in all places the great floods carry
waters heavily loaded with sediment, which causes all kinds of
difficulties.

We have even less information on erosion than on runoff.
The program of studies sketched above should be completed by

measures of sediment load and salinity of water. It is even more

104 THE FUTURE OF ARID LANDS

necessary than for runoff that there be a study of particular and
discontinuous phenomena. The intensity of rainfall 1s an even
more determining factor for erosion than for runoff, and statistics
of intensity should be established for periods less than or equal
to a season, because of the evolution of the vegetation in the
course of the year.

The more one moves toward desert regions, the more sediment
tends to be at the bottom of the streams and not in suspension.
Now everyone knows how difficult it 1s to evaluate flow at a
depth. It would seem that for deserts the topographical method
is capable of giving good results. Desert streams flow in general
only over a certain length of their course, and a topographical
survey would suffice to determine the solid deposit of a flood.
For reasons of economy this can be done only in strictly limited
cases.

Underground Waters

The rain divides itself after soil saturation into surface and
underground waters in a proportion difficult to determine, save in
extreme cases: very pervious limestone formations or quite
impervious basins, where the excess of rain goes almost entirely
to one of these two fractions. Particularly striking examples of
these extreme cases can be found in the Hydrological Year Book
of Israel. In the course of the hydrological cycle, there are some-
times transfers from one fraction to the other. In general, the
individuality of underground water bodies in arid zones is more
sharply defined to the degree that the deep aquifers are more
important than the shallow ones.

W ater-Bearing Formations

The permanent and autocthonous phreatic aquifers yield
important resources in semi-arid regions, but increasingly less
important ones as one progresses toward drier regions, because
these aquifers are subject to a more and more intense evaporation,
involving great soil depth. In very dry regions, the existing
phreatic water levels are generally fed by the underflow of great
streams, or by a resurgence of deep underground water. In

WATER RESOURCES IN ARID REGIONS 105

deserts, the hydrology of underground waters is rather a deep
hydrology.

Whatever their nature, the study of underground water can
begin only by ropes cal and geological recognition of water-
bearing formations, by the survey of springs, their depth (level),
flow, and all the hydrological indications on the soil surface.
Among these it is particularly useful to observe the deposits left
by evaporation of subterranean waters. These deposits are linked
with the circulation whose conditions could have been modified in
more or less recent times.

Geological, geophysical, and hydrological prospecting does
not dispense with boring. At present the system most employed
is drilling with rotary equipment, but it consumes a great amount
of water, and this increases costs.

For deep drillings with a rotary drill, it is dificult to identify
and localize the different layers encountered. Systematic use of
electric logging permits the determination of precise correlations
between the formations encountered in several borings. It is
hoped that this type of process will be more and more improved.

Water Balance

Recharge. Everything that has been said about the irregular
and discontinuous supply of runoff is true for the supply of under-
ground waters. The phenomena of saturation of alimentary out-
crops over a vast extent are particularly rare. However, we have
seen that they could take place, at variable intervals, according
to the aridity of the region under consideration, for instance at
intervals of the order of a century.

In semi-arid zones, the discontinuity of the supply is already
very obvious, as shown in Figure 3, which gives the fluctuations
of a well of the Menzel Bou Zelfa aquifer, where average rainfall
IS 450 mm per year.

This graph allows the identification of two very clear cases of
recharge.

In desert regions, rarer still are rains capable of saturating the
soil. There is another supply process which can take place at
much more frequent intervals. It is the supplying of underground

106 THE FUTURE OF ARID LANDS

Legend
¢ Observed depth
7 Observed movement _
--~ Supposed movement

Depth of water table in meters

1945 71946" 1947" 1948 ~911949)) 1950) 19510 = 1952S SSeoas
Figure 3. Movement of the water table, Larue Well.

aquifers by stream floods. These can occur several times per
year. It is clear that saturation and percolation can frequently
occur locally along stream beds and on the flooded adjoining
areas.

In Figure 3, we realize that current annual supplies are not
sufficient to offset pumping in the dry season. There must be
added to these the supply of exceptional sort which coincides
with an intense runoff. The efficiency of this supply process is
certain (22). The computation of the safe yield of one aquifer
cannot therefore be established without observations covering
long periods.

It is possible to obtain an idea quite rapidly of the relative
participation of these two processes of supply in semi-arid regions,
but for the great desert artesian basins one is reduced to mere
suppositions.

The methods that can be considered to increase the recharge
of aquifers differ according to the process involved. To obtain
scientific bases it would be necessary to follow over many years
the fluctuations in water levels by sinking piezometric tubes in
the zones of recharge, and mainly in the zones of privileged supply
which are neighboring to water courses and their alluvial fans.

But the increase of recharge can be faced in a more practical
way by utilization of runoff waters. Two processes have been
employed: the first consists in stopping the water in infiltration
basins, which leads to taking special caution for waters contain-
ing sediment to avoid clogging; the second consists in slowing
down the runoff in such a way as to increase the time spent in
passing over the recharge areas.

WATER RESOURCES IN ARID REGIONS 107

Finally there is another source of supply: overflows coming from
nearby aquifers. Thus in Tunisia there is a series of aquifers in
miocene sand formations whose water naturally springs along
watertight faults. The waters are partially used for irrigation.
The waters not used, joined with waters strained out and perco-
lated from the irrigations, reinfiltrate and contribute to the sup-
ply of a compartment downstream. Water is in this way trans-
ferred from compartment to compartment, but each passage to
the surface of the soil, or near it, corresponds to an increase in
salinity. Sometimes, therefore, it is better to use as well as possible
the waters in the upstream compartments, without troubling
much about the decrease in supply of downstream compartments.

Hydrological considerations of the salinity of the water have an
importance as great as the economic findings for the planning
of systems of irrigation. Suppose a cycle of events wherein the
water percolated from upstream irrigations is reused for other
irrigation. On hydrological grounds the full utilization of under-
ground water requires the construction of watertight irrigation
systems as soon as the excess waters come to percolate into water
strata which have too much saline content to be recoverable for
other irrigation. This is the case for the artesian aquifers of the
Tunisian south. As long as downstream water strata which receive
percolation from upstream irrigation can be reused without too
much salinity, the irrigation ditches need not be lined unless
otherwise warranted.

The recharge of an aquifer by another can also take place by
underground communication. If the aquifers are of small extent,
in regions not very arid, they can be studied by comparing the
chemical composition of the waters and by examining the hydro-
logical data (11). In the case of very extensive units such as the
miocene and cretacean basins in the south of the Saharian Atlas,
the problem becomes much more difficult and has not yet been
solved for lack of sufficient information concerning the geology,
the deep hydrology, and the topography of the land (3, 12).

Discharge. Apart from the underground water which returns
to the surface in a form rather easily measurable—springs and
wells—some is lost to the sea by underwater seepage or evaporates

108 THE FUTURE OF ARID LANDS

in regions where the water comes near enough to the soil surface
to be evaporated from the soil or transpired by the vegetation.
Evaporation, whether from the soil or by plants, results in soil
salinity.

In some regions, the ground water bodies lie below closed
topographical basins whose bottoms are occupied by salt lakes,
without surface water, for the better part of the year. These are
the chotts and the sebkhas of the south Mediterranean regions.
They are filled by very fine silt. Their flat surface is covered with
salt. When soundings are made, water saturated with salt is
found under the salty crust, but if deeper soundings are made,
it is not uncommon to encounter layers of less salty water. The
existence of underground circulations through the silt is proved
in Chott Djerid by the existence of little springs and of natural
chimneys whose appearance from the air is shown in Figure 4.

A volume of water is also lost by defective catchments and
springs more or less choked by sand. These waters feed phreatic
aquifers, often salty, and join those evaporated by the halophile
plants.

It is a well-known fact that old defective wells, whose casings
are corroded, must be sealed off. The sealing operation can gen-
erally take place with more or less difficulty. But certain types of
leakage are much more difficult to avoid. This is notably true
when imprudent operations have caused the collapse of the roof
of an artesian aquifer.

The evaluation of all these water losses is quite difficult. It 1s,
however, indispensable for a correct study of underground water
resources development.

Circulation Reserves. Underground formations sometimes oc-
cur in desert regions, in great basins with huge reserves. The
flow of the springs of the Djerid (Tunisia) is so constant that
all the measurements taken for 50 years gave results whose
differences are comparable to errors of measurement. The same
statement has been made in the Mzab (Sahara) (4). Given what
we know of the irregularity of the supply of these aquifers, that
signifies that the free water table has a very considerable extent.
Its position is unfortunately not well known on account of lack

WATER RESOURCES IN ARID REGIONS 109

Figure 4. Aerial view of Chott Djerid.

of sufficient explorations and because geological outcrops are
often hidden under aeolian formations.

Because of the importance of the reserves, a long period of
time is necessary to appreciate the effects produced by new water
developments and to determine the moment past which the
safe yield of the aquifer will be exceeded.

We saw that for the rainfall and the runoff, it is better to make
numerous observations than very precise ones because of the
extreme irregularity of these phenomena. For the study of water
levels and the flow of aquifers, it is necessary, on the contrary,

110 THE FUTURE OF ARID LANDS

to increase as much as possible the precision of the measurements
because of the slowness of their variation. Also one should make
use of a precise survey system, generally absent in desert regions.

One must not lose sight of the fact that desert oases support
an extremely dense population. In the oasis of the Djerid a
population of 45,000 lives on 40 square kilometers of irrigated
land. One must therefore be very prudent in the execution of
work capable of modifying the flow of water.

When working on a virgin or little-used aquifer, it is often pos-
sible to oye spectacular improvements without preparing
very precise bases, but beyond a certain degree of use the ameliora-
tion of yield requires greater and greater efforts and increasingly
precise scientific information.

For good use of the reserves, it is naturally better to take out
the water at the lowest possible part of the formations. The
usable reserve is thus increased, the losses diminished, and at
the same time the instantaneous flow is increased.

But the lowering of the takeoff level by pumping raises difficult
social and economic problems. It can lead to the elimination of
small owners and the concentration of property. In areas of
great population density, divided into small parcels of land,
increases in agricultural production should yield priority to drink-
ing needs. These needs satisfied, landowners may lack the re-
sources to pay additional pumping expense. Hence the interest of
utilizing local sources of energy and especially wind energy.

Pant eullesky advantageous is the case where it is possible to
lower the takeoff level of the water without the expense of pump-
ing. This is the case in Chott Chergui in Algeria (1) where utiliza-
tion of the waters is projected by taking them below the level of
the Chott, situated on a plateau 1,000 meters high, to use them
in the valley of Cheliff, several hundred meters below.

From our point of view this experience is of utmost interest.
It gives the opportunity for a program of studies now in progress
which will not fail to better our knowledge on all the points
covered in the present study. ;

Another advantageous case, because it is relatively easy to
study, is that of the very important water-bearing limestone
formations in the eastern Mediterranean basin. It is easier in

WATER RESOURCES IN ARID REGIONS 111

this case than in that of sandy formations to determine the ge-
ometry of the formations. The points of emergence of the water
are better localized and observation of them permits easier study
of the water balance.

Conclusion

The arid zone is characterized as much by the variability of
its climate as by lack of water. To enable a maximum population
to live in an arid region, there must be maximum exploitation
of all the resources, and compensation for the variations in climate
which could not be compensated within that region, by means
of exchanges with vaster and vaster territories, according to the
degree of compensation necessary. Thus one must study global
variations in climate over territories of greater and greater extent.

In order to know the ensemble of the resources, there is need of
longer observations than in the humid regions. This gives special
interest to archaeological and historical research and increases the
importance of international exchanges of information.

The networks of pluviometric and hydrological observations
are generally not extensive enough, especially in mountain and
desert areas. Instruments of observation need improvement in
order to be adapted to arid zone conditions: rain gages capable
of functioning automatically and withstanding sand winds;
gages fitted for streams which are dry part of the year; devices
for measuring sediment flow at the bottom of the streams.

For underground waters, the prolonged time of the observa-
tions requires great precision in the measurement of flow and
water levels, which presupposes a good leveling survey, providing
for quick detection of often slow fluctuations.

It is very important to determine the natural ground water
discharge that can be salvaged. In this respect there is a lack of
information on the losses from underground aquifers by evapora-
tion in saline soils and by halophyte vegetation.

Hydraulic resources are characterized by the discontinuity of
their supply. We believe it necessary to study hydrological
phenomena individually: rain, runoff, erosion, recharge and dis-

charge.

2 THE FUTURE OF ARID LANDS

Current hearsay is, ‘In Tunisia, every year is exceptional.”
Seemingly it should be the same in every arid region. Particular
situations demand study as much as and even more than av-
erage ones. It is advisable to investigate individual causes and
their statistical distribution.

REFERENCES

i, Jojermietal, |lo uOlSts Etudes sur les nappes souterraines des hautes plaines
oranaises. Congres des Irrigations, Algiers.

. Bois, C. Années de Disette—Années d’ Abondance—Sécheresse en
Tunisie de 846 4 1SSz. Revue pour |’étude des calamités, Fasc. 21,
Geneva, 1944.

3. Castany, G., Degallier, R., and Doumergue, C. 1952. Les grands
problémes ad hydrogéologie en Tunisie. Congres de Géologie, Algiers.

4. Cornet, A. 1952. Essais sur |’Hydrogéologie du Grand Erg Occi-
dental et des Régions limitrophies—Les Foggaras. Extrait des
travaux del’ Institut de Recherches Sahariennes, Vol. VIII.

5. De Marchi, G. 1951. La période de sécheresses 1942-1948 en Italie.
U.G.G.1.-A.I.H.S., Congrés de Bruxelles.

6. Drouhin, G. 1952. Les problémes de Peau en Afrique du Nord Ouest,
WINESEO:!

7. Dubief, J. 1953. Essai sur ?Hydrologie Superfictelle au Sahara,
Algiers.

8. Prospects for Agricultural Development in Latin America. FAO, 1953.

g. Gaussen, H., and Vernet, A. 1948. Carte pluviométrique de la Tuniste.
Ministére de |’Agriculture et Service Météorologique, Tunis.

10. Ginestous, G. 1927. Le Chéne Zeen d’Ain Draham. Bull. Direction
Générale de 1 Agriculture, Tunis.

11. Gosselin, M. 1951. L’Inventaire des Ressources Hydrauliques de la
Tunisie. Annales des Ponts et Chaussées Nos. 5, 6.

12. Gouskov, N. 1952. La Géologie et les Problémes de l’eau en Algérie—
Données sur l’Hydrogéologie Algérienne. XIX° Congr. Géol. Intern.
2, (Algiers). 1952.

13. Langbein, W. 1949. Annual runoff in the United States.

14. Martin, . 1954. Compt. rend. acad. sci. No. 20.

15. Migahid, Am., and Abdel Rahman, Aa. 1953. Studies in the water
economy of Egyptian desert plants. Pub. Inst. Désert d Egypte 3,
No. 1 Heliopolis.

16. Pramanik, S. K., Hariharan, P. S., and Ghose, S. K. 1953. Analysis
of the climate of the Rajasthan Desert and its extension. /vfern.
Symp. Desert Research UNESCO, Jerusalem.

17. Preciozi, P. C. 1954. Evapotranspiration—Bilan Hydrologic—Zones
climatiques, Tunis.

iw)

18.
Ig.
20.

24.

WATER RESOURCES IN ARID REGIONS 113

Shalom, W. 1953. La stabilité du climat en Palestine. Jutern. Symp.
Desert Research UNESCO, Jerusalem.

Société Hydrotechnique de France (Divers auteurs). 1954. Les
Fournées de l Hydraulique d’ Alger. .

Thornthwaite, C. W. 1948. An approach toward a rational classifica-
tion of climate. Geog. Rev. 38.

Thornthwaite, C. W., and Benjamin Holzman. 1942. Measurement
of Evaporation from Land and Water Surfaces. U.S. Dept. of Agr.
Tech. Bull. No. 877.

Thuille, G. 1951. Crues et vagues phréatiques dans le Haoux de Mara-
kech. U.G.G.1.-A.1.H.S., Brussels.

Tixeront, J., and Berkaloff, E. 1954. Méthodes d’études et d’évaluation
L Erosion en Tunisie. U.G.G.1.-A.1.H.S., Rome.

U.S. Africa, Dept. of Irrigation. 1951. Report on the relations between
rainfall and runoff, Vol. WI. U.G.G.1.-A.1.H.S., Brussels.

Data and Understanding

LUNA B. LEOPOLD

United States Geological Survey, Department
of the Interior, Washington, D.C.

In the year 1534 when Cabeza de Vaca escaped from the
aborigines of southern Texas by whom he had been enslaved for
six years, he made his way on foot from the vicinity of Galveston
to the west coast of Mexico. Although his Relacién was not
printed until 1542, the verbal report of Cabeza de Vaca gave
impetus to the growing interest in exploration of New Spain.
Estevanico, the black, one of de Vaca’s companions, served as
guide to Fray Marcos de Niza on the first Spanish reconnaissance
to reach the village of Zuni in New Mexico.

The earliest Spanish exploring parties hoped to find riches, but
expected to acquire, at the least, facts. These “gentlemen of high
quality,’ as Castafieda called them, wanted to see for them-
selves whether the cities of Cibola had streets of silver. Hearsay
was not enough. Rumor was to be replaced by first-hand knowl-
edge.

Without discounting the hope for personal gain, these men
presumably were fired with some further intellectual and spiritual
motivation, among which must have been the desire for facts
about these parts where we are assembled. Inscription Rock, only
a few miles west of Albuquerque, bears illuminating tidbits of
history. Don Diego de Vargas, says the carved inscription of
1692, came here “A su costa’’—at his own expense.

Weare attempting tosurvey and correlatesome of the facts which
people have gained about the nature of semi-arid lands. We are
better off than the early Spanish explorers, for in the intervening
period data and information have been accumulated in scope and
in detail beyond the imagination of our predecessors. We have

114

DATA AND UNDERSTANDING 115

available excellent maps, knowledge of the soils and of the rocks,
both at the surface and below the ground, measurements of
precipitation, descriptions of the vegetation, data on the flow of
streams, experience in the use, if not the husbandry, of the land.

It is true that for the purposes of our complex civilization, the
need for additional data has far outstripped the programs of
fact-finding. But it appears that an indefinite expansion of the
collection of routine measurements would still leave something
lacking. I draw the distinction between measurement data and
understanding; between the collection of facts and knowledge of
processes and interrelationships. Although we have a wealth of
data, our understanding of the semi-arid environment is poor.

Understanding the physical and biologic processes operating in
an environment is important for living in and with the land. As an
example, let us look briefly at the interrelation of the water and
sediment in ephemeral streams, and the problem of valley trench-
ing or arroyo cutting.

The Problem of Arroyo Cutting

Many of the alluvial valleys of New Mexico are gutted by
trenchlike gullies. The rapid growth of arroyos in southwestern
valleys in the United States accompanied the livestock boom of
the late nineteenth century. It is clear that the pressure of
livestock on the vegetation has materially contributed to the
growth of arroyos. But the problem is more complicated. The
first American reconnaissance teams of the Army of the West
traversed New Mexico in the fall of 1846, some 20 years before
American settlers and their livestock had any appreciable effect
on local vegetation. In August of 1846, Lt. Simpson marched
from Santa Fé to the Navajo country. He crossed the Rio Puerco
near Cabezon. So deep was the arroyo at that place that he had
to cut down the 30-foot banks to get his brass cannon across.
There is enough evidence of this kind (1, 5) to indicate that in
certain valleys large gullies existed before American settlement,
even in places far removed from heavy grazing by Spanish
livestock. It must be supposed that these arroyos were the result
of natural rather than human causes.

Geologic and archaeologic studies have demonstrated several

116 THE FUTURE OF ARID LANDS

post-glacial but pre-Columbian periods of erosion followed by
aggradation. The last period of erosion prior to the present one
can be dated by pottery buried in the alluvium as occurring
approximately in the period a.p. 1200-1400 (2). The concurrent
trenching of alluvial valleys in that period appears to have oc-
curred at least as far north as Wyoming (6) and south into
Texas (3).

The problem is further complicated by difficulty in assessing
the effect of fluctuations in climatic factors on the recent episode
of erosion. In the hundred years of rainfall record in central New
Mexico, no progressive shifts of annual totals are discernible.
But there has been a progressive change in the number of rains of
various sizes. The period 1850 to 1880 was characterized by a
deficiency in small rains and a relatively great proportion of rain
events of large magnitude. This might be interpreted to mean
that coincident with the wave of settlement and accompanying
pressure of livestock in the nineteenth century, climatic factors
were particularly adverse to maintenance of physiographic equi-
librium (4):

The upshot of these considerations is that in the last century
there has been a repetition of a physiographic episode which had
occurred more than once in post-glacial time. But the recent
valley trenching was influenced to more or less extent by activities
of man and his grazing animals. The presettlement periods of
valley erosion and subsequent alluviation presumably resulted
from changes in climatic elements.

The arroyos cut during the last century have radically changed
the contribution of sediment which the alluvial valleys provide
to the master stream. This change of sediment inflow has prob-
ably contributed to the fact that the bed of the Rio Grande at
Albuquerque has gradually risen in recent decades and now
stands only slightly below the level of the flood plain on which
the center of this city stands. The gullies have dissected the
valley flats which were the best agricultural parts of the hinter-
land. The cutting of an arroyo trench lowers the local ground
water table and cienaga grasses give way to less productive
vegetation. Water formerly could be diverted from the shallow

DATA AND UNDERSTANDING WZ,

channel by a simple brush dam, or even by felling a single tree.
With the water flowing in the bottom of a deep trench, a much
more elaborate dam is necessary, even to make diversion possible.
The flood peaks increase because of loss of natural valley storage,
and for this reason also, any diversion works must be more
elaborate.

Not everyone living in an arid region can depend on major
irrigation projects. To the subsistence homesteader who depends
mostly on his own axe, plow, cow, and horse to make a living off
the land, valley trenching, as it occurred in New Mexico, was a
major calamity.

For the earth scientist the arroyo problem poses many ques-
tions, among which are these: (1) Assuming that grazing use has
contributed materially as a causal factor in arroyo cutting, can a
change in land use, specifically, a reduction in grazing pressure,
slow down arroyo growth or perhaps reverse the trend and lead
to valley aggradation? (2) How much can small structures, water
spreaders, and other minor works retard gully development?
(3) What is the future trend of physiographic development in
these alluvial valleys under present conditions of land use? On
the answers to these practical questions depend a host of decisions
which would affect the welfare of many people.

Practical measures, including gully control, watershed treat-
ment, and grazing management, have been applied locally in
various degrees over a period of two decades. Additional data
have been collected to describe the vegetation, the soils, the
streamflow, and the sediment yield. Yet it appears that the
answer to these questions is not much closer than it was in 1933.

Need for Fundamental Research

What is lacking is a satisfactory understanding of the hydro-
logic, physiographic, and biologic mechanisms on which depend
the stability or instability of the alluvial valley. In the hope of
achieving practical answers, no provision for long-term research
in fundamental mechanisms has been made. Although some
excellent research was started at Mexican Springs in 1933, lack
of continuity of funds forced a curtailment of those efforts and

118 THE FUTURE OF ARID LANDS

finally their discontinuance. Individual investigations such as
the study of Polacca Wash (8) have not been followed up.

First it seems necessary to improve our understanding of the
hydrologic relationships between intensity and amount of pre-
cipitation, infiltration, and surface runoff, for combinations of
soils and vegetation in semi-arid areas. The small experimental
watersheds maintained by the Soil Conservation Service and
Forest Service in New Mexico and Arizona are a step in this
direction, but lack of adequate funds keeps this effort pitifully
small relative to the need for such information.

A second field of needed research is in the hydraulics of flow of
sediment-laden water. Particularly deficient is our understanding
of the nature of bed and bank roughness and the manner in which
sediment in transport affects hydraulic resistance. In most
ephemeral channels bed roughness is determined primarily by
the dunes or ripples formed by moving sediment. We have few
observations and no theoretical concepts on which to build an
understanding of this phenomenon.

A third broad field is in the mechanics of gully formation,
including hydraulic forces, phenomena in the realm of soil me-
chanics, and physiographic principles.

Our own recent work has been concerned with these problems,
and at least indicates some of the possible approaches which
appear fruitful. The work began as a study of interrelations of
discharge, width, depth, velocity, slope, and sediment in natural
channels. On some of these parameters a plethora of data exists in
the records of the regular stream-gaging stations. But in the
existing network of measuring stations few measurements have
been made on water and sediment flows in ephemeral streams
draining I to 10 square miles. To obtain measurements for
analysis, during three summers of work in Wyoming and New
Mexico we chased thunderstorms, trying to reach a storm center
in time to observe arroyos in flood. When flow was found we waded
out into the arroyo and measured the depth, velocity, and width,
and sampled the sediment load. Successive sets of measurements
were made during the falling stage of the flow. Later, measure-
ments of channel slope and bed material were made.

DATA AND UNDERSTANDING 119

This investigation led to results which added something to our
knowledge of interrelations of sediment and hydraulic factors in
ephemeral channels (7). It emphasized an unexpected similarity
between perennial channels in humid areas and the ephemeral
channels of semi-arid areas. Certain differences, however, were
demonstrated, particularly in sediment load characteristics. These
differences appear to be reflected in hydraulic factors, particularly
in flow velocity.

The current problems of sediment deposition, of arroyo cutting,
land management, and water supply emphasize a present de-
ficiency in our understanding of basic physical mechanisms in
this environment. Basic data are necessary for, but do not substi-
tute for, basic research.

We have not extracted all the knowledge it is possible to gain
even from records and data already collected. Rainfall measure-
ments must be interpreted with an eye to topography, vegetation,
and land use. The arroyo problem presents so complex an interre-
lation between soils, geology, vegetation, hydraulics, and history,
that no single discipline can take precedence over others if under-
standing is to be achieved. These are only two examples of
unsolved problems.

If we are to achieve understanding and not merely content our-
selves with the collection of facts, we must bring to the task the
zeal implied by the words on Inscription Rock, “A su costa.”’

REFERENCES

1. Bryan, Kirk. 1925. Date of channel trenching (arroyo cutting) in
the arid Southwest: Science 42, 338-44.

2. Hack, J. T. 1942. The changing physical environment of the Hopi
Indians. Peabody Mus. Nat. Hist. Papers 35, 3-85.

3. Judson, S. 1953. Geology of the San Jon site, eastern New Mexico
Smithsonian Misc. Coll. 121, 1.

4. Leopold, Luna B. 1951., Rainfall frequency: an aspect of climatic
variation. Trans. Am. Geophys. Union 32, 347-57.

§. Leopold, Luna B., 1951, Vegetation of Southwestern watersheds in
the nineteenth century. Geog. Rev. 41, 295-316.

6. Leopold, Luna B., and J. P. Miller. 1954. A post glacial chronology
for some alluvial valleys in Wyoming. U. S. Geol. Survey Water

Supply Paper 1110-A.

120 THE FUTURE OF ARID LANDS

7. Leopold, Luna B., and J. P. Miller. 1956. Ephemeral streams: hy-

ioe)

draulic factors and their relation to the drainage net. U. S. Geol.
Survey Prof. Paper 252-A.

. Thornthwaite, C. W., C. F. S. Sharpe, and E. F. Dosch. 1942. Climate

and accelerated erosion in the arid and semi-arid Southwest, with
special reference to the Polacca Wash drainage basin, Arizona.

U. S. Dept. Agr. Tech. Bull. 808.

Variability and Predictability
of Water Supply

FRANK DIXEY

Colonial Geological Surveys,
London, United Kingdom

This paper is concerned essentially with the variability and
predictability of water supply in the drier parts of certain of the
British Overseas Territories and Commonwealth countries,
mainly in Africa. A great deal of experience in these matters has
been gained in these territories, and in recent years much coor-
dinated investigation and development of water supplies has
been achieved.

Common Features

The drier parts of these territories are not all strictly arid in
the sense of the definition adopted by UNESCO (17). The basis
for the division employed in this definition is the system developed
by Thornthwaite (26), which uses an index based upon the ade-
quacy of precipitation in relation to the needs of plants, so that
precipitation and temperature data and various other factors
are employed. This is described on pages 74-80.

Data for the application of Thornthwaite’s system are not
ordinarily available, although in some territories, as in Tan-
ganyika, attempts have recently been made to obtain them locally.
Consequently, the territories to be discussed may be described
in general terms as arid or semi-arid, corresponding respectively
to the desert and steppe of many authors. Culturally, the arid
areas are those in which the rainfall on a given piece of land 1s
not adequate for crop production; in the semi-arid lands, rainfall

121

122 THE FUTURE OF ARID LANDS

is sufficient for certain types of crops, and grass is an important
element of the natural vegetation unless overgrazing has replaced
it by brush. In the absence of other data these criteria are often
of considerable value. Where rainfall data are available, even if
only from scanty records, arid areas are sometimes taken to
include those with less than 350 millimeters (13.8 in.) and the
semi-arid those with less than 750 millimeters (29.5 in.). In
North Africa Tixeront and Drouhin consider as arid lands those
with an annual rainfall of less than 500 millimeters (see page
85). An area is classified as “extreme arid” if in a given locality
at least twelve consecutive months without rainfall have been
recorded, and if there is not a regular seasonal rhythm of rainfall.

Some of the territories, e.g., Northern Kenya, Aden, parts of
Bechuanaland, are known to have an average annual rainfall of
less than 12 inches, while in most of them the average annual
rainfall is less than 30 inches. Moreover, in most of these territories
the annual precipitation occurs within a period of four months or
less, so that for the remainder of the year the conditions are often
truly arid. Some account of the water supply conditions in these
territories and of the water supply investigation and development
programs carried out in them in recent years has already been
given (5, 6, 7, 8, 10). But whether arid, semi-arid, or humid, the
principles involved in the hydrogeological investigation of these
territories are the same, and the lessons gained in any one of
them relating to the location, occurrence, or movement of ground
waters are very largely applicable to the others. Moreover, the
question of drought, a relative term, affects all of them, although
in the drier areas droughts are of fairly frequent, but irregular,
occurrence.

There are certain other features which are common to all the
drier territories to be described, namely, their isolation, poor
communications, the general lack of basic hydrological data relat-
ing to precipitation, runoff, evapotranspiration and percolation,
the high cost of labor, transport, and material required to amelio-
rate the unsatisfactory water supply conditions, and the in-
ability of the sparse population to pay for such amelioration
except where there is an economic outlet for their stock or for

VARIABILITY AND PREDICTABILITY OF WATER SUPPLY 123

hides and skins. These conditions, mainly sparse population and
low economic prospects, have in the past afforded but little incen-
tive toward the acquisition of basic hydrological data, and still
do to a large extent.

Arid regions are, as a rule, better adapted for quantitative
hydrological studies than humid regions, and many of the avail-
able methods of estimating ground water supplies have been
developed in arid regions. A wider range of methods is none the
less available in developed than in underdeveloped areas. A num-
ber of methods, however, do not depend at all on development,
and much successful quantitative work has been done in areas
that were virtually undeveloped (1g). The Chott Chergui
scheme of northern Algeria provides a good example of this (12).

There is one important difference between arid and semi-arid
areas which has been emphasized by Shotton (25). In those
fringe areas—southern Palestine and parts of Jordan are good
examples, as well as the semi-arid parts of Africa—where nature
has provided at some period of the year an adequate rainfall and
yet turns the countries to arid desert at other times, there is
every incentive to search for underground water and to use it to
balance the irregularities of the rainfall. The search may even
be long and difficult and the results must always conform to the
law that more water cannot be taken from the ground than
soaks into it; but, subject to these limitations, there is a future for
parts of the semi-desert earth which most of the true desert
cannot hope to share.

Climatic Variations

The question of variations of climate in respect to arid lands
often arises, particularly as to whether a given territory is drying
up or not. There are, of course, cycles or variations of widely
different amplitude, whether of 11, 30, several hundreds, or many
thousands of years, and, over a given period of a few years, it is
usually not possible to say whether an increasing aridity is part
of a progressive change in that direction or merely the downward
curve of a lesser cycle. In the levels of certain of the great lakes of
central Africa a periodic rise and fall has been observed which

124 THE FUTURE OF ARID LANDS

shows a measure of correlation with the eleven-year sunspot
cycle (g). In this region also within the past forty years or more
the levels of certain lakes have shown a progressive fall, and ice
caps have shrunk, indicating a climatic change. Wayland believes
that the Kalahari climate is slowly reverting to a more arid
phase, but at the same time he says, “It 1s unthinkable that the
change will be an entirely one-way process; moreover, these
climatic mergings take time—geological time” (28).

In southern Africa an increasing aridity since European occupa-
tion began is generally admitted, although opinions still differ as
to the extent to which this is due to climatic change or to man’s
interference with vegetation and other natural conditions.
Bosazza et al. (2) expressed the view that the desert encroachment
of South Africa is the result of the activities of man and his com-
panion animals, although others have seen in meteorological
records and in the altered regime of rivers evidence of late climatic
change.

Gevers (13), for example, described drying rivers in the north-
eastern Transvaal. He recorded that Dr. Schumann, Chief
Government Meteorologist, in 1934 showed that there had been
a noticeable decrease in rainfall over great parts of the Union
since the eighteen nineties, and he quoted figures showing a
steadily declining rainfall from 1906-07 to 1946-47 at the Duivel-
skloof and Ravenshill stations, in the vicinity of the rivers de-
scribed. He stated that there was no question of the widespread
desiccation in the area referred to, and that there was also little
doubt that the marked decrease in rainfall since 1926 was the
predominant cause. There is equally little doubt that the effects
of the incidence of rainfall have been greatly modified by man-
made agencies, such as despoliation of the vegetal cover (forests,
bush, scrub, and grasslands) with consequent removal of long-
conditioned absorptive topsoil and increase in runoff. It is quite
clear by comparison with former conditions and ocular evidence
of the present day that all streams not rising in protected or com-
paratively well preserved catchment areas have lost their staying
and recuperative powers and that many of them have been
turned into stormwater drains. Gevers considered also that the

VARIABILITY AND PREDICTABILITY OF WATER SUPPLY 125

area was ideal for prolonged research on the lines of the Jonkers
Hoek and Cathkin Peak Forestry Research Stations. The effects
of gum and other plantations as against indigenous tree and bush
cover, and again of tree cover in general as against grasslands,
could be studied with great profit in this region.

This question of rainfall is highly pertinent, too, to the outlook
of the Sahara; some writers believe it to be decreasing, but the
evidence is not conclusive. Probably there has been no significant
change in the northern Sahara since Roman times. Apparent
jes cacti can be traced to human rather than climatic causes,
particularly overuse of the native vegetation, overdraft of ground
water, and past raiding. There is widespread belief that the
southern Sahara is advancing, but here again the evidence as to
climatic change is conflicting (18).

Tixeront (pages go-g2) ascribes the disappearance of the
ancient agricultural civilization of the arid lands of northern
Africa, comparable with that of the arid lands of the Near East
and of India, to human rather than to climatic factors. He records
that in the recent development of Tunisia full use has been made
of archaeological and historic records relating to Roman and Arab
cisterns, wells, and other works which when reconstructed and
restored appear to function much as originally, and certain
springs in this region also seem to be flowing as well as in Roman

days.
The Search for Water

On the question of the search for water in arid lands, one must
pay tribute to the extreme efficiency and thoroughness with
which the earlier inhabitants sought out and exploited temporary
surface supplies in scarcely perceptible hollows and at the foot
of rock catchments, as well as ground waters in shallow well
fields. It is amazing to see the great herds of the larger and smaller
stock that can be maintained from these sources. Even with the
aid of modern techniques it is often very difficult to find such
occurrences of water which are not already being exploited or
have been used in the past, and frequently the only advance we
can make is to prove, by drilling, deeper supplies that were

126 THE FUTURE OF ARID LANDS

beyond the reach of those early people. The modern inhabitants,
like the Masai of Tanganyika, or the Somali of northern Kenya,
merely use, often without even troubling to maintain, the wells,
hafirs, and catchments that were discovered and constructed by
their long-vanished predecessors.

In the present-day search for water it is necessary to use every
modern aid. A thorough study of the geology is required, with
special reference to the absorption, retention, and transmission
of ground water, together with an investigation of all aspects of
the hydrogeological cycle, even though studied under more humid
conditions. In addition, the application of geophysical methods is
needed to determine the presence of dykes, faults, shears, fissures,
and other structures, and the depth to solid or impervious rocks,
and of aerial photography to assist in the determination of
geological, typographical, hydrogeological and vegetational fea-
tures. A study must also be made of the vegetation itself.

Where highlands of higher rainfall rise out of the deserts, the
runoff on the windward side often increases the deeper ground
water flow even at a considerable distance from the foot of the
mountains.

Accumulated local geological experience becomes of special
importance in these circumstances and sometimes leads to an
almost intuitive appreciation of the possible presence of ground
water. Experience shows that in any program of development of
water supplies in arid regions it is most important, and far more
economical, that water should be sought in places where it is
most likely to occur rather than where it would be most con-
venient, and drilling merely on a grid or interval basis is likely
to be very wasteful of boreholes. In some territories, as in much
of eastern Africa, this is already leading to an increasing difficulty
in finding good supplies, in that the more promising sites are
being taken up to an increasing extent.

Well Yields

The question of the maintenance of yield of boreholes in arid
places is a very important one, for even when the pumping pro-
ceeds at only 50 or 60% of the tested yield, the yields sometimes

VARIABILITY AND PREDICTABILITY OF WATER SUPPLY 274

gradually diminish. For this reason some observers, such as
Wayland (28), in the Kalahari, consider that such boreholes are
dependent upon “‘fossil” water, which accumulated in a past
humid period; others, such as Bosazza (1), consider that small
reservoirs underground in the Orange Free State and the southern
Kalahari are undoubtedly recharged over a matter of years but
can be depleted in a few months with excessive pumping. Martin
has expressed the view that in southwest Africa the recharge of
aquifers on any large scale takes place only during periods of
exceptional precipitation which occur approximately every
ten years.

By means of lysimeters it has been shown that in some cases
in arid countries only 2% of precipitation has penetrated more
than 4 feet below the surface. In southern Africa it has been
recorded that borehole water supplies are less difficult in rela-
tively dry areas with little vegetation than in moister areas
with denser vegetation (22).

The United States Geological Survey has recently completed a
paper on the qualitative aspects of the relation of soil structure
to infiltration and unsaturated flow of water above the water
table, and quantitative studies of this important subject are in
hand.

Infiltration is much affected by salts, including the nitrate
and ammonium fertilizers; low permeability is common in regions
where alkaline salts are present in soil or irrigation water, and
is due to base exchange taking place during infiltration. The
acidification of waters tends to increase the rate of infiltration

(Qn).
Water Quality

The ground waters of arid lands are normally highly mineral-
ized. Frequently the clay fraction in wells and rocks seems to be
a principal factor in inducing salinity; in southern Mozambique,
for example, saline waters have been found to be almost invari-
ably associated with beds of clay and very clayey sands, whereas
fresh waters were almost invariably associated with sandstone or

grit (1). Experience with the deep alluvia of the Nyasa-Shire

128 THE FUTURE OF ARID LANDS

rift shows that in the saline areas the beds of sand and gravel
interbedded with thick clays also yield saline water (14).

In some countries, as in northern Africa, bodies of shallow
saline water are sometimes due to the evaporation of ground
waters of usable quality, and there is room for further investiga-
tion into the occurrence and use of such ground waters.

For these and other reasons related to the nature and structure
of rocks and the relative freedom of movement of waters in them,
usable and unusable waters are frequently found only short dis-
tances apart, so that an area yielding unusable supplies should
not be given up too hurriedly before the possibilities of finding
local fresh waters have been exhausted.

A recent survey carried out by Bosazza (1) in the low rainfall
and former desert area of the Sul Do Save in Mozambique shows
that over an area of about 2,500 square kilometers, out of all the
boreholes drilled to 30 to 40 meters, 24% are of low enough sa-
linity for human consumption and a further 24% good enough
for stock. Thus 48% of the boreholes are usable, a percentage
that is very high for southern Africa; and these conclusions are
drawn from work on boreholes which have been drilled at inter-
vals throughout the area without any geophysical work to assist.

Shotton (25) has recorded that ideas on the standard of water
acceptable to man for drinking have changed considerably in
recent years. It may now be taken as a fact that water with a
salinity of 3,000 parts by weight of sodium chloride per million of
water can be drunk regularly by human beings in a desert climate,
that a figure of 4,000 unaccompanied by important quantities
of other salts 1s acceptable, and that for short periods even a
figure of 5,000 is endurable. Domestic animals are often more
tolerant of dissolved constituents than man though there is no
close agreement on the worst limits of quality. Much investiga-
tion on this question has been carried out in Australia. Jack (15)
in South Australia states that horses will thrive on water with
I ounce of sodium chloride per gallon (6,260 parts per million)
and sets upper limits for living as 7,800 for horses, 9,400 for cattle,
and 15,600 for sheep—unless magnesium sulfate is present,
when the figures must be lowered. Edgeworth-David and Browne

VARIABILITY AND PREDICTABILITY OF WATER SUPPLY 129

(11) giving figures expressed as total solids have set limits even
beyond those of Jack. It is clear that both man and his stock can
drink water which is of a quality not infrequently obtainable in
deserts.

To obtain water of a quality suitable for crop irrigation is a
far more difficult matter, for there are certain salts—the alkali
carbonates and bicarbonates (black alkali)—which are only
acceptable to plants in very small concentration. A practical
limit of only 700 parts per million has been given for the total
solids permissible in irrigation water, and unless man finds such
water in quantity, he can neither grow regular crops for their
own sake nor as fodder for his horses. Shotton’s experience in
northern Egypt and Libya during World War IT convinced him
that a water table could be found almost everywhere in this
desert, but usually of such high salinity that a random well had
small chance of finding drinking water and next to no chance of
water of irrgiation quality. He considers it a fair assumption
that irrigation quality water is not to be expected in a desert
from its own local and limited rainfall unless exceptional condi-
tions exist. Of a number of such conditions two are mentioned.
The first occurs when newly percolated rain, making its way to
the water table, finds difficulty in mixing with the general body
of saline water. In the Western Desert during the war many
water points were established through this cause, with salinities
from 200 to 2,000 parts per million in a vast area where normally
the salinity stood at 5,000 or 6,000 (i1.e., unpotable) and ex-
ceptionally went up to 60,000. Characteristic of such wells were
the thin depth of good water (typically only a few feet), the very
sporadic distribution of these patches (undrinkable water could
exist only 100 yards away), the gradual tendency to become
more saline with pumping, and the small yield which rarely
exceeded a few hundred gallons an hour. Indeed, the smallness
of yield is an inevitable corollary of the fresh water—a fissure or
pore system open enough to give a large yield would not permit
the fresh water to remain unmixed with the salt in the first place.
Such wells, therefore, have no importance in irrigation prospects.

The second possibility is that of perched water, where geological

130 THE FUTURE OF ARID LANDS

structure causes the holding up of water above and quite separate
from the main table. The controlling factor is often a bed of
shale or clay occurring as a lens or fold between the aquifers.
Several examples of this type were developed in the Western
Desert during World War II.

Wartime experience in the eastern Egyptian Desert (Red Sea
Hills), where rainfall is extremely small and sporadic, showed
that by careful attention to geology, aided by geophysical meas-
urements, underground reservoirs of drinkable waters could be
found (23); of ten wells with drinkable water, five were of irriga-
tion quality, but the yields were only a few hundred gallons an
hour with a limited life, and therefore useless for irrigation
schemes.

As in a number of well-known cases in northern Africa, the
geological structure of some deserts is such that deep boreholes
in them tap artesian supplies of fresh water derived from humid
regions far beyond the desert margins.

In general, the overpumping of ground water supplies in arid
lands, where it does not lead to exhaustion, gives rise to increasing
salinity; rare exceptions to this are known, as where the abstracted
saline water is replaced by a new acquisition of infiltrated rain
water. Shaw (24) has shown that at Ma’an in Trans-Jordan, over
a period of eight years, while the effect of pumping in certain
wells was to increase salinity either at once or after a time lag, a
complete recovery could take place even after considerable periods
of heavy pumping have raised the salinity to very high figures.

Apart from the development of rare surface and shallow ground
water supplies, the indispensable tool in the investigation and
development of the water supplies of arid lands is the water-
boring machine, used with proper regard to the prevailing geologi-
cal and hydrogeological factors. In the British Overseas and
Commonwealth Territories, for example, large sums have been
spent on these operations in recent years for the benefit of the
local inhabitants and for ranching and other projects, and in
every territory active teams are now busily engaged on extended
programs of amelioration and development (5-8). As far as the
local inhabitants are concerned, the beneficial use of the new

VARIABILITY AND PREDICTABILITY OF WATER SUPPLY 13]

water supplies is intimately bound up with the question of grazing
control, without which overgrazing and its attendant evils would
occur and the final state of the territory would be worse than the
first.

Use of Surface Waters

In arid lands generally the question often arises of the possibil-
ity of making use of the storm water flows of the ‘“‘dry”’ sandy
river and stream beds which normally run to waste, or ultimately
spread out over flats from which they are evaporated with little
benefit to ground water sources below. Temporary or even per-
manent local supplies are sometimes found in them, and in some
territories, as in Tanganyika, it has been found that in the dry
season the water in sandy river beds tends to be concentrated in
lenticular, but not necessarily connected, sand reservoirs, and that
the individual sand lenses are slowly draining downstream (4).

Sub-surface dams are frequently proposed with a view to hold-
ing up the sub-surface flow in sandy river beds. They sometimes
meet with marked success, but the common obstacle to greater
development of such supplies is the difficulty of finding sections
of stream channels that are sufficiently impervious along both
floor and sides, together with the difficulty of finding low-cost
materials and methods of construction in isolated places.

There is usually great scope for the construction of dams and
tanks of various kinds to take the runoff from natural rock and
other catchments and from artificial catchments, but the high
evaporation losses, which may amount to as much as 8 feet in a
dry season, have usually acted as a deterrent to such schemes,
since they tend to make it impracticable or unduly costly to pre-
serve water until late in the dry season. The recent investigations
in Australia, Kenya, and elsewhere on the reduction of evapora-
tion losses by the use of cetyl alcohol, and related compounds,
which form a thin film on a water surface and thereby substan-
tially reduce evaporation, raise new hopes for the conservation of
water supplies in arid lands. Experiments to date have proved
highly successful, and large scale trials are now in hand. The
chemicals used are harmless to animals and plants.

132 THE FUTURE OF ARID LANDS

Small-scale out-of-door tests of cetyl alcohol by C.S.I.R.O. in
Australia have now been in progress in Victoria for about eighteen
months, and the results show an average of 50% reduction in
evaporation. A large-scale test was conducted in the summer of
1954 on a town reservoir at Woomerland, Victoria. Two acres of
water were treated, and, although the results were complicated by
seepage, it seems likely that evaporation was reduced by 30%.
Additional larger-scale tests are in progress.

Recharge of Aquifers

Finally, there is the question of the recharge of aquifers which
normally takes place under natural conditions but can also be
effected artificially. Of any given rainfall the residue, after runoff,
evaporation, and transpiration by plants have been accounted for,
percolates downward through the soil and ultimately augments
the ground water. In arid regions the proportion of rainfall that
reaches the ground water is small, say 4 or 5%, and in extreme
cases, as in parts of the Kalahari, recharge 1s considered to be nil.
In the drier parts of Tanganyika, for example, the precipitation is
considered to be accounted for as follows: runoff 4 to 6%, evapo-
transpiration minimum 72%, average 85%, maximum go%, and
percolation the remainder, say about 10% (4). But the losses by
runoff and evapotranspiration can to some extent be controlled by
manipulation of the vegetation and surface conditions with a
view to increasing the recharge of ground water. Grass sometimes
reduces runoff as compared with natural vegetation, and some-
times evapotranspiration can be reduced if the natural vegetation
is replaced by grass or cultivated crops. In Tanganyika a consid-
erable rise of shallow ground water has been observed in some
areas as a result of the clearing of the bush for cultivation. Soil
erosion not infrequently develops at a later stage, and then runoff
is increased to such an extent that very little percolation takes
place and the shallow ground water reserves are destroyed.

An important natural recharge takes place by influent streams
in arid areas, and this can be increased by the building of dams
and weirs which increase the opportunity for infiltration along
the stream channel; this is sometimes employed for improving the
supplies to wells and boreholes along stream courses.

VARIABILITY AND PREDICTABILITY OF WATER SUPPLY 13s

In arid regions the losses of rain water as a result of runoff and
evaporation in pans are enormous, and the question of the possi-
bility of conserving these supplies for dry season use by means of
storage or recharge frequently arises. If the methods of artificial
recharge now commonly applied in humid and_ semi-humid
countries could be employed in arid regions they would be of
untold benefit, but so far there has been only limited scope for
such application. In the first place, water intended for artificial
recharge is generally pretreated so that it will not clog the
pores of the recharge basins, and this would ordinarily not be
practicable in arid countries. Secondly, the form and capacity of
the recharge aquifer and its suitability for retaining and yielding
up the water can generally be determined, whereas this would also
be difficult with the limited resources and data of the arid regions.
Again, in humid countries, it is often found that under certain
conditions of natural vegetation and soil a rapid absorption of
water spread over a surface can be effected. In general, therefore,
recharge in arid areas is possible only if the water can be applied
in its untreated form, if a suitable absorption surface can be found,
and if a suitable aquifer is readily available.

So far, little if any application of these methods has been prac-
ticable in the more arid parts of Africa, but some successful at-
tempts in this direction have been made in southern California,
where the exhaustion of aquifers used for the irrigation of crops
has been prevented by the storing of flood waters of mountain
streams in the ground for later use. This was done mainly by
surface flooding or by infiltrating water from ditches or basins into
the fans of gravel, sand, and silt which extend from the mouths of
canyons to the lower cultivated areas, as has been described by
Lane (16) and by Michelson and Muckel (20).

In the flood method, water is allowed to pass over the ground
as a thin sheet controlled by ditches and embankments. Experi-
ments and field observations have shown that the highest percola-
tion rates are obtained where the natural vegetation and soil are
least disturbed.

Where ditches are used, they are flat bottomed, 3 feet to 12 feet
wide and g inches deep. They are constructed in a variety of ways
so as to disperse the water over a large area, usually with a recep-

134 THE FUTURE OF ARID LANDS

tion ditch to return the silt and surplus water to the main channel.
The flow is controlled by sluices. This method 1s easily operated
and maintained, but the percentage of actual infiltration area is
relatively small.

For recharge by basins, water is impounded in areas of some
400 feet to 100 feet by earth dams or banks some 3 feet high,
built along the contours. The basins are arranged so that the
water can overflow from one to the other, silt being deposited
mainly in the highest one. The level of the water is generally
maintained at a depth of about 6 inches. Where silting occurs, the
surface is repeatedly broken by scraping or harrowing, as on the
spreading grounds of Los Angeles, where this is done after each
10 to 14 days (3).

Where in a limited unit area, as defined by geological or geo-
graphical conditions, data are available or could be acquired for a
complete study of the hydrogeological cycle, it is sometimes pos-
sible to make full use of the total water resources of the area. In
Tunisia, for example, Tixeront (27) has studied the water supply
resources of three limestone masses on which the towns of Tunis
and Bizerta depend. The masses are individually of limited but
known extent, and they are cut off by faults and impermeable
beds. The rainfall, evaporation, and outflow of streams are all
well known, and the movements of the water table are checked by
observation wells, boreholes, and other works. The annual volume
of the additions of water to the ground water reservoirs as ob-
served and calculated accord satisfactorily. The results obtained
enabled the authorities to follow the fluctuations of the reservoirs
in the aquifers studied, to regulate their exploitation, and to
predict the importance to the township supplies of any new
additions of water.

In British African territories much interchange of information
and coordination of effort has been achieved by Inter-Territorial
Hydrological Conferences which it is proposed to hold periodi-
cally. One such conference took place at Nairobi in 1gs0 and
another in Southern Rhodesia in 1954. The papers presented at
these conferences, as well as the publications of the Water and
Meteorological Departments, clearly indicate the wide range of

VARIABILITY AND PREDICTABILITY OF WATER SUPPLY 135

hydrological research and development now in hand in these
territories.

Outline of Situation in British Overseas and Commonwealth Ter-
ritories

The points suggested for discussion comprise the following:

1. How predictable is precipitation in an arid region?

2. Are there distinct drought cycles?

3. What are the prospects for usable ground water occurrence
in arid areas?

4. What is the practicability of locating and estimating volume
and rate of natural recharge of underground water supplies?

5. Within a given watershed, to what degree can the water
sources and water yield be determined?

I have considered these points in turn in relation to investiga-
tion and development now proceeding in various British Overseas
and Commonwealth Territories. Very brief, general answers are
given in this paper, detailed answers for each area in a paper
published separately in Colonial Geology and Mineral Resources.
I wish here to express my great indebtedness to all those officers
mentioned in the text who have supplied me with the information
on which this account is based.

1. How Predictable 1s Precipitation in an Arid Region? In the
territories considered precipitation is not ordinarily predictable
owing to the absence of adequate long-term data on which to base
statistical investigations.

2. dre There Distinct Drought Cycles? In the British Overseas
Territories considered droughts are of common occurrence, but
they do not recur in regular cycles.

3. What Are the Prospects for Usable Ground Water Occurrence in
Arid Areas? In the territories considered, in the most arid areas,
such as Somaliland, northern Kenya, parts of Tanganyika, and
Bechuanaland, usable ground waters are found only with diff-
culty, and the percentage of successful boreholes is low, 50%, or
less; waters are frequently too saline for use, and in parts of Tan-
ganyika and Kenya the fluorine content is also high. The waters
occur at depths of 300 to 400 feet, or even more. In the less arid

136 THE FUTURE OF ARID LANDS

areas, such as Karamoja District of Uganda, parts of Kenya and
Tanganyika, and Northern Rhodesia and Nyasaland, the ground
waters are usually potable, and are met commonly at depths of
100 to 200 feet. The proportion of successful boreholes, yielding
135 to 1,000 gallons or more per hour, commonly ranges between
80 and go%.

4. What Is the Practicability of Locating and Estimating Volume
and Rate of Natural Recharge of Underground Water Supplies? As
regards the locating of ground waters, in all the territories con-
sidered, sites are selected on the basis of geological, topographical,
and geophysical investigations, but in the end the actual locating
or proving is effected by means of the drill. Water-boring opera-
tions accordingly take pride of place in all schemes for the inves-
tigation and development of ground waters. In the less developed
countries Government has to take the initiative in providing
drillers and drilling equipment, but in other countries private
contractors are also available. Governments carry out water
boring for the indigenous peoples as part of programs of develop-
ment, but they drill also for European settlers on payment.

As an example of Government drilling for the benefit of settlers
may be quoted the following arrangements current in the Union
of South Africa, where, however, the operations are more highly
subsidized than in most other territories.

Boring for water, on application by individual farmers, is car-
ried on by the Irrigation Department throughout the Union,
while Government subsidies are also payable on boreholes drilled
by private contractors, where the water supplies are required for
domestic and stock-drinking purposes. These measures have un-
doubtedly contributed greatly to the utilization of semi-arid
stockfarming areas and have also been a potent factor in the fight
against soil erosion, by reducing the distance stock have to travel
to water. There are at present 217 Government drills in operation
and the work is heavily subsidized.

In 1946, in order to assist stock farmers in certain parts of the
country where arid conditions prevailed, it was decided to pro-
claim certain areas as drilling zones, and the gradual extension of
the areas so proclaimed has resulted in the Union being divided

VARIABILITY AND PREDICTABILITY OF WATER SUPPLY 137

into two areas for water-boring purposes. The difference in the
two areas is concerned chiefly with the charges levied for holes
drilled by government machines, drilling in proclaimed areas
being on a ‘“‘no water—no payment” basis, and when water is
found the rate payable is based on a sliding scale calculated on
the quantity of water obtained and the footage drilled. In non-
proclaimed areas drilling is charged for at a fixed daily rate, but
rebates are allowed which in practice make the actual charges
very reasonable.

As regards estimates of volume and rate of natural recharge of
ground waters, none of the territories considered is sufficiently
well developed to have acquired the necessary detailed geological
and hydrological data to enable general estimates of this nature to
be made; locally, however, the requisite data have been built up
for particular limited projects, and on the basis of these data
estimates of volume and rate of recharge have been prepared and
put to practical use in the development of various schemes for
township and other water supplies.

§. Within a Given Watershed, to What Extent Can the Water
Sources and Water Yield Be Determined? In the better developed
territories the necessary data have been acquired regarding water
sources and water yield for particular projects, such as town water
supplies and hydroelectric and irrigation works. In the less well-
developed territories the information of this nature available is
almost nil, and 1s confined to minor local projects and scanty
short-term records. In general, in relation to the vast areas and
potential involved, the amount of hydrological data at present
available is small indeed. Within the last few years the British
east and central African territories have greatly augmented their
hydrological staffs and have provided for the collection of standard
data on a scale never before attempted. This essential step has
been taken because of the rapid development of these territories
and of the urgent realization of the need for such data for many
purposes. The necessity to provide water supplies for the larger
townships, and to prepare the great hydroelectric projects asso-
ciated with Victoria, Albert, and Nyasa lakes, and the Zambezi
and Kafue rivers has proved an immense incentive to the develop-

138 THE FUTURE OF ARID LANDS

ment of plans for large-scale hydrological investigation and for
the acquisition of the requisite basic data.

REFERENCES

1. Bosazza, V. L. 1954. Problems of water supply in the arid areas.
Geogr. Ff. 120, 119-22.

2. Bosazza, V. L., R. J. Adie, and Brenner, S. 1946. Man and the great
Kalahari Desert. 7. Natal Univ. Sct. Soc., 5.

3. Buchan, S. 1954. Artificial replenishment of aquifers. 7. Inst. Water
Engrs. 9, 1-29.

4. Coster, F. M. 1954. Ground waters and development in Tangan-

yika. Inter-Terr. hydrol. Conf., Salisbury, S. Rhodesia.

5. Dixey, F. 1946. Rural water supplies in Africa. S. African

Geogr. Ff. 28.
6. Dixey, F. 1950. 4 Practical Handbook of Water Supply. London.
7. Dixey, F. 1951. Water deficiency areas and arid zones in the
British colonies. U.G.GJI., Assoc. Internat. Hydrol. Sct., Brussels
4, 21-26.
a Dixeye 7 1951. Subterranean water supply investigation in the
British colonies. U.G.GI., Assoc. Internat. Hydrol. Sci., Brussels 4,
26-45.
9. Bree F. 1952. Variations in lake level and sunspots. Co/on.
Geol. Min. Resources 3, 213-18.

10. Dixey, F. 1953-4. Some recent studies in ground water problems.
Colon. Geol. Min. Resources 4, 121-48.

11. Edgeworth, D. W. T., and W. R. Browne. 1950. The Geology of the
Commonwealth of Australia, Vol. 2, p. 514.

12. Gautier, M. 1952. La resource aquifére du basin du Chott Chergut
(Oranie). C.R.7g9° Congr. intern. géol., Alger, Sec. 8, L’hydrogéologie
des régions arides et sub-arides, Fasc. 8, pp. 233-42.

13. Gevers, T. V. 1948. Drying rivers in the N. E. Transvaal. S. African
Geogr. F. 30, 17-44.

14. Holt, D. N. 1955. The location of ground water supplies in Nyasa-
land. Colon. Geol. Min. Resources 5.

15. Jack, R. L. 1914. Bull. No. 3, geol. Surv. S. Australia.

16. Lane, D. A. 1936. Artificial storing of ground water by spreading.
F. Am. Water Works Assoc. 28, 1240.

17. Meigs, P. 1951. World distribution of arid and semi-arid homocli-

mates. Arid Zone Programme (UNESCO).

18. Meigs, P. 1954. The outlook for arid north Africa: The Sahara.

Focus, Arid North Africa, No. 5. Am. Geogr. Soc.

(oe)

VARIABILITY AND PREDICTABILITY OF WATER SUPPLY 139

19. Meinzer, O. E. 1932. Outline of methods for estimating ground-
water supplies. U. S. Geol. Survey, Water Suppl. Paper 638.

20. Michelson, A. T., and D. C. Muckel. 1937. Spreading water for
storage underground. U.S. Dept. Agr., Tech. Bull. 578, 1.

21. Musgrave, G. W. 1942. Some aspects of infiltration in relation to
run-off. Penn State Coll. School Eng. Tech. Bull. July 27.

22. Nel, L. T. 1948. Ground-water investigations in the Union of South
Africa. U.G.G.I., Assoc. Intern. Hydrol. Sci., Oslo. 3, Comm. Eaux
souterr., p. 266.

23. Paver, A. L. 1946. Water and Water Eng. 49, 653-62.

24. Shaw, S. H. 1947. A Trans-Jordan desert water supply. Water ana
Water Eng.

25. Shotton, F. W. 1954. The availability of underground water in
hot deserts. Bzology of Deserts. Institute of Biology, London. See
also 1946. Water and Water Eng. 429, 218-26.

26. Thornthwaite, C. W. 1948. An approach towards a rational classifi-
cation of climate. Geogr. Rev., 38, 55-94.

27. Tixeront, J., and E. Berkaloff. 1948. Note sur la percolation pro-
fonde des précipitations d’eau dans les massifs calcaires en Tunisie.
U.G.G.I., Assoc. Intern. Hydrol. Sct., Oslo. 3, Comm. Eaux souterr.

28. Wayland, E. J. 1954. Outline of prehistory and stone age climatology
in the Bechuanaland Protectorate. 4cad. roy. Belg. Colon., 25, No. 4.

General References

Cooper, W. G. G. 1950. Electrical side in water finding. Bull. 7, geol.
Surv. Nyasald. (Zomba.)

Davies, D. A. 1954. Artificial inducement of precipitation in the tropics.
Tech. Mem. No. 6, E. A. met. Dept. See also Nature 1951. 167, 614:
1952. 169, 1001: 1954. 174, 2°06.

Dixey, F. 1943. Hydrographic surveys of northern Kenya. (Nairobi.)

Frommurze, H. F. 1952. Hydrological research in arid and semi-arid
areas in the Union of South Africa and Angola. Rev. Res. Arid Zone
Hydrol., Paris (UNESCO).

Frommurze, H. F., and L. T. Nel. 1949. The occurrence, location and
exploitation of underground water in South Africa. African Reg.
Sci. Conf., Fohannesburg I, 56-59.

Glover, Robinson, and Henderson. 1954. Provisional maps of the re-
liability of annual rainfall in East Africa. 7. Roy. met. Soc. 80, No. 346,
602-9.

Hunt, I A. 1951. General Survey of Somaliland Protectorate, 1944-50
(Hargeisa).

140 THE FUTURE OF ARID LANDS

MacFadyen, W. A. 1950. Water Supply and Geology of parts of British
Somaliland (Hargeisa).

Mitchell, T. 1954. Water conservation in southern Tanganyika. Corona
6, Nov.

Sansom, H. W. 1955. Prediction of the seasonal rains of Kenya by means
of correlation. Weather 10, No. 3.

Wicht, C. L. 1955. Hydrological research in southern Africa. S. African
reg. Comm. Conserv. Util. Soil (Pretoria).

Zans, V. A. 1951. Economic geology and mineral resources of Jamaica.
Geol. Surv. Famaica Bull. 1, 38-42.

Fluctuations and Variability
in Mexican Rainfall

C. C. WALLEN

Swedish Meteorological and Hydrological Insti-
tute, Stockholm, Sweden.

The need for climatological research in arid and semi-arid
regions to the benefit of agriculture has been stressed very clearly
by Thornthwaite (pp. 67-84), especially as regards the study
of the water balance. There are other meteorological branches,
however, the study of which seem to be of similar importance to
the arid regions as practically all problems of such areas emanate
from atmospheric conditions.

In Table 1 are summarized the most essential meteorological
research problems to be studied in the arid zones. These might
be separated into two principal groups: general meteorological
and synoptic climatological studies related to the circulation
conditions on the one side, and studies of the water balance on
the other.

It is obvious that in applying to agriculture in arid and semi-
arid regions both direct precipitation and ground water it is
necessary to know in detail, ‘‘Why it rains, when it rains, where
it rains,” which therefore always must involve the basic problem
of all climatic research in such areas. In investigating the direct
water supply for vegetation and agriculture as well as in trying
to discover ground water resources, it is necessary to have a basic
knowledge of the precipitation mechanisms and _ circulation
conditions that give or prevent rainfall and also to know the
actual amount of precipitation, its frequency of occurrence, its
variability, and long-term fluctuations. It is therefore as im-
portant to study, for example, the classification of circulation

141

142 THE FUTURE OF ARID LANDS

I General and synoptic
meteorology, dynamic
climatology

Basic investigations

General circulation con-
ditions

Particular circulation
problems as ‘‘easterly
waves”

Hurricanes

Influence of westerlies, as
for instance how ‘‘jet
stream”’ position effects
winter rain

Dynamics of cumulus
clouds

Radiation from sun and
sky

Applied

Research on use of wind
energy

Research on use of solar
energy

Soil conservation studies

Long-range forecasting of
precipitation

TABLE 1

Climatic Research in Arid Lands

II Water balance

Macroclimatic investiga-
tions
Basic
Humidity conditions of
air masses involved
Causes for humidity
distribution
Cloud physics
Distribution of precipi-
tation; relation of

distribution to circula-

tion

Variability of precipita-
tion

Chemical content of
precipitation

Heaviness of precipita-
tion

Max. temp. in relation
to sunshine; in cases
frost studies

Applied

Artificial rainfall

Irrigation

Use of chemical content
in precipitation

Ground water and pre-
cipitation

Soil conservation

Microclimatic investiga-

tions

Basic

Exchange of heat, hu-
midity, and momen-
tum at the ground

Calculation of evapora-
tion from heat balance

Measurements of evapo-
transpiration

Water supply in the soil

Soil and ground tempera-
ture

Dew studies

Microclimate in forests

Applied

Prevention of evapor-
ation

Use of dew for different
plants

Influence of deforestation

Reforestation

Soil conservation

Plant ecology

Seeding days and harvest
days

Phenology

types, hurricanes, perturbations in the airflow, convective in-
stability in cumulus clouds, winter rainfall related to the “‘jet-
stream” position and the statistics of fluctuations of rainfall
created by these mechanisms as to find out in which strata of
rocks the ground water exists or how to irrigate dry lands by
surface water coming from surrounding areas of higher precipita-

tion or from rivers coming from such regions.

The second type of investigation cannot be successfully carried
out without the first one. Furthermore, careful investigations of

MEXICAN RAINFALL 143

rainfall and of which circulation conditions create convection and
heavy showers are extraordinarily important in relation to soil
erosion, which is a serious danger in these regions. It should
finally be kept in mind that inducing rainfall by artificial means
never could be successful unless one has a thorough knowledge
of the circulation mechanism that creates the necessary cloud
types in the areas concerned.

An ultimate goal of the statistical and synoptic climatological
studies mentioned in the first group in Table 1 should also be to
give methods for long-range forecasting (monthly, seasonal, or
longer) of rainfall. This is of particular importance to agriculture
and water supply in arid and semi-arid lands.

The suggested studies of the water balance (group //) in itself
may be separated into a macroclimatological and a microclimato-
logical type. The former type often serves as a bridge between the
investigations in general meteorology and synoptic climatology
of the first-mentioned group (I) and the detailed studies included
in the microclimatic type. These investigations, in other words,
should give more detailed information of rainfall mechanisms and
cloud physics as well as of temperature conditions than might be
obtained by the general investigations of the first group.

The aim of the microclimatic studies of heat and water balance
discussed by Thornthwaite is to understand how economically to
apply irrigation for the reduction of evaporation, soil conservation,
reforestation, etc., with due regard for results given by the
general investigations. It is not advisable to take up such micro-
climatic studies of the water balance without having laid a solid
foundation by investigating the first-mentioned general problems.

As an example of a basic statistical and synoptic-climatic study
of a partly semi-arid or arid country I shall discuss briefly a
study of fluctuations and variability of Mexican rainfall which
was performed during my work as UNESCO advisor in economic
climatology to the Mexican Government in 1954 (6).

Precipitation in Mexico

Detailed studies of more delicate characteristics of precipitation
in Mexico are still lacking, mainly because the scattered network

144 THE FUTURE OF ARID LANDS

of stations makes detailed studies difficult to conduct. Neverthe-
less it was possible to make some studies of fluctuations and
variability of rainfall, these characteristics being of special interest
to agriculture.

From the scattered network of stations, with often not too
accurate measurements, it was possible to select 52 stations
having in most cases reliable records for more than 20 years.
Tacubaya, D.F., with records since 1878, has also been used.

It was generally assumed that only small errors can be involved
in records from a certain area which all show the same general
interannual fluctuations. When the records showed doubtful
deviations from the general trends of fluctuation, the stations
were not accepted.

The northern and northwestern parts of Mexico show semi-arid
or desert conditions with an annual rainfall ranging from 200
millimeters in Sonora to about 500 in Durango and Coahuila
(Figure 1). On the other hand, rainfall reaches 3,000-4,000 milli-
meters in the southeastern parts of the country. Except for these
extremes the rainfall map shows a large triangular area of dryness
extending through the country southward and having its base at
the U.S. border.

Mexico is a country of seasonal rainfall with summer and
autumn the rainy seasons in all parts of the country except for a
small area in the northwest where the Mediterranean rainfall
regime of winter rain prevails. The May—October season gives
generally more than 80% of the annual precipitation. This figure
is somewhat smaller along the Gulf Coast and higher in the
southwest (4).

The following circulation factors regulate the Mexican rainfall
as far as we know them:

1. Seasonal fluctuations in the position of the intertropical
convergence zone.

2. Location, extension, and intensity of the subtropical high-
pressure cells, directing the influence of the trades.

3. Perturbations in the summer easterlies (“easterly waves’’).

4. Hurricanes, generally created in connection with the
“easterly waves.”

145

MEXICAN RAINFALL

‘(€$61-6161) uonviidioaid jenuur uvosyyl

“T ons]

Ove

146 THE FUTURE OF ARID LANDS

Precipitation in mm.

fo}

(eo)
1

500
1870-79 1880-89 1890-99 1900-09 1910-19 1920-29 1930-39 1940-49
YO @ &

Figure 2. Ten-year overlapping means of annual precipitation in

Tacubaya, D.F. (1878-1953).

