Front Cover
 Half Title
 Title Page
 Panel on promising technologies...
 Table of Contents
 Introduction and summary
 Part I. Water supply
 Part II. Water conservation
 Summary in Arabic
 Resume en Francais
 Resumen en Espanol
 Advisory committee on technology...
 Board on science and technology...
 Back Cover

Title: More water for arid lands
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00082050/00001
 Material Information
Title: More water for arid lands promising technologies and research opportunities : report of an ad hoc panel of the Advisory Committee on Technology Innovation, Board on Science and Technology for International Development, Commission on International Relations
Physical Description: vii, 153 p. : ; 23 cm.
Language: English
Creator: National Academy of Sciences (U.S.) -- Advisory Committee on Technological Innovation
United States -- Agency for International Development. -- Office of Science and Technology
Publisher: National Academy of Sciences,
National Academy of Sciences
Place of Publication: Washington
Publication Date: 1974
Copyright Date: 1974
Subject: Water-supply   ( lcsh )
Water conservation   ( lcsh )
Water-supply, Agricultural   ( lcsh )
Arid regions   ( lcsh )
Water Supply   ( mesh )
Irrigacao   ( larpcal )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Bibliography: Includes bibliographical references.
General Note: Introduction and summary in English, Arabic, French, and Spanish.
General Note: "Prepared ... for the Office of Science and Technology, Bureau for Technical Assistance, Agency for International Development, Washington, D.C."
 Record Information
Bibliographic ID: UF00082050
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 02522974
lccn - 74010058

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    Half Title
        Half Title
    Title Page
        Page i
        Page ii
        Page v
        Page vi
        Page vii
        Page viii
    Panel on promising technologies in arid-land water development
        Page iii
    Table of Contents
        Page ix
        Page x
        Page iv
    Introduction and summary
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
    Part I. Water supply
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
    Part II. Water conservation
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
    Summary in Arabic
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
    Resume en Francais
        Page 144
        Page 145
        Page 146
    Resumen en Espanol
        Page 147
        Page 148
        Page 149
    Advisory committee on technology innovation
        Page 150
        Page 151
    Board on science and technology for international development
        Page 152
        Page 153
        Page 154
    Back Cover
        Page 155
        Page 156
Full Text

9 More
? Water
i, for
f Arid
Lands 4
Promising Technologies
Research Opportunities

The cover motif highlights arid and semiarid regions where this report may find
application. The map is based on a projection developed in 1973 by German geograpber
Arno Peters. It accurately presents the relative areas of the world's land masses.

Promising Technologies
Research Opportunities

Promising Technologies
Research Opportunities

Report of an Ad Hoc Panel of the
Advisory Committee on Technology Innovation
Board on Science and Technology for International Development
Commission on International Relations

Rdsum6 en Francais
Resumen en Espaiiol
National Academy of Sciences
Washington, D.C. 1974

This report has been prepared by an ad hoc advisory panel of the Board on Science and Tech-
nology for International Development, Commission on International Relations, National Re-
search Council, for the Office of Science and Technology, Bureau for Technical Assistance,
Agency for International Development, Washington, D.C., under Contract No. csd-2584.

NOTICE: The project which is the subject of this report was approved by the Governing Board
of the National Research Council, acting in behalf of the National Academy of Sciences. Such
approval reflects the Board's judgment that the project is of international importance and
appropriate with respect to both the purposes and resources of the National Research Council.
The members of the committee selected to undertake this project and prepare this report
were chosen for recognized scholarly competence and with due consideration for the balance of
disciplines appropriate to the project. Responsibility for the detailed aspects of this report rests
with that committee.
Each report issuing from a study committee of the National Research Council is reviewed by
an independent group of qualified individuals according to procedures established and mon-
itored by the Report Review Committee of the National Academy of Sciences. Distribution of
the report is approved, by the President of the Academy, upon satisfactory completion of the
review process.

A French-language version of this report is available free on request to
Office of Science and Technology
Development Support Bureau
Agency for International Development
Washington, D.C. 20523

Second Printing, February 1976
Third Printing, February 1978
Fourth Printing, May 1979
Fifth Printing, March 1981
Library of Congress Catalog Number 74-10058


Little known but promising technologies for the use and conservation of
scarce water supplies in arid areas are the subject of this report. Not a
technical handbook, it aims to draw the attention of agricultural and
community officials and researchers to opportunities for development
projects with probable high social value.
The technologies discussed should, at present, be seen as supplements to,
not substitutes for, standard large-scale water supply and management
methods. But many have immediate local value for small-scale water
development and conservation, especially in remote areas with intermittent
rainfall. With further research and adaptation, some of the technologies may
prove to be economically competitive with standard methods of increasing
the water supply or reducing the demand.
For the convenience of the busy reader, each technology is presented in a
separate chapter, and the material is arranged under these topics:

Stage of Development
Needed Research and Development
Selected Readings (a short list of reviews and general articles)
Contacts (a list of individuals or organizations the panelists know to be
involved in relevant research)
NOTE: Neither the Selected Readings nor the Contacts are meant to be

Several points deserve emphasis:
The panel considers that all the technologies in this report have proved
themselves within the individual settings described. When these technologies
are applied elsewhere, consideration should be given to unique local
conditions that may affect their success. Questions should be asked that
cannot all be accounted for in a general report.

The particular choice of technologies examined in the report is not meant
to reflect on others, which may be equally worthy of attention. Selection was
based on technical merit and potential for application, particularly in
developing countries, as seen by the panel. No order of importance is implied
by the chapter sequence. Some methods selected are ready for widespread
application; for others, the fundamental principles are still being developed.
Although most of the ideas discussed are not new, they have as yet had little
In its discussion of the technologies, the panel took heed of their
economic parameters but could not consider this subject in specific detail.
Attempts to estimate future cost in the vastly different economic and
ecological environments of the several dozen countries beset with the
problem of aridity would have bogged down the discussions, as would
consideration of political, institutional, and social factors. Accordingly, this
report confines itself to a technical overview, leaving to the reader the task of
weighing the technical prescriptions in the light of his country's resources and
The Ad Hoc Panel on Promising Technologies for Arid-Land Water
Development formulated this report at a meeting in Tucson, Arizona, in
October 1973. Each selected technology was evaluated and written up before
the meeting by an individual committee member, in collaboration with the
NAS staff; each paper was reviewed by the others, discussed during the
meeting, and modified according to the will of the panel as a whole. This
document, therefore, reflects a consensus.
The panel is indebted to Tresa Bass and Mary Jane Koob, who acted as
administrative secretaries for the meeting and for production of the report,
and to A. Richard Kassander, Jr., and Jack D. Johnson of the University of
Arizona for local arrangements in Tucson. The report draft was prepared for
publication by Jane Lecht, and the Arabic translation by Mohammed Sageer.
This project is part of an experiment to determine ways scientists and engi-
neers can make a more effective contribution to economic-development acti-
vities, particularly by translating recent research results into a usable form for
decision makers. If you wish to comment on this report and especially if you
find it useful in your work, please communicate with the staff officer, Dr.
Noel Vietmeyer, National Academy of Sciences- National Research Council,
2101 Constitution Avenue, JH 215, Washington, D.C. 20418, USA.

Two systems of ancient agriculture in the Negev-narrow terraced wadis and
farm units with small watersheds-show a most rational and wise use of the
available natural resources. The ancient farmer fitted his artificially created
agricultural ecosystems into nature and used landscape and topography to his
best advantage without damaging his environment. He neither caused erosion
nor brought about salination of his agricultural soils. By using the runoff he
tamed the flood torrents and prevented the damage that uncontrolled floods
usually produce. He certainly did not overirrigate, because his water resources
were limited, and in this case as in many others, limitation is the mother of
good management. The methods of the ancient civilizations of providing
drinking water are another example of a most rational use of nature's
resources. The same is true of the [qanat], which merits our special
admiration because of the great technical skill and ingenuity involved in its
construction. In all these cases man learned from his natural environment and
applied what he had observed by imitating nature and sometimes improving
on it. This is most obvious in the case of runoff agriculture. Most of the plant
associations of the natural desert ecosystems live on runoff water. A good
observer will notice this and may apply this knowledge to grow cultivated
plants to his own benefit.
Michael Evenari, Leslie Shanan, and Naphtali Tadmor.
The Negev: The Challenge of a Desert.

Panel on Promising Technologies in Arid-Land Water Development

DEAN F. PETERSON, Vice President for Research, Division of Research,
Utah State University, Logan, Utah, Chairman
FALIH K. ALJIBURY, Soil Physicist, Agricultural Extension, University of
California, Parlier, California
BAHE BILLY, Executive Manager, Navajo Tribal Agricultural Products
Industry, Farmington, New Mexico
C. BRENT CLUFF, Associate Hydrologist, Water Resources Research Center,
University of Arizona, Tucson, Arizona
HAROLD E. DREGNE, Chairman, Department of Agronomy, Texas Tech
University, Lubbock, Texas
EARL A. ERICKSON, Department of Crop and Soil Sciences, Michigan State
University, East Lansing, Michigan
MICHAEL EVENARI, Botany Department, Hebrew University of Jerusalem,
Jerusalem, Israel
JEROME GAVIS, Department of Geography and Environmental Engineering,
The Johns Hopkins University, Baltimore, Maryland
ROBERT M. HAGAN, Department of Water Science and Engineering,
University of California, Davis, California
JACK KELLER, Agricultural and Irrigation Engineering Department, Utah
State University, Logan, Utah
FERNANDO MEDELLIN LEAL, Director, Institute de Investigaci6n de
Zonas Desdrticas, Universidad Aut6noma de San Luis Potosi, S. L. P.,
LLOYD E. MYERS, U.S. Department of Agriculture, Agricultural Research
Service, Western Region, Berkeley, California
SOL D. RESNICK, Director, Water Resources Research Center, University of
Arizona, Tucson, Arizona
JAN VAN SCHILFGAARDE, Director, U.S. Salinity Laboratory, Soil and
Water Conservation Research Division, A.R.S., U.S. Department of
Agriculture, Riverside, California
NOEL D. VIETMEYER, Board on Science and Technology for International
Development, Commission on International Relations, National Academy
of Sciences-National Research Council, Staff Study Director

JULIEN ENGEL, Head, Special Studies, Board on Science and Technology
for International Development




1 Rainwater Harvesting 9
2 Runoff Agriculture 23
3 Irrigation with Saline Water 38
4 Reuse of Water 45
5 Wells 52
6 Other Sources of Water 64


7 Reducing Evaporation from Water Surfaces 73
8 Reducing Seepage Losses 81
9 Reducing Evaporation from Soil Surfaces 87
10 Trickle Irrigation 97
11 Other Innovative Irrigation Methods 105
12 Reducing Cropland Percolation Losses 111
13 Reducing Transpiration 116
14 Selecting and Managing Crops To Use Water
More Efficiently 123
15 Controlled-Environment Agriculture 128
16 Other Promising Water-Conservation Techniques 135




Advisory Committee on Technology Innovation 150
Board on Science and Technology for International
Development (BOSTID) 152
BOSTID Publications 152

LEON BERNSTEIN, U.S. Salinity Laboratory, A.R.S., U.S. Department of
Agriculture, Riverside, California.
KEITH R. COOLEY, U.S. Water Conservation Laboratory, A.R.S., U.S.
Department of Agriculture, Phoenix, Arizona.
ALBERT K. DOBRENZ, Department of Agronomy and Plant Genetics,
University of Arizona, Tucson, Arizona.
A. RICHARD KASSANDER, JR., Vice President for Research, University of
Arizona, Tucson, Arizona.
L. K. LEPLEY, Office of Arid Lands Studies, University of Arizona, Tucson,
MARTIN A. MASSENGALE, Department of Agronomy and Plant Genetics,
University of Arizona, Tucson, Arizona.
JAMES J. RILEY, Environmental Research Laboratory, University of
Arizona, Tucson, Arizona.
W. T. WELCHERT, College of Agriculture, Cooperative Extension Service,
University of Arizona, Tucson, Arizona.
L. G. WILSON, Water Resources Research Center, University of Arizona,
Tucson, Arizona.

Introduction and Summary

Arid regions today face more difficult problems than ever before. The
world's sand deserts appear to be enlarging, and droughts are contributing to
the economic devastation of whole nations. The six drought-stricken Sahelian
nations provide an extreme illustration, but industrialized and developing
countries both suffer from the crisis. The southwestern United States, for
example, faces falling water tables and increasing groundwater salinity.
Nevertheless, arid lands have underexploited agricultural potential. We
should learn that this potential can be best developed by concepts and
methods specifically suited to dry regions. Water practices developed for
temperate climates may not work as well in arid regions for technological,
environmental, economic, and cultural reasons. We need fresh, innovative
approaches to water technologies, particularly those designed to meet the
needs of arid regions in the less developed world, where there has often been
improper application of practices developed in regions with higher rainfall or
more abundant water supplies. Also, we need to reconsider practices
developed in arid regions by ancient agriculturalists. Basically there are two
approaches: increasing the supply of usable water and reducing the demand
for water. Supply and demand, as well as delivery, have to be considered as an
integral system.
There are many possibilities for simultaneously increasing supply and
reducing demand that together will bring benefits to arid lands. A major
opportunity to save water exists in conventional irrigated agriculture, by far
the arid world's largest user of water. This is true for centuries-old, traditional
systems as well as for large, capital-intensive modem water-management
systems. Conventional systems are not the main topic of this report, because
they are extensively treated elsewhere (Selected Readings, p. 5). Still the
following points need to be made:
In some arid lands the greatest opportunity for increasing water
supplies is to improve existing water systems and thus make more water
available without a complete new installation. For example, replacing canals
with closed conduits (of plastic, concrete, metal, etc.) will reduce evapora-
tion, or lining canals will reduce seepage losses. (Most chapters in this report


deal in one way or another with improvements to existing water systems and
the maximum utilization of existing water supplies.)
Significant amounts of water can be saved by improving water manage-
ment on the farm, a topic agriculturalists in many areas often neglect. One
deficiency is the design of on-farm distribution and drainage systems between
farm fields. Storage, main, and diversion canals and main drains may be well
engineered (even down to turnouts serving one or two hundred hectares), but
most ditches serving farm fields are inadequate, sometimes even nonexistent.
Furthermore, the irrigator often mismanages application of the water.
In designing new systems and rehabilitating old ones, the needs of the
user should be paramount. The system must deliver the right amount of water
to the user at the right time. Frequently, an irrigation project fails to reach its
potential because the use-requirements for the water have not been suf-
ficiently considered. For example, the delivery system in irrigation should be
designed to permit changing the water supply as crop demands change with
weather and plant maturity. But water is often delivered in an arbitrary and
inflexible manner.
Where groundwater is available, surface and groundwater supply and
delivery systems should be considered in combination (conjunctive use) for
optimal use of the total water resource.
Universally, farmers tend to overirrigate when water is available. This
can lead to problems of waterlogging and salinity, and leaching of fertility.
Frequently, institutional arrangements (systems of delivery, scheduling,
water-rights laws, traditions, etc.) encourage overirrigation. Although over-
irrigation may be needed to remove accumulating salts, recent studies indicate
that the amount required may be much lower than was formerly believed
(chapter 3).
Conventional irrigation is neither cheap nor simple; complexities in the
design, construction, and efficient operation of standard irrigation projects
are frequently oversimplified or overlooked. Fields are often inadequately
leveled, and even small undulations can waste large amounts of water. Pre-
cision land-shaping and skilled labor are required. Grading land to a flat
surface usually requires high capital costs for equipment, fuel, and mainte-
The scarcer the water, the greater the need for technical and manage-
ment skills.

Chapters 1-6 of this report deal with technologies for enhancing water
supplies; the rest cover water conservation. A summary follows of the tech-
nologies in this report.


Rainwater Harvesting
Rainwater collected from hillslopes and man-made catchments can create
new supplies of low-cost, high-quality water for arid lands (chapter 1).

Runoff Agriculture
Runoff agriculture involves rainwater harvesting; the water is used directly
in agricultural systems specifically designed for the purpose (chapter 2).

Irrigation with Saline Water
Saline water is widely available but rarely used because it restricts plant
growth and yield. Evidence is now accumulating that with care and under
certain favorable conditions, saline water can be profitably used for irrigation
(chapter 3).

Reuse of Water
Increasing demands on water make it necessary to greatly increase water
reuse. Technical developments such as recycling an4 advanced waste treat-
ment may have great importance in the future (chapter 4).

Hand-dug wells, a technology begun thousands of years ago, is regaining
popularity with the help of new materials and construction equipment.
Qanats and horizontal wells, methods for tapping underground water without
using pumps, are also described (chapter 5).

Other Sources of Water

This chapter briefly mentions groundwater mining, desalting, solar distilla-
tion, the use of satellites and aircraft for detecting water in arid lands, rainfall
augmentation, the possibility of using icebergs as a source of water, and dew
and fog harvesting (chapter 6).

Reducing Evaporation from Water Surfaces
Because evaporation is invisible, it is seldom regarded as a serious drain on
stored water, but annual evaporation losses, particularly in arid lands, are very
great. Evaporation reduction merits increased attention as a way to conserve
water (chapter 7).


Reducing Seepage Losses
Seepage causes serious water losses in canals and impoundments. Modem
materials and techniques can reduce or eliminate seepage, but costs are still
high (chapter 8).

Reducing Evaporation from Soil Surfaces
Water losses resulting from evaporation from soil surfaces can be reduced
by covers or mulches. In many cases the covers also serve complementary
functions such as stopping desert encroachment or promoting runoff agri-
culture (chapter 9).

Trickle Irrigation
This newly developed irrigation method uses a system of plastic pipes
placed on the soil among the plants. Water carried in the pipes drips onto the
soil beside each plant at a rate carefully matched to the plant's needs. Com-
pared with conventional irrigation, excellent crop yields have been obtained
with a minimum amount of water (chapter 10).

Other Innovative Irrigation Methods
Some simple irrigation methods, neglected in technical manuals or text-
books, with potential benefit for arid lands are presented pictorially (chapter

Reducing Cropland Percolation Losses
Large areas of sandy soil in arid lands are not used for agriculture because
the water sinks below the root zone too rapidly and the extra irrigation water
needed to compensate for this problem is not available. Techniques are now
being developed to produce artificial underground moisture barriers to pre-
vent or restrict water and nutrients from percolating away (chapter 12).

Reducing Transpiration
About 99 percent of the water absorbed by plant roots is released into the
air from leaf surfaces. If practical means to reduce this process can be found,
major savings can be realized in the amount of water needed to raise a given
crop (chapter 13).


Selecting and Managing Crops To Use Water More Efficiently

Relatively little has been done on designing water-efficient systems for
arid-land agriculture. Numerous research opportunities from plant genetics to
engineering remain to be explored (chapter 14).

Controlled-Environment Agriculture

When crops are grown within watertight but transparent enclosures, the
amount of water normally lost can be greatly reduced, and the atmosphere
around the plants can be manipulated to maximize productivity. These are
costly systems, but high agricultural productivity can be achieved with small
amounts of water in very inhospitable regions (chapter 15).

Other Promising Water-Conservation Techniques

This chapter briefly mentions water-conserving soil amendments and artifi-
cial recharge of groundwater. (chapter 16).

Selected Readings
Arid Zone Research Liaison Officer, Commonwealth Scientific and Industrial Research
Organization. Annual. Arid Zone Newsletter. (Booklet provides representative cross-
section of arid-zone research in progress in Australia. Available from the authoring
institution, Canberra, ACT, Australia.)
Amon, I. 1972. Crop Production in Dry Regions. Volume 1: Background and Principles.
Barnes and Noble, New York; Leonard Hill Books, London. 650 p.
Bateman, G. H. 1971. A Bibliography of Low-Cost Water Technologies. Intermediate
Technology Development Group, Ltd. (Parnell House, 25 Wilton Road, London
SW1 V 1S, England, Price 1.00). 45 p.
Dixey, F. 1950. A Practical Handbook of Water Supply. 2nd ed. T. Murby, London.
International Hydrological Decade. 1969. The Progress of Hydrology. Vol. 1, New De-
velopments in Hydrology, Vol. 2, Specialized Hydrologic Subjects, VoL 3, Hydro-
logic Education and Discussions. Urbana, Illinois, USA. 1295 p. (Available from V. T.
Chow, Civil Engineering Building, Room 3118, University of Illinois, Urbana, Illinois
61801, USA.)
Israelsen, O. W., and V. E. Hansen. 1962. Irrigation Principles and Practices 3rd ed. John
Wiley and Sons, New York. 447 p.
Kaul, R. W., ed. 1970. Afforestation in Arid Zones. Dr. W. Junk N.V. Publishers, The
Hague, The Netherlands. 435 p.
National Water Commission. 1973. Water Policies for the Future. Stock No.
5248-00006. Superintendent of Documents, U.S. Government Printing Office, Wash-
ington, D.C. 20402, USA. 579 p. (US$9.30).
Office of Agriculture, Technical Assistance Bureau, U.S. Agency for International
Development. 1973. Improving Farm Production in Tropical and Subtropical Regions
of Limited Rainfall. Technical Series Bulletin Number 4. 25 p. (Available from
authoring institution, Washington, D.C. 20523, USA.)


Office of Science and Technology, Technical Assistance Bureau, U.S. Agency for Inter-
national Development. 1972. Desert Encroachment on Amble Lands: Significance,
Causes and Control. 55 p. (Available from authoring institution, Washington, D.C.
20523, USA.)
Office of Science and Technology, Technical Assistance Bureau, U.S. Agency for Inter-
national Development. 1973. Techniques for Assessing Hydrological Potentials in
Developing Countries: State of the Art and Research Priorities. Report No. TA/OST
73-17. (Available from authoring institution, Washington, D.C. 20523, USA.)
Paylore, P. 1967. Arid Lands Research Institutions: A World Directory. University of
Arizona Press, Tucson, Arizona 85721, USA.
Peterson, D. F. 1973. Irrigation Practices Seminars 1956-70: an Evaluation. Asia Bureau,
U.S. Agency for International Development, Washington, D.C. 20523, USA.
Rhodesia Agricultural Journal. 1972. Water in Agriculture. Technical Bulletin No. 15.
197 p. (Available from Department of Research and Specialist Services, Ministry of
Agriculture, P.O. Box 8108, Causeway, Salisbury, Rhodesia.)
Water for Peace: Proceedings of the International Conference on Water for Peace,
Washington, D.C. 1967. 8 Vol. Stock No. I-.2; W29-5/V1-8, Superintendent of
Documents, U.S. Government Printing Office, Washington, D.C. 20402, USA.
Water Research Foundation of Australia. 1968. Water on the Farm. Report No. 25. 59 p.
(Available from authoring institution, P.O. Box 47, Kingsford, New South Wales,
2032, Australia.)
White, G. F., ed. 1956. The Future of Arid Lands. Publication No. 43. American
Association for the Advancement of Science, Washington, D.C., USA. 453 p.

Part I

Water Supply

1 Rainwater Harvesting

Though rain falls infrequently in arid lands, it comprises considerable
amounts of water; 10 mm of rain equals 100,000 1 of water per ha. Harvesting
this rainwater (FIGURE 1) can provide water for regions where other sources
are too distant or too costly, or where wells are impractical because of
unfavorable geology or excessive drilling costs. Rainwater harvesting is
particularly suited to supplying water for small villages, schools, households,
small gardens, livestock, and wildlife.
Ancient desert dwellers harvested rain by redirecting the water running
down hillslopes into fields or cisterns (FIGURES 2-4). Modern farmers in
arid lands have seldom harvested rainwater in this direct way, though in 1929
a 2400 m2 catchment in an arid part of Australia (300 mm average annual
rainfall) provided adequate water for "6 persons, 10 horses, 2 cows, 150
sheep" even during the years of lowest rainfall.'
Today, researchers are working to increase water runoff by modifying the
surface of the soil.


Rainwater harvesting is possible in areas with as little as 50-80 mm average
annual rainfall. This seems to be the lowest limit, but during a year with only
24 mm of rain, a water-harvesting catchment in Israel still yielded a usable
runoff.2 In some arid regions, such as the southwestern United States, snow
and sleet also contribute to runoff water. Loess and loess-like soils, present in
most deserts, are ideally suited for rainwater harvesting because after even a
small rainfall they form a crust that promotes runoff.
Sometimes rainfall runoff can be collected from an untouched natural
catchment; one way is to dig ponds in small depressions where they can
collect runoff (for example, the pond in FIGURE 48, page 84). Often the
catchment needs modification, usually by making the soil surface more

1Kenyon. 1929. (See Selected Readings.)
2Evenari, Shanan, and Tadmor. 1971. p. 325. (See Selected Readings.)


FIGURE 1 The yield from rainwater harvesting can be surprising. Here in the Negev
Desert runoff collected from the background hills is channeled to the farm site and
distributed to separate fields. (N. Tadmor)

impermeable to increase the amount of runoff. There are many methods,
including the following:

Land Alteration

In some cases all that is needed to collect and convey runoff water are
ditches or rock walls along hillside contours (FIGURES 2 and 5). Clearing
away rocks and vegetation usually increases runoff water (FIGURES 3 and
4); compacting the soil surface can increase it, too (FIGURES 6 and 7).


FIGURE 2 Two thousand years ago Nabatean inhabitants of the Negev Desert built
this channel across a hillside to harvest rainwater runoff. The channel leads water to a
cistern at the right. A similar channel, disappearing to the right, drains the slopes on the
other side of the hill. (L. Evenari)

FIGURE 3 Ancient rainwater-harvesting system in the Negev. Gravel that covered the
slopes has been moved aside, leaving a catchment that directs rainfall runoff to a farm in
the valley. (M. Evenari)


FIGURE 4 Aerial view of ancient gravel mounds and strips still used to increase the
rainwater harvest from hillsides in the Negev. Unable to remove all the gravel, the
Nabateans mounded it (shown by dots). Conduits and channels are organized so that
each drains a small catchment. This system divides overall runoff into small streams,
thereby avoiding erosion and presenting the farmer with small, easy-to-handle flows.
(N. Tadmor)

With these simple systems erosion is the main problem; when it is not
excessive and low-cost hillside land is available, these land alterations can be
very economical ways to harvest rainwater in arid lands.

Chemical Treatment

A promising method for harvesting rainwater is to treat soils with
chemicals that fill pores or make soil water repellent.
Sodium salts, which cause clay in the soil to break down into small
particles (partially sealing the soil pores and cracks), can be used to increase


FIGURE 5 Harvesting runoff from rocky hill slopes. Total capacity of the two tanks is
110 m'; 75 mm of rainfall fills both tanks. Rowa African, Purchase Land, Rhodesia.
(Rhodesia Agricultural Journal)

FIGURE 6 Western Australia rainwater-harvesting system. Catchments are graded and
rolled and shed water with a minimum rainfall of 7.6 mm. They cost $30-$40 per acre
(1968). They are designed so that for only 4.45 cm of runoff 1.6 ha of catchment will
provide 800 cubic m of water. Catchments are cambered so that rainfall runoff quickly
goes to the side of the "road," where a ditch conveys it to the main-collector drain and
thence through a silt trap to the storage tank. (Taken from Department of Agriculture of
Western Australia, 1950. See Selected Readings.)


runoff from many clay-containing soils (chapter 8). Sodium salts are intrigu-
ing as soil sealants because of their low cost, ready availability, and retarda-
tion of weed growth (FIGURE 24, page 33).
Other commonly tried water-repellent chemicals are silicones, latexes,
asphalt, and wax (chapter 9). Although much more research remains to be
done, these sealants now appear feasible for use on stable soils that do not
swell with moisture.
Asphalt offers promise for building low-cost, impermeable catchments,
particularly because it can be easily applied by spraying. In the United States
hillslope catchments have been cleared of vegetation, smoothed, treated with
a soil sterilant and two coats of asphalt to make rainwater catchments. One
coat seals the pores; the other protects against weathering. Asphalt catch-
ments on suitable slopes have been found to last 4 or 5 years. Problems
caused by unstable soil conditions, oxidation, and penetration by germinating
plants have recently been overcome by reinforcing the asphalt with plastic or
fiberglass and covering the catchment with gravel.

FIGURE 7 A modem rainwater-harvesting catchment south of the Stirling Range in
Western Australia. The average annual rainfall in this area is 500 mm, falling during seven
winter months. Natural ground surface is sandy with a clay subsoil. The sand is moved
into rows; the clay exposed by the reading process is shaped and spread to cover the
whole surface. The ridges discharge runoff water into a channel which conducts it to the
square tank (capacity 3000 m3). Main advantages are that the system uses the existing
soil and can be built with readily available equipment. Smoothing and compacting the
steep "road" surfaces is most important, here achieved by tractor and rubber-tired roller.
The system appears suitable for many other arid and semiarid areas. (D. J. Carder and M.


Paraffin wax has recently been used as a soil-sealant (FIGURE 8).
Granulated wax spread on the ground melts in the sun and flows into the
pores to produce a surface that readily sheds water. The wax can also be
melted and sprayed on the ground. In tests,3 wax-treated plots yielded an
average of 90 percent of the rainfall as runoff, compared to 30 percent from
untreated plots. Runoff water from the wax plots had low salt content (less
than 50 mg per 1) and almost undetectable organic matter.

Soil Covers
Instead of making the soil itself the water-shedding surface, it may be
better in some situations to cover it with a waterproof cover. On porous or
unstable soils in particular, maintaining other methods would be too costly.
Plastic sheet, butyl rubber, and metal foil offer opportunities for building
low-cost rainfall catchments, but they are easily damaged by wind. Plastic
films covered with gravel (FIGURE 9) have proved more successful; the gravel
protects the underlying membrane against radiation and wind damage. These
catchments, if properly constructed and maintained, can be durable, with a
projected life of more than 20 years. They are useful where gravel is readily
available and maximum runoff is not required (gravel retains some of the

General Principles
Water-harvesting methods are site specific. Before a system is installed, one
must know:
the soil (especially characteristics of water-holding, runoff, and
the topography (slope and the direction followed by natural runoff);
the precipitation characteristics (amount, reliability, etc.); and
the climate (wind, sunlight, temperature, etc.).
Because rainfall in arid lands is intermittent, storage must usually be an
integral part of any rainwater-harvesting system. However, when water-
harvesting techniques are used for runoff farming (chapter 2), the water is
"stored" in the cultivated soil itself. Sometimes it is possible to build catch-
ments to feed existing-even ancient-water-storage structures (FIGURE 2).


Surprisingly, the livestock-carrying capacity of many arid rangelands is
limited more by a lack of drinking water than by a lack of feed. Rainwater
3Fink, et al. 1973. (See Selected Readings.)


FIGURE 8 Paraffin wax is shredded and spread for rainwater harvesting in Arizona,
USA. Melted by the sun, the wax flows into and seals soil pores, making a
water-repellent surface that sheds water efficiently. (U.S. Department of Agriculture)


40. 'M

FIGURE 9 Laying polyethylene sheet for a rainwater catchment in the Sonoran
Desert, Arizona, USA. Sheet was later covered with gravel for protection against
damage and sunlight. (C.B. Cluff)

harvesting may be the only source of extra water. Improving drinking-water
supplies on arid rangelands or other remote watershed areas increases the
value of these grazing lands and allows the available feed to be used more
A water-harvesting system, once installed, will provide water without
requiring fuel or power.
Recently, catchment construction costs have fallen sharply, and further
cost reductions seem possible. The more promising chemical treatments and
soil covers (such as wax, reinforced-asphalt membranes, and gravelled plastic)
provide sediment-free, high-quality water for less than U.S.$0.05 per m3
in a 300-mm rainfall zone in the southwestern United States.4 Under favor-
able conditions the land alteration methods are the least expensive of all and
could provide water suitable for most agricultural purposes at much lower
Arid developing countries that produce and refine crude oil could use
asphalt to construct water-harvesting catchments. Heavy petroleum fractions
such as asphalt have limited demand and are often persistent pollutants,
difficult to dispose of.
4Cluff, et al. 1972. (See Selected Readings.)



Because rainwater harvesting depends on natural rainfall, it is no more
reliable than the weather. Without adequate storage, the system will fail in
drought years. In locations with an average annual rainfall of less than 50 to
80 mm, rainwater harvesting will probably never be economically feasible.
In applying water-harvesting methods to a given area, care is needed to
minimize side effects. Poorly designed and managed rainwater harvesting can
lead to soil erosion, soil instability, and local floods. Soil erosion, a constant
concern, can be controlled if the slope is short and not too steep (and if
drains are suitably sloped). Slope also affects the quantity and quality of
runoff. The most efficient water harvest is from small, gently sloping (prefer-
ably 1 to 5 percent) catchments.
Today, little expertise is available for designing rainwater-harvesting
systems. Furthermore, in many arid areas data are lacking on rainfall intensity
and variability.
A rainwater-harvesting catchment must withstand weathering and occa-
sional foot traffic. Fencing may sometimes be required. Environmental
contamination must be constantly considered. Colored or contaminated run-
off water will require treatment before it can be used for human consump-
tion. (A simple system that uses a sand filter is shown in FIGURES 10 and
Most soil treatments (especially the cheaper ones) have a limited lifetime
and must be renewed periodically. They also require occasional maintenance
because of cracking caused by unstable soils, oxidation, and plants growing
up through the ground cover or treated soil. No one material has proved
superior for all catchment sites.

Stage of Development

Rainfall harvesting is almost 4,000 years old: it began in the Bronze Age,
when desert dwellers smoothed hillsides to increase rainwater runoff and built
ditches to collect the water and convey it to lower lying fields (FIGURES 2 -
4). This practice permitted agricultural civilizations to develop in regions with
an average rainfall of about 100 mm, an inadequate rainfall for conventional
modern agriculture.
In modern times, but before 1950, only a few artificial catchments were
built, mainly by government agencies to collect water for livestock and wild-
life on islands with high rainfall and porous soils (the Caribbean island of
Antigua, for example5). The cost was usually high. In the 1950s interest in

5Bateman. 1971. p. 11 (See p. 5.)