5. Middle latitude westerly troughs passing over the northern
part of the country in winter.

In winter when the intertropical convergence zone 1s far to the
south, the easterlies influence only the southernmost part of the
land area while the northern parts are under the influence of the
middle latitude westerlies, which give very little precipitation
except for the Gulf Coast area in connection with the so-called
nortes. In summer the whole country, except for the northwestern
very dry regions, is under the influence of the tropical easterlies,
and rainfall is frequent owing to convectional instability and even
more to perturbations in the easterlies and to hurricanes.

Long-Term Fluctuations in Mexican Rainfall

Smoothing of the fluctuations in annual and monthly precipita-
tion was made by calculating 10-year overlapping means. Figure
2 shows the Io-year overlapping means curve for annual rainfall
in Tacubaya, D.F. The curve indicates that there was a decrease
of rainfall in central Mexico from the period of 1878-87 to the
period of 1892—1go1 and then an increase up to around 1935.
Since then there has been a decrease again up to recent years.
The fluctuations are rather great and the maximum amount of

147

MEXICAN RAINFALL

0} WNUTIXeU WO1J UOTBIBA JUID Jag ‘uoneqidisasd jenuur ul suoinjenjony poriad-suo 7

“UUNUTTUT UT

1g amS1yp

148 THE FUTURE OF ARID LANDS

rainfall, 958 millimeters, was reached in 1925, the minimum value,
440 millimeters occurred in 1894.

Studies of the fluctuations of monthly precipitations have
revealed that the increase from 1900 until around 1920 was
caused by an increase of both summer and winter precipitation,
counterbalanced by a decrease in fall precipitation. The rapid
increase from 1920 until 1935 was caused by an increase of
precipitation in both summer and fall. The decrease thereafter
has been accounted for by a decrease in rainfall of both summer
and fall.

It is likely that the considerable increase of summer rainfall
trom the turn of the century until the middle of the 30’s must
have been associated with a strengthening of the summer easterlies
over Mexico. This in turn was caused by a general shifting north-
ward of the subtropical high-pressure cells and particularly by a
transition to the west of the summer cell in the Caribbean Sea.
Such a northward shift goes very well with the well-known
warming up of the winters of northern latitudes during the same
period.

The decrease of annual precipitation since the middle of the
30’s suggests that a gradual regression of the subtropical high-
pressure cells has taken place. This has also been suggested by
evidences in northern latitudes of a cooling off in recent years
and particularly since the end of the 30’s (5).

Figure 3 shows the fluctuations of annual rainfall since 1920 in
different areas of Mexico. There is an overall tendency in all
regions, except for the southernmost one, to show first an upward
trend and later a decrease of rainfall. The turning points of the
curves from an upward to a downward trend occur at different
times so that they generally appear earlier in the south than in
the north. This is even more obvious on the western side than on
the eastern side. This gradual displacement of the maximum, as
we proceed northward, fits in very well with the idea of a gradual
shifting northward of the subtropical high-pressure cells and a
gradual moving back.

Variability of Rainfall in Mexico

In our study of the variability of rainfall in Mexico we have
used several measures. Of most interest were the studies of the

149

MEXICAN RAINFALL

‘(€$61-6161) uonezidisaid yenuue jo AVIGViiwa enuuvsaqul savy “b ain3yy

jo!sysip uodoojodog

150 THE FUTURE OF ARID LANDS

relative interannual variability (la/’)* and the anomalies of the
relative variability (RY){ in comparison with the Conrad’s
standard curve showing the interrelation between amount of
precipitation and variability for the world as a whole (1). Such
anomalies are much better to study than the (R/) in itself as
they are no longer influenced by the actual amount of rainfall as
is the value of (RV). The relative interannual variability is
shown in Figure 4 and the anomalies in Figure 5.

The highest variability is found in the areas of low precipitation
in the northern and northwestern parts of Mexico with a maxi-
mum of around 50% in the extreme dry areas. A secondary maxi-
mum of around 40% is found in the northeast, where rainfall is
greater but the water supply conditions are difficult for agri-
culture. Third maximas of some 30% are found in the Papaloapan
district of comparatively high precipitation. The lowest values
are found on the central plateau and along Sierra Madre Occi-
dental, reaching 15-20%.

Looking at the anomalies of Figure 5 we note that the area of
Baja California shows high positive anomalies, indicating that
the variability is even larger than normal. The same is true in
northeastern Mexico, southern Sinaloa, Nayarit, and parts of
the Gulf Coast area. On the other hand, it is interesting to note
that the inner part of the dry state of Sonora has variability
values much /ower than the normal conditions would give.

It is to be expected that current investigations of the relation-
ship between the general circulation and precipitation in Mexico
will explain the reasons for these extreme areas of anomalies, but
some ideas may be expressed now.

Thus, it seems likely that the reason for the abnormal variability
in Baja California, Sinaloa, Tamaulipas, and the Gulf of Tehuan-
tepec are the hurricanes which frequently hit these coasts in fall.
Studies of the variability of October precipitation suggest this

TaV
x (Ia) ef (1 — pr») at (pe — ps) tee. tb (Pri — Pn) \/ Pro; (laV) rei = a . )

p = mean annual precipitation; 7 = number of years on record.

T{[RY] = 100° where ¢ = Bay where d; = pi — ?p.
n

—

MEXICAN RAINFALL 51

ZZ +00
a)
)

ZL
TE,

eT
Be

Uf.
Bs)
Qe
SS
yy
By
1.0
ZZ

ity of annual precipitation anomalies with regard to world normal

SSS
SS
xR SMA
EXER
\ 3 <SS oe \
CS Qh
0 SN SQ

LES \
La >
Ss \ S WS \

Isanomals of relative variabil
distribution of relative variability of precipitation, according to Conrad.

Figure 5.

2 THE FUTURE OF ARID LANDS

Individual
annual rainfall

1000

500

Mean
annual rainfall

500 1000

Figure 6. Probabilities of having individual rainfall amounts (ordi-
nates) in the central plateau of Mexico as a function of mean annual
rainfall (abscissas).

idea. In the areas of rather large precipitation, as in the Papa-
loapan district, local topographical conditions must be decisive
in explaining the variability, especially in connection with
fluctuations in the general direction of the rain-giving easterlies.

Frequency Distribution of Rainfall Amounts

In studying the frequency distribution of annual rainfall a
special risk diagram suggested by Landsberg (3) was applied.

153

MEXICAN RAINFALL

‘(oF 61) OOTX9]A] UI SO}ERS JO VOIR OF UOlv los Ul BOIB pes¥Ay[No jo UOIJNQIAISTC VL, dINS

S2-02 \G

Y Sp QA
YY SI-01 WA
OI-s 4

(Le

154 THE FUTURE OF ARID LANDS

Figure 6 shows the probability of having individual annual rainfall
amounts (ordinates) in the central parts of Mexico as a function
of mean annual rainfall (abscissas). To show how to use this sort
of diagram, which seems particularly useful in arid regions of
high variability, we assume, for instance, that a particular agri-
cultural plant requires on the latitudes in question at least 750
millimeters of annual rainfall in 80% of the years in order to be
profitable on a long-term basis. The diagram shows, from finding
the intersection between the 750-millimeter value on the ordinate
and the 20% probability line, that only areas having an average
annual precipitation of more than 880 millimeters meet the
requirements. It could be mentioned as a comparison that a
similar diagram for the Papaloapan district shows that only areas
having more than 1,040 millimeters annually meet the same
requirements.

Diagrams of this kind should be prepared for more limited
districts in Mexico as soon as enough data are available.

Agriculture and Variability of Rainfall

It is evident from the maps of rainfall variability that the areas
in Mexico with high variability coincide with those where it is
known that agriculture is difficult. One of the best known examples
is the northeastern area where, even if precipitation sometimes
is sufficient, it is so unreliable that agriculture is very hazardous.

It is also clear that the central plateau area which long has
been known as the best for agriculture in Mexico is the region of
lowest variability in all the country. The relation between agri-
culture and variability is shown from a comparison between the
maps in Figures 1, 4, and 7. The last one shows the percentage
relation of cultivated land to the area of states in Mexico in 1940.
Little relation is found between the amount of rainfall in itself
and the distribution of cultivated land, but a rather good inter-
relation exists between cultivation and variability. A good relation
could also be found between the distribution of cultivated land
and climatic provinces of Mexico based upon the classification of
climate according to Thornthwaite, in which evapotranspiration
is the essential factor. It seems, however, that in applying climatic

MEXICAN RAINFALL 15S

indices to show the usefulness of land areas for agricultural
purposes the variability of precipitation should be introduced
besides the evapotranspiration factor.

Work to Follow

In the recently established Institute of Applied Science in the
University of Mexico work is now going on to investigate the
precipitation mechanisms that create the above-mentioned con-
ditions of fluctuation and variability. A classification of weather
types has been made for surface and upper air circulation over
Mexico and frequencies of the different types in various parts of
the year have been calculated. The next step will be to study the
correlation between these circulation types and precipitation in
various parts of the country. It is our hope that these investiga-
tions will become a solid basis for continued work on methods
for long-range forecasting of precipitation as well as for extensive
studies of the possibilities for inducing artificial rainfall and of the
water balance of the country.

REFERENCES

1. Conrap, V. anp K. Potiak. 1952. Methods in Climatology. Cam-

bridge, Mass.

2. ConTRERAS Arras, A. 1942. Mapa de las provincias climatolégicas

de la Republica Mexicana. Sria. de Agricultura y Fomento.

3. LanpsBerG, H. 1951. Statistical investigations into the climatology
of rainfall on Oahu (T.H.). Hawaiian Rainfall Contributions. Met.
Monographs. I, No. 3.

. Pace, J. 1929. Climate of México. Monthly Weather Rev. Suppl. 33.

. WaLLEN, C. C., anp H. W.-Son AHLMANN. 1954. Some recent
studies in Sweden on the present climatic fluctuations. 4rch. Met.
Geoph. Bioklim. Defant Festschrift.

6. WaLtEN, C. C. 1955. Some characteristics of precipitation in Mexico.

Geografiska Annaler 31, 1-2.

nm &

Beneficial Use of Water

in Arid Lands

JOHN H. DORROH, JR.

Soil Conservation Service,
United States Department of Agriculture
Albuquerque, New Mexico

During the past several years the Soil Conservation Service has
studied the characteristic precipitation and water yields of Ari-
zona, Colorado, New Mexico, and Utah in an attempt to deter-
mine: (1) what the average annual precipitation is; (2) what part
of this original supply is available for irrigation, domestic, and
other uses; (3) what watershed or climatic factors are predom-
inant in the production of water yields; and (4) what may be done
to reduce losses of water to non-beneficial use.

During the course of the study, isohyetal and water yield maps
of the four states were developed on a scale of 1: 500,000 to permit
reasonable delineations of usual amounts of precipitation and
subsequent water yields throughout the four-state area. Weighted
amounts of precipitation and the apparent total yield of water to
downstream users were then determined. The annual precipitation
varied from less than 4 to about 60 inches, and water yields from
less than one-tenth inch to more than 30 inches. The figures for
water yield do not take into account flow into or out of the four-
state area; they are merely map measurements. However, they are
considered representative of existing conditions.

Although the four states were covered in their entirety, the
following discussion will be confined to those portions with an
annual precipitation of 18 inches or less. These, we believe, en-
compass the arid zones.

156

BENEFICIAL USE OF WATER IN ARID LANDS lpe7,

Our studies developed the following facts with regard to the
four questions we asked:

What Is the Average Annual Precipitation?

The 18-inch-maximum zones comprise some 335,600 square
miles or 80% of the total surface area of the four states. The
average annual precipitation over this zone is about 13 inches.
This represents some 225,800,000 acre-feet of water and is the
total water supply or precipitation occurring over the arid zone of
these states.

What Part of Original Supply Is Available for Irrigation, Domes-
tic, and Other Uses?

Map measurements of the various water yield zones indicate
that some 5,200,000 acre-feet of water subsequently becomes
available for other than natural watershed uses. Conversely, about
220,600,000 acre-feet, or 98%, of the water that falls in the form
of precipitation is either consumed where it falls or is lost in
transit to points of downstream use. In checking other informa-
tion and data concerning water yields of the various states, it
appears that the yields shown on the maps are, if anything,
generous. Therefore, the estimates of water available from arid
zones for downstream use are probably high.

What Watershed or Climatic Factors Are Predominant in Pro-
duction of Water Yields2

During the course of our studies, notes were made of the physi-
ographic, vegetational, and other watershed characteristics that
might influence the relation between precipitation, the runoff in
surface streams, and ground water accretions. Extensive tables
were prepared showing the soils, vegetation, topography, geologic
conditions, and numerous other factors for each watershed that
had been gaged. An analysis of these data was then made to
attempt to delineate the factors or combinations of factors that
profoundly influence water supplies. In certain instances the
presence or absence of deep soils with high water holding capaci-
ties was found to exert a strong influence on water yields. Other
watershed characteristics, such as topography (which may or may

158 THE FUTURE OF ARID LANDS

not encourage natural water spreading) and vegetation (which
can modify infiltration rates), were also found to be influential.
However, no single watershed characteristic that consistently
correlated with high or low water yields could be found. This may
be due in part to the fact that each watershed had a different com-
bination of soils, vegetation, physiography, and other factors, any
of which might play a greater or lesser part in one watershed than
in another.

Soon after the study was started, however, it became apparent
that precipitation, in combination with evapotranspiration poten-
tials, consistently correlated with watershed yields throughout a
wide range of watershed conditions. True, it was found that, all
other things being equal, such watershed conditions as have been
mentioned heretofore could materially influence the yield of water;
nevertheless, the apparent influence of the precipitation-evapo-
transpiration factor far overshadowed watershed influences. For
example, it was found that with low precipitation and high
evapotranspiration perhaps 1% of the annual precipitation might
be expected to appear as water yield. On the other hand, and con-
sidering all zones covered by the study, with high precipitation and
low evapotranspiration, perhaps 50% of the precipitation might
appear as water yield. It became obvious, therefore, that these
two factors were predominant in the yield of water to the extent
that watershed characteristics, as such, ‘‘fell out of the picture.”

As mentioned by Thornthwaite (1) there has been published a
generalized map of the evapotranspiration potentials of the four
states considered in this study. Although the map is not of such a
scale as to permit a high degree of accuracy, it was possible to
develop approximate evapotranspiration potentials from the map
for comparative purposes.

Considering only the arid zone, the weighted evapotranspira-
tion potential was found to be about 29 inches on the average,
annually. This can be compared with the weighted precipitation
of 13 inches. To phrase this another way: on the average, there
exists a potential for evaporation or transpiration of some 220%
of the average annual precipitation. Although this comparison
disregards seasonal fluctuations in evapotranspiration or precipi-

BENEFICIAL USE OF WATER IN ARID LANDS 159

tation during those times of year when precipitation may be high
and evapotranspiration low, or vice versa, the situation in the
arid zone would undergo no major change were this factor con-
sidered. Recognizing this normal preponderance of evapotrans-
piration potentials over precipitation, it is self-evident that any
precipitated water that does not leave an area through surface
runoff or through infiltration to ground water will ultimately be
consumed.

This fact brings to light a popular misconception concerning the
importance of the relative evapotranspiration potentials of various
types of vegetation in the yield of water from arid zones. It has
been commonly believed that if one plant uses more water than
another then there will be less water available for downstream
use. It is not usually recognized, however, that whenever evapo-
transpiration potentials exceed the supply, or precipitation, it makes
no difference what type of plant occupies an area, the water will be
removed. Such being the case, the replacement of non-beneficial
plants with forage producing plants may be accomplished with
little or no loss of water.

Another result of the normally high evapotranspiration poten-
tials in arid zones is the tremendous amount of water evaporated
or transpired before reaching a point of downstream use. In the
San Simon Valley of Arizona, for example, only some 20% of the
indicated water yield from small experimental watersheds ever
reaches the mouth of the valley; the remainder, for the most part,
is consumed by evaporation or transpiration.

What Might Be Done to Reduce Losses of Water to Non-benefi-
cial Use?

At this point it might be well to review several facts brought
out in the previous discussion. (1) There are some 220,600,000
acre-feet of water, or about 98% of the total precipitation, being
consumed naturally on our arid watersheds before reaching a
point of downstream use. (2) In the arid portions of our four-state
area, evapotranspiration potentials are usually sufficient to con-
sume any water not flowing out of a watershed. This is without
regard to the particular type or amount of vegetation that may be
present on the area. (3) Tremendous water losses occur through

160 THE FUTURE OF ARID LANDS

evaporation or transpiration between the points of origin and
downstream use.

Disregarding the relatively small additional amount of water
that we might, with diligent effort, cause to flow out of our
natural watersheds, we still have millions of acre-feet remaining
there. It may be well to ask: What forage is this water now pro-
ducing, compared with its potential productivity? Also, what
needed commodities are produced by the water now being lost in
transit between the point of origin and the point of downstream
use?

If we accept the premise that evaporation and transpiration
potentials are such that normally any precipitation not leaving a
watershed will ultimately be consumed by whatever type of
vegetation that may be present or by evaporation alone, then is it
unreasonable to assume that we could, through replacement of
undesirable plants with desirable plants, develop a forage produc-
tion that we do not have today?

If we are correct also in assuming that large quantities of water
are consumed by non-beneficial vegetation in stream channels, is
there not reason to believe that we can eftect large savings of
water by the removal of such vegetation?

In summary, is there not every reason to believe that the part
of the 220,600,000 acre-feet of water which is not now reaching
points of downstream use and which 1s not now being put to bene-
ficial use on natural watersheds, represents the real potential for
increased production of needed commodities in our particular
arid area?

If the answer to these questions 1s yes, it appears that we should
direct our attention, action, and our research toward ways and
means of changing the use of water on our natural watersheds
from non-beneficial to beneficial use.

Although this discussion has been directed toward a particular
segment of a particular country, reasoning applied here probably
is applicable to other arid areas.

REFERENCE

1. Thornthwaite, C. W., with H. G. Wilm and others. 1945. Report of
the Committee on Evaporation and Transpiration, 1943-44. dm.
Geophys. Union Trans. of 1944, pp. 686-93.

Geochronology as an Aid to

Study of Arid Lands

TERAH L. SMILEY

Laboratory of Tree-Ring Research, University
of Arizona, Tucson, Arizona

Researches into past climate have yielded information which
illustrates that local climate is ever changing. The degree to which
this change occurs varies in time and in area. First attempts to
define climatic areas, or areas wherein the climate is essentially
homogeneous, met with about as much success as first attempts
to define the size and shape of an amoeba, both being equally
fluid and equally alive, appearing never to repeat themselves in
exactly the same pattern or shape. Additional research brought to
light the fact that in both studies there is more than meets the
eye, there is cause and effect which has to be determined and
understood. Geochronology can aid in understanding the effects
of climate although it cannot help in determining the cause.

Geochronology 1 is operationally defined as a field of study en-
compassing all scientific methods which can be applied to the
dating of terrestrial events. Climatic change 1s an event which
falls within this category. These methods are used to date changes
which have occurred over long periods of time to determine the
duration of change, the length of visible climatic patterns, and the
recurring climatic cyclics so we may view them in their proper
perspective and weigh them accordingly. No evidence of what
might be called a true cycle has been discovered even with recently
refined dating techniques. All evidence indicates a variable (in
time and in intensity) cyclic-like pattern of change which can
only be interpreted on a relative scale. One specific approach, 1.e.,

161

162 THE FUTURE OF ARID LANDS

palynology, pedology, or dendrochronology, gives information
along a prescribed line, whereas another approach gives different
data. Through correlating and, or, superimposing one over the
other(s), the climatic picture begins to evolve.

Tree Rings and Relative Rainfall

The first method I should like to discuss is dendrochronology.
This method yields absolute dates on specific tree rings according
to the growth year, and it yields relative rainfall patterns. There
is, at present, no technique known in which tree-ring data can
give information along the lines of absolute quantities. For ex-
ample, we can establish departures from a running mean which in
itself is the average of the data in hand; we can determine if a
year, or a period of years, is below average, average, or above
average in growth, but no method is yet known that will allow
us to determine the absolute amount of rainfall represented by
that average. Because many factors influence the growth of a tree,
it is impossible to isolate completely one physiological or environ-
mental factor from all the others. Thus we are not positive that
we have a true or absolute representation of the rainfall patterns in
contrast to patterns representative of other growth factors.

In the course of our tree-ring studies, thousands of samples from
living trees as well as historic and prehistoric specimens from over
the southwestern United States have been analyzed. Working with
the archaeological material supplied in quantity by southwestern
excavators, we have developed regional tree-ring indices for the
major river drainages and their related geographic regions in this
general climatic area. These local chronologies extend into pre-
historic times. They are limited by the material available; conse-
quently their individual lengths of time vary from one region to
another. The study of these regional indices is profitable because
they indicate the less pronounced variability that has existed
from one region to another within the larger and relatively homo-
geneous climatic area. The tree-ring “droughts,”’ or periods of
deficient rainfall as exhibited in tree growth, can be considered as
“droughts” only inasmuch as they fall below the average.

Aside from the relative quality of these deficient periods, there

GEOCHRONOLOGY 163

is also the variable intensity of deficiency from one region to
another. For example, what might be exhibited as an extremely
deficient period in the Rio Grande region will be less severe (at
least the trees will have much more relative growth) in the Mesa
Verde region of southwestern Colorado, and the actual dates on
the two respective periods will agree only in a general way. The
evidence indicates that we should have a separate ring series for
each area because there is no single period of equal intensity and
duration over the entire Southwest.

It is impossible to determine the major overall trend for either
the individual region or the entire area at the present time,
although future studies may allow a partial solution of this prob-
lem. In other words, we cannot predict what will happen next
year, or the following year, or during any future year. As long as
there is enough moisture to support tree growth, we can determine
the departures from the average, but this average may be at a
minimum or a maximum or any point in between, although it must
be within the prescribed amount of rainfall oth which a par-
ticular species of tree can grow. Pinus ponderosa can grow, for
example, in areas where the average rainfall is around 14 inches,
and it also grows where there is as much as 35 or more inches,
wherein it begins to be crowded out by other species. One area,
near Bend, Oregon, has a good stand of ponderosa where the
average annual rainfall is but 12 inches; this growth is possible
because the moisture comes at the right time of the year and the
tree is able to utilize all that does fall. From these figures, we can
say that the average yearly indices can vary from 12 to 40 inches
at least. Other factors, biotic, edaphic, etc., can influence both
the maxima and the minima to the extent that these figures must
be considered as approximations.

The subjective diagrammatic deficiency period chart, Figure 1,
gives what I believe to be the best representation of the periods of
deficiency for five major areas in the Southwest. Two of these
areas, Durango and northeastern Arizona, are incomplete. From
this chart, one can see that there is a period from ca. 50 B.c. to ca.
A.D. 320 during which there were very few of these so-called
droughts. Beginning with a deficient period around a.D. 320,

164 THE FUTURE OF ARID LANDS

however, such periods were common and more intense until the
late 1200’s when a long deficient period occurred. The 1300’s were
mostly good, as were the 1400’s. Several periods occurred during
the 1500’s, with a severe one at the close of the century. From
1600 to the present, conditions have been favorable, and a short
interval during the turn of the twentieth century 1s about the only
really deficient period. The horizontal line in the deficiency curve
at the bottom of the figure has no quantitative meaning; it could
represent 12 inches, or 40 inches, or any point in between, and it
could, in itself, be a fluctuating average or curve which it probably
should be.

Possible changes in the average annual rainfall are indicated, I
believe, by the following bits of information. Samples obtained
from living trees growing on the higher mountain ranges of south-
ern Arizona rarely have chronologies extending to before A.D. 1600.
Detailed studies now beginning on species identification of woods
found in archaeological sites scattered over the Southwest indicate
that a high percentage of specimens from a site, say dating around
A.D. 500, are Pinus edulis, yet today the area is well covered with
Pinus ponderosa. Tentative studies indicate that the presence of
prosopis in archaeological sites located on the Sonoran Desert,
seems to be confined to only those sites which are fairly recent in
time, say the last 300 or 400 years. Charred hickory (Carya sp.)
was found in the Double Adobe site located about 12 miles
northwest of Douglas, Arizona. Antevs has placed an age of about
g,000 to 10,000 years on this site. The presence of hickory indi-
cates a far more moist climate prevailed than there is at the
present time. The closest hickory, today, is found in eastern
Texas, about 800 to 1,000 miles to the east.

It is interesting to note, although it is beyond the scope of our
purpose here to go into detailed analysis, that Figure 1 has much
meaning archaeologically speaking. Studies concerning the correla-
tion of shifting populations during times of deficient periods of
rainfall, as expressed in tree rings, are now being inaugurated.
Evidence based on past studies indicates that there was, during
such times of stress, considerable shifting of the population centers
from one region to another.

"ySOMYINOG 9Y} UT yaoi 30.1} Jo spotied ualyoq

AONZIOIS3Q HLMOYDS 4O SGOl¥3d M JF

£l

S3I¥3S ONIY JO HLINAT -——————‘“+4

ci]

fe}]

“I dINSI

2 Lf (0)
SSS SS
|

£ 2 ' (0)

(SAILO3rENS)
SAYND AONIIDNIS30

vayVv VNOZIYV 3N

vayuv OONVYNG

vau¥v 3J0Y43A VS3W

vauv sj4VLS9V13

vauv JONVYS OY

S31YNLNI9O

165

166 THE FUTURE OF ARID LANDS

Geochronological Approach

Antevs has done much work in determining the paleoclimatic
story in the Southwest. His researches, based on geologic-climatic
observation and study, are the basis for much of our interpretation
of paleoclimate during the pre-Christian eras. Much of his work
was conducted in conjunction with archaeological problems con-
cerning ‘‘early man”’ discoveries in the area. Thus we have some
spotty knowledge of climate and man for the last 12,000 years,
although there are large gaps in this period for which we have no
information. The data derived by Antevs are based on the rise and
fall of lake levels, arroyo cutting and filling, the formation of
caliche, the deposition of laminated beds, and other geologic and
climatic features.

The detailed study of archaeological chronology in the Southwest
is now being started at the University of Arizona through the use
of tree-ring dating and carbon-14 analysis. Many dates on various
stratigraphical correlations will be necessary for minute analysis
of this problem.

Identification of the flora and fauna found associated with such
archaeological sites will also be one of the major problems in this
study. Studies in palynology have been recently started for this
area, and preliminary runs indicate far better results than had
been expected. These runs indicate that ‘‘caves may be to the
arid regions what peat bogs are to the glaciated areas.” In caves
which have been occupied by man, and many have been say for
10,000 years, we have found not only wind-blown pollen but also
pollen from plants which were insect pollinated and which may
have been brought in by man for preparation of food. Other caves,
not occupied by man, give good wind-blown series plus grains from
plants which were probably carried in by animals. The open clay
deposits, or cienegas, in the valley bottoms also yield pollen
although not as much as do the caves.

It is hoped that the pollen spectra associated with archaeological
sites in which specific identification of plant and animal remains
can be made will give a more accurate understanding of detailed
climatic indices for such things as rainfall and temperatures. The
dating of these sites would enable us to locate them along the time

GEOCHRONOLOGY 167

scale so that our compilation of the paleoclimatic history for the
area can then be continued.

The detailed study of geomorphological conditions to determine
the aridity of the land and its associated plant cover is also 1m-
portant to our work. Such periods as those of arroyo cutting and
filling are normally marked, or located nearby, by archaeological
sites. Thus we have some method of determining their age.

The study of the migrations of small modern mammals, as well
as fossil remains, will greatly enhance our knowledge of the local
conditions which prevailed at the time the animals lived in the
area. This, coupled with the study of other types of fossils, should
throw some light on the question as to why certain species of
animals have either migrated or become extinct in this area.

Several preliminary tests made in the playa lake beds in the
Southwest indicate that detailed studies in paleolimnology and
associated fields of micropaleontology, paleobotany, palynology,
paleoecology, and non-glacial varve analysis would be most useful.
From such studies we should be able to deduce specific climatic
indices which could then be fitted to a time scale through the use
of either carbon-14 analysis, or tree-ring dating, or some other
method of dating. It is hoped that such studies can be extended
into the third interglacial times so that an understanding of true
interglacial conditions which prevailed in this area could be had.
It is also essential that the so-called thermal maximum period
(ca. 5500-2000 B.c.) be carefully studied for its detailed effects.

Figure 2 is a tentative summary of the past climatic conditions,
on a generalized basis, that have existed in the Southwest for the
last 10,000 years; beyond this is an extrapolated period extending
to the climax of the Wisconsin glaciation at ca. 25,000 B.c. This
diagram has been compiled from many sources, mainly Antevs,
Fries, Bryan, Ahlmann, and others. It is not intended that it be
considered accurate. Rather it is intended to represent a working
model to guide our future research. The weakest portion is that of
the flora for which there is but little local information. It is hoped
that with the study of palynology accelerating as it is, we soon can
have a more accurate picture of this aspect. Our intentions are
that, as we work on a locality and we incorporate the specific data

THE FUTURE OF ARID LANDS

168

JWOS AWIL
QOO'IT 000'ZT OOO'ET O00'PI O00'ST 000'9T O00'ZT 000'8I O00'6T 000'02 OO0'I2 000'22 CO0'EZ O000'r2 000'Gz

AYNLSION

VO Td -aoyatv

YNNV4
SYNLVYAdWALl
(SAZ.LNV)
Se eae ean ae SGM UIAONWH | oNOSaTLLVYA G1gIdDNIudS -N3AVH M3N| | -— HOWSNOWH
aMousWad ANN LNOWIYYT9 ISuaHWW NIVLINa
SF oa SE eo Se lee Soe

169

GEOCHRONOLOGY

((€z) Aaytug ur sivedde japour
dejiuis Y) ‘uonvuvidxa 10j 3x93 99g ‘sivak ooo‘Lz Isv] OY JO; Japour oNvUULIZeIP 9ANQYIUD] YW “st aANST
JIWOS SIWIL
<"av ‘O'a>
oo0z O00OT 0 O00T 0002 000€ 000% 0005 0009 0004 0008 0006 000'01 000'IT oOO'zt

JYNLSIOW

VuOls OILNVILY-ans qvayog-ans
YNAVW4 $al0adS NYAGOW SH10ddS «ANFOOLSI9Id WOldAL»
SYNLVYSdW3l
(SASLNY)
SdOlY¥3d IWWYSHLOIN TWWHYAHLIGSW WWYSHLYNY ane aga mh
SSAYVA NVOIMSINV 'N oe : ONIWVSINIL
SSAYVA NVAdOUNA 'N bas >

IVNWYAHLOAN

170 THE FUTURE OF ARID LANDS

in its proper time position, we will be able to correct this model
as we go along. One of the eventual aims of this program in
geochronology is to determine the specific climatic history of the
Southwest for the last 50,000 or so years in as detailed a fashion as

possible.

REFERENCES

1. Ahlmann, H. W. 1953. Glacier Variations and Climatic Fluctuations.
Bowman Memorial Series No. HI. The American Geographical
Society, New York.

2. Anderson, Roger Y. 19§5. Pollen analysis: a research tool for the
study of cave deposits. American Antiquity, Vol. XXI, No. 1,
pp. 84-85. Salt Lake City.

3. Antevs, Ernst. 1952. Arroyo cutting and filling. F. Geol. 60, 375-85.

4. Antevs, Ernst. 1953. Geochronology of the deglacial and neothermal
ages. F. Geol. 61, 195-230.

5. Antevs, Ernst. 1954. Climate of New Mexico during the last glacio-
pluvial. 7. Geol. 62, 182-91.

6. Antevs, Ernst. 1955. Geologic-climatic dating in the West. dmeri-
can Antiquity, Vol. 20, No. 4, University of Utah Press, Salt Lake
City.

ae Antevs, Ernst. 1955. Geologic-climatic method of dating. In Geo-
chronology in the Southwestern United States, Verah L. Smiley, editor.
Phys. Science Bull. No. 2, pp.151-169, University of Arizona, Tucson.

8. Byran, Kirk. 1954. The geology of Chaco Canyon, New Mexico in
relation to the life and remains of the prehistoric peoples of Pueblo
Bonito. Smithsonian Misc. Collections 122, No. 7.

g. Deevey, Edward S., Jr. 1953. Paleolimnology and climate. In
Climatic Change, Evidence, Cause, and Effects. Harlow Shapley,
editor. Harvard University, Cambridge.

to. Douglass, A. E. 1919. Climatic cycles and tree growth. Carnegie
Inst. Wash. Pub. 289, Vol. I.

it, Diol, alo 18, TORS, Mel, Wok Il

12) Douslassya5 Ee to3belCids vole Te

13. Fries, Magnus. 1951. Pollenanalytische Zeugnisse der spatquartaren
Vegetationsentwicklung, heuptsachlich der Waldegeschichte, im
nordwestlichen Gétaland (Stidschweden). (Swedish with German
Summary) Acta Phytogeographica Suecica, No. 29, Uppsala.

14. Hack, John T. 1942. The changing physical environment of the
Hopi indians of Arizona. Papers of the Peabody Museum. of American
Archaeology and Ethnology 35, No. 1.

15. Hunt, C. B. 1953. Pleistocene-recent boundary in the Rocky Moun-
tain region. U. S. Geol. Survey, Bull. 966-A, pp. 1-25.

16:

Wo
18.

19.

INO).

lle

DD.

GEOCHRONOLOGY WA

Sayles, E. B., and Ernst Antevs. 1941. The Cochise culture. Medal-
lion Papers No. 29. Gila Pueblo, Globe.

Sears, P. B. 1950. Pollen analysis in Old and New Mexico. Bull.
Gaol, Soe, Zips iks 1s, IN@, We, wi5/Ai-

Sears, P. B., and Kathryn H. Clisby. 1952. Two long climatic
records. Science 116, No. 3007, 176-79.

Schulman, Edmund. 1951. Tree-ring indices of rainfall, temperature,
and river flow. Compendium of Meteorology. Thomas F. Malone,
editor, pp. 1024-29. American Meteorological Society, Boston.
Schulman, Edmund. 1954. Dendroclimatic changes in semiarid
regions. Tree-Ring Bulletin 20, Nos. 3-4, 26-30. University of
Arizona, Tucson.

Smiley, Terah L. 1951. A summary of tree-ring dates from some
southwestern archaeological sites. Laboratory of Tree-Ring Research
Bull. No. 5. University of Arizona, Tucson.

Smiley, Terah L., Stanley A. Stubbs, and Bryant Bannister. 1953.
A foundation for the dating of some late archaeological sites in the
Rio Grande area, New Mexico: Based on studies in tree-ring methods
and pottery analysis. Laboratory of Tree-Ring Research Bull. No. 6,
University of Arizona, Tucson.

Smiley, Terah L., editor. 1955. Geochronology, with special refer-
ence to southwestern United States. Phys. Science Bull. No. 2,
University of Arizona, Tucson.

Summary Statement

REED W. BAILEY

Intermountain Forest and Range Experiment
Station, United States Department of
Agriculture, Ogden, Utah

Not only have variability and predictability of water supply
been discussed but also the broader subject of water supply in
general. The major overall problem of arid lands is the need for
more water. Lack of water has limited occupancy and develop-
ment of arid lands since the beginning of history. It is important
to understand the sources of water available to arid lands, the
amounts, quality, variability, and predictability.

Arid land economies must rely on three sources of water: pre-
cipitation on the arid lands themselves, ground water, and stream
flow from more humid areas. Each of these water supplies has
distinct characteristics and presents different problems.

Precipitation on Arid Lands

Starting first with arid land precipitation, we can make several
general observations. The precipitation on these areas is meager—
usually much less than the evapotranspiration potential. The
amount of precipitation varies greatly from year to year and from
place to place in the same year. Droughts are common and some-
times extend over several years at a time. Forage and crop pro-
duction is highly uncertain.

We are agreed, I think, that the major obstacle to greater and
more efficient use of this limited and variable water supply is the
lack of a reliable method of predicting the occurrence of favorable
and unfavorable moisture conditions. Though much has been
learned about the probable recurrence interval of wet and dry

172

SUMMARY STATEMENT 173

years over a period of time, we have yet to learn how to predict
conditions for specific years and seasons. Solution of this problem
calls for further knowledge about the causes of climatic fluctua-
tions. More research on the physics of the air, and especially on
air-mass movements, is needed if we are to obtain that knowledge.

What runoff there is from arid lands occurs mainly as overland
flow during brief but torrential rains. These storm flows com-
monly are of short duration. They sometimes develop into violent
floods and almost always carry great quantities of sediment. It
has been pointed out that the water in these discharges is therefore
of limited usefulness unless it can be impounded and desilted.
Furthermore, though the arid lands yield very little runoff, they
are a major source of damaging sediment. I should like to under-
line this point with some figures of my own. About 85% of the
lands in the Upper Colorado River drainage basin are arid. They
contribute less than 10% of the water to the flow of the river, but
they are the source of more than 85 % of the sediment that is being
deposited in Lake Mead behind Hoover Dam.

We may conclude from what has been said that the prospects
are not very promising for increasing usable amounts of runoff
from our arid lands. The main problems are to get better control
of the runoff, reduce erosion and sedimentation rates, and to con-
serve the meager supplies that are available.

Ground Water

The second source of water in arid regions is ground water.
Springs and wells have been one of the more important sources of
water supplies in arid regions down through the ages. They have
been the principal and the most reliable source of water for domes-
tic needs, for irrigation, and for stock-watering purposes. The big
question seems to be what are the possibilities of getting more
sustained and usable yield from ground water sources?

It seems to be agreed that virtually all the readily accessible
ground water supplies have already been located and put to use.
New supplies must therefore come from new discoveries. The con-
sensus is also that we are not likely greatly to increase the water
supply by new discoveries of underground sources. I might go

174 THE FUTURE OF ARID LANDS

one step farther and say that in some localities underground water
is being used at a much more rapid rate than it is being replen-
ished.

We now have at our disposal highly efficient sounding devices
for locating ground water supplies. We also have drills that can
tap those sources and pumps to extract the water. However, there
are three obstacles standing in the way of utilizing new water
supplies: (1) a lack of information about ground water origin;
(2) inadequate knowledge of ground water recharge rates; (3) lack
of knowledge about the usability of impure water and economical
ways of making it more usable. Research in these fields is already
under way, but there is need for more.

Water from Humid Areas

A third source of water for arid lands is from streams that origi-
nate elsewhere. Like most of the ground water of great conse-
quence, this inflow is from more humid areas. The origins of the
streams in some places are humid islands within the arid zone and
in others they are situated at great distances.

The streams that flow from these humid areas in many arid
areas are the principal source of water for irrigation, power, and
manufacturing. The Rio Grande Valley is truly arid, and the
scale of occupancy and development as we find it today would not
have been possible except for the water—both ground and surface
—that comes to it from the humid mountains in the headwaters
of the Rio Grande. The highly productive Imperial Valley in
Southern California would still be a desert if it were not for the
availability of water from the Colorado River.

There has been a great expansion in the development of moun-
tain-born waters for use in arid zones in recent years. Great
storage dams, diversion works, and canals are being constructed
on most of the great rivers of the world and still more are planned.
Therein lies the big opportunity of more water for developing the
arid regions.

Several ideas presented in the discussion are worth repeating.
This outside or inflow water supply has some important limitations.
It is subject to annual and seasonal fluctuations. Periods of flow

SUMMARY STATEMENT WAS)

do not always coincide with periods of need for water. Many
streams carry objectionable quantities of sediment. Moreover,
the normal regime of flow can be altered and its sediment load
greatly increased by unwise use of the watershed lands.

Inasmuch as most of the comments about inflowing water came
out of the discussion, rather than from the papers, I would like
to develop this aspect of the subject a little more fully.

Dams and reservoirs are a well-known means of regulating the
availability of water from streams. They do not, however, control
water on the land or maintain the soil in place on watershed
slopes. The greatest threats to the usefulness of streamflow as a
source of water for arid lands are sedimentation in downstream
water storage structures and floods. The control of these menaces
and the increasing of water yields pose many problems in the field
of upstream watershed management. Among them are:

1. Inadequate understanding of the source of sediment and of
soil formation and erosion processes.

2. The necessity of distinguishing between normal and accel-
erated erosion, for which adequate criteria have not been devel-
oped in many places.

3. The need for controlling accelerated erosion on slopes and in
channels for which effective measures are not yet known in many
places.

4. Prevention of watershed deterioration under the impact of
increasing demands for use of the timber and forage resources,
for which guides are also lacking.

5. Inadequate information about the quantitative effects of
watershed treatments on sediment production and streamflow
characteristics.

Pellek USE OF PRESENT
RESOURCES

Questions

What are the possibilities of increasing and maintaining sustained
production from grass and forest lands without accelerating erosion?
What are the consequences of utilizing arid lands beyond their
capabilities?

What constitutes wise allocation of available water supplies among
the various needs in arid land drainage areas?

How can production be increased from existing water supplies?
Can irrigated lands be occupied permanently ?

: ie
i, a
Ahk ‘
ie res
a
4

Grazing Resources

R. O. WHYTE

Plant Production Branch, Agriculture Division,
Food and Agriculture Organization of the UN,
Rome

In order to avoid generalizations on this very extensive subject
which may be applicable to one area and not to others, the discus-
sion will be limited to countries in regions in which experience has
been gained under current FAO activities. These include the
Working Party on Mediterranean Pasture and Fodder Develop-
ment, the Working Party on the Development of the Grazing and
Fodder Resources of the Near East, and the Co-ordinated Grass-
land and Fodder Research Scheme of the Indian Council of Agri-
cultural Research. These three activities cover a belt from Portu-
gal to East Pakistan, characterized by lands which have been used
and misused for centuries, by several different types of climate,
and by a wide range of social and land-use systems. Throughout
all of them aridity is a dominant factor.

Grazing Lands from Portugal to East Pakistan

The Mediterranean climate is primarily one of winter rainfall
and summer droughts, increasing in intensity from the west to the
south and the east as one proceeds farther away from the ameli-
orating influences of the Atlantic and comes nearer to the Sahara
and Arabian Desert. Certain regions of Spain and Turkey differ
in having a continental type of climate resembling that in parts
of North America. The degree of aridity in this region is related
primarily to the duration and intensity of the summer drought;
the problems are greatest in the countries of North Africa lying
along the desert fringe of the Sahara. Most of the countries

179

180 THE FUTURE OF ARID LANDS

covered by the Working Party on the Development of the Grazing
and Fodder Resources of the Near East are also the winter-rainfall
type, and here the aridity of the summers is intensified in relation
to their proximity to the Arabian Desert. The winter rainfall is
generally lower than in much of the Mediterranean area and is
more erratic and unreliable in annual total and in geographical
distribution.

Passing eastward through Pakistan to India one enters a
monsoonal environment in which, in the semi-arid regions of the
west, a low, erratic, wet monsoon occurs between mid-July and
mid-September and almost complete drought for the rest of the
year, combined first with low and later with high temperatures.

The history of land use in these regions extends far beyond the
time of scientific observation and control, when early civilizations
were destroying the original vegetative cover for grazing, cultiva-
tion, fuel, house and ship construction and by the waging of wars.
By a study of early writings and of the many fragments of evi-
dence which are becoming available, an attempt is being made to
recreate a picture of the vegetation of that time and to trace its
use, misuse, and deterioration through man’s employment of the
axe, firestick, the plough, and the grazing animal.

Although there certainly were in the region under consideration
extensive areas of true desert or semi-desert, these have been
gradually extended by misuse of land and vegetation. There has
thus been a great increase in the total area of land affected by
aridity and desiccation; these conditions, so unfavorable for plant
growth and agriculture, have extended into areas not truly semi-
arid in type and have seriously affected natural reforestation or
regeneration of vegetative cover on grazing lands, as well as the
cultivation of crops. We must therefore attempt to distinguish
between the true deserts and the man-made semi-arid lands which
are those most susceptible of improvement.

In this connection it should be noted that there is little evidence
of climatic change in these areas, at least within the period during
which the major part of the de-vegetation has taken: place, nor 1s
there any indication that regeneration of the vegetative cover of
forest, grass, or semi-desert scrub is likely to change the overall

GRAZING RESOURCES 181

climatic pattern of today. The efficiency of the rainfall has, how-
ever, been greatly reduced owing to the deterioration of the vege-
tative cover and the increased evaporation from the soil surface,
unprotected from the sun’s rays; the temperature, velocity, and
sandload of the searing desert winds which blow over these
denuded areas have a serious effect on the remaining vegetation,
crops and trees, and effectively inhibit or retard regeneration.

Although it may be a desirable objective to attempt to regen-
erate some type of vegetation on all semi-arid lands, attention
must obviously be directed first to those regions in which it is
possible to carry out practical, and, as far as possible, economic
measures within a reasonable period of time. The delineation of
such areas would have to bear some relation to the amount and
seasonal distribution of rainfall, the incidence of drought years,
the present stage of deterioration, and other factors. In defining
the area for immediate operations, it will be necessary to consider
current forms of land use, the grazing systems on the desert and
semi-desert lands, the extent of the practice of the cereal-fallow
rotation in systems of shifting and settled agriculture, and the
lower limits of rainfall efficiency at which simple systems of rota-
tions of crops or reforestation are practicable.

Throughout the semi-arid lands of the Mediterranean, the Near
East, and India are to be found many types of rural sociology,
village or tribal structure, religious belief and prejudice, and
methods of crop and animal husbandry, but perhaps the outstand-
ing factor which has to be considered, and which has played a more
important part in the destruction of the vegetation, the prevention
of its regeneration, and the increase of desiccation, is the socio-
logical distinction between the shepherd and the cultivator. One
writer in another part of the world has related this distinction to
the patrilineal and matrilineal types of family structure and in-
heritance. The shepherd has for centuries been a freedom-loving
nomadic type, owning no land and having no interest in the prac-
tice of proper land use for the sake of his village, tribe, or nation.
The cultivator is the sedentary type, satisfied with the land around
his home. Both have destroyed the vegetative cover for their
separate purposes. As populations and livestock numbers have

182 THE FUTURE OF ARID LANDS

increased, so has also the conflict between shepherd and cultivator
in the struggle for dwindling land resources in an environment of
increasing desiccation which is their common inheritance.

Possible Measures for Improvement

The present picture indicates the consequences of using arid and
semi-arid lands beyond their capabilities and of inducing aridity
in regions not actually arid or semi-arid climatically. What are
the possibilities of increasing and maintaining sustained produc-
tion from grazing, cultivated and forest lands in these climatic
zones, while at the same time ensuring optimal conservation of
soil and water and reducing desiccation?

Basic Surveys

It is considered to be essential in the first place to survey and
map the existing plant cover on the uncultivated land, and also
to map the cultivated lands in relation to their farming systems
and the crops grown thereon. This information, combined with
our present knowledge of the possible extent of improvement, will
contribute materially, as far as the arid and semi-arid lands are
concerned, to correct land classification and so to the resources
survey upon which FAO is likely to be engaged in the coming
years.

There is, in the regions under review, considerable activity at
the moment in the survey, analysis, and mapping of the natural
vegetation. Teams of ecologists trained in France are working in
French Morocco, Algeria, and Tunisia. Work is in progress in
Portugal, an FAO botanist-ecologist is working in the countries
covered by the Near East Working Party, and a grassland survey
team of the Indian Council of Agricultural Research is studying
the grass vegetation of that country with reference to botanical
composition, yield, and carrying capacity and the relation of the
existing grass vegetation to the potential grass sub-climaxes, to
soil type, and to the climax type of forest in each vegetation zone.
The data which are becoming available from these plant sociologi-
cal and ecological studies will be of great practical value in the
optimal development of land resources, by the use of vegetation-

GRAZING RESOURCES 183

cum-soil indicators in the limitation of livestock carrying capacity,
the selection of areas for reseeding, cultivation, reforestation, and
other forms of land development.

It is well known that there are many different methods of
analyzing vegetation in current use, as well as different standards
for nomenclature in plant associations, and different techniques
of mapping. A survey of the grassland and grazing resources of
the world as a whole or of any particular part or region of it will be
most difficult to conduct if some agreement on the methods of
analysis and presentation of results is not achieved. It is the ob-
jective of FAO to call together a group of specialists in this field
in order to obtain the desired uniformity, at least as far as the
vegetation on grazing lands is concerned.

Reducing Pressure of Livestock

Since the natural vegetation in this belt from Portugal to
Pakistan is primarily a grazing resource, attention must be de-
voted in the first place to reducing the pressure of livestock on the
land. These livestock are mostly maintained under various sys-
tems of free-range grazing, based on different forms of nomadism.
The types of domestic animals are those which can exist on pre-
carious supplies of fodder and water and under conditions of wide
climatic extremes. A change from more destructive to less de-
structive types and an improvement in the general quality and
productivity of livestock both depend primarily on improved
amounts and quality of fodders and the elimination of excessive
losses in critical seasons or years.

The pressure of people and livestock on the natural vegetation
of arid and semi-arid grazing lands, in which the plant cover is
held at a very low ecological stage due to excessive grazing or
cutting for fuel, can be reduced by the settlement of nomadic
people in selected and especially favorable areas. Although such
action may be possible in restricted areas and with certain types
of social structure, it must be questioned whether it is possible
and in fact even desirable for, for example, the desert Arabs of
the Sahara (la zone steppienne pré-saharienne), the Bedouins of

184 THE FUTURE OF ARID LANDS

the lands bordering the Arabian Desert, and the graziers of the
semi-arid lands of Baluchistan and Rajasthan.

The measures being adopted in Algeria under the Secteurs
d’Amélioration Rurale are probably more realistic under present
conditions and might with advantage be extended to many other
parts of the region. The centers set up by this organization provide
specialized advice and attention to animal health; breeding flocks
are maintained from which the shepherding community can ob-
tain improved rams to replace the low-quality ones in their own
flocks. Reserves of fodder are built up at these centers from crops
grown locally or imported from other parts of the country more
favorable for crop production. Through a system of insurance at
so much per sheep per year, the livestock owners may draw upon
these reserves and so avoid the catastrophic losses which occur in
years of extreme drought. In the severe winter of 1953-54, losses
among flocks not associated with these centers amounted to up
to 40% of adult sheep and 80% of the lambs, whereas losses
where such centers existed were not more than 15% for the pri-
vate owners and 3% for the flocks belonging to the centers.

Such measures, combined with appropriate facilities for market-
ing and distribution of the produce, can be of permanent value
only if livestock owners realize the necessity of maintaining the
same or even a lower number of livestock of improved quality than
before. They should not attempt to keep even greater numbers of
livestock because of the greater security which these centers and
fodder reserves provide. The whole program would in such a case
defeat its own ends and the destruction of the vegetative cover
would proceed even more rapidly.

Better Use of Water

There are, in the regions under consideration, possibilities for
the development and better utilization of water resources. Where
these exist in regions primarily devoted to livestock husbandry,
it would obviously be desirable to utilize some or most of the
water for the cultivation of fodder crops and for the provision of
stock watering points. The fodder so produced might be used to
build up reserves for desert flocks or might be the basis of a more

GRAZING RESOURCES 185

settled form of animal husbandry combined with the cultivation

of food crops.

Regeneration of Vegetation

The main objective in pushing back the desert from those
regions in which aridity is primarily man-induced and limiting
it to the areas with a true desert climate must again be the re-
generation of vegetation by natural and artificial means. This
work must be carried out in regions with ancient forms of land
use and against a background of increasing human and livestock
populations which must continue to obtain their livelihood and
sustenance from the land, while the improvement is going on. If
the biotic factor could be limited for some 5 or 10 years, it would
be possible to make full use of the remarkable capacity of de-
graded types of vegetation to revive, to climb again up the eco-
logical ladder, and to provide a superior type of vegetation. It
would then be possible to evolve new controlled systems of land
use, to provide better sources and types of food, fodder, or fuel,
without again causing the degeneration of the plant cover back
to its present condition. There are many examples within the
area of excellent regeneration of vegetation, even after a few
years’ protection and sometimes under very low rainfalls. Many
more areas protected against grazing, browsing, and cutting are
required to provide the ecologist with his field laboratory in
which he can obtain the data on botanical composition and suc-
cession required by the improver of land use practices.

Many specialists trained in range work in the western United
States have been posted in the countries of North Africa and the
Near East in particular. It is natural that they should attempt
to apply in these new environments the outlook and techniques
found to be successful in their home states. In particular, they
have attempted the reseeding of semi-desert rangelands, with or
without water spreading. With a few exceptions can it be said
that any success has so far been obtained by the reseeding tech-
nique. The difficulties which have arisen are associated chiefly
with the methods adopted and the availability of seeds of adapted
species.

186 THE FUTURE OF ARID LANDS

It appears that more experimentation is needed with regard to
methods of reseeding in the semi-arid lands under review. The
method is likely to be applicable only in limited areas. The
Foreign Operations Administration expert in Iran, Laurence R.
Short, states that reseeding is possible in only about 10% of the
rangeland in that country. In the Desert Range Project in the
Western Desert of Egypt, it has been suggested that wholesale
clearance of desert scrub prior to ploughing and reseeding is not
advisable. The existing vegetation may be cleared in strips and
ploughing, or preferably light disking, and reseeding carried out,
producing the alternation of scrub strip-grass strip rather like
the cereal-fallow strip system adopted in parts of Western Can-
ada and sited across the prevailing wind. Such a system does
not entail so drastic a change in the microclimate at plant level,
the grass stands are protected to some extent from the searing
effects of desert winds and a reservoir of seed plants of adapted
vegetation remains should the reseeding fail.

In French Morocco, where artificial reseeding has failed in
many places, it has been found appropriate to grow seed mother
plants in a nursery and to transplant these in the area to be
revegetated. If protected from grazing, these plants will produce
seed for the establishment of young plants around their bases
and in the protection they themselves provide what might be
called the “hen and chicken” method of reseeding.

Seeds of the adapted species of the semi-arid lands in these
regions are not available in any quantity. Indigenous species
which might well be collected and multiplied include: Dacty/is
hispanica, Stipa lagascae, Cynodon dactylon, Aristida ciliata, Hy-
parrhenia hirta, Avena barbata, and Lolium rigidum in the West-
ern Desert of Egypt, species of Agropyron and low-rainfall
ecotypes of Dactylis hispanica and Phalaris tuberosa on the
Arabian desert fringe, and Cenchrus ciliaris in the Rajasthan
semi-arid lands. A number of the Foreign Operations Adminis-
tration Missions are beginning to collect and employ these
adapted species and ecotypes. Israel has a large scheme under
way and has established a special seed multiplication nursery at
Migdal Askalon. FAO in 1954 in association with the Common-
wealth Scientific and Industrial Research Organization of Aus-

GRAZING RESOURCES 187

tralia carried out a collection of grasses and legumes in the
Mediterranean countries, particularly in North Africa; these 600
species and ecotypes are now being multiplied at Rome through
the cooperation of the Italian government and Professor U. de
Cillis, with a view to introducing them in due course into the
observation trial grounds throughout the Mediterranean, the so-
called Uniform Mediterranean Nurseries. It is hoped to adopt a
similar approach in the Near East in 1956 and so in due course
to have available ample supplies of seed of species adapted to
the conditions of the environments in which reseeding work is
considered to be practicable.

Water spreading is another American practice which is being
tried with varying success in several semi-arid regions of the
Near East. Although it 1s obviously desirable to attempt to hold
the water from the torrential winter rainstorms on the land on
which it falls for the production of fodder and grazing, the tech-
nique is still open to criticism in relation to the expense of con-
struction of the diversion and retention dams, and the erratic
geographical distribution of these rainstorms from one season to
another.

Forestry

The forester also has an important role to play in these arid
and semi-arid regions. Cutting of relict tree and scrub vegetation
for fuel leads to widespread destruction of fuel and ecological
regression. There is great scope for the planting of fuel lots,
shelter belts to provide protection for crop areas or livestock
concentrations, and fodder trees for supplementary feed dur-
ing periods of emergency. The grazier and cultivator must also
realize the ameliorative influence of a forest stand on extremes
of climate and in relation to soil and water conservation. The
forester is still losing ground rapidly on account of encroachment
of grazing in forest areas, the cutting of trees for fuel or build-
ing purposes, the insatiable demand for land for the production
of food and cash crops. An intelligent plan of land utilization
based on land classification must allow for an adequate percent-
age of forest cover, in the right places, which must be composed
of the most desirable species for economic utilization.

188 THE FUTURE OF ARID LANDS

Social Change and Needs

In this brief review it has been possible to say little about the
cultivated land in the arid and semi-arid regions under review.
Much of this land is marginal for crop cultivation, some is farmed
on a semi-desert type of shifting cultivation, and the remainder
can be regarded as suitable for crop cultivation with the adoption
of modern methods of husbandry, the application of fertilizers,
and the use of adapted species. Again we have to accept the fact
that peoples are living in these areas and have to produce their
own requirements locally.

And that is the note upon which this review may be concluded.
The technicians know in many cases what may theoretically be
done to maintain and even increase the production of resources
under semi-arid conditions or how to reduce the severity of these
conditions. But this is like trying to rebuild the highway while
the traffic is still moving on it. Throughout these regions are to
be found graziers and farmers with long experience of maintain-
ing themselves and their crops and animals under some of the
most difficult and rigorous conditions to be found anywhere.
They know fairly well the systems of animal husbandry and
migration to adopt under the circumstances, they know the rela-
tive value and availability of their semi-desert fodder plants,
they know the best grazing areas and go there as soon as news
comes along the desert grapevine that a rain has fallen and growth
is beginning. The technicians must at all times consider them
and their knowledge, customs, and needs, and adopt measures
of improvement which fit in with their social and land use sys-
tems and appear practicable to them. The people on their part
must lose their mistrust and suspicion of the technician, whether
he be sociologist, ecologist, animal husbandman, forester, or irri-
gation engineer. They must realize that what is being attempted,
always dificult and sometimes impossible, is for their own ulti-
mate good. If they cooperate and offer friendly advice and
criticism when they see the technicians going wrong, there is no
doubt that great progress can be made in the rehabilitation of
these semi-arid lands.

Water Resources

L. N. McCLELLAN

Bureau of Reclamation, United States Depart-
ment of the Interior, Denver, Colorado

A century ago, Daniel Webster said: ‘‘What do we want with
this vast worthless area—this region of savages and wild beasts,
of shifting sands and whirlwinds of dust, of cactus and prairie
dogs? To what use could we ever hope to put these great deserts
and those endless mountain ranges?”’

Daniel Webster was a great national leader. I repeat his familiar
quotation concerning the arid western United States not to dis-
parage his memory but to emphasize a point. A wise man was
unable in his day to foresee the wonderful developments that
human progress can bring. Here in the arid west, cactus has given
way to citrus. Potatoes grow where there were prairie dog towns.
Shifting sands and whirlwinds of dust have not prevented the
development of an economy that supports a population of 38
million people. Crop, livestock, and industrial wealth, undreamed
of a hundred years ago, has been created by the energies of man
and his skills in science.

This empire, which last year contributed many billions of dol-
lars to the national wealth, has been made possible largely by the
development of its water resources. Time will tell whether this
generation has better vision than did Mr. Webster’s generation,
and whether we will be able to develop properly water that is still
unused. At least we recognize our duty to try to plan for best use
of the resources given us.

The theme of this symposium, the better use of present re-
sources, obviously could include a variety of resources. This paper
is confined, however, to the water resource. More particularly,

189

190 THE FUTURE OF ARID LANDS

it will bear on the place of irrigation in “wise allocation of available
water.” It also will touch upon a few of the many opportunities
to increase use of existing water and thereby to increase
production.

I should like to preface my remarks on these subjects by recog-
nizing the impressive magnitude and difficulty of the tasks ahead
of us. While noting the great things accomplished, we must realize
that our future endeavor becomes ever larger and more difficult
in direct proportion to the increasingly severe demands on our
water resources. New research tools are needed, and we must make
better use of the old ones to meet the challenges of the future.
Sociologic and economic research and engineering and scientific
research must advance rapidly and keep pace with each other if
we are to plan properly for the complex conditions of the future.

Major Water Uses

There can be no challenge to the ultimate goal, in any arid realm
on the globe, to achieve maximum beneficial use of all available
water supplies. There is little latitude in the allocation of such
limited water resources—direct use of water by man for his direct
benefit is the highest possible beneficial use and comes first. Mu-
nicipal and domestic uses therefore usually assume the first claim
on available water supplies.

The second claim may be for industrial and agricultural use.
All these uses should be considered in the planning for utilization
of any water supply. At the time the engineering works are con-
structed to retain water through storage or to divert it to imme-
diate use, the irrigation allocation may be paramount. As time
passes, the municipal use may increase and this future possibility
should be recognized in planning water utilization projects.

Agricultural purposes are the largest single consumer of water
in the water-deficient areas of this country and of the world. Irri-
gation is a principal use of land in the 17 western states in this
country, where the land area totals 1.1 billion acres. Of these acres,
some 42 million are considered susceptible to irrigation, and some
25 million are already under irrigation. The total water supply
averages about 392 million acre-feet annually. Of this water, only

WATER RESOURCES 191

about 128 million acre-feet can be ultimately used for irrigation.
Presently about 78 million acre-feet have been put to use. In many
western streams, the entire water supply is fully used except for
infrequent years of high flood runoff.

In other words, about one-third of the total stream runoff can
be used for irrigation, and we are now using only one-fifth of the
total. But for the sake of progress toward the goal, it is essential
that all efforts be made to irrigate the remaining 17 million acres,
putting to work much of the remaining s0 million acre-feet of
available water. In many cases this will necessitate diversions of
water between major waterbeds.

For streams such as the Columbia and Missouri Rivers, there
is a large foreseeable flow that cannot be used for irrigation pur-
poses. In such areas, plans are being formulated for the controlled
and coordinated use of water for non-consumptive purposes such
as navigation and power development.

Problems of Long-Term Planning

Whatever allocation of the water resource is made to whatever
purposes, the greatest final benefit to all interests will necessitate
adequate storage. Increasingly as years pass greater use will be
made of both surface and ground water storage. I am not seeking
an argument when I say that surface storage developed to its
maximum is the prime means of developing and utilizing our sur-
face water supplies. Some water should and will be put to work
where it falls—head water, small dam storage has a proper place
in planning full beneficial water development. The fact remains,
however, that reservoirs, whether main stream or tributary, must
be the principal basis for conserving the waters of our surface
streams.

Certain benefits of surface storage are so obvious as to need no
recitation here. One benefit that I would like to bring out is that
storage is a necessity to correct the variable characteristic of
stream flow. Such variability is of a cyclic nature. It is marked by
a tendency for high stream flow to occur in periods of from five
to ten years, and for low flow to extend over periods of from five
to twenty years. Capacities must be provided to retain water from

192 THE FUTURE OF ARID LANDS

high flows to and through droughts. Otherwise we cannot accom-
plish more complete utilization of our limited water. Such facili-
ties must be planned with care so as to afford greatest yields.
They must also be planned and built well in advance of their
actual need to ensure full reservoirs prior to the droughts of the
future.

We are considering broad-scale water development, which es-
sentially is planning for the future. The length of time required
for orderly planning and construction of water conservation works
should be stressed. On the Columbia Basin Project in the State of
Washington, for example, water was first available for irrigation
in 1952. The first report envisioning future development of this
project was made by the Corps of Engineers in 187g—76 years
ago. The Reclamation Service launched its first studies of this
project in 1904. The Columbia Basin Project is one of the world’s
largest and most expensive, but it is typical of large projects. It
illustrates the inevitable passage of time from dreams to reality
in matters of this sort.

Probably every country of the world is experiencing an increase
in population because of many factors, especially including im-
proved medical progress and higher levels of living. The increases
in the United States during recent decades have consistently ex-
ceeded advance estimates. As recently as 1950 the Bureau of the
Census was forecasting that the national population in the year
1975 would be 1go million. Currently this Bureau is cautiously
making its estimate for the year 1975 with a maximum and a
minimum—a range of 198 million to 221 million persons. Whatever
figure proves to be correct, the point is inescapable—water re-
quirements will continue to rise.

Population increase is a greater problem in the western states
where water is scarcer than it is in the rest of the country. The
rate of growth during the decade that ended in 1950 was 25.8%
in the 17 western states, and was 14.5% for the nation as a whole.
The disparity in population growth between the western one-
third of the nation and the other two-thirds has become greater
since 1950.

WATER RESOURCES 193

Allocation of Water Supplies

Man’s domestic requirements inevitably will be satisfied first
and finally, up to the point at which consumption of all available
water sets a limit on expansion of a community. (This is the situ-
ation numerous American cities are approaching today, in the east
as well as in ane west.)

In assessing “‘wise allocation” for types of use other than do-
mestic, more latitude in judgment and planning exists. For exam-
ple, irrigation may be criticized as an extravagant use of water.
Nevertheless, the construction of irrigation works that are efh-
ciently planned and soundly financed is justified by the increase
in food and fiber which this kind of agriculture produces.

One major justification for irrigation is the contribution it makes
to the wealth of the nation. Besides the farm families themselves,
whole communities derive their support from irrigated agriculture.
For each individual living on a farm, there are two more individ-
uals in a nearby community whose support is directly or indi-
rectly due to irrigated agriculture.

The dramatic examples of the new farms and of one-to-two
ratio are in America’s scattered “last frontier.’’ This frontier is a
source of inspiration to those who visit the plateau above the
Columbia River, the previously empty desert lying between Phoe-
nix and the Colorado River, or the prairies and benchlands in the
Missouri Basin receiving irrigation water for the first time.

To measure in dollars part of the value of irrigated agriculture,
I mention that the value of crops irrigated on 69 Bureau of
Reclamation projects was $786,000,000 during 1953 and
$935 ,000,000 1n 1952. Income of such dimensions came from about
6 million acres, or only about one-fourth of the land in the United
States which is irrigated through various organizations.

The same storage structures which make irrigation possible in
many instances bring about other important benefits. The power
that helps to finance the irrigation also contributes to the eco-
nomic base of the west, and the flood control features prevent
destruction of wealth.

Incidentally, the total of values, over and above the direct

194 THE FUTURE OF ARID LANDS

values of crops, livestock, and power, far exceed the portion of
construction that is reimbursable under law.

There are those who advocate reduction of irrigation’s share of
limited water resources because the United States as a nation
produces a surplus of some crops. To these, I say that crop sur-
pluses should not be considered to be a serious liability so long as
there are human beings on the edge of starvation anywhere in the
world.

The crop surpluses of recent years in this country are transient.
Without unceasing efforts in the direction of expansion, by 1975
our surpluses as we know them today will disappear simply under
the impact of increasing population in the United States itself.
Per capita consumption of food also is rising. The quality of our
diet has steadily improved for decades and will continue to im-
prove under the impetus of our rising level of living.

The limitation on tillable acres is real and recognizable. We
know that through simple multiplication of cultivated acreage we
cannot meet the requirements of the future, even under the con-
cept of 100% utilization of the water that is available for irriga-
tion. We must count, however, on continued scientific improve-
ment in agricultural methods, which have yielded such spectacular
successes especially during the last decade.

Irrigation is in the west where the population is growing at the
fastest rate, and efficient distribution methods demand maximum
production close to points of consumption. A final point is that
irrigated land in the west is the source of many specialty off-
season crops which are not competitive with production elsewhere.

Irrigation, then, must have equal consideration in any alloca-
tion of our limited water supplies. Although my references have
sometimes applied specifically to the western part of the United
States, the principles apply equally to any part of the world’s arid
realm.

Increasing Production from Existing Supplies

Many of the world’s rivers, including major ones flowing to the
sea through dry and underdeveloped areas, are not yet put fully
to beneficial use. Such use must be the legacy of science and

WATER RESOURCES 195

engineering to the well-being of the present and future generations.
We must try to do several things—put to use water not now used,
make better use of water now used, and improve the quality of
water.

Many possibilities now are being explored to make better use of
water and to improve its quality. I cannot hope to discuss these
in any detail, and therefore will content myself witha quick sketch-
ing of some of the areas of research which hold promise.

Tremendous quantities of water diverted from streams or reser-
voirs are not actually used. Tens of thousands of miles of irrigation
canals, for example, suffer seepage and other losses up to as much
as one-half the amount of water diverted. A variety of impervious
linings are being placed in these canals in an endeavor to reduce
these losses. In the six years ending in 1952, about 25 million
square yards of linings were placed in more than 750 miles of
canals and laterals on Federal Reclamation projects, saving an
estimated 700,000 acre-feet of water annually. The cost of these
linings is justified by the value of the water saved.

Excessive weed growth in canals increases losses from seepage
and transpiration. Experiments with miscible oils led to the de-
velopment of a plant poison that is toxic in canals but not toxic
on land. The problem is to kill weeds without killing beneficial
plants. An aromatic oil that comes out of suspension and evapo-
rates before the water is diverted to the fields is a partial solution.

As the cost of water increases, western industry is finding it
advantageous to plan for maximum practical reuse of water, even
though the action requires expensive equipment or plant additions.

Planners in the field of agricultural irrigation are ever more
taking into account the potentialities of reuse—sometimes re-
peated reuse of return flows. Water subject to reuse often requires
improvement in quality. The techniques for improvement vary
according to whether the purposes are urban, industrial, or agri-
cultural.

The Department of the Interior is engaged in a pioneer saline
water research program. Although the greater public attention
has been given to conversion of sea water to produce a quality
acceptable for industrial or even domestic use, the research en-

196 THE FUTURE OF ARID LANDS

compasses also saline waters in interior portions of the west.
Improvement of these interior waters will enable not only certain
new uses but also certain reuses. At present, costs of sea water
conversion run upward from $450 per acre-foot of usable water
yielded. It would of course be desirable to reduce this unit cost if
possible to do so. At the same time the present cost is not too much
above the sum needy communities are willing to pay; the city of
Colorado Springs recently offered $350 an acre-foot for water
presently being used for agricultural purposes.

Many cities are learning they no longer can be profligate in
their consumption of limited water resources, even in times of
normal precipitation and storage. They are encouraging or even
requiring their residents to be more economical in water use and
are becoming more rigid in enforcement of regulations governing
the water use by industrial plants. States are inclining more and
more toward requiring treatment of industrial waste and sewage
so as to improve waters that are subject to reuse.

In the field of agriculture, great promise lies in research to
discover the minimum quantities of water required to produce
maximum crops. The Department of Agriculture and the Land
Grant Colleges have made rather substantial progress in this field.
I for one will be pleased when the laboratory experiments and
limited field tests have been expanded into general application of
the principles, and the results are known. Our irrigation districts
and public authorities can do much to educate farmers along the
lines of voluntary water conservation which is in their own in-
terest.

A somewhat related thought is that by persuasion, and perhaps
ultimately by law, farmers must become convinced that parallel-
ing canals, competing irrigation districts, and excessive individual
water applications are luxuries the nation cannot afford.