FIGURE 10 An experimental rainwater catchment with a sand-filled water-storage
tank. The sand reduces evaporation and filters the water as it enters and is withdrawn,
making water suitable for drinking. Tank is lined with thin plastic; storage capacity is
increased by constructing beehive-shaped cells out of stacks of plastic sausages, described
in FIGURE 12. (Intermediate Technology Development Group)

rainwater harvesting increased, and some lower cost treatments were installed.
One of the most extensive is in Western Australia, where several thousand
hectares of shaped, compacted earth catchments supply water for both house-
holds and livestock.6 (See FIGURES 6 and 7.) Their performance is good
when they are properly maintained. Approximately 240 ha of asphalt or
asphaltic-concrete catchments have also been constructed to furnish water for
32 small towns in Western Australia.7
Currently, rainwater harvesting is for small-scale use, for farms, villages,
and livestock. The land-alteration method is ready for immediate worldwide
use. Australia and Israel use it already; in the Sudan and Botswana (FIGURE
12) catchment tanks have been introduced in technical assistance programs.
Chemical treatments and ground covers, though still mainly experimental,
are used on a worldwide, but slight, scale. Proved to be technically feasible
and successful, they are not yet economically attractive enough to generate
widespread adoption.

Needed Research and Development

No method of rainwater harvesting has been subjected to a long-term
economic analysis. Large field trials in different areas are needed to build up a
data base that could lead to a better understanding of the economic viability
of different methods in different economic environments. Developing
countries particularly need the data, because most of the technology was
designed for Israel, Australia, or the United States. With adaptive research to
fit the needs, economics, and materials of developing countries, rainwater-
harvesting methods may be of exceptional and immediate value.
6Carder. 1970. (See Selected Readings.)
7Kellsall. 1962. (See Selected Readings.)


FIGURE 11 Capping 2-m-tall internal water-storage cells with soil-cement-filled
sausages (FIGURE 12). The area will next be back-filled with sand. This technology re-
quires only a spade, funnel, carving knife, and mallet. Kordofan Province, Sudan.
(M. G. Ionides)

FIGURE 12 Catchment tank at Palapye Central School, Botswana, paved with mud
and flat stones and sealed with a plastic liner. Revetment is made of
"sausages"-thin-plastic tubes filled with soil-cement. Tubes are sealed at one end, and
soil containing a small amount of cement is poured in. Tubes are then pricked with nails
and placed in a shallow pan of water. Before the cement sets the tubes are stacked in
place. No formwork is needed; the soil-cement is self-curing. (See also chapter 9.) The
sausages are a modem, cheap, do-it-yourself technology. Water-storage tanks or cisterns
with a catchment alongside to run the rainfall in are ancient devices, often forgotten
today. (Intermediate Technology Development Group)


The major technical-research need is to reduce the costs of sealing catch-
ment soils and to make the treatment practical for a wider variety of soils and
situations. Industry is continually formulating new materials which should be
continually monitored and evaluated for use in rainwater harvesting.

Selected Readings
(See also Selected Readings, chapter 2.)
Carder, D. J., and G. W. Spencer. 1971. Water Conservation Handbook. Soil Conserva-
tion Service, Department of Agriculture (Jarrah Road, South Perth, Western Australia
6151, Australia).
Chinn, S. S. W. 1965. Water Supply Potential from an Asphalt-lined Catchment near
Holualoa, Kona, Hawaii. Geological Survey Water Supply Paper 1809-P. U.S. Govern-
ment Printing Office, Washington, D.C., USA.
Cluff, C. B. 1971. Plastic catchments for economical harvesting of rainfall. In Proceed-
ings of the Tenth National Agricultural Plastics Conference, ed., J. W. Courter. (Avail-
able from J. W. Courter, National Agriculture Plastics Association, Dixon Springs
Agriculture Center, Simpson, Illinois 62985, USA.) pp. 192-202.
Cluff, C. B. 1974. Plastic reinforced asphalt membranes for precipitation harvesting and
seepage control. In Proceedings of the Eleventh National Agricultural Plastics Confer-
ence, ed., J. W. Courter. (In press.) (Available from J. W. Courter; see Cluff, 1971.)
Cluff, C. B.; G. R. Dutt; P. R. Ogden; and J. K. Kuykendall. 1972. Development of
Economic Water Harvesting Systems for Increasing Water Supply, Phase II. Office of
Water Resources Research Project No. B-015-ARIZ. (Available as Report No. PB-214
128, National Technical Information Service, U.S. Department of Commerce, Spring-
field, Virginia 22151, USA. US$4.50.)
Department of Agriculture of Western Australia. 1950. Roaded Catchments for Farm
Supplies. Bulletin No. 2393. (Perth, Western Australia.)
Evenari, M.; L. Shanan; and N. Tadmor. 1971. The Negev: The Challenge of a Desert.
Harvard University Press, Cambridge, Massachusetts 02138, USA.
Fink, D. W.; K. R. Cooley; and G. W. Frasier. 1973. Wax treated soils for harvesting
water. Journal of Range Management 26:396-8.
Frasier, G. W.; L. E. Myers; and J. R. Griggs. 1970. Installation of Asphalt-Fiberglass
Linings for Reservoirs and Catchments. U.S. Water Conservation Laboratory. Report
No. 8 (Agricultural Research Service, U.S. Department of Agriculture, Phoenix,
Arizona 85040, USA).
Intermediate Technology Development Group, Ltd. 1969. The Introduction of Rain-
water Catchment Tanks and Micro-irrigation to Botswana. (Parnell House, 25 Wilton
Road, London SWIV IJS, England.) 74 p.
Jefferson, J. H. K. 1953. A Note Based on Field Experience in Planning Hafir Excavation
Programmes. Memoirs of Field Division No. 4, Agricultural Publication Committee
(Ministry of Agriculture, Khartoum, Sudan).
Kellsall, K. J. 1962. Construction of Bituminous Surfaces for Water Supply Catchment
Areas in Western Australia. Hydraulic Engineers Branch, Public Works Department
(State Government of Western Australia, Perth, Western Australia.) (Mimeographed
notes for field staff.)
Kenyon, A. S. 1929. The ironclad or artificial catchment. Journal of the Department of
Agriculture of Victoria 27:86-91.
Myers, L. E., and G. W. Frasier. 1969. Creating hydrophobic soil for water harvesting.
Journal of the Irrigation and Drainage Division, Proceedings of the American Society
of Civil Engineers 95, Number IR1: 43-54.
Myers, L. E.; G. W. Frasier; and J. R. Griggs. 1967. Sprayed asphalt pavements for water
harvesting. Journal of the Irrigation and Drainage Division, Proceedings of the
American Society of Civil Engineers 93, Number IR3:79-97.


Robertson, A. C. 1950. The Hafir: What-Why-Where-How. Bulletin No. 1. Agricul-
tural Publication Committee (Ministry of Agriculture, Khartoum, Sudan).
Tadmor, N. H., and L. Shanan. 1969. Runoff inducement in an arid region by removal of
vegetation. Soil Science Society of America Proceedings 33:790-94.
Velasco Molina, H. A., and O. Aguirre Luna. 1972. Una estimaci6n del costo de captar y
almacenar agua de lluvia en regions deserticas y semi-deserticas del norte de Mexico.
(An estimation of the cost of catching and storing of rainwater in arid and semiarid
regions of North Mexico.) Agronomia, Mexico 145:74-9.

(See also Contacts, chapter 2.)
Animal Husbandry Farms, University of Sydney, Camden, New South Wales, 2570,
Australia (H. J. Geddes)
Botany Department, Hebrew University of Jerusalem, Jerusalem, Israel (M. Evenari, L.
Departamento de Suelos e Ingenieria Agronoma, Instituto Tecnologico de Estudios
Superiores, Monterrey, Mexico
Department of Soil and Water Science, Faculty of Agriculture, Hebrew University of
Jerusalem, Rehovoth, Israel (D. Hillel)
District Soil Conservationist, Western Soil Conservation District, Soil Conservation
Service of New South Wales, P.O. Box 118, Condobolin, N.S.W. 2877 Australia
Doxiadis lonides Associates of London, Ockham Mill, Ripley, Surrey, U. K.
Hydraulic Engineers Branch, State Public Works Department, Perth, Western Australia
Irrigation and Water Supply Commission, P.O. Box 74, North Quay, Brisbane, Queens-
land 4000, Australia (A. M. Carmichael)
Ministry of Agriculture, Khartoum, Sudan
Northam District Office, Department of Agriculture of Western Australia, Northam,
Western Australia (D. J. Carder)
Soils Division, Department of Agriculture of Western Australia, Jarrah Road, South
Perth 6151, Australia (I. A. F. Laing)
Soil Conservation Service of New South Wales, Flotta Lauro Building, 18-24 Pitt Street,
Sydney, N.S.W. 2001, Australia (J. C. Newman)
University of Western Australia, Department of Civil Engineering, Nedlands, Western
Australia (M. Hollick)
U.S. Department of Agriculture, Agricultural Research Service, Berkeley, California
94705, USA (L. E.. Myers)
U.S. Water Conservation Laboratory, Phoenix, Arizona 85040, USA (K. Cooley, G.
Water Resources Research Center, University of Arizona, Tucson, Arizona 85721, USA
(C. B. Cluff)

2 Runoff Agriculture

Once rainwater runoff has been harvested from slopes (see chapter 1), it
can be used for crop production (FIGURES 13 and 14). The combination is
known as runoff agriculture. FIGURES 15-24 show ingenious, simple
runoff agriculture systems in various parts of the world.
Runoff agriculture was developed almost 4,000 years ago to permit crop
production on lands receiving as little as 100 mm average annual rainfall.
Extensive investigations reveal that ancient farmers in the Middle East cleared
hillsides to increase runoff water and built rock walls along the contours to
collect it and ditches to convey it to lower lying fields (FIGURES 2-4).
These systems allowed agricultural civilizations to survive in desert regions
that today support only a small human population and produce few crops.
Warfare and political upheavals resulted in mismanagement and neglect of the
ancient farms, but the techniques of runoff agriculture are still applicable
today. Runoff farms, using modern technology and crop varieties selected for
local conditions (chapter 14), could benefit many desert regions. Artichokes,
asparagus, flower bulbs, some fruits and nuts, barley, sorghum, pearl millet,
and forages all are potentially important crops for runoff agriculture; most
are now grown in large runoff-agriculture field trials in the Negev Desert.


In runoff agriculture the principles and practices depend on rainwater
harvesting (described in chapter 1). The basic need is a rainwater catchment
that provides enough water to mature the crop. Obviously, the crop's own
water requirements (chapter 14) and general water conservation techniques
(chapters 7-16) are crucial to a successful harvest. Poor crop yields in
drought years are usually offset by production in good years.
The type of farming practiced must make the best use of the water. In
general, perennial crops with deep root systems adapt better to runoff
agriculture, because they can use runoff water stored deep in the soil, safe


FIGURE 13 A good barley crop produced by runoff agriculture in the Negev Desert.
(L. Evenari)

from evaporation. Some deep-rooted, drought-resistant fruit trees can be very
successful. Shorter lived crops can also be grown; grains, such as pearl millet,
that mature rapidly and require only one rainfall hold particular promise.8
Plants that become dormant during dry periods and begin growing when
water becomes available are particularly suited to runoff agriculture.
The desert soils and climate of the Negev have been found suitable for a
variety of crops under runoff agriculture. Excellent yields have.been obtained
from pasture plants, field crops, and orchards, well above those of dryland
farming and comparable to yields in irrigated farming (TABLE 1).

8A companion report on tropical plants, now in preparation, describes some lesser
known grains that can be grown to maturity with a single flood irrigation. (See BOSTID
publication 16, p. 153.)


FIGURE 14 Wadi Mashash, Israel (annual rainfall 100 mm). Left: runoff-treated
pasture. Right: overgrazed natural pasture. (U. Nessler)

Runoff Farming

Ancient runoff farms in the Negev (FIGURES 15 and 16) had several
cultivated fields, fed by watersheds of 10 to 50 ha. The watersheds were
divided into small catchment areas of 1 to 3 ha that allowed runoff water to

TABLE 1 Yields from Runoff Agriculture, the Negev, 1971

Crop Tons/Hectare

Peaches 6-12
Apricots 3- 8
Grapes 12 -15
Figs 6- 8
Almonds (dry shelled) 0.3- 1.8
Barley 1.3- 4.8
Wheat 1.1- 4.5
Peas (seeds) 5.4- 6.9
Sunflowers (seeds) 2.2 2.7
Alfalfa (Medicago sativa, fresh weight) 16 37.7
Wild Oats (Avena sterlis, fresh weight) 20 31.2
Pistachios 0.4- 1.8

Source: Evenari, Shanan, and Tadmor. 1971 (See page 21.)


FIGURE 15 Reconstructed ancient farm at Avdat in the Negev. Farm is, and was,
watered by runoff from surrounding slopes and wadis. In the foreground are four
reconstructed terraces; in the background, four reconstructed channels lead runoff to the
farm. To the right are traces of three channels that once carried runoff water to the
lower terraces. Ratio of catchment to cultivated area is 20:1-each ha of cultivated land
receives runoff from 20 ha of slopes, as well as direct rainfall. Cultivated area receives
water roughly equal to a rainfall of 300-500 mm from actual rainfall of 100 mm. (L.

be collected in easily constructed channels on the hillsides and were small
enough to prevent uncontrollable amounts of water (FIGURE 1). Channels
directed the water to cultivated fields which were terraced and had stone
spillways so that surplus water in one field could be led to lower ones.
Farmers dammed the small channels between the catchment and the fields
with rocks; by removing strategic rocks from the channel walls they could
guide the water to different fields at will.
A form of runoff farming that utilizes water from small, deliberately built
catchments has been practiced in Botswana.9 The water is used on school
vegetable gardens. The catchments have included school playgrounds, roads,
etc. (FIGURE 12).
Intermediate Technology Development Group, Ltd. 1969. p. 70 (See p. 21.)


FIGURE 16 Orchard in Negev. Rainwater falls on the slope behind and runs down to
strategically located ditches that convey it to the trees. In temperate regions agriculture
is based on direct precipitation and on practices, such as plowing, that encourage rain to
infiltrate the soil. Runoff agriculture is an indirect method suited to arid lands; it collects
rain from a larger area and concentrates it on a smaller, cultivated area. (L. Evenari)

FIGURE 17 Sketch of water-spreading dikes built in Pakistan. Zigzag pattern slows the
torrent of floodwater and allows it to penetrate the soil. Crops are then planted in the
wetted areas behind the dikes. (Adapted from French and Hussain. 1964. See Selected

~-- -- Y ~V--l

J -


In arid areas the limited rainfall usually falls during short, intense storms.
The water swiftly drains away into gullies, and then flows, sometimes for
many miles, toward the sea or an inland lake. Water is lost to the region, and
floods caused by this sudden runoff can be devastating, often to areas other-
wise untouched by the storm.
Water-spreading is a simple irrigation method for use in such situations:
floodwaters are deliberately diverted from their natural courses and spread
over adjacent floodplains (FIGURE 17) or detained on valley floors
(FIGURES 18 and 19). The water is diverted or retarded by ditches, dikes,
small dams, or brush fences. The wet floodplains or valley floors can then be
used to grow crops. Water-spreading is also frequently practiced on range and
pasture lands.
Water-spreading systems need a careful design and engineering layout to
withstand floodwaters. Potential sites are found on many arid and semiarid
ranges, sometimes (as in the rainshadow of a mountain range) where floods
are more common than rain. They must be selected with full consideration
for topography, soil type, and vegetation. Two requirements are essential:
Runoff waters, available for spreading, produced by an upstream drain-
age area that gives at least a few water flows each year; and

FIGURE 1.8 Man-made terraces (reinforced by unpalatable bushes) built in ancient
times to slow down and capture floodwaters in this wadi in the Negev. Built thousands
of years ago, some are still used for farming by Bedouins. Terrace walls are 10-50 m
apart and about 30 cm high. Rain brings wild flooding; surplus water pours over the
terraces; but the walls retain a pool of water that slowly sinks into the soil. Experiments
here have shown that these walls are high enough to fully moisten enough soil to get
crops such as barley or wheat to maturity. Water spreading may predate irrigation.
Ancient farmers built many such systems throughout the Middle East, South Arabia, and
North Africa. (L. Evenari)


Floodplains or gently sloping areas where the soils are suitable for crop
Inherently more risky than standard irrigation, the system depends on
fairly regular rainfall and on soils (e.g., loess) that facilitate runoff. A con-
stant concern is that sediment and gravel carried by floodwaters may adverse-
ly affect the crop land.

Microcatchment Farming

A plant can grow in a region with too little rainfall for its survival if a
rainwater-catchment basin is built around it, forcing rainfall from a
larger than normal area to irrigate the plant. This practice is called microcatch-
ment farming. The previously described principles apply to this microscale
runoff agriculture; many of the same soil treatments mentioned in chapter 1
can be used.

FIGURE 19 In 1972 near the small town of Tchirozerine (close to Agades), Niger,
West Africa, Touareg nomads build a rock wall to capture flood waters. The soils here
absorb little moisture, and the rainfall runs away in flash floods. Using stones gathered
from the fields, Touareg workers built eight 1-m-high walls across the plains so that rain
is retained, and absorbed by the soil. When summer showers fell in 1973, the water
retained by stone dams and walls flooded nearly 1 square mile on the plain and grass
flourished-extraordinary events in that area. (Oxfam-America)




FIGURE 20 Plan and cross-section of a microcatchment. Arrows indicate direction of
runoff flow. Cultivated plot (c-d) is placed at the lowest point of the natural terrain
within the catchment; its position varies. Walls are 15-20 cm high; c-d is about 40 cm
below the catchment, holding seeping water close to the plant; root-zone soil must be at
least 1.5 m deep; the a-b distance can be less than 5 m or more than 30 m, depending on
climate and crop. (M. Evenari)

Microcatchments used in the Negev Desert range from 16 m2 to 1,000 m2.
Each is surrounded by a dirt wall 15-20 cm high (FIGURES 20-23). At
the lowest point within each microcatchment a basin is dug about 40 cm deep
and a tree planted in it. The basin stores the runoff from the microcatchment.
The size of the basin is matched to the water harvest expected.
The basins are fertilized with manure, and, unlike the catchment area,
their soil surface is kept loose to encourage water penetration. A mulch may
also be used to decrease water evaporation from the soil (see chapter 9).
On an otherwise barren desert plain microcatchments provide enough
additional water to ensure the growth of fruit trees and forage plants. Micro-
catchments and variations of this method are used in Tunisia for growing
olives-and apparently have been since ancient times.
In the Negev microcatchment construction costs are very low-from
US$5 to US$20 per ha, depending on the catchment size. The cash return
from crops repays their construction costs within a few years.' o
Microcatchments are more efficient than large-scale water-harvesting
schemes (chapter 1) because conveyance losses are minimized. In light rains,

'OEvenari, Shanan, and Tadmor. 1971. (See p. 21.)


FIGURE 21 Microcatchment farming in the Negev. Pomegranate trees grow in 500-m'
microcatchments in a 100-m-rainfall region. The only soil treatment is shaping. The
orchard is less dense than those in temperate climates; 40-60 pomegranate trees per ha
are planted. Smaller trees, such as grapevines, can use smaller catchments (80-100 per
ha); catchments of just over 30 m' (320 per ha) are enough to grow a saltbush plant and
guarantee a supply of forage even in severe drought. (L. Evenari)

they provide runoff water when others will not. It is much cheaper to convert
a certain area into microcatchments than to construct a runoff farm because
microcatchments do not need channels, conduits, terraces, and terrace walls.
Also, microcatchments can be built on almost any slope, including almost-
level plains, enabling the farmer to use large, flat areas unsuited for runoff

Desert Strip (or Contour Catchment) Farming
Desert strip, or contour catchment, farming is a modification of micro-
catchment farming. It employs a series of terraces that shed water onto a
neighboring strip of productive soil. They are often tiered up a hillslope



(FIGURE 24),but on level terrain an artificial slope for the catchment can be
made by mounding soil between the strips.
The catchment section can be left in a natural state or cleared of rocks and
vegetation, planted with range grasses, or made impervious by the sealants
described in chapter 1. Desert strips are, in general, even easier to install and
maintain than microcatchments. These methods are being tested in Arizona; in
Wadi Mashash, Israel, it is used to produce grazing for sheep (FIGURE 14).


Runoff water can allow plants to grow in otherwise too arid habitats.
Highway edges often illustrate the principle: because the road acts as a catch-
ment, roadside vegetation on the lower side is greener and more dense. It has
even been proposed that water-storage tanks be built beside road pavement at
the foot of suitable hillslopes to collect water.
Runoff agriculture can be used to make new agricultural lands where water
is otherwise inadequate to support agriculture. And yields from already
cultivated areas can be increased without installing costly irrigation projects
to bring in water from a neighboring region. It has particular promise for
marginal areas; runoff agriculture can lower the risk of crop failure.
Runoff used to grow forage can relieve grazing pressures on nearby
rangelands. Overgrazed areas can be revegetated and the carrying capacity of
grazing land greatly increased. For example, the weighted average produc-
tivity of an 80-ha water-spread area at "Conneybar," Byrock, New South

FIGURE 22 Microcatchments in Botswana with 2-year-old apricot trees. (U. Nessler)


FIGURE 23 Participants in the international training course in the Negev Desert
preparing microcatchment plots (see chapter 2 Contacts). (U.'Nessler)

FIGURE 24 Desert strip runoff farming at Page Experimental Ranch near Tucson,
Arizona, USA. Grapes are cultivated in drainage ways. Catchment area treated with
sodium chloride remains bare; untreated foreground area has returned to native grasses.
Bottom left-hand corner shows part of an expanded-polystyrene floating cover designed
to reduce evaporation from the pond (see chapter 7). (C. B. Cluff)


Wales, Australia, from 1968 to 1973 was 2.66 sheep per ha. Without water-
spreading the carrying capacity of this region is 0.18 sheep per ha (as
measured over the 25 years, 1947-1972). Seasonal feed shortages still occur
in poor rainfall years at Conneybar, but their consequences are much less
than they would have been without water-spreading. High grazing intensities
have been applied for short periods. Up to 586 sheep per ha have been grazed
on a 28 ha pasture for periods of up to 4 days. Runoff-irrigated fields are
used as special fields to increase control over, and minimize losses of, newly
born animals and to hold stock during shearing, dipping, mating, etc.11
Runoff agriculture can extend the season during which forage is succulent
and nourishing, providing green forage when it is especially needed.
Water-spreading can provide erosion control, for it deflects the torrent of
water and dissipates its energy.


Runoff agriculture requires a deep soil that can store water between rains.
It works best for deep-rooted crops, such as trees and shrubs, which can tap
stored water and depend less on frequent rainfall. Annual crops, in contrast,
need rain at the beginning of the growing season and sometimes at intervals
The method is enhanced by plant varieties able to withstand intermittently
wet and dry soil. As in normal agriculture, yields also depend on insect and
disease control.
Environmental prerequisites are
A minimum mean precipitation of 80 mm per rainy season if the
rainy season coincides with the cold period of the year, more than 80 mm if
it occurs during summer when evaporation is greater;
Crust-forming or impermeable soils on the catchment areas;
Soils in the cultivated areas with high water-storing capacity;
Not more than 2-3 percent salinity in the cultivated soil; and
A minimum of 1.5-2 m of soil depth in the cultivated area (unless
water-storage facilities are available).
In runoff agriculture the water must be distributed evenly over the
cultivated area to prevent prolonged ponding, overirrigation, or deep
percolation losses. In some cases the area to be cultivated can be constructed
so that any excess spills to a lower collection level. The cultivated area must
be uniform, without gullies or ridges. Before deciding on runoff agriculture,
one needs to consider
The water-use characteristics of plants to be grown;
"Cunningham. 1973. (See Selected Readings.)


Their yields;
Their ability to resist drought;
Whether the soils in the cultivated areas can store enough water to
mature the crop; and
The amount of evaporation from the soil surface.

Stage of Development

In ancient times runoff agriculture was widespread over the whole arid
region of the Middle East, southern Arabia, and North Africa. On many
thousands of hectares of the Negev Desert, it was the basis for civilization
Runoff agriculture has been shown to be technically sound for modem
use, too. Its rebirth as a systematic method took place in the Negev Desert in
Israel, where large-scale experiments have been conducted for the past 15
years (FIGURES 15, 16, and 21) and a training school for developing-country
personnel is located at Wadi Mashash (FIGURE 23). Some microcatchment
farming occurs today in several other arid countries such as Mexico, Botswana
(FIGURE 22), India, Pakistan, and Australia. The microcatchment method is
used to grow wheat and fruit trees over a 70,000-ha area in Khost province,

Needed Research and Development

Runoff agriculture can be used today if care is taken in selecting the site,
designing the system, and selecting the crop. With good management it can
make arid wasteland productive and can be an economically sound invest-
ment. Modem experience, however, is limited to a few isolated projects.
Intensive technoeconomic evaluations in several regions of the world with
different climates, soils, and crops are needed to identify its potential for the
To make runoff agriculture more effective, there is a need to develop crops
better suited to it (further discussed in chapter 10). For example, if crops
matured in 60 instead of 80 days, the soil would not have to store so much
water, the risk of crop failure would be lessened, the system would require
less rainfall, and management requirements would be reduced.

In microcatchment farming the crucial problem is still the optimal size of the
microcatchment for each species. It is obvious that this parameter is relative
not only to each species but also to precipitation, soil quality, and steepness
of gradient. We have much to learn about such matters. Other problems
concern optimal depth and the size of the basin in relation to the size of the


catchment area. These factors are most important because they determine,
inter alia, the size of the surface area wetted by the flood and the volume and
depth of the water column in the soil. These in turn affect the time during
which the soil containing the root system is waterlogged and soil and root
aeration is bad. A knowledge of these factors may even lead to different
patterns of constructing the basins and the placing of the trees-perhaps on a
knoll inside of the basin. There may also be the possibility of increasing
runoff volume by pretreating the soil surface of the microcatchments in
different ways.12

Technoeconomic studies are particularly needed for runoff agriculture
using chemically treated and ground-covered catchments.

Selected Readings
(See also Selected Readings, chapter 1.)
Cull, J. K. 1964. Water spreading at Lanheme. Queensland Agricultural Journal
Cunningham, G. M. 1970. Waterponding on scalds. Journal of the Soil Conservation
Service of New South Wales, 26:146-71.
Cunningham, G. M. 1973. Waterspreading and Waterponding on Australian Rangelands
Paper presented at the Alice Springs Symposium on Water in Rangelands, October
1973. (Available from the author; see Contacts.)
Evenari, M., et al 1968. Runoff farming in the desert. I. Experimental layout. Agron-
omy Journal 60:29-32. (See also Shanan, et al.; Tadmor, et al. below.)
French, N., and I. Hussain. 1964. Water Spreading Manual. Range Management Record
Number 1, West Pakistan Range Improvement Scheme, Lahore, Pakistan.
Hauser, V. L., et al. 1968. Conservation bench terraces. Tansactions of the American
Society of Agricultural Engineers 11:385-98.
Jones, R. M. 1966. Tree establishment on scalds in the Hay Plain. Journal of the Soil
Conservation Service of New South Wales 22:2-7.
Michelson, R. H. 1966. Level pan system for spreading and storing watershed runoff.
Soil Science Society of America Proceedings 30:388-92.
Newman, J. C. 1963. Waterspreading on marginal arable areas. Journal of the Soil
Conservation Service of New South Wales 19:49-58.
Newman, J. C. 1966. Waterponding for soil conservation in arid areas in New South
Wales. Journal of the Soil Conservation Service of New South Wales 22:18-28.
Proceedings of the Water Harvesting Symposium 26 28 March 1974, Phoenix, Arizona.
1974. Published by Agricultural Research Service. (While supplies last, single copies
available free from U.S. Water Conservation Laboratory, 4331 E. Broadway, Phoenix,
Arizona 85040, USA.)
Quilty, J. A. 1972. Soil conservation structures for marginal arable areas: diversion
spreader banks and tank drains. Journal of the Soil Conservation Service of New
South Wales 28:157-68.
Shanan, L., et al. 1970. Runoff Farming in the desert. III. Microcatchments for improve-
ment of desert range. Agronomy Journal 62:445-9.
Tadmor, N. H., et al. 1970. Runoff farming in the desert. IV. Survival and yields of
perennial range plants. Agronomy Journal 62:695-9.
Tadmor, N. H., et al. 1971. Runoff farming in the desert. V. Persistence and yields of
annual range species. Agronomy Journal 63:91-5.

12Evenari, Shanan, and Tadmor. 1971. Op. cit. p. 228.


Zing, A. W., and V. L. Hauser. 1959. Terrace benching to save potential runoff for
semiarid land. Agronomy Journal 51:289-92.

(See also Contacts, chapter 1)
Central Arid Zone Research Institute, Jodhpur, India (H. S. Mann, Director)
College of Agriculture, University of Arizona, Tucson, Arizona 85721, USA
(W. G. Matlock)
Department of Agronomy and Plant Genetics, University of Arizona, Tucson, Arizona
85721, USA (N. F. Oebker)
German Agricultural Group, P.O. Box 183, Kabul, Afghanistan
Intermediate Technology Development Group, Ltd., Parnell House, 25 Wilton Road,
London SW1V 1JS, England
Oxfam, B. P. 489, Ouagadougou, Haute Volta, Africa (M. Behr)
Soil Conservation Service of New South Wales, Condobolin, New South Wales 2877,
Australia (G. M. Cunningham)
Wadi Mashash, a runoff-agriculture training center in the Negev Desert for trainees
from arid lands all over the world (M. Evenari, Botany Department, Hebrew
University of Jerusalem, Jerusalem, Israel). Also contact: Wadi Mashash Information
Center, 61 Darmstadt, Paulusplatz 1, West Germany (0. Schenk and U. Nessler).

3 Irrigation with Saline Water

Beneath many of the world's deserts are reserves of saline water, and many
surface waters-estuaries, coastal lagoons, land-locked lakes, and irrigation
return flows-contain fairly large amounts of salt. If saline water could be
used for irrigation, more desert land could be cultivated; the nonsaline water
now used in agriculture could be released for human consumption, reducing
the need for expensive desalination schemes now contemplated for supplying
urban areas.
Today, new appreciation of plant physiology and soil science and new
irrigation techniques are showing that with careful management, saline waters
can be used to grow a variety of crops.


The salt resistance of crops largely determines the suitability of saline
irrigation waters. The salt tolerance of crops has now been studied
intensively, and adequate information for selecting crops of suitable
resistance to saline waters is becoming available. Although only a few crops
such as cotton, barley, wheat, sugar beets, rye grass, Bermuda grass, and the
wheat-grasses Agropyron elongatum and A. desertorum are known to be salt
tolerant, they are important in developing countries because they form the
basis of much agricultural production. Salt-tolerant trees include the date
palm, olive, pomegranate, and pistachio.13
In general, irrigation waters whose total dissolved solids are below 600
milligrams per liter (mg/1) may be used on almost any crop. If leaching
(discussed later) and drainage are adequate, water from 500 to 1,500 mg/l
can be, and is, widely used on all but the most salt-sensitive crops. Water of
1,000 2,000 mg/1 can be used for crops of moderate tolerance, especially if
frequent irrigation is employed. Water of 3,000- 5,000 mg/1 will produce
high yields only from highly tolerant crops, such as those listed earlier.
Despite claims to the contrary, irrigation with undiluted seawater has not
13Some lesser known salt-tolerant plants are described in a companion report on
neglected tropical plants, now in preparation. (See publication 16, p. 153.)


proved practical for producing crops. Seawater has a total salt content of
about 35,000 mg/1, greatly exceeding the tolerance of even the most salt-
tolerant crop studied to date-Suwanee Bermuda grass Cynodon dactylon,
which can tolerate about 12,000 mg/1.
The type, as well as concentration, of salt in the water is important. For
example, the relative concentrations of sodium to calcium and magnesium
affect a water's suitability because high sodium ratios affect soil structure
(compare FIGURES 46 and 47) and plant nutrition. The anion of the salt,
e.g. chloride or sulfate, may also be important.
In the practice of saline irrigation, the basic premise is that proper
irrigation and drainage management prevent salts from building up in the soil.
It is essential to avoid this buildup by leaching: applying more irrigation
water than the plant requires so that the extra water carries the salts down

FIGURE 25 Date palms in Southern Tunisia that have been irrigated for 4 years with
water containing 2,000 mg/1 salts. The extensive drainage system is required to
facilitate leaching. (J. W. van Hoorn)


below the plant's roots. The suitability of saline water for irrigation is thus
also governed by the leaching characteristics of the environment, i.e., whether
they facilitate or retard the removal of salts from the root zone (FIGURE
25). If the leaching characteristics of the soil particles or the overall drainage
of the area are insufficient, soil salinity will increase, and the final result may
be a barren wasteland. Light-to-medium-textured soils that are not subject to
structural changes that restrict water flow are more likely to be successfully
irrigated with saline waters.
According to recent findings, new irrigation methods can increase a crop's
tolerance to salinity. Compared to furrow irrigation, trickle irrigation
(chapter 10), for example, has been shown to improve yields of crops
irrigated with saline water (2,000 2,500 mg/1). The salt stresses on a plant
are aggravated as the soil dries out and the salt concentration increases.
Frequent irrigations (as in the trickle method) minimize such stresses.


Saline ground, surface, and estuarine waters are widely available but not
often used for irrigation; new research findings make it possible to use them
more widely for agriculture, landscaping, etc.
The cost of using saline water, especially from aquifers near the surface, is
not likely to be excessive. At Kibbutz Mashabei S'deh in the Negev Desert

FIGURE 26 Irrigating cotton with brackish water (2,500 ppm dissolved salts) caused a
stunted plant, but the cotton yield was 59 percent higher than with freshwater irrigation.
Spraying plants with the transpiration-suppressing plant hormone abscisic acid (see
chapter 13) further increased cotton yield. (M. Twersky)


FIGURE 27 Maize and fodder sorghum in Tunisia irrigated with sweet water
containing only 200 mg/1 of salt (left side) and with saline water containing 3,500 mg/1
(right side). Respective yields of maize-9,000 and 5,000 kg of grains per ha; of fodder
sorghum (foreground)-90 and 50 tons of green matter per ha. (J. W. van Hoorn)


S.'. '.