Operators of storage reservoirs, among whom the Bureau of
Reclamation is the largest in the United States, have only
scratched the surface of the subject of reservoir evaporation.
Means of reducing evaporation are in the experimental stage. We
are actively interested in the principle of using polar compounds
which have the effect of a film on water surfaces. Household

WATER RESOURCES 197

detergents, not oil, are familiar polar compounds. A monomolec-
ular layer, if it could be maintained, would have the effect of
greatly reducing reservoir evaporation, but its harmful effects, if
any, are unknown, and must be carefully studied before seriously
considered for use.

Reforestation, controlled livestock grazing, regulated recreation,

and other accepted practices to avert land erosion have the addi-
tional main purpose of conserving water. The saving of valuable
land and water resources is a goal in itself, but an attendant
benefit is the reduction in the volume of sediment poured into
stream beds. Sediment is a tremendous nuisance, as it reduces
water quality, aggrades channels, makes deposits on cultivated
fields during irrigation, and shortens the life of reservoirs. In
planning projects in areas where sediment is a serious problem,
adequate consideration, from the physical and financial view-
points, must be given to sediment problems.
One of the important measures needed for putting our water
supplies to most beneficial use is improved hydrological tech-
niques and more comprehensive hydrological and meteorological
data. The installation of additional observation and recording
stations during recent years and the accumulation of records
covering longer periods will pay off in terms of more accurate
planning of engineering structures. Recent improvements in
methods for computing maximum probable large and small floods,
together with more precise hydrological data, provide a firmer
basis for saving water.

More accurate estimates of water yield from snow cover will
result in improved operation of irrigation projects. The precision
of flood control operations can be enhanced by better estimates of
the rate of runoff from snow melt. An illustration of the benefits
of better data lies in the research completed and underway at Lake
Hefner, Lake Mead, and the Bruning Air Base which is providing
more precise methods for estimating evaporation and transpira-
tion. These are aids toward more exact use of reservoir capacity,
irrigation, and operating methods.

Evaporation is a major waster of water, but is not first in this
unenviable distinction. Evaporation consumes an estimated 15

198 THE FUTURE OF ARID LANDS

million acre-feet a year in the 17 arid western states. A greater
waster is the class of plants known as phreatophytes. Tamarisk or
salt cedar is one of the better known of these water-loving useless
plants. It has been estimated that in the west these consume more
than 20 million acre-feet of water a year. The importance of the
loss is illustrated by the fact that water developments in Cali-
fornia, which have cost hundreds of millions of dollars to provide
for a population of 12 million, assure a water supply totaling only
21 million acre-feet. The previously used chemical spraying and
mechanical methods of controlling phreatophytes in reservoir
deltas and elsewhere undoubtedly can be improved upon.

Ground water pumping in too many places has been overde-
veloped. Improved state laws, or in some states the passage of
ground water codes, and better enforcement of laws certainly are
needed to improve control of ground water use. Beyond law,
however, we need also additional engineering information about
ground water. Water development planning is giving constantly
greater attention to combined development of surface and ground
water resources.

Important principles are involved not only in the technological
problems sketched above, but lie also in the broader considera-
tions of comprehensive resource development. Such planning is
intended and I am sure will be accomplished by the joint efforts
of local, state, and federal agencies. Beyond local or state bound-
aries there is need for regional research, for analysis of regional
needs properly related to national needs, for analysis of econom-
ical development, and for balanced area development.

The days of quick and inexpensive exploitation of any natural
resource are in the past. Our situation today is such that important
progress can be made only through the unified effort of the many
interests involved and the planning for any given area.

There can be no room for jealousies or enmities between de-
velopmental groups, public or private. We cannot afford compe-
tition between sciences; teamwork is essential between chemists,
geologists, engineers, economists, hydrologists, and agricultural
specialists.

Ill-conceived development is an extravagance no nation in the

WATER RESOURCES 199

world is wealthy enough to afford. Yet water is a renewable re-
source, but it has a value measurable in dollars, and each usable
gallon that flows unused to the sea is an unforgivable waste.

The arid areas of the west and of other parts of the world have
water resources to develop as a foundation for urban centers, for
industry and for the agricultural hinterlands. Research, planning,
and construction should proceed at an increasing rate to make the
promise of today the actuality of tomorrow. There is much to be
done, and we should set ourselves to the task of doing it promptly,
cooperatively, and vigorously.

Geography’s Contribution to the
Better Use of Resources

HILGARD O'REILLY STERNBERG

Centro de Pesquisas de Geografia do Brasil,
Faculdade Nacional de Filosofia, Universidade
do Brasil, Rio de Janeiro

This paper has a twofold purpose. In the first place, it attempts
to furnish a few instances of how the work of geographers can con-
tribute to the better use of present resources. But, since the geo-
graphical approach can best be understood by references to a spe-
cific example, the paper also highlights some aspects of the dry
section of the Nordeste, or northeastern Brazil, afficted with
recurring periods of drought.

The Nordeste

The region we are concerned with consists essentially of an
extensive, but not very elevated, plateau (200-300 meters, or
650-1,000 feet), where most of the moderately upwarped pre-
Cambrian basement, a maturely eroded surface, has been stripped
of its sedimentary covering. Along a considerable section of the
eastern seaboard, the crystallines—under the regional designation
of Borborema plateau—terminate in a much dissected front,
which rises from the 40- to 60-kilometer (roughly 25- to 40-mile)
wide coastal belt and is particularly prominent in the states of
Pernambuco and Paraiba. On the west, approximately at the
state boundary between Ceara and Piaui, the oldland plunges
under an extensive pile of westward-dipping strata. These break
off in an imposing escarpment, the Ibiapaba, which dominates
the exhumed plain by as much as 800 meters (2,600 feet). Exten-

200

GEOGRAPHY’S CONTRIBUTION 201

~ Bs a

Figure 1. The Serra do Pereiro, rising abruptly out of the crystalline
plain, eastern Ceara. Partially destroyed mature landscape of upland
gives way to steep flanks of serra.

sive patches of the once widespread sedimentary blanket have
been left behind as tablelands on the resurrected erosion surface
of the basement complex. More or less isolated massifs of resistant
igneous and metamorphic rock also rise abruptly as mountains
(serras) above the undulating plain (Figure 1). They represent
remnants of a higher, even older, erosion surface or, in some cases,
fault blocks.

Broadly speaking, moisture decreases rapidly from east to
west, hence the threefold, longitudinal division of the Nordeste in
climato-botanical bands, recognizable when striking inland from,
say, Recife or Joao Pessoa: (1) the so-called zona de mata, or
forest zone, a humid coastal region; (2) a transition belt known as
the agreste; and (3) the vast, drier backlands, or sertao, where
precipitation, after reaching minimum values, picks up again
toward the west. The area of more abundant and dependable
rainfall, which parallels the eastern coastline, will not be of im-
mediate interest in this discussion.

In the section with which we are concerned, seasons are sharply
defined by precipitation, not temperature: winter, so-called,

202 THE FUTURE OF ARID LANDS

Figure 2. The Jaguaribe River by the town of the same name, in
Ceara. Notice the square, board-lined water hole dug in the dry river

bed.

roughly the first six months of the year, 1s wet; summer, the
second semester, 1s dry. Thus, for example, data from the ten
meteorological stations maintained in Ceara state by the Servi¢go de
Meteorologia indicate that g1 % of the aggregate annual precipita-
tion normally falls in the winter months. The 3,000 kilometer
(2,000 mile) long Sao Francisco River, rising in the humid moun-
tains of Minas Gerais, flows through the southern section of the
region under consideration. All other streams, fed directly by
surface runoff, are subject to violent floods and, in the dry season,
are sectioned into isolated pools. The Jaguaribe River basin, in
Ceara, is typical. Although it drains an area one and a half times
that of the Rio Grande upstream of Elephant Butte, it is an
intermittent stream, often facetiously referred to as the largest dry
river in the world (Figure 2). At intervals, the rainy season sets in
late and/or shows a marked downward deviation from normal.
When acute moisture deficiencies occur two or more years 1n suc-
cession, all economic activities are disrupted—the séca (drought)
has struck once more.

The fact that the island-like eminences rising above the level
of the old basement—be they crystalline ranges or sedimentary
tablelands—as well as the dissected eastern slopes of the Borbo-
rema, and the Ibiapaba escarpment to the west, are all favored by

GEOGRAPHY’S CONTRIBUTION 203

much more abundant and reliable precipitation influences the
population pattern to no little extent. But it 1s highly relevant to
point out that the drier sertzo, although more sparsely settled, is
by no means uninhabited and commonly sustains a major part of
the total population. Unlike the peoples of more permanently
aggressive environments, who face uninterrupted water deficien-
cies, the sertanejos, who need no special ability to graze their
herds and plant certain crops on this land during normal years,
time and again are taken unawares by the séca. It is hard enough
to develop a system of land use which fully considers ordinary
conditions; to contend with unusual circumstances really calls for
additional determination, organization, skills—and capital. The
region is certainly not organized to make full use of available
water supplies. It would appear, in fact, that the water resource
problems of the Nordeste, are man-made to a larger degree than is
commonly assumed (11).

Delimitation of Drought Area

Accelerated erosion, the signs of which are obvious throughout
the region, signifies not merely soil lost for agriculture, but reduc-
tion in water storage capacity (Figure 3). That outright desert
landscapes may be created by man, even in warm temperate
rainy climates, is a well-known fact, which may be illustrated

Figure a Wilted cornfield on gullied lands of western Ceara. Poti
River Valley.

204 THE FUTURE OF ARID LANDS

Figure 4. A quarter century of misuse created this desert landscape
out of forest-covered lands in Municipio of Cerqueira Cesar, Sao Paulo
State.

with a recent example from the pioneer coftee lands of southern
Brazil (Figure 4). Incorrect agricultural practices that aggravate
floods and droughts are certainly not restricted to the northeastern
section of Brazil, but are the rule throughout the country (G, mo).
It is not surprising, therefore, that much confusion exists con-
cerning the true extent of the lands afflicted by recurrent droughts.
The problem of defining their area is no mere point of academic
interest, for the Brazilian Constitution establishes that:

In implementing plans to provide for protection from the effects of
the so-called séca of the Nordeste, the Union shall expend yearly, on
works and services of economic and social assistance, a sum equivalent
to not less than three per cent of the tributary income.

Paragraph one. One-third of this amount shall be set up as a special
fund with which to succor the populations stricken by the calamity;
this reserve, or part of it, may be applied at moderate interest rates, ac-
cording to law, in loans to farmers and industrialists established within
the area subject to the drought.

Paragraph two. The states lying within the ‘area of the drought”
shall apply three per cent of their tax income in the construction, on a
cooperative basis, of reservoirs, and in other services which may be
necessary to their populations.*

* Law 1005, sanctioned December 24, 1949, establishes rules for the
application of the constitutional provision.

GEOGRAPHY'S CONTRIBUTION 205

Now better use of present resources undoubtedly involves wise
allocation of such funds as are available for the rehabilitation and
development of the semi-arid lands. An overextension of the
bounds drawn for such purpose is contrary to the spirit and to the
letter of the Constitution. By spreading financial resources thin,
it seriously impairs the solution of the specific problems en-
visaged. Here then we have a field of inquiry for the geographer—
and the task of drawing boundaries is one of his least challenged
attributions.

The record shows several attempts to delimit the semi-arid
Nordeste, but I am afraid none of them is very convincing.

The area at present receiving official recognition as water-
deficient is called the ‘“‘Drought-Polygon.”’ The formula by which
it was delineated derives its origin from a report by the head of
the federal agency dealing with droughts. Published some twenty
years ago, immediately after a sequence of years with diminished
rainfall, it is based on the assumption that ‘‘observations carried
out during the dry years 1930, 1931, and 1932 permit a scrupulous
and sufficiently exact delimitation ... of the area subject to the
great periodic droughts, that is to say, the area defined as dry
(séca)” (14). The following procedure was recommended:

After drawing the isohyets corresponding to this triennium, the dry
zone may be circumscribed by a polygonal line which accompanies the
600 mm [23.62 in.] isohyet in such a fashion that [advancing] from the
periphery towards the center, the 300 mm line is always found.

This, the report explains, excludes the semi-arid zone of southern
Bahia, where rainfall did not fall below the 300 mm [11.81 in.] limit in

1932 (14).

The ensuing geometric boundary, with minor variations, was
subsequently written into law.* The same law provides that:

The established boundaries may be altered by law if further observa-
tions should reveal the occurrence of the sécas in other zones of the
northern states with the same characteristics already observed in the
area delimited in the present article.

* Lei No. 175 de 7 de janeiro de 1936, Regula o disposto no artigo 177
da Constituicao.

206 THE FUTURE OF ARID LANDS

It is immaterial to speculate whether these “characteristics”
refer to the zones or landscapes where the droughts obtain or to
the sécas themselves: neither one nor the other has been charac-
terized with any degree of precision. This perhaps may be one
reason why there has been relatively little opposition to subse-
quent amendments, which have enlarged the “‘legal”’ (but obvi-
ously arbitrary) drought area to the point, for instance, where
the boundary has advanced some 700 kilometers (more than 400
miles) in a southwesterly direction (Figure 5).

The way in which a single element was considered in setting up
this operational region may be contrasted with the attention
commonly dispensed by geographers to the interrelation of all
pertinent phenomena occurring in a certain area. Whereas the
amount of precipitation is unquestionably a basic control, it is
not clear why the 600-millimeter isohyet was invested with the
significance of a critical limit. Furthermore, the knowledge that
rainfall operates in the presence of other factors should caution
against the facile acceptance of the simplified sequence of cause
and effect which the adopted criterion implies. Not only must
other climatic components be considered, but it is necessary to
take into account additional influences, such as slope and water-
holding capacity of soils. Field work appears thus as an indispen-
sable ally (not to say part) of climatological studies.

Even supposing agreement had been reached concerning the
value of precipitation to be considered critical in defining the area,
the wisdom of drawing boundaries on the strength of rainfall
measured during one single, exceptional period, such as 1930-32,
remains open to question. The more customary procedure, of
course, would utilize mean or normal conditions, on the basis of
records extending back through the greatest possible number of
years. One might, however, challenge the advisability of obscuring
the unusual experiences in long averages, since, especially in the
Northeast, it is the unusual conditions which are the greatest cause
for concern. How then can one attempt to sharpen the technical
meaning of the expression ‘‘érea da séca,” current in non-technical
usage? A simple and apparently satisfactory approach might be
based on the frequency of the occurrence of drought-producing

GEOGRAPHY’S CONTRIBUTION 207

42° 2° acarab OCE 4 y 380

~\( (E
nt tCAMOcIM Z

R. G.¥DO NORTES”

a H runt | 7

+. JOAO |PESSOA

a } ¢ aa 34°

le Va
pe Ree

2 s :

Sp
©)
|!
/!
|
UPCA R A-I-B-Ary? fir®

MB U\C 0;

~, 7

: Wo '
MARANHAQO ¥ i

{--__--~ ,OEIRAS

\ yy) \ O~

p>
OBA

ACAJU

' xe
IAS BARRETO
Q

\TUCANOG,

wer~

OALAGOINHAS

SIFEIRA DE SANTANA
i u SALVADOR
i iS)

AMARGOSA
1)
/
/
/

140

AS i
Ss.-* IBIPETUBA bl
A
i
U

I
l
Il

Vi ova, 2A )POCOES =<
; rORIA DA CONQQISTA

N
N
z
=<
=
‘=

38°

a Osr. DIVISA
SH yer, ra

\ Td
RIO\PARDOM
is MINAS Xi xg

/. If
onTES L. 7

SME OMINAS NOVAS
LAROS V7 ~~{-OTURMALINA

/ gr socsitva BOUNDARIES OF THE
fe “DROUGHT POLYGON”

(PIRAPORA 420 BRAZIL

Figure 5. Progressive enlargement of the area officially recognized
as subject to droughts. Note overlap with Sao Francisco Valley, to the
development of which the federal government shall also apply not less
than 1% of the tributary income.

208 THE FUTURE OF ARID LANDS

conditions, somewhat along the lines suggested by Russell’s
climatic year concept (7).

It seems evident from the foregoing that the establishment of
boundary lines implies an investigation of the internal structure
of the area to be delimited. The interpretation of regional differ-
ences and local patterns brought to light in the course of such an
investigation, a routine task for the geographer, prepares a val-
uable framework for regional planning. Having an area compar-
able in size to that of France or Spain, the Nordeste could be
expected to include several greatly contrasted natural landscapes.
As already indicated, the region does indeed present considerable
topographic and climatic diversity. There can be no single devel-
opment scheme here, and any long-range program should aim to
extract all possible advantages from the variety of existing con-
ditions. This calls for optimum use of every single tract of land
and a combined development of all resources. The required in-
ventory and appraisal of the possibilities offered by the environ-
ment furnishes the planner with a basis to judge the claims of
competing uses. It also enables him to obtain a better perspective
in the appraisal (or reappraisal) of the significance of what may be
no more than one-sided solutions for the problems of the region.

Water Storage

The traditional approach to the question of insuring a perma-
nent provision of water in the Northeast rests essentially on the
construction of large surface storage reservoirs. Elsewhere, I have
indicated some of the limitations of this solution as applied to the
region (11). However, since it is customary to point to the bene-
fits which the establishment of large reservoirs have brought to
lands of even more acute moisture deficiencies, I should like to
refer to yet another phase of the geographers’ work, the compara-
tive study of widely separated geographical regions. Although the
Nordeste, in respect to certain elements or element complexes,
may be bracketed in the same general type as other arid and semi-
arid lands, an examination of some notable differences shows the
region to be unique in many respects. One example will suffice
here. With a single exception, all its streams rise in the dry zone

GEOGRAPHY’S CONTRIBUTION 209

itself (and every year cease to flow for months on end), whereas
the majority of large storage reservoirs established in the arid
lands with which a parallel is drawn impound waters of rivers
rising in rainy or even snow-covered country. The upper Rio
Grande furnishes an opportune example. Although its basin is a
typical arid zone watershed, the river rises among the snow-clad
peaks of Colorado and is fed by mountain streams.

That reservoirs in the Nordeste have produced even punier
results than could be expected (with around 2,000 hectares or less
than 5,000 acres under irrigation in 1950) is due essentially to (1)
lack of research regarding the multiple factors involved in the
establishment of reservoirs and (2) failure to utilize even what
irrigation potential has been obtained—most agricultural activity
having to do with reservoirs is carried out in the storage basin
itself, where the drop of the water level exposes moist land. How-
ever, these are matters for the hydraulic and irrigation engineers.

The point I want to make here may be a rather surprising one.
But a survey of the existing official literature on the Nordeste and
discussions with professionals responsible for public policy have
led me to conclude that a large part of the opposition to a realistic
appraisal of the remedial measures for the Nordeste stems from
an erroneous interpretation of the landforms in this area. One of
the outstanding features of the region is the large number of
gorges carved through ranges scattered throughout the sertao
(Figure 6). The general belief is that almost all these water gaps
are the result of erosion by the outlets of lakes where waterfalls
poured over the impounding ranges (14). Some interpretations are
hazy as to the origin of these hypothetical lake basins. Others
attribute them to crustal movements; at least one lake is ascribed
to an upwarping during the Caledonian revolution (which took
place some 300 million years ago) (1). All such interpretations
lend comfort to the notion that “it is enough to rebuild the
ranges, breached by erosion, in order to detain the rivers which
escape through these gaps” (2). Once the mountain ranges have
been patched up, man will have reestablished the lakes (which
became extinct by the progressive cutting down of their outlets),
thus providing large volume reservoir storage and even exerting

210 THE FUTURE OF ARID LANDS

Figure 6. Watergap carved by the Jaguaribe River through steeply
dipping quartzite beds of the Serra dos Orés.

a beneficial effect on the climate of the area (3). Although no
proof is presented, the existence of these lakes is unquestioned
among those responsible for the plans formulated and put into
operation in the Nordeste.

To anyone familiar with the study of landforms, evidence that
extensive sedimentary strata blanketed the crystalline undermass
in late Tertiary times suggests an entirely different origin for the
watergaps. This interpretation may be outlined as follows. The
drainage system, freshly initiated upon the unbroken sedimentary
covering, conformed to the surface irregularities of the stratified
beds, being, of course, entirely independent of the topography and
structure of the buried crystalline base. In time, the streams cut
into the underlying basement and sawed across the buried ranges,
from which the covering was ultimately stripped (Figure 7).

It is true that whatever their origin may be, the gorges generally
provide good dam sites. Their abundance, however, hardly justi-
fies the description of this region as “‘the ideal land for the con-
struction of reservoirs” (5). In fact, areas of adequate topography
for large storage systems are not plentiful. In the absence of
indispensable technical studies, the notion that they were to take
the place of former lakes may have contributed to the premature
establishment of some large scale reservoirs. In challenging the
lake theory, the physiographer does not pretend that his broad
interpretations are a substitute for detailed engineering studies,
but they might contribute to avoid construction of dams, before

GEOGRAPHY'S CONTRIBUTION 211

such studies have been properly eftected. There is no substitute,
for instance, for a detailed survey of proposed storage basins, but
the student of landforms who interprets most of the sert@o as a
resurrected fossil plain, a worn-down, old-age surface, can predict
that the terrain, broadly open and unobstructed, is not likely to
provide much depth in proportion to the free water surface of the
reservoir. In short, the geographical approach will indicate that
there is little justification for the belief that construction of such

Figure 7. Suggested explanation for the watergaps carved through
the Serra de Santa Catarina, Paraiba State, by Piancé River and its
tributary, the Aguiar. Sequence of strata in the sedimentary cover is
assumed analogous to that exposed in the residual Araripe tableland,
some 125 kilometers to west. Following superposition on the under-
mass, drainage shows tendency to adjust to structure.

212 THE FUTURE OF ARID LANDS

reservoirs should be promoted almost in terms of a logical response
to conditions of the physical earth.

Promoting Better Use of the Land

The geographer, furthermore, concerns himself with the region
as a whole. He is thus susceptible to the fact that the impounding
of water for irrigation purposes, when unaccompanied by other
measures (as, by and large, has been the case in the Nordeste), at
best benefits only the irrigable lands downstream from the reser-
voir. With the exception of a narrow tract immediately contiguous
to the water line, it is indifferent to the fate of the soils in the
watershed, yet some of these may be counted among the more
promising for agriculture and support a considerable portion of
the rural porno.

The crystalline serras, which rise above the gently undulating
sertao are a case in point. They receive the benefit of abundant
and, what is particularly important, more dependable precipita-
tion. Although the location of meteorological stations, as a rule,
is not such as to favor the portrayal of local climatic contrasts
resulting from the topography, it is obvious that sharp transitions
occur within a very few miles. Observe, for instance, that rainfall
at Meruoca, an upland locality (elevation about 670 meters or
2,200 feet), is more than double that of Sobral, a lowland town
(elevation about 70 meters or 230 feet), only 23 kilometers (14
miles) away: 1,732.3 millimeters (67.56 inches) and 852.4 milli-
meters (33.24 inches), respectively. The higher parts of the relief
seem to derive additional benefit from the fact that they rise into
and are probably dampened by the cloud cover which sweeps
tantalizingly across the low country. The mantle of weathered
rock and soil is thicker, as much as 25 meters (82 feet) depth
having been reported for the Serra da Baixa Verde (13), and the
relics of former forests, still to be found in some of the serras
(Figure 8) may be contrasted with the dry caatinga of the sur-
rounding plains (Figure 9).

Although handicapped by less advantageous topography, the
serras, with their smaller holdings, are devoted largely to farming
—corn, beans, sugar cane, fruits and coffee being typical products.
Saegeed of their former vegetation and improperly cultivated

GEOGRAPHY'S CONTRIBUTION 213

Figure 8. Agricultural pattern on summit of Serra do Meruéca,
western Ceara.

ME PE ana ee

Figure g. An aspect appearance of the caatinga during the dry season.

(Figure 10), they have been ravaged by accelerated soil erosion.
Having destroyed the natural conditions which permitted ample
water storage, man has not replaced them by appropriate mechan-
ical and vegetative controls. Not only is soil impoverished but

214 THE FUTURE OF ARID LANDS

Figure 10. Corn planted on steeply sloping hillsides in the Serra de
Sao Pedro, southern Ceara. -

also increased surface runoff results in drought conditions, even
with high annual precipitation. A great number of springs, which
in the beginning of the last century watered the slopes of moun-
tains such as the Uruburetama and other elevations in the vicinity
of Fortaleza, Ceara, are said to have disappeared when the
hillsides were cleared and planted to cotton and coffee. Although
the area of each of these humid ‘“‘islands,” with deeper, more
fertile soil, is small in comparison with the sertdo, their total area
is considerable, particularly in Ceara.

When the 1950 Census was taken, the numerous serras, in the
state of Ceara alone, supported a population of about 500,000. It
is Interesting to compare this number of people, already estab-
lished on the land, with the total of 400,000, which are to be
settled, in a remote future, according to plans based on the com-
pletion of four major systems of storage and irrigation in various
sections of the Nordeste (4). It appears to me that the compara-
tively small areas of relatively good soil, such as that of the
serras, must be regarded as very precious indeed and that develop-
ment schemes should not merely contemplate the establishment
of artificial nuclei of settlement, with the construction of large
reservoirs, but should make every effort to promote full and

GEOGRAPHY'S CONTRIBUTION 215

competent use of existing, natural cores of more intensive land
occupancy. Since present water deficiencies in the serras are
largely the result of improper use of the soil, such areas, as Whyte
has pointed out, (pp. 182-188), should also be the most susceptible
to improvement—those tracts, that is, where the soils have not
been washed away entirely.

Most of the general ideas advanced with regard to the poten-
tialities of the serras apply to the eastward-facing slopes of the
Borborema plateau, formerly hung with the Atlantic rain forest,
but degraded by centuries of wasteful exploitation.

In discussing the possibilities of increased production from
present water resources, special reference should be made to the
agricultural activities carried on at the foot of the imposing
Ibiapaba escarpment, on the west, and at the base of a large
isolated patch of the once continuous strata—the Chapada do
Araripe.

Rising to an altitude of the order of 1,000 meters (3,300 feet),
the [biapaba is the seat of considerable condensation of moisture.
In addition to farms strung along the crest and, to a great extent,
dependent on rain agriculture, a strip of irrigated cropland

Figure 11. Belt of cropland at foot of Ibiapaba escarpment, in the
vicinity of the town of Ipu, Ceara.

Sakae aa:

216 THE FUTURE OF ARID LANDS

Figure 12. Fern-shaded spring on the flanks of the Araripe tableland

near Crato, Ceara.

extends along the foot of the sedimentary scarp and 1s sustained
by a number of small streams which have nicked and notched the
steep inface and tap the westward dipping aquifers (Figure 11).

The Chapada do Araripe, the main tabular remnant east of the
Ibiapaba, reaches an average altitude of around g60 meters
(3,150 feet) and acts as an immense underground storage reservoir.
Watered by more than 150 springs (Figure 12), a green belt
several miles broad extends along the foot of the tableland in
southern Ceara and suggests a veritable oasis to the traveler
approaching through the dry caatinga. Here, where an intensive
use of the soil is possible, a quarter-million people are concen-
trated. Productivity and value of the land are in direct proportion
to the possibility of it being irrigated. Access to water determines,
for example, just how many hectares of cane an operator can
cultivate.

Becausethe techniques employed in the utilization of the water
issuing from the sandstone beds of the Ibiapaba and Araripe
uplands leave much to be desired, there is unfortunately a con-
siderable wastage of water. It has been estimated that losses in
the open /evadas (irrigation ditches) at the foot of the Araripe

GEOGRAPHY’S CONTRIBUTION PAA

plateau represent more than 50% of the discharge of the springs
and may even exceed 70% (6). As there is more land suitable for
irrigation than water available for that purpose, it follows that
the use of available water at maximum efficiency could lead to
doubling the producing cropland. It is relevant to add that, in
some instances, irrigation water, flowing directly down bare
slopes causes soil depletion and erosion.

The Geographical Approach

To the geographer, a complex province such as the Nordeste
signifies much more than the mere sum of its component units.
Geography does not stop, therefore, with the analysis of each
partial landscape, but tries to clarify the relation of the parts
among themselves. A geographical study will bring out, for in-
stance, the integration of farming in the serras and pastoral activi-
ties in the sertao. It will show how cattle 1s moved to the hills after
the harvest, to graze what residues are available, and how, in
times of drought, both man and beast may find refuge in the
mountains. It might go on to mention that some of the leather
bags carried by the pack trains one meets on trails leading to the
highlands contain manure gathered in the corrals of the sertao,
which also furnishes a market for the produce grown on the serras

(Figure 13).

Figure 13. Pack train descending Ibiapaba escarpment near Ubajara,
Ceara, with produce grown on humid rim of upland.

218 THE FUTURE OF ARID LANDS

Cy: Fertilizer
insecticides ~All :

Maren | Fertilizer Income Phosphate
ae loiranenia eS Equipment eR

| AGRESTE

SERTAO
(Cattle, etc.)

<= Irrigation

ATLANTIC OCEAN

(Food, etc.), =

eltmegation poe | « GE a
= . a

Irrigation | 2 MATA Oo

eS ad SAA (Sugar) =

Labour

om oe Meat

PAULO AFONSOMPN oe) tena Se |

Figure 14. Highly idealized diagram of a development scheme sug-
gested for Pernambuco by Hans W. Singer.

Such an investigation of existing patterns of areal interrelation
furnishes a valuable basis for development plans. Economists
point out that combined projects, which consider the various
diversified parts of a region, are likely to be more productive than
those that are confined to one particular type of landscape.

The geographical approach can also provide a fuller compre-
hension of the spatial relations obtaining between the drought-
afflicted backlands and other areas. The study of such interre-
gional patterns are fundamental to a precise formulation of long
range development schemes of the type tentatively outlined by
Hans Singer for the state of Pernambuco (8). This United Nations
economist has put forward a plan, based on what are avowedly
initial impressions, which endeavors to take advantage of the
different potentialities of the zona da mata, where sugar cane pro-
duction is concentrated; the agreste, fairly heavily populated,
with much production of food crops; and the sertao, with pastoral
activities and specialization in cotton and other dry land crops. The
scheme takes into account the phosphate deposits (an estimated 40
million tons of high grade calcium phosphate), recently discovered
near Recife, an excellent port, and the energy supplied by the Paulo
Afonso hydroplant, inaugurated January 1955 on the Sao Fran-
cisco River. The accompanying, highly idealized, diagram is self-
explanatory (Figure 14).

Conclusion

I hope to have brought out the fact that geography’s contribu-
tion to the solution of problems of arid and semi-arid regions can
be many sided and wide ranging. Geographers have been fully

GEOGRAPHY'S CONTRIBUTION 219

aware of their responsibility in providing more factual knowledge
“to serve as a proper foundation for schemes of improvement and
development, especially in those areas which are commonly re-
garded as underdeveloped.”’* Both as a body professional and as
individuals, geographers have brought their efforts to bear on the
specific problems of water resources and arid lands. But the solu-
tion of these problems can be accomplished only by a concerted
action of specialists in many fields. Geography, not only can
furnish the necessary background against which to plan such
integrated action, but has also accumulated valuable experience
in the unique function of bringing together, in a coherent and
organic body of knowledge, the facts imparted by specialists in
the most diverse fields of the social and the natural sciences.

Nore on Itiusrrations: With one exception, all photographs were
taken by the author in July, 1951. Figure 4 was taken in January, 1951,
by Roberto Maia, and is reproduced by courtesy of O Cruzeiro.

REFERENCES

1. Aguiar, Francisco Gongalves de. 1941. Acude Curema-Mae D’Agua,
Bol. da Inspectorta Federal de Obras Contra as Sécas 15, No. 1, p. 6.

2. Almeida, José Américo de. 1923. 4 Parahyba e seus Problemas,
Parahyba, Imprensa Official, p. 325.

3. Almeida, Jos¢ Américo de. 1954. Report read before the Chamber of
Deputies, Didrio do Congresso Nacional, (Secao I), (January 14),
p. 64.

4. Berredo, Vinictus. 1950. Obras Contra as Sécas, Rio de Janeiro,
Ministério da Viacao e Obras Publicas, Departamento Nacional de
Obras Contra as Sécas, p. 31.

5. Lisboa. Miguel Arrojado. 1913. O Problema das Sécas, Annaes da
Bibliotheca Nacional do Rio de Faneiro 30, 132.

6. Netto, José Oriano Menescal. 1942. Irrigagdo no Cariry Cearense,
paper presented to the Segunda Reunido Regional de Economia Ru-
ral, Fortaleza (January).

7. Russell, Richard J. 1932. Dry Climates of the United States: IT.
Frequency of Dry and Desert Years, 1g01-1920. Univ. Calif. Publ.
Geogr. 5, 245-74.

Russell, Richard J. 1934. Climatic Years, Geogr. Rev. 24, No. 1
g2-103.

* Conclusions and proposals from the December 1949 meeting of the
Commission appointed by the International Geographical Union, to
consider the institution of a world-wide survey of land use.

220

THE FUTURE OF ARID LANDS

Russell, Richard J. 1945. Climates of Texas, dun. Assoc. Am.
Geogr ao, wNOl 23752.

. Singer, Hans W. 1953. Reports Concerning the Economic Develop-

ment of the North-East, VIII: Economic Impressions from the
North-East (mimeographed). Rio de Janeiro, Banco Nacional de
Desenvolvimento Econémico, pp. 14-18.

. Sternberg, Hilgard O’Reilly. 1949. Floods and Landslides in the

Paraiba Valley, December 1948. Influence of Destructive Exploita-
tion of the Land. Compt. Rend. XVI Congr. Intern. Geogr., Lisbon.
pp- 335-64. Also in Conservation in the Americas, No. 8, April 1950,

p- 2-20. In Portuguese with English summary, in Rev. Brasil.
Geogr. Ml INow 211949), 224.261.

. Sternberg, Hilgard O’Reilly, 1951. Brasil Devastado. Bo/. Geogr.,

s
25 INO: UCS, AS=2O),

. Sternberg, Hilgard O’Reilly. 1951. Aspectos da Séca de 1951, no

Ceara, Rev. Brasil. Geogr., 13, No. 3, 327-69.

. Sternberg, Hilgard O’Reilly, 1953. 4 Paisagem WNordestina e o

Problema das Sécas, Rio de Janeiro, Escola Superior de Guerra.

. Vasconcelos, Sobrinho. 1949. ds Regides Naturais de Pernambuco,

0 Meio e a Civilizacgao, Rio de Janeiro, Livraria Freitas Bastos. S.A.,

p. 69.

. Vieira, Luis Augusto da Silva. 1934. Relatério dos Trabalhos Reali-

zados no Triénio 1931-~33?, Rio de Janeiro, Ministério da Viagao e
Obras Publicas, Inspetoria Federal de Obras Contra as Sécas, 1934,
Volk Types:

Agricultural Use of Water Under
Saline Conditions

L. A. RICHARDS

United States Salinity Laboratory, Riverside,
California

The question of whether irrigation agriculture can persist per-
manently has been included among other subjects for considera-
tion at these meetings. The principal difference between irrigated
and nonirrigated agriculture, when considered in relation to per-
sistence or permanence, arises from salinity. All waters used for
irrigation contain more or less soluble material. It is only a matter
of time until these solutes will accumulate in the root zone to the
extent that plants will not grow, unless some leaching occurs. This
generalization appears final and pessimistic, but, of course, it
would be difficult indeed to practice irrigation without some net
transfer of water downward through the root zone. Adequate
leaching often occurs on irrigated farms without thought or atten-
tion by the farmer. If this is not the case, salt will accumulate,
and eventually soil and water management operations will have to
be altered accordingly.

Water Conductance of Soils

Leaching is accomplished by the downward movement of water
through soil. To be adequately drained, any water table that tends
to form in irrigated land must be kept well below the root zone.
This requires that outlets be available for the ground water and
that water transmission in the subsoil be appreciable. Because of
its relation to water application methods and to leaching, the rate
at which a soil will transmit water is one of its most important

221

PPP) THE FUTURE OF ARID LANDS

physical properties as far as irrigation agriculture is concerned.
The hydraulic conductivity of soil is related to texture. It is
generally higher in coarser soils, but it is also influenced by struc-
ture and can be profoundly altered by soil management operations
and the exchangeable cation status.

Open ditches or covered drains can be used if necessary to
control the water table, but the cost of such facilities will be
prohibitive if the water conductance of the soil is too low. In
subhumid climates, a shallow, stable water table may be a great
economic asset for carrying crops through occasional drought pe-
riods but, for irrigation agriculture in an arid climate, a shallow
water table is a serious salinity hazard. Given an adequate water
supply and reasonably favorable soil texture, it might well be
argued that in arid climates the hazards from a water table near
the root zone considerably outweigh any advantages connected
with subirrigation.

The drainage requirements of any given irrigation project will
relate to the permissible depth and mode of variation of the water
table and the volume of water that the drains must convey in a
given time. These, of course, will depend on a whole complex of
factors related to soil, crop, climate, quality of irrigation water,
and farm management. In any case, the underground drainage
must be adequate to carry away the excess water and salt added
during irrigation if long-time profitable operation is to be attained.

Water Quality

The quality of irrigation water enters importantly in deter-
mining irrigation feasibility and permanence. There have been
no recent discoveries of new hazards to irrigation agriculture
among the soluble constituents of irrigation water. Some constitu-
ents like boron are toxic in minute concentrations. The main prob-
lems, however, appear yet to be the accumulation of soluble salts
and exchangeable sodium in soil.

The salinity of irrigation water has a direct bearing on such
factors as crop selection, method of application of the water, and
the leaching required to control salt accumulation in the soil. All
these in turn are subject to constraints imposed by drainage
conditions.

WATER UNDER SALINE CONDITIONS 223

Progress is being made on methods for measuring and expressing
the effect of saline conditions on plants. Reliable information is
accumulating on the salt tolerance of various crops. This informa-
tion is important as a guide for leaching operations and makes it
possible to continue agricultural production with salt tolerant
crops under conditions where it may not be feasible to maintain
low levels of soil salinity.

Application of Irrigation Water

The method of application of irrigation water is important. At
low salinity, furrow irrigation can be used. At higher levels of
salinity, the requirements for land preparation are more exacting.
The shape of the ridge and the position of the seed line with respect
to the water line in the furrow may require consideration. At high
salinity, flood irrigation must be practiced.

Excess soluble salts can be removed from soil only by the action
of water moving downward through the soil profile. Surface flush-
ing operations for salt removal are generally ineffective. Some-
times a trend toward salt accumulation can be reversed by minor
changes in the method of irrigation or the amount of irrigation
water applied. In extreme cases, ponding of water on the soil
surface tor long periods is required. In all cases, adequate drainage
is a controlling factor.

Principles governing a field leaching operation are simple and
straightforward. Soil tests can be used to indicate the need. The
process involves covering the soil surface until sufficient water has
passed through the soil to remove the soluble salt. The time re-
quired will depend on soil properties and drainage conditions. If
the exchangeable sodium is high, chemical amendments should be
applied. Soil tests are available for indicating the soluble calcium
required to reduce the exchangeable sodium to permissible levels.

Leaching Requirement

When the salinity of the irrigation water is high, leaching should
be accepted as one of the controlling factors in farm management
operations. Although soil salinity tests can be used as a guide, it is
helpful to have an estimate of the amount of water to be allocated
for leaching purposes. The leaching requirement is the fraction of

224 THE FUTURE OF ARID LANDS

the irrigation water that must be leached through the root zone to
control salinity at any specified level. It depends on the salinity
of the irrigation water and the salt tolerance of the crop. The
leaching requirement may be readily estimated from salt balance
considerations, and for steady state or long-time average condi-
tions it is equal to the electrical conductivity of the irrigation
water divided by the electrical conductivity of the soil solution at
the bottom of the root zone. The latter would be 8 millimhos per
centimeter for alfalfa, for example. If the conductivity of the
water is 1 millimho per centimeter, then the leaching requirement
would be one-eighth or 12.5% of the irrigation water. Hundreds
of thousands of acres have been irrigated in western United States
for more than ahalf century with water of this salinity and with no
apparent decline in productivity. However, in areas of very fine-
textured soil where water of this salinity was used, it was not
feasible to provide adequate drainage under the existing economic
conditions. Consequently, leaching was inadequate, the soil be-
came saline, and the farms were abandoned.

Sodium Hazard

Some irrigation waters cause exchangeable sodium to accumu-
late in the soil. This is usually due to a high proportion of sodium
to the other cations in the water, but other processes are involved.
The concentration of solutes that enter the soil in the irrigation
water is increased in the soil solution by evapotranspiration. Some
of the components such as the alkaline earth carbonates and cal-
cium sulfate may precipitate during this process, thus decreasing
the salinity hazard but increasing the sodium hazard. Research is
yielding a more complete understanding of the chemical and
physical processes taking place in the soil-water system. Recent
progress has been made toward finding a general relation between
the exchangeable sodium percentage and the relative activity of
sodium in the surrounding liquid phase or in a water extract of
soil. Classification schemes for water quality take these exchange-
able sodium hazards into account, but the prediction of future
soil amendment requirements from water analyses is somewhat
complicated. Further experience will be required before the rate

WATER UNDER SALINE CONDITIONS 225

of accumulation of exchangeable sodium under field conditions
can be reliably estimated from a chemical analysis of the irrigation
water. Nevertheless, much of the guesswork has been removed
from the agricultural use of water under saline conditions. Tests
are available for appraising the salinity and sodium status of soil
and for estimating the amount of amendment for reducing ex-
changeable sodium to an acceptable level. Progress is being made
toward putting this information to practical use by farmers.*

* Agricultural research dealing with salinity, exchangeable sodium,
drainage, leaching, salt tolerance of crops, and quality of irrigation
water has been conducted by numerous agencies in this country and
abroad. Information on these subjects has been summarized in the U.S.
Department of Agriculture Handbook No. 60 entitled “Diagnosis and
improvement of saline and alkali soils.’ Also see Circular, “Tests for
salinity and sodium status of soil and irrigation water.” These are ob-
tainable from the Superintendent of Documents, U. S. Government

Printing Office.

Consequences of Using Arid Lands
Beyond Their Capabilities

CYRIL LUKER

Soil Conservation Service, United States
Department of Agriculture, Washington, D. C.

“Using land within its capability” is a phrase that has come to
have special significance in the soil and water conservation work
of the United States Department of Agriculture. It means recog-
nizing the physical limitations affecting each acre of land and being
guided by those factors into the kind or kinds of use for which the
land is best suited, and under which it can produce most efficiently
on a sustained basis.

The consequences of using any land beyond its capability are
generally adverse. This is especially true of the arid lands. In arid
climates the balance between land, water, and people 1s more
critical than in any other agricultural province. When the balance
is upset the consequences are generally more severe than in humid
areas because recovery from the effects of overuse usually takes
longer in arid zones.

In the technical assistance program of the Soil Conservation
Service in the United States, soils are grouped into eight capability
classes for purposes of dealing with land-use problems. This pro-
vides an especially effective approach in arid and semi-arid areas
where land condition and production are so sensitive to land use.

These eight classes range from Class I, which is the best kind of
farm land to Class VIII, which is land with such severe limitations
that it cannot be used for forestry, grazing, or cultivated crop
production.

226

USING ARID LANDS BEYOND CAPABILITIES 227

The classes fall into two general groups—those suitable for
cultivation and those unsuitable for cultivation. Each class is
distinguishable from the others by the relation of definite physical
features to the intensity of use (cultivation, woodland, pasture,
range, and wild life) and the intensity of soil management required
for safe and sustained use. Classes I, II, and III are suitable for
general agricultural production of the region. Class IV should be
restricted to limited cultivation, and Classes V, VI, VII, and
VIII are suited for grazing, woodland, wild life, and recreational
purposes, in descending order of intensity of use.

Land Abandonment

Presentation of the remainder of this subject can best be done
by reference to research and experience in part of the southwestern
United States. Here are some classical examples of wide-scale land
misuse. In this general area one tragic consequence of land misuse
is land abandonment or nonuse. This denotes land use failure.
Land abandonment, of course, may be due to physical land failure,
or to financial or economic distress. In either case the basic prob-
lem is land use failure.

Although the two conditions do not divide sharply, it was found
in a recent survey and study of land abandonment and rehabilita-
tion in a section of this area that about 60% of land abandonment
was due to financial distress and about 40% to severe erosion.
The lands that were severely eroded were, in much of the area,
lands that were being used beyond their capability. Frequently,
the farmer continued to crop shallow, sloping lands in an effort to
obtain the occasional rewarding crop; but the usual results were
low production for a period and finally abandonment.

The principle of capability use is usually not violated on range
lands, and complete abandonment seldom takes place there. How-
ever, if grazing is heavy enough to prohibit maintenance of a
satisfactory condition of range plants for long periods, the result
is low return from livestock and the acceleration of soil erosion. In
the end it may mean permanent soil losses that so reduce produc-
tivity that the land might as well be abandoned.

228 THE FUTURE OF ARID LANDS

Erosion Damage

Studies in the southern Great Plains indicate that lands being
used beyond their capability deteriorate more rapidly than any
others.

Lands used beyond their capabilities are likely to be sandy or
shallow phase soils, have steeper slopes, and, in the western part
of the plains, have lower annual precipitation. Under such condi-
tions the production of vegetative cover for the land is frequently
not adequate to keep the soil in place and both wind and water
erosion take a high toll. Such damage is destructive to any soil,
but on shallow phase soil the damage is more likely to be irrepa-
rable than on the deeper better grades of land. And yet because in
wet cycles these lands often produce high yields, there is a strong
tendency to gamble on them each year. This results in their being
used beyond their capability.

In the southern Great Plains experience has proved that shal-
low, moderately sandy, and deep loose sand-hill soils of steep or
even gentle slopes cannot be kept permanently productive under
cultivation in the 14- to 20-inch rainfall belt. Shallow, fine-
textured soils under cultivation on both flat and sloping fields
have failed in all areas of less than 18 inches of rainfall.

On the other hand, level to nearly level medium depth, moder-
ately sandy soils have stood up well under cultivation with suitable
practices under rainfall as low as 16 inches.

Deep, nearly level, heavy soils have stood up well and can be
farmed with suitable precautions wherever found in the southern
Great Plains. This kind of soil does not occur, however, in large
amounts in areas of less than 16 inches of rainfall.

Research findings, which are verified by the experiences of many
farmers in this area, emphasize one of the most serious conse-
quences of using the land beyond its capability—permanent dam-
age to the land. Productive soil is the most important part of the
capital structure of farm businesses, and it cannot be replaced.
This capital loss showed up in a very pronounced way in lower
crop yields after the damaging soil blowing of the 1930’s.

Low plant vigor was generally evident in the variation of green
coloring of the plants from different fields. This variation empha-
sized the sharp variation in the amount of nitrogen available.

USING ARID LANDS BEYOND CAPABILITIES 229

This unnecessary fertility depletion is more serious in arid land
areas where we have not yet learned how to utilize commercial
fertilizers effectively, particularly in years of low rainfall. The
problem of rebuilding fertility in eroded and depleted soils is
further complicated by the limited number of adaptable legumes
suitable for inclusion in the cropping systems in arid areas.

Effects on Farm Credit

Soil erosion and other land damage has an adverse effect on
farm credit, both to individuals and large areas where land capa-
bilities are being ignored. Land farmed with regard to its limita-
tions is a better credit risk in every instance than land farmed
beyond its capabilities. Land used beyond its capabilities creates
difficulties in the establishment of necessary reserves farmers must
build up during favorable years to tide them over the lean years.
Feed reserves are recognized as especially essential to livestock
operators who must maintain a flexible position if they are to
avoid financial catastrophe.

Sound land use is also required if lending agencies are to provide
long range credit instead of the seasonal type which is usually
inadequate under such conditions. Increasing numbers of lending
institutions are recognizing these principles of conservation farm-
ing in providing sound loan services in the arid belts.

Economic and Social Implications

Just a few miles east of Albuquerque, New Mexico, in the 14-
inch precipitation belt, there are sizable acreages devoted to the
production of beans. Much of the soils there are shallow, steeply
sloping, low in organic matter, and are in small farms. Incidence
of crop failure is high, and alternative crop choices are very lim-
ited. These conditions combine to create a serious long-range
economic problem which will grow progressively more acute until
a range livestock economy is developed to replace the precarious
dryland bean farming. During the past few years many of the
occupants of these small farms have supplemented their incomes
through employment in near-by national defense projects, indus-
trial developments, and on large ranches and farms.

Population adjustments necessary for a livestock economy in

230 THE FUTURE OF ARID LANDS

such an area are never easy for people to make, although some
progress has been made in recent years toward the adjustment.
Some of the more important roadblocks in the way of making this
kind of adjustment at a faster rate include the need for resettle-
ment of people; the necessity for obtaining adequate credit for
enlarging and changing the farm business; acquiring needed educa-
tion in a new type of farming; and inevitable inertia and resistance
to change.

In any event, a change from cash crop farming, whenever it
exists in such a hazardous zone, to livestock grazing should be
speeded up by every practical and feasible means. Two principal
objectives will be accomplished: first, it will help stop further
irreparable damage to the land; and second, it will establish a
sound basis for a permanent agriculture. Only such an approach
will enable people to conduct a stable farm business. The alterna-
tive is not a pleasant one—further soil resource destruction,
blighted hopes and ambitions of people, economic and social stag-
nation, and finally abandonment of land to the forces of nature,
whose rehabilitation job is slower than we can afford.

Role of a Conservation Program

The use of appropriate conservation practices will slow down
the process of land damage, even when land is being used for
purposes for which it is not suited. However, the cost of applying
measures that will safeguard lands under such misuse is too high
and soon reaches the point of diminishing returns, especially when
an entire farm is being used beyond the capability of its soils. So,
in these cases, a conservation program is largely a delaying action
that buys time in which to get the land used in harmony with its
physical, climatic, and economic limitations. In these cases where
suitable conservation measures are not employed, the rapid dete-
rioration of the land makes ultimate solution much more difficult,
or even virtually impossible.

Wherever soils are suitable, and adequate water supplies are
available, irrigation is an effective means of stabilizing farming in
arid areas. Many examples in the Southwest prove the value of
this type of land use. Some caution is required, of course, espe-

USING ARID LANDS BEYOND CAPABILITIES 23)

cially where underground water is being used. In many areas these
supplies are already being depleted faster than they are being
replenished.

Without irrigation, however, successful farming in the arid and
semi-arid lands of the southwestern United States requires mainly
that the farmers accept highly variable and fluctuating rainfall as
a normal phenonemon. Once they understand this as a cropping
limitation, they can develop their farm enterprises in harmony
with the pattern. This means gearing the farm program to the dry
phase of the weather cycle rather than to the wet periods, as so
many have done in the past. The first course is an approach to
successful farming, because it recognizes the capability of the land.
Any other course courts almost certain failure.

The attainment of these objectives over a significantly wide area
of the arid lands requires exercise of local leadership that will move
people to act as rapidly as facilities and resources will permit.
Farmers throughout the United States now have suitable admin-
istrative machinery for this in the form of locally organized soil
conservation districts. Soil conservation districts now cover most
of the agricultural lands of the nation, and include within their
boundaries about 80% of these arid and semi-arid lands in the
southwestern United States. Districts are regarded as highly effec-
tive mechanisms for solving soil and water problems in an arid
region as well as elsewhere in the nation.

Needed Research

Information gained from further research should prove helpful
in keeping arid land use within the limits of capability. Such
additional investigation might well be conducted along the follow-
ing lines:

1. Secure information that will make it possible to define more
sharply the differences between the characteristics of arid land
suitable for cultivated crop production and those unsuited for such
use.

2. Develop more adaptable grasses and legumes for these areas.

3. Work further on the possible uses of commercial fertilizers
for such lands.

232 THE FUTURE OF ARID LANDS

4. Determine the most desirable size of farm and ranch units
from an economic standpoint.

Summary

The basic physical soil conservation objective of the U. S.
Department of Agriculture, ‘““The use of each acre of agricultural
land within its capabilities and the treatment of each acre of
agricultural land in accordance with its needs for protection and
improvement,” is the best approach to solution of the complex
land problems we confront in the arid areas of the United States
and elsewhere. Sustained production and permanent occupancy
of these lands cannot be achieved unless agriculture of the arid
areas is based on such a concept.

Possibilities of Increasing and Main-
taining Production from Grass and
Forest Lands without Accelerating
Erosion

RAYMOND PRICE

Rocky Mountain Forest and Range Experiment
Station, United States Department of Agricul-
ture, Fort Collins, Colorado

Use of grass and forest in arid lands for sustained production
without accelerating erosion requires great managerial skill. A
delicate adjustment of vegetation to environment exists. In this
adjustment the balance between constructive and destructive
forces is sensitive and easily shifted toward destruction.

The climate consists of extremes with great fluctuations from
season to season and year to year. The adverse climatic factors
are: the wide range of temperature and precipitation in a given
place; the great variance in seasonal and annual precipitation;
and the occurrence of extremely intense storms after long dry
periods. Climatic variation produces much more extreme differ-
ences in plant growth and erosion hazards in arid than in humid
regions. Because of the delicate adjustment of vegetation to en-
vironment, a succession of dry years can weaken vegetation so
that heavy, or even moderate, rains can cause severe erosion.

Many mistakes have been made because of wishful, optimistic
estimates of climatic changes. The actual shifts of an erratic cli-
mate in arid lands must be accepted as they come. We must learn

233

234 THE FUTURE OF ARID LANDS

to take advantage of favorable periods and to guard against the
unfavorable weather events which are bound to happen. Perhaps
the old frontier rule of preparing for the worst while hoping for
the best is still a good guide for arid land residents.

In the southwestern United States there are many examples of
the improper use of arid lands. Present day problems in the Rio
Grande Valley are to a large extent the result of poor land use in
the past. Historical evidence clearly shows that a general decline
of watershed lands and resources began during the 1880’s. This
decline was largely brought on by damage to the natural plant
cover through poor grazing management, promiscuous wagon
trailing, and injudicious dry Penta hs evidences of decline
are: the decreased grazing capacity of range lands; deep and con-
tinuous gullies slashed through alluvial valleys; shifting sand and
sand dunes; and the silting up of river channels and storage
reservoirs. Much of the water yield is now from surface runoff and
is silt laden. Water once held in the soil to support plant growth
is now drained out by deep gully channels. Irrigation diversions
on tributaries have been undercut and destroyed. The Rio Puerco
is a terrifying example of the destruction caused by accelerated
erosion. The lower valley of the Rio Puerco between San Luis and
Cabezon and in the vicinity of Salazar once supported a prosper-

HUMID ARID

SOIL

Figure 1. Diagrammatic illustration emphasizing that.a more deli-
cate adjustment of vegetation, rainfall, and soil exists in arid lands than
is common in humid lands. This vegetation-environment relationship
must be recognized in the sustained use of grass and forest lands in arid
regions.

INCREASING PRODUCTION WITHOUT ACCELERATING EROSION 235

ous farming and livestock enterprise as shown by ruins of once
substantial ranches and homes. Now the valley is gashed by enor-
mous straight-walled channels 20 to 40 feet deep.

Lower down on the Rio Grande, the river has not been able to
carry the enormous load of sediment dumped into it by its muddy
tributaries. As a result, adjacent valley lands have become water-
logged and are no longer cultivated. The raising river bed also
brings an increased flood threat to bordering cities and farms.
Sand and mud bars furnish a foothold for salt cedar (Tamarix
spp.) and other water-wasting vegetation.

Improvement of the land and water situation of the Rio Grande
Valley and other similar arid land is not impossible. Restoration
of plant cover on the watersheds is a necessary step in this im-
provement. Unfortunately, many are puzzled when the need for
cover improvement and erosion control is emphasized. Often dete-
riorated conditions are accepted as normal. The great difference
between present day vegetation and that of an earlier day is not
fully realized. Demonstrations provided by some ranch lands,
national forests, and other well-managed areas show what can be
and has been accomplished. Research and experience have shown
many ways to maintain and improve production on grass and
forest lands.

Management of Forage Resources Adjusted to Rainfall Deficiency

A perpetual problem in range land management is how to bal-
ance animal numbers with forage supplies. This is troublesome
enough anywhere but particularly difficult in dry lands, which are
almost universally characterized by large fluctuations in precipi-
tation between seasons and from year to year. Man, misled by
optimism or driven by short-range necessity, has all too often
overused the ranges. Recovery proceeds more slowly and less cer-
tainly than in humid lands. If deterioration is severe, unfavorable
conditions may persist in the absence of any use unless the natural
forces tending toward recovery are given assistance.

Average annual rainfall in arid lands is of little value in land
management. Extremes of rainfall occur more frequently than do
mean conditions. In southeastern Arizona and southwestern New
Mexico of the United States, rainfall varies exceedingly from one

236 THE FUTURE OF ARID LANDS

year to the next, but periods of continuous excess or deficiency
may persist for 4 or § years. Grazing management systems must
be designed to meet the large changes in forage production that
result from variations in rainfall (8, 11).

Stocking Based on Below-Average Forage Production

The wide variation in rainfall and the consequent variability in
annual forage production makes proper stocking essential. At-
tempts to adjust stocking each year to the highly variable amount
of forage, unless properly done, may have disastrous results. A
breeding herd built up to use the forage crop produced in good
years cannot be maintained in subsequent dry years. The practical
way to utilize the highly variable forage crop is to stock conserv-
atively.

A margin for safety is proy.ded by stocking at a level less than
that which the average forage production will support. Ungrazed
forage produced in favorable years provides a forage reserve for
lean years and also protects the soil. In southern New Mexico,
stocking at 65% of average forage production is recommended
for black grama (Bouteloua eriopoda (Torr.) Torr.) and tobosa
grass (Hilaria mutica (Buckl.) Benth.) ranges (g). Even at this
seemingly conservative rate of stocking, severe reductions of live-
stock numbers would be needed during prolonged periods of
drought such as have occurred three times during the last four
decades. Obviously, the desirable levels of stocking will vary widely
with climatic regions and forage species, but stocking somewhat
below the average forage crop should be practiced on arid ranges.
Moreover, the basic breeding herd should be kept low—s5 to
60% of the total number grazed. The remainder of the herd
should be made up of younger, more salable animals that can be
disposed of quickly when dry periods and short forage supplies
occur. This practice will safeguard the most valuable breeding

herds (3).

Seasonal Grazing for More Forage

Yearlong grazing is the most punishing form of use. If contin-
ued, this practice reduces the proportion of desirable forage species
and the amount of usable forage; it will also result in thinning of

INCREASING PRODUCTION WITHOUT ACCELERATING EROSION 237

the plant cover and cause accelerated soil erosion. Fencing or herd-
ing so that grazing use is rotated between areas increases produc-
tion and also provides more vegetation for soil protection. Allow-
ing part of the range to rest during the growing season every
second or third year results in a wider variety of the better forage
grasses as well as greater forage production.

In many cases, it is possible to take advantage of differences in
the growth and palatability of forage species. In southern New
Mexico, when black grama is grazed only during the non-growing
season, almost twice as much herbage is produced as under year-
long use. Under yearlong use the valuable black grama is replaced
by worthless snakeweed (Gufierrezia spp.). It 1s possible to with-
hold black grama areas during the growing season by grazing
cattle on tobosa grass flats, which provide abundant palatable
forage during the summer. Tobosa grass is not harmed by judi-
cious summer grazing and is not palatable to cattle during the
winter months. This pattern of seasonal use increases range pro-
ductivity above that possible under yearlong use of either grass

eyes (hs)
Proper Distribution of Livestock

Obviously, wells and stock tanks for water should be provided
wherever possible so that animals can make effective use of the
range. Well-spaced watering places aid in (1) distributing animals
over the range, (2) preventing local overgrazing, and (3) keeping
livestock in better condition. The number of cattle watered at a
place should not exceed the grazing capacity of the area within a
114- or 2-mile radius of the watering point.

Judicious placing of salt grounds aids in drawing cattle to other-
wise little-used parts of the range (12). The use of a mixture of
cottonseed meal and salt as a self-feeding supplement promises to
be an even more effective method of obtaining better distribution
and thereby avoiding overused areas (2).

Control of Noxious Plants

The invasion of grassland by worthless or undesirable plants
has resulted in a critical problem throughout much of the south-
western United States (18). Increases in low-value shrubs on

238 THE FUTURE OF ARID LANDS

former grassland sites are responsible for lowering grazing capacity,
especially where grazing pressure has been heavy. Forage produc-
tion can be increased by removing these shrubs. For example, in
southern Arizona removal of mesquite (Prosopis juliflora var.
velutina (Woot.) Sarg.) on grassland sites when accompanied by
conservative grazing has increased perennial grass forage produc-
tion fourfold. There are other important benefits from mesquite
control. Erosion ordinarily active under mesquite is reduced by
the reestablishment of grasses. Cattle are also more easily worked
and losses from screwworm are reduced (10).

Considerable research is being done to find control methods for
many other undesirable plants. Much progress has been made. It
must always be kept in mind, however, that there is no perma-
nent advantage in the mere killing of low-value plants unless
their elimination is followed. by grazing practices which permit
increases in desirable forage species.

Reseeding of Adaptable Grasses

Heavy grazing, drought, the spread of noxious plants, and the
combination of these factors have depleted the grasses and other
forage plants on large areas. Restoration of much of this land
within a reasonable length of time depends on the development
of practical economic methods for range reseeding. To date, suc-
cessful methods have been worked out for reseeding many arid
sites. However, much remains to be done, and an expanded re-
search program is justified.

Most success with range reseeding on arid sites has been with
the use of introduced grasses. Several species of lovegrasses
(Eragrostis spp.) introduced from South Africa have been used
successfully in Arizona and New Mexico. Mild winters, hot sum-
mers, and annual precipitation of at least 11 inches, mostly
received in the growing season, characterize these areas (1, 13).
Grasses from Asia and the Middle East such as crested (/gro-
pyron desertorum (Fisch.) Schult.) and intermediate (4. inter-
medium (Host.) Beauv.) wheatgrasses and Russian wildrye
(Elymus junceus Fisch.) have proved well adapted to areas where
winter temperatures are lower and more of the precipitation occurs

INCREASING PRODUCTION WITHOUT ACCELERATING EROSION 239

in the winter. A few native gramas (Bouteloua spp.) and muhlen-
bergias (Muhlenbergia spp.) and other species have given con-
siderable promise but have not been so widely used as the intro-
duced species (6, 14).

Through use of adapted introduced grasses, range reseeding has
often increased forage production ten- to twentyfold. With the
realization that the great advance in knowledge and widespread
use of reseeding on arid lands in the United States is largely the
result of research during the last 20 years, considerable optimism
is justified.

Better Management of Shrub and Woodland Areas

Shrubs and trees make up the cover on large areas of the south-
western United States. Even the true desert is not really treeless
but contains here and there scattered along the drainages and
rocky slopes dwarf trees and treelike plants such as yuccas (Yucca
spp.), paloverdes (Cercidium spp.) and cacti. On foothills and
mesas are broad zones of orchardlike woodlands of pinyons
(Pinus edulis Engelm.), junipers (Funiperus spp.) and evergreen
oaks (Quercus spp.) often intermingled with chaparral made up of
shrubby species such as manzanitas (4rctostaphy/los spp.), sumacs
(Rhus spp.), shrub live oaks (Quercus spp.), and ceanothuses
(Ceanothus spp.).

Although the desert trees and shrubs furnished food, medicines,
fiber, fuel, and other necessities to the early inhabitants, they are
of little economic use now. The development of uses for these
unique plants would increase productivity of the desert lands with
little risk of increasing erosion. Very likely the increased returns
would make it possible to provide greater protection from floods
and erosion.

A similar opportunity to find or develop merchantable products
is present in the chaparral and woodland types. The present think-
ing is that portions of these types should be replaced with grasses
wherever soil and slope conditions make tree removal and grass
establishment practical. Present research is concerned mostly
with methods of killing or removing chaparral and woodland
species. Markets for the woody material removed are needed to

240 THE FUTURE OF ARID LANDS

offset the cost, and some research is being done to develop products
such as pulp for paper, chemical extractives, and others. For the
comparatively level lands where shrub and tree control is now
practiced, the change to grasses is desirable for both forage pro-
duction and erosion control provided that grazing is carefully
managed. In the shrub and woodland zone, erosion is often active
because of insufficient effective grass cover to protect the soil.
Unless the increased growth of trees and shrubs is reduced, grass
cover cannot be increased because the roots of woody plants domi-
nate even the open areas and extract available soil moisture. The
necessity for proper control of grazing after brush is removed
must be emphasized.

Management of Higher Forested Areas in Arid Regions

At higher elevations, the mountain ranges and plateaus are
covered with valuable forests of tall pines (Pinus spp.), spruces
(Picea spp.), and firs (4dzes spp.).

Thinning and improvement cutting in the ponderosa pine,
(Pinus ponderosa Laws.) and spruce-fir forests do not present
erosion hazards in themselves. The danger of erosion comes from
the skid trails and logging roads built and used to remove the prod-
ucts. In thinning, the general recommendation is to remove the
slowest growing trees to concentrate the capacity of the site upon
the most vigorous trees. When stands are mature or overmature,
as in much of the spruce-fir zone, harvest cuttings are made to
replace the slow-growing old trees with young fast-growing
trees (5).

To prevent excessive erosion, roads must be carefully located
to avoid steep pitches and soils that erode easily. Drainage must
be provided to prevent water from concentrating on the road sur-
face and in side ditches. Cuts, and particularly fill banks, must be
protected. Much of the high country is too steep to be logged by
methods now in common use without excessive erosion. For these
locations, cable logging, chutes, or other methods are needed for
transporting logs with less road building (16).

Improvements in utilization and the development of new mar-
kets for wood products will make possible better management to

INCREASING PRODUCTION WITHOUT ACCELERATING EROSION 24]

UPPER ANCHOR

¥
WinCH” USN N
AND MOTOR Po
ak
CoE ND RID LINE
id as:
= ¥ +h nese

SON
ayue dees <I VLINE STOP

SPAR TREE

INTERMEDIATE
SUPPORT

i Ne Ss

Ga a, meaty
~ Vga) Ne

Ze
——S

DN)

Figure 2. A cable logging system that appears satisfactory to harvest
timber on steep watershed lands with a minimum of erosion and other
destruction to the watershed.

improve productivity and also to prevent watershed damage.
Destructive exploitation is always tempting when risks are high
and the profit margin low. Sustained, forward-looking operations
are encouraged when products removed are suited to a diversity
of uses and when reasonable profit can be obtained without
excessive risks.

Effects of Land Management on Water Supplies

Comparatively little streamflow is yielded by the low-lying
grasslands and woodlands where most rains are absorbed by the
dry soil. Although flash floods follow cloudburst storms, often such

242 THE FUTURE OF ARID LANDS

flows are heavily laden with mud. The aim of management in such
low-lying areas is to reduce sediment movement. Most of the large
water reservoirs lie within this zone and catch the debris dumped
into the the tributary streams. Although the scanty rainfall sup-
plies plants, they are not competing with man for their supply
but rather are racing evaporative forces to take water from the
soil before it returns to vapor. These plants provide soi! protec-
tion, forage, and other products from soil moisture that would
otherwise evaporate. Phreatophytes are an exception. These
water-loving plants are not dependent on local rainfall but grow
along streams and rivers where they steal water that could be put
to other uses. Our justified fight against these water thieves must
not be carried to the upland vegetation (4).

The permanent rivers, on which the Southwest depends for its
water supply, tap the higher elevations where precipitation is
more abundant and water to feed streams is in surplus, at least
at certain seasons. Just as on lower lands, the protective effect of
vegetation is important. Where plant cover is reduced, the capac-
ity of the soil to absorb and store water is diminished. Flash floods
and mud flows are all too common after excessive grazing use,
careless logging, or wild fire. Where the protective cover remains,
streams run clear and contribute little sediment to aggravate
downstream troubles. The good quality of mountain waters is
appreciated by irrigators as well as by domestic and industrial
users.

In some situations, mountain lands can be managed to increase
water production. Enough research has been done to show the
possibilities of increasing streamflow, but not enough has been
done to make general recommendations. Where winter snow is
heavy, dense stands of evergreen trees hold snow on their foliage
and let it evaporate without reaching the ground. In snow coun-
try, thinnings in young forest, selection cutting, and patch clear
cuttings to open up areas of dense forest have been shown to in-
crease the amount of snow finally contributing to streamflow.
These measures are good forest management practices and can
be carried out so that both wood and water production are bene-
fited. Another possibility is the replacement of deep-rooted species
of trees with shallow-rooted vegetation. This would reduce the

INCREASING PRODUCTION WITHOUT ACCELERATING EROSION 243

fall soil moisture deficit on areas where soils are deep and less
water would be required to prime the soils before runoff
occurred (17).

As a general rule, the more arid the climate the less the oppor-
tunity to increase water yield by thinning or removing vegetation.
With the exception of plants growing along watercourses, the
water used by vegetation comes from soil moisture that can be
removed from the soil only by plant roots or by evaporation. In
dry climates, this moisture is lost from the soil with or without
plants. Changes in vegetation are expected to provide more water
where precipitation is mainly in the form of snow, or where it falls
in large enough quantities to saturate the soil at frequent
intervals (7).

Growing Demands; Limited Resources

Demands upon arid lands in the southwestern United States
have increased and will continue to increase. Recently, there has
been a decline of subsistence farming and ranching and a tre-
mendous growth of urban populations. The concentration of popu-
lation increases the strain on water resources because no one
locality has sufficient water naturally available to it to support its
present size and future hopes. Water is no doubt a resource that
will continue to be limiting. Managers of forest and range lands
have an obligation to protect water supplies. Fortunately, prac-
tices that maintain and improve forage and forest production are
also beneficial in controlling erosion and reducing sedimentation.
A first need on much of our watershed lands in the arid Southwest
is repair and rehabilitation. Past damage has been severe, but de-
terioration need not continue. Improvement has already been
made in many areas. Human needs and demands can be expected
to fluctuate and vary from one product to another in the course
of time. We must safeguard all arid land resources so that future
needs can be satisfied.

REFERENCES

1. Anderson, Darwin, Louis P. Hamilton, Hudson G. Reynolds, and
Robert R. Humphrey. 1953. Reseeding desert grassland ranges in
southern Arizona. Ariz. Agr. Expt. Sta. Bull. 240.

244

103},

T4.

5
16.

18.

THE FUTURE OF ARID LANDS

. Ares, Fred N. 1953. Better cattle distribution through the use of

meal-salt mix. F. Range Mangt. 6, 341-346.