FIGURE 28 Rhodes Grass (Chloris Gayana Kunth) irrigated with 2,600 mg/1 saline
water. (Kibbutz Mashabei S'deh, Israel)


(100-mm rain) a 4,800 m3/day electrodialysis plant desalts underground
brackish water. For comparison, water from the same well (2,600 mg/1) has
been used to irrigate cotton (FIGURE 26), wheat, maize, sorghum (FIGURE
27), Rhodes grass (FIGURE 28), Bermuda grass, etc., and various vegetables.
After 3 years, economic and general-utility considerations favor direct use of
the saline water over the electrodialysis; the saline aquifer is now used for this


Although saline-water irrigation holds exciting possibilities for the future,
it does not promise the conversion of vast stretches of arid land into
cultivated fields. Many crops cannot tolerate it; indiscriminate use may
severely damage the soil; and suitable soil and climate do not always coincide
with suitable water. Furthermore, the greater management skills necessary
may not be available.
To irrigate with saline water and maintain good yields demand good water
management by trained specialists. The type and mix of salts and their con-
centrations in the water need to be investigated before a decision is made to
use saline waters. Saline-water irrigation will increase the salinity of ground-
water and possibly make it unsuited for other uses. If groundwater is the
source, the project may have a short useful life. If poorly managed, irrigation
with saline water may seriously damage the soils and even make them barren.
When irrigation water contains 5,000 mg/1 or more of salts, the leaching
requirements for even highly tolerant crops may be substantial. For example,
more than 25 percent extra water may be needed just to move the salts below
the root zone; if less is used, salination of the soil occurs because salts are not
removed as fast as they are added.
Even highly tolerant crops may go through salt-sensitive stages when they
need low-salt water. For example, seedlings of small grains and sugar beets are
sensitive, though the adult plants are not; saline water may impair their
One must be careful in applying test results from temperate regions to
tropical arid lands. The same crop grown in temperate or humid regions can
tolerate more highly saline water than it can in arid regions because rains (and
soil waters) dilute the irrigation water. Lower temperature may also increase
a plant's salt resistance.

141nformation provided by J. Schecter, Acting Director, University of the Negev-
Research and Development Authority.


Stage of Development

The use of highly saline water for irrigation has been fairly limited. As
already mentioned, large-scale experiments are under way using brackish
water for irrigation on sand and sandy loam soils in the Negev Desert. The use
of brackish water for irrigation has been studied during 7 years of field
research at six experimental stations in Tunisia. Major objectives of this
UNESCO-sponsored program were to determine the optimum use of the avail-
able saline surface and ground waters and to control soil salinity through
improved irrigation techniques. As a result, saline river water containing
2,000 3,500 mg/1 of salt is today used in Tunisia for irrigation on a large
scale in the Medjedah Valley (and other locations) on medium-to-heavy-
textured soils fitted with an extensive tile drainage system (FIGURES 25 and

Needed Research and Development

Breeding and selection of plants that can use water of higher than normal
mineral content is very much needed. The greatest success will undoubtedly
be with crops already showing some tolerance to salinity, e.g., coastal
Bermuda grass, barley, cotton, wheat, sugar beets, and perhaps the more
salt-tolerant vegetables. Some varieties with greater salt tolerance than usual
are known and may become the basis for breeding stock.
Using saline water for irrigation requires sophisticated management, but
detailed management requirements are not fully understood and need careful
investigation. The following need particular attention:
Determining the relationship between saline water and the physiological
stress performance of plants;
Alleviation of stresses by different irrigation, fertilization, soil aeration,
leaching practices, nutrients, hormones, chemical and physical treatments,
etc.; and
Field application of the knowledge gained.

Selected Readings
Amnon, I. 1972. Crop Production in Dry Regions. Volume 1: Background and Principles.
Barnes and Noble, New York and Leonard Hill Books, London. 650 p.
Bernstein, L. 1964. Salt Tolerance of Plants. Agriculture Information Bulletin No. 283.
U.S. Government Printing Office, Washington, D.C. 20402, USA. 23 p. (US$0.20).
Casey, H. E. 1972. Salinity Problems in Arid Lands Irrigation: A Literature Review and
Selected Bibliography. Arid Lands Research Information Paper Number 7, University
of Arizona, Office of Arid Lands Studies, Tucson, Arizona 85721, USA. 311 p.
FAO. 1971. Salinity Seminar Baghdad. Irrigation and Drainage Paper Number 7. Rome.
254 p.


Levitt, J. 1972. Responses of Plants to Environmental Stresses. Academic Press, New
Stylianou, Y., and P. I. Orphanos. 1970. Irrigation of Shamouti oranges with saline
water. Technical Bulletin Number 6. Cyprus Agricultural Research Institute, Nicosia,
Twersky, M. 1971. Factors of Chemical Fertilization in Saline Water Irigation: A
Review. The Negev Institute for Arid Zone Research, Beer Sheva, Israel.
Twersky, M., and D. Pasternak. 1972. Crop Irrigation with Brackish Water in Mashable
Sadeh. Preliminary Report, The Negev Institute for Arid Zone Research, Beer Sheva,
UNESCO. 1972. Hydrological Aspects of Saline Water Resources: A Provisional
Annotated Bibliography. Distribution No. SC/WS/438. Paris. 66 p.
UNESCO/FAO. 1973. Irrigation, Drainage and Salinity: An International Source Book.
Hutchinson. 510 p. (Available from Unipub, Box 433, New York, New York 10016,
UNESCO. 1970. Recherche et formation en matiere d'irrigation avec des eaux sales en
Tunisie. Paris. August 1970.
van Schilfgaarde, J., et al. 1974. Irrigation management for salt control. Journal of the
Irrigation and Drainage Division Proceedings of the American Society of Civil
Engineers. (In press.)
Waisel, Y. 1972. Biology of Halophytes. Academic Press, New York. 395 p.

Central Salt and Marine Chemicals Research Institute, Bhavnagar, Gujarat, India
Central Soil Salinity Research Institute, Kamal, Punjab, India
Centre de Recherches pour l'Utilisation de l'Eau Salee en Irrigation, Route de Soukra,
B. P. 10, Ariana, Tunisia
Agricultural University, De Nieuwlanden, Nieuwe Kanaal 11, Wageningen, The Nether-
lands (J. W. van Hoorn)
Kibbutz Revivim, Negev, Israel (J. de Malach)
Ministry of Agriculture and Natural Resources, Nicosia, Cyprus
Negev Institute for Arid Zone Research, Beer Sheva, Israel (J. Schecter)
U.S. Salinity Laboratory, U.S. Department of Agriculture, Agricultural Research Service,
P.O. Box 672, Riverside, California 92502, USA (J. van Schilfgaarde, Director)
Water Resources and Development Service, Land and Water Development Division, Food
and Agriculture Organization of the United Nations, Via delle Terme di Caracalla,
00100 Rome, Italy.

4 Reusing Water

Reusing water can greatly lower the overall demand for water resources.
Wastewater can be used for irrigation, for industry, for recharge of ground-
water; in special cases, properly treated wastewater has been used for munic-
ipal supply. With careful planning, various industrial and agricultural demands
may be met by purified wastewater, thereby freeing freshwater for
municipalities, which require better water, suitable for human consumption.
Water reuse may have greater impact on the future usable water supply in
arid areas than any of the other technologies discussed in this report.


Agricultural Use
Spreading wastewater on marginal land to create new farmland may prove
particularly important in arid countries. In such areas reclaimed water will
probably be used first for irrigation. Filtering wastewater through soil
removes all particulate matter: most cations and some anions (including
phosphates) are strongly adsorbed, and organic matter is decomposed by soil
bacteria. These actions can contribute plant nutrients to soil (FIGURES 29
and 30).
Using municipal wastewater for irrigation is especially attractive where
agricultural lands are located close to cities, because the plant nutrients in
sewage would otherwise go to waste. Some biological treatment of sewage
should precede land application, but for many crops the degree of treatment
required is so low that little technology and capital investment are required.
(Mexico City uses vast amounts of untreated sewage as irrigation water.)
Where irrigation systems are already in use, connecting them to municipal
systems is fairly simple, though institutional arrangements may prove dif-
ficult. The American Public Works Association has recently stated that

... on the basis of the exhaustive study which was undertaken, it must be
concluded that the land application of wastewaters offers a viable alternative


^io ^^ t.r.
*S~i <-.;^d /W

FIGURES 29, 30 Two maize crops on a 10-ha plot at Rhodesia's Salisbury sewerage
works, 1971-72 season. Tall maize on left was given a particularly heavy preirrigation
with treated sewage. Stunted maize at right was irrigated almost entirely by rainfall. No
additional fertilizer was added to either crop. The contrast shows very clearly the
fertilizing effect of nutrients in the sewage. (Rhodesia Agricultural Journal)

to advanced treatment processes and deserves serious consideration by many
communities and industries throughout the United States. Land needs, when
taken in perspective with total land uses, are not unreasonable and may, in
fact, play a desirable social role by providing green belts and open areas, and
preserving rich farm lands and cloistered areas. The conclusions of the report
point to the almost unqualified success of this method of application, both in
this country and throughout the world, when the facility has been properly
operated and efforts have been made to apply sound engineering, geological
and farming expertise to design, construction and control procedures.5

Since 1892, Melbourne, Australia, with a present population of almost
2 million, has disposed of its wastewater in irrigation at the 109 km2 Board
of Works Farm at Werribee (FIGURE 31). A total of 4,200 ha of the farm is
employed for irrigated pasture of which 1,370 ha are used for grazing 15,000
head of cattle through the year. Forty to fifty thousand sheep are fattened
during spring and summer. Health restrictions are imposed only on the sale of

15American Public Works Association. p. viii. (See Selected Readings.)


t.i ~

FIGURE 31 This Melbourne and Metropolitan Board of Works Sewerage Farm at
Werribee, Australia, has long used treated sewage from the city of Melbourne as
irrigation water. This formerly barren, arid, windswept plain is the major sewage-
disposal facility, serving more than 1.5 million people. (Sewage flow averages 360
million 1 per day, 950 mid during rainy periods.) The farm grazes 15,000 cattle a year;
40,000-50,000 sheep are fattened during the spring and summer, resulting in the sale of
5,000 cattle, 36,000 sheep, and 250 bales of wool during an average year. (Melbourne
and Metropolitan Board of Works)


cattle and sheep for slaughter-but the 0.02 percent condemnation rate of
cattle carcasses is the same as that for the surrounding area. No higher
incidence of disease among farm employees has been found to result from
their employment.16 In less well-operated sewage irrigation projects in India,
however, the operators have been found to have abnormally high loads of
Recycling irrigation water runoff by pumping it back to the head of the
system is another way to reuse water for agriculture, but salinzation of the
soil is a serious hazard (see chapter 3). Industrial wastewater may also be fit
for irrigation, but it may require treatment when the industrial process adds
chemicals detrimental to plant growth or public health.

Industrial Use
Municipal wastewater from secondary treatment plants can be used for
cooling, ore separation, and other purposes that do not have severe water-
quality requirements. Use as process-water requires advanced treatment. The
degree and kind of treatment depend on demands and economics of the
application. For pulp and paper production the use of wastewater after only
limited advanced treatment has been found to be economically feasible.18

Municipal Use
Municipal use places the highest demands on water quality. Wastewater
must usually undergo secondary and tertiary treatment to make it potable.
Processes for removing ammonia, nitrates, and phosphates are available;
residual, potentially toxic compounds and dissolved organic substances can be
reduced to very low levels by adsorption on activated carbon. Dissolved
mineral matter can, if necessary, be reduced to acceptable levels by ion
exchange, electrodialysis, or reverse osmosis; however, adding these processes
can double or triple the capital and operating costs of a conventional treat-
ment plant.
Producing water of the necessary quality requires a large investment in
capital equipment, power, and chemicals. The cost of such water is relatively
high, but may be lower than that of desalted seawater (chapter 6); in arid
lands it may be lower than the cost of developing alternative supplies of
water, though not if treatment is needed to remove dissolved mineral salts.
Windhoek, South-West Africa, a metropolitan area of 84,000, meets its
water needs by treating and recycling into the potable water supply 4 million
16Ibid. Section V.
17Ibid. p. 147.
18Personal communication from National Institute for Water Research; Pretoria, South
Africa (see Contacts).


FIGURE 32 City of Windhoek, South-West Africa, (population 84,000) on the edge of
the Namib Desert cycles its municipal wastewater through this treatment plant and back
into the potable supply. (National Institute for Water Research, Pretoria, South Africa)

i/day of its sewage, which represents one-third of the total daily supply
(FIGURE 32).


The important advantage of water reuse is that it can, if properly managed,
reduce by severalfold the demand on water from natural sources. Contin-
uously recycling 50 percent of the wastewater in effect doubles the water
In some arid locations reusing wastewater in industry may provide
additional water needed to permit industrialization that would not otherwise
be feasible.


In any reuse scheme these major constituents of wastewater have to be
Pathogenic bacteria and viruses
Parasite eggs
Heavy metals


Pathogenic bacteria can be killed by chlorine disinfection, but the remain-
ing hazard from viruses after advanced waste treatment is not known. Al-
though few outbreaks of epidemic virus diseases have been attributable to
transmission in water supplies, the prevailing view is that present known
treatment processes remove the hazard only if very carefully controlled.19
Too much uncertainty exists to certify water from even advanced treatment
processes as virologically safe. Though this increases reluctance to use treated
wastewater for drinking or for irrigating vegetables that are eaten fresh, it
should not hinder its use for less critical purposes.
To reuse water without causing environmental disaster calls for good
management and a good understanding of the user's requirements. Systems
can easily be mishandled and cause serious disease or harm to the environment.
If more than 50 percent of the water supply is wastewater, salt accumulation
can cause serious problems whether the water is for agriculture, industry, or
municipal use.

Assimilative disposal of wastes on the land, which does no permanent
damage, is clearly a very different thing from uncontrolled dumping, which
destroys the soil and may lead to serious pollution of ground water. Far too
few engineers have a feel for proper rates of application to the soil. They
forget that a little too much can change dilution to pollution. If you want to
use a farm or a forest as a dis osal site, there is no substitute for a good
farmer or forester to manage it.

The cost and difficulty of reusing water depend on the treatment processes
needed. Some secondary and most tertiary treatments require large capital
investment and trained, capable personnel. Operating costs are high, in many
cases too high for water reuse to be feasible. In arid regions the cost structure
is more favorable. Direct reuse for potable supplies may have to overcome
aesthetic objections, even if the water is demonstrably pure. Furthermore,
people may object to eating food grown with human waste. Reusing water
will often require that all sectors-agriculture, industry, and urban adminis-
trations-be integrated in management and policy.

Stage of Development

Reuse of water has been practiced ever since people have taken water from
rivers. Thus, in a sense, it is not new. Along rivers such as the Ganges, Nile,
and Mississippi individuals, communities, and industry reuse the water many
times over. There is no evidence that this causes harm. This chapter deals,

19Malina and Sagic. 1974. (See Selected Readings.)
20Dean. 1971. (See Selected Readings.)


however, with the planned and deliberate reuse of water, which is increasing.
As demands for existing resources in water-scarce areas become even greater,
its potential is highly encouraging. Although the technology for reuse is now
available, economic considerations will probably limit its use initially to
specialized locations or purposes. Even so, such uses may release natural
sources of water for potable supplies. Ultimately, as acceptance grows, wide-
spread recycling of wastewater into the potable supply is a definite
possibility. Already, some sanitary engineers recommend adding properly
treated wastewater to the potable supply, but most are cautious because of
present uncertainties over the danger from viruses and heavy metals. 1

Needed Research and Development

Because water reuse will undoubtedly be highly competitive with alter-
natives, it deserves a high research priority. The considerable amount of
research under way has not emphasized reusing water to increase a nation's
supply of water as much as at least one of its competitors, desalination.
Research should center on ways to reduce cost, to combine secondary and
advanced treatment processes, and to answer concerns about virological
The importance of continued research to develop treatments to reduce
virological hazards, and research to determine the residual hazard after treat-
ment, cannot be overstressed.
Research is needed to reduce the cost of tertiary treatments and to
develop alternative, less expensive treatment processes. Electrodialysis and
reverse osmosis show promise for removing many kinds of dissolved
impurities, but better antifouling techniques and membranes that require less
pretreatment are needed. Improved biological processes are needed for
removing ammonia and nitrates in secondary effluents, as are new low-cost
specific ion exchangers for removing mineral salts.
In the agricultural use of wastewater our biggest lack of knowledge is on
the effects of long-term application. We do not fully understand to what degree
continuous application of wastewater to land alters the nature of soil. We do
not know the capacity of soils to absorb different metals (boron for instance)
without permanent damage. Nor have we learned how to prevent or restore
altered soils.
Research is also needed to develop improved management techniques and
institutional arrangements whereby effluent can be substituted for potable
water now used in agriculture and industry.

21American Society of Civil Engineers Committee on Environmental Quality
Management. 1970. (See Selected Readings.)


Selected Readings

Agriculture and Land Spreading
American Public Works Association. 1973. Survey of Facilities Using Land Application
of Wastewater. Report Number EPA-430/9-73-006. Office of Water Program
Operations, U.S. Environmental Protection Agency, U.S. Government Printing
Office, Washington, D.C. 20402, USA. 377 p. (Comprehensive survey of 100 facilities
worldwide where domestic or industrial wastewater effluents are applied to the land.)
Bouwer, H. 1968. Returning wastes to the land, a new role for agriculture. Journal of
Soil and Water Conservation 23:164-8.
Cluff, C. B.; K. J. deCook; and W. G. Matlock. 1971. Technical and institutional aspects
of sewage effluent-irrigation-water exchange, Tucson region. Journal of the American
Water Research Association 7:726.
Dean, R. B. 1971. Ultimate disposal of industrial waste: an overview. Massachusetts
Institute of Technology, Technology Review 73(5):20.
Kirby, C. F. 1967. Irrigation with waste water at the Board of Works Farm, Werribee.
Symposium on Water on the Farm. Water Research Foundation of Australia Report
No. 25. (P.O. Box 47, Kingsford, New South Wales 2032, Australia.)
Malina, J. S., and B. P. Sagic. 1974. Virus Survival in Water and Wastewater Treatment.
Center for Research in Water Resources, The University of Texas, Austin, Texas (In
Metcalf and Eddy, Inc. 1973. Wastewater Treatment and Reuse by Land Application.
(2 Vol.) Report No. EPA-660/2-73-006. U.S. Government Printing Office, Wash-
ington, D.C. 20402, USA. (Vol. 1: U.S.$1.10; Vol. 2: U.S.$2.40)
National Association of State Universities and Land-Grant Colleges. 1973. Recycling
Municipal Sludges and Effluents on Land. Report No. PB 227-184A/S. (National
Technical Information Service, U.S. Department of Commerce, Springfield, Virginia
22151, USA. U.S.$6.00.)
Sopper, W. E., and L. T. Kardos, eds. 1973. Recycling Treated Municipal Wastewater and
Sludge through Forest and Cropland. Pennsylvania State University Press, University
Park and London. 479 p.
Special Report on Land Disposal. 1973. Journal of the Water Pollution Control
Federation 45:1465-1507.
Wolman, A. 1948. Industrial water supply from processed sewage treatment plant
effluent at Baltimore, Maryland. Sewage Works Journal 20:15.

Municipal Recycling
American Society of Civil Engineers Committee on Environmental Quality Management.
1970. Engineering evaluation of virus hazard in water. Journal of the Sanitary
Engineering Division, American Society of Civil Engineers 96(SAL):111.
Stander, G. J., and J. Funk. 1969. Direct cycle water reuse provides drinking water in
South Africa. Water and Wastes Engineering 6(5):66.

Culp, G. L., and R. W. Culp. 1971. Advanced Wastewater Treatment. Van Nostrand
Reinhold Co., New York.
Fair, G. M., and J. C. Geyer. 1968. Water and Wastewater Engineering Volume 2. John
Wiley, New York.
Gavis, J. 1971. Wastewater Reuse. (Prepared for the National Water Commission.)
National Technical Information Service, Springfield, Virginia 22151, USA. Accession
No. PB 210-535.


Metcalf and Eddy, Inc. 1972. Wastewater Engineering, McGraw Hill, New York. 782 p.
National Water Commission. 1973. p 306-315. (See Selected Readings, p. 5.)
U.S. Department of Housing and Urban Development, Office of International Affairs,
1970. Sewage Lagoons for Developing Countries. Ideas and Methods Exchange
No. 62.35 p. (Available from the authoring institution, Washington, D.C. 20410, USA.)
Weber, W. J., Jr. 1972. Physiochemical Processes for Water Quality Control. Wiley Inter-
science Co., New York.

American Public Works Association, 1313 East 60th Street, Chicago, Illinois 60637,
Bureau of Sanitary Engineering, State of California, Department of Public Health,
Berkeley, California 94720, USA (E. Sepp)
Central Public Health Engineering Research Institute, Nagpur, Maharashtra, India
(G. B. Shende)
College of Agriculture, University of Arizona, Tucson, Arizona 85721, USA (C. B. Cluff,
A. D. Day, and T. C. Tucker)
Civil Engineering Department, University of Nairobi, Nairobi, Kenya (R. Jones)
Civil Engineering Department, University of Texas, Austin, Texas 78712, USA
(J. F. Malina)
Community Water Supply and Sanitation, Division of Environmental Health, World
Health Organization, Geneva, Switzerland (L. A. Orihuela)
Department of Soil Science, University of California at Riverside, Riverside, California
92501, USA (A. L. Page)
Department of Water Affairs, South-West Africa Branch, Windhoek, South-West Africa
(Namibia) (S. W. Burger)
Department of Watershed Management, University-of Arizona, Tucson, Arizona 85721,
USA (G. S. Lehman and L. G. Wilson)
Environmental Health Laboratory, Hebrew University of Jerusalem, Jerusalem, Israel
(H. I. Shuval)
Institute for Research on Land and Water Resources, Pennsylvania State University,
University Park, Pennsylvania 16802, USA (W. E. Sopper)
Institute for Wate4 Berlin, Germany (W. Niemitz)
Institute De Inginiero Sanitaria, Universidad De Buenos Aires, Buenos Aires, Argentina
(R. A. Trelles)
Melbourne and Metropolitan Board of Works, 625 Little Collins Street, Melbourne,
Victoria, Australia
Municipal Treatment and Reuse Section, Advanced Waste Treatment, Environmental
Protection Agency, Cincinnati, Ohio 45268, USA (R. B. Dean)
Muskegon County Department of Public Works, Muskegon, Michigan 49440, USA
(J. C. Postlewait)
National Institute for Water Research, Council for Scientific and Industrial Research,
P.O. Box 395, Pretoria, South Africa
Negev Institute for Arid Zone Research, Beer Sheva, Israel (J. Schecter)
Office of Water Programs Operations, U.S. Environmental Protection Agency, Wash-
ington, D.C. 20460, USA (B. L. Seabrook)
Robert S. Kerr Environmental Research Laboratory, Environmental Protection Agency,
P.O. Box 1198, Ada, Oklahoma 74820, USA (C. Harlin, R. Thomas)
U.S. Water Conservation Laboratory, U.S. Department of Agriculture, ARS, Phoenix,
Arizona 85040, USA (H. Bouwer)

5 Wells

Dug wells, qanats, and horizontal wells are discussed in this chapter.

Dug Wells
Hand-dug wells (FIGURE 33) have been used for thousands of years but
have become less popular with the advent of tube wells.22 Today, interest in
dug wells is reviving, and they still hold much promise for arid lands. Modern
materials, tools, and equipment may transform crude holes in the ground,
hosts for parasitic and bacterial diseases, into more safe, soundly engineered,
hygienic, and reliable sources of water.23 Dug wells are inexpensive and easy
to construct and maintain by fairly unskilled labor. They provide storage for
water, as well as a source.
In most cases dug wells will be superseded by tube wells, but they provide
an important transition step, and in some cases, dug wells will always be best,
e.g., for shallow, low-yielding aquifers and for inaccessible regions where
transporting drilling equipment is difficult.
In Afghanistan and India, dug wells are being seriously reconsidered.24
Since about 1954, when the use of air compressors and rok drills became
common, many existing dug wells on the Deccan plateau of central India have
been deepened by digging through lava flows that had blocked previous
equipment. In the last 10 years India has also improved many dug wells solely
by adding pumps. Powered by internal-combustion engines or electric motors,
inexpensive centrifugal or turbine-type pumps, installed on platforms 1-2 m
above the water level, boost the water up to ground level. Suitable pumps are
now made in many developing countries, including India and Pakistan.

Dug wells, however, do have distinct limitations:
They cannot be used to reach groundwater deeper than 20-30 m.
Their water production is usually low.
Well-digging technology is understood and used in most countries, but
the art of lining wells has regressed, and there is an important need for
improved linings. The liner protects against caving and collapse and prevents

22Gibson and Singer. 1969. (See Selected Readings.)
23Wagner and Lanoix. 1959. (See Selected Readings.)


&LV .1

FIGURE 33 Excavation of a dug well in the Negev Desert. (U. Nessler)


polluted surface water from entering the well. The main problem is lining the
walls below the level of the water table.
Another need is for safer, more rapid, more efficient digging techniques.

A qanat is, essentially, a horizontal tunnel that taps underground water in
an alluvial fan and, without pumps or equipment, brings it to the surface so
that it can be used.
A qanat system is composed of three essential parts (FIGURE 34):
One or more vertical head wells dug into the water-bearing layers of
the alluvial fan to collect the water.
A gently downward-sloping underground horizontal tunnel leading the
water from the head wells to a lower point at the surface. (In part, the
tunnel acts as a subsurface drain to collect water.)
A series of vertical shafts between the ground surface and the tunnel,
for ventilation and removal of excavated debris (FIGURE 35).
Because qanats deliver water without pumps and pumping costs (FIGURE
36), they can be used where pumped wells are too expensive to operate.
Qanats vary greatly in length, depending on the depth of the aquifer and
the slope of the ground. The conduit from the head well to the mouth may
extend 1-4 km; one in southern Iran is more than 28 km. Commonly, the
length is 10-16 km. The water obtainable from individual qanats also varies;
a study of 200 in the Varamin plain southeast of Teheran, Iran, measured
the largest yield at 270 liters per second and the smallest at 1 liter per
About 3,000 years ago the Persians learned to dig qanats (an ancient
Semitic word and ancestor of the word canal) to bring mountain groundwater
to arid plains. Their qanats were built on a scale rivaling the great Roman
aqueducts. The system has since been used in various places from Pakistan to
North Africa. In Afghanistan and Pakistan they are called karezes, in North
Africa, foggaras, and in the United Arab Emirates, falaj. Though new qanats
are seldom built today, many old ones are still used, especially in Afghanistan
and in Iran, where there are some 40,000 qanats comprising more than
270,000 km of underground channels that supply 35 percent of the country's
In Iran areas with only 150-250 mm of annual rainfall grow their own
food and even produce cotton, dried fruits, and oilseeds for export-
achievements made possible by qanats. Until recently (before the Karaj Dam
was built), the 2 million inhabitants of Teheran tapped the foothills of the

24Government of India, Ministry of Food and Agriculture. 1962. (See Selected

- Farmland

FIGURE 34 Exploiting the gradient of the natural terrain, qanat system is a series of wells dug into an underground water supply and
connected by underground channels that convey water by gravity to the ground surface at lower levels. (Based on a diagram in Scientific
American, Wulff, 1968. See Selected Readings.)


FIGURE 35 Row of craters, each marking the mouth of a qanat ventilation shaft,
across an arid plain in western Iran. Crater walls protect the shafts and the tunnel below
from erosion caused by the inflow of water during a heavy desert rainstorm. (FAO)

Elburz Mountains with qanats for their entire water supply. The agricultural
production made possible by qanats repays the investment cost for construc-
tion and maintenance. In 1967 the return on investments from the sale of
water and crops ranged from 10 to 25 percent per annum depending on the
size of the qanat, the yield of water, and the crop.25
A recent innovation now used in Iran is a hybrid between a dug well and a
qanat. A dug well is excavated to below the water table and then horizontal
galleries are bored out, using the excavating methods of the qanat builders. In
the dug-well shaft a centrifugal pump is then installed to pump to the surface
the water collected by the horizontal galleries.

Qanats have limitations:
They usually flow continuously and year-round, so unused water is
wasted. Flow is maximum during the rainy season when the demand for
irrigation is least and minimum during the major irrigation period (summer).
Qanats may dry up altogether in drought years.
They serve the lower elevations of alluvial fans, which tend to have
more saline and poorer soils than do the higher elevations.

25Wulff. 1968. (See Selected Readings.)


FIGURE 36 Water flowing out the main qanat of Dusadj village, Iran, is used mainly
for irrigation. A qanat can deliver water in otherwise very arid terrain. (FAO)

Qanat water is often of poorer quality than water from wells drilled
higher in the alluvial fan.
Qanats are expensive and dangerous to build by the primitive hand-
tunneling methods of the past, and in recent years construction costs have
increased along with rising standards of living and labor costs. However, if
modem engineering, geology, hydrology, and remote tensing are applied, the
qanat principle could play a role in future water production in arid lands.
Research is needed on safe construction methods, on qanat linings that
increase safety and decrease maintenance, and on ways to shut off the water
flow when it is not needed.

-~, ;.
~r.rE; ~.....


FIGURE 37 A completed horizontal well on test yields over 50 gpm. Plumbing head
includes two tees, a shutoff valve to regulate the flow, a vacuum relief, and a pipe
reducer. (W. T. Welchert)

FIGURE 38 Drilling a horizontal well. A standard well casing is drilled on a slight
downward slope. Equipment is light and portable and easily transported even in remote,
rough terrain. (W. T. Welchert)


Good sites for
horizontal wells
are dike forma-
tions, impervious
geologically tilt- Water Table
ed clay, or rock
walls that form
a natural dam. Horizontal
(W. T. Welchert) Wel

Horizontal Wells

In the development of water supplies, small springs are often neglected.
Yet in many remote and arid mountain regions, springs are the safest and
most dependable source of water for domestic use. The horizontal well
system, an improved spring-development process, has great potential for
providing and conserving sanitary water in geologically appropriate areas.
A horizontal well is a "cased" spring (FIGURE 37). A horizontal boring
rig (FIGURE 38) is used to drill a hole and install a steel pipecasing into a
mountain or hillside to tap a trapped water supply (FIGURE 39).
Tapping water from springs is an ancient art. Conventionally, when a seep
or spring is located, it is either dug or dynamited to expose the water-bearing
rock. Results are erratic and always carry a risk of damaging the natural
barrier that dams the underground reservoir. The flow, once established this
way, is almost impossible to control and may result in rapid depletion of the
Horizontal wells virtually eliminate these hazards. They are drilled at
promising sites where springs, seeps, or traces of water are found. Occurrence
of phreatophytes (chapter 13), dried-up springs, and favorable geology are all
indicators used to select the drilling site. A horizontal well can tap the aquifer
with precision and safety. Furthermore, it protects against contamination by
animals, dust, erosion, etc. No pumps are needed. Maintenance costs and
problems are insignificant in comparison to those of other systems for
harnessing springs.
If the flow is very low, a storage tank can be added to accumulate water
during the night or off-season. With adequate storage, spring sites that flow
only during a few weeks in the year may be useful.


During the last 15 years about 2,000 successful horizontal wells have been
drilled in arid areas of the southwestern United States. A 1967 University of
Arizona study indicates that a serviceable water supply was obtained at 45
out of 53 locations tried during one program. Successful yields varied from
1-230 1/minute; most were in the 10-40 1/minute range. Drilling time
averaged 32.3 hours per producing well.26
Horizontal drilling equipment is currently manufactured; it is simple,
portable, and dependable. The drilling process involves a rotary, wet-boring
horizontal drill stem rig (FIGURE 38), a carbide-tipped or diamond-core drill
bit, a small recirculating water pump, a cement slurry pressure tank, a drill
water supply, and a few standard plumbing tools and supplies.
Horizontal-well drilling is quite a different technology from vertical drill-
ing. Skill, patience, and field experience are required to master it.

Selected Readings

Dug Wells
Bennison, E. W. 1947. Ground Water, Its Development, Uses and Conservation. E. E.
Johnson, Inc., 319 N. Pierce St., St. Paul. Minnesota 55104, USA.
Gibson, U. P., and R. D. Singer. 1969. Small Wells Manual. Office of Health, 633 PP,
U.S. Agency for International Development, Washington, D.C. 20523, USA. 156 p.
Government of India, Ministry of Food and Agriculture. 1962. Handbook on Boring and
Deepening of Wells. Report Number Agr. 77, New Delhi, India.
U.S. Department of Agriculture. 1971. Water Supply Sources for the Farmstead and
Rural Home. Farmers Bulletin No. 2237. 18 p. (Superintendent of Documents, U.S.
Government Printing Office, Washington, D.C. 20402, USA. U.S.$0.15.)
Wagner, E. G., and J. N. Lanoix. 1959. Water Supply for Rural Areas and Small
Communities. World Health Organization, Geneva, Switzerland.


Butler, M. A. 1933. Irrigation in Persia by kanats. Civil Engineering 3 (2):69-73.
U.S. Agency for International Development. 1973. NESA Irrigation Practices Seminars-
An Evaluation. Office of Agriculture, Technical Assistance Bureau, U.S. Agency for
International Development, Washington, D.C.
Wulff, H. E. 1968. The qanats of Iran. Scientific American 218:94-101.

Horizontal Wells
Freeman, B. N. Horizontal Wells: New Water Source. Five-minute motion film produced
by Deere and Company as a public service. Contact: Advertising Department, Deere
and Company, Moline, Illinois 61265, USA.
Welchert, W. T., and B. N. Freeman. 1973. 'Horizontal' Wells. Journal of Range Manage-
ment 26(4):253-6. Also available in Spanish as: Pazos Horizontales. Selecciones del
Journal of Range Management 2(4). Agosto 1973.

26Welchert and Freeman. 1973. (See Selected Readings.)