. Ares, Fred N. 1952. Size and composition of the herd. 4m. Cattle

Producer. December, pp. 14-18.

. Fletcher, H. C., and L. R. Rich. 1955. Classifying southwestern

watersheds on the basis of water yields. 7. Forestry 58, 196-202.

. Gaines, Edward M., and E. S. Kotok. 1954. Thinning ponderosa

pine in the Southwest. Rocky Mountain Forest and Range Exp. Sta.
Paper No. 17.

. Hull, A. C., Jr., and W. M. Johnson. 1955. Range seeding in the

ponderosa pine zone in Colorado. U. S. Dept. Agr. Circ. 953.

. Martin, W. P., and L. R. Rich. 1948. Preliminary hydrologic re-

sults, 1935-48, “‘base rock” undisturbed soil lysimeters in the grass-

land, Arizona. Soi/ Sct. Soc. dm. Proc. 18, 561~567.

. Nelson, Enoch W. 1934. The influence of precipitation and grazing

on black grama range. U. S. Dept. Agr. Tech. Bull. 4o9.

. Parker, K. W. 1944. Adjust range stocking for heavier animals.

hes Gattleman 3, (3) 17 aise 385 39:

. Parker, K. W., and S. Clark Martin. 1952. The mesquite problem

on southern Arizona ranges. U. S. Dept. Agr. Circ. 908.

. Reynolds, Hudson G. 1954. Meeting drought on southern Arizona

rangelands. 7. Range Mangt. T, (1), 33-40.

. Reynolds, Hudson G. 1955. Do you use fully the levers at hand to

obtain best livestock distribution? Western Livestock F. Yearbook
33, 38-39, 109-111.

Reynolds, Hudson G., F. Lavin and H. W. Springfield. 1949. Pre-
liminary guide for range reseeding in Arizona and New Mexico.
Southwestern Forest and Range Expt. Sta. Research Rept. No. 7.
Reynolds, Hudson G., and H. W. Springfield. 1953. Reseeding
southwestern rangelands with crested wheatgrass. U. S. Dept. Agr.
Farmers’ Bull. 2056.

Southwestern Forest and Range Experiment Station. 1951. The
Fornada Experimental Range. (Processed.)

United States Forest Service, California Region. 1954. A guide to
erosion reduction on National Forest timber sale areas. (Processed.)
Wilm, H. G., and E. G. Dunford. 1948. Effect of timber cutting on
water available for stream flow from a lodgepole pine forest. U. S.
Dept. Agr. Tech. Bull. 968.

Young, Vernon A., Frank R. Anderwald, and Wayne G. McCully.
1948. Brush problems on Texas ranges. Texas Agr. Expt. Sia. Misc.
Pub. 21.

Land Reclamation and Soil
Conservation in Indian America

PEDRO ARMILLAS

Instituto Indigenista Interamericano, Mexico

Since I am an archaeologist who specialized in the study of pre-
Columbian America, I want to bring to the attention of the
specialists in other fields of knowledge a summary of data on the
relationships between man and land in arid and semi-arid zones
of America in the past. This summary will provide illustrative
examples of the kind of information which archaeology can provide
on this subject.

The American Indians are classified, from the viewpoint of the
relationships between man and natural resources, as follows: (a)
pre-agriculturists, getting their livelihood by gathering of plants,
hunting and fishing; (4) proto-agriculturists, farmers using the
shifting-field system; and (c) intensive agriculturists.

The pastoral way of life which depends on the raising of grazing
animals and which might have notoriously harmful effects on soil
conservation, was not found in pre-Columbian America.

Although the activities of the human groups belonging to the
first two categories may affect natural resources, their low density
of population tends to make the damage done, if any, less serious
than that resulting from the action of the large, concentrated
populations of intensive agriculturists.

Intensive Agriculture in Indian America

At the time of Columbus the intensive agriculturists extended,
although not in continuous distribution, from the American South-

245

246 THE FUTURE OF ARID LANDS

west to northwestern Argentina. This zone included the areas of
native civilizations of Mexico-Central America (called Meso-
america by the anthropologists) and Peru-Bolivia (called 4ndean)
as cultural climaxes. Because of limitations of space, and also
because of the reason noted above, I shall concentrate my atten-
tion on these two areas.

Mesoamerica

Mesoamerica includes central and southern Mexico, Guatemala,
British Honduras, E] Salvador, and western Honduras. Owing to
a diversity of factors, the most important of which are great differ-
ences in altitude and broken relief, it is far from being a geographi-
cal unit. In the lowlands, which include the coasts and some of the
interior depressions, tropical rain forests and monsoon forests,
tropical savannas, and tracts of hot steppe and cactus desert are
found. In the mountains and plateaus there are temperate forests,
mesothermal savannas and cold steppes.

In the semi-arid zones, constituting most of the area, the degree
of aridity depends on the duration of the winter dry season and the
intensity, reliability, and efficiency of the summer rains. Also, the
regularity of the beginning of the rains—as a determinant of the
planting period—is an important factor in the highlands, where
early frost might destroy the unripe crops when planted late.

In the rain-deficient sections and in many of those having a
markedly seasonal distribution of rains, canal irrigation or other
specialized forms of horticulture were important, although not
exclusive of rain farming, in pre-Columbian times. The distribu-
tion of irrigated lands included the highlands, the intermountain
depressions, the Pacific coast and the semi-arid zone of central
Veracruz. Hydraulic engineering was most developed in the valley
of Mexico. Elsewhere the irrigation works were built on a small
scale, possibly with the resources of single local communities, but
requiring in many basins centralized control for the allocation of
water rights.

The other specialized forms of horticulture to which I referred
were:

(a) Gardens irrigated with water manually elevated from wells,
dug within these plots.

INDIAN AMERICA 247

(6) Chinampas. These so-called floating gardens are actually
artificial islands built near the shores of shallow lakes by piling up
layers of aquatic plants and silt from the bottom of the lakes. The
porous soil of the chinampa is perpetually moist as a result of the
infiltration from the surrounding waters. Additional moisture is
applied directly to the individual plants by lifting water manually
from the canals surrounding the plot, by means of long-handled
cloth buckets. As the water is muddy, this amounts to the addition
of new soil. Furthermore, aquatic plants and other fertilizers are
used.

The Indian methods of land reclamation and soil conservation
also included various types of adjustments to farming on slopes,
these are:

(a) Rock walls built on the contour to form level benches irri-
gated with elaborate aqueducts, as were the gardens of the Tex-
cocan king Nezahualcoyotl (fifteenth century). Actually, the
royal gardens were only a part of an extensive project of land
reclamation through terracing and irrigation in the foothills of
the mountains to the east of Texcoco. There is an early colonial
document concerning the water rights of the Indian farmers culti-
vating the reclaimed lands. Later, the upper aqueducts and some
of the terraces were abandoned and erosion carried away the soil
on the top of the hills. However, the middle and lower level ter-
races are still well kept and cultivated.

(6) Rain farming terraces. (1) With retention walls—the distri-
bution of these extended from the valley of Mexico to British
Honduras. Most of them are now abandoned; (2) without reten-
tion walls but hedged with maguey—these are built to reduce slope
gradients, to retain soil and moisture. This system is widely
practiced at present in Central Mexico.

Although fairly widespread, the techniques of soil conservation
were by no means universal in pre-Columbian Mexico. There is
no doubt that in many districts the exploitation of land might
have resulted in serious depletion and destruction of soils, but the
wholesale damage seems to have been a result of the technological
changes introduced with the Spanish Conquest. In the first place,
large-scale mining operations demanded huge quantities of wood
for timber and fuel, thus bringing about the destruction of forests,

248 THE FUTURE OF ARID LANDS

to a greater extent than before. Other causes of the destruction of
soil were the introduction of herds of grazing animals and the
change from intensive cultivation to extensive cultivation with
the introduction of the plow. Also, the transfer of manpower from
agriculture to other activities and the decrease in population re-
sulting from the epidemics introduced by the Spaniards brought
about the abandonment of pre-Spanish irrigation systems.

These technological changes, however, were not always detri-
mental. The introduction of the plow, which, as previously noted,
was conducive to the destruction of soil in some areas, permitted
the opening up of new lands to cultivation, in regions of heavy soils
not cultivatable with the native digging stick. The introduction
of new plants also extended cultivation beyond the limits of the
pre-Columbian crops.

Andean America

The Peruvian coast and the Peru-Bolivia highlands comprise
the central Andean area. The coast is a desert interrupted by the
valleys of the rivers flowing from the Andes to the Pacific. In
one of these cases, the Viru Valley, archaeologists have found evi-
dence of occupation by sedentary farmers going back to the third
millenium B.c., some 4,500 years ago. At first, farming probably
depended on flood waters, but since the first millenium B.c. the
agricultural potentialities of the valley soils were fully exploited
by means of canal irrigation. More than 2,000 years ago the irriga-
tion system had snondet to its maximum potentiality, permitting
the cultivation of 40% more land than at the present time (24,200
acres against 17,300 in 1946). There is some evidence of population
decline in the north coast of Peru during the late pre-Spanish pe-
riods (after A.D. 1200), but the possible causes are not yet known.
They may have been socio-political in nature or may have in-
volved the diminishing efficiency of the irrigation systems, or both.

In the valleys of the south coast, the flow of water available
for irrigation is smaller than in those of the north coast, and there
are reasons to assume that it was the same in the past. Archaeology
shows that the density and concentration of population, the de-
gree of urbanization, and the complexity of the socio-political

INDIAN AMERICA 249

structures were directly proportional to the respective water
resources of each area.

In the semi-arid and broken highlands, the Indians built huge
systems of irrigated farming terraces. The beginning of terracing
for agriculture may go back to before the time of Christ, but it
seems that its maximum expansion took place in Inca times. After
the Spanish conquest the terraces were neglected and most of them
fell into ruin.

Applications of Archaeology

Dr. Whyte called attention to the need to ‘‘map the cultivated
lands in relation to their farming systems and the crops grown
thereon” (p. 182). I suggest that this study must be made within
the historical context. In many zones the knowledge of the past,
the local story of trial and error in land management, might help
in planning for the future.

Dr. McClellan asks for sociological and economic research keep-
ing pace with engineering and scientific research (p. 198). Anthro-
pologists have explored the relationships between technical change
and society. Man is a social animal and the alteration of the bal-
ance between technology and the societal forms, through induced
changes of the first, tends to start chain reactions which are as far
reaching in their effects as those resulting from the alteration of
the ecological balance in nature. This is to be kept in mind by
those who have the responsibility of planning the future of arid
and semi-arid lands and of the people living on them.

Furthermore, there is the problem of acceptance or rejection.
In underdeveloped areas where people live in traditional ways it
is absolutely necessary to bear in mind the cultural situations in
any project of improvement. Only under that condition will it be
possible to guarantee the social acceptance of the technical im-
provements which have to be introduced. It 1s a principle well
established in cultural anthropology that whatever may be the
technica] value of programs of aid, these may run the risk of failure
if they do not take into account the cultural conditions.

Better Use of Present Resources:
Concluding Remarks

KANWAR SAIN

Central Water and Power Commission, Ministry
of Irrigation and Power, New Delhi, India

Through countless centuries there has been built up a balanced
relationship among water, land, and the vegetative cover. Each
dependent on and helpful to the others, they have developed
together, through physical, chemical, and biological processes, to
create and maintain useful and abundant resources for the habita-
tion and sustenance of man. But man has not used the resources
properly; in his ruthless exploitation of land and water resources,
man has violated the basic arrangements in a manner that upsets
the fruitful balance which created and maintained the land and
water resources.

No doubt a great part of the arid and semi-arid regions are
man-made in so far as the vegetation has been cut down or burnt
for purposes of cultivation, or else has been used for grazing
purposes. Though the original cause of the Rajputana desert in
India can be traced to geological events, its further deterioration
can definitely be attributed largely to human causes. The forests
were utilized beyond their natural recuperative powers. The
vegetation, wherever it existed, became a grazing ground. All
these resulted in accentuating the formation of desert character-
istics.

Permanency of Irrigation

Richards comes to the conclusion that, ‘‘All waters used for
irrigation contain more or less soluble material. It is only a matter

250

CONCLUDING REMARKS 251

of time until these solutes will accumulate in the root zone to the
extent that plants will not grow, unless some leaching occurs.”

I am afraid I find it difficult to accept this pessimistic generali-
zation in its entirety. I can quote cases from India and China,
where land has been under irrigated agriculture for more than a
thousand years. Richards’ remarks would of course apply, when
either the water contains salts, or the salts exist in the soil profile,
which are brought to the surface during the process of repeated
irrigation and evaporation. This situation has been successfully
handled in India by crop rotation and leaching the soil, as Richards
suggests.

International Cooperation in Reclamation

Whyte has outlined the activities of the FAO who are working
on all promising methods for the reclamation of arid regions.

In India an ad hoc committee was appointed by the Government
to recommend measures for the immobilization of the Rajputana
desert. The recommendations included a Desert Afforestation Re-
search Station to study the silviculture of the various species
already growing in the desert with particular reference to their
succession, to consider the possibility of introducing exotic desert
species from other countries and from other parts of India, to
maintain a number of seed stores for distribution through de-
partmental agencies, and to propagate vegetation. In addition,
they suggested creation of a five-mile wide forest belt several
hundred miles long to withstand the onslaught of winds. The
committee also recommended the establishment of nurseries for
experiments with and distribution of suitable species of plants,
shelter belts transverse to the direction of winds, increase in the
proportion of area under forest, and adoption of improved agri-
cultural practices by the cultivators. A Desert Research Station
has already been set up at Jodhpur and work is in progress on
the best method of afforestation and creation of oases for the
spread of vegetation.

In 1954 the World Forestry Congress met in India to discuss
the place of forests in the land economy of the country. The
Congress decided to recommend to the 47 governments the cre-
ation of an international commission on desert control and afforest-

252 THE FUTURE OF ARID LANDS

ation of arid zones. This new organization, it was proposed, would
study problems affecting the dry areas and serve as a center for
the collection and dissemination of information to all FAO
members and other interested countries. I believe the FAO has
already undertaken to collect results of the experiments and
observations at present available from several countries and to
analyze such results in order to advise other countries in matters
of land utilization, conservation of forest, and afforestation poli-
cies. | have no doubt that these developments will make a great
advance in the direction of a methodical and scientific reclamation
of arid and semi-arid lands.

But maximum benefit can be achieved by bringing under culti-
vation vast areas in arid and semi-arid regions. Moisture is essen-
tial for this purpose. In a number of regions of the world both land
and water exist. What is needed is to marry these for the good of
mankind.

Arid Lands as a Source of Food

The total land area of the world, excluding ice-covered regions,
is about 33,500 million acres. Less than 10 per cent of this is
cultivated, giving a per capita figure of about 1.25 acres. Our
Indian vegetary dietary can be produced from about 0.8 acre per
head, although the Chinese have been managing with 0.5 acre per
capita. The British require 1.4 acres per head to satisfy their
needs, whereas the people with higher standards of nutrition will
require two or more than two acres per capita.

One of the chief maladies of present day agriculture has been
that in many countries the area under the plow has been de-
creasing instead of increasing owing to greater demand on land
from other quarters like expansion of cities, highways, canal
systems, industries, and other matters of defence or strategic im-
portance. The world stands in great need of an increase in crop
acreage and better exploitation of water resources. This can be
done by reclamation of arid wastes, marginal and saline lands.
In India, for example, there is a considerable area of unculti-
vated land which can be utilized to advantage. Out of 810 million
acres, about 330 million acres are sown at present. Only 12 per
cent of the surface flow rivers have so far been used.

In any arid land, the problem is definitely the limitations of

CONCLUDING REMARKS PIS)

water and the available land fit for irrigation. We must first know
our assets before making a scientific plan for their utilization. I
suggest that this study can be done in four steps as follows:

1. The assessment of overall water available for irrigation and
of land resources fit for irrigation.

2. The assessment of the water and land resources at present
being utilized.

3. The possibility of using the balance of water and land to the
optimum limit for irrigation.

4. The possibility of finding new sources of water.

In the arid land of Rajasthan in India it has been found that
out of the total utilizable surface and ground water potential of
about 12 million acre-feet only about 4.3 million acre-feet have
been utilized. Thus there is great scope for further utilization.
Similar will be the case in quite a few of the arid regions of the
world. The water potential of any arid or semi-arid region being
meager, the problem would generally be to make a judicious,
economical, and wise utilization of the supplies available. The more
economical schemes could be taken up first, but there should be a
long-range plan for fully utilizing the resources.

Better utilization of the water now being used can be achieved
in two ways. One is to reduce all possible wastage and the other,
to put every cubic foot of water to the maximum utility. Wastage
occurs by evaporation, seepage, and transpiration. These can be
minimized by taking the water in lined canals and closed conduits
and by irrigating the fields by subsurface methods or by sprinkling.
For getting the maximum out of water, crop pattern in the region
should be studied and modified if necessary. It would be advisable
to adopt a crop scheme so that the moisture of one crop helps the
succeeding crop. Again, the optimum water requirements of crops
should be studied. I am of the opinion that in regions with similar
type of soil and climate the total water requirements of a given
crop inclusive of utilizable rainfall would generally be the same.

India has done valuable work on water requirements of plants,
which can be studied by other countries to advantage. After care-
ful study of climatic conditions, rainfall, water table, type of
soil, local agricultural practices, dust and wind storms, high
temperatures during the summer and other conditions, we have
evolved crop patterns and water requirements for crops in differ-

254 THE FUTURE OF ARID LANDS

ent tractsin the Bhakra Project area which is a typical arid region.
It is seen that in addition to the above points water requirements
for various crops depend upon: (2) nature of crop, (4) intensity of
irrigation, (c) Kharif-Rabi ratio, (d) size of holding.

With regard to the above points, it would be useful to establish
research stations and agricultural farms in the particular regions
concerned to conduct field experiments, collect information, and
render the necessary advice.

Other Improvements

The re-use of water arises only in the case of domestic and
industrial needs. As McClellan has suggested, surely all possible
ways must be considered for the re-use of water. I may add that
after domestic and municipal use the same water, after suitable
treatment, can be re-used for purposes of irrigation. This has been
tried in some parts of India with good results.

I venture to make another suggestion for supplementing the
water resources in arid and semi-arid regions. We should study
the feasibility and economics of diverting the supplies from some
rivers to basins in the neighboring regions. In India the Rajasthan
desert is supplied with water for irrigation and other purposes
from neighboring states.

Because water is so scarce in arid regions it would be advisable
to collect rain water for domestic use even on house tops. Yet
another source is dew. Israel has developed some techniques to
take advantage of dew for meeting in part the water requirements
of plants. Experiments for producing artificial rain should be
pursued in an effort to make it a commercial proposition.

Concluding, I would say that the socio-economic conditions of
the agricultural population have to be improved if better use of
the limited resources available in the arid regions is to be made.
Cooperatives for financial assistance and for proper distribution
of land should be set up. The colonization of such areas must
simultaneously be taken up with financial assistance from the
State. Communication and transport tacilities must be provided.

I have a firm belief that proper utilization of present resources
in arid and semi-arid lands is one of the more effective solutions to
provide food for the increasing population of the world.

PROSPECTS FOR ADDITIONAL
WATER SOURCES

Questions

How practicable 1s it to demineralize saline water?
How practicable 1s it to reuse waste waters?
How practicable 1s it to induce precipitation?

IV hat are the social and economic implications of these programs?

Demineralization of Saline Waters

SHEPPARD T. POWELL

Consulting Engineer, Baltimore, Maryland

It is widely acknowledged that there is a great need for fresh
water in many areas of the world. The ability of man to provide
food and other bare necessities of life, even on a subsistence level,
is dependent in large part on adequate supplies of water. To raise
his living standards above the minimum level, man must have
water of good quality in relative abundance. From early antiquity,
obtaining water has been one of the most pressing problems facing
man. Regardless of intellectual, economic, and political progress
over the centuries, the problem has not been solved in many
areas of the world, and, in fact, has become intensified with
population growth.

Even a cursory review of recent attempts to solve water
supply problems reveals a widespread consciousness of the needs.
In this respect there are no divergent opinions. However, the
general agreement on the need for water does not extend to the
means of solving the problem.

The conversion of salt or brackish water into usable fresh
water has a strong appeal to people in arid regions. In fact, it is
being considered as a means to increase supplies in many areas
having reasonable amounts of rainfall. The degree of interest and
hope of accomplishment are, in general, proportional to local
needs and to the shortness of available fresh water supplies.

Unfortunately, there are misconceptions concerning the com-
plexity of the problems involved in saline water conversion. There
are unjustified expectations of early success in this field. The
non-technical press has been overly optimistic in dramatizing
the developments in the conversion of saline water. This optimism

257

258 THE FUTURE OF ARID LANDS

has led many to believe that there will soon be ample quantities
of fresh water, produced at a cost comparable to that of natural
fresh water supplies. Although this belief is at present unrealistic
for the continental United States, the production of usable fresh
water from saline sources is neither visionary nor impractical.
In some areas of the world, even present day conversion processes
can produce fresh water at lower costs than those of natural
supplies. There is every indication that demineralization of saline
waters, through research and development backed by adequate
financial support, and by pooling of knowledge, will help solve
the water shortage problem in many arid regions and at a cost
commensurate with the resulting benefits.

Current Research Activities

Research in desalting processes by various groups and indi-
viduals in this country, scattered and sporadic over the years,
is now proceeding at an accelerated rate, largely as a result of
coordinating efforts and financial support by the Saline Water
Conversion Program of the United States Department of the
Interior. In 1952, the Congress of the United States, recognizing
the need for new sources of fresh water in arid and semi-arid
regions in this country, passed Public Law 448. The purposes of
the act are clearly stated:

To provide for the development of practicable low-cost means of
producing from sea water, or from other saline waters, water of a quality
suitable for agriculture, industrial, municipal, and other beneficial con-
sumptive uses on a scale sufficient to determine the feasibility of the
development of such production and distribution on a large-scale basis,
for the purpose of conserving and increasing the water resources of the
nation.

For those who may not be aware of the specific provisions of
Public Law 448, they are summarized here. The Secretary of the
Interior is authorized to:

(2) Conduct research and _ technical dlevelgyimen: work by
means of grants and contracts,

(0) ‘Teresinance methods for recovery of and marketing of by-
products,

DEMINERALIZATION OF SALINE WATERS 259

(7) Engage, by contract or otherwise, chemists, physicists,
engineers and other persons to conduct research and development
work,

(e) Cooperate with federal, state and municipal agencies and
other organizations in effectuating the purposes of the act,

(f) And, as may be appropriate, to correlate and coordinate
the research activities of private organizations engaged in this
field.

To carry out these provisions, an expenditure of $2,000,000 was
authorized for a five-year period. An initial appropriation of
$125,000 was made for the year ending June 30, 1953, followed by
a supplemental appropriation of $50,000 for specific use in award-
ing research contracts. To finance the program for the following
year, Congress appropriated $400,000, one-fifth of the amount
authorized in the law for the five years. An equal amount was
made available for use during the current year, which ends June
So 55-

The Congress designated the Secretary of the Interior to carry
out the terms of the act. Because the work and the results relate
to several of the bureaus in the Department, it was found that the
work could best be administered by a small group known as the
Saline Water Conversion Program, within the office of the Secre-
tary, under the Assistant Secretary for Water and Power. To
advise the Secretary in broad policy matters, a group was named,
consisting of nine qualified persons in various fields related to the
program. Provision was made, also, for liaison with other federal
agencies having interests in the conversion of saline water. A
departmental committee from the Bureau of Mines, Bureau of
Reclamation and Geological Survey provides technical and policy
assistance.

In carrying out the program, consideration has been given to
all known processes for demineralizing saline waters. Support
has been given to research groups for investigating various
schemes which show promise of economical fresh water produc-
tion.

260 THE FUTURE OF ARID LANDS

This sponsored research has been most fruitful in acquiring
fundamental data needed to evaluate the many proposed desalting
methods. The five-year plan of research and development being
administered in this country by the Saline Water Conversion
Program is, as far as possible, being coordinated with similar work
elsewhere in the world. In several countries abroad, private and
government supported technical groups are engaged in research
on saline water conversion methods. Among these are the Na-
tional Council for Applied Scientific Research (T.N.O.) in The
Netherlands, the Admiralty Materials Laboratory in England,
and others in France, Sweden, Germany, Switzerland, North
Africa, the Middle East, Italy, Australia, Union of South Africa,
and elsewhere. Widespread interest in such research has been
promoted throughout the world by the United Nations Educa-
tional, Scientific, and Cultural Organization and related activities.

In many respects the general program of research in other
areas parallels the extensive research now in progress in this
country. Much has been accomplished both here and abroad.
Regardless of these accomplishments, the financial aid provided
has been meager, considering the magnitude of the water needs
in arid and semi-arid regions throughout the world.

Status of International Cooperation

The need for cooperative action between nations on matters
of saline water conversion is recognized. This is evidenced by the
sponsorship by UNESCO and others organizations of meetings,
such as the present assembly, for the exchange of ideas and in-
formation.

In addition to these sessions, there is currently a fairly active
exchange of data by technical press releases, by individual
correspondence, and by reporting of research activities in many
countries.

An example of realistic cooperative effort was the recent survey
trip made by a United States mission, representing the Saline
Water Conversion Program, to Europe and Northwest Africa.

The purposes of the visit were: (2) to examine the work under-
way and that proposed in connection with the Organization for

DEMINERALIZATION OF SALINE WATERS 261

European Economic Cooperation, cooperative research projects
by the three nations, the Netherlands, the United Kingdom, and
France for the purpose of ascertaining the extent, if any, to which
the United States might participate; and (@) to exchange in-
formation on salt water conversion techniques, needs, and eco-
nomics with those engaged in this field in Europe and northwest
Africa.

The report of the mission (3) is highly informative on current
research and progress in the areas visited. The observers’ findings
warrant thoughtful study for guiding saline water conversion
activities elsewhere, since they highlight the merits and limitations
of many processes. European knowledge and progress in the
design and efficiency of certain apparatus may be more advanced
than our own along some lines.

Because of divergent viewpoints and approaches to the overall
programs, close cooperation and coordination of effort between
all research groups in this field should be encouraged and strength-
ened. This should apply not only to exchange of fundamental
data, but should be extended to cross-licensing of patentable
devices and through other contractual arrangement in the best
interests of all groups. It may be that an international commission
or patent committee can be established and directed by UNESCO,
or other coordinating agency, to bring about a reasonable and
acceptable program for mutual action in this field.

Conversion Processes—Past, Present, and Future

To many people, especially in this country, the idea of con-
version of sea water into fresh water is an Aladdin’s lamp concept
and is of recent origin. This has tremendous popular appeal. This
attitude ignores the many complications involved and the high
cost of realizing this objective. It is impossible accurately to
chronicle the history of man’s effort to separate fresh from salt
water, but undoubtedly it dates back to antiquity. Credit is given
by Hample (2) to Sir Richard Hawkins for the first successfully
operated distilling apparatus, as early as 1593. Undoubtedly, un-
recorded efforts to secure usable water by miscellaneous means
long predated this recorded accomplishment.

262 THE FUTURE OF ARID LANDS

Inquisitive search for usable fresh water over the years has
resulted in many workable processes for desalting saline waters.
Most of the known procedures for desalting have been classified (1)
broadly into three major groups according to the type of process
used, namely, physical, chemical, and electrical. Within each of
these general classifications are many specific methods of water
conditioning. These include vaporization, crystallization, sub-
limation, adsorption, ultrasonics, osmosis, ion exchange, electro-
ion migration, and numerous other processes and phenomena.

Of particular interest at the present time is the method of
electro-ion migration utilizing membranes which selectively
remove cations or anions of the dissolved salts in saline waters.
This process has been the subject of much experimentation.
New synthetic membranes being developed give hope of increas-
ing the efficiency of this method to a practical range.

Although it was at first thought that distillation had been
advanced about as far as possible, that is not true, and much
research is now being carried on under the Salt Water Conversion
Program on various distillation processes. Those being studied
critically are vapor compression stills, vacuum distillation, mul-
tiple effect evaporators, critical pressure separation and various
combinations of these schemes. As a result of these studies,
valuable data have been acquired on potential means for reducing
cost. One of the most recent developments in this field is Hick-
man’s modification of conventional vapor compression, which
takes advantage of fundamental principles of boiling to increase
the rate of evaporation. This modification, although still in the
research stage, gives hope of greatly increasing the capacity over
conventional evaporators.

Research in these and other methods is revealing the funda-
mental principles of operation and is pointing out the advantages
and disadvantages of many of them from a practical standpoint.
The time limitation precludes even superficial discussion here of
the merits of these processes. They are mentioned merely to
indicate potential avenues for further developments as applicable
to specific needs in various areas based on local conditions and
available energy sources.

DEMINERALIZATION OF SALINE WATERS 263

Generalization of any world problem is grossly misleading and
unreliable. This is certainly true of the separation of fresh water
from saline supplies. The physical, sociological, economic, and
political status of peoples and the meteorological and geograph-
ical environments in widely separated regions are so varied that
there can be no single solution to the problem. These and other
factors must be considered before any decision is reached as to
the desalting process most suitable in a particular area.

What Cost Saline Water Conversion?

The crux of realistic accomplishment in demineralization of
saline water anywhere rests wholly on permissible cost.
Everett W. Howe (6) has wisely pointed out, however, that “the
cost is never too high when human life itself is at stake.” Deter-
mination of the cost of desalted water involves a careful survey
of the possibilities of selective water use. Comprehensive statis-
tical data are needed for appraisal of the economic practicability
of partial or total use of desalted water for miscellaneous regional
needs.

The permissible cost of separating fresh water from saline
supplies anywhere depends on the urgency of the existing needs,
whether they be for irrigation, industrial, municipal, or other
uses. Local conditions must govern the acceptable cost of using
all types of saline supply. Comparison of treatment cost in
different areas is misleading unless allowance is made for local
influences. In areas where no fresh water is available, the accept-
able cost of demineralization bears little relation to that of areas
having relatively abundant natural fresh water supplies.

In semi-arid areas, selection of water supply rests upon the
comparative costs of converted water and fresh water imported
from remote sources.

In regions where saline water only is available and all fresh
water must be obtained by importation or conversion, strict
conservation of all fresh water used must be enforced. Scarcity of
fresh water promotes greater tolerance of lower quality in water
for many uses which in non-arid areas would be considered more
demanding. The economics of any particular situation will always

264 THE FUTURE OF ARID LANDS

govern the choice between fresh and saline supplies. Untreated
or partially desalted saline water must be utilized where and when
possible to limit consumption of costly demineralized water, thus
minimizing the overall water cost to consumers.

Presently available unit costs of demineralized water from
various processes are based upon laboratory research and small-
scale pilot plant operations, and should be considered only
approximations of ultimate costs. These costs may change con-
siderably in large-scale operations. Research data are invaluable,
however, for comparing cost of various desalting processes under
specific regional conditions, but may not be broadly applicable
except where comparable conditions are known to exist. Accept-
able costs for desalted water for military operations in areas
wholly devoid of fresh water, or where fresh water is in short
supply, would be prohibitive in areas where water from this source
is in competition with imported water or where reuse of existing
local water 1s practicable.

Little effort has been made, individually or collectively, to
grade quantity and quality of water for man’s specific require-
ments. Intelligent planning and wise use of salt and fresh water
would, we believe, produce results fully commensurate with the
effort and cost involved.

At the beginning of the Saline Water Conversion Program, the
cost of converting sea water to fresh water by the best known
process was estimated at $400 to $500 per acre-foot. For economic
evaluation of demineralization processes, two arbitrary cost goals
were set, one for municipal water and one for irrigation water.
These goals were $125 and $40 per acre-foot, respectively (38 and
12 cents per 1,000 gallons). For these criteria, no distinction is
made between sea water and brackish water as the supplies to be
converted: (rma):

No process has been developed up to the present time which
has met the desired initial cost goals. In fact, all schemes proposed
or in actual pilot-plant or full-scale production now show costs
in excess of these criteria. However, the results of current research
indicate that the goal of producing fresh water at a cost which
municipal and industrial consumers can pay seems to be in sight.

DEMINERALIZATION OF SALINE WATERS 265

For example, it is estimated that simple solar distillation appara-
tus, located in the southwestern area of the United States, will
produce water at a cost of from $1,600 or $1,700 per million
gallons, or slightly less. It is believed that the design of solar
stills can be improved to produce water from saline sources for
one dollar per 1,000 gallons, or $1,000 per million gallons. Costs
estimated recently for other desalting methods are $1.15 per 1,000
gallons for vapor compression and $1.75 for a six effect flash
evaporator.

For many uses, the ultimate cost to the consumer can be re-
duced by blending the distillate produced with the raw saline
supply. For instance, if the raw water contained total dissolved
solids of 4,000 parts per million, and if water of 2,000 parts per
million were usable for certain requirements, then the cost of
delivered water would be cut in half.

The primary demand for fresh water in arid and semi-arid
regions is to supply water for domestic use and for irrigation. It is
the latter use of water which is of the greatest importance in the
adaptation of treatment of saline supplies to make such water
suitable for the growth of crops in regions where there is absence,
or at least a shortage, of fresh water supply for irrigation.

In many cases there is a lack of realistic evaluation of this
phase of the problem. The quality of the water necessary for
irrigation varies widely with a number of factors. These include
total dissolved solids in the irrigation water and the ratio of
various salts contained in the supply. Of great importance, also,
is the soil condition in the selection of crops to be grown. In
studying this problem, H. E. Hayward, in his many publications,
has amassed and published voluminous data on the subject.

In determining the cost of conversion of saline waters to produce
water of satisfactory quality for irrigation purposes, one must
evaluate all these factors, and the choice of crops grown will be
influenced by the degree of treatment required in specific cases.
It is important that permissible water quality be evaluated in the
light of these limiting conditions.

The research program financed by grants and aids made possible
through the provisions of U. S. Public Law 448 is already highly

266 THE FUTURE OF ARID LANDS

productive. In the short period of less than three years, assembled
data have revealed the merits and estimated cost of demineralized
saline supplies and have made it possible to predict ultimate
costs of water production by various methods which are being
studied.

Constructive research now in progress gives more hope of the
possibility of reducing costs to acceptable values than in any
previous period.

Application in Arid and Semi-Arid Regions

The role which power or energy plays in demineralization cannot
be overemphasized. Davis S. Jenkins, Director, Saline Water
Conversion Program, Department of the Interior, has so appro-
priately remarked, ‘‘for processes using an external energy source,
the cost of energy alone for converting sea water will be, at the
minimum, in the order of $20 an acre-foot (at 5 mill power). Thus
it becomes important that nonconventional energy sources such
as solar energy, wind power, and geothermal energy be explored
vigorously in connection with process development and use.”’

There are many arid and semi-arid regions near the sea or other
saline water sources where demineralization systems could be
developed. Augmentation of fresh water supplies in many of these
locations is necessary because of industrial growth. The establish-
ment of industries in areas where existing water supplies are
inadequate to meet the civil, agricultural, and industrial needs
also depends on an additional source of fresh water. Conditions of
this type are found in many places in this country and throughout
the world. At such locations, economic salt water conversion
offers a form of relief from water shortage difficulties.

The conditions cited are illustrated in a number of rapidly
growing industrial communities in this country, such as the Texas
City area in Texas, in southern California, and even along the
eastern seaboard. In these areas, ample supplies of saline water
are available and at some locations within these areas, sources of
waste heat could be utilized for desalting purposes. —

Such conditions justify careful investigation to determine the
cost of producing fresh water for industrial uses for comparison
with the cost of importing water from remote sources.

DEMINERALIZATION OF SALINE WATERS 267

It is obvious that in areas where no industries exist and where
there are no local sources of energy, production of fresh water
from saline supplies utilizing waste heat is not possible. However,
it may be entirely possible to develop industries in some such
regions by the installation of small multipurpose plants designed
to produce power, steam for industrial purposes, and fresh water
as needed for industrial and other uses. For areas lacking con-
ventional fuels, nuclear energy might be utilized as a source of
heat for such a combination plant. Although not now attractive
cost-wise for demineralization alone in the United States, the
development of practical reactors designed for power generation
with provision for heat recovery would open many arid regions
for selective industries, especially those having low unit water
consumption. The recoverable heat from such units could be
utilized for saline water conversion in a number of separation
processes. Such a multipurpose project offers an intriguing ap-
proach to the problem of water supply in many arid and semi-
arid areas.

The need for saline water conversion is not limited to coastal
areas. Many inland areas which do not have adequate fresh
water have saline sources available. Many inland water sources
usually considered to be fresh water are in reality fresh only
seasonally, and for parts of each year should be classified as
saline.

Most tidal rivers are subject to variations in quality on account
of penetrations of salt water upstream during periods of low
river flow. Many inland streams are intermittently too saline to
permit using the water for irrigation, industrial uses, or for human
consumption. Ground waters often contain high concentrations
of dissolved salts, impairing their use for various purposes. Many
of these waters are far less saline than sea water, and therefore
can be demineralized more cheaply.

In addition, exchanges of water near coastal points of use with
water being diverted there from inland fresh water sources may
be feasible in some areas. As set forth in a recent report by the
Secretary of the Interior (5), this would result in indirect benefits
from saline water conversion to certain regions too far from the
sea for economical direct service with converted sea water.

268 THE FUTURE OF ARID LANDS

Recommendations

Our measure of the needs for saline water conversion and its
permissible cost cannot be reduced to any common denominator.
There is much confusion in this respect, since the degree of treat-
ment and imperative needs for specific purposes are seldom fully
defined. Before such matters may be intelligently evaluated,
there must be critical specifications with respect both to the
proven needs of an area and permissible economic balance between
capital investment, operational cost, and benefits accruing there-
from.

In a recent address, Thorndike Saville (4) made a statement on
the need for a national water policy in this country which in many
respects is applicable to other nations. Although proposed for
conservation of fresh water, the fundamental principles stated
are equally pertinent to the use of saline water. Following are
excerpts from his address:

The needs for water, some of them competitive, for all purposes, must
be forecast in the light of all available facts. The means to supply these
needs must be canvassed; existing and new sources, re-use, treatment
of wastes, ground water storage, importation from other drainage areas,
etc.,... planning is essentially a local and regional function and should
involve to the highest degree possible the participation of those who are
to be benefited (or injured) by the program.

The most reasonable approach and optimistic hope for the
future will depend upon some of these basic principles.

The increased efficiency of known desalting processes accom-
plished through research and development is encouraging, but
continued improvement in this respect must follow the laws of
diminishing returns, and beyond certain limits no future cost
reduction can be expected. As earlier indicated, there are, how-
ever, other avenues of exploration which give promise of reducing
the cost of desalting processes by taking advantage of potential
credits accruing from operating ingenuity. Some of these are:

(a) Use of low grade fuel available in the area or reasonably
near the site of the conversion plant.

(6) Combination of processes involving maximum utilization
of potential energy sources, including nuclear power.

DEMINERALIZATION OF SALINE WATERS 269

(c) Utilization of off-peak (dump) power, at reduced rates,
especially in areas supplied with hydrogeneration and atomic
power.

(d) Drastic conservation of high grade demineralized water by
reserving it for preferred selective uses.

(e) Re-use of all available sewage and industrial wastes, with
or without treatment.

(f) Grading of water quality for selective uses and maximum
use of saline water for all permissible industrial, agricultural,
and municipal requirements.

(g) Programming of all desalting projects as multipurpose
developments, depending upon the economic Justification for
such work.

(z) Miscellaneous potential credits, including those resulting
from recovery of by-products from concentrated brine, particu-
larly trace elements.

We would not presume to present here any detailed plan,
realizing the ramifications of such programming of cooperative
activity, but suggest the following action:

(2) Pooling of knowledge through world efforts as has now been
intelligently organized and planned by the United Nations
Economic and Social Council, UNESCO, the United States
Government, and other allied groups. This initial program of
proven worth should be extended and amplified.

(4) Continuation and acceleration of the existing United States
Saline Water Conversion Program’s method of cooperative
action between governmental agencies and private industries, and
extension of this program or adoption of similar programs in
other countries.

Industrial, agricultural, and municipal needs for practical
saline water conversion point to an opportunity for private
capital in a potentially lucrative enterprise. There is undoubtedly
an extensive world market for exploitation, especially in present
underdeveloped lands. Up to the present, industries developing
and manufacturing equipment have manifested only passive
interest in such world markets.

A recently published release by the United Nations Economic

270 THE FUTURE OF ARID LANDS

and Social Council, entitled ‘““The Development and Utilization of
Water Resources,’’ concisely detailed the world’s needs for
solution of water problems. The document was prepared by the
Secretary General of the United Nations in compliance with the
resolution submitted by the Council. Although this report was
not specific as to the world saline water conversion needs, many of
the recommendations contained therein are applicable to de-
sired cooperative action for solution of the water supply need in
arid and semi-arid regions everywhere. The descriptions of world
conditions presented in the document and the suggested pro-
gramming of future action deserve thoughtful study as they relate
to the pressing demand for fresh water in areas where only saline
supplies are available.

These general recommendations can be carried out only if
backed by trained administrative and technical organizations
furnished with adequate funds for financing continued research,
statistical studies, design development, and allied endeavors.

Voluminous data on saline water separation processes have
been acquired from the research sponsored by the Saline Water
Conversion Program. Valuable as this program is proving to be,
it is merely one step leading to realistic development of practical
processes applicable to specific conditions. The knowledge being
acquired is pointing out the potential merits of several processes
being investigated. Invaluable fundamental and practical in-
formation has been, and is being, obtained which should be
translated into the design of commercial or semi-commercial
installations as rapidly as possible. There still is need for further
research in this field, but sufficient knowledge has now been
acquired in one or two processes to minimize the financial risk
of proceeding to the next step of installing pilot or demonstration
plants for further development of these schemes. The need of
further action is realized and one process is now under test on a
small pilot model in Arizona and South Dakota.

The Saline Water Program should be extended by Congressional
action beyond the termination date of July, 1957.

The overall program should also include cooperative action by
state grants and other assistance from industry and public

DEMINERALIZATION OF SALINE WATERS 271

utilities in areas requiring the solution of water shortage problems,
either actual or potential.

Summary

To people in areas not penalized by inadequate fresh water,
the water shortage difficulties of arid regions may be of casual
interest only. The need for solving these problems, however,
presents an international task of the greatest importance. The
availability of water to support human life in any region has
great significance in world affairs. A practical method of making
fresh water available to the arid regions of the world would have
a beneficial and stabilizing effect on the social, economic, and
political life of all nations.

REFERENCES
1. Chapman, O. L., G. W. Lineweaver, and D.S. Jenkins. October 1952°

Demineralization of Saline Waters. Department of the Interior
Washington, D. C.
2. Hample, C. A. Fresh water from the sea, 1948. Chem. Eng. News 26,
IN@s Hii UCI's
3. Inspection of Research and Developments in Saline Water Conver-
sion in Europe and Northwest Africa. September, 1954. Department
of the Interior, Saline Water Conversion Program, Washington, D. C.
. National Water Policy. May 1954. Proc. dm. Soc. Civil Eng. 80.
. Secretary of the Interior. January, 1955. Third Annual Rep. on
Saline Water Conversion.
6. UNESCO, Arid Zone Programme. IV. Reviews of research on
problems of utilization of saline water. Page 73, Utslization of Sea
W ater.

an Bb

Demineralization as an Additional

Water Source for Arid Lands

WILHELMUS F. J. M. KRUL

University of Delft, and Government Institute
for Water Supply, The Hague, The Netherlands

In this paper a general outline will be given of the actual
desalting processes and their possible development in the near
future. At the end are some references to recent literature, in
which technical details may be found.

In arid zones saline water often occurs in the form of: (a) sea
water at the shore, (4) brackish water in salt inland lakes, (c)
more or less mineralized ground water.

Economically acceptable processes of demineralization may add
to the possibilities of solving arid lands problems.

Up to the present demineralization has been practiced in
special cases at relatively high costs, particularly by: (a) distilla-
tion with vapor compression—small units in sea vessels, e.g.,
submarines, and big installations in some places like Kuweit
(2,650 tons per day) Curacao and Aruba (3,200 and 1,600 tons
per day); (6) ion exchange—on a rather large scale for water
softening in industry (boiler feed water), and only on a small scale
for demineralization; (c) distillation by solar energy—small units
for domestic use.

Coordination of Research

All over the world research is being carried out to improve the
economy of different demineralization processes. Coordination
and cooperation have developed fast in recent years. In the
United States it has been carried out under the Saline Water
Conversion Program, Department of the Interior, Washington;

272

DEMINERALIZATION 273

in Europe under the O.E.E.C. (Organization for European
Economic Cooperation), Paris, which is part of the Economic and
Social Council of the United Nations. OEEC has established a
Working Party No. 8 (W.P.8) on “demineralization of salt and
brackish waters,” in which several European countries (Belgium,
Denmark, France, Germany, The Netherlands, Norway, Sweden,
United Kingdom) are represented. This cooperates with Australia,
the Union of South Africa, and with the United States through the
director of the Saline Water Conversion Program (David S.
Jenkins) and the Rockefeller Foundation.

Since I am chairman of Working Party No. 8, I shall give an
outline of the actual status of that committee’s activities.

In 1953 an ad hoc group of experts prepared a report for Work-
ing Party No. 8 on existing methods and possibilities for further
development from a technical and economic standpoint. On the
basis of this report W.P.8 recommended further investigations in
four directions, namely, distillation with vapor compression,
electrodialysis, ion exchange, and solar energy distillation.

Four countries, the United Kingdom, The Netherlands, French
Algeria and French Morocco proved to be willing to form a re-
search center for one of these directions, in cooperation with
other countries that would be willing to share the costs and profits.
Meanwhile some of these proposals have been worked out. The
actual situation is as follows.

Status of Current Research
Distillation with Vapor Compression

This needs complicated machinery and skilled labor. Total
costs depend largely upon fuel prices and are almost independent
of the salinity of the raw water. Hence this process is especially
to be recommended for the treatment of sea water and for large
units, leading to lower investment per ton of product.

The crucial point is the forming of deposits of calcium and
magnesium salts in the apparatus, since periodical descaling
causes very expensive maintenance. Incrustration can be pre-
vented by adding special chemicals, affecting the pH, but at the
same time corrosion under high temperature must be prevented.

It is also possible to add dispersing agents, modifying the

274 THE FUTURE OF ARID LANDS

physical nature of the scale-forming agents, thus rendering them
non-adherent to the metal heating surfaces, or to add sequestering
agents, maintaining scale-forming solids in solution.

Much research already has been done in this field, especially
in England. The United Kingdom has now set up a two-year
program for additional investigations under the guidance of the
Department for Scientific and Industrial Research, to be carried
out at the Admiralty Materials Laboratory.

Other European countries have been invited to share in the
costs, the know-how, and other profits. The work will be super-
vised by a steering group, in which the collaborating countries
are represented.

W.P. 8 has quoted the target value for the total cost, including
capital investment, for reducing sea water (20,000 ppm Cl’) to a
salt content of 300 ppm Cl’ (by mixing sea water with the dis-
tilled water) for a capacity of 10 tons per day at $1.30 per
ton. For a big installation of 1,000 tons per day the cost would be
30 cents per ton. It should be borne in mind that these figures have
no absolute value; they give only an idea of the economic level of
the process, in comparison with other solutions.

Electrodialysts

This process has been studied especially in the United States
(Ionics, Inc., Boston, Mass.) and in the Netherlands (National
Council for Applied Scientific Research, T.N.O.).

The process of desalting by direct current, passing through a
three-compartment apparatus with 2 membranes, e.g., of cello-
phane, has been known for a long time. It is, however, hardly
applicable to sea water, since a rinsing liquid is needed of similar
or lower concentration than that of the fresh water to be obtained.

However, quite recently a new possibility has been opened as a
result of the introduction of highly selective membranes, which
permits the use as rinsing liquids of salt solutions of a considerably
higher concentration than the dialysate. Moreover, great progress
has been made by constructional improvement, especially by
placing a great number of selective membranes in shunt with
extremely narrow compartments (0.5 millimeter).

DEMINERALIZATION 275

Some three years ago a T.N.O. apparatus on a semi-technical
scale for desalting a water with 1,000 ppm Cl’ to a content of
300 ppm Cl demanded 18 killowatt-hours per ton. At the actual
moment the energy consumption can be as low as 0.5 kilowatt-
hours per ton and even less. The cost of investment in such
apparatus is low in comparison with distillation or ion exchange.

Under the auspices of W.P. 8 a certain number of countries have
already decided to take part in this project.

In the United States a small electrodialysis apparatus of this
kind has been delivered by [Ionics to various consumers in order
to get practical experience under different circumstances.

In the Netherlands, the National Council for Applied Scientific
Research has planned a two-year research program, in which
desalting of waters of different composition will be practiced in
large-sized units. The principal aims of the investigations will
be to find out the optimal dimensions of desalting units for
various applications and constructional improvement of the
installations.

In 1953 W.P.8 estimated the target value for a daily production
of 10 tons of a water of 300 ppm Cl’ from brackish water with
1,000 ppm Cl’ at a total cost of 30 cents per ton. Up till now no
indications have been found that under proper conditions the
results will not come up to the expectations.

For a capacity of 1,000 tons per day the expected cost would
decrease to 8 cents per ton. The cost level largely depends upon
the price and the durability of the membranes and the price
of electric current. This may explain the great differences in the
figures, given above, from those published in the 1953 Report on
Saline Water Conversion as a result of the work of Ionics, Inc.
According to Ionics the total cost of desalting from 1,000—300 ppm
Cl’ for a capacity of 1,000 tons per day would only be 1 cent per
ton.

It should, however, be acknowledged that desalting problems
mostly occur in regions with relatively expensive electric power.

The Lon-Exchange Process

This demands high expenditure for chemicals for regeneration.
The amount of chemicals needed is almost proportional to the

276 THE FUTURE OF ARID LANDS

amount of salt to be removed, so that the application is practically
restricted to brackish water.

It may be possible to decrease the production cost by improving
the regeneration process. To this end ion exchange can be prac-
ticed in stages, several resin beds being used in succession, accom-
panied by counterflow regeneration. Electric current can also be
applied in regeneration to save chemicals.

Much research has already been done in Algeria, since this
method of demineralization is especially adapted to a country
where skilled labor is scarce and power comparatively costly.
France has set up a project for further research in Algeria and is
willing to cooperate with other interested countries under the
auspices of W.P.8.

The experts group of W.P. 8 have estimated the total cost of
desalting in the range from 1,000-300 ppm Cl’ at $1.60 per ton
for a capacity of 10 tons per day.

Solar Energy

In the same way the French protectorate of Morocco intends
to do more research in the field of distillation with solar energy,
a process that must be restricted to small quantities in view of the
large installations required. It is cheap in operation, as it does
not require any fuel or electric energy, nor any chemicals.

Factors Affecting Application

In view of possible applications in the further development of
arid regions, we must distinguish between: (a) water for human
consumption, (4) water for animal husbandry, (c) water for
agricultural purposes. Other factors are the availability of electric
current and skilled labor.

Human consumption asks for water of high quality (300-500
ppm Cl’) in relatively small quantities, whereas production cost
can be rather high.

For cattle breeding the salt content may be 500-1000 ppm Cl’.

For agricultural purposes large quantities of medium quality
(750-1,000 ppm Cl’) must be available at a low price.

It would seem that in the near future demineralization of
brackish ground water in arid zones will become a practicable

DEMINERALIZATION PLULT/

solution for human consumption. For very small quantities, dis-
tillation with solar energy may be recommended. For larger
quantities electrodialysis may prove to be the best process, and
for desalting of sea water both distillation with vapor compression
and electrodialysis may be successful. Ion exchange will also be
practicable for brackish ground water in smaller quantities.

For desalting highly mineralized ground waters (5,000 ppm
Cl’ or its equivalent), as occur in many parts of the world, e.g.,
Algeria, Morocco, Australia, and South Africa, electrodialysis
may prove to be the unchallenged best solution, at least for
application to animal husbandry.

The present status of technical development does not yet
permit applying demineralization to the large quantities of
water needed for irrigation. However, in a more remote future
the results of the research which is now going on in so many
places and, perhaps, the availability of cheaper energy resources
may lead to more favorable conditions.

Consideration should therefore be given to means and ways to
foster further investigations in this field, if possible on an inter-
national basis. I therefore propose as a basis for discussion that a
lasting contact should be established between the UNESCO
Advisory Committee on Arid Zone Research, the United States
Saline Water Conversion Program, and Working Party 8 of
O.E.E.C. in order to promote worldwide coordination.

REFERENCES
General

Annual reports of the Secretary of the Interior, U. S. A. “On saline
Water Conversion,” 1952-1953-1954.

Deéminéralisation des eaux salées, Presse/D(53)10, O.E.E.C., Service
de Presse 4-8-1953, Chateau de la Muette, 2, Rue André Pascal,
Paris XVle.

Ellis, C. B. 1954. Fresh Water from the Ocean, Ronald Press, New
York.

Electrodialysts

1. Arnold, M. H. M. and L. R. Seaborne. Ind. Chem. p. 295 (July).
2. Wegelin, E. 1953. Bull. Centre Belge d’Etude et de Documentation des
Wai aIcceenOy Zep te 2—loo.

278 THE FUTURE OF ARID LANDS

3. Wegelin, E. 1954. Ion exchange and its applications, ‘“Conference
Soc. of Chem. Ind., London, pp. 122-128.
4. Winger, A. C., G. W. Bodamer, and R. J. Kunin. 1953. 7. Electro-

chem. Soc. 100, 178.
g. Winger, A. GC; GW. Bodamer, Ko), Kunin, (C2 Siuigerana:
G. W. Harmon. 1955. Ind. Eng. Chem. 47 (1), 50-60.

Distillation by Vapor Compression
6. Langelier, W. F. 1954. Mechanism and control of scale formation in
sea water distillation. 7. 4m. Water Works Assoc. 46 (5), 461-469.
7. Latham, Jr., A. 1945. Compression distillation. Petroleum Refiner 24,

515.
8. Reports of the Admiralty Materials Laboratory, Holton Heath,
Poole, Dorset, England.

Ton Exchange
g. Ion exchange and its applications, Conference Soc. of Chem. Ind.,

London, 1954.
10. Ion exchange in water and waste water treatment. Symposium,

1955. Ind. Eng. Chem. 47 (1), 46-101.

The Salinity Factor in the
Reuse of Waste Waters

H. E. HAYWARD

United States Salinity Laboratory,
Riverside, California

In the arid zones of the seventeen western states and other
arid regions throughout the world, it is generally recognized that
the total area of irrigable land far exceeds the acreage that may
be served with available supplies of irrigation water. For example,
the 1950 Census of Agriculture reports over 24,000,000 acres of
irrigated land in the seventeen western states, and the future
program for irrigation agriculture calls for the use of as much as
30,000,000 acres of land. An even higher estimate of 51,500,000
acres of irrigable land in the seventeen western states was made
by the National Resources Board in 1936.

Paulsen has stated (2) that the “total average annual water
supply in the 17 Western States is about 390 million acre-feet as
compared to the present withdrawal use of about 100 million
acre-feet. Except for irrigation, only a small part of the water
withdrawn for use is consumed, estimated by some to be as little
as 10 per cent.” This statement suggests a very favorable situa-
tion, but the distribution of water supplies is such that some areas
have surplus water at all times whereas other areas seldom have
sufficient water. In the Southwest, the average annual runoff is
less than one-fourth inch, and no streams flow out of the Great
Basin. This indicates the importance of planning for irrigation
agriculture so that there may be a more effective use of all avail-
able water resources. Such a program involves, among other

279

280 THE FUTURE OF ARID LANDS

measures, the reuse of drainage and return flow water from irri-
gated land and the use of sewage effluent and industrial wastes
whenever possible.

Bowen and Powell have discussed two major possibilities of
additional water supplies: the induction of precipitation, and the
demineralization of saline waters. I wish to direct my attention
to one aspect of the third question proposed as a subject for
discussion at this technical session: ‘“‘How practicable it is to reuse
waste waters?’ This is the salinity factor in the reuse of waste
waters. Three points will be considered: (1) the characteristics
which determine the quality of water for irrigation use; (2) the
characteristics of return flow water and drainage as they relate
to the quality of water for agricultural use; and (3) the conditions
under which saline waters may be used to augment the water
supplies essential to the maintenance and extension of irrigation
agriculture in arid lands.

Quality of Irrigation Water

Although the quantity of available water is the primary con-
sideration in the development of irrigation agriculture, quality
of water becomes a more and more critical factor as the supplies
of surface and ground waters are depleted. Four characteristics
determine quality of water for irrigation use: (1) the total con-
centration of soluble salts, (2) the concentration of sodium and
the proportion of sodium to calcium plus magnesium, (3) the
concentration of bicarbonate, and (4) the occurrence of micro-
elements such as boron in toxic amounts.

In many cases, the total concentration of soluble salts is the
best single index for evaluating the quality of irrigation water.
The salt content of most irrigation waters ranges from 0.1 to
5 tons of salt per acre-foot of water (approximately 70 to 3500
ppm). The amount of salts in river waters in western United
States varies from as low as 70 ppm in the Columbia River at
Wenatchee, Washington, to 740 ppm in the Colorado River at
Yuma, Arizona, 1,574 ppm in the Sevier River at Delta, Utah,
and 2,380 ppm in the Pecos River at Carlsbad, New Mexico. The
salt concentration in river waters may vary materially, depending
upon the sampling site, and this factor is important in relation

SALINITY FACTOR 281

to return flow downstream. Thus, the Rio Grande has a range in
salt concentration from approximately 180 ppm to 1,775 ppm
at different points on the river.

Ground waters from pumped wells constitute the principal
source of irrigation supplies in many areas, and the range in salt
concentration may be much greater than for surface waters. For
example, analyses of a large number of wells in the Coachella
Valley, California, indicated a range in soluble salts varying
from 130 ppm to 8,500 ppm. In addition, wells in close proximity
to each other may have very different salt concentrations. Two
wells in this valley within a half mile of each other have salt con-
centrations of approximately 400 and 8,500 ppm, the former
well being 65 feet deep, the latter 180 feet.

The principal ions found in natural waters are the cations
calcium, magnesium, and sodium, and the anions bicarbonate,
sulfate, and chloride. Potassium, nitrate, fluoride, boron, and other
constituents may be present in low concentrations. Sulfate and
chloride salts usually predominate, but occasionally waters may
be high in bicarbonates and less frequently in nitrates.

The sodium factor in irrigation water 1s related to the alkali haz-
ard and is determined by the absolute and relative concentration
of the cations. If the proportion of sodium is high, the alkali
hazard is high and, if calcium and magnesium predominate, the
hazard is low. The soluble cations in the irrigation water have a
pronounced influence on the distribution of the exchangeable
cations in the soil, and it is for that reason, in part, that the so-
dium content in irrigation water is important. If sodium consti-
tutes less than one-half of the cations in the irrigation water,
there is ordinarily very little danger of unfavorable sodium soil
conditions developing from the use of the water but, as the
proportion of sodium increases, the hazard increases.

In some instances, waters may be low in total salts but high in
bicarbonate. This condition tends to aggravate the sodium prob-
lem in soil where the amount of bicarbonate is considerably in
excess of the calcium plus magnesium. In such a case, residual
sodium carbonate 1s present and, as the irrigation water evaporates
from the soil, calctum and magnesium carbonates precipitate, and
the sodium percentage of the soil solution increases. This is

282 THE FUTURE OF ARID LANDS

followed by the replacement of calcium by sodium on the soil
particles, the exchangeable sodium percentage of the soil increases,
and the physical condition and permeability of the soil are likely
to be impaired. In addition, the hydrogen ion concentration of the
soil may decrease and organic matter may be dissolved, result-
ing in the dark color which is characteristic of a so-called black
alkali soil.

The Salinity Laboratory has proposed a scheme of classification
in which waters are divided into four classes based on salt con-
centration and into four other classes with reference to the prob-
able extent to which soil will adsorb sodium from the water and
the length of time required to affect the soil adversely. These are
designated as the salinity hazard and the sodium hazard. The
salinity hazard is measured in terms of electrical conductivity
expressed in micromhos per centimeter at 25°C. Class 1 water
ranges up to 250 micromhos per centimeter; Class 2, from 250 to
750; Class 3, from 750 to 2,250; and Class 4, in excess of 2,250.

The relative proportion of sodium to other cations in an irriga-
tion water has usually been expressed in terms of soluble sodium
percentage, but it appears that the sodium adsorption ratio which
is simply related to the adsorption of sodium by soil, has some
advantage for use as an index of the sodium or alkali hazard of
water. This ratio is defined by the equation

SAR Ne
- = (Cas ie Me**) /2

where sodium, calcium, and magnesium represent the concentra-
tions of these ions in milliequivalents per liter. The sodium
hazard is largely determined by the proportion of sodium to cal-
cium plus magnesium present, together with the total salt content
as indicated by electrical conductivity. Thus, the curves are given
a negative slope to take into account the relation of the sodium
hazard to total concentration. For example, a water with an
SAR value of g and a conductivity of less than 168 would be an
Si water, from 168 to 2,250, an S2 water, and greater than 2,250,
an S3 water. This system is somewhat arbitrary and tentative,
but field and laboratory observations appear to support it.

SALINITY FACTOR 283

TABLE 1

Permissible Limits of Boron for Several Classes of Irrigation Waters (5)

Boron Class Sensitive Crops Semitolerant Crops Tolerant Crops
(ppm) (ppm) (ppm)

Excellent <0. 33 <0.67 <1.00

Good 0.33 to 0.67 0.67 to 1.33 I.00 to 2.00

Permissible 0.67 to 1.00 TG GEEOLZEOO 2.00 to 3.00

Doubtful I.00 to 1.25 2.00 to 2.50 3.00 to 3.75

Unsuitable SiO 22.59) = B.915

Boron is a minor constituent of practically all natural waters,
and irrigation waters should be analyzed for this element if there
is any reason to suspect its presence at toxic levels. Although
boron is an essential microelement for plant growth, it may be
toxic at concentrations only slightly in excess of those needed for
optimum growth. Toxicity may develop with boron-sensitive
crops when the concentration is as low as 1 ppm but, for most
crops, water containing I to 2 ppm is satisfactory, and waters up
to 3 ppm may be used with the more boron-tolerant crops. Water
containing in excess of that amount of boron is, in most cases,
unsuitable for irrigation purposes. The permissible limits of boron
for several classes of irrigation waters considered on the basis of
the relative sensitivity of the crop to boron are given in Table 1.

Characteristics of Return-Flow Water

The pollution of streams by irrigation residues and the charac-
teristics of return-flow water have been discussed by Scofield
(3), Howard (1) Wilcox (6), and others.

In some cases, the water available for irrigation is unfit for
irrigation use in its natural state, owing to unsatisfactory quality,
but there are other instances where the quality of the irrigation
supplies has been impaired by drainage and return flow down-
stream. The major effects of use and reuse of irrigation waters as
related to quality are (1) an increase in the total amount of
dissolved solids, (2) the loss of calcium, magnesium, bicarbonate,
and sulfate by precipitation, and (3) an increase in the quantity
and proportion of sodium and chloride in the water.

284 THE FUTURE OF ARID LANDS

ABE Re2

Discharge and Salt Burden for Seven Stations on the Rio Grande above Fort
Quitman, Texas

(Annual means and total for the year 19497)

I 2 3 4 5 6
Total
Concentration
‘ Condace s Dissolved
Station Miles Dis- tivity Diseiengs Solids
solved ae? (acre-feet)
: (micro- (tons)
Solids
mhos
ee per cm)
Otowi Bridge, N. M. ° 206 320 1,323,000 370,440
San Marcial, N. M. 184 353 520 1,0$4,000 505,920
Elephant Butte, N. M. 240 404 610 813, 500 447,425
Caballo Dam, N. M. 268 441 670 GON INTE)
Leasburg Dam, N. M. 318 489 730 689 , 143 458,280
El Paso, Tex. 37/5 750 1160 463,540 472,811
Fort Quitman, Tex. 456 2631 4030 134,030 479,827

2 These data are assembled from records of the U. S. Geological Survey, the U. S. and Mexico In-
ternational Boundary Commission, the U. S. Bureau of Reclamation, and the U.S. Salinity Laboratory

Data for the Rio Grande illustrate these effects. Table 2 gives
the discharge, salt burden, and total concentration of salts for
seven stations on the Rio Grande above Fort Quitman, Texas.
These data are for a representative year and illustrate the effect
of return flow on the quality of irrigation water.

The distance from the Otowi Bridge, New Mexico, to Fort
Quitman, Texas, is approximately 450 miles, and in this stretch
of the river there are many diversions to irrigated lands. The
four major irrigated areas are the Middle Rio Grande Project
between Otowi and San Marcial, the Rincon and Mesilla Valleys
between Caballo Dam and Courchesne, and the irrigated areas
in Texas west of Fort Quitman which include the irrigation in El

Paso County and that in the Hudspeth District.

Middle Rio Grande Valley Project 67,000 acres
Rincon and Mesilla Valleys 8§,000 acres
E] Paso County 57,000. acres
Hudspeth District 12,000 acres

Total 221,000 acres

SALINITY FACTOR 285

These values are estimates and, in past years, there have been
wide fluctuations in the amount of irrigated acreage owing
primarily to variations in available water supplies. For example,
in E] Paso County, the irrigation acreage reached a peak of over
67,000 acres in 1950 and dropped back to slightly less than 57,000
acres in 1954.

The first point to consider is the salt burden of the river, which
is expressed as dissolved solids in tons (column 6). These data
indicate that the salt burden from Otowi Bridge to San Marcial
increases significantly, but, from the reservoir area at Elephant
Butte to Fort Quitman, Texas, it is very uniform, the differences
probably being within the limits of measurable error. On the other
hand, the discharge figures (column 5) indicate that there is a
marked decrease in stream flow until essentially all the water
has been used below El Paso. This decrease in flow is the result
of diversions to the irrigated lands along the river above Fort
Quitman. The effect on quality of water of a nearly constant
salt burden and a decrease in stream flow is reflected in columns
3 and 4 which show the total concentration of dissolved solids in
parts per million and as conductivity expressed in micromhos per
centimeter. There has been an approximate sixfold decrease in
discharge from Elephant Butte to Fort Quitman and slightly in
excess of a sixfold increase in total concentration of salts over the
same stretch of the river. Thus, flow decreases and the concentra-
tion of salt increases, while the salt burden is virtually constant.
This indicates that the tonnage of salt which is carried back to
the stream in the return-flow drainage is approximately equal to
the amount of salt diverted from the river in the irrigation
water. If this were not the case, the tonnage of dissolved solids
would probably decrease in a downstream order.

Although the total dissolved solids do not change materially
from Elephant Butte to Fort Quitman, there is a marked increase
in the content of sodium and chloride present (Table 3). These
data illustrate the effect of use and reuse of water on composition.
The tonnage of sodium has increased threefold and the sodium
concentration, over 30 times; chloride tonnage has increased
twelvefold and concentration, 117 times. On the other hand, there

286 THE FUTURE OF ARID LANDS

ABERSS

Sodium and Chloride for Seven Stations on the Rio Grande above Fort Quitman, Texas
(Annual means and totals for the year 1949)

Sodium Chloride
Station
ppm meq/! tons ppm meg/l tons
Otowo Bridge, New Mex. WG) O84 4, 70m PGW 13,3018
San Marcial, New Mex. AE TOR OE 25 24 1) 45 CS
Elephant Butte, New Mex. Be Dacl9) OU OHO air Sy) | Gyan UO
Caballo Dam, New Mex. GH DAR COSAos PV 1b OYA ZB ex)
Leasburg Dam, New Mex. TOUS LOMMO Opp G2 Aye OR OAS
FE! Paso, Texas 147 6.41 92,940 ONS TS BoB SSC
Fort Quitman, Texas 601 26.15 109,546 874 24.65 159,298

@ From same sources as Table 2.

has been a large reduction in the amounts of sulfates and bicar-
bonates.

On the basis of the data for discharge, salt burden, salt concen-
tration, and composition of Rio Grande waters, it is clear that
the quality of irrigation water downstream may be affected
adversely where there are diversions for irrigation that deplete
the flow. The data also indicate the importance of an accurate
knowledge of the quality of water if it is to be used for irrigation.

Use of Saline Waters for Irrigation

With accurate data on quality at hand, the next consideration
is: can waters of low quality be used without inducing unde-
sirable effects on the properties of the soil to which they are
applied and the crops which are to be grown?

If waters of high salinity are used for irrigation without proper
management or adequate drainage, the salts accumulate in the
root zone and the concentration of the soil solution increases.
As the salt concentration or osmotic pressure is built up, there is
a decrease in the ability of the plant roots to absorb water in
adequate quantities. Experimental work has shown that retarda-
tion of plant growth is virtually linear with an increase in osmotic
pressure of the soil solution and, in most cases, it is largely
independent of the kinds of salt present. When the osmotic
pressure of the substrate has increased sufficiently, the entry of
water into roots will cease, and most crop plants will die.

SALINITY FACTOR 287

When the irrigation water applied is high in sodium and the
soil becomes partially saturated with sodium, the clay particles
are highly dispersed and may move downward through the soil
to lower levels. This results in an unfavorable soil structure in
which the first few inches of the soil profile may be relatively
coarse textured, but lower in the profile where the clay has
accumulated, there is a dense layer that is frequently very low
in permeability. When such soils are wet, they tend to “run”’
together; when dry, they form hard clods, with large cracks on the
surface.

In addition to the adverse effects of saline and high-sodium
waters on soil characteristics, and the reduction in the intake of
water by plant roots where the substrate becomes excessively
saline, some salts are toxic to crop plants when they occur in
soils in excess amounts. The ions which are most likely to cause
toxic reactions are chloride, sodium, bicarbonate, and sulfate.
Among the microelements, excess boron most frequently produces
symptoms of injury.

Since undesirable soil conditions develop and unfavorable
crop responses are likely to occur when the quality of irrigation
water is unsatisfactory, the question arises: under what conditions
and to what extent can saline waters to be used to supplement
available supplies in areas where water shortage is a major prob-
lem? Time does not permit me to discuss in detail the major
aspects of this question, but the following lines of approach to
the solution of the problem are suggested for consideration. First,
selection of suitable land for irrigation; second, use of proper
water-management practices; and third, selection of salt-tolerant
crops which are adapted to local climatic conditions.

Information regarding the salinity and sodium status of the
soil, its hydraulic conductivity or water-transmission properties,
the texture of soil and character of substrata, water-table condi-
tions, drainability, and the topography of the area is necessary
in considering the possibilities of using water of poor quality.
Preleaching is required if the soil is saline, and leaching plus
amendments may be indicated if the soil is saline-alkali and non-
gypsiferous. Drainage is essential to remove excess salts, to

288 THE FUTURE OF ARID LANDS

prevent the accumulation of salts by irrigation, and to prevent
the occurrence of high water-table conditions.

With respect to water management, irrigation should be con-
trolled in such a way that a favorable salt balance will be main-
tained. This occurs when the output of salts for a given area
exceeds the input (4). Enough water, in excess of the consumptive
use or evapo-transpiration requirements, should be applied to
remove from the irrigated area approximately as much salt as is
transported onto the land by the irrigation water. Since plants
absorb water from the soil solution but take in only a small
proportion of the dissolved constituents, there will be a gradual
increase in the salinity of the soil unless the amount of water
applied to the land is sufficiently in excess of the plant require-
ments and the losses of water by surface evaporation so that
salts in solution are carried out of the root zone and into the under-
drainage.

Successful water management involves two operations which
are opposed to one another. In the first instance, sufficient water
must be applied to insure the movement of salts through the
profile so as to prevent an accumulation of salts and maintain a
favorable salt balance. On the other hand, excessive use of water
must be avoided in order to prevent the development of high
water-table conditions and the consequent drainage problems.

A second consideration in water management is the possibility
of using “blending”? methods in cases where some of the water
supplies are too saline, but other supplies of good quality are
available. If soil properties and drainage conditions are satisfac-
tory, waters of high salinity may be used by “blending,” or
mixing, waters of poor and good quality in such proportions that
the salinity of the water applied to the land has been reduced to
reasonably satisfactory limits.

The selection of appropriate crops may serve to ameliorate a
situation where the use of saline water has resulted in saline
conditions. Field and plot tests at the Salinity Laboratory and
elsewhere have demonstrated that there is a wide difference in
the relative salt tolerance of the crops grown in the western states.
For example, the chances for satisfactory yields of field crops

SALINITY FACTOR 289

grown under saline conditions will be greater if crops such as
sugar beets or cotton having high salt tolerance are planted rather
than field beans, which have a very low salt tolerance. Likewise
with forage crops, better results may be expected if some of the
salt-tolerant grasses are grown instead of salt-sensitive clovers
such as Alsike, red, and Ladino varieties. With vegetable crops,
garden beets, asparagus, and spinach will be more successful than
green beans or celery. In selecting salt-tolerant crops for a given
area, it is important to understand that climatic factors may pro-
foundly influence the response of plants to salinity. For this
reason, the choice of suitable salt-tolerant varieties will depend
upon local climatic conditions.

Summary

The quality of water supplies becomes a more and more
critical factor with the increasing need for additional water for
irrigation in semi-arid and arid areas. The major characteristics
which determine water quality are: total concentration of soluble
salts, concentration of sodium, concentration of bicarbonate, and
the occurrence of minor elements such as boron in toxic amounts.
The reuse of waters returning to the stream after diversion for
irrigation commonly results in a significant increase in total
salinity. In some cases, increased concentrations of sodium,
chloride, and other elements may affect soil characteristics
adversely or may be directly injurious to plant growth. For these
reasons, it is important to determine the characteristics of all
questionable waters. With this information at hand, the possible
solution of the use of such waters depends upon the use of appro-
priate soil- and water-management practices and the selection of
salt-tolerant crops.

REFERENCES

1. Howard, C. S. 1953. Irrigation and water quality. Reclamation Era,
30, 1-2, A,

2. Paulsen, C. G., Chief Hydraulic Engineer, U. S. Geological Survey.
1954. “‘What to do about our growing demands for water.’ Presented
at the 12I1st meeting of the American Association for the Advancement
of Science, Berkeley, California.

290 THE FUTURE OF ARID LANDS

3. Scofield, C. S. 1932. Stream pollution by irrigation residues. Ind.
Eng. Chem. 24, 1223-1224.

4. Scofield, C. S. 1940. Salt balance in irrigated areas. 7. dgr. Research,
61, 17-40.

5. United States Salinity Laboratory Staff. 1954. Diagnosis and im-
provement of saline and alkali soils. U. S. Dept. Agr. Handbook
No. 60.

6. Wilcox, L. V. 1953. Irrigation water quality as affected by use and
reuse. Presented at Symposium, ‘‘Water use and conservation policy”.
123rd National Meeting of the Am. Chem. Soc., Los Angeles, Cali-

fornia.

Induced Precipitation

E. G. BOWEN

Division of Radiophysics, Commonwealth Scien-
tific and Industrial Research Organization, Syd-
ney, Australia

It is assumed that the reader is familiar with the processes of
inducing rain to fall from clouds by introducing materials like dry
ice, water, or silver iodide into the clouds at an appropriate stage
in their development. No attempt will be made to review the
previous history in this field, but an account will be given of two
new developments which may have an important bearing on the
future of weather control. The first of these deals with some new
discoveries in relation to silver iodide seeding, the second with an
unexpected contribution from a field of science normally thought
to be outside the bounds of cloud physics.

Silver lodide Seeding

A vast number of seeding operations have been carried out in
the past few years in which silver iodide has been dispensed into
the atmosphere from smoke generators on the ground. It is safe
to say that the net result of these operations has been to produce
more controversy than they have rainfall, despite the fact that in
the laboratory, silver iodide is unquestionably a highly efficient
freezing nucleus.

Looked at from a physical point of view, and quite apart from
the question of whether there has been any effect on rainfall, two
obvious points require checking: (a2) whether the silver iodide
remains effective as a freezing nucleus when exposed to the at-
mosphere; and (4) whether it can attain the requisite height,
which in summer might be of the order of 15,000 or 20,000 feet.

291

292 THE FUTURE OF ARID LANDS

Distonce Downwind from Smoke Generotor (km)

Figure 1. The concentration of active freezing nuclei downwind from
a silver iodide ground generator.

Until recently these points had never been investigated and it
was therefore decided to study them experimentally. Tests have
now been carried out in Australia by Smith and Heffernan (2) and
by Smith, Heffernan and Seely (3) with the following results. When
a silver iodide generator emitting 10'® nuclei per hour is run in
flat terrain under typical convective conditions, the distribution
of effective nuclei downwind from the generator is as shown in
Figure 1. It is seen that the freezing nucleus concentration drops
to a low level at a distance of 10 to 12 miles downwind and that
appreciable concentrations do not extend above heights of 2,000
or 3,000 feet. The rapid decrease in concentration cannot be
accounted for simply by diffusion of the freezing nuclei and cal-
culation shows that they must be subject to a rapid decay in
activity. Figure 2 gives the decay rate for two typical burners,
showing in one case a decay of 10% times in 30 minutes.

Bolton and Qureshi (1) conducted a separate investigation to
arrive at the physical reasons for this decay and found that the
most important single factor controlling the decay rate is the
ambient temperature of the atmosphere. The rate of decay is in

INDUCED PRECIPITATION 293

Decay factor of silver iodide freezing nuciei

O:OOO!

O:0000I6 30 40 6O 80 100—s«*I2O 140
Duration of exposure of silver iodide agcroso! to the
atmosphere (minutes)

Figure 2. The decay in activity of silver iodide particles from a
hydrogen burner and a kerosene burner respectively.

fact a very steep function of air temperature as shown in Figure
3. While the silver iodide might decay by a factor of 10 in a few
minutes at ambient temperatures between 20° and 30° C, the
decay is only a factor of 10 in several days at a temperature of
Solin 1G,

This result explains the failure of many ground seeding opera-
tions which have been carried out at relatively high temperatures
on flat terrain. What is more important, it suggests the methods
by which silver iodide might be successfully used for producing
effects over wide areas.

294 THE FUTURE OF ARID LANDS

Conclusions

The results above suggest immediately several experiments in
silver iodide seeding which might be of outstanding importance
in the whole problem of weather modification.

1. Silver iodide seeding from aircraft operating at moderate or
at high altitudes. Under these conditions the decay rate of the
silver iodide might be low and the material is already at approxi-
mately the height where it can become effective in clouds.

2. A seeding experiment in which silver iodide is dispensed
from a high mountain peak where the ambient temperature is
near or below o°C.

The results also bring out the importance, in any experiments
of this kind, of independent measurements being made to verify:

lO-Ka +

N N
NS SS Ss
go EON
w~ NN ‘
Ne N SS XS
560 N SS
+ NX \
19) e aN N+
(e
a SN
SOMES
5 Ra
AACE S < N
6 S NeETTES
t oy
s e\A, +
E NEN
< \
20 x
N
aN
» Ne
re XS
L a ft n sui tee
O3 Ke) 3 10 30 100 300

Time for a decay to one-tenth concentration (mins) at
atmespheric pressure
Figure 3. The decay in activity of silver iodide particles as a function
of air temperature.

INDUCED PRECIPITATION 295

(a2) that the silver iodide is being emitted in an active form; (%)
that it diffuses in such a way as to enter appropriate cloud sys-
tems; (c) that it does not suffer a high decay rate on exposure to
the atmosphere.

Influence of Meteoric Dust on Rainfall

During the latter part of 1954 and early 1955 some new results
were obtained from an unexpected direction which may have a
crucial bearing on the whole problem of artificial stimulation of
rainfall.

It has become almost an axiom of meteorology that if in any
given month the values of a meteorological quantity are totaled
for an adequate number of years and over an adequate area, they

§ 3 NANA

€c

‘4

3 I

cw)
10 20 3)
Janvary

Figure 4. The cloudiness measured at 9 A.M. in Sydney for each day
in January over the period 1900-53.

will average out and give a mean which (apart from seasonal
trends) does not vary substantially from one day to the next. A
good example of this is the cloudiness of Sydney for the period
1900-53 for each day in January which is shown in Figure 4.
This has a mean value of 0.33 with variations about this value
of not more than a few per cent.

If the daily rainfall of Sydney for the month of January and
the first few days of February is plotted in the same way, the
day-to-day variations are very much greater and in some cases
show departures of 2 to I on one day as compared with another.
In Figure 5 is given the daily rainfall totals for Sydney for the
period 1859-1952, showing distinct maxima on January 13, 22,
and February 1.

296 THE FUTURE OF ARID LANDS

Total rainfall (inches)
o)

10 20 3I
January February

Figure 5. The total rainfall of Sydney for each day in January over
the period 1859-1952.

One’s first reaction is that these variations are due to statistical
fluctuations. If they arise from purely random causes they would
normally smooth out as records of more and more rain gages are
totaled. As a next step Figure 6 shows the January rainfall for
20 stations in the state of New South Wales, all more than 100
miles from Sydney. The peaks are still present and, furthermore,
they occur on precisely the same dates as those in the Sydney
record. It begins to appear, therefore, that this is a real phenom-
enon and the rainfall in this area and for this particular period

300
G
s 13 oe
£200 ° \
~ ee / 4 |
— A e e Ps
lo} ° \
fe WY \ s Ip Ne i * VA
5
£100 r
o
+
i
[ake [eae eee eee P| vee Si
O 10 20 31
Janvary February

Figure 6. The January rainfall of twenty stations in New South
Wales for the period 1890-1946.

INDUCED PRECIPITATION 297

ul
O

Total of heaviest fall Gins)

January

Figure 7. The January rainfall of fifty stations in New Zealand for
the period 1900-52.

was considerably greater on January 13 and 22 and February 1
than on other days in January.

Turning now to an adjacent geographical region, namely New
Zealand, which is some 1200 miles away, the results of 50 stations
for the period 1900-52 are given in Figure 7. Again there are peaks
in the rainfall which happen in each case to be one day later than
those in New South Wales.

Turning to the northern hemisphere, Figure 8 shows bulk rain-
fall records, again for the month of January, for four widely
separated regions in the northern hemisphere, namely Great
Britain, The United States, Japan, and the Netherlands. These,
too, show characteristics almost identical with those in the south-
ern hemisphere. _

It appears therefore that there must be some worldwide influ-
ence on rainfall which leads to more rain than usual occurring all
over the world on particular calendar dates.

If, in fact, these variations in rainfall are a worldwide phenom-
enon, they cannot have their origin in moving weather systems as
normally conceived of. They must be due to an effect which can
act simultaneously over the whole globe, namely an extraterres-
trial influence. In addition, it must be an influence which is tied
to particular calendar dates. The only extraterrestrial phenomena
known to the author which satisfy these requirements are meteor
showers, the majority of which recur each year on the same dates.
The meteor particles exist in vast elliptical orbits around the sun,

298 THE FUTURE OF ARID LANDS

ce €

> E 1 a4 2

> = 100 I rae ras
2 5 ri es tect ages” :
G £

z D 50 JAPAN

3 32 STATIONS

E Cc 1876-1950 (c)

3 g

z =

10 20 3!
January

ee
@ 150 \ 2 A
5 a e, ee co, | ifs)
& 2 2 gloe ou uunone 3
7) is ro NS of ng oe e\ aN
y 100 We Pe 3 \ af

Ss e e 5 ri) = G WW q / »
£ d 6 50 as vd

USA
ince 48 STATIONS g PSS UCTORS .
= - bd
3 1I8G9- 1950 (b) & 1901-1950 (d)
c |
10 20 3I ie) 20 3i
January Janvary

Figure 8. The total rainfall for four regions in the northern hemi-
sphere.

and the earth passes through these streams year after year as it
revolves in its own orbit.

There is only one prominent meteor shower during the month
of January, namely the Quadrantids on January 3. Going back
into December, however, there are the Ursids on December 22
and the Geminids on December 13. It will be seen at once that
these occur almost exactly 30 days before the world peaks in
rainfall which have already been discussed. The hypothesis has
been advanced, therefore, that the phenomenon is due to the
effects of meteoric dust falling into cloud systems in the lower
atmosphere, the time difference of 30 days being accounted for
by the rate of fall of the dust through the atmosphere.

If this hypothesis is substantiated it might have the most pro-
found effect on our concepts of artificial control of the weather.

INDUCED PRECIPITATION 299

It implies that the fundamental rainmaking process in the atmos-
phere is a seeding process and that the atmosphere is much more
free from rain-forming nuclei than has previously been supposed.

Conclusion

These results suggest that the arrival of dust in the upper
atmosphere and its descent to the ground might turn out to be
one of the most important factors controlling rain formation. It
focuses attention on several broad fields of study which would
be important in this regard:

1. The study of meteors in all their aspects, particularly the
neglected field of meteoric dust.

2. The physics of dust falling through the atmosphere.

3. The properties and distribution of freezing nuclei in the
lower atmosphere.

From the point of view of artificial weather modification, it
suggests: (a) that the effects might be very much greater than
was otherwise supposed; (4) that the most effective seeding opera-
tions might be those which are designed to influence the weather
on a continent or hemisphere wide basis, rather than operations
intended to cover a few square miles of territory only.

REFERENCES

1. Bolton, J. G., and N. A. Qureshi. 1954. The effects of air tempera-
ture and pressure on the decay of silver iodide. Bull. Am. Met.
Soc., 35, 395-399.

2. Smith, E. J., and K. J. Heffernan. 1954. Quart. F. Roy. Meteorol. Soc.,
80, 182-197.

eepomich b) |. Ke - Hefernan, and Ba Ke Seely. 1955. dhe decay,
of ice nucleating properties of silver iodide in the atmosphere.
F. Meteorol., 12, 379-385.

Some Relationships of Experimental
Meteorology to Arid Land
Water Sources

VINCENT J. SCHAEFER

The Munitalp Foundation, Inc.,
Schenectady, New York

Since the fall and early winter of 1946, when our first cloud
seeding experiments were brought to a successful conclusion,
many things have happened to clarify as well as confuse the pic-
ture of the present state of experimental meteorology. With about
eight and a half years of experience behind us, what conclusion
can be drawn in relation to successes or failures?

As in any other new science, advance has been rapid as well as
slow. Startling new discoveries have occasionally emerged from
the less spectacular plodding advance as new facts in atmospheric
physics have become established.

Variation in Concentration of Ice Nuclei

With the 1946 demonstrations that extensive cloud systems not
only exist for considerable periods as supercooled clouds (1) but
may also be profoundly and rapidly modified by relatively small
amounts of dry ice fragments (23), the question arose as to how
variable the concentration of ice nuclei was in the natural atmos-
phere. To gather statistical data in this regard three hourly
observations of the concentration of ice nuclei were inaugurated
at the Mt. Washington Observatory. These studies have been
continued from January 1, 1948 to date. The more than 18,000
observations show that natural ice nuclei concentration may vary

300

EXPERIMENTAL METEOROLOGY 301

by a factor of a million (25); there are periods of a half day to
several days when values are relatively high, but there are other
days when the concentrations are very low. During the six and a
half years of record, a trend is noticeable toward higher concen-
trations of moderate counts (§ X 10? to 1 X 104 per cubic meter)
with lower levels of both high and low counts. It is not easy to
determine whether this trend is caused by increased cloud seeding
activities or, what is more likely, an increase in area of the drought
stricken regions of the Southwest United States. Air trajectory
studies have shown a high correlation between the high counts at
Mt. Washington and dust storms and related air movements
from the Southwest.

Importance of Ice Nuclei in Controlling Precipitation

In some of the early flight studies of Project Cirrus, particularly
during February 1948 when a series of exploratory flights were
conducted over and near Puerto Rico, we were able to determine
conclusively that ice crystals were not unique in causing heavy
precipitation (24). Subsequent studies by Langmuir (13), Wood-
cock (32), Bowen (5), Mordy (20), d’Albe (1), and others have
demonstrated that quite often in the subtropical regions of the
world the precipitation cycle may be controlled, if not completely
dominated by the presence or absence of large salt nuclei. The
presence of the “‘trade wind inversion,” which normally limits the
vertical development of clouds to levels which never cool below
o°C., together with the relatively long life cycles of individual
cloud elements, is conducive to the development of this ‘‘warm
cloud” rain. In almost all cases, however, there are exceptions to
the rule. Large cumulus clouds break through the inversion or
develop when the inversion is absent, reach high altitudes, and
produce excessive rain.

Thus, although it is likely that a certain concentration of large
salt nuclei may initiate the precipitation cycle even in such large
supercooled clouds, there is no reason to believe that the precipi-
tation pattern will be dominated by the salt particle effect if
optimum concentrations of ice nuclei are present. Much further
research remains to be done in separating these two distinctly
different effects.

302 THE FUTURE OF ARID LANDS

Since it is such cloud systems which produce the Kona Storms
of Hawaii, the ‘‘blow down” storms of Central America, the hurri-
canes of the Caribbean and Atlantic seaboard regions, and the
major rain-producing storms of regions like the southwestern
United States, it is of great importance to learn the techniques
for “taking away”’ the effects of natural atmospheric nuclei and
replacing them with controlled amounts of more dominant arti-
ficially introduced nuclei. Until we learn how to do this effectively,
any talk of weather control is unrealistic.

Importance of Ice Nuclei in Regions Like New Mexico

Starting in 1948 and continuing for three years, cloud-seeding
studies were conducted in New Mexico by Project Cirrus. The
flight studies were primarily of an exploratory nature and were
not planned to attempt to produce economical amounts of rain.
That some of the seeding flights were followed by substantial
amounts of precipitation in the seeded regions was incidental to
the primary objective, which was to establish the nature of the
clouds and the effects that might be induced by using various
amounts and varieties of seeding materials (16).

For the New Mexico studies dry ice, silver iodide and water
ice, water and gaseous ammonia were used. A total of twenty
seeding experiments were conducted in 1948, 1949, and 1950 (27).
Since these were experimental studies, the experiences resulting
from each flight were used to modify each subsequent flight study.
Based on the results obtained, the following general conclusions
were established.

1. Cumulus clouds in New Mexico commonly contain large
amounts of supercooled cloud masses.

2. The bases of such clouds under summer conditions vary
between 12,000 and 14,000 feet msl.

3. The freezing level commonly occurs at 16,000 feet.

4. The most spectacular effects of seeding with dry ice or silver
iodide occurred on days when the first clouds appeared over the
cloud breeding spots not later than 1000-1100 o’clock.

5. The most striking seeding effects by aircraft occurred in
clouds which were growing vigorously and were seeded when their

EXPERIMENTAL METEOROLOGY 303

TABLE 1
Flight Seeding

No. Date Agent Results

45 October 1948 CO./AgI Development of large storm.

106 July 1949 CO; Removal of large cloud from line of
cumulus

108 July 1949 CO, Towering, followed by consolidated rain-
storm.

Ke) July 1949 AgICO, Development of large storm.

168 July 1950 CO, Development of cirrus overcast.

172 July 1950 CO, Towering, followed by consolidated rain-
storm.

total vertical thickness was between 8,000 and g,0o0o feet, and
thickness of supercooled cloud was 4,000 to 5,000 feet. Seeding
was effected at the top of the clouds with temperatures of about
—12°C. Crushed dry ice fragments with no particles larger than
I cubic centimeter were scattered through the top portion ot the
cloud at a rate of about 2 pounds per mile.

6. Under these conditions initial radar echoes followed the
seeding operation by 14 to 21 minutes. The echoes invariably
brightened rapidly and showed close relationship to the upper air
motions related to the areas seeded.

7. Heavy rains occurring with a sparsity or complete absence
of lightning were the most striking effects observed after dry ice
seeding. Observations of the appearance of the cloud tops asso-
ciated with the experiments showed regions where spectacular
towering occurred, followed by a rounding off and consolidation
of the ice crystal tops, especially where rains persisted for 2 to 4
hours.

8. The largest effects followed combined seeding of dry ice and
silver iodide with the latter introduced either by airborne or
ground-based generators.

g. When cumulus cloud developments started in the afternoon,
cloud dissipation was the common result of seeding activities.

10. Large cumulus clouds which do not develop to the precipi-
tation stage are of relatively common occurrence in regions like
New Mexico and Arizona. When precipitation fails to occur,

304 THE FUTURE OF ARID LANDS

much of the cloud shifts to false cirrus streamers of ice crystals,
which often dominate the afternoon and evening sky.

11. During periods of dust storm activity, clouds are commonly
seen to shift completely to ice crystals when still quite small. At
other times clouds in one region of the visible sky will show
profound modification, while clouds in another region will develop
in a normal manner. In some instances these effects are probably
due to tongues of air containing natural ice nuclei. In some cases
such effects may be attributed to the effect of silver iodide ground
generators.

Precipitation Pattern Due to Orographic Clouds in New Mexico

Since the bases of most large cumulus clouds that form in New
Mexico are at altitudes several thousand feet above the highest
mountain summits, naturally occurring precipitation generally
falls on the down wind side of the mountains. The cultivated areas
and range lands which depend on this natural rainfall delineate
this precipitation pattern in considerable detail. An excellent
example of this effect is to be seen in the bean-growing area of
New Mexico, situated east of the Sandia and Manzano Moun-
tains, to the southeast of Albuquerque. In direct contrast, the
orographic effects of mountains in subtropical regions and in
places where the mountain summits rise higher than the cloud
bases normally produce precipitation patterns on the upwind side
of the mountain slopes. In such regions the area on the lee side
of the mountains often shows regions of rainfall deficiency and
actual ‘“‘rain shadows.’’ Good examples of this pattern are to be
seen in the rain forest west of Hilo in the Hawaiian Islands and
the rain forest on the western slopes of the Olympic Mountains
in northwestern Washington.

If methods are developed which effectively modify the “nat-
ural” rainfall from orographic cumulus in places like New Mexico,
attention should be directed to this phenomenon since the end
result might cause trouble. If the precipitation pattern from
seeded clouds tends to appear further upwind than the ‘‘natural”
rain, the end result might be of little benefit to anyone. This
danger may occur in any region of marginal rainfall.

EXPERIMENTAL METEOROLOGY 305

Based on the few seeding experiments made with Project Cirrus
the rain areas developed over the regions which ordinarily receive
rain. Thus, if any conclusions might be drawn from these few
experiments, the type of seeding we conducted might be expected
to produce an augmented total yearly rainfall rather than a
redistribution of precipitation.

Studies of this sort should become an essential part of any long
range plan of arid lands research concerning atmospheric water
sources.

By proper control it might be feasible, for example, to aug-
ment ground water supplies by causing more precipitation on
mountain slopes with drainages running into thick gravel depos-
its.

Movement of Air in Valleys Bordered by Mountains

One of the problems involved in seeding clouds from ground-
based silver iodide ground generators is that of getting the silver
iodide particles into the cloud to be modified. Several significant
studies of this phenomenon were made southeast of Albuquerque
during July 1949.

On July 14, 1949, an attempt was made by S. E. Reynolds and
the writer to introduce dry ice into an orographic cumulus, using
a large pilot balloon. The target cloud was a relatively small but
vigorously growing orographic cumulus above the western edge
of the Manzano Mountains about five miles east of the launching
site. Upon release, the balloon carrying about a half pound chunk
of dry ice started moving in a southerly direction and was soon
lost to view. Reynolds and Schaefer then decided to get closer to
the cloud base, so took another balloon with dry ice over to the
edge of the mountain nearly under the cloud base. Upon release,
the balloon, observed with binoculars, immediately started rising
toward the cloud without deviation. Just before this second bal-
loon entered the cloud base, the first balloon was seen to join it,
drifting up rapidly from the south!

Within less than thirty minutes after the balloons entered the
cloud a radar echo due to precipitation formed, which was then

followed by rain.

306 THE FUTURE OF ARID LANDS

A day and night study of air movements in the Rio Grande
Valley of New Mexico during the month of July showed that
mountain, valley, and drainage winds fit into a recurring diurnal
pattern. These air motions may be studied by watching smoke
plumes, smog patterns, zero lift balloons, and ordinary pilot
balloons.

The early morning situation is dominated by a drainage wind
flowing down the river valley from the north. This continues until
the mountain slopes become sufficiently heated to start convective
movements, at which time the wind swings to a westerly and then
a southwesterly direction. This wind then stops blowing in the
evening. It is then followed by an easterly mountain wind which
carries the cold air produced by radiative cooling down into the
valley. These air movements are of sufficient magnitude to impress
their pattern on a microbarograph. These traces are the most con-
vincing evidence of the recurrent diurnal fluctuations in air flow of
the area.

On July 21, 1949, B. Vonnegut of the Project Cirrus group
started a silver iodide smoke generator at 0530 and at the same
time began releasing a series of zero lift balloons at the research
station south east of Albuquerque. The air flow followed the de-
scribed pattern. At about 0830, an isolated orographic cumulus
cloud started forming about 25 miles south of the station. Visual
and photographic observations of this cloud showed that its top
grew at the average rate of 160 feet per minute between 0830 and
0957. At 0957, the summit of the cloud was at 26,000 feet and its
temperature about —23°C. At that time the top of the cloud
started growing at the rate of 1,200 feet per minute. This rapid
rate of growth continued for fifteen minutes. At 1012 it had
reached an altitude of 44,000 feet, with a calculated temperature
of —65°C. At this altitude, the rate of growth slowed down
considerably.

The first radar echo was observed at 1006 at 20,500 feet, where
the temperature was —9°C. This echo occurred at 25 miles and
an azimuth of 165°C. With the generator located at 5,600 feet
msl, the smoke plume from the silver iodide generator moved
toward the Manzano Mountains in such a manner that it was

EXPERIMENTAL METEOROLOGY 307

probably close to the cloud base at 0800. Observations of visible
smoke plumes in the Rio Grande Valley show that, while there is
a tendency for an early morning inversion to stabilize the air, this
condition rapidly changes as the sun begins heating the valley
floor. Together with the orographic effect of the mountains, it is
very likely that the stability of the air in the early morning would
have prevented a rapid dilution of the stream of silver iodide, so
that an effective seeding reaction probably occurred just above
18,500 feet, where the temperature was colder than —4°C. Assum-
ing a rise within the cloud of 120 feet per minute, this would have
the silver iodide producing an effect at about og4o.

Our other experiences in New Mexico have shown that seeding
effects may be expected to produce an initial radar echo within
15 to 25 minutes. Thus, the initial radar echo at 1006 would be in
very good agreement with the mechanisms that might be expected
to control the precipitation cycle. The initial precipitation area
covered about one square mile and was deep within the cloud. It
is very unusual for precipitation to develop at such low altitudes
in New Mexico. Within four minutes the precipitation echo had
increased to seven square miles, and two minutes later it had ex-
tended upward to 34,000 feet, where the temperature was — 43°C.
The rapid vertical growth with an average of 1,160 feet per minute,
which was first observed at 0957, continued until to12. The out-
ward manifestation of this upheaval shows a remarkable linearity.
Extrapolation of this line shows an excellent relation of the
probable triggering effect of the silver iodide.

This single local storm developed continuously into a large zone
of precipitation.

In the early afternoon, Flight 110 was activated with two Proj-
ect Cirrus B-17’s. Twelve separate seeding operations were carried
out between 1445 and 1530, utilizing from one-third to four and
a half pounds of crushed dry ice, depending on the size of the cloud
masses seeded. During the afternoon, 1.2 inches of rain fell at the
radar site. A remarkable fan-shaped pattern of precipitation
developed in the region downwind from the seeded region. This
has been analyzed in considerable detail by Langmuir (15), who
concluded that much of the rain that fell in New Mexico on July

308 THE FUTURE OF ARID LANDS

21 could be related to the silver iodide and dry ice seeding opera-
tions. The writer flew into this storm in the early afternoon and
observed that it consisted of a great mass of cumulus with heavy
precipitation moving in a northeasterly direction.

These observations present strong evidence that the convective
clouds, which daily form over the mountains in the Rio Grande
Valley in the summer, tend to serve as chimneys which lift the
valley air to higher levels of the atmosphere. Silver iodide or any
other material introduced into the valley region is carried by the
converging air and effectively introduced into cloudy regions
where they may affect the cloud structure, if still active as either
condensation or ice nuclei, or both.

Effectiveness of Silver lodide as a Nucleus for Ice Crystal For-
mation

A great deal has been written about the role of silver iodide as
an ice crystal nucleus. Although much research has been conducted
since Vonnegut’s original paper (30), silver iodide still remains
the most effective foreign particle for ice crystal nucleation. Many
different factors have been shown (4, IT, 21, 31) to affect its poten-
tial activity. It has such complex crystal habits, however, that
much caution is required in concluding that all or any of these
factors cause irreversible effects. Recent studies by the writer
show that silver iodide is most effective if it first serves as a con-
densation nucleus in the liquid water zone of a cloud (26, 29). If
this occurs, the particles serve as freezing nuclei at a threshold
temperature between —4° and —s°C. If the silver iodide particle
enters a liquid water droplet, there is the possibility that the effects
of sunlight, low humidity, temperature, certain impurities, and
other factors which have been cited for their deactivation effects
may be nullified. The active program of laboratory study and
field evaluation currently underway in many parts of the world
may be expected to shed much light on the properties of this im-
portant cloud seeding material within the next few years.

Use of Silver lodide for Increase of Snow Pack
One of the most important aspects of water supply in regions

of submarginal rainfall is the snow pack of the high mountains.
With improved methods of moving water by pipe line and canal,

EXPERIMENTAL METEOROLOGY 309

it is often possible to have this source of water hundreds of miles
from the arid regions where it is to be used.

Seeding operations in high mountainous areas have excellent
potentialities. Since most mountains rising above 8,000 to 10,000
feet msl are higher than the dew point level of winter air masses,
the upper levels are commonly covered with clouds. These are
often supercooled and, except where seeded by blowing snow near
the surface, are deficient in ice nuclei. Silver todide generators
placed at convenient locations are often operated with ambient
temperatures close to or even colder than o°C. Under such con-
ditions deactivation by the temperature effect is of little or no
concern.

Wintertime orographic clouds rarely have the vertical thickness
or lifetime of summer clouds. They occur for longer periods, how-
ever, and if effectively seeded may be forced to yield much of the
condensed cloud water in the form of snow crystals.

Although considerable progress has been made in the study of
snow pack increasing (10, 12), much remains to be done. By con-
trolling the particle size of silver iodide, it should be feasible to
control the temperature of nucleation. The smaller the particle,
the colder the threshold temperature of nucleation. Thus, it may
be possible to move the lower level of the snow pack area by con-
trolling either or both the concentration of ice crystals and the
temperature at which they form. By producing silver iodide par-
ticles of highly uniform size, the effectiveness and efficiency of
smoke generators might be greatly improved in the same manner
as with our artificial fog generators developed during World War
II. This may be achieved by effective control of the quenching
speed of the jet of vaporized silver iodide. By delaying the intro-
duction of cold air into the vaporized stream, the smoke particles
will grow much larger, but they will be less numerous than when
the vapor is rapidly diluted. It should be feasible to form mono-
dispersed particles over the size range of 0.004 to 1.0 micron diam-
eter. Thus, under nearly the same operating conditions the number
of particles may be varied by a factor of 100 million.

This is a field of research in experimental meteorology which has
hardly been touched but it may yield rich dividends in commercial
operations as well as basic research.

310 THE FUTURE OF ARID LANDS

Need of Basic Research in Atmospheric Physics

Much basic work remains to be done in the exploratory phases
of experimental meteorology. Despite the extensive and wide-
spread commercial cloud seeding which has been underway for
the past 4 to § years, certain aspects of the problem have been
seriously neglected.

Perhaps the simplest and yet one of the most neglected fields
has been the cloud survey. Despite some localized studies (3, 7, 17,
22) great gaps remain in our knowledge of these fundamental
features of the lower atmosphere. There has been a tendency in
the past to assume that cloud behavior was similar in most places.
However, a little study in the field of what may be called ‘‘com-
parative cloud structure” quickly demonstrates the danger of
such extrapolation.

While it is possible to discover similarities in the clouds which
form near and over the Hawaiian Islands and those of Puerto Rico,
a cumulus forming in such trade wind regions is far different in
properties, appearance, and precipitation mechanism from one
over a mountain bordering the Rio Grande of New Mexico. In
turn these two will have characteristics entirely different from a
cumulus forming near the Priest River Valley of northern Idaho.

Fortunately, cloud surveys are becoming more common. Recent
studies in the midwest (g), Puerto Rico (8), Sweden (1g), and the
northwestern United States (2) have established techniques
which may be adapted to local conditions and available personnel
and equipment. These procedures vary from exact measurements
of many individual clouds to the patterns of cloud systems forming
localized thunderstorms.

Supplementing these local observations is the current prepara-
tion of what might be termed a dynamic cloud atlas. Time lapse
cloud movies are being obtained in many parts of the world under
sponsorship of our Foundation (28).

A widespread photographic coverage of cloud systems in the
United States and its surroundings is underway through the co-
operation of such groups as the U.S. Forest Service, the National
Park Service, various universities, the Boeing Airplane Co.,
private industrial meteorological organizations, the U. S. Weather
Bureau, and cooperative individuals. In addition, similar cloud

EXPERIMENTAL METEOROLOGY 311

movies are being obtained through the cooperation of the Air
Weather Service of the USAF in such scattered places as Sweden,
England, Germany, Italy, Japan, Hong Kong, and Bermuda.
From the more than 30,000 feet of film now on hand, a series
of movies will be prepared to illustrate the variations and similari-
ties which occur in clouds that form under a wide variety of geo-
graphic, topographic, and climatologic situations. Such films wil!
be made available for educational purposes by the Foundation.

Future Prospects of Increased Water Supplies from the Atmos-
phere

Although it must be admitted that there is no currently avail-
able easy solution to the tapping of unlimited water from the sky
rivers which flow over arid lands, there are still many research
angles to probe. Under certain physical conditions such as have
been described in this paper, it is fairly certain that some addi-
tional water may be secured above that subject to the variations
of natural conditions.

The future possibilities of securing additional water supplies in
arid regions from atmospheric sources depend on several factors.

The most immediate may be expected from improvements in
current techniques of cloud-seeding activities. The most remote 1s
the natural variation in the general world circulation, which in
the past produced so much rain and snow as to cause the great
alluvial deposits that cover the Rio Grande Valley hundreds of
feetaeep)

The most important in the foreseeable future is the promise
indicated by the studies and hypotheses of Langmuir (18) and
Bowen (6), which suggest that optimum concentrations of effec-
tive ice nuclei may play a dominant role in controlling hemispheric
weather. Although their findings are still the subject of consider-
able controversy, it seems reasonable to expect that a proper
understanding of trigger effects in the atmosphere, which may
cause singularities, abnormal weather patterns, extensive storms,
and the persistencies, which lead to drought or flood, will eventu-
ally lead to a certain degree of weather control.

The time which will be required to reach this goal will depend
to a major degree on basic research and the vision, enthusiasm,

S12 THE FUTURE OF ARID LANDS

curiosity, and imagination of a few good research scientists, aided
by the cooperation and good will of a much larger segment of inter-
ested people. Groups such as those comprising the Arid Lands
Conference are in a vital position to be of much help in this regard.

REFERENCES

1. d’Albe, Fournier. 1955. Giant hygroscopic nuclei in the atmosphere
and their role in the formation of rain and hail. drch. Meteorol.
Geophys. Broklimat.
2. Barrows, J. S., V. J. Schaefer, and P. B. MacCready. Project Sky-
fire: a progress report on lightning fire and atmospheric research.
Research Paper 335, Intermountain Forest & Range Experiment
Station.
3. Battan, L. J. 1953. Observation of the formation and spread of pre-
cipitation in convective clouds. 7. Mereoro/. 10, 311.
4. Bolton, J. G., and N. A. Qureski. 1954. The effects of air tempera-
ture and pressure on the decay of silver iodide. Bul/. Am. Meteorol.
Soc. 35, 395.
. Bowen, E. G. 1950. The formation of rain by coalescence. 4ustralian
F. Sct. Research A8, 193.
. Bowen, E. G. 1955. Induced precipitation. This volume, pp. 291-299.
» Braham, Jin, ROR S2E: Reynolds, and]: EE tlarrelle neGneraneloud
census in New Mexico. ¥. Meteorol. 3, 416.
8. Byers, H. R. 1955. Geographical differences in cloud populations.
Arch. Meteorol. Geophys. Broklimat.
g. Byers, H. R., R. R. Braham, Jr., and L. G. Battan. 1955. Results of
some randomized seeding experiments in cumulus clouds. Presented
at 136th Natl. Meeting. 4m. Meteorol. Soc., Washington, D. C.

to. Elliott, R. D., and R. F. Strickler. 1954. Analysis of results of a
group of cloud seeding projects in Pacific slope watershed areas.
Bull. Am. Meteorol. Soc. 35, 171.

11. Inn, E. C. Y. 1951. Photolytic inactivation of ice forming silver
iodide nuclei. Bull. dm. Meteorol. Soc. 32, 132.

12. Krick, I. 1954. Progress in weather control. 7. dm. Water Works
Assoc. 46, 803.

13. Langmuir, I. 1948. The production of rain by a chain reaction in
cumulus clouds at temperatures above freezing. 7. Meteorol. 5, 175.

14. Langmuir, I. 1948. The growth of particles in smokes and clouds and
the production of snow from supercooled clouds. Proc. dm. Phil.
Soc. 92, 167.

15. Langmuir, I. 1950. Progress in cloud modification by Project Cirrus.
Occ. Rep. No. 21, Project Cirrus Rept. No. RL 357, General Electric
Research Laboratory, Schenectady.

Sy ey

16.

18.

1g.
IXO)-

Die

DB.

24.

Oke
26.

OFT
28.
29.
20:

Bit

ies)
iw)

EXPERIMENTAL METEOROLOGY 313

Langmuir, I. 1950. Results of the seeding of clouds in New Mexico.
Occ. Rept. No. 24, Project Cirrus Rept. No. RL. 364, General Electric
Research Laboratory, Schenectady.

Langmuir, I. 1gg0. Studies of tropical clouds. Occ. Rept. No. 25,
Project Cirrus, Rept. No. RL 365. General Electric Research Lab-
oratory, Schenectady.

Langmuir, I. 1953. Final report project Cirrus: Part II. Analysis of
the effects of periodic seeding of the atmosphere with silver iodide.
Rept. RL 785, General Electric Research Laboratory, Schenectady.
Ludlam, F. H. 1955. The problem of planning and assessing seeding
experiments. Arch. Meteorol. Geophys. Broklimat.

Mordy, W. A., and L. E. Eber. 1954. Observations of rainfall from
warm clouds. Quart. F. Roy. Meteorol. Soc. 80, 48.

Reynolds, D. E., B. Vonnegut, e¢ a/. 1951. Effect of sunlight on the
action of silver iodide as sublimation nuclei. Bull. dm. Meteorol. Soc.
32, 47.

Schaefer, V. J. 1948. The possibility of modifying lightning storms
in the Northern Rockies. Station Paper No. 19, Northern Rocky
Mt. Forest & Range Experiment Station. Missoula, Mont.
Schaefer, V. J. 1948. The natural and artificial formation of snow in
the atmosphere. Trans. dm. Geophys. Union. 29, 492.

Schaefer, V. J. 1949. Report on cloud studies in Puerto Rico. Occ.
ep. No, 12, Project Cingus Rept. No. RIE 190 General Blectric
Research Laboratory, Schenectady.

Schaefer, V. J. 1950. The occurrence of ice crystal nuclei in the free
atmosphere. Proc. First Natl. dir Pollution Symp. 1, 26. (Pasadena.)
Schaefer, V. J. 1952. Formation of ice crystals in ordinary and nuclei
free air. Ind. Eng. Chem. 44, 1300.

Schaefer, V. J. 1953. Final Report Project Cirrus: Part I. Labora-
tory, Field and Flight Studies. Rept. No. RL 785, General Electric
Laboratory, Schenectady.

Schaefer, V. J. 1953. Cloud photography project. Weatherwise 6, 72.
Schaefer, V. J. 1954. Silver and lead iodides as ice crystal nuclei.
JF. Meteorol. 11, 417.

Vonnegut, B. 1947. The nucleation of ice formation by silver iodide.
F. Applied Phys. 18, 593.

Vonnegut, B., and R. Neubauer. Recent experiments on the effect
of ultraviolet light on silver iodide nuclei. Bul/. dm. Meteorol. Soc.
32, 356.

. Woodcock, A. 1949. Sampling atmospheric sea salt nuclei over the

ocean, 7. Marine Research (Sears Foundation) 8, No. 2177.

Some Problems in Utilizing
Water Resources of the Air

GLENN W. BRIER

United States Weather Bureau,
Washington, D. C.

Two pioneers (pp. 291 and 300) in the field of weather modifica-
tion have discussed some of the problems that lie ahead in man’s
attempt to extend his control over nature. In view of the wide
differences of opinion that still exist on the question of “‘rain-
making,” it may appear to some that little or no progress has been
made. However, progress has been made—at least in the sense of
finding out what some of the most important problems are and
what areas need further exploration. There seems to be general
agreement that more facts and more data are needed. And the
necessity and importance of getting them was brought out at the
Mid-Century Conference on Resources for the Future. Meteorolo-
gist Athelstan Spilhaus stated, ‘“To summarize in one sentence:
If the control of the sea and the air and the useful things in them
is kept clearly as the objective in this search, it will then be an
objective large enough and with such a tremendous pay-off that
it should attract the best scientific minds’”’ (8).

It will be the purpose of this paper to discuss some problems
connected with the collection and interpretation of the facts which
are needed before an intelligent answer can be given regarding the
potentialities of weather control as a solution to the arid lands
water problem.

Survey of Moisture Resources Available in the Atmosphere

Precipitation is the end product of a large number of events on
both the macro and micro scale. The amount of precipitation de-

314

UTILIZING WATER RESOURCES OF AIR 315

pends upon the water content of the air. It would therefore seem
reasonable to determine for various regions the available supply
of moisture, primarily in the form of clouds, since they represent
the most likely source material for rain. The frequency of occur-
rence of clouds of various types and thickness, water content, etc.,
could be determined and estimates made of possible precipitation
increase providing the moisture could be made to precipitate out.
“Census” counts of clouds precipitating naturally could be made.
In some areas, studies like these have been started and the World
Meteorological Organization Technical Report No. 1 ‘Artificial
Inducement of Precipitation with special reference to the arid
and semi-arid regions of the world” (12) describes a preliminary
survey reviewing the general atmospheric conditions necessary
for cloud seeding operations and the likelihood of finding such
conditions in the arid and semi-arid areas. However, in many areas
the basic meteorological data are inadequate. To fill these gaps
it will be necessary to find economical ways of extending the ob-
servational network to obtain the information that will have
direct bearing on the particular questions to be answered.

Distribution of Condensation and Freezing Nuclei in Time and
Space

In addition to the usual meteorological observations, there is a
need for more data regarding the nature and distribution of the
particles in the atmosphere on which water droplets and ice crys-
tals form. Studies by the previous speakers and others indicate
there are large fluctuations in the number of these particles from
time to time and from one location to another. Aside from the
problems of instrumental errors, important sampling questions
arise here regarding the representativeness of the measurements.
How large a sample is necessary to represent adequately the aver-
age concentration over an area? These and other questions might
be answered by sampling experiments and statistical theory.

Regarding the nature and source of these particles, a Japanese
investigator (3) identified the central nucleus of natural ice crys-
tals as belonging to the clay mineral group in about ten out of
fifteen individual cases studied. Is this typical, or would another
sample taken at a different time or place show something else?

316 THE FUTURE OF ARID LANDS

Schaefer (5, 6) reported that some soil and dust particles have
nucleating properties. Bowen (1) suggested a relationship between
rainfall and variations in meteoritic dust although the evidence on
this point is circumstantial in nature as pointed out by Neumann
(4) and Swinbank (g). Without question, more direct measure-
ments of the quantities concerned are needed before most of the
theories proposed can be confirmed or rejected.

Collecting such data is expensive and no program for indis-
criminate collection of such data is recommended. Initial efforts
are likely to be confined to those areas where the role of nuclei in
the rain-making process seems to be of greatest importance or
interest for one reason or another. As more data become available
and are examined, unexpected relationships will be uncovered,
some of which will be fortuitous and therefore misleading. There-
fore a program for collecting the data should also include plans
for objective and unbiased evaluation of the data, but these plans
must be flexible enough to permit promising clues to be recognized
and followed-up effectively. The methods of sequential analysis
used successfully in other scientific fields might be used here. Maxi-
mum effectiveness in this program requires the coordinated efforts
of the climatologist, meteorologist, cloud physicist, chemist, hy-
drologist, statistician, and astronomer.

Analysis and Interpretation of Field Data

If the collection of the field data is supplemented by laboratory
experiments and studies of physical theories, it should be possible
to estimate the relative importance of vertical motion and the
supply of condensation and freezing nuclei in the production of
natural rain. Data on the natural variation of these particles could
be correlated with rainfall occurrence, cloud structure, and electri-
cal effects. These data might shed new light on Bowen’s meteoritic
dust hypothesis or reveal other important effects that influence
precipitation. For example, the role played by atmospheric elec-
tricity is not understood, but other pioneers in cloud modification
including Gunn (2) of the Weather Bureau and Vonnegut (11) are
giving careful attention to this subject.

Because of the absence of a complete physical theory on the

UTILIZING WATER RESOURCES OF AIR 317

formation of rain and necessary supporting data, most evaluations
on rain-making to date have been statistical in nature and insensi-
tive in detecting changes of less than about 20%. When used
properly these methods can be powerful tools in extracting infor-
mation from the data, but they cannot make good data out of poor
data. It is unfortunate that nearly all the rain-making evaluations
have been made on commercial cloud-seeding operations, which
are not primarily designed as experiments for the purpose of ob-
taining scientific information.

Data on the natural distribution of condensation and freezing
nuclei should also be useful to those engaged in commercial cloud
seeding. As mentioned earlier, there is evidence that the natural
supply of nuclei is variable (7). At the present time most commer-
cial cloud seeders operate on the assumption that the natural
supply is inadequate, and attempt to overcome this deficit by
adding more. Suggestions have been made that adding too many
artificial nuclei to the atmosphere may result in “‘overseeding”’
and inhibit rainfall. In view of the possibly large fluctuations in
the concentration of natural nuclei due to variations of wind blown
dust, volcanic ash, meteoritic dust, sea salt, etc., it may be reason-
able to ask how Nature knows the optimum concentration for
rain and how she prevents ‘‘overseeding.”’ Obviously this leads
one to speculate on the interesting possibility that there may be
some situations in which rainfall would be increased by removing
nuclei from the air.

The data discussed above should also be useful in determining
the economic value of weather modification in the arid lands. The
monetary value of a rain-increasing project will depend upon the
additional water resources available in the air and the cost of
putting some of this water on the ground. If this cost is relatively
low and if there is a reasonable chance of success, an intensive
effort is justified. In this country, President Eisenhower has ap-
pointed an Advisory Committee on Weather Control which plans
to give particular attention to the economic implications and
scientific basis of weather modification in their report, which is
expected to be available in 1956. On the international level, the
World Meteorological Organization’s Committee on Aerology (10)

318 THE FUTURE OF ARID LANDS

has prepared a report on the possibilities of artificial control of
weather.

Conclusion

Time does not permit a discussion of numerous other problems.
The large number of questions that remain unanswered may sug-
gest that little progress has been made. In relation to the total
task to be accomplished, progress has been small, but in relation
to the effort that has been expended, progress has been satisfactory
or rapid. Progress has been overshadowed by wild speculations,
unsupported claims, and extreme statements which have raised
false hopes. It is clear that the immediate solution to weather
modification is not in sight, but from a long range point of view
there is no reason to be pessimistic. Man is gradually learning to
control such things as the atom and cancer, and there is no good
reason to think that weather is beyond his reach. The failure to
achieve satisfactory control by presently known methods does
not prevent other and better methods from being found. At one
time it was thought that there was one precipitation mechanism.
Now several other methods are being considered. In other words,
there is an indication that, if Nature does not make it rain one
way, she provides another. Perhaps it will be the same for man.
If we do not control the weather in one way, we will ultimately
control it in some other way. This will require greater discoveries
which will depend upon our vision and willingness to approach
the problem with an open mind.

REFERENCES

1. Bowen, E. G. 1953. The influence of meteoritic dust on rainfall.
Australian F. Phys. A. 6, 490.

. Gunn, Ross. 1954. Electric field regeneration in thunderstorms.

Gio WIGHEORO, Vil INI@; 2; WRSHUBe.

Tsono, K. 1955. On ice-crystal nuclei and other substances found in

snow crystals. 7. Meteorol. 12, No. 5, 456-462.

4. Neumann, J. 1954. Fluctuations of long period accumulations of
daily rainfall amounts. Australian F. Phys. T, 522-526.

5. Schaefer, Vincent J. 1950. Experimental meteorology. 7. Applied
Math. Phys. 1, 153-236.

i=)

ies)

Io.

1192.

UTILIZING WATER RESOURCES OF AIR 319

. Schaefer, Vincent J. 1951. Snow and its relationship to experi-

mental meteorology. Compendium of Meteorol. 221-234.

. Schaefer, Vincent J. 1954. The concentration of ice nuclei in air

passing the summit of Mt. Washington. Bull. dm. Meteorol. Soc.
Bop NOs gp SIO. 34.

. Spilhaus, Athelstan. 1953. ‘“The way ahead for research in sea and

air resources” in The Nation Looks at Its Resources. Report of the
Mid-Century Conference on Resources for the Future, Washington,

Da CeeDecemben 2-4, 19535 Pagel gio.

. Swinbank, N. C. 1954. Comments on a paper by E. G. Bowen.

See reference 1, pp. 354-358.
United Nations. 1954. Rainmaking: A study of experiments. De-
partment of Public Information, New York.

. Vonnegut, Bernard, 1955. Possible relationships between thunder-

storms, electrification, and rain. 135th National Meeting of the
American Meteorological Society, New York, January 24-27, 1955.
World Meteorological Organization. 1954. Artificial inducement of
precipitation with special reference to the arid and semi-arid regions
of the world. Technical Report No. 1, Geneva.

The Economics of Water Sources

LOUIS KOENIG

Southwest Research Institute,
San Antonio, Texas

I have been asked to make a statement, in the nature of an
afterthought at a natural science conference, on the economics of
water sources, and, in a general way, on the economics of arid
lands. One of the characteristic distinctions between the natural
and the social sciences is that the former have a superabundance
of facts and data, whereas the latter have a notable deficiency of
these. Since facts and data are frequently, if begrudgingly, capable
of cutting short and bringing to an untimely end an otherwise
healthy discussion, it is not surprising that natural scientists are
noted for their taciturnity, while social scientists are blessed with
a superabundance of talk. This no doubt accounts for the time
limit which has been placed on my remarks, as well as the fact
that a natural scientist has been chosen to talk on economics,
possibly with the hope that he would stumble into some facts that
would bring his discussion to a timely end.

W ater Economics
Attention has been given to the prospects of additional water

resources: artificial precipitation, demineralization, and water
re-use.

Artificial Precipitation

A diligent search for economic data on artificial precipitation
leads only to the conclusion that economic statements on artificial
precipitation cannot be made at this time. One can cite several
individual cases, for example that of Shreveport, Louisiana, which

320

ECONOMICS OF WATER SOURCES 321

recently obtained a year’s water supply after 5 years of drought
for a $20,000 contract with a commercial cloud seeder. The great
majority of the people of that city no doubt feel that $20,000 next
year, and the next, will keep them in water. The economist’s con-
clusions, however, will wait the adding up of all the $20,000’s that
have been or can be spent without success, which in effect is what
the President’s Advisory Committee on Weather Control and the
World Meteorological Organization’s Committee on Aerology are
now engaged in.

Demineralization

For demineralization the picture is much better.

1. Competitive Prices. To answer the economist’s first question—
competitive prices. (I must apologize to our international friends,
that I speak only of United States conditions, and to our Republi-
can friends, that I use the 19§2 dollar.) Bulk municipal water is
priced at $50-80 per acre-foot, irrigation water from the govern-
ment at $2-6, and from private companies up to $40. In marginal
areas or critical situations prices are higher, for example, one
Texas town was hauling water at $1000 per acre-foot. The cost to
the user of pumping underground water from wells is $4-5 per
acre-foot per Ioo feet of lift. In the area studied users are pumping
150-400 feet, thus paying $6-16 per acre-foot, and this sets the
competitive price.

2. Production Costs, the economist’s second question. The de-
mineralization processes on which engineering estimates have been
made (some of them of the most rudimentary sort) may produce
from sea water at costs of from $114 to $1000 per acre-foot and
from brackish waters as low as $20. In detail (all estimates except
as noted):

Dollars/acre-foot

Using solar energy for multiple effect distillation 1069
(It will never get below) 700
Simple solar evaporation 775
(It may get to) 400
Triple effect distillation 700
Vacuum distillation using waste diesel heat 528

Ten effect distillation (actual cost) 415

322 THE FUTURE OF ARID LANDS

Vapor compression distillation 429
Supercritical distillation 148
Vapor compression distillation (Hickman) Te
Electrolytic membrane
885 ppm feed 4
4,635 ppm feed 20
10,000 ppm feed 40

Conclusions. Brackish water conversion can be competitive now
when water costs or supply are marginal. Sea water conservation
is not there yet, but the outlook is hopeful.

3. Transportation Costs, the economist’s third question.

Water should be a cheap commodity, and it is, as chemical com-
modities go. A cheap bulk chemical will sell for about $15 per ton,
that is, $20,000 per acre-foot. Quarried bulk limestone, one of the
cheapest chemical raw materials, will cost $1.20 per ton, that is,
$1650 per acre-foot; salt in brine about 30¢ per ton, that is, $400
per acre-foot. The $40 per acre-foot we are shooting for is equiva-
lent to 3¢ per ton. Think of that. If we had to put a postage stamp
on it to transport a ton it would double the price! And that brings
us to the problem of transportation costs. This subject needs
serious study by engineers. We have two operations that are com-
parable: (a) aqueducting of water under essentially gravity flow,
and (4) pipelining of water and liquid petroleum products.

In the former category are the piped aqueducts such as the
Owens, the Colorado, the Delaware. The Colorado aqueduct de-
livers water to Los Angeles, a distance of 273 miles, for $30 per
acre-foot, about 11¢ per acre-foot mile. Also included are the
canals, which are cheaper, for example, the Gulf Coast Canal is
estimated to deliver water for $8 per acre-foot, about 2¢ per acre-
foot mile. So if we were to transport water by gravity flow, say
so miles from the converting plant at the seashore, or 50 miles
from a brackish source, the transportation cost would be $1 to 5
per acre-foot. It would double the price of present irrigation water,
increase the cost of converted brackish water by 10%, and be
insignificant if one were already using converted sea water at
current estimated costs. i

Unfortunately most transportation of converted water will
not be by gravity flow but will be uphill all the way—there’s no

ECONOMICS OF WATER SOURCES 325

way of damming up the ocean. This means resorting to a pipeline,
pumping against a head of water every foot of the way. This is
much more costly, running $3 to 7 per acre-foot mile. Thus the
s0-mile haul would cost $150 to 350 per acre-foot.

The magnitude of these figures may surprise some who have
been talking only of water production costs. I hope that it will
stimulate some serious engineering study of water transportation
costs Over various typical distances and types of terrain.

4. Markets, the economist’s fourth question. Is there a market
of sufficient magnitude for converted water? As we have just seen,
since the ratio of value to transportation cost is low, the marketing
area for water is limited to the vicinity of the production area.
Pilot studies have been made in certain water-deficient areas, such
as California and Texas 50 miles from the sea and under 500 feet
elevation, showing that an additional 11,000,000 acre-feet could
be consumed and 10,000,000 of present consumption displaced. At
$40 per acre-foot, that is an $800,000,000 business. Only two chem-
ical companies and only thirty manufacturing companies have
larger sales. There is most certainly a market, but the exact extent
and location of it will depend on precise market research to deter-
mine how much water will be used where at successively lower
selling prices.

§. Soctal and Economic Magnitude, the economist’s fifth ques-
tion. How seriously would it affect the present economy? Answer:
Staggeringly. Consider now the world’s largest chemical plant—it
produces 3000 tons per day of product. In the year 2000, one area,
southern Texas, will be able to utilize 12.3 million acre-feet of
water. How many water conversion plants will it take to convert
this water? 10? 100? 1000? No. It will take 15,000 plants the size
of the world’s largest. One plant every 150 feet from the Sabine to
the Rio Grande; wall to wall for 450 miles. Twelve million
acre-feet is in tonnage about 4o times the entire nation’s petroleum
production and about 400 times the chemical production.

6. Value, the economist’s sixth and last question and the philos-
opher’s first. Assessing the value of water is a treacherous under-
taking. The smaller the group assessed the more likely the error
and the more certain the chance for differing opinions. Very few

324 THE FUTURE OF ARID LANDS

disputes over conclusions developed when the Bureau of Reclama-
tion appraised the water situation in southern Texas and con-
cluded that a $1.1-billion investment in a canal plus a $too-million
annual operating expense would ultimately increase the gross
annual income of the area by $6 billion over the present income.
Not many have denied the value of that water. At the next level,
when a water-short community 1s deciding whether to encourage
agriculture or manufacturing there will be a considerable argu-
ment and permanent uncertainty over the decision. At the lowest
level, when an individual farmer, for example, is deciding which
crop to grow or whether to put an additional acre-foot on his pres-
ent crops, there is very little scientific certainty and a great possi-
bility for error. In fact, farmers, being in general poor at cost
accounting and not having the facts to begin with, very frequently
make other than optimum choices. Depending on the purpose of
the compiler, water values may be based on gross income, on net
profit, on payout period, on total employment, even on esthetic
values, among other things. In the United States, where under
scientific management net profit usually controls the decision,
there are not enough available experimental data on the physical
results of water use to set a proper value on it. What is needed is
a careful study by economists and engineers of the value of water
used for various purposes and under various conditions; and es-
pecially if irrigation is to continue a major user of water, studies
are needed by agronomists and agricultural economists to place a
definite value on water used for specific agricultural purposes.

Water Re-use

In general, topics of competitive prices, transportation costs,
markets, and values have been discussed in the preceding section
on demineralization in order to provide an orientation for the de-
mineralization costs there cited. Those factors apply equally well
to other methods of securing additional water sources such as
water re-use.

The re-use of water in the United States is much more common
than most people realize. In the United States 70 million people
drink from sources which are used also for sewage disposal. Prob-
ably 69,990,000 of them are not aware of it. A properly designed

ECONOMICS OF WATER SOURCES 325

sewage treatment plant can, and many do, produce an effluent of
better quality chemically and biologically than many public raw
water supplies. In the United States about 125 municipal sewage
plants utilize the effluent in agriculture. In addition, in at least
a half-dozen cases, sewage effluent 1s used for industrial operations.
Sewage effluent in general will cost about $10 per acre-foot, and
one industrial plant is purchasing it for $6. However, municipal
effHuents would in general supply only a fraction of the industrial
requirement. Most industries make only one use of water, but in
almost every case it would be possible to reduce water use greatly
if a water shortage should develop. One plant in the Texas-Gulf
Coast area reduced its requirement from 672,000 acre-feet per
year to 44,000 by re-use; another from 224 acre-feet per year to 4.
But there is no possibility that even combined municipal and in-
dustrial reclaimed effluents can make more than a fractional
contribution to irrigation supplies.

Full and economic utilization of industrial waste waters will
require further intensive research and development to provide
suitable processes and a major educational campaign to persuade
managements of the necessity.

Generalized Economics of Arid Lands

In making a few general remarks on the economics of arid lands,
I take the liberty of the social scientist who, out of the necessities
of the present stage of development of his science, must make
broad generalizations from what the natural scientist would judge
inadequate or faulty data. If I am in error, blame it on the natural
scientist who dares to navigate these tricky waters, rather than on
the economists who by much practice have learned to shoot the
rapids without spilling a drop.

Land Value Fluctuations

Our economic systems and theories have been developed quite
largely in lands enjoying adequate and regular rainfall. Under
these conditions the value of the land, first for agricultural, then
for industrial purposes, depends upon the inherent quality and
productivity of the soil, on topography, and on the location rela-
tive to the markets and raw materials. Land, therefore, even after

326 THE FUTURE OF ARID LANDS

the industrial revolution, became a property with a quite perma-
nent value. Fluctuations in value occurred only through discovery
of mineral rights, general fluctuations in the economy, and through
an encroachment of urban areas. In arid lands, however, all these
factors are overshadowed by the overwhelming importance of
water to any utilization of the land. Accordingly the value of
lands in arid regions is subject to another major fluctuation,
namely, that in the availability of water. The predictability of
land, values, therefore, unlike those in humid regions, becomes as
uncertain as long range predictions of weather and climate, and,
as we have seen in more than one short range cycle in our own
time, economic displacements of great magnitude can occur
through this unpredictability. When an economy contains a sig-
nificant segment as an arid land component, the fluctuations of
the latter may initiate sympathetic fluctuations of more serious
consequence in the general economy.

“Oasis Economy”

Except for some variation in place utility and in inherent pro-
ductivity, in the humid regions the value of land changes gradually
from point to point and provides a broad gridwork on which
human activities may be placed with considerable freedom of
choice. In the arid regions, however, as we have just discussed,
except for mineral values, the value of the land is insignificant in
the absence of water, and thus the condition arises that the value
of a given ptece of land is not inherent in the land but ts the value of
the available water at that point. Despite this major economic fact
there has been no readjustment of the economic principles de-
veloped in humid regions, in that our laws and economic system
still consider land the primary value in all areas.

More particularly, unlike humid areas where the value of land
varies gradually from point to point, in arid areas the value of
land changes abruptly at the boundaries of water availability.
Although not caused completely by availability of water, this
characteristic of arid land population distribution 1s one of the
first things noted by the humid-land traveler in arid regions. Arid
land population and activity distribution is characterized by a few
spots of population separated from each other by great distances

ECONOMICS OF WATER SOURCES 327

having practically no population. I use the term ‘‘oasis economy”’
to describe the situation.

There are social and economic consequences of this oasis econ-
omy which, time permitting, could be further explored. As an ex-
ample of a minor sociological consequence I shall cite the restric-
tion in dwelling environment offered in the oasis economy. In the
humid regions the degree of urbanization, being high in the center
of the metropolitan areas, gradually decreases toward the periph-
ery, typically reaching the rural level before beginning to rise
again as the outskirts of the next city are approached. This gradual
change allows the dweller in those regions a wide variety of
choices in balancing the rural and the urban in his dwelling en-
vironment. Not so in the arid lands, for example the Southwest.
Here, although it is true that the degree of urbanization decreases
outward from the center of the city, at the city boundary, namely
at the termination of water, gas, and other utility services, there
is a very abrupt transition to a completely rural environment.
Thus the dweller in the arid regions has essentially only two
choices; either urban residential living or semi-isolated, ranch
type, in many cases supplying his own water, gas, and even power.

Oasis Economy and Advancing Technology

It should be pointed out that solutions such as are discussed in
this volume on prospects for additional water resources are solu-
tions which may intensify the oasis economy rather than alleviate
it. Any solution which increases the availability of water at one
spot, possibly at the expense of another spot, will aggravate the
situation. Technical advances will be made in water supply and
transportation of water, and most of these, if only for efficient
utilization, will intensify the oasis economy. For that reason it
should be stressed, especially for the engineers, that the better use
of present resources and the better adaptation of plants and ani-
mals present possibilities for eliminating the oasis economy and
making arid lands generally available for habitation as we of the
western world understand the normal pattern of land use. I make
this point particularly because it happens that the solution to
additional water resources lies so largely in applied research and
engineering, whereas the solution to better use of present resources

328 THE FUTURE OF ARID LANDS

and better adaptation of plants and animals lies more in the com-
paratively neglected area of fundamental and long term research.

Appropriate Use

Still another anomalous situation arises through the historical
entry into arid lands by humid land people. For the past several
centuries the pattern for development of new lands has been that
when first opened up they constituted the hinterland of civilization
and were considered useful only for agriculture and other extrac-
tive industries. Then as civilization and _ industrialization
advanced, these rural areas were gradually taken over and con-
verted to urban and industrial uses. This situation was satisfactory
as long as adequate rainfall permitted appropriate use of the lands
for either agriculture or industry. This is definitely not true in the
twentieth century arid lands. And yet, up to this time we have
been largely continuing the historical sequence of wasteland,
farm land, city land. I submit the proposition that the use of irri-
gation in the arid lands of the twentieth century is not an appropri-
ate use of that valuable resource, water, but it is an attempt to
follow a historical precedent and increase the value of the land.
Actually from the standpoint of water use, agriculture is a mar-
ginal use of water. In the United States the water that will support
one worker in arid land agriculture will support about 60 workers
in manufacturing.

Those who are used to our present economy will ask: “Who is
going to supply the food.”” The answer is: The railroad trains run
in both directions, let the humid regions ship food into the arid,
since with the arid lands reaching the limit of their water supply,
the humid regions can produce more efficiently anyway and have
by no means reached the limits of their capabilities. The arid
lands should then look toward industrial rather than agricultural
expansion. If we can ship cotton, we can ship cameras; if we can
ship radishes, we can ship radios; if we can ship watermelons, we
can ship watches. Only then can the arid lands provide employ-
ment and a continued high standard of living both for the present
inhabitants and for the many who are swelling their numbers
seeking the favorable climate and living which the lands afford.

Better Adaptation of Plants and
Animals to Arid Conditions

Questions

What screening procedures would lead to the selection of more
productive plant and animal species for arid regions?

What are the genetic and physiological bases for drought resist-
ance in plants and animals?

What are the prospects of increasing drought resistance through
genetic research?

How can we develop a program of revegetation?

What are the economic possibilities in the development and utili-
zation of arid land plants and animals?

What are the possibilities of maintaining larger human popula-
tions in arid areas?

Adaptation of Plants and Animals

OMAR DRAZ

Desert Range Development Project, Desert In-
stitute, Mataria, Egypt

The growing world population, together with the need to raise
the standard of living for millions of people who suffer hunger or
malnutrition, makes the increase in world food production an
urgent and vital requirement.

The Food and Agricultural Organization annual reports show
that shortages in animal products have always been more evident
than those of other food materials (6). It 1s quite obvious that
control of livestock diseases and parasites, systematic animal
breeding programs, and improved agricultural methods, together
with building of more irrigation projects, would augment animal
production. Producing abundant cheap animal feed, however,
will always remain the main factor in materially increasing animal
production. This was proved on a country wide basis in New Zea-
land. After 24 years of work it was found that raising the plane
of nutrition among the animal population caused more production
in annual butter fat per cow than selection, grading up of herds,
or elimination of poor producers (10). Better use of our arid and
semi-arid lands that cover millions of acres could help in provid-
ing this essential cheap animal feed.

Continuous overstocking of arid or semi-arid zones has greatly
reduced their productivity. Through ages of misuse, man has in
the most unwise way exploited millions of acres by extensive and
uncontrolled grazing. This has caused denudation of plant cover,
which, hand in hand with erosion, produced aridity and in many
cases man-made deserts.

331

332 THE FUTURE OF ARID LANDS

Consideration of better use of such land is gaining more interest
by certain international and governmental organizations as well as
by many scientific bodies and institutions.

Most of the efforts for improvement of arid zones have been
directed at studying the plant side. Search has been made for
adapted species, methods of planting have been developed, and
considerable areas have been reseeded. Although animals were
responsible for aridity, less effort was devoted to studying grazing
management. Moreover, less attention has been given to the
human factor which often is the real cause of the lack of balance
between the fodder reserves and the misuse of the grazing lands.
Studies of all phases of the problem must go hand in hand if we are
to find a solution.

Plants and animals adapted to any given arid environment will
differ from place to place. To obtain maximum productivity in
arid zones we have to use the best adapted animals that can graze
drought-resistant plants. The range should be kept in a state of
productivity to be used by the present and future generations.

For the selection of plants and animals better suited to arid
zones, a thorough understanding of the ecological, genetic, and
physiological basis for drought resistancy has to be borne in mind.

Ecologists are mainly interested in adaptation to external
environments. Physiologists are primarily concerned with equilib-
rium of functions within the organism. Geneticists are concerned
with the hereditary makeup of the organism. Range technicians
must make use of the relation of all these factors to one another.

Drought-resistant perennial plants are equipped with morpho-
logical and physiological characters that enable them to withstand
drought. The most important of these are: (1) deep penetration
and extensive development of the root system, (2) relatively high
osmotic pressure of the cell sap, and (3) higher capacity for regu-
lating transpiration (g).

Certain annual plants can complete their life cycle in a com-
paratively short time, thus avoiding, rather than withstanding
drought. This subject is dealt with by Shantz.

Animals adapted to arid zones are the result of interaction of
two main factors, namely heredity and environment. Under arid

ADAPTATION OF PLANTS AND ANIMALS 333

conditions animal life is characterized by extreme ecologic and
physiologic adaptation to feed scarcity and to absence of free
water. Since climatologically most of the arid zones are extremely
hot at least for a certain season of the year, animals to be adapted
to such conditions should have a high degree of heat tolerance.
Although better range management practices might partially
improve the environment by providing better feed, shade, wind-
breaks, etc., this usually takes a long time.

Selection of Animals

As possibilities of modifying environment conditions are
limited, the best adapted animals for the use of the limited avail-
able feed and water reserves must be selected.

Fluctuation in weather conditions does not materially change
body temperature of warm-blooded animals. It activates thermo-
regulatory apparatus to counteract the external change. Accord-
ing to hereditary capacity, animals differ in ability to dissipate
body heat. Heat dissipation can be increased through the effect
of several factors such as increase of surface area, change in color
of body coverings, vasodilatation of subcutaneous vessels and/or
increased vaporization through sweating, and acceleration of
respiration rate.

Species capacity to counteract any external change rests upon
a genetic basis, but the environment may stimulate or limit its
expression (1).

Wright (13), in discussing general principles governing form
and function in different classes of domesticated animals, has
shown that the environmental temperature and the rainfall are
the most important factors which influence the distribution of
domesticated animals. He also brought together certain broad
principles formulated by the earlier zoologists, which were found
to affect form and function.

As to direct climatic effects he referred to Bergmann’s rule,
which implies that animals tend to be larger in size in cold cli-
mates than do comparable species in warmer regions. This rule
was amplified by Allen’s rule which states that in general the
peripheral parts of the same species tend to be enlarged under

334 THE FUTURE OF ARID LANDS

conditions of high temperature with a substantial increase in sur-
face area and heat dissipation. Wright introduced what he called
Wilson’s rule. In describing the coats of mammals, Wilson dis-
tinguished between the bristly external hair covering and the soft
woolly fibers which are commonly hidden beneath these coarser
hairs. He stressed the fact that in cold countries animals tend
toward the woolly coat, whereas in warm regions the tendency 1s
toward more development of hair and disappearance of wool.

Animals inhabiting warm and humid regions show greater
melanin pigmentation than the same species in cooler and drier
regions. In arid desert regions the skin is characterized by the
accumulation of yellow and reddish brown phaeomelanin pigment
(Gloger’s rule). Wright considers that rainfall, through its effect
on quality and seasonal growth of vegetation, acts indirectly upon
size, conformation, and density of the animal population as well
as on their grazing habits and structural formation.

Knowledge of the fundamental physiology of farm animals re-
garding heat tolerance and capability for withstanding unfavor-
able conditions is not yet adequate to set up scales for comparing
different breed adaptability (5).

Rhoad (12) developed the Iberia heat tolerance test for calculating
a heat tolerance coefficient for cattle. This simplified device, being
a field test, has its advantages. Its results however, are liable to
be affected by many external factors. Comparisons of tropical and
half-tropical crossbreeds with the pure temperate breeds on the
basis of this test show a high heat tolerance with concentration of
the tropical blood. The fact that such coefficients do not evaluate
other physiological factors in relation to productivity has always
to be borne in mind.

The Arabian camel (Camelus dromedarius) is a distinguished
illustration of the effect of environmental conditions on a species
(9). Its size and length of limbs and neck coincide with Berg-
mann’s and Allen’s rules. Its hairy coat and yellowish brown
coloration is a good proof for Wilson’s and Gloger’s rules. The
flexible sole of the foot, the callous horny pads that bear most of
the animal’s weight during the sitting position, the water sacs in
the rumen, the cloven upper lip, the mouth which is almost un-

ADAPTATION OF PLANTS AND ANIMALS 335

affected by injury of thorny bushes or trees, as well as many struc-
tural modifications, all operate to make the camel best suited to
arid conditions and food scarcity. The camel’s hump was devel-
oped to store a food reserve in the form of fat. Babcock (2) has
pointed out that fat metabolism produces not only energy but also
free water, which exceeds in volume the volume of the original fat.
This has a profound significance in metabolism, more especially
in the camel, which has to exist for long periods in waterless re-
gions.

Similar combinations of exo- and endo-adaptations could be
traced in other domesticated animals that were developed under
arid conditions such as the Arabian horse, certain strains of fat-
tailed sheep and zebu cattle.

Conditions in the deserts and semi-deserts usually result in a
certain amount of topographic isolation of animals. This provides
enormous potential variability through occasional crosses between
isolated populations. Natural selection under such conditions as
Allee (1) has stated ‘‘will act as a sieve that eliminates the unfit
and allows the fit to pass through. In another sense, natural selec-
tion is a causative force that determines the pattern of hereditary
units through selective sorting after recombination.”

It is most probable that the Arabian horse, the horse of the
desert with its noble qualities of endurance and speed, as well as
its specific anatomical constitution, has been developed through
such a pattern of inheritance.

Breeding of Animals

Phillips (11), in discussing methods of breeding that can be
applied in underdeveloped areas, recommended the use of the fol-
lowing methods: (1) selection within the native types, (2) grading
up with already improved types or breeds from other countries,
(3) development of new types out of animals that are graded up
only a part of the way to the improved type.

A detailed discussion of these methods has appeared in a recent
Food and Agricultural Organization publication, showing the use
and limitations of each method (11).

Compared with artificial selection, natural selection is very

336 THE FUTURE OF ARID LANDS

slow in establishing new combinations of characteristics through
mutation and recombination of genes. Haldane has estimated
that it takes 10,000 years for a favorable combination of fifteen
genes to become established, whereas under artificial selection this
could be accomplished in relatively few years. This emphasizes
the importance of artificial selection for improvement of animals
and plants in arid zones.

There is, however, a fourth method of improvement that can
be used in arid zones, that 1s the z#troduction of preadapted species.
Certain strains of goats, fat-tailed sheep, and zebu cattle that
have originated in arid environments are likely to be successfully
adjusted for introduction to similar arid regions.

In planning for such trials, Wright (13) recommends two meth-
ods for comparing homoclimatic areas, Le., climatographs and
hitherographs. The former is based on the sdaiion of air tempera-
ture to relative humidity, and the latter, on the relation of tem-
perature to rainfall.

Examples of preadaptation are more prominent in the plant
kingdom. Crested wheatgrass (4gropyron cristatum) is a native
of the wide plains of Siberia and Central Asia, yet its merits in
reseeding depleted areas was first proved in the northern Great
Plains of the United States and not in the eastern hemisphere
where it originated.

Harding grass (Phalaris tuberosa) and subterranean clover
(Trifolium subterraneum), both Mediterranean species, are now
considered the best adapted to many areas in southern Australia.

Recently, Kochia indica (Wight), a native of India and Pakis-
tan, has been successfully established in the coastal belt of the
Egyptian western desert, under less than 6 inches annual rainfall.
The species, although originally a salt soil plant, proved to have
enough drought resistancy to withstand the semi-arid conditions
of the new habitat. Although useless in its original country, its
high palatability and easy establishment made the species of some
use under Egyptian conditions.

In comparing improved breeds and local breeds, Hagedoorn (7)
expressed the fallacy of looking down upon local “unimproved”
breeds of domestic animals and plants just because each individual

ADAPTATION OF PLANTS AND ANIMALS S7/

compares unfavorably with a corresponding individual of an
“improved” breed. He repeated that there is only one real measure
of superiority in domestic animals, that is adaptation to the con-
ditions of agriculture into which the breed must fit. In discussing
the suitability of animals to arid zones, he stated: “One sheep
weighing a hundred pounds has only one head and one set of legs;
two sheep, weighing a hundred pounds together, have two heads
and two sets of legs so that they can be in two different places to
hunt the scanty herbs, and for this reason, in conditions where the
sheep of fifty pounds can just live, a hundred-pound sheep must
necessarily starve.” He was referring to differences in comparative
adaptability of breeds. However, in any one breed or line, it is
obvious that fewer, larger, thriftier animals kept in good flesh are
certainly better than a greater number of small undernourished
ones.

Cooperative Planning in One Country

In planning for improvement of production under arid condi-
tions it is essential to treat all phases of the problem. A. T. Semple
believes that cooperative work could be accomplished if plant
scientists handle the problems of plant production and study the
effect of animals on plants. Meanwhile, animal scientists would
handle the livestock problems and determine the effects of plants
upon animals. This division of responsibility, if carefully planned
within a research organization, would maintain cooperative work
and lessen both duplication and the chance of overlooking any part
of the problem. Tackling a group of problems from all angles at
one time might be the best screening procedure for selection of
better adapted plants or animals for arid regions.

The establishment and adequate financing of one or more arid
zone experiment stations may thus be a good approach to the
problem. The plan for the Ras el Hekma Desert Range Research
Station, which covers 25,000 acres in a 6-inch rainfall area,
includes the following points:

1. Analysis of background information to indicate potentialities,
covering climatological studies, topographic mapping, soil mois-
ture, land survey, and classification. Aerial surveying through

338 THE FUTURE OF ARID LANDS

interpretation of air photos was found to be of great value to the
program.

2. Study of the human factor in relation to land use and land
tenure as well as to livestock numbers in all phases of the program.
Unless the recommended practices are acceptable to the people
and in harmony with their customs and way of life, the whole
program of improvement is bound to fail. The extensive damage
that has been done in most of the desert and semi-desert areas
resulted from factors beyond the control of the people who are
always held to blame. The instability of life and lack of property
rights are the real causes of over-grazing and nomadism. Nomadic
life is a forced way of living and will be dropped when a good
alternative is available.

2. Species adaptability tests, including methods of establishing
plant species by reseeding and their response to grazing, is ihe
means of pointing the way for large-scale reseeding. We have
tested almost 300 species of native and introduced plants. Only a
few proved to be of some use such as Phalaris tuberosa, Ehrharta
calycina, Agropyron desertorum, Medicago sativa, Melilotus species,
Sanguisorba minor, Atriplex species, Kochia indica and Prosopis
juliflora. Many natives hold much promise once seed supply is
developed; the following are examples: Dactylis glomerata var.
hispanica, Stipa lagascae, Oryzopsis miliaceae, Cynodon dactylon,
Potertum varicosum, Plantago albicans, Moricandta nitens, Poly-
gonum spp., and Lotus creticus.

It is worth while to mention that it was comparatively easier at
the beginning to obtain foreign seed and information from already
established stations abroad than to collect our own.

From observations of how many species failed to establish
themselves against how much less succeeded, one may very soon
observe and recognize the better adaptation of the native species.

Selection of sites needs careful study. In our case, one screening
nursery and twelve outplanting plots have been established along
a 300-mile strip of land.

4. Study of soil and water conservation methods and applica-
tion of certain practices, including water spreading systems.

5. Natural recovery through protection and management.

ADAPTATION OF PLANTS AND ANIMALS 339

This includes taxonomic as well as ecological work. Fencing of
several areas representing different conditions of soil proved its
efficiency in quick restoration of plant cover. In a two-year period
of protection, the number and size of native plants already men-
tioned under (3), growing in fenced areas, have considerably
increased in comparison with those in unprotected areas.

6. Study of livestock health, breeding, feeding, and manage-
ment to insure that animals will be best adapted and fully produc-
tive. In many instances improvement of the ranges was followed
by the spread of certain animal diseases, mostly parasitic infes-
tations. The study of existing parasites is carried out to help in
finding the correct rotation of grazing. Range and livestock
management should be handled in a way that will control such
infestations and insure maximum productivity.

7. Study and use of available concentrates from the Nile Delta
to allow possibilities for using all the roughage produced in the
desert.

8. Water development for domestic and livestock use and for
limited irrigation including supplemental forage production.

g. Establishment of a training center for range management,
serving also as an extension center.

Fruitful Lines of Action

In summary, it may be concluded:

1. With available information, more work must be conducted
in animal physiology, to be able to set scales for drought resist-
ancy in animals. Yet the use of existing breeds of cininnell origi-
nated under arid conditions and having certain characteristics
exo- and endo-adaptive to these conditions has its practical and
scientific significance.

2. As most of the characteristics that will counteract environ-
mental changes rest upon a genetic basis, and as the patterns of
inheritance in breeds of animals under desert and semi-desert
conditions usually help in developing such characteristics, the use
of these breeds in improvement programs 1s highly recommended.
The extent of such use has to be governed only by the results of
experimental studies on comparative productivity.

340 THE FUTURE OF ARID LANDS

3. Inasmuch as plants, the grazing animal, and the human race
should be considered as a community living in a symbiotic group,
screening procedures for selection of more productive species and
development of programs for revegetation of arid regions should
be tackled from different angles at one time, through the estab-
lishment of well-equipped experiment stations adequately staffed
and financed.

4. Fodder production generally is increased under proper man-
agement. Water development paves the way for better use of the
range. The present destructive uncontrolled rights of use now
prevailing in arid and semi-arid lands should be subjected to the
fulfilment of certain land improvement requirements. Fencing
of a certain percentage of land within a reasonable time, followed
by proper use of the range, should allow for complete property
rights on the land. Loans to encourage and support land owner-
ship in desirable types of use might materially increase produc-
tion in arid lands.

5. Nomadism is a symptom more than a disease. Once a chance
of stable life is given through proper land use and ownership and
through controlled grazing, there will be little reason for its ex-
istence. With improvement of forage and water resources and the
development of a settled way of life, a number of technicians,
laborers, and livestock producers could be absorbed to help de-
velopment and formation of proper desert communities.

This, in general, is the approach and pattern which one single
country like Egypt might be able to develop to solve partly the
problems of its 22 million, rapidly growing population, who live
crowded in 6 million acres surrounded by deserts. However, there
are certain other fields where work on arid land problems might
be advanced most fruitfully through international collaboration.

International Cooperation

Most of the stations where arid land improvement is carried
out at present are field stations conducting empirical trials. Fu-
ture programs on arid land development will need and have to
depend more on basic research in such fields as physiology of
drought resistance, genetics, soil science, ecology, solar energy, and

ADAPTATION OF PLANTS AND ANIMALS 341

other fields which need intensive laboratory studies conducted
over a long period of time. These obviously are too big a task for
individual countries.

Information about arid land problems is not always available
to most of the interested people. The dissemination of this type
of information is another field that needs more international co-
operation.

Successful introduction of grasses and legumes into different
countries, although quite recent, has been in many cases out-
standing. The establishment of an international unit to help
procurement and proper distribution of plant material might facil-
itate exchange of seeds and information from different arid lands
in the world.

Meetings, such as the international arid lands meetings, pro-
vide a very worth-while means for exchange of ideas among
world experts. More of this type should be planned in different
parts of the world and at regular intervals.

REFERENCES

ie Allee. Or lbPark, Wl. Park) Ac Emerson, and Ke Sehmidt. 1949:
Principles of Animal Ecology, W. B. Saunders Company, Philadel-
phia and London. Pages 590-602, 631-647.

2. Babcock, S. M. 1912. Wisconsin Agr. Expt. Sta. Research Bull. 22.

3. Curasson, G. 1947. Les Chameau et ses Maladies. Vigot Freres,
Paris. Pages 23-35.

4. Draz, Omar. 1954. Some desert plants and their uses in animal feed-
ing. Publications de l'Institut du Desert d’Egypte, Heliopolis,
Egypt.

5. Findley, F. D., and W. B. Beakley. 1954. Progress in the Physiology
of Farm Animals (Hammond, Editor). Butterworth’s Scientific
Publications, London.

6. Food and Agricultural Organization. ig52. Second World Food
Survey.

7. Hagedoorn, A. L. Animal Breeding. Crosby Lockwood and Son Ltd.
London. Pages 71-82.

8. Leese, A. S. 1927. 4 Treatise on the One-Humped Camel in Health
and Disease. Hayenes and Son, Lincolnshire, England.

g. Migahid, A. M., and A. A. Abd el Rahman. 1953. Studies in the
water economy of Egyptian desert plants. Bull. Inst. Desert
ad Egypte 3, No 1.

10. New Zealand Dairy Board. 1950. Nineteenth annual report and

342 THE FUTURE OF ARID LANDS

statement of accounts for period of 12 months ended 3 July 1943.
(FAO Development Paper No. 8, Agr. 1950).

11. Phillips, R. W. 1953. Breeding livestock adapted to unfavorable
environments. FAO Agr. Studies No. 1, p. 153.

12. Rhoad, A. C. The Iberia heat tolerance test for cattle. Tropical Agr.
21, 194.

13. Wright, N. C. 1954. Progress in the Physiology of Farm Animals
(Hammond, Editor). Butterworth’s Scientific Publications, London.

Better Adaptation of Plants
to Arid Conditions

R. MERTON LOVE

University of California, Davis, California

Stated briefly, the task is to explore the fundamental and ap-
plied aspects of research in order to suggest ways and means of
transforming low-production-per-acre areas into more profitable
agricultural use.

At the outset, it must be realized that we are confronted with
a soil-climate-plant-harvest complex. It is an agricultural prob-
lem. It has been with us ever since Man began his attempts to
supplement that which Nature provided by domesticating ani-
mals and plants. If it differs at all from problems facing other
phases of agriculture, it differs in degree only.

Since, in general, production is low in arid areas, it is likely that
animals will be used for a long time to harvest the forage crop
grown on non-irrigable land. Thus, we substitute ‘‘animal”’ for
“harvest,” and so we have a soil-climate-plant-animal complex.

Drought Resistance

Ting capacity of plants to survive periods of drought with little
or no injury 1S usually termed ‘drought resistance.” The term
“drought” is not in itself subject to any rigid definition. Certainly
it includes soil drought and atmospheric drought, the latter in-
volving high temperature, low humidity, and high wind velocity,
singly or in combination.

It has not been sufficiently emphasized that the term drought
resistance is a very inclusive term comparable in its scope to the

343

344 THE FUTURE OF ARID LANDS

term disease resistance. Plants may be susceptible or resistant to
smuts, rusts, mildews, and so on. As research on plant diseases
developed, it was found that there were races of many of the
disease organisms and resistance to one race of rust, say, was no
indication of the reaction to be encountered with another race.
Disease resistance has many facets.

Use of the term drought resistance is even more nebulous be-
cause the term has not yet been satisfactorily defined. Water is
an essential factor in photosynthesis since in the manufacture of
food the carbon of carbon dioxide is united to the water brought
up from the soil. This water remains in the plant until the food is
oxidized or broken down, when it may be released. In addition,
(1) water is an essential component of protoplasm, (2) it serves as
a solvent for oxygen and carbon dioxide, (3) it has a high specific
heat and absorbs much excess heat energy, (4) it aids in the
transportation of raw materials and foods, and (5) it maintains
turgor in living cells. In addition to satisfying all these needs the
plant requires large amounts of water to replace that lost by
bleeding, through the action of glands, by guttation, and by
transpiration (55).

Arid land species may be ephemerals, which are ‘‘drought-
escaping”’ (60); succulents, which are a distinctive group of plants
not only in structure but in metabolism and water economy (61);
or “drought-enduring” species, whose cells can endure a severe
reduction in water content for extended periods of time without
serious injury, for example, creosote bush Larrea tridentata (59).
Only plants of the latter group are truly drought resistant.

Physiological Bases for Drought Resistance

Attempts to explain drought resistance on a purely morpho-
logical basis have proved inadequate, although certain structural
features of plants undoubtedly aid in their survival in dry habi-
tats (50). It seems clear that one of the basic factors in drought
resistance of plants is a capacity of the cells to endure desiccation
without irreparable injury. Desiccation of the protoplasm in itself
does not result in death of the plant, according to Iljin (27, 28),
but rather the mechanical disturbances such as pressure, stretch-
ing, and tearing, which result from dehydration of the cell.

BETTER ADAPTATION OF PLANTS 345

Similarly, attempts to explain drought resistance under con-
trolled environments on a purely physiological basis have not to
date been entirely satisfactory, although certain facets have been
explored. Aamodt (1) and others are of the opinion that drought
resistance must be determined by direct testing of the plants
under controlled conditions. The difficulty has been that any one
investigator has limited his study to one or two factors, whereas
in the field many factors may be operative at the same time. For
instance, high temperature is one facet of drought, but the ability
of a plant to withstand the effects of high temperature is not
necessarily correlated with the ability of a plant to survive in
arid conditions. Still, high temperature is one factor and studies
by Julander (29) and others have indicated that in some species
resistance to high temperatures may be taken as an index of
drought resistance.

In regions where plants are subjected to lengthy periods of
summer drought and heat, perennial plants must in some manner
withstand the severity of the summer period. During such a period
plant growth becomes very limited or ceases entirely. Laude (32),
using 20 perennial grasses, conducted a critical study of the nature
of this condition which he termed ‘‘summer dormancy.” With its
Mediterranean type climate a dry summer period of 5 to 6 months
prevails at Davis, California. Normally, all 20 species are summer
dormant under natural rainfall conditions. Thirteen of the 20
species continued vegetative growth throughout the summer when
watered at weekly intervals; four species ceased vegetative growth
even though supplied with water, although they did retain green
tissue; and three species did not even retain green tissue (Table 1).

Another test was devised to allow a period of summer dormancy
and then supply water to the seven species that failed to continue
growing in the first test. Four of the seven commenced growth
after a short dormancy period, but two species of Stipa did not
respond to water until September, and Poa scabrella did not break
dormancy until cool weather prevailed with the first fall rain.

In still another test, Laude severed the lateral roots 18 inches
from the center of the rows, and the vertical roots at 2- or 4-foot
depths by cuts under the blocks. This was done in late July. Soil
moisture samples taken August 3 indicated that the soil was dry

346 THE FUTURE OF ARID LANDS

TABLE 1
Effect on Vegetative Growth of Supplying Soil Moisture

Supplied at weekly intervals throughout the dry summer period to twenty perennial
species in the field at Davis, California, 1947-1950 (32).

Species Ceasing Vegetative Growth

Species Continuing Vegetative Growth Retaining green Retaining no

tissue green tissue
Agropyron desertorum Melica californica Poa bulbosa
Bromus carinatus Poa nevadensis Poa scabrella
Bromus catharticus Stipa cernua Poa secunda
Bromus stamineus Stipa pulchra

Dactylis glomerata

Ehrharta calycina

Elymus glaucus

Festuca arundinacea

Lolium perenne

Oryzopsis miliacea

Phalaris tuberosa var. stenoptera
Poa ampla

Poa compressa

to the permanent wilting percentage to a depth of 6 feet. Undis-
turbed plants showed considerable foliage development by Sep-
tember 12, whereas those with roots severed at the 2- and 4-foot
depths did not resume growth until after rain had fallen late in
October.

In Poa scabrella, however, the initiation of summer dormancy
is associated with long photoperiods and high temperatures. Cool-
ness and availability of water break this dormancy.

It is seen from these studies that different environmental
factors, singly or in combination, are associated with the initiation
and the breaking of summer dormancy: moisture availability,
temperature, and day length.

The drought resistance of a plant, then, is a reflection of the
interaction of a number of physiological and morphological re-
sponses to the many environmental factors opereting during the
life history of the plant.

One aspect, to my knowledge, has not yet been sufhiciently
investigated. I refer to the effect of soil type on the survival capa-
bilities of different species in arid areas. For example, our field

BETTER ADAPTATION OF PLANTS 347

plots and seedings in California have demonstrated that EArharta
calycina produces well in light soils in the 18-inch winter rainfall
belt in the Great Central Valley of California, but on “‘tight”’ soils
plants will survive only a year or two at the most. Phalaris tub-
erosa var. stenoptera does very well on these ‘“‘tight” soils. In very
sandy soils of coastal California the performance of E. calycina is
remarkable, whereas P. ¢uberosa var. stenoptera survives only in
the lower, more favorable, swale sites. This will be referred to
again later.

Many studies have been conducted, of course, on mowing or
grazing effects on the capacity of plants to withstand drought
conditions (64). In general, plants higher in food reserves are more
tolerant of heat injury. According to Julander (29) large accumu-
lation of colloidal carbohydrates, especially levulosans, is associ-
ated with drought resistance. Heavily clipped plants do not
accumulate food reserves during drought and are less resistant.

It would seem, in summary, that physiological investigations
on the nature of drought resistance should be intensified and cor-
relative studies should be conducted on the sites and in the en-
vironments where the range plants are used.

Genetic Bases for Drought Resistance

Olmsted (53) reviewed his work on photoperiodism of native
range grasses of the midwestern United States. He found con-
siderable variation, both interspecific and intraspecific, in their
photoperiodic behavior. Intrastrain behavior also differed. A
southern Texas strain of Bouteloua curtipendula showed less intra-
strain diversity than the others. A strain from central Oklahoma
exhibited the most intrastrain diversity in all respects, including
photoperiodic behavior.

Knowles (30) determined the extent of variation within Bromus
mollis and the relation of this variation to the environment. An
adventive species not present in California before 1870, it was
abundant in most regions of the state by 1900 (56). In the short
span of 70 years two distinct ecotypes are recognizable in Califor-
nia: a late coastal ecotype and an earlier maturing interior ecotype.
In addition, there is some genetic diversity within the ecotypes.

Turesson (67) and Clausen, Keck, and Hiesey (5, 6) have done

348 THE FUTURE OF ARID LANDS

outstanding experimental work on plant ecotypes. Turesson
showed that if collections of a species from different habitats were
grown together under the same environmental conditions their
distinctive characteristics are frequently heritable. Clausen et ai.
by means of reciprocal transplants exhibited intrinsic differences
between strains of wild species.

Gregor and Sansome (18) and Stapledon (62) reported pros-
trate habit of growth to be more common in coastal forms of
bunchgrasses, but this appeared to be related to years of close
grazing. Turesson (67) found the same situation in perennial
dicotyledons from the maritime areas of western Europe. And
Love (unpublished) found it to exist in Stipa pulchra, a native
California bunchgrass.

Love (35) reported that interspecific Stipa hybrids remained
green longer than the parental species. These hybrids are sterile,
and no doubt failure of seed production contributed to a longer
period of green feed. He stated that it may become possible to
interplant strains of two species which would cross and produce
a population of vigorous hybrids interspersed among the parents.
Technological difficulties involved in harvesting and processing
Stipa seed have stopped this particular project. However, the
fundamental idea may well be applicable to other closely related
cross-pollinated species.

Genetic studies with maize (Zea mays) were conducted by
Heyne and Brunson (23) on the reaction of seedlings to high tem-
peratures. In selfed lines and crosses between them, they studied
linkage relations in eight of the ten linkage groups. Close associa-
tion was found between heat tolerance and the Su su (sugary vs.
starchy) and Pr pr (aleurone color) loci, and possibly including
the Cc locus (color factor affecting Pr). Hybrid vigor had no effect
on heat tolerance of the seedlings.

Stebbins (63) reviewed his research work on interspecific
hybridization of grasses. Since the production of new types by
this process is a long and difficult process, he concluded: “‘If a par-
ticular grassland region already possesses many productive
forage species which are ideally adapted to it, the improvement
achieved by this method is probably not worth the time and effort

BETTER ADAPTATION OF PLANTS 349

that must be put into it. But if desirable and well-adapted species
are not available for the regions concerned, a condition which 1s
more or less true for most of the regions of the earth having a Mediter-
ranean-type climate, with long, rainless summers, then this radical
method of breeding may in time be rewarding.” (Italics mine.)
As will be pointed out later, I cannot agree with the italicized
portion of the above statement.

There is general agreement with Stebbins’ statement that at
least 70% of grass species are allopolyploids, the result of wide
crosses. But there is also general agreement that newly developed
(the so-called raw) amphidiploids are quite dissimilar to old es-
tablished allopolyploids, especially in their cytogenetic behavior
(7). This does not mean that, ultimately, benefits will not be
reaped from interspecific hybridization. It does mean that results
from such a program are likely to be very much in the future.

Love (36, 38a) called attention to the difficulties inherent in
such radical hybridization programs. In 19§2 he outlined the fun-
damental research needed to put such a project on a sound, sci-
entific basis after reviewing this field of endeavor. The review
showed beyond a doubt that the best chances of success would be
obtained from beginning hybridization work at the diploid level.
He also called the attention of the world’s grass breeders to an
entirely new development (17, 58).

They found that ‘colchicine treatment of full sibs of a true-
breeding variety of sorghum gave variants possessing a number
of ancestral characteristics of which some bred true immediately.”
This treatment of emerging shoots does not necessarily cause
polyploidy, but may also cause somatic reduction. Huskins (26),
who pioneered this work, wrote: “If the production of homozygous
diploid tissues could be induced frequently enough, it would have
some advantages and few disadvantages for the plant breeder
over the production of haploids.”

Clausen (4) and his colleagues have pioneered a new field by
hybridizing apomictic species of Poa that belong to distinct sec-
tions of the genus. They have crossed forms so widely separated
geographically that they previously had no opportunity for hy-
bridization. They have obtained transgressive segregation and

350 THE FUTURE OF ARID LANDS

greater tolerance to differences in environment. The program is
based on the fact that in a predominantly apomictic plant, occa-
sional hybrid seeds may be obtained, which in turn develop into
plants that are essentially apomictic. This obviates the necessity
of spending years to stabilize a desirable type. Just how much this
type of program will contribute to production of arid lands re-
mains to be seen.

In the self-pollinating cereals, Hordeum vulgare and Triticum
aestivum, somatic variability is carried between lines only, and the
population consists of a small number of homozygous biotypes.
Mixtures of such biotypes planted, harvested, and reseeded for
several years at a number of locations result in a preponderance
of plants of one biotype, not necessarily the same one, at each
station, although no biotype is completely eliminated (1g, 31, 65)
In other words, a local inbreeding population has diverse biotypes
available for response to changed environments that may occur
in the future.

Cooper (12) reported the results of a comprehensive ecological
and genetical study on heading responses in local populations of
cross-pollinating species of Lo/ium. He had shown earlier (11)
that the flowering responses in these species are closely adapted
to local conditions of temperature and day length. Cooper’s mate-
rial satisfied Mather’s (47) two criteria: gene effects must be
additive on the average and the non-heritable variance must be
independent of the genotype. In each local population heading
behavior is uniform under the conditions for which the population
has been selected, but genetic diversity may be revealed under
changed environmental conditions.

These results support Mather’s statement that the population
can possess high immediate fitness for its present environment
and yet maintain a reserve of variability for evolutionary change.
And as Cooper (12) wrote: “Such a population structure provides
the genetic basis for local ecological adaptation.”’

Frandsen (16) pointed out that natural selection may perhaps
act on the improved strains when grown in practical agriculture.
Often soil conditions are better at the breeding stations. This and
other favorable factors “may cause a shift in the genetical com-
position of the improved strains due to natural selection, a fluc-

BETTER ADAPTATION OF PLANTS 351

tuation which could go in the direction of less productive types.”’
He emphasized that effective breeding work must be combined
with a continuous improvement of the environmental conditions
which man can influence.

Although few genetic studies have been conducted on drought
resistance, especially in forage plants, one can expect that genetic
diversity exists just as it does for heading response to photoperiod
and temperature in cross-pollinating species of Lo/ium. Similarly,
there can be little doubt that self-pollinating species have a reser-
voir of diversity in the homozygous biotypes that is available for
selection by the plant breeder.

Screening Procedures Leading to Selection of More Productive
Plants for Arid Regions

Before discussing screening procedures, I should like to describe
briefly what has occurred naturally in California in the short span
of 185 years. As a matter of record, great changes occurred in the
floral composition of some parts of California in as few as 30 years.
I believe this brief review will provide some interesting ideas on
what might be done to improve the more rapid utilization of
introductions and new types of plants.

The majority of alien plants of California have been introduced
unwittingly, and according to Robbins (56) “many have become
highly undesirable, constituting our worst weeds.’’ According to
Parish (54) “It will be safe, then, to assume a very definite date
for the beginning of that foreign invasion which since has so
greatly modified the plant population of the state. For it must
have been a virgin flora that greeted the eyes of Fr. Serra and his
companions, when, on the 14th day of May, 1769, they reached
the bay at San Diego....The few previous explorers... had
made but transient landings, but the followers of Saint Francis
brought with them flocks and herds, and in the careful prepara-
tions for their expedition they had been particularly charged to
provide themselves with a store of seeds of useful plants... .”

The first mission was established, then, in San Diego in 1769.
The last of the chain of missions was built 800 miles to the north,
near Sonoma, Sonoma County, in 1823.

Hendry (20) and Hendry and Bellue (21, 22) made admirable

35/2 THE FUTURE OF ARID LANDS

studies of the seeds found in adobe bricks from the walls of old
buildings, including missions, whose construction dates are known.
If time allowed, it would be interesting and enlightening to dis-
cuss in detail the introduction and spread of weeds, species by
species. Only a few can be mentioned, but these will serve to point
up the problem. Hordeum murinum L. (common foxtail) is one of
the most widely distributed grasses on range and pasture lands.
Seeds were found in adobe bricks from missions constructed in
1775 and in 1780. Brewer and Watson (3) reported it abundant in
the south coastal area by 1860, in 1890 Hilgard (24) described it
as a “fearful nuisance” in Central California, and in 1902 Davy
(13) reported it rapidly coming in on the ranges of northwestern
California.

According to Robbins, twenty important alien grasses were
fairly well established in the state by 1860. These include annual
Bromus spp. and Hordeum murinum with obnoxious seed charac-
teristics, and desirable annual species of Bromus, Avena, Lolium,
Medicago, and Erodium. From 1860 to 1g00, Bromus mollis and
several obnoxious annual Hordeum species became well estab-
lished.

The rapidity with which these alien annuals invaded an area is
attested by Davy (13). From Sherwood Valley, Mendocino
County, he gained an idea of the primitive flora. When the valley
was first settled in 1853, Danthonia californica, a perennial bunch-
grass, was the dominant and most valuable grass of the hillside
and valley floor. In 1902 the prevalent grasses were Bromus race-
mosus, Hordeum murinum, Festuca myuros, and A1ra caryophyllea,
all introduced species except the latter. Reports indicate that one
good forage grass, Bromus mollis, was not present in the state be-
fore 1870, but in the short span of thirty years it was very abun-
dant in most regions of the state (56). Talbot et a/. (66) again
called attention to the “great extent to which the native vegeta-
tion over vast areas in California has been replaced by plants
introduced from the Old World.”’

What is the explanation for the rapid spread of these annual
plants in California? They have proved to be extremely well
adapted to the environmental conditions in the state, having
superceded, under past and present management, the original

BETTER ADAPTATION OF PLANTS 353

bunchgrasses that were abundant before settlement by Europeans
(8).

Oryzopsis miliacea was introduced into California in 1879.
Attempts to use it in range seeding programs were unsuccessful
Not until it was tried in the ash of brush burns was its place in
revegetation recognized (37, 41). Similarly, in the midwestern
United States, Zgropyron desertorum (formerly called 4. cristatum)
was introduced in 1898 and again in 1go06. Not until the 1930’s
were commercial seedings really successful (14).

In California Medicago hispida has spread over much of the
range land. On two terrace soils, however, it has never become
well established. Trifolium hirtum from Turkey, introduced by
the U. S. Department of Agriculture, and 7. incarnatum were
first tried on these soils in 1944. Trifolium hirtum was so successtul
that in five years it was in seed production, and many thousands
of acres were seeded to it and Trifolium incarnatum and T. sub-
terraneum (38, 44). For the last two species, seed supplies were
already available. The recent introduction and spectacular adap-
tation and usefulness of Trifolium hirtum in California is in rather
direct contradiction to Whaley’s (68) statement. He wrote that
although imported plants ought to be subjected to continuing
investigation, “‘We must, however, face the fact that the world’s
flora is now well known and that the likelihood of discovering a
plant that will grow satisfactorily in such regions as those with
which we are concerned, and produce in quantity substances of
considerable usefulness, is at best slight.’’

This brings to mind the work of Duisberg (15). By means of
alcohol extraction he produced an edible livestock feed from the
desert shrub, Larrea divaricata. If industrial use could be made of
associated resins and acids, as well as the feed constituents, per-
haps the latter could be supplied to the livestock industry at
economically practical costs. This type of research on plants other
than grasses and forbs might well pay dividends.

Experience with Oryzopsis miliacea and Trifolium hirtum indi-
cate that these species are adapted to certain ecological niches.
The screening procedure necessary is to find these ecological
niches. The 7. hirtum story is worth pursuing further. It was
known that these terrace soils were extremely deficient in phos-

354 THE FUTURE OF ARID LANDS

phorus. The obvious approach was to apply sufficient phosphorus
so that the desired plants could be grown. However, all attempts
to establish Medicago hispida failed, although many elements were
tried. Such fertilizer plots overseeded in later years with trifoliate
clovers gave remarkable forage yields, averaging three to four
times that of the unimproved range. In fact, T. Airtum succeeded
even without additions of phosphorus.

The lesson to be gained is that one plant will succeed on a soil
where another has failed. This idea is not new, for it is the very
basis of successful agriculture, but perhaps the idea has not been
carried far enough in our thinking about range lands.

In addition, then, to climatic adaptation we have established
the fact that two other factors must be considered: soil and the
plant species.

If we think of the plants on the range as a crop, the next con-
sideration is how to harvest it. In other phases of agriculture,
special harvesting methods are generally required, depending on
the crop. For example, it is readily appreciated that lettuce, bar-
ley, and cotton have different harvesting requirements.

In most agricultural crops a weed 1s recognized as a weed, e.g.,
Amaranthus spp. in cotton. This also applies to some of our range
weeds. It is recognized, too, that the weeds reduce crop yields by
competing with the crop for water and nutrients. A poisonous
plant such as Ha/ogeton, and extremely undesirable ones such as
Hypericum perforatum, and some of our brush species stand out
as special problems and their control may be attacked in orthodox
ways.

In general, however, the problem of weeds and weediness on our
ranges is not so simple. It is often one of degree.

On much of California’s foothill ranges the cover includes the
following winter annuals. They may be grouped into two classes:

Undesirable Desirable
“the weeds” “the crop”
Gastridium ventricosum Erodium botrys
Festuca spp. Bromus mollis
Hordeum spp. Avena spp.
Bromus rubens Trifolium spp.
B. rigidus Erodium cicutarium

Medicago hispida

BETTER ADAPTATION OF PLANTS 355

Everyone may not agree with this order, but I have placed what
I consider the most undesirable first on the list. Bromus rigidus
is a borderline grass. it is palatable and nutritious when young,
but the ripe panicles are obnoxious because of the barbed awns on
the seeds which do not readily shatter. For this reason it is placed
in the “undesirable” column.

In some fields, undesirable species may comprise 90% of the
ground cover. What method of weed control can be used to change
this balance? No chemical has yet been developed that will per-
form this miracle.

It was customary in the early days in California, as elsewhere,
to turn animals into a field or area and leave them there as long
as possible. In the fall there was a scarcity of feed, but in the early
spring, an excess. Since the introduced annuals spread so rapidly,
this grazing system must have favored them. Given free choice,
animals will select and choose their forage plant by plant. As the
more palatable plants are consumed, livestock turn to the less
palatable ones.

Continuous use, then, favors the less desirable species. It should
be obvious, then, that 1f we want to discourage such weedy an-
nuals, we must change our harvesting program.

Experiments to test this idea were reported by Love (34) and
Jones and Love (33). Both mowing and grazing were used. In a
clipping experiment half of each of twenty-four plots of Spa,
seeded in the fall of 1942, were mowed April 1, 1943, and the other
half, May 13, eae about six weeks later. Counts made in Janu-
ary, 1944, showed a reduction of 30% in stand in the first set
compared with a 71% reduction in the late-clipped plots.

In the second experiment fall seedings were subjected to inten-
sive grazing by sheep at two different dates, beginning April 2,
1943, and April 20, 1943. Plant counts were made September 23,
1943. Eight perennial grass species in plots totaled 255 plants in
the first field and only 75 in the second. Apart from actual num-
bers, those plants in the first field were healthy and vigorous at
the end of the dry season in 1943, whereas those in the other field
were poorly established with weak root systems, barely holding
the crowns in contact with the soil. Love wrote (34): “‘The fact
that during this critical period (1.e., the first spring following

356 THE FUTURE OF ARID LANDS

seeding) the grazing animals did not damage the seedlings, but,
on the contrary, reduced the competition provided by the annuals,
is a fundamental one and points the way to the improvement of
the California range.”’

Eight years of tests on a livestock ranch confirmed the prelimi-
nary tests (38, 39, 40, 45). Thus it was that Murphy, Love, and
Berry (52) were able to recommend grazing methods following
the biological control of Hypericum perforatum.

These experiments have brought into focus the fourth element
of our complex climate-soil-plants-livestock.

Some of the ecological interrelationships of range plants were
learned from these experiments. For instance, good stands and
growth of desirable winter annuals, particularly legumes, reduced
the succeeding population of summer weeds such as Hemizonia
spp. and Trichostema lanceolatum (38, 44). The latter can grow
only in fields where undesirable winter annuals mature before all
subsurface moisture is exhausted, leaving some available for the
summer weeds.

Admitting the importance of increasing drought resistance and
improving screening procedures, perhaps these are not of the most
immediate urgency. Even more urgent now is the need to create
a more favorable environment for the adapted plants already
available. The way to create this more favorable environment is
by the application of agronomic principles of crop production to
range improvement. Consider the complex of range species as a
crop (with weeds, to be sure) and learn how to harvest this crop
by livestock. Manage the area to encourage the crop and discour-
age the weeds. Love and Sumner (45) defined this agronomic
aspect of range improvement as ‘‘the process of replacing a rela-
tively undesirable population of plants with a more desirable
type of forage.”

In the last decade, particularly, men trained in fields other than
ecology have taken an increasing interest in range improvement.
New concepts are being developed. A typical purely ecological
definition is that of Sampson (1952 ): “Range management is the
science and art of procuring maximum sustained use of the forage
crop without jeopardy to other resources or uses of the land.”

BETTER ADAPTATION OF PLANTS 357

Sampson goes on to say: “A knowledge of agronomy and animal
husbandry, however, is highly useful to the range manager, though
these subjects apply less directly to the understanding and solu-
tion of range problems than such subjects as plant physiology,
ecology, and taxonomic botany.” Evaluating the range in terms
of its climax status presupposes that the climax vegetation is the
most productive or most economical (64). It is encouraging to see
different ideas expressed by some workers in America (40, 45, 48),

New Zealand (49), and Uruguay (57).

How Can We Develop a Program of Revegetation?

This question may have been directed to the broad program
after basic research information is available. It will be discussed
in two parts: research and development.

Avenues of Research

There are several aspects to revegetation, any one of which 1s
more applicable to some land than to others. On some ranches a
combination may be used with great benefit to the whole opera-
tion. These aspects are fertilization, use of competitive plants,
seeding of long-lived species, and special problems.

1. Fertilization of Otherwise Unimproved Range. Research in
California has shown two ways that fertilization can be used to
improve forage yields: first, the use of phosphorus or sulfur to
increase the yield of resident legumes (2, 9, 10, 38); second, the
use of nitrogenous fertilizers alone or in combinarian with phos-
phorus and sulfur to increase total production and also to provide
feed earlier in the season (25, 33, 38, 46). In extensive tests Martin
and Berry found that yields of meat could be increased two- to
fivefold, the better soils providing the lower increases and the
poorer soils, the higher increases. Conrad (10) has developed an
“exploratory” test method for this purpose (Figure 1).

2. Fertilization of Improved Seeded Range. his will be dis-
cussed in the following paragraphs.

Use of competitive plants. The fact that on many soils annual
legumes respond to phosphorus or sulfur fertilization makes possi-
ble an important development in range improvement. A dense

358 THE FUTURE OF ARID LANDS

Agee ee B C D
HITTIN
| SSSASX_EE
NS [2 aT SINT NINN NISSEN
ae PL NNN NINN SR
SS SS NINNNINNNIS SE
) ee Be 2 I INISINISINISSISSS SNS 9
ee ea La NIN NINN
a Le NINN NO
ee SS SSSA SNNNNNSSNSS SSNS
ee Las INI NI NINN NIN SN
VA MAMMA VAVAVAVAVAVAVAVA VN OOOOOAOA
LLLLLL LLM MV VV PPP PP PRPRPLQQRLG
LLLLLLL LLY VV VV VV IRAP PPP RK
3 ZZZZZZZ VV VV VV VV IIA RRR KR] 3
VLLLLLL LMM PVP PPP PIT
ZLLL ZL LAN V VV VV VRP RAR RRR
LLL LLL LEV VV VV RRNA RRS
EEE EE EER pret braecccs
HORE Sess
LAV VAVAVAV AV AVA OOO OOO OG
WADED RPA Po
PPP PPP LR RRR
OOOO AD cicrcrririnininicecocececestcee
WO RS nl seatatetaraconel
y >
MAMA Shes
B C D

Figure 1. Exploratory layout to determine nutrient deficiencies on
the range by plant response.

The following materials and amounts are suggested for each 60-by-80-foot strip:

Horizontal strips: 1-1, nothing applied horizontally. 2-2, 8 pounds treble superphosphate.
3-3, 8 pounds treble superphosphate and 8 pounds muriate of potash in turn applied uni-
formly. 4-4, 8 pounds muriate of potash.

Vertical strips: A-A, nothing applied vertically. B-B, 16 pound, gypsum. C-C, 16 pounds
gypsum and 4 to 8 pounds ammonium nitrate. D-D, 4 to 8 pounds ammonium nitrate.

Just before the stakes are pulled if two or three handfuls of borax are spread in a circle
of about 2 feet diameter around each stake as indicated by an open circle, the killed growth
makes the plot easy to find later.

population of annual legumes, together with concentrated grazing,
will reduce the population of undesirable annual grasses and sum-
mer weeds to a remarkable extent. Therefore, fertilizers and soil
amendments should be tried along with the annual legumes.

BETTER ADAPTATION OF PLANTS 359

Annual legumes are perhaps the most promising species for im-
proving and increasing returns per acre in arid areas. They are
more nutritious than annual grasses, whether green or mature, and
they provide more flexibility to the year-long livestock operation.
In some instances seed can even be broadcast into resident sod
without seedbed preparation (69).

Seeding of long-lived species. Begin tests on arable land, if at
all possible, and use the best agronomic methods to establish a
stand (42, 43). We have found that a summer crop such as sudan-
erass (Sorghum vulgare var. sudanenstis) 1s ideal preparation for a
dryland pasture seeding. The sudan provides some summer feed,
and the grass and legume seeds can be planted directly into sudan
stubble without further seedbed preparation. Actual seedling
counts at the University of California’s Hopland Field Station
indicated that 19 times as many seedlings of Phalaris tuberosa var.
stenoptera were established following sudan as following wheat.
This is primarily due to the reduction in competition of resident
species by the sudan crop.

Adapted perennial grasses are worth striving for since they
have a much longer season of use than annual forage (33, 34, 38,
40, $1). Again, fertilization together with judicious grazing in-
creases yields of perennials. In an experiment at the University
of California’s Hopland Range Field Station 400 pounds per
acre of ammonium-phosphate-sulfate fertilizer increased peren-
nial forage production four times over the 200 pounds per acre
application. Fertilization the second year after seeding yielded
two-tenths pound per day of sheep weight gains compared with
one-tenth pound unimproved, unfertilized range at the Station.

Fertilization, combined with controlled grazing, not only
increases forage production but may also improve its quality by
creating a more favorable environment for the desirable species.
Many of the species on lands of low fertility are “invaders”
which by their very nature are the least desirable plants from the
standpoint of livestock production. They may be highly desirable
from the standpoint of soil conservation, but man cannot afford
to wait for the successional trends that over the centuries will
result in an improved cover.

A rotational system of grazing should be designed to harvest

360

THE FUTURE OF ARID LANDS

Water ‘‘O”

Early Close Grazing
First Year

Water “O”

Early Close Grazing
Second Year

Water ‘‘O”

Early Close Grazing
Third Year

Figure 2. Standard seasonal grazing rotation plan.

the crop when most advantageous. This calls for at least three
subdivisions of the range. Each subdivision is grazed heavily one
year out of three at the time of maximum flush of growth to con-
trol the undesirable species as outlined by Jones and Love (33).
This system has been further discussed by Love (40) and Murphy
et al. (§2) (see Figure 2). Jones (formerly Extension Agronomist)
and Love made a survey of hundreds of test plots in California
in 1942 and 1943. They had been seeded in the years 1937 to
1942 by the Agricultural Extension Service under a program
designed by B. A. Madson and Jones. (It is a pleasure here to pay
tribute to these two men who started to teach Californians that
there are differences in grasses). This survey showed that many
more failures could be attributed to exclusion of livestock (i.e.,
protection) than to excessive grazing. In the protected plots the
weeds gained control. These observations have been amply veri-
fied since.

Special problems. In California, as in some other regions,
brush is an undesirable plant that reduces forage yields. Fire,
herbicides, and mechanical equipment may be used alone or in
combination to get rid of the brush in preparation for revegetation
by desirable grasses and legumes (41).

Development of Program of Revegation

Two aspects of a development program are private and public.
The individual owner must decide on the program to be de-

BETTER ADAPTATION OF PLANTS 361

veloped on his property. This will be determined by his needs and
opportunities, and of course will have to fit in with other farm
operations such as grain growing and irrigated pastures. Inci-
dentally, experience has shown that in California about ten acres
of unimproved range are required to supplement each acre of
irrigated pasture (33).

Non-arable land may be fertilized to provide early feed. Arable
land that cannot be irrigated may be seeded to legumes to provide
late green feed or feed during the dry season (44). The better soils
may be seeded to mixtures of perennial grasses and legumes to
extend the green feed period.

Governments may assist in a number of ways. More money
put into research programs would undoubtedly pay dividends.
At the farm level, such programs as the U. S. Department of
Agriculture’s Agricultural Conservation Program is helpful in
assisting farmers financially.

What Are the Economic Possibilities?

A few examples will be given from California experience.

In 1950 when the first farmer began his project of converting
poor land to annual clovers his cost for land preparation, fertiliz-
ing, and seeding were about $20.00 an acre. Carrying capacity the
first spring following seeding was trebled, returning at grazing
rental rates $15.00 an acre. Aftermath grazing added another
$5.00 an acre to the returns. By 1952 his costs were down to about
$15.00, primarily because the new clover seeds were more abun-
dant and cheaper. Another farmer seeded and phosphated 350
acres in 1953 at a cost of $8.00 per acre. His net return (based on
cattle gains) was about $10.00 an acre the first year. A herd of
700 animals had access to the 350 acres for 50 days in the spring
and averaged 46 pounds more than a comparable herd on unim-
proved range throughout the season.

A perennial grass-annual clover mixture produced 79.3 pounds
of beef per acre July 12 to October 3, 1953, in addition to a grazing
period in the spring and about 120 sheep-days per acre cleanup
later that same fall. The gross return that year alone was well
over $20.00 per acre. With trial plots as a guide, there is no ques-

362 THE FUTURE OF ARID LANDS

tion of economic returns resulting from fertilization of unimproved
ranges.

What Are the Possibilities of Maintaining Larger Human Popula-
tions in Arid Areas?

The answer to this question seems to follow directly from the
preceding discussion. It has been shown that arid lands in a
Mediterranean type climate can be made from two to ten times
more productive by improved agricultural practices. This makes
the livestock economy a more efficient one, and if one acre can
be made to produce what now takes two to ten acres, a larger
human population can be supported.

It was pointed out, above, that one of the difficult problems
confronting the range manager is efficient harvesting of the forage
crop. Because animals are “selective’’ in their grazing habits, that
is, when they have a choice they will first graze the most palatable
plants. On a so-called overgrazed range, reducing livestock
numbers for the season 1s of little avail since the animals can still
select the most palatable plants first. Such a policy will result only
in continuing the trend of range depletion.

A three-field, three-year rotation grazing plan for California
foothill ranges has been described earlier in this paper.

The California Forest and Range Experiment Station con-
ducted five years of research on a five-field, five-year rotation
grazing plan at their Burgess Spring Experimental Range on the
Lassen National Forest in Northeastern California (A. L. Hor-
may, unpublished). This is in the Great Basin zone where the
grazing season is June 1 to September 30.

The California Region, V, of the U. S. Forest Service has put
this plan into operation on a 32,352-acre allotment, of which 20,
645 acres are usable as range. The allotment is fenced into five
fields of approximately equal carrying capacity. During the §-year
period each of the range units receives a different grazing treat-
ment. The timing of heavy grazing and resting is based on the
growth requirements of Festuca idahoensis (Idaho fescue)—the
key species in this allotment. This plan has been put into effect
with no reduction in livestock numbers, and it is confidently
expected that grazing capacity will be increased in later years.

BETTER ADAPTATION OF PLANTS 363

This additional example has been given to encourage others to
plan similar rotational grazing systems based on the growth
requirements of the key species. Intensive grazing at appropriate
periods, followed by resting, will assuredly result in more efficient
use of the range resource.

Conclusions

Everyone interested in the development of higher producing
grasslands and the more efficient utilization and conversion of
the feed grown should read C. P. McMeekan’s talk presented at
a plenary session of the Sixth International Grassland Congress
(49). He presented the philosophy underlying the interdependence
of grassland and livestock. He concluded: ‘“‘The most effective
use of grassland frequently involves the temporary abuse of both
pasture and animals, and [I] make a plea for the clear recognition
of this by those interested primarily in the pasture itself. Grass-
land improvers must also never forget that pasture is useless un-
less usable. I believe the recognition of these two facts is essential
if further development of our grasslands is to result in healthy,
productive swards in terms of animal use.”

REFERENCES

1. Aamodt, O. S. 1935. A machine for testing the resistance to injury
by atmospheric drought. Canad. F. Research C12, 788-795.

2. Bentley, J. R., and L. R. Green. 1954. Stimulation of native annual
clovers through application of sulfur on California foothill range.
F. Range Management T, 25-30.

3. Brewer, W. H., and S. Watson. 1876. Geological Survey of California
Botany, Vols. I and II. John Wilson & Son, University Press,
Cambridge, Mass.

4. Clausen, J. 1952. New bluegrasses by combining and rearranging
genomes of contrasting Poa species. Proceedings 6th Intern. Grass-
lands Congress 1, 216-221.

5. Clausen, J., D. D. Keck, and W. M. Hiesey. 1940. Experimental
studies on the nature of species: I. Effects of varied environments
on western North American plants. Carnegie Inst. Wash. Pub. 520,
Taig oe

6. Clausen, J., D. D. Keck, and W. M. Hiesey. 1945. Experimental
studies on the nature of species: II. Plant evolution through am-
phiploidy and autoploidy, with examples from the Madiinae.
Carnegie Inst. of Wash. Pub. 564.

364

Wi-

119),

THE FUTURE OF ARID LANDS

Clausen, R. E. 1941. Polyploidy in Nicottana. dm. Naturalist 15,
291-306.

Clements, F. E. 1920. Plant indicators: the relation of plant com-
munities to process and practice. Carnegie Inst. of Wash. Pub. 290,
388.

. Conrad, J. P. 1950. Sulfur fertilization in California and some related

factors. Soil Sct. T0, 43-54.

. Conrad, J. P. 1951. Fertilization of range forage. Calif. Agr. 5, 2:

34:

Cooper, J. P. 1951. Studies on growth and development in Lo/ium:
II. Pattern of bud development on the shoot apex and its ecological
significance. 7. Ecol. 39, 228-270.

Cooper, J. P. 1954. Studies on growth and development in Lo/ium:
IV. Genetic control of heading responses in local populations. 7.
Ecol. 42), 2: 521-556.

. Davy, J. B. 1902. Stock ranges of northwestern California. U. S.

Bur. Plant Ind. Bull. 12, 1-81.

. Dillman, A. C. 1946. The beginnings of crested wheatgrass in

America. F. 4m. Soc. Agron. 38, 237-250.

. Duisberg, P. C. 1952. Development of a feed from the creosote bush

and the determination of its nutritive value. F. Animal Sci. 11,
174-180.

. Frandsen, K. J. 1952. Theoretical aspects of cross-breeding systems

for forage plants. Proc. 6th Intern. Grasslands Congress 1, 306-313.

. Franzke, C. J., and J. G. Ross. 1952. Colchicine induced variants in

sorghum. ¥. Heredity 43, 107-115.

. Gregor, J. W., and F. W. Sansome. 1927. Experiments on the ge-

netics of wild populations: I. Grasses. 7. Genetics 17, 349-364.

. Harlan, H. V., and M. L. Martini. 1938. Effect of natural selection

on a mixture of barley varieties. 7. dgr. Research 51, 189-199.

. Hendry, G. W. 1931. The adobe brick as a historical source. Agr.

JalGlin Diy BE WUC=OM.

. Hendry, G. W., and M. K. Bellue. 1925. The plant content of adobe

bricks. Calif. Hist. Soc. Quart. 4, 361-373.

. Hendry, G. W., and M. K. Bellue. 1936. An approach to south-

western history through adobe brick analysis. Sympostum on Pre-
historic Agriculture, Flagstaff, Ariz. University Press, Albuquerque,
INE Vi Racesti— 3:

. Heyne, E. G., and A. M. Brunson. 1940. Genetic studies of heat and

drought tolerance in maize. 7. 4m. Soc. Agron. 32, 803-814.

. Hilgard, E. W. 1890. The weeds of California. Calif. Agr. Expt. Sta.

ING Dis AIOE,

. Hoglund, O. K., H. W. Miller, and A. L. Hafenrichter. 1952. Appli-

BETTER ADAPTATION OF PLANTS 365

cation of fertilizers to aid conservation on annual forage range. F.
Range Management 5, 55-61.

. Huskins, C. L. 1948. Segregation and reduction in somatic tissues:

I. Initial observations on Allium cepa. F. Heredity 38, 310-325.

. Ijin, W. S. 1930. Der Ursachen der Resistenz von Pflanzenzellen

gegen Austrocknen. Protoplasma 10, 379-414.

. Lyin, W. S. 1933. Uber Absterben der Pflanzengewebe durch Aust-

rocknung und ihre Bewahrung vor dem Trockentode. Protoplasma
19, 414-442.

. Julander, Odell. 1945. Drought resistance in range and pasture

grasses. Plant Physiol. 20, 4: 573-599.

. Knowles, P. F. 1943. Improving an annual bromegrass, Bromus

mollis \.., for range purposes. 7. dim. Soc. Agron. 35, 7: 584-594.

. Laude, H. H., and A. F. Swanson. 1942. Natural selection in varietal

mixtures of winter wheat. 7. 4m. Soc. Agron. 34, 270-274.

. Laude, H. M. 1953. The nature of summer dormancy in perennial

grasses. Botan. Gaz. 114, 3: 284-292.

. Jones, B. J., and R. M. Love. 1945. Improving California ranges.

Callie lint. Ser, G17. 129, 1-48.

. Love, R. M. 1944. Preliminary trials on the effect of management

on the establishment of perennial grasses and legumes at Davis,
California. 7. Am. Soc. Agron. 36, 699-703.

. Love, R. M. 1946. Interspecific hybridization in Stipa L.: I. Natural

hybrids, 4m. Naturalist 80, 189-192.

. Love, R. M. 1947. Interspecific and intergeneric hybridization in

forage crop improvement. 7. 4m. Soc. Agron. 39, 41-46.

. Love, R. M. 1951. California range plants. Calif. Agr. 5, 1: 8, To.
. Love, R. M. 1952. Range improvement experiments at the Arthur

Brown Ranch, California. 7. Range Management 5, 120-123.

. Love, R. M. 1952. The value of induced polyploidy in breeding.

Proc. 6th Intern. Grassland Congress 1, 292-298.

. Love, R. M. 1953. The use of livestock and competitive plants to

control weeds on annual ranges. Proc. 5th Annual Calif. Weed
Conference.

. Love, R. M. 1954. Range management standards. The Appratsal F.

22, 409-414.

. Love, R. M., and B. J. Jones. 1947. Improving California brush

ranges. Calif. Expt. Sta. Circ. 371, (rev. 1952).

. Love, R. M., V. P. Osterli, and L. J. Berry. 1953. Hardinggrass for

eneXe THeSeclhate, (CWI haps U5 WR. 5, els

ave whe ME Ve Ee Ostenliivandes | Berry. 195¢-. lmprovemyour

range with Harding. Calif. Ext. Serv. Leaflet.

. Love, R. M., and D. C. Sumner. Rose clover, a new winter legume.

Calif. Expt. Sta. Circ. 407, 1-12.

366

THE FUTURE OF ARID LANDS

. Love, R. M., and D. C. Sumner. 1952. California range plants.

Calif RAG SO i Qe ale

. Martin, W. E., and L. J. Berry. 1954. Will range fertilization pay?

Univ. of Calif. Agr. Ext. Serv. Mimeo. 1-31.

. Mather, K. 1949. Biometrical Genetics. Methuen, London.
. Mcllvain, E. H., and D. A. Savage. 1954. Progress in range manage-

ment. Advances in Agronomy 6, 2-65.

. McMeekan, C. P. 1952. Interdependence of grassland and livestock

in agricultural production. Proc. 6th Intern. Grassland Congress 1,
149-161.

. Meyer, B. S., and D. B. Anderson. 1941. Plant Physiology. D. Van

Nostrand Co., Inc., New York.

» Muller a. W.; OF K> Hoglund, and” Av Ic. Hatenrichtemsgsa

Reseeding to aid conservation of annual forage range. 7. Range
Management 6, 414-422.

. Murphy, A. H., R. M. Love, and L. J. Berry. 1954. Improving

Klamath weed ranges. Calif. Expt. Sta. Circ. 487, 1-16.

. Olmsted, C. E. 1952. Photoperiodism in native grasses. Proc. 6th

Intern Grassland Congress 1, 676-682.

. Parish, S. B. 1920. The immigrant plants of southern California.

S. Calif. Acad. Sci. Bull. 19, 4: 3-30.

. Raber, O. 1928. Principles of Plant Physiology. The Macmillan Co.,

New York.

. Robbins, W. W. 1940. Alien plants growing without cultivation in

California. Calif. Expt. Sta. Bull. 637.

. Rosengurtt, B. 1952. Las praderes naturales del Uruguay. Proc. 6th

Intern. Grassland Congress 1, 602-609.

colchicine-induced variants in sorghum. Agron. F. 46, 10-15.

. Runyon, E. H. 1936. Ratio of water content to dry weight in leaves

of the creosote bush. Botan. Gaz. 97, 518-553.

. Shantz, H. L. 1927. Drought resistance and soil moisture. Ecology

8, 145-157.

. Spoehr, H. A. 1919. The carbohydrate economy of cacti. Carnegie

Inst. Wash. Pub. 287.

. Stapledon, R. O. 1928. Cocksfoot grass (Dactylis glomerata L.)

ecotypes in relation to the biotic factor. 7. Eco/. 16, 71-104.

. Stebbins, G. L., Jr. 1952. Species hybrids in grasses. Proc. 6th

Intern. Grassland Congress 1, 247-253.

. Stoddart, L. A., and A. D. Smith. 1943. Range Management, Mc-

Graw-Hill Book Co., Inc., New York.

. Suneson, C. A., and G. A. Wiebe. 1942. Survival of barley and

wheat varieties in mixtures. ¥. dm. Soc. Agron. 34, 1052-1056.

66.
O7E
68.

69.

BETTER ADAPTATION OF PLANTS 367

Talbot, M. W., H. H. Biswell, and A. L. Hormay. 1939. Fluctua-
tions in the annual vegetation of California. Ecology 20, 394-402.
Turesson, G. 1930. The selective effect of climate upon the plant
species. Hereditas 14, 99-152.

Whaley, W. G. 1952. Arid lands and plant research. Scientific
Monthly 16, 228-233.

WallitamsseVVe AU URS IMS Wove) anci|ee Conrad 1056.) Rance
improvement in California by seeding annual clovers, fertilization,
and grazing management. 7. Range Management 9, 28-33.

Animals and Arid Conditions:

Physiological Aspects
of Productivity and Management

KNUT SCHMIDT-NiELSEN*

Duke University, Durham, North Carolina

The rapidly increasing amount of scientific information about
animals in arid climates now permits some general conclusions
which can provide a fruitful basis for further research on the
practical aspects of animal management and production.

From results obtained in recent research emerge principles
which could be easily applied to the thinking and the projects of
the practical research worker, and also by those concerned with
planning the future of arid lands and their populations. It is
evident that no more can the fate of man be left to chance. The
scientist’s responsibility is to make his conclusions available so
that his hypothesis can be put to the hard and rigorous test of
practical use. This paper is presented in the hope that it may
clarify how simple reasoning may transform scientific facts into
practical considerations.

How Can Animal Production Be Increased2
The yield of animal products can be increased in a number of
different ways, well known, but not always well practiced where

* The research on the physiology of the camel was done as teamwork
with my wife, Dr. Bodil Schmidt-Nielsen, and our collaborators Dr.
T. R. Houpt and Dr. S. A. Jarnum. It received financial support from
UNESCO, the John Simon Guggenheim Memorial Foundation, and
from scientific agencies of the United States Government.

368

ANIMALS AND ARID CONDITIONS 369

=
w
>
WwW
=
ww
2)
Zz
<
z
a)
| oa
=
<
2

EXPENDITURE

FOOD INTAKE

Figure 1. Production of animal tissue (cross-hatched area, labeled
“Storage””) can take place only when the level of feeding exceeds the
maintenance level. The line labeled “Basal Metabolism” refers to the
metabolic level when no food is taken in. Food intake raises the me-
tabolism, and the quantity of food required for maintenance will there-
fore exceed the basic metabolic level. With increasing food intake, the
energy expenditure increases steadily as indicated by the line ‘‘Actual
Expenditure.” The ‘Maintenance Level’ is where the “Food Intake”
equals the “Actual Expenditure.”

This is a simplified illustration and does not refer to any specific data.
The slope of the curve “Actual Expenditure” includes expenditures for
obtaining the food, for digestive processes, and waste. If the vegetation
is scattered, more work is expended to obtain it, and production is
decreased.

they are most needed. Production (which in the following refers
to production of consumable animal foodstuffs) can be increased
in two ways, either by increasing the amount of feed or by a better
utilization of the feed available. The first problem, increasing the
feed, is a problem for hydrologists in irrigation, for agriculturists
in selection and improvement of plants, and in fertilization, etc.
In this paper I shall mainly concern myself with the second prob-
lem, that of utilizing the feed available as natural vegetation on

370 THE FUTURE OF ARID LANDS

lands where agricultural improvements cannot easily be made.
However, I shall return to the question of increasing the amount
of available feed in a somewhat different way, with stress on the
word ‘‘available.”

Colonel Draz, in his paper in this symposium, has pointed out
the importance of raising the level of nutrition of the animals for
increased production. It is obvious that when animals are fed
just enough for maintenance, no production can take place. By
cutting down on the number of animals, the same amount of feed
will serve for production as well as maintenance for the rest of the
animals. How production increases with increasing level of feed-
ing, but only above a certain level, is shown in Figure 1.

One desirable way to increase actual production on a limited
feed supply would be to indoctrinate the animal owners in this
very simple fact of production physiology.

In Norway, as a young student, I tried to convince a country
woman of the poor economy of keeping chickens too long. She was
unwilling to kill any hen that was still producing some eggs and
was angele to see that they consumed an ever increasing amount
of feed as compared to her return in eggs. It must be equally
difficult to teach principles of production economy to populations
where wealth and social prestige is proportional to the number of
livestock.

It is well known that overstocking reduces productivity also
in another way, which has more serious proportions because of
its more far-reaching effects. I am referring to the effects of
overgrazing on depletion of vegetation, and the consequential
reduction of plant production, soil erosion, increased aridity, etc.
It may prove less difficult to inform people of this chain of events
than of the more subtle arguments of production economy.

The Physiological Role of Water

In arid lands feed is not the only limiting factor; water is a
precarious commodity that in the hotter part of the year takes on
the role of a limiting factor of all-encompassing importance. Here
also there are two ways out: providing more water or being more
economical with the available water.

ANIMALS AND ARID CONDITIONS 37]

Obtaining more water is not always simple. Large scale drilling
for water has been practiced in the southwestern United States
with effects on the water table that have been described as disas-
trous. A similar decrease in the water table in some Old World
arid areas would involve a much more precarious situation if wells
in villages and oases were starting to dry up. Obviously, large
scale irrigation cannot be started without careful consideration
of long range effects, but when found possible, it is a most efficient
way of handling the production problem.

The other possibility is that of using the available water to
better advantage. In the following I shall present some general
considerations of this problem, and in order to illustrate principles
I shall simplify matters as much as possible.

In winter, the water content of the vegetation is high. At the
same time the animals need relatively small amounts of water
because the temperature is low and water is not used for heat
regulation. Animals do not return frequently to the wells, and
may therefore graze over large areas. They can utilize well the
vegetation which is at its maximum productivity. The range of
the animals is restricted by management and herding problems,
and a nomadic (or semi-nomadic) management seems more
advantageous because it will permit the utilization of areas far
away from human settlements.

The amount of water in the plants may be so high that at least
the camel becomes completely independent of drinking water in
the winter time. We have offered water to camels that had been
without it for two months, and they would not drink. Subsequent
examination of blood, tissues, stomach, etc., showed them nor-
mally hydrated. In the summer the situation is different. The
vegetation dries up, and at the same time the animals need water
for heat regulation.

In all deserts one finds some mammals, mostly small rodents,
that seem completely independent of water. The American kan-
garoo rats and the Old World jerboas, for example, do not drink
water and can thrive indefinitely on only dry food.

Even the driest seeds contain some absorbed water, but a
larger quantity is formed by the oxidation of the food in the body.

372 THE FUTURE OF ARID LANDS

On oxidation one gram of starch yields 0.6 gram of water and one
gram of fat yields almost 1.1 grams of water. By exercising the
greatest physiological economy with water expended for urine
and feces and for evaporation (which cannot be completely
avoided because the expired air is saturated with water vapor),
those small rodents can just manage on the oxidation water,
being independent of intake of free water.

These small animals do not use water for heat regulation. They
are nocturnal and remain in their underground burrows through-
out the hottest part of the day. They are an ecological paradox,
living in the desert without being exposed to the rigor of desert
heat.

The large animals cannot escape the desert heat by hiding
underground. To avoid undue rise in the body temperature they
evaporate water from the surface of the skin (sweating) or from
the moist respiratory surfaces (panting).

The oxidation water which goes a long way for the small
rodents would not go far in the water economy of an animal
using water for heat regulation. The amount of oxidation water
formed in man is about one-quarter liter per day (on a metabolism
of 2,000 keal) and is insignificant when sweating rates may be
over I liter per hour or Io to 1¢ liters or more per day.

Water and the Camel’s Hump

This seems to be the place to discuss a widespread miscon-
ception with respect to the role of oxidation water in the water
balance of the camel. It has been said that since oxidation of fat
yields more than its weight in water (1.07 grams of water per
gram fat),-a camel that walks into the desert with a hump with
40 kg fat actually carries a potential water supply of more than
40 liters. This, of course, is true, and has led to the deceptive idea
of “water from the hump.” What was forgotten is that oxygen is
required to oxidize the fat. This involves ventilation of the lungs
and loss of water in the expired air. The amount of evaporation
from the lungs is of the same magnitude as the quantity of water
formed. In very dry air it would exceed the oxidation water, and
even in moderately dry air there will be no appreciable gain.

ANIMALS AND ARID CONDITIONS 373

TABLE 1

Comparison of Foodstuff Used, Oxygen Taken up, and Water Formed
Metabolic level 10,000 keal

Oogles Ohideiion EOS Nees

Foodstuff used for Oxidation Water ae from
kg of Food Formed De etee
liter kg Disy Bute
kg
Fat 1.06 2130 Teng ots
Starch 2.39 1980 TGS 17

One further and rather striking fact is that fat yields almost
twice as much oxidation water as starch. Seemingly fat is more
advantageous to the water economy than starch. However, fat
also yields more calories. This means that when the water yield
is related to metabolic rate, the apparent advantage disappears.

Let us make a simple calculation: A camel has a metabolic
level of, say, 10,000 kcal, and if he uses exclusively fat or exclu-
sively carbohydrate we can use the comparison presented in
Ialole i,

For a given metabolic level one finds that more water is formed
if starch is metabolized, than would be formed by metabolism of
fat. Also, the evaporation from the lungs, which is proportional
to the oxygen consumption, is higher when fat is burned. Al-
though water is formed in the oxidation of fat, the conclusions
must be that: (a) incidentally to the necessary oxygen uptake
water is evaporated from the lungs in an amount similar to that
formed in the oxidation process, (4) the evaporation from the
lungs is slightly higher when fat is metabolized, and (c) at a given
metabolic level starch would yield more water than fat. The bio-
logical truth is that fat 1s the most widespread form of energy
storage in the animal kingdom, and in this sense the camel is no
different from other animals. Fat gives more energy per weight
unit than other foodstuffs, and the economy in carrying the re-
serves as lightly as possible is indeed very useful, particularly to
an animal that may be deprived of an adequate food supply for
extended periods of time. The water economy of the camel, then,

374 THE FUTURE OF ARID LANDS

is not located in the fat of the hump. Strict economy in water
expenditure is the all important factor.

Water Economy of the Camel

It was mentioned before that the sweating rates of man in the
hot desert may be as high as 16 liters per day or more. Compared
with this quantity, the water lost in urine is quite insignificant.
The minimum urine output in man is about 300 ml per day, and
if the kidney were more efficient and could produce twice as con-
centrated urine, it would be possible to save 150 ml of water, or
one per cent of the amount evaporated. While such an efficient
kidney is essential to the kangaroo rat, the relative saving is unim-
portant in a large animal that uses water for heat regulation. The
quantity of water used for evaporation may be tremendous, and
if some economy could be accomplished in this amount it would
be of great significance.

When the temperature of the environment rises above that of
the animal (more precisely, that of the skin surface) heat will be
conducted to the animal from the hot surroundings by conduc-
tion from the air and by radiation from the sun and the hot
ground. Body temperature can be kept from rising only by evapo-
ration of water.

It is a consequence of the simplest physical laws that less heat
will reach the animal surface if there is an insulating layer be-
tween the heat source and the body. There is a great deal of
physiological truth in the old saying that the Arab wears so
many garments to exclude the desert heat. The woolly coat of
the camel has a similar function in the summer. In our recent
experimentation we showed that the camel’s fur was an efficient
factor in reducing water loss in the summer. Of course, the fur is
also an efficient insulation against loss of body heat in the winter.

Another and major factor in the water economy of the camel
is the variation in his body temperature. In the summer a camel
may have a morning temperature of 34° and an afternoon maxi-
mum of 40.6° to 40.7°. Man, when exposed to a hot environment,
will, by evaporation of water, maintain a practically constant
body temperature of about 37°. The variability of the camel’s

ANIMALS AND ARID CONDITIONS 375

body temperature has a double advantage. In hot environments
the camel, instead of evaporating water, permits the body tem-
perature to increase to a maximum of about 40.6°. This increase
can be regarded as a storage of heat in the body, heat which can
be dissipated in the cooler night without expense of water. But
there is one further important advantage to the high body tem-
perature. The difference in temperature between the hot environ-
ment and the body will be smaller, and since the movement of
heat is proportional to the temperature difference, less heat
reaches the body and less water is required to prevent a further
increase in body temperature.

This reaction, which is a well-regulated physiological mecha-
nism, effects a very considerable economy with water in the two
ways Just mentioned.

Length of Time Without Drinking

The time an animal can tolerate lack of water in the desert
depends on the rate of water loss, and the limit to which actual
desiccation of the body can be tolerated. We have just seen some
of the physiological mechanisms which result in an exceptionally
low rate of water loss in the camel. They are so efficient that the
rate of water loss in this animal in the summer is less than one-
third that of the donkey, which is also well adapted to the desert
climate.

The camel can also withstand an exceptional degree of dehy-
dration of the body. In one case a camel had lost more than 40%
of its body water before it was allowed to drink. This should be
compared with the situation in man and other mammals which
die from explosive heat rise when 15-20% of the body water have
been lost in hot surroundings. However, the donkey can tolerate
as much depletion of the body water as the camel, showing that
this animal too is exceptionally well adapted to desert life.

The camel, since it has a low rate of water loss and tolerates
considerable depletion of the body water, can go for a long time
without drinking. How long he can go without water depends
not only on external conditions of air temperature, wind, solar
radiation, etc., but also on how much he works, how far he has

376 THE FUTURE OF ARID LANDS

to walk and at what speed, what load he carries, what feed he
eats and its water content, what grazing he can do, etc. It is
therefore meaningless to discuss whether a camel can go for 5
days or 10 days without water; as we have seen he can go entirely
without water in the winter.

Drinking Capacity

When the camel drinks he fills up with water in a short time.
On one occasion a camel that weighed 325 kg (when dehydrated)
drank 103 liters of water in less than 10 minutes. The water be-
comes evenly distributed in the body in less than two days. The
blood and tissue fluids become rapidly diluted to an extent that
would not be tolerated by other mammals, which would die from
water intoxication at a much lower water intake. This difference
poses a number of important physiological questions that are
being further investigated.

One important observation is that the camel, as well as the
donkey, drinks an amount of water corresponding to the amount
of water depletion, but they do not drink more than that needed
to bring the water content of the body back to normal. In other
words, there is no extra intake of water that could be regarded
as a storage, or a supply to be drawn on when need arises.

The legendary structure of the camel’s stomach, which as early
as in Pliny’s Historia Naturalis was interpreted as serving water
storage, serves no such purpose. When carefully examined it 1s
clear that such a function would be extremely unlikely. The
glandular rumen sacs that are supposed to hold the water, could
not possibly store a significant amount. Furthermore, they con-
tain more solid food than the major part of the rumen, and the
fluid which can be obtained from them has the same salt concen-
tration as the general body fluids. However, this does not contra-
dict the widespread tale that an Arab in an emergency will kill
his camel and drink the fluid in the stomach. The fluid would
serve well in an emergency, and one finds abundant amounts of
fluid in the rumen of the camel, just as in other ruminants. The
mistake is made in implying that it 1s stored water.

ANIMALS AND ARID CONDITIONS 377

Nitrogen Utilization on Low Grade Feed

The rumen, however, serves another important function in the
camel. Investigations on the physiology of renal function in the
camel revealed an extremely low excretion of urea when the feed
was low in protein. In one particular camel fed on dry dates and
hay, the total urea excretion was less than 1 gram per day, cor-
responding to the metabolism of about 2.5 grams protein per
day. There is no reason to believe that the actual protein metabo-
lism would be that low in such a large animal. If one remembers
that urea (or ammonium compounds) when fed to cattle is syn-
thesized into amino acids by the bacterial flora of the rumen, and
that cattle can be fed a major part of their “protein” supply in
the form of urea, one sees what may happen in the camel. Urea
retained from secretion in the urine may be re-used as protein
via bacterial synthesis in the rumen.

When urea was injected in large amounts in a camel, less than
one-tenth was recovered in the urine. Stomach samples showed
an increased urea and ammonia content, and it is reasonable to
conclude that protein was synthesized in the rumen as it 1s in
cattle fed urea.

The value of this ability to re-use the protein nitrogen is ob-
vious to any animal husbandry man who works with low grade
pastures. It is not unlikely that other ruminants would behave
like the camel, and we are now in the process of investigating
whether the sheep can utilize a similar “urea cycle” when fed an
extremely low protein diet.

Role of Physiology in Production Considerations

If one summarizes the physiology of the camel the following
features stand out:

1. The camel can withstand an unusually high degree of water
depletion, and it also has a very low rate of water loss; in com-
bination, these two factors mean that the camel can go for ex-
ceptionally long periods without drinking.

2. The camel (as well as the donkey) has a very unusual drink-
ing capacity, which means that, when water is available, it can
fill up in a very short time.

378 THE FUTURE OF ARID LANDS

3. The camel (and the donkey) maintains its appetite in spite
of dehydration. While other animals will lose their appetites when
deprived of drinking water, the camel continues to eat until the
water depletion becomes severe.

The value of maintaining the appetite will be clear when one
considers that tolerance to water deprivation permits the animal
to graze over a large area, and continued food intake is the
essential to the utilization of the available grazing.

The effect of the range of the animal on the area available for
grazing is illustrated by the diagram in Figure 2. The areas that
can be covered by different animals are encircled, using as radii
the number of days the animals can go without water in the
severe desert summer (Sahara). (The figure for sheep is only an
estimate because we lack sufficient data.) In reality, the animals
do not remain away from water until the limit of endurance. In
the western parts of the Grand Erg Occidental of the Sahara, the
sheep are returned to the wells for watering every day in the
summer. The camels stay away for several days but will return on
their own initiative within four days. This is in the same propor-
tion as the maximal endurances, as indicated in the diagram by
the numbers on the radii.

(The actual areas covered may be even more to the advantage
of the camel because of the much greater speed of this animal.)

I would like to discuss separately the meaning of this diagram
to maintenance and to production.

In summer, vegetation and water supply are at a minimum,
and an important problem is to maintain an animal stock suffi-
ciently large so that later the winter grazing can be efficiently
utilized. In this case the problem is one of maintenance rather
than production. Assuming that the vegetation density is unt-
form over the areas represented in the diagram, one can calculate
the approximate relative numbers of animals that can be main-
tained, and likewise, as given in the diagram, the total weight of
meat maintained.

When the time of abundant feed comes, production is resumed.
At this time one might prefer a rapidly reproducing and growing
animal in order to utilize maximally the food available. However,
this reasoning is fallacious. It is true that production is more

ANIMALS AND ARID CONDITIONS 379

(MAX. 12 DAYS)

(MAX. 4 DAYS)

7 DONKEYS
630 KG

(MAX. 3 DAYS?)

is har

47 TPH Se
WOO
\ aay aL
‘ i
/

\
~~ FE

~N os
mae (0) SHEER 19 CAMELS
250 KG 8500 KG

Figure 2. Relative grazing areas for camel, donkey, and sheep.
The radii of the circles are in the same proportions as the number of
days (approximate) that the animals can go without water.

The relative numbers of animals that could be maintained on these
areas with a given constant vegetation density throughout the area,
have been calculated. For this calculation only maintenance require-
ments have been considered (and not growth or meat production), since
during the summer time satisfactory maintenance of a relatively large
stock is more important in order to have a large size herd for production
when vegetation is more easily available in the winter. (In this con-
sideration the question of overgrazing has been disregarded.) It is in
the summer that the grazing area is restricted by the access to water,
and this coincides in time with the lowest availability of vegetation in
the same area. In winter the grazing area of the camel is not limited by
water supply, and its range will then be determined by the practical
limitations of the human herders.

rapid in a small animal (sheep compared to camel), but the main-
tenance expense per kilogram is also higher. It has been shown
that the production per unit feed is approximately independent
of body size of the animal; in other words, for a given amount of

380 THE FUTURE OF ARID LANDS

feed, camels should produce approximately the same amount of
meat as sheep. Due to the larger grazing area, more feed is avail-
able to the camel and productivity should be correspondingly
higher.

It therefore seems that from the standpoint of production as
well as maintenance the camel would be far superior to the sheep.
(The donkey, now used only in small numbers, can be left out of
consideration because its meat, for religious reasons, usually is
not eaten.) The tremendously higher estimated production of the
camel is a point I would like to emphasize as a most important
subject for future practical research in order to test the validity
of the ideas presented here. From the theoretical considerations
it seems amazingly clear that the camel offers a most obvious
solution to increased meat production in arid zones with a low
natural vegetation density that cannot easily be increased. On
this point practical research is urgently needed.

Human and Social Factors in Management

The possible success of practical attempts to utilize the prin-
ciples just outlined for increased production will depend entirely
on the management of the herds with respect to avoidance of
overstocking. Some further human and social factors are also of
great importance.

In the areas I know from personal experience, camels are of
great importance for meat production. At the present time camels
are butchered when they are very old or when, for other reasons,
they cannot be kept without difficulties (for example, vicious
males or sick animals may be butchered young). When meat
production is the main purpose nothing is gained from maintain-
ing fully grown animals. It would be better economy to butcher
the animals young and maintain only the still productive ani-
mals. For this purpose, of course, all reproducing females are
productive, and it would seem most efficient to maintain a herd
of females with a sufficient number of mature (but young) males
for breeding. Other males should be regarded as having completed
the role of production. It may require much thoughtful educa-
tional work before such viewpoints gain general acceptance in
these areas.

ANIMALS AND ARID CONDITIONS 381

The following can be used as illustrative examples. I have seen
a vigorous and well-fed young male camel butchered because it
was too vicious to be handled. It brought the owner a price of
nearly 50,000 francs. I have also seen a very old scraggly female
camel being brought to butchery by its owner because unex-
pectedly it became pregnant and was not expected to survive it.
It brought 18,000 francs, probably half of what it could have
brought a few years earlier.

It would also be important to investigate the effect of castra-
tion of the males on the final yield of meat and fat, and whether
castration should be done in the very young animal or, for ex-
ample, a year before butchering.

Where low grade pasture cannot be readily improved, it can
be utilized to advantage along the principles outlined above only
in connection with nomadic or semi-nomadic life. The availability
of cheap labor for herding is a requisite condition for this type
of production.

It has often been said that modern means of transport, in par-
ticular the automobile, are rapidly making the camel superfluous.
This is true for long distance transport, where now even the
Sahara is crossed by a network of roads with rather regular bus and
truck traffic. On a smaller scale the camel still maintains great im-
portance in transport, but its use may in the future yield more to
the truck. However, in meat production on vast areas of low
grade arid lands the camel may maintain its role. In production
of meat the truck can never complete.

The truck may, however, depose the camel in this field, too,
in an indirect way. As a representative of the mechanized culture
the truck represents higher wages, fostering settlement and aban-
donment of primitive nomadic life. The situation is not unlike
the migration of the rural population to the big cities in so many
Western countries.

Many oases, at least in the Sahara, depend on the nomads for
meat production. Areas inaccessible to the settled populations can
be used for production by the nomads. Settlement leads to over-
grazing of the area immediately surrounding the settlement, unless
means and resources are available for adequate irrigation. One of
the great dangers in planning the future of arid lands is the belief

382 THE FUTURE OF ARID LANDS

that the Western way of life and a mechanized culture are always
a step forward to peoples now living close to a marginal existence.

REFERENCES

Specific references have been omitted in this paper. For questions of
production physiology, the interested reader is referred to S. Brody,
Bioenergetics and Growth, Reinhold Publishing Corporation, New York,
1945; for water balance questions to K. Schmidt-Nielsen and B. Schmidt-
Nielsen: Water metabolism of desert mammals, Physiol. Revs. 32, 135

(1952).

The Locust and Grasshopper
Problem in Relation to the
Development of Arid Lands

B. P. UVAROV

Anti-Locust Research Centre, London

Grasshoppers and their larger swarming relations, the locusts,
are the oldest and most serious enemies of agriculture. Since
their depredations are particularly great and persistent in the
regions with arid or semi-arid climate, it should be of interest to
present a brief outline of the locust and grasshopper problem as it
already exists in arid lands, and as it may be affected by ac-
celerated development of them.

Mosaic Vegetative Cover

An essential feature of the ecology of grasshoppers is that dif-
ferent stages of their life are passed in different environments.
Eggs are laid in the ground, usually in bare spots, whereas the
young hoppers and adults require abundant vegetation for food
and shelter. Therefore, a complete environment must include
two kinds of habitat: oviposition habitat and food-shelter habi-
tat. This implies that a patchy, mosaic vegetative cover is more
favorable to these insects than uniform vegetation.

The mosaic type of vegetation is most typically met with
where two major vegetation zones are in contact, e.g., forest and
grassland, prairie and semi-desert, or savanna and desert. It is
in fact in such transitional belts that grasshoppers and locusts
generally reach their maximum economic importance. Thus, in

383

384 THE FUTURE OF ARID LANDS

Siberia, the species of grasshoppers which occur commonly over
the steppe zone, become persistent pests mainly on its northern
fringe where it comes in contact with forests and a mosaic of
grassland and forest is formed. In Australia, the areas particu-
larly favorable to grasshoppers are characterized by a transitional
mosaic of low shrubs and bunchgrasses.

In Central Asia and in tropical Africa, the migratory locust
thrives on the edges of river flood-plains overgrown with dense
tall reeds and grasses but bordering on semi-desert grassland with
sandy bare patches. The mosaic structure of vegetation is, on the
other hand, often directly due to human activities, or very much
accentuated by them.

In Siberia, the clearance of forests for settlement and cultiva-
tion has created favorable conditions for the extension of steppe
grasshoppers into the former forest zone, while the fallow system
of agriculture, with disturbed land lying idle for several years,
has produced a mosaic of weeds, grass, and bare patches emi-
nently suitable for mass outbreaks of these pests.

In the dry hill pastures of Mediterranean countries, excessive
grazing and destruction of perennial shrubs for fuel and by goats
have resulted in a cover of short bunchgrasses, with abundant
bare spots, and here the Moroccan locust is enabled to thrive.
Similar examples of human activities favoring grasshoppers and
locusts are found in South Africa, Australia, and North America.
In the Philippines and Indonesia, deforestation and seasonal
burning have resulted in forest-grassland mosaic, enabling the
migratory locust to live and multiply where it could not even
exist before. In the Sudan, mechanized cultivation in the transi-
tion zone between the tall grass and short grass savanna has
caused several native grasshoppers to become important pests.

Effects of Human Activities

In all these cases, the general effect of human interference has
been to destroy the original uniform structure of vegetation and
to create a mosaic in which the more arid habitat was particu-
larly encouraged to expand, together with its grasshopper popu-
lation. The importance of such increase in the ecological aridity

LOCUST AND GRASSHOPPER PROBLEM 385

of the area lies in the fact that outbreaks of grasshoppers are
usually associated with a series of dry years. The effects of dry-
ness in favoring grasshoppers are, naturally, greater in an envi-
ronment which is partly arid naturally, or made so by man. A
feature of mosaic habitats, particularly those created by man, is
their instability, as the relative extent of their arid and more
humid parts is likely to vary from year to year. This has an 1m-
portant effect on grasshoppers, which are very mobile insects.
Even in most equable conditions, they move from the food-shel-
ter habitat to the oviposition sites, but when the contrasts are
very strong such movements extend and become migrations,
leading to a concentration of the insects in crops.

The result of migrations is especially striking in the case of
locusts. A relatively favorable season may cause a great increase
in a local population of locusts; if this is followed by drought,
this population would move and concentrate in the most favor-
able places. This creates crowding, to which locusts respond in a
most characteristic way—they acquire gregarious habits and
travel in dense masses. As the direction of flight of locust swarms
is largely dependent on winds, the swarms arising in one area
may travel great distances and invade fertile lands far from their
birthplace. An extreme case of this kind is offered by the desert
locust of Africa and western Asia. The home of this species is in
deserts which are generally extremely arid, but are liable to lo-
calized rainstorms. Such rains bring forth abundant ephemeral
vegetation, on which locusts can multiply rapidly. In a few weeks,
however, the plants wither away and the locusts, if they have had
time to grow up, must migrate elsewhere or perish.

Swarm flights follow seasonal winds and generally end in an
area where such winds bring rain, so that locusts can produce a
new generation many hundreds of miles away from the previous
one. The survival of such nomadic insects is clearly dependent
on the chance of swarms reaching an area where they can feed
and reproduce. In this respect also, man is already beginning to
make their life less hazardous; some areas in the Sudan and
Arabia, where natural vegetation is green for only a few months
in the year and that only if rain happens to fall, have been irri-

386 THE FUTURE OF ARID LANDS

gated for cultivation, and they now harbor locust populations
which otherwise would not have been able to survive without
migration. Irrigated crops in the southern United States, Argen-
tina, and Central Asia suffer from grasshoppers whose existence
in these arid lands was made much easier by man.

Development of Semi-Arid Lands

This much too brief survey of the problem should be sufficient
to show, first, how erroneous was the old belief that grasshopper
and locust plagues are a feature of undeveloped lands and almost
belong to the past. On the contrary, there is no doubt that the
grasshopper plagues in semi-arid lands have been created by their
much too rapid and thoughtless development, in the same way
as the deterioration of soil and its erosion. Moreover, even now
and even in the better developed semi-arid countries, the prob-
lem tends to become more and more acute, and ever increasing
efforts and expenditure are required to reduce losses from grass-
hoppers by chemical methods, which alone will never provide a
satisfactory solution. This experience of the already developed
semi-arid countries should serve as a warning to others where
schemes for the opening of vast new semi-arid lands are being
considered.

Development of Arid Lands

The development of the truly arid lands is only beginning, but
we have seen that it is almost certain to benefit such inhabitants
of the desert as locusts. Their life is full of hazards at present,
but should the ‘‘islands’” of cultivation in the desert become
large and numerous, they would save many a nomadic swarm
from starvation and provide conditions in which local popula-
tions of locusts could multiply.

Even apart from encouraging locusts directly, cultivation of
desert lands is certain to increase the risk of crop losses as a
consequence of making more crops exposed to locust attack.

All this does not mean, of course, that development of arid
and semi-arid lands should be discouraged. My object is merely
to draw attention to the need for taking into account its possible

LOCUST AND GRASSHOPPER PROBLEM 387

hazards, among which locusts and grasshoppers have already
proved to be outstanding. These particular hazards can be
avoided if they are carefully studied before it is too late. One
hopes that the future development of arid lands will not follow
the pattern of the semi-arid ones, where a rapidly expanding ex-
ploitation of the land preceded an understanding of its long-
range consequences. In the typical semi-arid country of North
Africa, the Moroccan locust 1s called in Arabic djerad-el-adami—
man’s locust. This name is appropriate to most locusts and grass-
hoppers, as it has been man who has encouraged and continues
to encourage them by short-sighted land usage.

REFERENCES

1. Andrewartha, H. G. 1943. The significance of grasshoppers in some
aspects of soil conservation in South Australia and Western Aus-
tralia. F. Dept. Agr. S. Australia 46, 314-322.

2. Bei-Bienko, G. Y. 1936. The geographical distribution on zones of
economic importance of the Moroccan Locust (Dociostaurus maroc-
canus Thnb.) in U.S.S.R. [In Russian.] [tog? Rab. vsesoyuzn. Inst.
ZLashch. Rast. 1935, 16-20.

3. Brown, E. S. 1947. The distribution and vegetation of egg-laying
sites of the Desert Locust (Schistocerca gregaria Forsk.) in Tripoli-
tania in 1946. Bull. Soc. Fouad Ent. 31, 287-306.

4. Canizo, J. Del, and V. Moreno. 1950. Biologia y ecologia de la
langosta mediterranea e marroqui (Dociostaurus maroccanus
Thunb.). Bol. Pat. veg. Ent. agric. Madr. 17 (1949), 209-242.

5. Chetyrkina, I. 1936. The geographical distribution and zones of
economic importance of the Italian Locust Calliptamus italicus L. in
Kazakhstan. [togi Rab. vsesoyuzn. Inst. Zashch. Rast. 1935, 20-22.

6. Davies, D. E., 1952. Seasonal breeding and migrations of the Desert
Locust (Schistocerca gregaria Forskal) in north-eastern Africa and
the Middle East. Auti-Locust Mem. 4, 56 pp.

7. Donnelly, U., 1947. Seasonal breeding and migrations of the Desert
Locust (Schistocerca gregaria Forskal) in western and north western
Africa. Anti-Locust Mem. 3, 43 pp.

8. Eig, A., 1935. Ecologie du Criquet Marocain on Iraq. Bull. En-
tomol. Research 26, 293-309.

g. Filip’ev, I. N. 1929. The locust question in Soviet Russia. [Vth
Intern. Congr. Entomol. 2, 803-812.

10. Fortescue-Foulkes, J. 1953. Seasonal breeding and migrations of the
Desert Locust (Schistocerca gregaria Forskal) in south-western Asia.
Antt-Locust Mem. 5, 35 pp.

THE FUTURE OF ARID LANDS

. Gunn, D. L. 1952. The Red Locust. 7. Roy. Soc. Arts 100, 261-

284.

. Joyce, R. J. V. 1952. The ecology of grasshoppers in East Central

Sudan. Aniti-Locust Bull. |4 +] 97 pp.

. Key, K. H. L., 1954. The Taxonomy, Phases and Distribution of

the genera Chortoicetes Brunn. and dustroicetes Uv. (Orthoptera:

Acrididae). Canberra, Div. Ent., C.S.I.R.O. 237 pp. Multigraph.

. Knechtel, W. K., 1938. Ueber die Wanderheuschrecke in Rumanien.

Bull. Entomol. Research 29, 175-183.

. Morant, V. 1947. Migrations and breeding of the Red Locust

(Nomadacris septemfasciata Serville) in Africa, 1927-1945. Anti-
Locust Mem. 2, 59 pp.

. Olsuf’ev, N. G. 1930. Zur Frage uber die Periodizitat der asiatischen

Heuschrecke. [In Russian with German summary.] Bull. P/. Prot.
ening ie Eents) leno i W477.

. Pasquier, R. 1934. Contribution a |’étude du Criquet Marocain,

Dociostaurus maroccanus Thnb., en Afrique mineure. (Ire Note.)

Bull. Soc. Hist. nat. Afr. N. 25, 167-200.

. Pena, F. de la 1942. Presente y future de la plaga de langosta en

Espana. Pudl. Serv. Lucha contra la langosta 16, 23 pp.

. Predtechenskui, S. A. 1935. The annual cycle of the Desert Locust

(Schistocerca gregaria Forsk.), its migrations and periodicity in
Persia and adjacent countries of tropical and sub-tropical Asia.

Bull Pl Prot. Leningrs (i. Ent.) No. 12; 35. pp:

. Predtechenskui, S. A. 1936. The geographical distribution and zones

of economic importance of the Migratory Locust (Locusta migratoria
L.) in U.S.S.R. Itogi Rab. vsesoyuzn. Inst. Zashch. Rast. 1935, 13-15.

. Rainey, R. C. 1951. Weather and the movements of locust swarms:
y 9

a new hypothesis. Nature 168, 1057-1060.

. Rao, Y. R. 1942. Some results of studies on the Desert Locust

(Schistocerca gregaria Forsk.) in India. Bull. Entomol. Research 33,
241-265.

. Remaudiere, G. 1954. Etude ecologique de Locusta migratoria

migratorioides Rch. & Frm. (Orth. Acrididae) dans le zone d’inun-
dation du Niger en 1950. Locusta No. 2.

. Richards, O. W., and N. Waloff. 1954. Studies on the biology and

population dynamics of British grasshoppers. duti-Locust Bull.
17 [6 +] 182 pp.

. Rubtzov, I. A. 1939. Climatic characteristics of the reservation

grounds and periods of mass increase of Siberian Acrididae. [In
Russian with German summary.] Prod/. Ecol. Biocenol. 4, 35-88.

. Tsou, T. L. 1935. The distribution of the Migratory Locust and

ecological study of its breeding ground in China. [In Chinese with
English summary.] dgricultura sinica 1, 239-272.

D7 fe
28.

29.

30.

ile

gon

34-

35:

36.

37:

LOCUST AND GRASSHOPPER PROBLEM 389

Uvarov, B. P. 1928. Locusts and Grasshoppers. A handbook for their
study and control. Imp. Bur. Entomol., London.

Uvarov, B. P. 1936. The Oriental Migratory Locust (Locusta mi-
gratoria manilensis, Meyen 1835). Bull. Entomol. Research 27, g1-
104.

Uvarov, B. P. 1951. Locust research and control 1929-1950. Col.
Research Pub. 10, H.M.S.O., London.

Uvarov, B. P. 1954. The Desert Locust and its environment: 85-89.
In J. L. Cloudsley-Thompson [Ed.] Biology of Deserts, Inst. Biol.,
London.

Vesey-Fitzgerald, D. F. 1955. The vegetation of the outbreak areas
of the Red Locust (Nomadacris septemfasciata Serv.) Anti-Locust
Bull. 20, 30 pp.

Waloff, Z. 1946. Seasonal breeding and migrations of the Desert
Locust (Schistocerca gregaria Forskal) in eastern Africa. Anti-
Locust Mem. 1, 74 pp.

Wet, W. J. de, and D. van V. Webb. 1952. Field observations on the
behaviour of hoppers of the Brown Locust in the swarming phase.
Sci. Bull. Dept. Agr. S. Afr., 337, 38 pp.

Zakharov, L. Z. 1932. The Locust problem in the Northern Caucasus.
(The present status and future prospects.) [In Russian.] Trad.
sev.-kavk. Inst. Zashch. Rast. 1, (8) (1), 3-13.

Zimin, L. S. 1934. On the biology and ecology of Calliptamus
turanicus Tarb. in Middle Asia. [In Russian.] In Acrididae of Central
Asia. Tashkent, Central Asiatic Inst. Pl. Prot., 82-112.
Zolotarevsky, B. N. 1933. Contribution a l’é ‘tude biologique du
Criquet Migrateur (Locusta migratoria capito Sauss.) dans ses foyers
permanents. Ann. Epiphyt. 19, 47-142.

Zolotarevsky, B., 1946. Les phases acridiennes et l’invasion du
Criquet Migrateur dans la Gironde. 4un. Epiphyt. 12 (n.s.), 1o1—

II4.

Desert Agriculture:
Problems and Results in Israel

MICHAEL EVENARI ann DOV KOLLER

The Hebrew University of Jerusalem, Israel

Phytogeographically and climatically, 36% (9,468 km?) of the
area of Israel belongs to the Mediterranean, 16.3% (4,288 km?) to
the Irano-Turanian, and 45% (11,835 km?) to the Saharo-Sindian
territory (17, 18). This means that even the best agricultural re-
gion of Israel, the Mediterranean region, is semi-arid, and the
rest of the country 1s classified as either arid or very arid. Conse-
quently, our agriculture had, from the very beginning, to deal
with arid zone problems, and we can consider the whole country
as a large-scale experiment in problems of aridoculture.

The ever increasing population of Israel has prohibited the lim-
itation of our agriculture to the best regions, and forced its
extension into our Irano-Turanian steppes and Saharo-Sindian
deserts.

That this was really done is clearly seen from the figures for
the Negev* (Table 1).

Whereas in the northwestern part of the Negev agriculture is
based on supplementary irrigation with water piped there from
northern Israel, it was decided that in the rest of the arable Negev
all agricultural practice will be based primarily on the use of local

* The Negev is the southern part of Israel and roughly comprises the
Jrano-Turanian steppe region in the north and the Saharo-Sindian desert
region in the south. The whole region receives only winter rainfall. The
yearly average for the Irano-Turanian region fluctuates between ca. 150
and 300 mm, for the Saharo-Sindian between ca. 25 and 125 mm. As in
all desert regions, rainfall is very uncertain as to amount and season.

390

PROBLEMS AND RESULTS IN ISRAEL 391

TABLE 1
Number of Agricultural Settlements in the Negev between 1945 and 1955
Phytogeographical region 1945 1950 1955
Irano-Turanian II 38 46
Saharo-Sindian 2 8 15

rain water. The latter is what we call ‘‘desert agriculture” in the
following pages.

The reason for using only the natural precipitation is twofold:

1. The relative scarcity of one of the most important natural
resources of Israel, i.e., water, necessitates its more rational use.

2. Irrigation water from the north cannot economically be
raised the necessary 500 meters and conducted over great dis-
tances to the few arable patches scattered here and there in the
Negev.

The whole project aims at trying to collect information about
the possibility of maintaining a larger population in an extremely
arid area, by the sole use of its natural water resources, and is,
therefore, highly experimental. The results which this project
would achieve may be of universal importance in similar arid
zones elsewhere.

The first results of five years of experimental desert agriculture
carried out by a large team of research workers* will be summed

* The main cooperating agencies are:

1. The Israeli Ministry of Agriculture (Departments of Water
Utilisation, Soil Conservation and Ecology).

2. The Israeli Ministry of Development.

3. The Israeli Government Department of Antiquities.

4. The Israeli Government Meteorological Service.

5. United States Operations Mission in Israel, Sections of Range
Development and Water Spreading.

6. The Hebrew University of Jerusalem and its Departments of
Botany, Geology, Geography, Archaeology and Agriculture.

7. The Technion in Haifa (Divisions of Hydraulics, Agricultural
Engineering, and Soil Conservation).

8. The Weizmann Institute of Science in Rehovot.

g. The Agricultural Experiment Station in Rehovot, especially its

departments of Soil Science and Field Crops.
(Footnote continued on page 392.)

392 THE FUTURE OF ARID LANDS

up here. Special stress will be laid on the problems. Most of these
could not be seen at the start, and evolved only during the course
of the work. Based on the experience gained during this work, it
will be possible to evaluate more realistically the future possibili-
ties of desert agriculture and its contribution to the solution of the
human food problems.

The first lesson to be learned—important in the past and even
more so for future work—was that desert agriculture is possible
only when preceded by planned, thorough, scientific fact finding
about the area involved. Geological, pedological, biological (es-
pecially phytogeographical, phytosociological, and plant physio-
logical), meteorological, and archaeological surveys must be
carried out. These surveys have to be done cooperatively. Vital
meteorological data, for example, which could not, in our case, be
based on decades of records, could be worked out only on the
evaluation of ecological and phytosociological observations. For
this purpose the method of “shifts in amplitude,” using the IE-
amplitude of certain indicator plants (7-11) or the distribution
and limits of distribution of certain plant associations (g-I1, 42,
45, 46, 48) are very useful. We may point out here that ecology,
phytogeography, and phytosociology are very important auxiliary
disciplines for all surveys in general. Thus, for instance, no pedo-
logical survey is complete without the donthrien map of the
main plant associations (42, 48), and the most important demarca-
tion of the borderlines of arable zones can be done only by ob-
serving the yearly fluctuations of the annual vegetation (10, 47).
These surveys should, therefore, never be an ad hoc compilation of
facts but an integrated product of scientific teamwork, without
emphasizing only immediate practical needs.

10. The agricultural settlements of Sde Boker (in the Negev High-
lands), Mashabei Sade, and Revivim (on the loess plains), Yotvata and
Fin Yahav (in Nahal Arava).

Special thanks are due to the Ford Foundation. A great part of the
work reported here could be carried out only with the help of the generous
grant given us by the Ford Foundation. By doing so the Ford Foundation
tes greatly furthered our knowledge of desert agriculture with all the
practical consequences thereof.

PROBLEMS AND RESULTS IN ISRAEL 393

As a result of these surveys we possess today, inter alia, a sound
geological knowledge, a detailed phytogeographical, ainuogne ite
logical, and pedaloctcal map of the Negev, transforming an area,
which twenty years ago was largely terra incognita to science,
into a scientifically better known area than even certain parts of
Europe.

One may be astonished to find archaeology in the above-men-
tioned list of necessary surveys. As will be shown later, the
archaeological survey has made a most valuable contribution to
the solution of practical agricultural problems. The study of
archaeology should, therefore, never be neglected in regions where
previous civilizations existed in desert areas.

The practical importance of these integrated surveys as the
first essential in desert agriculture cannot be overemphasized.
Whenever we tried to jump ahead and to find empirical solutions
without possessing the necessary data, we were faced with costly
failures that could have been avoided had necessary scientific
information been collected and analyzed before taking the first
practical step.

The second essential, we learned, was that any team working on
a project of desert agriculture has to be extremely versatile and
elastic. Some of the main practical problems could not be foreseen
when the working plan was made. They emerged, sometimes in
the most unexpected way, only during the course of the practical
work. The research team, therefore, has to be ready and scien-
tifically capable of dealing with all kinds of unexpected problems
by either tackling them themselves or by enlarging the circle of
cooperating agencies. Since most of the problems which have to
be solved are scientifically fascinating as, for instance, the problem
of germination, it is sometimes difficult to remember that practi-
cal results are the goal and that we cannot allow ourselves to get
lost in most interesting, purely scientific research.

In the following pages we shall deal with some of the main
problems of desert agriculture. Emphasis will always be laid, in
accordance with the main theme of this conference, on the lesson
to be learned from it for future work and not so much on the
results achieved.

394 THE FUTURE OF ARID LANDS

Germination and Vegetative Propagation (27, 28)

It became clear, when we started, that for all reseeding work a
thorough research on the germination conditions of the seeds
involved is of paramount importance. We can state today with
certainty that any reseeding project in desert areas is doomed to
failure when it does not provide for a special systematic research
on germination and propagation problems. We should like to
illustrate this point by citing a few examples.

In the course of an investigation on the use of the wild growing
plants of the Negev for industrial purposes it was found that
Funcus maritimus, a perennial hydrohalophyte, produces excellent
raw material for the paper industry. In order to propagate this
plant, seeds were collected, and field sowing was carried out in the
Negev. It failed completely, as the seeds did not germinate. The
problem was then handed over to our germination laboratory
where it was found that 7. maritimus is an obligatory light ger-
minator (between 20° and 30°C). Treatments with thiourea and
potassium nitrate, which are known to promote dark germination
in light-requiring seeds, did not abolish this need for light. It was
concluded that ¥. maritimus could be propagated by seed only
by means of surface sowing, in spite of the agrotechnical difh-
culties involved.

A method was therefore worked out to sow the seeds in shallow
ditches on top of continually moistened soil. This was carried out
in Yotvata (Nahal Arava) with very satisfactory results.

It is interesting to note that a chloride content of ca. 4,800 ppm
in the water did not affect germination. In sodium chloride solu-
tions containing 7,000 ppm chloride, germination percentages
decreased, but the germinated seedlings were not injured, al-
though the chloride concentration had eventually increased twice
to threefold by evaporation during the tests.

During the search for good native pasture plants it was found
that Colutea istria, a perennial legume growing wild in the dry
stream beds (wadis) of the Negev highlands, is an excellent pasture
plant. Its germination is, however, inhibited in two ways: the
seed coats are impermeable to water and they also contain a
water-soluble growth inhibitor which considerably retards seed-

PROBLEMS AND RESULTS IN ISRAEL 395

ling growth. This retardation of growth is a very important factor,
as germination and seedling growth have to be very rapid in order
to establish this plant in its climatically unstable natural sur-
roundings. A practical method had, therefore, to be developed
for the removal of the coats before sowing the seeds.

Oryzopsis miliacea 1s a perennial grass, valuable both for
pasture and soil conservation, various ecotypes of which are
native to nearly all regions of Israel. In the northern part of the
country reseeding of the plant in fall was practiced. The seeds
germinated gradually during early spring. For reseeding under
desert conditions, where soil moisture is available for much
shorter periods, such prolonged germination 1s, however, out of
the question. A method for hastening germination had to be
worked out. It was found that the seeds contain a water-soluble
germination inhibitor which is removed by treating the seeds
with sulfuric acid. Seeds thus pretreated may be dried without
impairing their germinability. This is important, as after the acid
treatment the seeds are wet and cannot be sown by mechanical
means.

In addition, it was found that the acid-treated seeds need for
optimal germination either a daily alternating temperature of 20°
and 30°C, independently of light, or a constant temperature of
20°C with the addition of light. This would have meant that,
under constant temperature conditions, the seeds should be sowed
very superficially with all the resulting hazards. However, it was
found that dark germination could be increased to nearly 100%
when the dark is interrupted by a short period of illumination at
any time after the first day. A method was then developed by
which the seeds were illuminated for five minutes after twenty-
four hours of imbibition in the dark, and then dried. Seeds thus
treated germinated well in darkness under all temperature condi-
tions, and our germination problem was solved.

Many more such examples could be cited all of which would
prove the one point already made, namely that no desert agricul-
ture is possible without collaboration of a germination laboratory
where the germination conditions of the seeds involved may be
investigated. The reasons are obvious. The seeds of most desert

396 THE FUTURE OF ARID LANDS

plants contain different mechanisms which regulate germination
by means of inhibition. Under natural conditions this results in
inhibited germination which either fractionates germination over
many years or limits it to coincide with the infrequent suitable
conditions for the species. These are modifications of survival
value for the species insuring it against extinction in a year in
which incipient good germination conditions are followed by a
period of drought. But in agricultural practice we must attain just
the opposite, i.e., very rapid, simultaneous germination at the time
of sowing. As the time of sowing has to be chosen very carefully—
and the question of when to sow is one of the main problems
confronting desert agriculture—one has to be absolutely sure that
germination is rapid and uniform. This can be achieved only when
the germination of a given seed has been carefully studied, its
germination mechanisms explored, and its requirements for ex-
ternal germination conditions are known.

The establishment of a germination laboratory in connection
with each local project for desert agricultural research is not
enough. There must be a central international organization where
all the information obtained about germination of desert seeds in
different regions of the world is collected and correlated. The
reasons for this suggestion are obvious. The local germination
laboratories must serve a practical need and, as pointed out
before, should not get lost in the purely scientific research prob-
lems involved. On the other hand, usually, only part of the work
done there on different seeds is published in scientific journals.
This information should also be fed into a central collecting
agency and stored there where it can be available to everybody.

The propagation of species suitable for desert agriculture is
concerned not only with reseeding but also with vegetative
propagation.

When the vegetative propagation of the important pasture
plants Hordeum bulbosum and Phalaris bulbosa by transplanting
of bulbs was tried, it was found that bulbs removed from the soil
at the end of the growing season lose their viability completely
after four to seven days. This makes their storage impossible. An
investigation was undertaken, therefore, to find out if there are

PROBLEMS AND RESULTS IN ISRAEL 397

developmental stages where the bulbs keep their viability during
longer periods of storage.

Another problem arose with Atriplex halimus, an excellent
pasture plant. After extensive research it was found that this
plant can be propagated by seeds after proper pretreatment, but
cuttings can be used as well. The problem here was to see if
there is any periodicity in the rooting response of cuttings. In this
field, too, the lesson learned was that only by physiological re-
search can the proper methods for vegetative propagation be
found. This can be done only through the cooperation of a lab-
oratory of plant physiology.

Selection of Plants

Here, naturally, the main problem is the finding of suitable
pasture and other economically important species for reseeding
purposes. The greatest part of our efforts was concentrated on
local species and arid ecotypes, which are able to withstand
prolonged periods of drought.

Selection of Pasture Plants (12, 13)

In the selection of pasture plants for desert agriculture consider-
ations other than those for moister climates have to be taken into
account:

(a) Grasses utilize soil moisture up to a depth of 120-150 cm.
In some of the main wadis in the Negev, moisture penetrates 300
to 400 cm. This deep lying moisture may be utilized if deep-
rooting pasture shrubs are found. Such a mixed grass-shrub
pasture would afford a more rational utilization of available soil
moisture.

(4) In the desert the growing seasons for most grasses coincide
with each other and last throughout late winter and spring. This
limited grazing season may be extended by reseeding with ever-
green shrubs like Atriplex halimus.

Consequently, great advantage may be gained from a mixed
pasture consisting of grasses (perennials and annuals), deep-

398 THE FUTURE OF ARID LANDS

rooted pasture-shrubs, and legumes. The most promising species
were found to be:

Local perennial grasses: Hordeum bulbosum, Oryzopsis miliacea,
Cynodon dactylon.

Perennial grasses introduced from neighboring regions in Is-
rael: Phalaris bulbosa, Oryzopsis holciformis, Agropyrum junceum.

Annual grasses: dvena sterilts.

Annual broad-leaved herbs: Medicago spp., Trifolium spp.
Malva spp.

The extremely drought-resistant Panicum turgidum deserves
special mention as it is at the same time an excellent binder of
shifting sands and a recognized pasture plant.

Of the shrubs suitable for pastures we mention Atriplex hali-
mus, an omni-Mediterranean meso-halophyte, well eaten by all
animals. It may also be used for human consumption as a boiled
vegetable or salad. The bush branches out profusely from its
base, and can be used well in soil conservation and for the pre-
vention of flash-flood damage. Large-scale reseeding with this
species has already been successfully attempted (especially for
reinforcement of dykes and terraces).

Calligonum comosum is a Saharo-Sindian species which, to-
gether with Haloxylon persicum, forms the climax vegetation of
sandy soils situated on the borders between steppe and desert
(g). It is apparently a good prospect as a pasture plant. Experi-
mental reseeding has already given promising results.

Colutea istria, a Saharo-Sindian leguminous species, is consid-
ered an excellent pasture shrub, and has already been used for
reseeding of depleted desert pastures.

From pasture plants introduced from abroad we mention:
Panicum antidotale and various Atriplex species like Atriplex
semibaccata, A. spongiosa, A. vesicaria. Our A. halimus seems,
however, to be more suitable under local conditions.

Oryzopsis miliacea and Phalaris bulbosa strains, which origi-
nated in the Mediterranean area, were improved upon in the
United States, and then were reintroduced to Israel. As far as
desert agriculture is concerned our own local ecotypes seem to
be superior.

PROBLEMS AND RESULTS IN ISRAEL 399

Selection of Industrial and Pharmaceutical Plants

Mention has already been made of Funcus maritimus as raw
material for our paper industry. Israel possesses a modern paper
mill which uses imported raw material. Funcus maritimus has
been investigated thoroughly from a technical point of view and
first class paper has been produced from it (2g). Its planting and
sowing in salt marshes of the Negev is still in the experimental
stage.

As regards pharmaceutical desert plants, we are carrying out
a systematic survey of our desert flora, searching for plants which
contain pharmaceutically important substances. At the same
time, plants known to contain drugs are being investigated as
regards the amounts of the active constituents and their seasonal
fluctuations (37, 38). With the exception of Hyoscyamus muticus,
large-scale culture of these plants has not yet been attempted.
Our main prospects are:

Hyoscyamus muticus for atropine.

Anabasis haussknechtii for anabasine.

Artemisia herba-alba and Artemisia monosperma possess anthel-
mintic properties. From the essential oils of 4. monosperma a
new substance, furoartemone, was isolated, a furan compound
possibly of great importance.

Peganum harmala tor harmine and harmaline.

Periploca aphylla, Daemia tomentosa, Leptadenta pyrotechnica
for various glucosides.

Gypsophila rokejeka for sapogenin.

Capparis spinosa for heart alkaloids.

Achillea fragrantissima and A. santolina for essential oils.

Agave sisalana, introduced as an industrial fiber plant, con-
tains in the unused parts of its leaves hecogenin, which can be
used as a starting material for the synthesis of steroidal hor-
mones.

Special mention has to be made of Citrullus colocynthis. This
belongs to the Saharo-Sindian area, but penetrates into the Medi-
terranean territory and grows on sandy soils. Its fruit contains
the well-known and long-used colocynthin, and its seeds contain
20 per cent oil. This interesting drought-resistant plant could be

400 THE FUTURE OF ARID LANDS

used for the breeding of either perennial or drought-resistant
watermelons, or both.

Selection of Woody Desert Plants for Various Purposes. (25)

Woody species find their uses in deserts: (a2) for shade, which
is so essential to the well-being of livestock in the scorching noon-
day desert sun; (4) for permanent dune fixation by plants which
resist oversanding and blowing out; (c) for shade and beauty
wherever human beings settle.

A good local candidate, but not yet used on a practical scale,
is Haloxylon persicum which is to be found as remnants of former
larger ‘“‘desert-forests” in Nahal Arava. It is a good “fodder-tree”’
and represents the dominant plant of a climax association.
“Large-scale artificial reestablishment in its original area of dis-
tribution is highly recommended” (Q).

The local Tamarix species represented a special problem. This
is one instance where considerable financial loss was experienced
when trying to proceed empirically without proper scientific
preparation. In the earliest stages of this work Tamarix was
planted indiscriminately on a large scale. It was found that the
species used was ecologically the wrong one, and the whole plan-
tation failed in one year’s time. Nine species with widely diverse
ecological requirements are to be found in Israel, but their tax-
onomy is most difficult, as everybody knows who ever tried to
identify the different closely related species. The thorough taxo-
nomic survey, which is the first precondition for their proper use
in desert afforestation, is now being undertaken.

Local Retama and Pistacia species are at present also being
studied taxonomically as only after determining the subspecies
and varieties which differ much in their respective ecological am-
plitude can they be profitably propagated.

Of the introduced species, Eucalyptus rostrata may be impor-
tant. Fourteen years ago a big plantation was made in the Negev
on a locality with 1oo mm yearly rainfall. Only few specimens
survived, apparently representing a drought-resistant strain.
They are now being propagated.

PROBLEMS AND RESULTS IN ISRAEL 401

Selection of Plants for Sand Dune Fixation (12)

Great areas of the Negev lands are composed of shifting and
semi-mobile sand dunes. In order to fix those dunes and at the
same time derive some economic profit from the plants which
are being used for the fixation, the ecology and economic value
of all the plant communities growing naturally on sand dunes of
our area were studied.

Testing their resistance against blowing out, oversanding, and
salinity, the following plants were found to be most useful: Fun-
cus maritimus, fF. acutus, Scirpus holoschoenus, and Eragrostis b1-
pinnata, which can all be used in paper manufacture. Artemisia
monosperma as a pharmaceutical plant, dgropyrum junceum,
Convolvulus secundus, C. lanatus, and Calligonum comosum as
fodder plants. Here, too, the absolute necessity for close coopera-
tion with a laboratory for plant physiology was experienced, es-
pecially in questions of germination and propagation, and with a
technological laboratory for the testing of the respective plants
for their usefulness as raw material for paper manufacture or for
pharmaceutical value.

In summing up the results obtained regarding the selection of
suitable plants for desert agriculture, it is important to point out
that we relied nearly exclusively on the selection of pre-adapted
ecotypes either of the region to be developed or of neighboring
phytogeographical territories of Israel. The approach to the prob-
lem was—with success—an empirical one. Here much has to be
done in the future—and is now being done—based on very long-
range scientific planning which has to be not only regional but
also international. The various ecotypes have to be tested for
their different qualities under constant controlled conditions.
Genetic strains of these ecotypes have to be developed and hy-
bridization experiments to be carried out.

All this necessitates, again, a well organized international clear-
ing house for systematic exchange of information and its correla-
tion. In addition, it is obvious that a small country like Israel
cannot afford installations which are expensive to build and ex-
pensive to maintain—where genetic strains developed from local

402 THE FUTURE OF ARID LANDS

ecotypes can be tested under controlled conditions like those
available in the Earhart greenhouses in Pasadena. The old meth-
ods of testing under natural conditions are much too laborious
and slow for the rapid tempo needed in a progressive desert agri-
culture.

Use of Natural Precipitation (6, 30, 31, 40)

As already mentioned, the only source of irrigation for our
desert agriculture is the natural precipitation, and the runoff
water derived thereof. We learned how to utilize this, mainly by
studying the remnants of the ancient agricultural systems of the
Nabataeans and Byzantines who once populated the central and
southern Negev. There are today six dead cities (1) which were
once thought to have thrived on caravan trade only. After an
archaeological survey we know that they practiced a highly devel-
oped desert agriculture and possessed a very elaborate system of
utilizing practically every drop of precipitation for agricultural
purposes.* After the destruction of these desert cities in the sev-
enth century A.D. (they were founded in the second century B.c.)
the intricate system of perfectly and laboriously constructed
dams, spillways, and terraces slowly crumbled, and for many
centuries they were maintained very amateurishly by desert no-
mads. Today we partly use the same reconstructed dams and
terraces, the same rock-hewn cisterns which the Nabataeans had
built over 2,000 years ago. It is thrilling to see time and time
again how the present day dispositions of highly complex irriga-
tion systems, calculated by trained specialists, with the latest
technical aids, coincide with remnants of ancient irrigation sys-
tems on the same spot.

There is today a consensus that in historical times there was
no fundamental change of climate in our region (31) and that the
Nabataean and Byzantine agriculture in the Negev faced essen-
tially the same problems we are facing today.

* Similar agricultural systems are known from ancient civilizations in
the desert areas of southern Arabia and North Africa. A comparative
study of such systems all over the world might provide valuable practical
and historical information.

PROBLEMS AND RESULTS IN ISRAEL 403

The first problem in the utilization of rain water in the desert
is how to deal with desert floods and transform them from a
destructive force into a creative agent. Even though the rainfall
in our deserts is very scant, the limited percolation of water in our
desert soils, and the resulting excessive runoff, cause sudden
tremendous flash floods in the wadis. After a rainfall of as few as
10 mm on the watershed, for instance, it is no rarity to see a flood
of 30,000 m*/hr crashing through a wadi draining a 50-sq km
catchment basin for four to five hours. These great quantities of
water are practically the only source of irrigation water. But
unless they are harnessed, they are lost, and all they cause is exten-
sive soil erosion. Various types of dams, dykes, terraces, and spill-
ways were, and still are, the answer to this problem. Topographical
considerations are the decisive factor in the choice of the system
to be employed.

After ten years of observation we have reached the conclusion
that this type of flood causes a rainfall of a few millimeters to in-
crease soil moisture in the wadis to the equivalent of several
hundred millimeters of precipitation. This transforms the wadis
and the catchment basins, however small, into productive soil,
especially when the natural moistening by floods is augmented by
various flood control measures, as described later.

The second problem is an edaphic one. Real soil in our desert
area is rare. Most of our terrain cannot be used at all for lack of
soil (21). Only where loess, sandy loess, or sand is to be found, or
can be induced to deposit by flood control, is agriculture possible.
This again means that desert agriculture 1s possible in strips and
patches found only in the above-mentioned wadis and catchment
basins, or in sandy plains.

The utilization of water and the agricultural technique are very
different in tributary wadis, in the major drainage channels, and
in flood plains. Accordingly, we shall treat each one separately.

Tributary Wadis

The first consideration with this type of wadi is that, owing to
its narrowness and steepness, a tributary wadi has to be controlled
as a unit from the watershed down. Only thus may the destructive

404 THE FUTURE OF ARID LANDS

Figure 1. Flood water irrigation. I. Tributary wadis (Schematic).
Il. Tributary wadis (Plane view). III. Broad flat wadis (Plane view).
(Direction of flow is from right to left.)

SS—Stone shelves on contour lines; BR—Bush reinforcement; CT—
Cultivated terrace in water course; PA—Ponded area in water course;
ED—Earth dyke on contour lines; SP—Stone spillway with reinforce-
ment of dyke tip.

force of the flood be harnessed so that the work done in the lower
part of the wadi may not be totally demolished. The total length
of the wadi is terraced by a series of broad masonry shelves into
level plots, each lower than the preceding one—like the stairs of a
staircase. These shelves are quite low (25-50 cm). The wadi is
thus completely transformed into a series of terraces, narrow (a
few meters wide) in the upper parts of the wadi and broader in
its lower parts (Figure 2).

The velocity of the water is thus considerably reduced, the

PROBLEMS AND RESULTS IN ISRAEL 405

percentage of percolating water is increased, and greater depth of
water penetration is achieved (Figure 3).

Soil particles brought by the run off water from the slopes are
deposited on the terraces, the soil quality and topography of which
are thus slowly improved, and the terraces further leveled.

The stone dams can be reinforced by the planting of bushy
plants like edible Atriplex halimus, on inedible Thymelaea hirsuta,
etc., in front of each dam. A “‘bush-dam” alone may be sufficient
and at times even better than a stone dam. The considerations are:
(a) The bush-dam is pliable and may be better able to withstand
the force of the flood than solid stone or earth dams. (4) The dam
would grow in size concurrently with the silting processes. (c) The
bush-dam filters the flood water, thus enriching the soil with or-
ganic matter. (7) The costs of construction are lower. (e) Conse-
quently, more dams may be planted per unit area. (f) The plants
themselves may provide grazing, but as this may prove detri-
mental for the dam, inedible species may have to be used to
replace the edible ones.

Main Wadis

Floods in main wadis are very strong and cannot be completely
stopped by any sort of dam. They can, however, be slowed down.
Thus floods cannot be prevented but have to be brought under
control in some other way.

According to the quality of the beds of the main wadis, they are
agriculturally treated in different ways.

(a) Broad, flat beds are covered by a deep layer of loess. An
earthen dyke is built across the wadi on the contour. The dyke
does not, however, reach across the wadi, but terminates a few
meters before reaching the opposite bank, and is terminated by a
broad spillway. The next dyke downstream is built just the other
way around with its spillway on the opposite bank. The threshold
of the spillway is constructed several decimeters higher than the
wadi bed. The water is thus forced to follow a much longer zigzag
course down the wadi, instead of running straight down it, and a
shallow pond forms behind each dyke. This artificially creates a
much gentler slope and slows down the velocity of the water con-

406 THE FUTURE OF ARID LANDS

Figure 2. Flood water irrigation in tributary wadis in the Negev
Highlands. Two tributaries, completely terraced up to the watershed,
stone shelves reinforced with inedible bushes (Thymelaea hirsuta,
Haloxylon articulatum, etc.).

siderably. Danger of erosion is thus reduced, and at the same time,
the water has more time to percolate and moisten deeper layers of
the ground. The design of these structures is planned on topo-
graphic maps (Figure 1, III).

(6) Broad loess wadis are longitudinally dissected by a steep
stony watercourse. In this case the shape of the watercourse pre-
vents any flood control. However, it may be, and has been in an-
cient times, utilized by diversion and spreading on the loess banks,
and on higher lying fields on the slopes. That the Nabataeans and
Byzantines were masters in this type of agriculture is proved by
the very extensive system of terraces left by them and by the
excellent, solid quality of the stone walls built round the terraced
fields. These fields, which accompany the wadi bed on both sides
along its course, are leveled and their edges are protected by low
stone walls. The main terrace wall runs perpendicular to the course

PROBLEMS AND RESULTS IN ISRAEL 407

Figure 3. Dense plant cover obtained by efficient water spreading.
Terrace thickly covered by Avena sterilis; slopes carry Zygophylletum
dumosit. Uncovered patches on terraces received insufficient water

(drought year!).

of the wadi. At the side of the field adjoining the wadi bed a stone
wall protects the plot from erosion by the flood water. The terraces
thus formed are very broad, completely level, and one only slightly
lower than the next. Onto these terraces water is diverted and
spread by channels running obliquely from the watercourse.
However, the force of the water often demolishes these channels.
The ancient civilizations overcame this difficulty by constructing
a solid low stone dam, a few meters thick, at a point below the
opening of the channel. The height of this dam controlled the
height of the flood water at the mouth of the channel, all excess
water flowing over thedam. The force and amountof water entering
the channel were controlled by the shape and size of its opening.
We have developed a type of detention dam at a point above
the channel calculated to catch most of the flow. The dam is
pierced by a large diameter pipe, which permits the flow of a

408 THE FUTURE OF ARID LANDS

limited volume of water (about 10,000 m*/hr). This serves a
double purpose. First, this controls the maximal pressure of water
reaching the channel, and secondly, this constantly reduces the
water pressure on the detention dam, and thus helps to preserve it.
The earth dam is built sufficiently strong to detain the total
flow until it is completely drained off by the outlet pipe at a con-
trolled rate.

Flood Plains

In many places in the Negev small or large level plots of land
can be found surrounded on all sides by gently inclined slopes.
After a rainfall, the runoff water of these slopes is collected on the
flood plain, where it slowly sinks into deeper soil layers. Small,
stone-reinforced channels, built obliquely on the surrounding
slopes, greatly increase the amount of catchment water.

Before concluding, we should like to emphasize that all our
experience so far bears out the well-known truth that there is no
soil conservation or flood control without permanent plant cover
on the water-spreaded lands.

Agrotechnical Problems (6)

There are a great number of agrotechnical problems involved in
desert agriculture. One of the main agrotechnical difficulties en-
countered in desert agriculture is the rapid formation of a hard
crust on the upper surface of the loess soil, whenever the soil has
been wetted and starts to dry out. This crust greatly obstructs
reseeding by preventing seedling emergence, except from cracks
formed in the drying out crust, and by tearing of roots when the
crust separates from the subsoil. This problem is being attacked by
trying to prevent crust formation by adding soil conditioners
(Kryllium and other artificial soil conditioners specially developed
in Israel (26) and natural soil conditioners like straw and sand) to
the soils. Another type of attack is made via germination. The
study of the germination mechanisms of each seed often leads to
methods of greatly accelerating germination, so that uniform
seedling emergence is obtained before crust formation starts.
This is another example of the need for close cooperation between
various research agencies, as under these extreme conditions it is

PROBLEMS AND RESULTS IN ISRAEL 409

sometimes of vital importance to hasten germination and seedling
growth even by a few hours.

The time of sowing is another problem. It is technically easier to
sow before the first rain or flood, but there are reasons for not
doing so: (1) crust formation; (2) the danger of the seeds being
washed away with the top-soil by the first flood; (3) danger of
germination after a small rainfall sufficient to bring about germina-
tion but insufficient to maintain further growth; (4) on account of
the rather erratic occurrence of effective rainfall, the impossi-
bility of deciding on the species to plant until the rain has actually
fallen.

All these considerations lead to the conclusion that it is practica-
ble to sow only after the soil has been wetted by flood. This raises
the basic agrotechnical problem of methods of sowing in wet clay.
This has to be done rapidly, and not be delayed even for one day
as the drying of the topsoil is very rapid.

Utilization of Available Saline Water

Most of the available ground water is highly saline. There are,
however, a number of plants which thrive on saline water. To
cite only a few of the many species on which such work is being
done: (1) Funcus maritimus as mentioned above; (2) Phoenix
dactylifera, which is planted extensively at Revivim and along the
salt marshes in Nahal Arava, between Yotvata and Elath on the
Red Sea shore; (3) vegetables, such as beets and spinach; (4) salt-
resistant strains of forage crops, such as alfalfa.

Dew

A yearly water balance of certain desert plants shows that they
do not rely only on the water taken up from the soil (20, 22, 23).
On the other hand, it seems today a well-established fact that dew
is taken up by leaves, and may even be transported from the
leaves to the roots where it can be secreted into the soil (16, 35).
At the same time it seems to be certain that plants growing under
arid conditions benefit decisively by the uptake of dew (16). Are
there any practical conclusions to be drawn for desert agriculture
from those facts?

Based on ten years’ measurements of dew, made with Duvde-

410 THE FUTURE OF ARID LANDS

vani’s dew gage (14), we possess today figures about the average
number of dewy nights per month, and about the amounts of dew
formed in different topographical localities and at different heights
above the soil surface (3, 12, 24). It will perhaps be possible in the
future to choose special localities, rich in dew, for certain crops
profiting especially from dew, and to create artificially better
microclimatic conditions for the formation of dew on plants and
in soils.

In addition, the application of limited quantities of additional
moisture in the form of artificial dew might prove useful in con-
servation of irrigation water resources, especially when performed
at night, when plant growth is at a maximum and loss of water by
evaporation is at a minimum.

Conclusions and Outlook for the Future

Desert agriculture is definitely one of the many different ways
of increasing populations in arid areas. It can be successful only
under three conditions:

1. Prior to any practical work, a broad, thorough scientific
survey of the area has to be carried out.

2. The surveying and practical work has to be done by a closely
knit cooperating team of scientists of all branches of the natural
sciences. This teamwork has to be very elastic as constantly new
problems arise, necessitating the cooperation of new agencies.

3. The practical work in its pilot plant stage has to center
around experimental settlements situated inside that desert area
where desert agriculture is being tried out.

It is felt that the establishment of an international center for
the storage and exchange of information on desert research 1s
urgently needed. At the same time this center could refer major
research projects, which cannot be carried out by small nations
for lack of the needed equipment and scientific manpower, to the
few specialized, well-equipped, experimental research institu-
tions. It is much too early to discuss the purely economic side of
desert agriculture, as we do not yet possess the necessary data.
But when taking Israel as an example, it can be said that by using
the methods described here, 3 to 5% of our desert area, which up

PROBLEMS AND RESULTS IN ISRAEL 4ll

now produced nothing, can be converted into productive

agricultural land.

REFERENCES

. Amiran, R., 1954, The cities of the Negev. Israel Explor. F., 4, 50.
. Amiran D. H. K., 1954, The geography of the Negev and the south-

ern limit of settlement in Israel, Zsrae/ Explor. F., 4, 51.

. Anonymous, 1947/48, 1949, 1950, 1951, 1952, Annual dew summaries,

Israel Meteorological Service.

. Bloch, M. R., Kaplan, D. and Schnerb, J., 1954, Funcus maritimus,

a raw material for cellulose, Bud/. Research Council Israel, 4, 192.

. Bonne, A. 1953, Land resources and the growth of world population,

Proc. Symp. Desert Research, 464, Jerusalem.

. Bogoslav, D., 1954, Sde Boker—A summary of the agricultural

work done during the period 1953/54, Hassadeh, 35, 87, (Hebrew).

. Boyko, H., 1945, On forest types of the semi-arid areas at lower

altitudes, Palest. F. Bot. Rehovot, 5, 1.

. —, 1947, A laurel forest in Palestine, Palest. F. Bot. Rehovot, 6, 1.

g. —, 1949, On the climax vegetation of the Negev. Palest. F. Bot.

Rehovot, 7, 17.

. —, 1949, On climatic extremes as decisive factors for plant distribu-

tion, Palest. F. Bot. Rehovot, 7, 41.

. —, 1951, Les bases écologiques de la régénération de la végétation

des zones arides, Un. Int. Sci. Biol. Ser. B, 9.

. —, 1954, Report to the Ford Foundation on pasture research and

ecology of dune plants, Manuscript.

. and Tadmor, N., 1954, An arid ecotype of Dactylis glomerata L.

in the Negev. Bul/. Research Council Israel, 4, 241.

. Duvdevani, S., 1947. An optical method of dew estimation, Quart.

7. R. met. Soc., 13, 282.

. —, 1953, Dew gradients in relation to climate, soil and topography,

Proc. Symp. Desert Research, 136, Jerusalem.

. —. Effects of dew on plant growth under arid conditions, 4m.

7. Botany (In press).

. Eig, A., 1931-32, les éléments et les groupes phytogéographiques

auxiliaires dans la flore palestinienne, Fedde’s Rept. spec. nov. Beth.,
63, I.

. —, 1938, On the phytogeographical subdivision of Palestine, Pa/est.

F. Bot. Ferusalem, 1, 4.

. —, 1946, Synopsis of the phytosociological units of Palestine,

Palest. F. Bot. Ferusalem, 3, 183.

. Evenari, M. and Richter, R., 1938, Physiological-ecological investi-

gations in the Wilderness of Judaea, 7. Linn. Soc., 51, 333.

412

Dale

THE FUTURE OF ARID LANDS

—,and Orshansky, G., 1948, The Middle Eastern hammadas,
Lloydia, 11, 1.

—, 1949, Ecologia de las plantas del desierto, Rev. argentina agron.,
G5 1D,

. —, 1953, The water balance of plants in desert conditions, Proc.

Symp. Desert Research, 266, Jerusalem.

. Gilead M. and Rosenan, N., 1954, Ten years of dew observation in

Israel, Israel Explor. F., 4, 120.

. Goor, A. Y. and Zohary, M., 1954, Report to the Ford Foundation on

research in afforestation and forest tree ecology in the Negev, Manu-
script.

. Katchalsky, A., 1954, Report to the Ford Foundation on the study of

the interaction of polyelectrolytes with soil, Manuscript.

. Koller, D., 1954, Germination regulating mechanisms in desert seeds,

Ph.D. Thesis, The Hebrew University of Jerusalem, (Hebrew).

.—, 1955, Report to the Ford Foundation on germination research,

Manuscript.

7 Eevina VES tTesan Desert alanis | in Israel as potential sources of cellu-

lose, Proc. Symp. Desert Research, 346, Jerusalem.

: Lomdesamillle We Ge, 1O53, Mwaals in deserts, Proc. Symp. Desert

Research, 365, Jerusalem.

. —, 1954, The use of flood waters by the Nabataeans and Byzan-

tines, Israel Explor. F., 4, 50.

. Oppenheimer, H. R., 1950, Geobotanical research in Palestine 1938—

1950, Vegetatio, 3, 301.

. Ravikovitch, S., 1953, The aeolian soils of the Northern Negev,

— Proc. Symp. Desert Research, 404, Jerusalem.

39:

40.

. —, 1954, Report to the Ford Foundation on the investigation of the

soils of the Southern Negev, Manuscript.

. Stone, E. G. and Schachori, A. Y., 1954, Absorption of artificial dew,

CalipeAgric., o)-

. Weitz, J., 1954, The settlement of the Northern Negev, Jsrae/

Explor. F., 4, 59.

. Weizmann, A., 1954, Report to the Ford Foundation on the work done

by the natural products research group, Manuscript.

. Zaitschek, D. V., 1953, Some useful plants of therapeutical value

from desert regions in Israel, Proc. Symp. Desert Research, 350, Je-
rusalem.

Zohary, D., 1953, Ecological studies in the vegetation of the Near
Eastern deserts III. Vegetation map of the Central and Southern
Negev, Palest. F. Bot. Ferusalem, 6, 27.

—, 1954, Notes on ancient agriculture in the Central Negev, srae/

Explor aieyes lie

41.
42.

43;

44.

45.

46.

47.

48.

PROBLEMS AND RESULTS IN ISRAEL 413

Zohary, M., 1945, Outline of the vegetation in Wadi Arabia, 7.
Ecol., 32, 204.

—, 1947, A geobotanical soil map of Western Palestine, Pa/est. 7.
Bot. Ferusalem, 4, 24.

—, and Orshansky, G., 1949, Structure and ecology of the vegeta-
tion in the Dead Sea region of Palestine, Palest. F. Bot. Ferusalem,
4,177.

— and Feinbrun, N., 1951, Outline of vegetation of the Northern
Negev, Palest. 7. Bot. Ferusalem, 5, 96.

—, 1952, Ecological studies in the vegetation of the Near Eastern
deserts I. Environment and vegetation classes, [srae/ Explor. F. 2,
201.

— and Orshan, G., 1954, Ecological studies in the vegetation of the
Near Eastern deserts V. The Zygophylletum dumosi and its hydro-
ecology in the Negev of Israel, Vegetation, 5/6, 340.

—, 1954, Report to the Ford Foundation on geobotanical and phyto-
sociological survey of the vegetation in the Negev, Manuscript.

—, 1955, Geobotany, Sifriath Hapoalim, Merhavia, (Hebrew).

Problems in the Development and
Utilization of Arid Land Plants

PM RUETZ

Agricultural Research Service, United States De-
partment of Agriculture, Beltsville, Maryland

Native plants of the desert region of the southwestern United
States and northern Mexico, in addition to contributing much to
the health and well-being of the inhabitants, have long held the
attention of chemists and biologists as potential sources of indus-
trial raw materials. Although attention has been directed to
many of the species as containing exploitable quantities of essen-
tial oils, medicinal alkaloids, gums, fibers, rubber, tannin and
other products (4, 5), very few of them have been studied thor-
oughly to determine their chemical constituents. An even smaller
number, represented by guayule (Parthenium argentatum) for
rubber (7), and canaigre (Rumex hymenosepalus) for tannin (6),
have been investigated by chemists, plant breeders, and engi-
neers attempting to improve the plants and the methods of
processing to the point where they can be utilized in commerce,
either as wild plants or cultivated crops.

Conditions of Commercial Use

Unless the plant product involved is scarce or has unique
properties demanded by a particular industry, it must be pro-
duced on a competitive cost basis with similar products from
other sources. Furthermore, it must be available in sufficient
quantity to furnish a fairly continuous supply to commercial
users. Candelilla wax, from species of Euphorbia and Pedilanthus,
is an example of a desert plant product which has been in com-

414

ARID LAND PLANTS 415

mercial use for many years. The continuous utilization of can-
delilla wax is based in part on distinctive properties not present
in other materials and in part on its price in relation to other
similar waxes, such as carnauba (2). The introduction into com-
merce of any new product from arid land plants will be difficult
unless such information is known.

Natural stands of some potentially valuable desert plants, such
as canaigre and jojoba (Simmondsia chinensis) (3), are not exten-
sive enough to furnish material for commercial utilization. Other
species are sufhciently abundant to provide an adequate and
perpetuating supply only if they are managed on a sustained
yield basis. For example, excessive harvesting of candelilla plants
in Mexico without proper provision for regeneration of the plants
after cutting, has resulted in depletion of natural stands which
appeared to be inexhaustible (2).

Means of Improvement

Cultivation of arid land plants, either for the purpose of sup-
plementing supplies from natural sources or for furnishing the
entire supply, offers promise in many cases. Certain species may
be adapted to low-cost operations, carried on in conjunction with
water conservation practices where sufficient water is not avail-
able for crops with high water requirements. The work of various
public agencies in reseeding deteriorated areas of range land sug-
gests methods that may be used. Species that can be readily
adapted to cultivation, such as canaigre, guayule, and plantago
(1), have definite promise as crops in irrigated areas or in areas
with marginal supplies of irrigation water. The ability of arid
land plants to survive under conditions of extreme water stress
makes it possible to grow them under a much wider range of
conditions than can be done with other crop plants.

Any program for developing desert plants for commercial utili-
zation should also include thorough investigation of the possibil-
ity of developing superior strains through selection and breeding.
Substantial improvement in the yield of a desired constituent
can often be made simply by selecting desirable plants from nat-
ural populations. Investigations carried on with guayule (7) and

416 THE FUTURE OF ARID LANDS

plantago (1) have indicated great promise of improvement by
means of interspecific hybridization.

Canaigre: An Example

Recent experimental work with canaigre (Rumex hymenosepa-
lus) by the Special Crops Project, Agricultural Research Service,
Department of Agriculture, may be used to illustrate briefly some
of the points mentioned.

Canaigre, a distinctive species of Rumex, with tuberous, tan-
nin-bearing roots, grows in scattered stands from west Texas to
California and from Colorado and Utah to northern Mexico. In
many locations the plants are not sufficiently concentrated to
furnish roots for commercial exploitation. Furthermore, roots
from plants in the extensive and accessible stands in eastern
Arizona, New Mexico, and Texas are relatively low in tannin
and difficult to extract. Any industrial use of canaigre for tannin
will, therefore, necessarily depend upon roots from cultivated
plants to maintain an adequate supply.

Distinct variations in root, leaf, and seedstalk characteristics
have been found in strains collected in various locations. Roots
from different locations varied in tannin content from less than
20 to more than 40% (dry weight basis). Similar wide variations
between strains were found in non-tannin extractives which affect
processing quality. When all the collections were grown under
the same environmental conditions in southern Arizona, many
distinctive characteristics were found to be heritable.

The range of variation in the wild material is so great that
practically every characteristic needed in a cultivated crop ap-
pears to be available. Some selected lines consistently produce
high yields of roots containing more than 35% of readily extract-
able tannin. Other lines are characterized by desirable root shapes
for mechanical harvesting, abundant seed production, and dis-
ease resistance. It appears that through the use of appropriate
breeding procedures, varieties can be produced that are distinctly
superior to the wild types and productive enough for use as a
cultivated crop. The use of F; hybrids for commercial production
also appears to be a promising possibility.

ARID LAND PLANTS 417

The development of effective methods of production also has
an important bearing on the success of any program for convert-
ing wild plants into cultivated crops. Early experiments with
canaigre indicated that two full growing seasons were necessary
to produce economic yields of roots with satisfactory quality. In
recent experiments, 1§ tons of roots per acre, equivalent to ap-
proximately 1,800 pounds of 100% tannin, has been produced in
one 10-month growing season. Other results indicate that even
higher yields may be obtained from improved strains, using
either seed or crowns for planting stock.

Soil moisture is the major factor which regulates the volume
of roots produced by wild canaigre plants. The storage roots sur-
vive long periods of extremely low soil moisture, increasing in
size and number only during favorable seasons. As a cultivated
crop in southern Arizona, canaigre requires a limited amount of
irrigation to produce satisfactory yields. Adequate soil moisture
must be available during October and November to establish the
plants, and during March and April to promote root growth and
tannin storage. Further study of the seasonal development of
canaigre in relation to soil moisture may be expected to lead to
more effective methods of using limited supplies of irrigation
water.

The processing of canaigre roots into high quality tanning ex-
tracts presented several major problems because of the presence
of sugar and starch in the roots. Research on processing, con-
ducted concurrently with the breeding and production investiga-
tions, has resulted in the development of methods for producing
tanning extracts which compare favorably with commerical ex-
tracts in tannin content and tanning properties (6). Tonnage lots
of canaigre extract will be available in the near future for use in
semi-commerical tanning tests.

The progress that has been made in developing canaigre for a
source of vegetable tannin for industry illustrates one method of
attacking the problem of utilizing arid land plants. Such progress
can be achieved only through expensive research on the part of
chemists, plant breeders, agronomists, and engineers. Ideally, it
should be a cooperative enterprise involving appropriate public

418 THE FUTURE OF ARID LANDS

agencies and industries interested in the product to be obtained.
Without such a concentrated effort in which scientists of several
disciplines cooperate, many potentially useful arid land plants (1,
3, 6) will not become crops or even be extensively utilized.

REFERENCES

1. Chandler, Clyde, 1954. Improvement of plantago for mucilage pro-
duction and growth in the United States. Contribs. Boyce Thompson
Inst. 11(8), 495-505.

2. Daugherty, P. M., ef a/. 1953. Industrial Raw Materials of Plant
Origin: III. Survey of candelilla and candelilla wax. Bull. Eng. Expt.
Sta. Georgia Inst. Technol. 16.

3. Daugherty, P. M., et a/. 1953. Industrial Raw Materials of Plant
Origin: IV. A Survey of Simmondsia chinensis (Jojoba). Bull. Eng.
Expt. Sta. Georgia Inst. Technol. 17.

4. Duisberg, Peter C. 1953. Chemical components of useful or potentially
useful desert plants of North America and the industries derived from
them. Desert Research, Proc. of the International Symposium held in
Jerusalem, May 7-14, 1952.

5. Krochmal, A., et a/. 1954. Useful native plants in the American south-
western deserts. Econ. Botany 1 (8), 3-20.

6. Rogers, J. S., and L. M. Pultz. 1952. Canaigre: A possible domestic
source of vegetable tannin. Shoe and Leather Reptr. Dec. 27, 1952.

7. Tysdal, H. M., and R. D. Rands. 1953. Breeding for disease resistance
and higher rubber yield in Hevea, Guayule, and Kok-Saghyz. Agron.

¥. 45 (6), 234-242.

Plants, Animals, and Humans

in Arid Areas

ENRIQUE BELTRAN

Instituto Mexicano de Recursos Naturales Re-
novables, Mexico

Gambling on the climate may be possible
in semiarid regions, but the dweller in an
arid region has to play safe or perish—

R. W. Bailey. Yearbook of Agriculture,
1941.

In discussing the possibility of improving conditions for pro-
duction and the standard of living in arid zones, major emphasis
is usually placed on bettering and propagating domestic plants
and animals.

This is, without doubt, especially important, but I believe that
within the whole it 1s necessary to fit the arid zone-wild plants and
animals-human population-domestic plants and animals complex.
Taking into consideration the four components of this complex,
one should try to discover the elements which may serve to balance
them from a desirable economic and social viewpoint.

Arid Zone

This is the basic element, the stage on which the drama is set,
and its principal characters will be largely determined by the
degree of aridity surrounding them. This element does not notice-
ably change throughout history.*

* T do not forget the profound change which may take place in an arid
zone if it is supplied with water; but, in that case, it takes on new char-
acteristics and, therefore, it is no longer necessary to consider it an
mAtIGeZonen:

419

420 THE FUTURE OF ARID LANDS

Wild Plants and Animals

In the above-mentioned setting, the flora and fauna which
naturally live there have become adjusted to the hostile environ-
ment and get along reasonably well. The variety of species and
the number of individuals which can exist in arid zones is remark-

able.

Human Population

In a strictly biological sense, the human race is only one more
animal species and, to a certain extent, it thrives in arid zones.
But, given man’s capacity to act upon his environment, it is
necessary to reserve a separate place for him and to consider him
as a new element which will fundamentally modify the ones al-
ready mentioned.

Domestic Plants and Animals

Man partly maintains himself from the wild plants and animals
which exist in arid zones. However, as soon as he establishes
himself, he introduces domestic types which constitute a new
element, altering the zone’s natural dynamic equilibrium, that is,
the equilibrium existing when the impact of human action is not
felt.

Balance of the Four Components

Undoubtedly, there are many possible ways to improve the
domestic plants and animals in arid zones by applying scientific
methods of selection, breeding, and cultivation. Nonetheless, the
attainment of these possibilities is generally accompanied by an
increase in the local population, making new demands on a natu-
rally feeble environment and further altering the native flora and
fauna.

It must be considered that the prevailing conditions in arid
zones are really borderline for organic life, and even though it is
possible to improve them somewhat by human action, it is still
easier, by the same token, to increase its unfavorable conditions
until they are not adequate for human existence.

Therefore, in the question “‘What are the possibilities of main-

PLANTS, ANIMALS, AND HUMANS 42]

taining larger human populations in arid zones?” there is implied
a great potential danger, since an immoderate increase in the
population of such zones may be encouraged, with all the risks
that this may involve.

Conditions exist in Mexico which make it necessary to consider
this point very carefully. It has been estimated that between
50% and 80% of the country’s national territory are arid and
semi-arid zones, depending on the criterion used, and it has been
roughly calculated that, in spite of adverse conditions, those
zones support about fifteen million inhabitants, representing
more than half of the Republic’s total.

This population has traditionally been sustained by the desert
it lives in, either availing itself of the native animal and vegetable
elements or supplementing their possibilities by means of uncer-
tain farm crops and an extremely limited livestock. The meager-
ness of the products obtained gives rise to extremely poor living
conditions for the inhabitants, above all in zones of greater water
scarcity. And what happens in Mexico 1s repeated in many other
regions of the world.

With the aid of modern science and technology, it will be pos-
sible to improve the living standards of such communities, espe-
cially if the effort is directed toward balancing the utilization of
native resources with what may be obtained from crop plants
and domestic animals, improved as much as possible.

But it would be dangerous to apply to those zones a demo-
sraphic policy of maximum increase in local population or of
transportation from other areas in an attempt to benefit them
or to solve resettlement problems.

It is evident that many arid zones could materially increase
their yields in order to achieve a substantial improvement in the
living standards of the groups living there at present. But if the
population density goes beyond a certain maximum, which can-
not be very high, not only will it have a greater impact on the
feeble environment which supports it, but the existing living
standards will surely fall.

Since the time available does not permit a fuller or more de-
tailed discussion, the basic aim of this brief explanation is to

422 THE FUTURE OF ARID LANDS

emphasize the following points, which I feel should be clearly
established.

1. Arid zones offer intrinsic conditions which necessitate the
greatest caution in considering any project affecting them.

2. The rise in local living standards should not depend exclu-
sively on the improvement of domestic animals and plants nor
on the introduction of new ones; it should also be based, as far
as possible, on the rational ecological utilization of the elements
native to the region.

3. Any tendency to a substantial increase in the population
should be carefully examined before encouraging it, for it could
have disastrous results. As Bailey states so well in the words
which preface these comments, the margin of tolerance in arid
regions is so small that any error may have fatal consequences.

By always bearing in mind the existence of the arid zone-wild
plants and animals-human population-domestic plants and animals
complex, one may get a general view of the problems relative to
the improvement of those regions. In order to ensure permanent
yields, it is essential to base any action onsound ecological grounds,
with a clear conservationist orientation.

Perhaps the preceding observations may appear pessimistic,
although in reality they are not. Basically, they tend to guarantee
a high living standard to the desert population groups by yre-
venting the feeble environment which supports them from further
deterioration as the result of an intensive misguided exploitation
producing only temporarily favorable yields.

Nor do we state categorically that an increase in the human
population of those zones is impossible. We only point out the
obvious potential danger, that an immoderate increase in popula-
tion may provoke a real catastrophe in a naturally adverse en-
vironment.

If we accept, and we believe it is impossible not to do so, the
existence of the arid zone-wild plants and animals-human popula-
tion-domestic plants and animals complex, it is evident that it will
be necessary to seek a reasonable ecological equilibrium among
those four components. Any omission of one of these, or the mis-

PLANTS, ANIMALS, AND HUMANS 423

taken assumption that one may be independent of the rest, might
cause an irreparable maladjustment.

Any intensification of the utilization of wild plants and ani-
mals or any increase in the cultivation and breeding of domestic
plants and animals, beyond the natural or improved possibilities
of the environment, can be dangerous. In turn, any unplanned
population increase will immediately make greater demands on
the natural resources, which can be satisfied only at the risk of
grave consequences.

Summary Statement

OLAF S. AAMODT

Agricultural Research Service, United States De-
partment of Agriculture, Beltsville, Maryland

The authors have adequately summarized the major points in
their own papers. This summary by the Chairman will consist in
brief statements in answer to the six basic questions used by the
participants in the development of their subject.

What screening procedures would lead to the selection of more
productive plant and animal species for arid regions?

1. Selection in natural environments of preadapted types.

. Introduction and reselection in environments to be served.
3. Basic research on biotic factors influencing plant response.
4. Laboratory for basic studies under controlled environments

of (a) salinity, (2) climate, (c) diseases and other pests, (d) nutri-

tion, (e) heredity, (/) etc.
5. International cooperation on a coordinated testing program.
6. Regional technical conferences to screen ideas, as well as
methods and procedures, and to promote new lines of reasearch.

‘What are the genetic and physiological bases for drought re-
sistance in plants and animals?

1. Physiological. Basic research on water and irrigation re-
quirements.

2. Morphological. Characters that will not interfere with uti-
lization. ;

3. Habitats. (a) Escape of critical climatic periods. (4) Asso-
ciation with other species.

424

SUMMARY STATEMENT 425

4. Growth patterns and cycles of development adjusted to cli-
matic limitations.
5. Cultural patterns and management.

What are the prospects of increasing drought resistance through
genetic research?

1. Improvements in drought resistance can be made but will
be limited to inherited potentialities of available species, native
or introduced.

(a) Evaluation of indigenous material.

(2) Mode of inheritance of important characters.

(d) Development of controlled selective environments for
testing and evaluation.

(e) Selection of adapted superior plants under field condi-
tions.

(f) Ability to survive natural conditions and special treat-
ments.

How can we develop a program of revegetation?

1. Historical study of practices, past and present, and causes of
failures.
2. Analyze limiting micro-climatic factors in environment at
representative stations.
3. Collect preadapted races and evaluate basic adapted ma-
terial.
4. Introduce adapted foreign species and varieties.
5. Determine cultural requirements of promising species and
association of species.
6. Use vegetative cover for increasing ground water recharge.
7. Develop establishment procedures for revegetation with im-
proved species and strains.
8. Control grazing adjusted to plant requirements.
g. Determine number and kind of livestock.
10. Develop feed reserves for use during critical periods.

426 THE FUTURE OF ARID LANDS

11. Manage vegetation and livestock to balance forage with
livestock requirements.
12. Explore new patterns of land use and management.

What are the economic possibilities in the development and utili-
zation of arid land plants and animals?

Excellent with basic program of research, teaching, and exten-
sion.

1. Increase in knowledge and understanding of basic principles
involved.

Intensification of education and extension programs: (a)
translating knowledge at hand into action, (4) expansion of dem-
onstration areas of improved practices, (c) cooperation of public
and private agencies.

3. Modify land tenure and credit systems to meet needs of a
sound program of production and use.

4. Establish objectives and needs of programs on a national
basis.

5. Promote culture of food trees: olive, almond, carob, pis-
tachio, fig, and date.

6. Regional understanding and agreements on management
and use.

7. Agreements on relative values of competitive uses of water.

8. Classify and control livestock for more efficient land, use.

What are the possibilities of maintaining larger human popula-
tions in arid areas?

1. Possibilities will vary according to present population num-
bers, resources, and possibilities for development. Some areas al-
ready have population and resources out of balance. |

2. Some areas are favorable for expansion but improved stand-
a are more important as a first step.

3. Demineralization of water for human and livestock watering
will permit more efficient utilization of vegetation in some areas.

4. Stabilize population through integration of social and eco-
nomic factors.

Recommendations by the
Socorro Conference

Following the symposium at which the foregoing papers were
presented, 71 participants Joined in the field trip and then met at
the New Mexico Institute of Mining and Technology at Socorro
to examine the implications of these findings for new research. In
an informal atmosphere the group sought to identify the major
gaps in knowledge and possible ways of closing them. Stress was
placed upon lines of research that appeared to require cooperation
across national or disciplinary frontiers.

The first day was devoted to a review and evaluation of the
problems and development of the Rio Grande Valley as a typical
arid area and of problems and procedures involved in planning and
conducting integrated surveys of semi-arid and arid zones.

On the second day, the conference separated into three working
groups with the following general assignments: new approaches
needed in meteorology and applied climatology; the concept of the
water budget and its areal application; and closing the gap be-
tween scientific knowledge and its application to arid lands de-
velopment. Each group formulated a series of recommendations
to be considered by the conference as a whole on the final day. The
conference considered each recommendation, combined, modified,
and clarified some, and approved those that follow.

Anthropology, Archaeology, and Geography

1. A bibliography of our present knowledge of biological
adaptations of man and the cultural patterns in arid climates,
past and present, is needed to promote specific research in these
areas. Such research would contribute to the betterment of living
conditions and to planning for greater safe use of arid areas.

2. Further research, in addition to the diffusion of information,

427

428 THE FUTURE OF ARID LANDS

is needed concerning the history of land use, especially agriculture,
in arid and semi-arid regions. Information in this field has practical
applications in land-use planning, and our present knowledge is
very sketchy. The UNESCO Advisory Committee on Arid Zone
Research is urged to consider means of furthering research on this
subject and to consider the publication of a volume dealing with
agriculture of the past in the arid and semi-arid lands of the world.

3. Exploration is needed of possible new patterns of resource
use and practice with local participation in the studies to insure
the public understanding so necessary for achieving any change,
even on a gradual basis. There is a tendency to encourage the
maintenance of existing patterns, even when it is realized that
existing patterns have been inherited from conditions quite dis-
similar to those of the present. Land-use histories may be valuable
in dramatizing climatic hazards; there is need for long-term 1m-
provement in management; and even statistical data on climatic
change can be effectively and convincingly presented if they are
properly organized.

Meteorology and Climatology

4. The conference notes with satisfaction the recent action of
the UNESCO Advisory Committee on Arid Zone Research in
planning to devote the next arid lands symposium to climatologi-
cal problems of arid lands and urges sponsorship of continued re-
search on arid land climatology by the committee.

5. It is recommended that the program of the International
Geophysical Year, which previously emphasized polar observa-
tions, be extended in 1957-58 to include, to the maximum extent
possible, the arid belt of the world, and, in addition, that arid
zone countries involved be asked to participate in this program.
Although the original plan of the International Geophysical Year
has been expanded, the vast arid and semi-arid areas of Africa,
Asia, Australia, North America, and South America—30°N to
30°S principally—are still poorly represented in the list of longi-
tudinal and latitudinal sections fixed for intensive observations.

The observations of solar radiation and of other meteorological

RECOMMENDATIONS 429

elements on the surface and in the upper atmosphere in the arid
countries, and specifically along a parallel of latitude through as
many as possible of the world’s deserts, should be useful in the
solution of arid and semi-arid zone problems. Intensification of
observations at the national level will enable arid zone countries
to benefit even more from the international aspects of the Inter-
national Geophysical Year.

6. More effective climatic studies require an increase in density
and improvement in representativeness of meteorological stations
(both at sea level and at higher elevations) for surface and upper
air observations in all arid areas.

7. Careful attention should be paid to current research studies
concerning relationships of solar emanations and _ terrestrial
weather patterns, with particular attention to the effects that
may bear on arid land problems.

8. Synoptic and dynamic climatological studies, in different
arid and semi-arid regions, are essential. Emphasis should be
placed on interrelationships between the general circulation of the
atmosphere at upper levels as well as at the surface and the
precipitation in different parts of the areas and at different times
of the year. With such studies as a basis for the development of
understanding, prediction of precipitation within the area in
question may follow.

g. Inasmuch as the matter of the evaluation of the results
following attempts to modify weather and weather processes is
recognized as offering great possibilities for the peoples of arid
lands, every effort should be made to develop improved techniques
for statistical evaluation of weather modification experiments and
to use the best available present techniques and data in the
analysis of the results of such experiments.

1o. An international cooperative program of synoptic observa-
tions should be instituted to determine the concentration of ice-
forming nuclei throughout the world, especially during periods of
the earth’s passage through meteoritic streams. These observa-
tions should be supplemented by measurements, synoptic if pos-
sible, of the concentration of naturally and industrially induced

430 THE FUTURE OF ARID LANDS

condensation nuclei (including giant hygroscopic nuclei) by
studies of the chemical composition of precipitation and by the
ooo. of cloud surveys.

» AL fNOIRS vigorous study of all possible aspects of periodic
ce seeding is imperative.

10% Dreger: knowledge of nucleation properties of silver iodide
as affected by the methods of generation and dispersal is inade-
quate. Further studies, with particular reference to the decay of
the nucleating activity of silver iodide with increasing time of
exposure in the atmosphere, are recommended.

13. Closer integration of the sciences of climatology and hy-
drology can be fostered through better exchange of information
and collaborative analyses aimed at improving joint methodology.
The lack of such collaboration between climatologists and hy-
drologists has contributed in the past to inadequate estimation of
available water resources in some arid-zone projects.

Recommendations 17, 19, 25, and 29 are also applicable.

Hydrology, Geology, and Soils

14. The importance of ground water in arid zones calls for
continued research on the following aspects of this subject: (1)
methods of exploration and estimation of the volume of ground
water bodies; (11) methods of increasing ground water recharge
and of estimating rates of recharge; (111) the relation of vegetation
and other biological factors to ground water recharge; (iv) the
geomorphological aspects of the occurrence and chemistry of
ground water.

15. The precipitation occurring on drainage basins should not
be regarded in terms of utilization for irrigation alone, and more
consideration should be given to planning for the beneficial use of
water that is not reaching points of downstream use.

16. Continued study is needed of the factors and practices
modifying soil structure under various land-use practices, such as
grazing, dry land farming, and irrigation farming, recognizing the
importance of soil structure and its maintenance in relation to
permeability and to prevention and abatement of erosion.

17. The work of hydrologists and climatologists would benefit

RECOMMENDATIONS 431

greatly by the fullest possible use of vegetation studies, specifically
by the consideration of the role of vegetation as a factor in the
hydrology of dry lands and plant species and communities as
indicators of climates, past and present.

18. Further attention should be given to the study of the geo-
morphic dynamics of landscapes for application to regional and
land-type appraisal and land-use planning.

Also see recommendations 13, 21, 25, 28, and 30(i1).

Biology, Ecology, and Conservation

1g. Intensive studies of the microclimatic environments of
plants and animals should be encouraged and pursued. The rela-
tionships between the data usually recorded by meteorological
stations and the microclimatic effects of these phenomena in
different sections of typical arid environments should be subjected
to intensive study.

20. Research on plant and animal ecology, improvement, and
management in arid areas should be intensified. Emphasis should
be placed on preadapted species and races, on understanding of the
physiological factors in the selection of characteristics desired in
breeding, on water requirements of the various species and breeds
as related to production, and on utilization of available soil and
climatic resources by plants and of available vegetation by various
species and breeds of animals.

21. Additional research is needed on the methods of determina-
tion and the estimation of the water requirements of plants in
arid regions, especially on the efficiency of transpiration, the rela-
tionship between transpiration and photosynthesis, and the regu-
lation of transpiration. The suggestion is made to the UNESCO
Advisory Committee on Arid Zone Research that it compile a
review of information and research studies currently available on
this subject.

22. Intensified research should be undertaken, and research
results should be applied in the management of grazing, because
of the paramount importance of grazing management in the con-
servation and improvement of arid grasslands.

23. A thorough investigation should be made of indigenous

432 THE FUTURE OF ARID LANDS

plants of arid and semi-arid regions with a view toward deter-
mining their usefulness and adaptability to grazing and culti-
vation.

24. There is reason to believe that studies of pharmaceutical
and industrial uses of desert plants would be justified.

25. Intensified studies should be made on the formation,
measurement, and utilization of dew to determine its potentialities
as a supplement to rainfall in arid regions. Such studies should
encompass the utilization of dew by plants and the selection of
plants most efficient in such use; the relationship of dew to soil
moisture; and the establishment of physical relationships for ex-
tracting dew from the atmosphere.

26. Natural arid land ecological communities of indigenous
animals and plants in their original habitats are essential for
educational and scientific purposes. Areas of adequate size should
be acquired and preserved in the various arid land countries.

Also see recommendations 1, 2, 3, 14, 16, 17, 18, and 30(, 11).

Organization, Communication, and Interdisciplinary Programs
27. The UNESCO Advisory Committee on Arid Zone Research

is urged to revise and reissue its list of national and international
scientific institutions concerned with arid land problems. The re-
vised list should be as comprehensive and as up to date as possible
and should include addresses and fields of interest in order to
serve effectively for intercommunication among workers in various
disciplines.

28. Permanent cooperation in connection with studies on the
demineralization of salty and brackish water should be main-
tained among the UNESCO Advisory Committee on Arid Zone
Research, the U.S. Saline Water Conversion Program, and Work-
ing Party No. 8 on demineralization of salt and brackish waters
of the Organization for European Economic Cooperation (OQEEC)
with the objective of adapting technical possibilities to local needs
and economic resources.

29. Interdisciplinary studies should be promoted in order to
sharpen the concepts used in defining, delimiting, and classifying

RECOMMENDATIONS 433

arid and semi-arid lands, with special emphasis on the variability
of precipitation.

30. A demand for the application of scientific and scholarly
knowledge in arid areas should be created by means such as
those indicated in items 1-1v. It should be noted that in many
situations the driving force necessary for getting available knowl-
edge applied to improvement of land and water utilization 1s
lacking. This driving force is essentially public demand or social
pressure. Creating such demand is the most effective method of
attaining the desired end. (1) Expansion of demonstration areas,
even though they have some disadvantages. They stress manage-
ment by practical operators, such as commercial or family farmers,
where demonstration of practical value are sought. However,
demonstrations designed for the promotion of specific under-
standing of resource problems and techniques by business and
political leaders, and even by technical men themselves, should be
maintained. Cooperative interdisciplinary demonstrations on
single resource management problems are useful. They have been
successfully extended to include treatment of the entire resource
pattern in areas of some size. Complex demonstrations of inte-
grated resource management on a scientific basis not only are
proved but also deserve more intensive use. Demonstrations may
be supported entirely by public funds, partly by public funds, and
entirely by private funds. The possibility of extending the useful-
ness of the demonstration technique under private auspices is a
relatively new subject that deserves further attention. (11) Re-
search in the social sciences, exploring limiting factors that have
tended to keep knowledge from application. Much more knowl-
edge is needed concerning the social and economic factors that
influence the development and application of science, and con-
cerning the art of persuading people to take action in resource de-
velopment to their own and their community’s long-term benefit.
Special attention should be paid to economic and social studies
that can throw light on the relative values of competitive uses of
water. However, the cooperation of engineering and natural
sciences in such investigations will be essential, together with the

434 THE FUTURE OF ARID LANDS

collaboration of geographers and anthropologists. (111) Enlistment
of local leadership and local interests in support of both basic
research and practical studies designed to advance present knowl-
edge and to transfer such knowledge into a form for direct applica-
tion. Local groups will profit by taking an active part in and by
supporting these programs. In many fields in some parts of the
world institutes have developed techniques for enlisting the aid
of private industry in some broad public programs. These tech-
niques may be applicable in areas other than the limited ones
where they have been applied up to the present, but applications
will vary with cultures and local conditions. (iv) Greater attention
by scientists to their relations with the press, radio, television, and
other channels of public information. Such relations involve an
opportunity and a dual responsibility—responsibility to science
for the presentation of an accurate and complete report and to the
particular audience for framing scientific material in a form that
will most effectively reach the public whose interest in resource
development they hope to arouse and to inform. Also, more
attention should be paid to the art of communication and its
demands in the training of scientists.

31. More effective interdisciplinary pooling and dissemination
of information should be developed for the purposes of advancing
science, as well as public understanding of scientific matters, by
the following. (1) Establishment of local and national committees
on arid zone problems to enlist public interest in support of
studies of arid lands and the dissemination of information on
results of such studies in each country. The nature of such com-
mittees and their method of formation ought to be locally deter-
mined in order to meet special conditions in each country, but it 1s
strongly urged that they be broad in scientific disciplines and in
representation from both private and public agencies. These com-
mittees should operate in a manner best suited to the interests
and possibilities of each country and should be aimed at encourag-
ing research and spreading information, utilizing UNESCO as a
clearing-house in this field. (ii) Creation of a preliminary project
to explore the feasibility of an abstracting service on arid zone
literature. A periodical, patterned after existing successful ab-

RECOMMENDATIONS 435

stracting journals, would include, as soon after original publica-
tion as possible, abstracts of technical economic, and social litera-
ture related to arid zone problems and research. Consideration
should be given to the desired business and production organiza-
tion, the volume of material to be included, the subject matter
divisions, the availability of abstractors, the cost of publication,
the required subscription price, and so forth. A target date for
the report on this feasibility study should be 1 year from the
adoption of this recommendation by some agency capable of
committing funds. (111) Encouragement of the formation of re-
search organizations, comprehensive in discipline and concerned
with the best use of specific limited resources. Such organizations
should be encouraged in all arid lands through adequate and
broadly based financial support and through organized community
interest.
Also see recommendations 2, 3, 4, 5, 10, 13, 18, and 21.

Committees for the Meetings

American Association for the Advancement of Science Planning
Committee

Chairman:

Gitsert F. Wuite, President, Haverford College; United States mem-
ber of the UNESCO Advisory Committee on Arid Zone Research

Watiace W. Arwoop, Jr., Director of the International Division,
National Academy of Sciences—National Research Council

Joun A. BEeHNKE, Associate Administrative Secretary, American As-
sociation for the Advancement of Science

Peter C. DutsBerG, Southwestern Irrigated Cotton Growers Associa-
tion and Desert Products Company; Chairman of the Committee
on Desert and Arid Zone Research of the Southwestern and Rocky
Mountain Division of the AAAS

Gove Hampipce, Regional Representative, Food and Agriculture
Organization of the United Nations

GeorceE R. Puixitps, Soil Conservationist, Soil Conservation Service,
U. S. Department of Agriculture

Water M. Rupotpu, Assistant to the Science Advisor, Office of the
Science Advisor, U. S. Department of State

Exvin C. Straxman, Emeritus Professor of Plant Pathology, University
of Minnesota, St. Paul, Minnesota

E. C. Sunpertin, Deputy Director, National Science Foundation

Hersert Tuo, Chief Climatologist, U. S. Weather Bureau

Dart Wo tre, Executive Officer, American Association for the Ad-
vancement of Science, ex officio

Conference Committee

Chairman:
Gitsert F. Wuire, President, Haverford College, Haverford, Pennsyl-
vania

Georces AusBeEert, Chef du Service des Sols de l’Office de la Recherche
436

COMMITTEES 437

Scientifique et Technique Outre-Mer, Paris, France; Member of
UNESCO Advisory Committee on Arid Zone Research

B. T. Dickson, Retired Chief, Division of Plant Industry, Common-
wealth Scientific and Industrial Research Organization, Canberra,
Australia; Member of UNESCO Advisory Committee on Arid
Zone Research

Luna Leopo.tp, Water Resources Division, Geological Survey, U. S.
Department of the Interior, Washington, D. C.

Paut B. Sears, Professor of Conservation, Yale University, New
Haven, Connecticut; President Elect of AAAS

Joun A. BEHNKeE, Associate Administrative Secretary, AAAS, Wash-
ington, D. C. (ex officio)

The Southwestern and Rocky Mountain Division of the AAAS
Local Committee

Chairman:

Perer C. Duissperc, Southwestern Irrigated Cotton Growers Associa-
tion and Desert Products Company; Chairman of the Committee
on Desert and Arid Zone Research of the Southwestern and Rocky
Mountain Division of the AAAS

Anton Berkman, Head, Department of Biology, Texas Western College,
E] Paso; Chairman of Finance Committee and Treasurer

E. F. Castretrer, Dean of the Graduate School, and Chairman, De-
partment of Biology, University of New Mexico, Albuquerque,
New Mexico; Chairman in charge of Symposium arrangements

J. L. Garpner, Research Conservationist, Agricultural Research Service,
Soil and Water Conservation Research Branch, U. S. Department
of Agriculture, State College, New Mexico; Chairman in charge of
Field Trip

Frank E. E. Germann, Professor of Chemistry, University of Colorado,
Boulder; Executive Secretary of the Division

Errx K. Resp, Regional Chief of Interpretation, National Park Service,

_ U.S. Department of the Interior, Sante Fé, New Mexico

Joun L. Hay, Hydraulic Engineer, Engineering Department, City of
El Paso, Texas; Chairman in charge of Publicity and Education

H. L. Srannxe, Head, Department of Biological Science, Arizona State
College, Tempe, Arizona

E. J. Workman, President, New Mexico Institute of Mining and Tech-
nology, Socorro, New Mexico; Chairman in charge of Conference
arrangements

438 COMMITTEES
Southwestern Committees

Finance

Chairman:
Anton H. Berkman, Texas Western College, El Paso

Joaguin R. BusraMeEnTe, Comision Internacional de Limites y Aguas,
Juarez, Chihuahua, Mexico

Herp FL LetcHer, Forest Service, U. S. Department of Agriculture,
Tempe, Arizona

Joun S. Hawkins, Glenncannon Engineering Company, El Paso,
Texas

Witiiam J. Koster, University of New Mexico
Epwin B. Kurtz, University of Arizona, Tucson

Frep Lavin, Agricultural Research Service, U. S. Department of
Agriculture, Tempe, Arizona

Eart M. Movtton, Albuquerque Chamber of Commerce
DanteEL Rosinson, Arizona State College, Tempe

Leo ScuusTeER, JR., Schuster & Skipworth Insurance Agency, El Paso,
Texas

Moras L. Suusert, University of Denver, Denver, Colorado
HERBERT L. STaHNKE, Arizona State College, Tempe
Icnacio ZunNzuNEGUI, Anderson-Clayton Co., Juarez, Mexico

Education and Publicity

Chairman:
Joun L. Hay, Engineering Department, City of El Paso, Texas

Witu1aMm B. Ackerman, City Planning Department, City of El Paso,
Texas

C. W. Borxin, Las Cruces, New Mexico

J. H. Buets, Radio-TV Station KROD, EI Paso, Texas

Crark Cuampie, Jefferson High School, El Paso, Texas

Joun R. Cropton, University of Colorado, Boulder

Bertua P. Durron, Museum of New Mexico, Santa Fé, New Mexico

Harotp B. Ermenporr, Soil Conservation Service, U. S. Department
of Agriculture, Albuquerque, New Mexico

G. Warp FEntey, University of New Mexico (In charge of Press Room)
W.S. Foster, County Agricultural Agent, El Paso, Texas

COMMITTEES 439

Lawrence FREEMAN, Soil Conservation Service, U. S. Department of
Agriculture, El Paso, Texas

Met Geary, El Paso Times, El Paso, Texas

Harper Griswo 1p, Tree Surgeon, El Paso, Texas

MarsHa.t Hatt, El Paso Herald Post, El Paso, Texas

Jesse A. Hancock, Texas Western College, El Paso (Speakers’ Bureau)
Mrs. Witiarp C. Korrxe, Farmington, New Mexico

Pepro Picasso, Secretaria de Agricultura y Ganaderia, Juarez, Mexico
James McCreary, Arizona State College, Tempe

Victor H. ScHorreLMAYER, Glendale, California

Field Trip

Chairman:

J. L. Garpner, Agricultural Research Service, U. S. Department of
Agriculture, State College, New Mexico

BrewsTeER Batpwin, New Mexico Bureau of Mines and Mineral Re-
sources, Socorro

Frank E. Korrtowsx1, New Mexico Bureau of Mines and Mineral Re-
sources, Socorro

Erik K. Reep, National Park Service, U. S. Department of the Inte-
rior, Santa Fé, New Mexico

Albuquerque Committee on Arrangements

Chairman:

E. F. Casretrer, University of New Mexico

Howarp J. Dirrmer, University of New Mexico

Mrs. W. J. Eversoxe (Ladies’ Entertainment)

G. Warp Fen ey, University of New Mexico (Press Room)
Martin W. F eck, University of New Mexico

J. L. Garpner, Agricultural Research Service, U. S. Department of
Agriculture, State College, New Mexico

Bert Hurrman, Albuquerque Chamber of Commerce

Joun E. Kircuens, University of New Mexico

Witutam J. Koster, University of New Mexico

Sruart A. Norrurop, University of New Mexico

J. L. Riessomer, University of New Mexico

E. J. Workman, New Mexico Institute of Mining and Technology,
Socorro

440 COMMITTEES

Socorro Committee on Arrangements

Chairman:
Cray SMITH

J. E. ALLEN, Housing

S. E. Reyno.tps, Refreshments
H. E. Sy.vesrer, Stenographic Service

A. J. THompson, Local Transportation

V. Vacguier, Linguist

R. H. Weser, Auditorium Arrangements

(All of New Mexico Institute of Mining and Technology)

Participants in the Socorro
Conference

Aamopt, O. S. Technical Specialist, Plant Sciences, Agricultural
Research Service, U. S. Department of Agriculture, Beltsville,
Maryland

ACKERMAN, Epwarp A. Director, Water Resources Program, Re-
sources for the Future, Washington, D. C.

ARMILLAS, PEDRO Professor, National School of Anthropology, Mexico,
D. F.; and Technical Collaborator, Inter-American Indian Insti-
ture, Mexico, D. F.

AvuBert, GeorGes’ Chef du Service des Sols de |’Office de la Recherche
Scientifique et Technique Outre-Mer, Paris, France

Barttey, Reep W. Director, Intermountain Forest and Range Experi-
ment Station, Forest Service, U. S. Department of Agriculture.
Ogden, Utah

BeEHNKE, Jonn A. Associate Administrative Secretary of AAAS, Wash-
ington, D. C.

Bettran, Enrique Director, Instituto Mexicano de Recursos Natu-
rales Renovables, Mexico, D. F.

Boxe, Rrcuarp L. Reynolds Metals Co., Louisville, Kentucky
Bowen, E.G. Chief, Division of Radiophysics, Commonwealth Scien-
tific and Industrial Research Organization, Sydney, Australia

Boyxo, Mrs. ExizasetH Rehovot, Israel

Borxo, Huco Rehovot, Israel

Branam, Roscoe R., Jr. Associate Professor, Department of Meteor-
ology, University of Chicago; and Director, Institute of Atmos-
pheric Physics and Associate Professor of Physics, University of
Arizona, Tucson

Brier, Grenn W. Chief, Meteorological Statistics Section, Weather
Bureau, U. S. Department of Commerce, Washington, D. C.

Curistian, C. S. Officer-in-Charge, Land Research and Regional Sur-
vey Section, Commonwealth Scientific and Industrial Research
Organization, Canberra, Australia

CrauseEN, Jens Department of Plant Biology, Carnegie Institution of
Washington, Stanford, California

Dickson, B. T. Retired Chief, Division of Plant Industry, Common-
wealth Scientific and Industrial Research Organization, Canberra,
Australia

44]

442 PARTICIPANTS

Dixey, F. Director of Colonial Geological Surveys, Imperial Institute,
London, England

Dorrou, Joun H., Jr. Hydrologist, Soil Conservation Service, U. S.
Department of Agriculture, Albuquerque, New Mexico

Draz, Coronet Omar Director, Desert Range Development Project,
Desert Institute, Mataria, Egypt

Duisperc, Peter C. Southwestern Irrigated Cotton Growers Asso-
ciation, and Desert Products Co., E] Paso, Texas

Epersote, Gorpon K. Chief, Training Administration Branch, Bu-
reau of Reclamation, U. S. Department of the Interior, Washing-
ora, 1D), C.

E.-AsHKAR, MauMoup, A. Associate Professor of Soil Science, Uni-
versity of Alexandria, Egypt

Evenari, Micuaet Professor of Botany and Vice President, Hebrew
University, Jerusalem, Israel

FourniER D’AtBeE, E. M. Mision de la UNESCO, Universidad Na-
cional de Mexico, Mexico, D. F.

GaRDNER, J. L. Research Conservationist, Agricultural Research Serv-
ice, U. S. Department of Agriculture, State College, New Mexico

GERMANN, Frank E. E. Professor of Physical Chemistry, University
of Colorado, Boulder

GREENE, HerBert Adviser on Tropical Soils to the British Colonial
Office, Rothamsted Experimental Station, Harpenden, England

Hickox, Roserr B. Senior Hydraulic Engineer and Project Super-
visor, Southwest Watershed Hydrology Studies, Agricultural Re-
search Service, U. S. Department of Agriculture, Albuquerque,
New Mexico

Koenic, Louis Associate Director, Southwest Research Institute, San
Antonio, Texas

Krut, W. F. J. M. Professor, University of Delft; and Director, Gov-
ernment Institute of Water Supply, The Hague, Netherlands

Lanemutr, Irvine Consultant, Research Laboratory, General Elec-
tric Company, Schenectady, New York

Leopotp, Luna B. Water Resources Division, Geological Survey, U.S.
Department of the Interior, Washington, D. C.

Lewis, M. R. Consultant to the Institute of Water Utilization, Uni-
versity of Arizona, Tucson

Love, R. Merton Professor of Agronomy and Agronomist in the Ex-
periment Station, University of California, Davis

Maztovum, Sousut Director of Irrigation and Hydraulic Energy, Minis-
try of Public Works, Damascus, Syria

McCretran, L. N. Assistant Commissioner and Chief Engineer, Bu-
reau of Reclamation, U. S. Department of the Interior, Denver,

Colorado

PARTICIPANTS 443

McGinntes, Witi1am G. Director, Central States Forest Experiment
Station, Forest Service, U. S. Department of Agriculture, Colum-
bus, Ohio

Meics, Pevertt Chief, Regional Research Section, Quartermaster Re-
search and Development Center, U. S. Department of the Army,
Natick, Massachusetts

Morenaar, ALDErT Irrigation and Drainage Specialist, Agriculture
Division, Food and Agriculture Organization of the United Nations,
Rome, Italy

Monon, THEODORE Professeur Au Muséum National d’Histoire Natu-
relle, Paris, France

Munoz, Cartos Director, Departemento de Investigaciones Agricolas,
Ministerio de Agricultura, Santiago, Chile

Nagvi, S. N. Director, Pakistan Meteorological Service, Karachi,
Pakistan

NicuHotson, N. L. Acting Director, Department of Mines and Tech-
nical Surveys, Ottawa, Canada

Puitiips, GEorGE R. Assistant Director, Planning Division, Soil Con-
servation Service, U. S. Department of Agriculture, Washington,
ID). (Ce

Picui-SERMOLLI, Ropotro E.G. Curator of the Herbarium, Botanical
Institute, University of Florence, Florence, Italy

Price, Raymonp Director, Rocky Mountain Forest and Range Ex-
periment Station, Forest Service, U. S. Department of Agriculture,
Fort Collins, Colorado

Reyno ps, S. E. Research and Development Division, New Mexico
Institute of Mining and Technology, Socorro

Satin, Kanwar Chairman, Central Water and Power Commission,
Ministry of Irrigation and Power, New Delhi, India

Sayre, A. Nerson Chief, Ground Water Branch, Geological Survey,
U. S. Department of the Interior, Washington, D. C.

ScHAEFER, VINCENT J. Director of Research, The Munitalp Founda-
tion, Schenectady, N. Y.

Scumipt-NIELsEN, Knut Professor of Zoology, Duke University, Dur-
ham, North Carolina

Sears, Paut B. Professor of Conservation, Yale University, New
Haven, Connecticut

SHantz, Homer L. Santa Barbara, California

SHieLtps, Mrs. Lora M. Head, Department of Biology, New Mexico
Highlands University, Las Vegas

SMILEY, TERAH L. Geochronologist, Laboratory of Tree-Ring Re-
search, University of Arizona, Tucson

Sniper, Rogpert G. Director of Research and Vice-President, Con-
servation Foundation, New York, N. Y.

444 PARTICIPANTS

STERNBERG, HircarD O’Rettty Professor of Geography, and Director,
Centro de Pesquisas de Geografia do Brasil, Universidade do Brasil,
Rio de Janeiro, Brazil

Sunpstrom, R. W. District Engineer, Ground Water Branch, Geologi-
cal Survey, U.S. Department of the Interior, Austin, Texas

Swarprick, James A. Deputy Head, Division of Scientific Research,
UNESCO, Paris, France

Tuo, Hersert Chief Climatologist, Weather Bureau, U. S. Depart-
ment of Commerce, Washington, D. C.

TrxERoNT, JEAN Ingenieur en Chef des Travaux Publics, Tunis,
Tunisia

Uvarov, B. P. Director, Anti-Locust Research Centre, London,
England

VaLENTINE, K. A. Associate Animal Husbandman, Department of
Animal Husbandry, New Mexico College of Agriculture and Me-
chanic Arts, State College, New Mexico

Watten, C. C. Assistant Director, Swedish Meteorological and Hy-
drological Institute, Stockholm, Sweden

Went, F. W. Professor of Plant Physiology, Earhart Plant Research
Laboratory, Division of Biology, California Institute of Technology,
Pasadena, California

Watre, Girpert F. President of Haverford College, Haverford, Penn-
sylvania

Wiccins, Ira L. Professor of Biological Sciences and Director, Natu-
ral History Museum, Stanford University, Stanford, California

Witcox, L. V. Asistant Director, U. S. Salinity Laboratory, Agricul-
tural Research Service, U. S. Department of Agriculture, River-
side, California

Workman, E. J. President and Director of Research and Development
Division, New Mexico Institute of Mining and Technology, So-
corro, New Mexico

Index

Aamodt, O. S., 345
Abstracting service, proposed, 434
Acanthosicyos horrida, 17
Achillea fragrantissima, 399
Achillea santolina, 399
Adaptation, of animals, 12-155,
33142
of plants, 7-12, 332, 338, 343-67
Admiralty Materials Laboratory,
research, 260, 274
Adropogons species, §
Advisory Committee on Arid Zone
Research (see UNESCO)
Africa, 16
arid area, 121-22
basic data collection, 137
climatic maps, 77
climatic variation, 123-25
East, use of wells, 126
location of ground water, 136-37
locust areas, 384-85
North, 34, 35, 51
recharge of aquifers, 132-35
South, 17, 52
surface waters, 131-32
usable ground water, 135
water budget, Zeerust, 75
well yields, 126-31
Agave sisalana, 399
Agricultural settlement,
398
Agropyron species, 186
Agropyron cristatum, 336
Agropyron desertorum, 238, 338,
346, 353
Agropyron junceum, 398, 401
Ahlmann, H-W., 167

Negev,

Aira caryophyllea, 352
Alfalfa, 60
Algeria, 14
demineralization research, 276
ecological study, 182
hydrological studies, 123
rural improvement, 184
salinity research, 273
Aleem We G23
Anabasis haussknechtit, 399
Ancient man, 18, 31, 51, 402
Andean America, 248
Animals, adjustments, 12-15
ecology, research needs, 431
physiology, 370-80
production improvement, 22-23,
337-39, 368-79, 377-82
selection, 61, 333-36
Antevs, =, 166, 167
Anthropology, needed
427
use, 249
Aquifer (see Ground water)
Archaeology, 51-52, 92-93, 98,
392-935, 4
needed research, 427-28
Arid zone, 3~7, 27, 77) 432
of Africa, 121-22
of Brazil, 203-8
extent, 48
Tunisia, 85
Aristida ciliata, 186
Arizona, 5
control of worthless plants, 238
introduced grasses, 238
precipitation, 156, 235, 162-70
water need and yield, 81, 156

research,

445

446
Arroyo cutting, 11§—-17, 166, 234—

35
Artemesia herba-alba, 399
Artemesia monosperma, 399, 401
Asia, 16 (see also India)
Central, 51
locust areas, 384
Atriplex halimus, 397, 398
Atriplex sembibaccata, 398
Atriplex species, 338
Atriplex spongiosa, 398
Atriplex vesicaria, 398
Amoiserllial, As, Ady Ag, 58, OS
grasshopper area, 384
meteoritic particles, 295-99
seeding of clouds, 292
surface waters, 131
Avena barbata, 186
Avena species, 352, 354
Avena sterilis, 398

Babcock, W. M., 335

Bailey, R. W., 419

Bellue, M. K., 351

leary, IL. |o5 ASO, Bz

Bolton, J. G., 292

BOSE, Vo Nt UA TG), WDE

Bouteloua species, 239

Bouteloua curtipendula, 347

Botany, needed research, 22, 231,

431

Bowen, E. G., 280, 301, 311, 316

Brazil, better land use, 212-17
drought polygon, 205-8
Nordeste, 200-5
water storage, 208-12

Breeding, of animals, 335-37, 431
of plants, 24, 39, 349, 414-18

Brewer Ve lleen 350:

Bromus rigidus, 354, 355

Bromus rubens, 354

Bromus species, 346, 347, 352, 354

Browne, W. R., 128

Brunson, A. M., 348

Bryan, K., 167

INDEX

Cactus, 38
Caldwell, M. C., 22
California, evaporation, 71-72
field tests on range, 345-47
grass experiments, 359-62
recharge of aquifers, 133
solar energy, 63
vegetation changes, 351-57
Calligonum comosum, 398, 401
Camel, 334
grazing area, 377-80
use of water by, 372-80
Canada, 45, 186
Canaigre (see Rumex hymenose-
palus)
Candelilla wax, 414-15
Cappario spinosa, 399
Carbon-14 dating, 29, 166-67
Cenchrus species, 60
Ceylon, 49
Chaffey, G., and W. R., 47
Citrullus colocynthts, 399
Clausen, R. E., 347, 349
Climate, 4-5, 28-30, 67-84, 87-94
Climatic changes, 29, 51, 93, 123-
Noe Wole—FAit, WO), NOP)
Climatology, needed research, 28—
29, 74, 142, 155, 428-30
Cloud surveys, 310, 430
Colorado, 5, 6, 19, 20, 28
precipitation and water yield,
156
water need, 81
Colorado River, 173
Columbia River, 191, 192
Commonwealth Scientific and In-
dustrial Research Organiza-
tion of Australia, 186
Colutea tstria, 394, 398
Combined resource use, 44, III,
IgO-g2, 217-18, 267, 420-23
Conrad oy,
Convolvulus species, 401

Cooper, J. P., 350

INDEX

Cooperative research, 417-18
Cooperative surveys, 182, 391-92
Cynodon dactylon, 186, 338, 398

Dactylis glomerata, 338, 346
Daemia tomentosa, 399
Pulls, 18.5 On
Danithonta californica, 352
Davie Bs. 352
de Cillis, U., 187
de Martonne, E., 6
Demonstration areas, 433
Denmark, wind power, 62
Desert agriculture, Israel, 390-413
Desert Symposium in Israel, 35
Deserts, man-made, 21, §9, 124-25,
IVO=BM, DIOR, Qi
Dew, 409-10
needed research, 432
Directory of institutions, 432
Donkey, grazing area, 377-380
Dominican Republic, water budget,
7
Diaz O- Ol, 370
Drought, 52, 343
enduring plants, 12, 344
enduring animals, 15, 334, 372-80
escaping animals, 13
escaping plants, 11, 344
evading animals, 13
evading plants, 11-12
ECIEEOM, FO, GOO, NAG, We
73
resistant animals, 14-15
resistant plants, 12, 343, 424-25
Drought resistance, genetic bases,
My 5e
physiological bases, 344-47
Drouhin, G., 122
Dry-land farming, 18-20, 42-43,
95, 98, 227-32, 390
Dubief, J., 88
Duishercomes C2022 aK
Duvdevani, S., 409

447

Ecology, 392, 431
Economics of water, 263-66, 274—-
76, 320-28
Economics, needed research, 324
Edgeworth, D. W. T., 128
Erharta calycina, 338, 346, 347
Elymus glaucus, 346
Eragrostis species, 238, 401
Erodium species, 352, 354
Erosion, accelerated, 31, 117, 175,
228, 234, 242
Esparto, 22
Eucalyptus rostrata, 400
Euphorbia species, 414
Evaporation, 69-73, 197-98
Evapotranspiration, 52-53, 67-73,
95-96
conditions affecting, 73
potential, 53, 67, 70, 73, 158
Experimental settlements, 392, 410
Egypt, range experimentation, 186,

SOSo
water budget, Gaza, 75

Farm credit, 86, 229
Fertilizers, 357-58
Festuca arundinacae, 346
Festuca idahoensts, 362
Festuca myuros, 352
Flood frequency, 89, 101, 118
Flood plains, 116, 408
Florida, evaporation, 72
Food and Agriculture Organiza-
tion, 49, 59, 179, 182, 331, 335
Forests, relation to grazing lands,
187, 242, 251
Frandsen, K. J., 350
French Morocco, demineralization
research, 276
range improvement, 182, 186
saline research, 273

Fries, M., 167

448

Gastridium ventricosum, 354
Genetics, needed research, 22, 24,
425
Geochronology, 29, 161~71
Geography, 200-20
needed research, 427-28, 431-33
Geology, 392
needed research, 430-31
Geomorphology, 30-32, 210-12
needed research, 32, 117, 119
Geven, U2 Wo, 138
Gibrat, G., 88
Grasshopper, 383-89
Grazing (see Range)
Great Plains, 19, 228
Gregor, J. W., 348
Ground water, 23, 32-34, 53-54,
IOS), WAR, WIS)
Africa, 135-37
exploration, 125-26
needed research, 34, 56, 174, 430
recharge, of aquifers, 56, 105-6,
131, 132-34
salinity, 107, 127-31, 267
use and misuse, I0g-II, 126-27,
198
Guayule, 22, 414
Gunn, R., 316
Gypsophila rokejeka, 399

Hagedoorn, A. L., 336
Halogeton, 354

Haloxylon persicum, 398, 400
Hample, C. A., 261
Hawkins, R., 261
Hayward, H. E., 57, 265
lefterninanep hesn| 2g
Heinrich, R., 9

Hemizonia, 356

Hendry, G. W., 351
Heyne, E. G., 348

Hiesey, W. M., 347
Hilgard, E. W., 352
Homestead laws in U.S., 19

INDEX

Homoclimates, 29, 74, 336

Hopi Indians, 16-17

Hordeum bulbosum, 396, 398

Hordeum murinum, 352

Hordeum vulgare, 350

Horse, 335

Horton, R. E., 81

Howard, C. S., 283

lakoyes Jen We, B77, 202

Human health, 40, 61-62

Hydrophytes, 7-9

Hydrology, 85-103, 123, 134
needed research, 141-43, 197,
ASS

Hyoscyamus muticus, 399

Hyparrhenia hirta, 186

Hypericum perforatum, 354, 356

Ice nuclei (see Precipitation-in-
duced)
Ula, Wile Ss, Qua
India, 51, 180-82
Council of Agricultural
search, 182
Desert Research Station, 251
Rajasthan desert, 250, 253
solar energy, 62
Indus River, 47, 49, 54
Interdisciplinary programs, 432-35
Interior drainage, 6
International cooperation, 45, 63,
MGV MNt7) ASOLO, Aj 7%,
396, 401, 428-35
International Geophysical Year,
428
International
need, 410
Inter-Territorial Hydrological Con-
ference, British Africa, 134
Rontess ines .2774— 75
Irrigation, 43
ancient works, [6,0 2lgOsmgle
98, 246, 248, 402
boron limits, 283

Re-

research center,

INDEX

canal losses, 19§

drainage, 222

flood, 17, 403-8

India, 48

justification, 43, 193-94, 326-28

leaching requirement, 223-24,
251

permanency, 221, 250

plant nutrients, 37

potential, 49, 55, 279, 190-91

relation to grazing, 184

return flow, 283-86

salinity problems, 56-57, 129,
DDI=DS, PHO, PS eI, Pheyne
285-88

sodium factor, 224-25, 281, 282,
286

soil characteristics, 35, 403

United States, 279

water management, 80-81, 196,
DSR. O'S

Israel, Negev desert, 57
development, 390-408

acke Res) 1285 129

Jenkins, D. S., 266, 273

jjomes, 18, Wa, BEG, sao

Julander, O., 345, 347

Funcus acutus, 401

Funcus maritimus, 394, 399, 401,
409

Kalahari desert, 17, 127
Kangaroo rat, 14-15, 371
Keck= DE Dr 347
Kellogg, C. E., 58
Knowles, P. F., 347
Kochia indica, 336, 338

Laboratory of Climatology, 77
Land capability, concept, 226-27
Land use, drought risk, 203, 229-
30
range lands (see Range)

449

relation to grasshoppers and
locusts, 386-87
social factors, 229-30
NKaeleel iENezTEN, OS—A, Aoy/,
431, 433
Land values, 325
Landsberg, H., 152
Langmuir, I., 301, 307, 311
anew Dae ia3
Larfrére, L., 6
Larrea divaricata, 353
Larrea tridentata, 344
Laude, H. M., 345
Leopold, L. B., 32
Leptadenia pyrotechnica, 399
WibbysaVe ress
Libya, 3
Local and national committees,
434
Locust, 383-89
Lolium species, 60, 350, 351, 352
Lolium perenne, 346
Lolium rigidum, 186

Madson, B. A., 360

Maize (Zea mays), 16-17, 348
Malva species, 398

Martin We Ba 357;

Mather, K., 350

McClellan, L. N., 249, 254
McKeekan, C. P., 363
Medicago hispida, 353-54
Medicago sativa, 338
Medicago species, 352, 398
Mediterranean area, 28, 31, 179-

82
locust areas, 384
Meigs, P., 4, 74

Meliotus species, 338

Melica californica, 346
Mesoamerica, 246-47
Mesophytes, 7-9

Meteoritic dust, 295-99, 316, 429

450

Meteorology, needed research, 142,
TS PCV, AKO), WO, Bris = 116),
428-30

Mexico, 45

industrial raw material, 414-15

population-supporting capacity,
247, 421

precipitation variations, 143-55

Meyer, A. F., 81

Michelson, A. T., 133

Microclimates, 186, 431

Mid-Century Conference on Re-

sources, 314

Middle East, 42, 45, 51, 179-82

Millemale ence

Missouri River, 191

water deficiency, 79

Mordy, W. A., 301

Moricandia nitens, 338

Muhlenbergia pungens, 5

Muhlenbergia species, 239

Murphy, A. H., 356, 360

National Council for Applied Re-
search, Netherlands, 260, 274—

National Resources Board, 279
Navajo Indians, 22
Netherlands, the, saline research,
27/3
Neumann, J., 316
New Mexico, 5
arroyo cutting, 115-18
ice nuclei, 302-4
introduced grasses, 238
land use and misuse, 229, 236-37
precipitation, 156, 235
water need and yield, 75, 81,
156, 235
water quality, 284-86
Nigeria, 61
Nile River, 18, 47, 49, 54
Nomadism, 61, 86, 183-84, 381-82

INDEX

Oasis economy, I10, 326-27

Olinsted? GE eu7;

Organization for European Eco-
nomic Cooperation, Working
Party, Nos 8; 27gsy277eeag2

Oryzopsis holciformis, 398

Oryzopsis miliacea, 338, 346, 353,
395, 398

Pakistan, 51

Thal development, 54-56
Panicum antidotale, 398
Panicum turgidum, 398
Papago Indians, 16
Parish, 55 Bs, 35

Parthentum argentatum, 22, 414,

415
Paulsen, C. G., 279
Pedilanthus species, 414
Peganum harmala, 399
Periploca aphylla, 399
Phalaris bulbosa, 396, 398
Phalaris tuberosa, 336, 338, 346,

3475 359
Phillips, R. W., 335
Phoenix dactylifera, 409
Phreatophytes, 52, 198
Physiology, needed research, 22-
24, 39; 397, 431
range animals, 333-34
relation to drought, 344-47
Phytogeography, 392
Phytosociology, 97, 392
Pima Indians, 16
Pinus edulis, 164
Pinus ponderosa, 163
Pistacia species, 400
Plantago, 415-16
Plantago albicans, 338
Plant, adaptation, 343-67
adjustments, 7-12, 23
breeding, 39, 416-18
distribution, 4-5, 37-38, 383-84
indicators, 97, 183, 392

INDEX

industrial and pharmaceutical,
22, 399, 414-18, 432
introduction, 60
salt-tolerant, 223, 289, 409
selection, 38, 60, 351-57, 397-
402, 415, 424-25
spread of alien, 351-52
Plant ecology, needed research, 431
Plant physiology, needed research,
431
Poa scabrella, 345
Poa species, 346, 349
Pollen analysis, 166-67
Polygonum species, 338
Population-supporting capacity,
27, 48-49, 214-18, 252, 362-
63, 380-82, 386-87, 391, 419-
23, 426
Poterium varicosum, 338
Powell, J. W., 18
eowellenS. ll 5) 535.280
Prairie oe 22
Preciozi, P. C., 95
Precipitation, estimation, Ci, UGG
orographic, 304
prediction, 135
variability, 77-79, 86-94, 143-
54, 162-70, 295-97
Precipitation, induced, 30, 50-s1,
Day AQUGDs, JOOS
GORE, Qi, Be
{CemUuICleIQOO_4, {us — 10)
prospects for, 299, 311, 318
silver iodide seeding, 291-95,
385-9
Preservation of habitat, 432
Primitive man, 15-17
Project Cirrus, 302-4, 306-8
Prosopis juliflora, 238, 338
Prosopis species, 164 ©
Psoralea lanceolata, 5
Public information, 434

45]

Quality of water (see Water)
Oureshrm INA AL 292

Range improvement, 22, 38, 41-42,
160, 179-88, 182-88, 360-63,
BO), AGUS =I)

centers, 184

costs, 361

fertilization, 357, 359

needed research, 357-60, 431

regeneration, §9, 185-87, 425

relation to irrigation, 43, 361

social factors, 42, 181, 188, 338,
340, 380-82

water supply, 184

Redfieldia flexuosa, 5

Retama species, 400

Research (see also Cooperative

research; _— Interdisciplinary
programs)
public support for, 434
Reservoir evaporation, 131-32,
196-97

Reservoir silting, 32, 173, 175, 197
IRhoadiAN@ x 334
Richards, es A. 5 PI DIS, No}
Rio Grande, 174, 235

salinity, 284-86'"
Robbins, W. W.,; 351, 352
Rockefeller Foundation, 273, 368
Rodents, 15, 23, 371
Roman settlement, 52
Rumex hymenosepalus, 414-418

SHIMarA, LL, 16
animal range, 378
climatic stability, 125
Saline Water Conservation Pro-
BEM, 57, WIA6O, WX, AOA,
266, 269, 272, 277, 435
Salerolerant plants, 223, 289, 409
Sand dune fixation, 401
Sansome, F. W., 348
Samusie, (C. Oe, 1s

452

Saville, T., 268

Schaefer, V. J., 316

Schmidt-Nielsen, K., and _ B.

Schmidt-Nielsen, 13, 14

Scirpus holoschoenus, 401

Scofield, C. S., 56, 283

Sediment movement, 103-4, 116-

L/S

Seed germination, 394-97, 408

Seely, B. K., 292

Semple Ace 3877

Sewage effluent, 269, 325

Shantz, Fineye92

Shaw, S. H., 130

Sheep, grazing area, 378-80

Short, L. R., 186

Shotton, F. W., 123, 128, 129

Siberia, grasshopper area, 384

Silver iodide (see Precipitation-

induced)

Simmondsta chinensts, 415

Smiths Es) 202

Social sciences, needed research,

42, 43, 45, 198, 433

Socorro conference, 427-35

Soil, 6, 34°37) 58-59, 403
conservation program, 230-231
crust problem, 408
irrigated, 286-288
AMONG IRS, O, J=WO, AO, 75 Jip
needed research, 430-31
surveys, 36, 41, 392

Soil Conservation Service, U. S.,

118, 1G6).2265, 2011

Solar energy, 62-63, 276

Sorghum vulgare, 359

- South America, 49 (see also Brazil;

Andean America; Mesoa-
merica)

Southwestern U.S., 11, 16, 21
climatic indices, 167-70
irrigation, 191
native plants for industry, 414-

INDEX

salinity, 224, 280-81
tree-ring studies, 162-65
use and misuse, 234, 242
Soviet Union, 35
Spanish explorers and _ settlers,
114, 351
Spilhaus, A., 314
Stapledon, R. O., 348
Stebbins, G. L., 348, 349
Stipa lagascae, 186, 338
Stipa species, 345, 346, 348, 355
Sumner, D. C., 356
Swinback, N. C., 316

Talbot, M. W., 352
Tamarix species, 400
Thal Valley development, 54-56
Thornthwaite, (Cl WeeaeGeemaa
95, 121, 141, 143, 154, 158
Tigris-Euphrates rivers, 47, 49
Vixeront, 2,515,525 122s
Tree-ring studies, 51, 162-65
Arizona, University of, 166
Tunisia, go-92
Trichostema lanceolatum, 356
Trifolium species, 354, 398
Trifolium hirtum, 60, 353-54
Trifolium incarnatum, 353
Trifolium subterraneum, 38, 60,
336, 353
Tritium content of water, 33-34
Triticum aestivum, 350
Tube wells, 56
Tunisia, 51-52, 85
agricultural planning, 86-87
archaeology, 98, 125
ecology, 182
precipitation and runoff, 89-103
recharge of aquifers, 134
salinity, 99
water balance and needs, 94-
g6, 105-11
Turesson, G., 347, 348

INDEX

UNESCO, Advisory Committee
on Arid Zone Research, 57,
Ti, NI, AC), Gaal, Los, ane,

432, 434
Uniform Mediterranean Nurseries,

MEd Kingdom, saline research,
273

United States (see also Arizona;
California; Colorado; South-
western U.S.; Utah; Wyo-
ming; Columbia River; Mis-
sourt River; Rio Grande)

climatic zones, 78

U.S. Bureau of Reclamation, 193-
196

WES: ee tment of Agriculture,
DVS), Ar, Bice Neon, ZLi(6

U.S. Department of the Interior,
195

U.S. Forest Service, 118, 362

U.S. Geological Survey, 56, 127

U.S. Salinity Laboratory, 282,
288

U.S. Weather Bureau, 310

Urbanization, 21, 266, 327

Utah, precipitation, 156

water need and yield, 81, 156

Vegetation (see also Plants), 4-5
mosaic distribution, 383-84
planting, 60, 239-40, 251

Vonnegut, B., 306, 308, 316

Water, appropriate use, 44, 159-
60, 190, 193, 266-67, 328
balance, 74-83, 94-95, 142
bookkeeping, 80

453

improved use, 194-98, 223
oxidation, 372-75
quality, 56-58, 98-99, 127-30,
221-25, 279-80
re-use, 195, 254, 324-25
spreading, 187, 402-08
storage, 86, 131, IgI-92, 208-12
supply in Southwest, 157
supply in U.S., 99-103, 190-91,
)
yield, ssp USS ESTO, EIS
Water demineralization, 23, 40,
57-58, 195-96, 257-71, 272-78
GOSS, AAO, Dy, DIG DFG,
BOT 28
processes, 261-63, 273-76
research needs and coordination,
ASCO, AQLI=JO, DID=W3,, AD
Watson, S., 352
Wavylandhtia |i. 7,
Weather Control, Advisory Com-
MMNCTS OM, QUG, FOU
Webster, D., 189
Well drilling, 56, 106-10, 136-37
Whaley WeiG 208033
Wind power, 40, 62
Woodcock, A., 301
Whyte; Re Or 60; 2793 a5
\vootedmtes IN; (Coy BRR, ARO
WhilcoxenIb Ve, 283
Working Party No. 8 (see Organi-
zation for European Eco-
nomic Cooperation)
World Health Organization, 49
World Meteorological Organiza-

HON; AO). SO, BUFy HOA
Wyoming, arroyo cutting, 118

Xerophytes, 7-11

vcr Oa
feria

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "The future of arid lands : papers and recommendations from the International Arid Lands Meetings"

See other formats