Dug Wells
Ministry of Agriculture, Government of India, Krishi, Bhawan, New Delhi, India
(J. K. Jain, Irrigation Adviser)
Service especial de Saude Publica, Government of Brazil, Brasilia, Brazil
University of the Negev, Beer Sheva, Israel (A. Issar)

Horizontal Wells
Cooperative Extension Service and Agricultural Experiment Station, University of
Arizona, Tucson, Arizona 85721, USA (W. T. Welchert)

6 Other Sources of Water

This chapter briefly mentions some other sources of water in arid lands
that are either more than adequately described in literature elsewhere or too
highly speculative to warrant more specific treatment in a general report.

Groundwater Mining
Many productive aquifers in arid regions of the world are storing large
quantities of water that can be tapped by well-known deep-well extractive
techniques. Aquifers of this type are widespread beneath deserts in northern
Mexico, southwestern United States, North Africa, eastern Saudi Arabia, the
Sinai, and elsewhere in southwest Asia. The water in such aquifers is not
naturally replaced once it is withdrawn; it must be considered a "wasting
asset" like any other mineral commodity. Its development must be under-
taken with the full understanding that usually within a few decades the
supply will be depleted and capital investments must be amortized within
that time. Under some conditions these aquifers may provide interim supplies
to generate capital to underwrite expensive systems that bring in a more
permanent supply of water from outside the region. Development of such
aquifers is now proceeding in several parts of Algeria, Libya (FIGURE 63),
Egypt, Saudi Arabia, Mexico, and the United States.

Low-cost desalting of seawater would be a boon to arid lands bordering
seas or salt lakes; over the past few decades many proposals to build huge
distillation plants to produce water for agriculture have been widely
advertised and intensively promoted. Although new and improved desalting
methods using membranes and ion exchange have been developed, no method
can yet promise truly low-cost freshwater. Current promises of cost reduction
for distillation plants are based on the assumption that the cost of the
product decreases as the size of the plant increases. But a practical limit
exists to cost reduction achieved by this means. Also, there are problems in


disposing of huge quantities of hot brine, pumping and conveying desalted
water to the point of use, and storing it until needed for irrigation. Another
fundamental problem is that desalting plants require large amounts of energy
to construct and operate.
Most proponents of desalting schemes now agree that the water will be too
expensive for use in irrigation as practiced today. However, desalination of
sea or brackish water could prove economically rewarding in special situations
such as tourist centers.
Distillation plants producing up to several million gallons per day are
commercially available and are already used for domestic and industrial
purposes in some very arid regions where the local economy can afford it.

Solar Distillation
In solar distillation the sun's radiation passes through a transparent cover
on to a source of brine; water evaporates from the brine; and the vapor
condenses on the cover which is arranged to collect and store it. It was first
used in Chile's Atacama Desert in 1872 in a plant supplying drinking water
for livestock used in nitrate mining. The plant reportedly operated for 30
The process today is generally in the pilot stages, though small
community-scale stills are close to extensive commercial application. Durable
designs requiring little day-to-day attention and operating with minimal
maintenance have been developed in the United States, France, Spain, and
Australia (FIGURE 40). Solar distillation is now used on a small commercial
scale to supply small towns in isolated areas of Australia and small
communities in the Mediterranean basin and the Caribbean. Modem research
into solar distillation is emphasizing new materials and designs for economical
and durable construction to reduce product-water cost.

Remote Sensing for Detecting Water
The use of photographs from satellites and high-flying aircraft is a newly
developing tool that is proving useful for planners in arid lands. The following
are a few successful uses of such remote sensing.

Open-water oases have been detected, and repetitive imagery has been
used to determine their transience.
r Geological information for determining the shape and extent of under-
ground aquifers has been obtained from most types of remotely obtained
imagery, including radar, thermal, and optical images from aircraft or


FIGURE 40 Solar still at Caiguna, Western Australia. Waste heat from a nearby
internal combustion engine supplements the solar heat, markedly increasing the still's
efficiency. (Commonwealth Scientific and Industrial Research Organization, Highett,
Victoria, Australia)

satellites. Airborne geophysical sensors, such as magnetometers, are also use-
ful in mapping underground structures that control groundwater movement
and distribution.
The extent of springs, seeps, and shallow groundwater in alluvial
channels is often mappable, even from satellites, because trees and shrubs
associated with moisture can be detected.
Effects of rainstorms on the desert can be detected; after a storm, the
darker tones of the moist soil can be seen, and a few weeks later vegetation
responding to the moisture is revealed.
Plumes of freshwater from offshore submarine springs can be detected
with the use of thermal scanners because they usually differ in water tempera-
ture from the surrounding sea.
The extent and movement of floodwaters have been mapped.
The snow cover on mountains has been determined in order to estimate
the amount of runoff water that will be available at lower altitudes after the

The following topics are included solely to give some cautions and a sense
of the range of current speculation about water supplies for the distant


Rainfall Augmentation

Certain cloud formations contain supercooled water; rainfall augmentation
hastens precipitation of this water. Adding ice, frozen carbon dioxide, and
silver iodide-whose crystal shapes promote condensation (nucleation) of
supercooled water-causes dramatic ice-crystal formation and produces rain.
This method is known as cloud seeding.
To produce rain by cloud seeding, one must always have the right
meteorological conditions. Even then the seeding "can sometimes lead to
more precipitation, can sometimes lead to less precipitation, and at other
times the nuclei have no effect."2 7
Though there is keen interest in cloud seeding in arid lands, experience
indicates that its best opportunities for increasing precipitation are in
areas where cold, wet air masses are swept upward over mountain ranges.
Prospects for increasing precipitation over low-lying arid lands do not seem
promising, primarily because of the scarcity of water-rich clouds. Arid lands
that benefit from cloud seeding will probably be those fed by streams
originating in mountains.
The results of cloud seeding are difficult to predict because of the still
imperfect knowledge of the physical processes causing precipitation and
because of engineering difficulties in getting seeds into clouds in optimum
amounts and at the right time and place. A related uncertainty is whether
seeding clouds in one area modifies precipitation in another. Detailed physical
analysis of some cloud systems may, in the future, allow one to predict the
effects of cloud seeding, but research is in the very early stages.

Eighty-five percent of the world's freshwater is trapped as ice in the polar
regions, but it is generally considered unusable. Engineers, glaciologists, and
physicists are now speculating on whether it could be profitably recovered by
towing icebergs to water-short regions. According to some, "the idea appears
both technically feasible and economically attractive and merits serious
consideration."28 Data on size and distribution of icebergs indicate that the
supply is more than adequate, and satellites can be used to select suitable
icebergs. Prime producing sites in the Antarctic could supply icebergs for
Australia, the Atacama Desert in Chile, and other arid regions of the
Southern, and perhaps even the Northern, Hemisphere.

27Committee on Atmospheric Sciences, National Academy of Sciences-National
Research Council. 1973. (See Selected Readings.)
2Weeks and Campbell. 1973. (See Selected Readings.)

The icebergs are there; the problem is to move them. According to one
report,2" a hypothetical supertug, which could be built with present
technology, could tow icebergs up to 16 km long, 3.5 km wide, and 200 m
thick. On delivery, the water in such icebergs would be worth hundreds of
millions of dollars.
Melting losses are important because transit times at reasonable towing
speeds may exceed 100 days and water temperatures at the delivery sites
exceed 150C. Unfortunately, detailed calculations and adequate field observa-
tions to accurately predict towing speeds and melting rates still need to be
Two inherent problems are getting the ice to melt at the end of the tow,
and pumping the water from sea level into the supply system. Perhaps the ice
could provide a "cold sink" for coastal powerplants; thus, waste heat would
be utilized for melting ice and the thermodynamic efficiency of the
powerplant increased.
As with other water-resource exploitation proposals, critical technology
assessment must precede any decision that iceberg harvesting is a realistic,
desirable possibility. In particular, iceberg-harvesting proposals raise interna-
tional legal and political questions concerning resource rights-matters of
uncertainty in the delicate Antarctic Treaty regime and matters of very
intense negotiation in the emerging restructuring of the law of the sea.
Additionally, as the importance of the polar regions to world climate has
become increasingly evident, it has become equally evident that not enough is
known of the physical basis of climate to permit confident prediction of the
extended effects of even apparently small modifications of the polar ice

Dew and Fog Harvesting
The possibility of condensing water from the atmosphere by some simple
scheme has intrigued several investigators. Some have suggested that ancient
civilizations accomplished it for agricultural purposes. "Dew mounds" in the
Negev (FIGURE 4) and the ancient "aerial wells" of Theodosia in the Crimea
were piles of rocks that supposedly cooled during the night and condensed
the early morning dew. Experiments have been unable to show that this
process occurs to any significant extent, and the supposed dew mounds of the
Negev have now been shown to result from soil-smoothing operations to
increase rainfall runoff29 (chapter 1). Dew will condense on piles of rock, but
not in harvestable and usable quantities.

28Weeks and Campbell. 1973. (See Selecte4 Readings.)
29Evenari, Shanan, and Tadmor. 1971. p. 127. (See Selected Readings.)


Selected Readings

Groundwater Mining
United Nations, Department of Economic and Social Affairs. 1973. Ground Water in
Africa. Report No. ST/ECA/147. (UN Publication Sales No. e.71.II.A.16.) United
Nations, New York. 170 p.
Vohra, B. B. 1972. Ground Water Comes of Age: Some Policy Implications. Ministry of
Agriculture, New Delhi, India. January 31, 1972. 8 p.

Clawson, M.; H. H. Landsberg; and L. T. Alexander. 1969. The economic impractica-
bility of desalting seawater for large-scale agriculture. Science. 164:1141-8.
Fried, J. J., and M. C. Edlund. 1971. Desalting Technology for Middle Eastern
Agriculture; an Economic Case. Praeger Special Studies in International Economic
and Development. Praeger, New York. 113. p.
United Nations. 1964. Water Desalination in Developing Countries. United Nations, New

Solar Distillation
Proctor, D. 1973. The use of waste heat in a solar still. Solar Energy. 14(4):433-49.
Talbert, S. G.; J. A. Eibling; and G. O. G. Lof. 1971. Manual on Solar Distillation of
Saline Water. Prepared for Office of Saline Water, U.S. Department of the Interior.
Battelle Memorial Institute, Columbus, Ohio 43201, USA.
United Nations Department of Economic and Social Affairs. 1970. Solar Distillation as a
Means of Meeting Small-Scale Water Demands. Report No. ST/ECA/121. (U.N.
Publication Sales No. E 70.II.BI) United Nations, New York. 86 p.

Remote Sensing
Goddard Space Flight Center. 1973. Symposium on Significant Results Obtained from
Earth Resources Technology Satellite-i, Report No. X-650-73-155. (Available from
Center, Greenbelt, Maryland 20770. USA.)
International Hydrological Decade. 1969. Vol. 1. pp. 61-125. (See Selected Readings,
Thomson, K.; R. Lane; and S. Csallany, eds. 1973. Remote Sensing and Water Resources
Management. American Water Resources Association (Box 434, Urbana, Illinois
61801, USA price U.S.$15.00).

Rainfall Augmentation
Committee on Atmospheric Sciences, National Research Council. 1973. Weather and
Climate Modification: Problems and Progress. National Academy of Sciences,
Washington, D.C. 258 p. (U.S.S6.25)
Center of Scientific and Technological Information. Artificial Rainfall Newsletter. A
monthly review with abstracts from technical literature. Available from the authoring
institution, P.O. Box 20125, Tel Aviv, Israel (U.S.$18.00/year)


Hult, J. L., and N. C. Ostrander. 1973. Antarctic Icebergs as a Global Fresh Water
Resource. Report No. R-1255-NSF. National Science Foundation, Washington, D.C.,
Hult, J. L., and N. C. Ostrander. 1973. Applicability of ERTS for Surveying Antarctic
Iceberg Resources. Report No. R-1354-NASA/NSF. Goddard Space Flight Center,
Greenbelt, Maryland 20770. USA.
Weeks, W. F., and W. J. Campbell. 1973. Icebergs as a fresh water source: an appraisal.
Journal of Glaciology. 12:65.

Dew and Fog Harvesting
Evenari, M.; L. Shanan; and N. Tadmor. 1971. The Negev: The Challenge of a Desert.
Harvard University Press, Cambridge, Massachusetts, USA. pp. 126-47.
Gindel, I. 1965. Irrigation of plants with atmospheric water within the desert. Nature.

Part II

Water Conservation

7 Reducing Evaporation

from Water Surfaces

Reservoirs and canals in arid lands are subject to heavy evaporation losses,
but because evaporating water is invisible these losses are often not
recognized. From small reservoirs, stock tanks, and farm ponds with large
surface areas open to the air (compared to the volume of water stored)
evaporation losses often exceed the amount of water used productively.
Reducing evaporation is an important way to increase the supply of water.
It increases reservoir capacity without new construction; in arid regions it
may mean the difference between a dry reservoir and a filled one.


Generally, the method for reducing evaporation has been to cover the
water surface with a barrier that inhibits vaporization. On small tanks, a cover
or roof is an obvious choice, but for ponds and larger reservoirs the solution is
less simple. For the latter the methods tested include the following: liquid
chemicals that automatically spread out, forming a sealant layer across the
surface; blocks, rafts, or beads that float on the water surface and reduce the
area where vaporization can occur; and storing the water in sand- and
rock-filled dams.

Liquid Chemicals
Aliphatic alcohols, e.g., cetyl alcohol, are long slender molecules that align
themselves side-by-side on a water surface, covering it to form a film 1
molecule thick. Considerable research and publicity have been devoted to the
possibility of using such films to reduce evaporation from water surfaces.
Unfortunately, the films do not reduce the amount of solar energy the water
absorbs, and they decrease the amount of heat normally lost from the water
because inhibiting evaporation also inhibits the cooling effects of vaporiza-
tion. Although evaporation decreases where the alcohol layer is intact, the
higher water temperature increases evaporation at any part of the water


FIGURE 41 An experiment in Arizona, USA, using wax to suppress evaporation.
Blocks of wax have been added to the tank (foreground). Sun's heat melts the wax to
form the continuous film visible on the tanks in the background. During 4 years this film
has suppressed over 85 percent of the normal evaporation. (K. C. Cooley)

surface the barrier does not cover. Furthermore, an intact alcohol barrier is
impossible to maintain because of wind and wave action. These problems
make the method impractical today.
The aliphatic alcohol molecular monolayer is, nonetheless, a tempting and
potentially rewarding concept: it requires only small amounts of materials
(less than 60 g per ha of water surface); it reportedly does not restrict the
transmission of oxygen to the water; and it consists of materials that are
nontoxic to fish or humans. Attempts are now under way to circumvent the
problem of maintaining a continuous film on the water surface. One approach
uses a plastic net to restrict the drift and disruption of alcohol layers.


Wax is an unusual, recently tested evaporation suppressant. Floating
blocks of wax are added to the water; in sunlight they soften and flow to
form a flexible, continuous film. In Arizona, a wax cover on a small tank
(FIGURE 41) is still in good condition after 4 years; the evaporation-
suppressing efficiency is over 85 percent. Even if the film cracks and breaks
during cold weather, the sun's heat subsequently re-forms it.


=_.-.. w .--. jf -.

FIGURE 42 Protective covering of molded, lightweight, concrete slabs cut evaporation
by an estimated 80 percent recently in full-scale tests on two reservoirs in Ovamboland,
South-West Africa. Engineers cover reservoirs with 24-in2, 2-in thick floating slabs of
polystyrene, sand, and concrete. Exposed surface of each slab is painted white to reflect
the intense sun rays, which hasten evaporation.
The specific gravity of the material (0.8) keeps about 80 percent of the floating slab
submerged, reducing the possibility of piling up in heavy winds. Submerged portion of
the slab is coated with bitumen for durability and to prevent the material from affecting
the taste of the water. Slabs have rounded comers to permit the large body of water
underneath to breathe properly when covered with the densely packed slabs. (National
Institute for Water Research, South Africa)

Solid Blocks

Floating materials, covering the water surface, reduce the area where
evaporation can occur. Blocks of lightweight concrete, polystyrene, wax,
rubber, and plastic are under trial as evaporation retardants. Floating concrete
blocks, made with lightweight aggregates, have been used to reduce
evaporation from a 10,000 m2 reservoir in South-West Africa30 (FIGURE
To overcome the water heating that is an inherent problem in evaporation
suppression, researchers are working with floats made of insulating and
light-colored reflecting materials that prevent solar energy from entering the
water. For example, sheets of inexpensive and highly insulating expanded-
polystyrene, 2.5 cm thick and coated with asphalt and gravel have been

30Concrete slabs cut reservoir losses. 1966. (See Selected Readings.)


FIGURE 43 Floating foam-rubber sheet (9 m diameter, 5 mm thick) covers a water-
storage tank near St. George, Utah, USA. Experience to date suggests that the cover will
last 10 years with an evaporation-control efficiency of 80-90 percent. Estimated cost of
the water saved in a 120-cm/year evaporation zone is US$1.80-$2.00. (U.S. Department
of Agriculture)

tested. Coupled together with inexpensive clamps to form large rafts up to
160 m2, they can form excellent energy and vapor barriers. (See FIGURES 24,
p. 33; and 43.) Foamed butyl rubber can also be effective; though expensive,
it may last for over 10 years.

Sand-Filled Reservoirs
Evaporation can be controlled by filling reservoirs with sand and loose
rock. Water is stored in the pores between the particles, and the water level is
kept more than 30 cm below the surface to shield it from evaporation (see
gravel mulches, chapter 9). Recently, two small, plastic-lined tanks were built
near Safford, Arizona, and filled by commercial rock-picking machines. The
rocks reduced the tank's volume by 55 percent, but they reduced evaporation
by 90 percent.31 Another related technology is the sand-filled water storage
tank developed in the Sudan for use with rainwater harvesting (FIGURE 11).
Small sand-filled dams have been used in the Namib Desert since 1907 for
supplying drinking water to livestock. They can store water for long periods,
much longer than conventional open storage. They can provide water during
31 Cluff, et al. 1972. (See Selected Readings.)

, N_'-ft t*c~~~ i


FIGURE 44 Sand-filled storage dam in South-West Africa built of concrete. Staged
construction is clearly visible. Well shaft behind the dam wall is used to extract the
water. The dam size can be judged from the persons at right. (0. Wipplinger)

years of total drought; when the water table is 1 m below the sand surface,
evaporation ceases for all practical purposes.32 The water is drawn off by a
drainage pipe through the dam wall or by a well dug into the sand (FIGURES
44 and 45).
The dam wall is built across the riverbed during the dry season; later, the
river floods will deposit the necessary sand and gravel. Normally, the soil
carried by flood waters is 3/4 sediment and mud and only 1/4 sand and
gravel. A dam trapping all of this mixture would quickly silt up. To ensure
that only sand and gravel are deposited, the dam wall is heightened in stages
of only 1 m (though the first is usually about 2 m). Floodwaters deposit
heavy gravels and sand, but silt and soil are carried over the top by the speed-
ing waters. Each 1-m stage is added when the dam is filled with sand and
gravel (which may require a complete wet season) until the operating height
of 6-10 m is reached.
Sand dams are particularly effective when built over fissures that lead
underground to natural aquifers, for the dam can slow rushing floodwaters
enough to get the aquifer recharged (chapter 16).
A simplified approach to this method is to use a pump and a well-shaft
system that can be readily sunk into dry riverbeds to extract water stored
naturally in the sand a few feet below the surface.33

32Wipplinger. 1958. (See Selected Readings.)
33Ball and van Rynveld. 1972. (See Selected Readings.)


FIGURE 45 Sand dam in South-West Africa built of rock and mortar. (0. Wipplinger)


Evaporation control is a particularly important way to conserve water
because it usually requires little new construction and the additional water
becomes available without construction delays. In many cases it will cost less
to reduce evaporation than to collect and store an equivalent amount of
water from other sources. Some of the better materials have provided water
for less than U.S.$0.025 per m3 in a 200-cm-per-year evaporation zone.31
Evaporation is greatest during the driest seasons, which are also the peak
periods for water use. Controlling only dry-season evaporation with short-
lived methods could have economic significance in arid lands.
Suppressing evaporation in impounded waters also suppresses the increase
in salt concentration that occurs with evaporation. Floating evaporation-
control materials cut off the light, thereby reducing the growth of undesirable
algae and submersed aquatic weeds.


At present, evaporation suppression is limited to small storage facilities
such as ponds, tanks, troughs, and oases. In practical terms large reservoirs,

31 Cluff, et al. 1972. (See Selected Readings.)


lakes, and rivers are still beyond the reach of available technology because it
is very difficult for the evaporation-suppressing system to survive heavy
winds, storms, and floods.
The effects of evaporation-control methods on animal life in the water
may not be important in small containers, but they should be considered in
Sand storage dams can be built only where the geology permits. The
floodwaters must contain gravel and sand (coarse or fine granite, quartzite,
microschist, and dune sand all work well), and the dam site must be made
absolutely watertight (e.g., cement-grouted to stop seepage if necessary). To
build a sand dam requires patience, time, and logistics that can cope with
delays while each stage of the dam wall is built, because often only one stage
can be added each year. For this last reason the technology has not yet been
widely accepted.
Seepage control (chapter 8) is usually easier and cheaper than evaporation
control; therefore, evaporation control should not be considered for a
reservoir or water-distribution system until seepage is controlled.

Stage of Development

Hydrologists and engineers have long been aware of the quantity of water
in arid lands lost each year by evaporation, but evaporation suppression is still
experimental. By far the greatest number of studies have been concerned with
alcohol films. Although research conducted on the use of solid floating
materials is limited, progress has been considerable. This method appears to
offer the most promise for small storage. Sand dams have been built by the
score in South-West Africa during the past 50 years. A few have also been
built in Kenya and other countries of East Africa.

Needed Research and Development

No economical method exists for reducing evaporation from large,
multipurpose reservoirs; research in this area is desperately needed.
Extended field trials are needed to test the practicality of the floating
covers, such as expanded-polystyrene rafts or wax layers on water reservoirs.
These do not require any special equipment or skills and can be readily
introduced into arid lands for pilot testing.
Research is needed to overcome the mechanical difficulties of stabilizing
any evaporation-control system on water surfaces subject to wind, wave, and
currents. This is particularly so for the alcohol monolayers.


It is likely that some radically different approaches to evaporation control
await discovery; research into novel methods is encouraged.
The sand storage dam is a technology that needs field testing in many arid
areas. Research is needed into its rational design, e.g., for the height of stages
in relation to the extent of the catchment.

Selected Readings
Ball, J. S., and J. A. van Rynveld. 1972. The development of water-supplies from sand
riverbeds. In Rhodesia Agricultural Journal, 1972. Water in Agriculture, pp. 27-36.
(See page 6.)
Cluff, C. B. 1966. Evaporation reduction investigations relating to small reservoirs.
Technical Bulletin. 47 p. Agricultural Experiment Station, University of Arizona,
Tucson, Arizona 85721, USA.
duff, C. B.; G. R. Dutt; R. R. Ogden; and J. K. Kuykendall. 1972. (See page 21.)
Concrete slabs cut reservoir losses. 1966. Engineering News-Record 177(15): 140.
Cooley, K. R. 1970. Energy relationships in the design of floating covers for evaporation
reduction. Water Resources Research. 6(3):717-25.
Cooley, K. R., and L. E. Myers. 1973. Evaporation reduction with reflective covers.
Journal of the Irrigation and Drainage Division of the American Society of Civil
Engineers. 99:353.
Frasier, G. W., and L. E. Myers. 1968. Stable alkanol dispersion to reduce evaporation.
Journal of the Irrigation and Drainage Division, American Society of Civil Engineers.
Ionides, M. G. 1967. Water in dry places. Engineering (London, England) 27(Oct):662-6.
(Describes sand-filled reservoirs and use of plastic-tube revetments.)
Roberts, W. J. 1969. In International Hydrological Decade Vol. 2: 666-93. (See page
Wipplinger, O. 1958. The Storage of Water in Sand. South-West Africa Administration.
Windhoek. (Describes sand dams.)


Department of Agricultural Engineering, Oklahoma State University, Stillwater, Okla-
homa 74074, USA (F. R. Crow)
Director of Water Affairs, Kaiser Street, Windhoek, South-West Africa.
U.S. Water Conservation Laboratory, Agricultural Research Service, U.S. Department of
Agriculture, Phoenix, Arizona 85040, USA (K. R. Cooley)
University of New South Wales, Kensington, New South Wales, Australia (G. J. Wiesner)
Victorian State Rivers and Water Supply Commission, Victoria, Australia
Water Resources Research Center, University of Arizona, Tucson, Arizona 85721, USA
(C. B. Cluff)

Sand Storage Dams
Department of Civil Engineering, University of Stellenbosch, Stellenbosch, South Africa
(0. Wipplinger)
Department of Conservation and Extension, Ministry of Agriculture, P.O. Box 8108,
Causeway, Salisbury, Rhodesia (J. S. Ball)
Hydraulic Research Laboratory, C.S.I.R., P.O. Box 395, Pretoria 0001, South Africa
Illinois State Water Survey, Urbana, Illinois 61801, USA (W. J. Roberts)

8 Reducing Seepage Losses

For economic reasons most arid lands are forced to use earthen canals and
reservoirs. Because their soils are often porous, many of these storage
facilities and conduits suffer serious water losses through seepage. In diverting
water from its intended purpose, seepage can also cause serious waterlogging,
salinification, and erosion of neighboring soils.
Seepage can be reduced if the walls of reservoirs and conduits are made
watertight. Recent technology has produced many inexpensive, waterproof
materials that may prove valuable for this purpose.
Of the various ways to reduce seepage, only some of the newer, low-cost
methods are discussed in this report.


Many ways to make soils impervious for rainwater harvesting (see list in
chapter 1) can also be used to reduce seepage in waterways-soil compaction,
chemical treatment of soil, and soil covers such as butyl rubber, sheet plastic,
asphalt reinforced with plastic or fiberglass, and ferrocement.34
Much seepage is caused by calcium in the soil. Calcium causes clay to
bunch up (aggregate), forming cracks and a porous structure that lets water
seep through easily. In this situation seepage can be greatly reduced by
treating the soil with a sodium salt such as sodium carbonate (FIGURE 46).
Sodium breaks up clay aggregates (FIGURE 47) and causes clay particles to
swell and plug the soil pores (FIGURE 48). This method can be successful
only where conditions are favorable (e.g., the soil must have a minimum of 15
percent clay, be at least 30 cm deep, and have the chemical capacity to
exchange calcium for sodium ions.)
Shallow earthen, but rock-free reservoirs (less than 3 m deep) can be made
watertight with low-cost polyethylene and polypropylene films, but deeper
reservoirs or those built on stony soils require thicker and tougher films of
vinyl or reinforced polypropylene. In both cases the films should be
34See, for example, Ferrocement: Applications in Developing Countries. (Available
without charge. See publication 8, p. 152.)


FIGURE 46 Cracked soil structure indicates high calcium content in a dry pond bed in
Coconino National Forest, Arizona, USA. Water seeps away through the cracks; original
seepage loss in this 1,000 m2 stock-watering pond was 5-12 cm per day. The dry pond
was cleared of rocks and weeds and 1 ton (about US$80 worth) of sodium carbonate
applied by hand. This was then mixed with the soil to a depth of about 8 cm with a
small tractor and disk, and ...


FIGURE 47 ... in this soil from the same pond after treatment sodium carbonate has
caused the clay aggregate to break down into fine particles that retard seepage....

protected with a covering of soil or gravel. In Europe and the United States
butyl rubber is increasingly used to line reservoirs, water channels, and
storage tanks. It is strong, durable, weather and pest proof, but it is an
expensive way to reduce seepage.
Concrete-based materials, such as cement-stabilized soil, ferrocement, and
concrete-filled fabric also have potential.
One problem encountered in lining reservoirs is stabilizing the banks. Soil
is usually unstable on slopes greater than 1:3. Steeper slopes are desirable,
however, to maximize storage, minimize evaporation losses, and inhibit
weeds. Promising methods of providing a revetment to protect steep banks
include the use of used rubber tires, ferrocement, and soil cement packed into
sausage-shaped plastic bags (FIGURES 10-12, pp. 19-20). Near-vertical slopes
can be constructed with sausage revetments. Local soil is blended with a small
amount of cement and moistened through pinholes made in the plastic tube.
The sausages are stacked in place tightly and allowed to set.3 This technique
is also suitable for canal and ditch revetments.


Any reduction in seepage provides additional water without requiring new
equipment or facilities, and most methods do not interfere with use of the
3SIntermediate Technology Development Group, Ltd. 1969. (See Selected Readings.)


FIGURE 48 ... Seepage then dropped to 0.4 cm per day. After a "booster shot" of
200 kg of sodium salts (33 months after the initial treatment) this low seepage rate has
been maintained in the pond for 5 years. (U.S. Department of Agriculture)

impoundment for recreation, fishing, etc. In some areas reducing seepage
prevents associated problems such as waterlogging and salinification of the
surrounding soil. Lining canals also reduces maintenance and weed-control
In general, reducing seepage losses in reservoirs and conduits is easier and
more economical than reducing evaporation losses (chapter 7).


The primary disadvantage of seepage control is its cost.
In most methods maintenance of the lining is a constant concern
because even small holes can allow large amounts of water to drain away,
especially if the surrounding soil is porous.

36For a discussion of aquatic weed problems see publication 11, p. 153.


Stage of Development

Seepage control has been practiced in arid lands since the beginnings of
civilization. However, most of the techniques discussed in this chapter-
including the use of sodium salts, plastic sheeting, and butyl rubber-have
been used within the last 20 years and are commercially available. Reinforced
asphalt has been under test for approximately 10 years. The use of
inexpensive polypropylene is a recent development, not yet extensively tested
or utilized. Ferrocement, virtually untested for this purpose but known to be
watertight, holds great promise because it can be constructed by unskilled
labor using materials available in developing countries.

Needed Research and Development

There is a major need for widespread field trials, particularly in arid
developing countries, to test and compare the effectiveness of different
systems and consider the economics of their application.
New, more economical materials are needed, for they could make seepage
control possible worldwide. As new watertight membrape materials become
available, their use in this application should be evaluated.
Improvements in sealing and laying underground moisture barriers
(chapter 12) could bring major breakthroughs; a watertight seal could be laid
directly in the soil where it is protected from mechanical damage and

Selected Readings
Board on Science and Technology for International Development. 1973. Ferrocement:
Applications in Developing Countries. National Academy of Sciences. Washington,
D.C. (See publication 8, p. 152.)
Boyer, D. G., and C. B. Cluff. 1973. An evaluation of current practices in seepage
control. Hydrology and Water Resources in Arizona and Water Resources in Arizona
and the Southwest 2:125-51. Arizona Section of the American Water Resources
Association, Tucson, Arizona. (Complete volume available from Water Resources
Research Center, University of Arizona, Tucson, Arizona 85721, USA.)
Intermediate Technology Development Group, Ltd. 1969. The Introduction of
Rainwater Catchment Tanks and Micro-irrigation to Botswana. 74 p. (For ITDG
address, see Contacts.)
Kraatz, D. B. 1971. Irrigation canal lining. Irrigation and Drainage Paper Number 2.
Water Resources and Development Service, FAO, Rome. 170 p.
Myers, L. E., ed. 1963. Seepage Symposium, Phoenix, Arizona, Proceedings. Agricultural
Research Service Report A.R.S. 41-90. U.S. Department of Agriculture, Washington,
D.C., USA. 180 p.
Reginato, R. J.; L. E. Myers; and R. S. Nakayama. 1968. Sodium Carbonate for
Reducing Seepage from Ponds. Report Number 7, U.S. Water Conservation Labora-
tory, A.R.S., U.S. Department of Agriculture, Phoenix, Arizona, USA. 6 p.


Second Seepage Symposium, Phoenix, Arizona, Proceedings. 1968. Agricultural Re-
search Service, Report Number A.R.S. 41-147. U.S. Department of Agriculture,
Washington, D.C., USA. 150 p.
The University of Arizona, Cooperative Extension Service and Agricultural Experiment
Station. 1965. How to Make a Plastered Concrete Water-Storage Tank. Bulletin A-41.
Tucson, Arizona, USA. 12 p.

Plastics (and agricultural) supply houses in some countries should have in stock
materials suitable for lining ponds and conduits. Some agricultural supply houses also
carry sodium and phosphate salts for pond sealing. The following are involved in research
Bamangwato Development Association, Radisele, Botswana
Doxiadis lonides Associates, Ltd., Ripley, Surrey, England
Intermediate Technology Development Group, Ltd., Parnell House, 25 Wilton Road,
London SW1V 1JS, England
Water Resources and Development Service, Land and Water Development Division, Food
and Agriculture Organization of the United Nations, Via delle Terme di Caracalla,
00100, Rome, Italy
Water Resources Research Center, University of Arizona, Tucson, Arizona 85721, USA
(C. B. Cluff)

9 Reducing Evaporation

from Soil Surfaces

Evaporation from soil surfaces wastes large amounts of water-an
important consideration in arid lands where low humidity greatly encourages
evaporation. From one-fourth to one-half of the water lost from a crop is
evaporated from the soil surface.37
This loss can be reduced and irrigation water saved by placing watertight
moisture barriers or water-retardant mulches on the soil surface.38 In many
cases these barriers will also stabilize loose soils, stop desert encroachment,
allow runoff agriculture, aid in landscaping, or reduce salinity buildup.
Suppressing evaporation from the soil conserves water where its effect is
great: within the root zone of the plant. In an arid region small water savings
here may be more important to a crop's survival than large improvements at
earlier points in the water supply.


Some soil-surface moisture barriers are made of nonporous materials such
as paper, asphalt, latex, oil, plastic film, or metal foil, but a 5-25 mm thick
layer of porous material can also substantially reduce evaporation. Water and
water vapor move so slowly through a dry, porous material that soil moisture
is retained. In practice, suitable porous materials are plant residues such as
straw, sawdust, wood bark, or cotton burs, as well as gravel, sand, or cinders.

Plant Residues
Planting directly into the standing residue of the previous crop is one way
to retard evaporation (and reduce erosion) because the residues and hard soil
surface provide a better moisture barrier than the loose surface left after

3Viets. 1966. p. 270. (See Selected Readings.)
3Also by creating windbreaks of trees, fences, or taller growing plants-a well-known
and highly site-specific method not dealt with in this report. (see also FIGURE 50)

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs