water harvesting - doc-developpement-durable.org€¦ · key words:water harvesting, drier...

Water HarvestingIndigenous Knowledge for the Future of

the Drier Environments

Theib OweisDieter Prinz

Ahmed Hachum

International Center for Agricultural Researchin the Dry Areas (ICARDA)

The authors: Theib Y. Oweis is Senior Irrigation and Water Management Scientist at theInternational Center for Agricultural Research in the Dry Areas (ICARDA), Dieter Prinzis Professor of Agricultural Engineering at the University of Karlsruhe, Germany, andAhmed Y. Hachum is Professor of Irrigation and Drainage Engineering at the Universityof Mosul, Iraq.

Recommended Citation: Oweis, T., D. Prinz and A. Hachum. 2001. Water Harvesting:Indigenous Knowledge for the Future of the Drier Environments. ICARDA, Aleppo,Syria. 40 pages.

Key words: water harvesting, drier environments, water scarcity, micro-catchment,macro-catchment, desertification, afforestation, runoff farming.

Front CoverTopThe traditional Jessour water-harvesting system built to support palm trees, olives andfigs in the dry areas of southern Tunisia.

BottomAn indigenous water-harvesting system, which still provides drinking water for peopleand livestock in the WANA region.

Back CoverTopIn Western Sudan, under conditions of extreme water scarcity, water harvesting is themain source of water for human and live stock consumption.

MiddleEyebrow system on fig trees concentrate runoff water and help to overcome the long, drysummer in North West Egypt.

BottomA small runoff-basin system supporting shrubs, at a demonstration site for micro-catch-ment water harvesting at the ICARDA research station, Tel Hadya, Syria.

All photos by T. Oweis except where indicated otherwise.

ICARDAP.O. Box 5466, Aleppo, Syria. Tel.: (963-21) 2213433, 2225112, 2225012. Fax: (963-21) 2213490/2225105/2219380. E-Mail: [email protected] site: http://www.icarda.cgiar.org

Foreword

The steppe, known as al-badia in Arabic, covers vast areas of land in West Asia and NorthAfrica (WANA), and is characterized by harsh climatic conditions. People who live in thesteppe are economically disadvantaged, being totally dependent on the availability ofgrazing for their livestock—their main source of income. Drought and overgrazing areseriously eroding the capacity of the steppe to support the livestock feeding. As a result,migration from the steppe to urban areas is increasing, but this entails social and environ-mental problems. At the same time, production of livestock and therefore of meat anddairy products is declining, which has implications for their price in the market and forhuman nutrition.

In view of these realities, the importance of supporting research and development inthe drier environments cannot be underestimated. The cost of doing nothing, and so let-ting this cycle of degradation and migration continue, cannot be predicted using conven-tional economics, but is undoubtedly very high. If people's living standards are to beraised, and the migratory trends reduced, it is essential to improve the management ofnatural resources in the dry areas.

The most important natural resource in the drier environments is rainfall. Despite itsscarcity, rainfall is generally poorly managed, and much of it is lost through runoff andevaporation. Capturing the rainwater and making an efficient use of it is crucial for anyintegrated research and development project. Water harvesting can play an importantrole in meeting the objectives of such projects.

Water harvesting is not a new concept; indeed, we have inherited a wealth of indige-nous knowledge about this ancient practice. ICARDA (International Center forAgricultural Research in the Dry Areas) recognizes the importance of this knowledge andhas developed a project to blend the indigenous with modern knowledge for the benefitof the WANA region. The project is exploring the potential of different techniques, andadapting them to local conditions. Known as "On-farm Water Husbandry in West Asiaand North Africa," this project covers 11 WANA countries.

This publication distils the knowledge and experience of water harvesting gained byICARDA, national research groups, and advanced institutions over the years and pre-sents them in a non-technical language. Although it emphasizes the techniques most suit-able for the steppe areas of WANA, the principles outlined are applicable in dry areasworldwide.

We hope this publication will prove useful to decision makers in national govern-ments, national researchers, donor agencies, and to other stakeholders in agriculturalresearch and development.

Prof. Adel El-BeltagyDirector General, ICARDA

iii

Contents

Foreword iii

IntroductionThe Challenge of Dry Areas 1Purpose and Scope of this Document 2

Principles and Benefits of Water Harvesting 3What is Water Harvesting? 4An Ancient Practice 4How Can Water Harvesting Help? 5Components of Water-Harvesting Systems

Overview of Water-Harvesting SystemsMicro-catchment Systems 6

On-Farm Systems 7Rooftop Systems 14

Macro-catchments and Floodwater Systems 14Wadi-bed Systems 15Off-wadi Systems 18

Planning, Design and ImplementationSocioeconomic Aspects 23Designing Water-Harvesting Systems 24

Selecting the Site and Technique 24Selecting the Crops 25Designing the System 25

Implementation 28Operation and Maintenance 29

Prerequisites for Success 29

ICARDA'S Research Activities Planning Water Harvesting Using Remote Sensing and GIS 31Indigenous Water-Harvesting Systems 32Micro-catchments: Success Stories 32Water Harvesting for Supplemental Irrigation 32Water Harvesting from Greenhouse Roofs 33Runoff Inducement 34

Future Challenges 34

Selected Bibliography 36

iv

Rainfall in the drier environments isgenerally insufficient to meet basicneeds for crop production. As it ispoorly distributed over the growingseason and often comes in intensebursts, it usually cannot supporteconomically viable farming. In theMediterranean areas, for example,annual rainfall is generally less than250-300 mm and comes mainly insporadic, unpredictable storms.Even this water is mostly lost inevaporation and runoff, leaving fre-quent dry periods during the grow-ing season.

Runoff can occur even in relative-ly flat areas, where unfavorable soilsurface conditions prevent infiltra-tion. Here, most of the rain collectsas puddles, before flowing intostreams and then into swamps or“salt sinks,” where it loses qualityand evaporates; only a small portionjoins the groundwater. On its way itmay cause substantial gully erosion.

Loss of what little rainfall there iscauses severe moisture stress in

growing crops, reducing yield dras-tically if, indeed, any yield is pro-duced at all. The problem is exacer-bated by other unfavorable naturalconditions, such as extreme temper-atures during cropping periods andshallow soils of poor quality.

Although social and cultural fac-tors may be implicated, the loss ofwater without benefit for agricul-tural or domestic use, and the mis-management of land, are significantfactors in the process of desertifica-tion and the increasing poverty ofthe dry areas. For agriculture to

1

Introduction

The Challenge of Dry Areas

Typical degraded lands in the Jordanian badia; note the poor vege-tative cover.

Runoff water in the badia causes substantial erosion before beinglost in “salt sinks.”

Rainfall is mostly lost in evaporation immediately after it falls.Photo from the Syrian badia near Palmyra.

2

Water Harvesting: Indigenous Knowledge for the Future of the Drier Environments

have a chance of success, rainwa-ter must be husbanded and landmust be properly managed. Waterharvesting is the key to makingbetter use of rainwater for agricul-tural purposes: it increases theamount of water available perunit of cropping area, reduces theimpact of drought, and usesrunoff beneficially.

Purpose and Scope of this Document

As water shortage in the dry areas is a recurrent crisis, people have a great need forinformation on how to capture and use every available drop of water efficiently. Waterharvesting is an effective and economical means of achieving this objective and infor-mation on its various systems and techniques is in great demand.

Although many publications focus on water harvesting, most of them are technicalpapers in scientific journals with limited value to the average water-harvesting practi-tioner. Furthermore, each document covers only selected aspects of water harvestingfor a specific region or environment. Other documents on water harvesting are pub-lished as country reports with very limited circulation and are usually site-specific,mainly for sub-Saharan Africa.

Unlike technical papers in scientific journals, this publication covers the essentialaspects of water harvesting, concisely and simply, in a manner that responds to theneeds of ICARDA’s national research and development partners. The knowledge andtechniques most suitable and adaptable to the WANA region are presented. Also sum-marized here, are some of the findings of the adaptive research conducted by ICARDAand its national partners, and indicates future research needs. The approach adopted inclassifying, presenting and relating water-harvesting systems to field conditions isunique in the literature, and is believed to be appropriate to the nature and needs ofpractitioners in the region.

The main purpose of this publication is to increase awareness, among those workingdirectly with farmers, of the potential of water harvesting, and its technical and socioe-conomic aspects. It should also prove helpful to policy-makers in charge of developingwater resources, as well as to all those with an interest in dryland agriculture. In manyplaces previously published material is used. A list of major sources of information andreferences for further reading is provided.

Typical annual rainfall with great variability and low amounts in the drier

environments of WANA. Data forMatrouh, northern Egypt.

What is Water Harvesting?Water harvesting is based on the principleof depriving part of the land of its share ofrain, which is usually small and non-pro-ductive, and adding it to the share ofanother part. This brings the amount ofwater available to the latter area closer tocrop water requirements and thereby per-mits economic agricultural production.

For example, an area of four hectares inan arid zone receiving 150 mm of annualrain cannot normally produce an economiccrop. If two hectares forming one half ofthe area add their share of 150 mm of rainto the other half, then the latter will have atotal of 300 mm. This amount may beenough to support drought-resistant crops.Furthermore, if three hectares of the landcontribute their quantity of rain to theremaining one hectare, this quarter wouldhave a total of 600 mm, i.e. its own 150 mmshare of rain and the shares of the otherthree hectares (450 mm). If well distrib-uted, this may be enough to support quitea wide range of crops. Of course, in reality,only a portion of this water may be divert-ed easily and at low cost. Such concentra-tion of rainwater iscalled water harvest-ing, which may bedefined as “theprocess of concentrat-ing precipitationthrough runoff andstorage, for beneficialuse.”

Water harvesting may occur naturally orby intervention. Natural water harvestingcan be observed after heavy storms, whenwater flows to depressions, providingareas for farmers to cultivate.

Water harvesting by interventioninvolves inducing runoff and either collect-ing or directing it, or both, to a target areafor use. Besides being applied to agricul-ture, water harvesting may be developedto provide drinking water for humans andanimals as well as for domestic and envi-ronmental purposes.

3

The indigenous jessour sys-tem of Tunisia, using stonewalls, has supplied fig andolive trees with sufficientwater in a very dry environ-ment for hundreds of years.

Rainwater naturally harvested in depressions in MarsaMatrouh in northern Egypt, where it supports figs andolives.

Principles and Benefits of Water Harvesting

An Ancient PracticeAs long as people have inhabited the dry-lands and have cultivated crops, they haveharvested water. Ephemeral streams(wadis) and water collected in wadi bedsand cisterns, supported people’s liveli-hoods in the arid and semi-arid areasmany thousands of years ago, and allowedthe growth and development of cities.Millions of hectares of land in the dry partsof the world must once have been cultivat-ed using water harvesting, but, for a vari-ety of reasons, this practice has steadilydeclined.

The importance of West Asia and NorthAfrica in the development of ancientwater-harvesting techniques is unques-tioned. In southern Jordan early water-har-vesting structures are believed to havebeen constructed over 9,000 years ago.Evidence shows that simple water-harvest-ing techniques were used in southernMesopotamia as early as 4,500 BC.

Runoff agriculture in the Negev desertcan be traced back as far as the 10th centu-ry BC. In Yemen a system dating back to atleast 1,000 BC diverted runoff water to irri-gate 20,000 ha, producing crops that mayhave fed as many as 300,000 people. In

southern Tihama, Yemen, runoff agricul-ture is traditionally used for sorghum pro-duction. In Baluchistan, Pakistan, thekhuskaba and the sailaba systems, describedhere later, were applied in ancient times,and are still in use today.

Even in pre-Roman times, water-har-vesting techniques were applied extensive-ly in North Africa. Archeologists haverevealed that the wealth of the “granary ofthe Roman Empire” was largely based onrunoff agriculture. In Morocco’s Anti-Atlasregion a great variety of harvesting tech-niques still exists. In Tunisia water-harvest-ing systems known as the meskat, the jes-sour and the mgoud, using slopes and walls,have a long tradition and are still prac-ticed. In Egypt, the northwest coast and thenorthern Sinai areas have a long traditionin using cisterns and wadi-bed runoff culti-vation.

How Can Water HarvestingHelp?

Water harvesting is particularly advanta-geous in the following circumstances:• In dry environments, where low and

poorly distributed rainfall normallymakes agricultural produc-tion impossible. Providedother production factors suchas soils and crops are favor-able, water harvesting canmake farming possibledespite the absence of otherwater resources.• In rainfed areas, where

crops can be produced butwith low yields and a highrisk of failure. Here water-harvesting systems canprovide enough water tosupplement rainfall andthereby increase and stabi-lize production.

4


Water-harvesting reservoirs dug in the rocky mountains of Al Baida,near Petra in southern Jordan, supported civilizations several thousandsof years ago.

5

• In areas where water supply for domes-tic and animal production is not suffi-cient. These needs can be satisfied withwater harvesting.

• In arid land suffering from desertifica-tion, where the potential for productionis diminishing, due to lack of propermanagement. Providing water to theselands through water harvesting canimprove the vegetative cover and canhelp to halt environmental degradation.

The specific benefits listed above lead inturn to many other non-tangible and indi-rect socioeconomic gains. These include thestabilization of rural communities; reducedmigration of rural people to cities; use andimprovement of local skills; and improve-ment in the standard of living of the mil-lions of poor people living in drought-stricken areas.

Components of Water-Harvesting SystemsThe main components of water harvestingsystems are:

Catchment area: the part of the land thatcontributes some or all its share of rainwa-ter to a target area outside its boundaries.The catchment area can be as small as afew square meters or as large as severalsquare kilometers. It can be agricultural,rocky or marginal land, or even a rooftopor a paved road.

Storage facility: the place where runoff

water is held from the time it is collecteduntil it is used. Storage can be in surfacereservoirs, subsurface reservoirs such ascisterns, in the soil profile as soil moisture,and in groundwater aquifers.

Target area: where the harvested water isused. In agricultural production, the targetis the plant or the animal, while in domes-tic use, it is the human being or the enter-prise and its needs.

Overview of Water-Harvesting SystemsAs water harvesting is an ancient traditionand has been used for millennia in mostdrylands of the world, many different tech-niques have been developed. Most of theseare for irrigation purposes, while others areto conserve water for human and animalconsumption. The same techniques some-

times have different names in differentregions and others have similar names but,in practice, are completely different. Water-harvesting methods are classified in severalways, mostly based on the type of use orstorage, but the most commonly used clas-sification is based on the catchment size.

6


Classification of water-harvesting systems

Micro-catchment Systems

Micro-catchment systems are those inwhich surface runoff is collected from asmall catchment area with mainly sheetflow over a short distance. Runoff water isusually applied to an adjacent agriculturalarea, where it is either stored in the rootzone and used directly by plants, or storedin a small reservoir for later use. The targetarea may be planted with trees, bushes, orwith annual crops. The size of the catch-ment ranges from a few square meters toaround 1000 m2. Land catchment surfacesmay be natural, with their vegetationintact, or cleared and treated in some wayto induce runoff, especially when soils arelight. Non-land catchment surfaces include

the rooftops of buildings, courtyards andsimilar impermeable structures.

Diagram of micro-catchment water-harvesting systems

Water-Harvesting Methods

Micro-catchment methods

Macro-catchment andfloodwater methods

On-farm Systems Rooftop Systems Wadi-bed Systems Off-wadi Systems

Contour ridges Semi circular/trapezoidalbunds

Small pits Small runoff basins(Negarim)

Runoff strips Inter-row systems

Meskat Contour-bench terraces

Small farm reservoirs Water-spreading

Wadi-bed cultivationsLarge bunds, Tabia

Jessour

Hafaer, Tanks & Liman

Cisterns

Hillside conduits

On-Farm Systems

On-Farm micro-catchment systems aresimple in design and may be constructed atlow cost, making them easily replicableand adaptable. They have higher runoffefficiency than macro-catchment systemsand do not usually need a water con-veyance system. They allow soil erosion tobe controlled and sediments to be directedto settle in the cultivated area. Suitableland-based micro-catchment techniquesexist for any slope or crop. However, thesesystems generally require continuousmaintenance with a relatively high laborinput.

Unlike macro-catchment systems, thefarmer has control within his farm overboth the catchment and the target areas.All the components of the system are con-structed inside the farm boundaries. This isan advantage from the point of view ofmaintenance and management, butbecause of the loss of productive land it isonly in the drier environments, wherecropping is most risky, that farmers arewilling to allocate part of their farm to acatchment.

The most important land-based micro-catchment or on-farm water-harvesting

systems in the dry areas of WANA aredescribed below.

1. Contour ridges

These are bunds or ridges constructedalong the contour line, usually spacedbetween 5 and 20 m apart. The first 1–2 mabove the ridge is for cultivation, whereasthe rest is the catchment. The height ofeach ridge varies according to the slope’sgradient and the expected depth of therunoff water retained behind it. Bunds maybe reinforced by stones if necessary.Ridging is a simple technique that can becarried out by farmers. Ridges can beformed manually, with an animal-drivenimplement, or by tractors with suitableimplements. They may be constructed on awide range of slopes, from 1% to 50%.

The key to the success of these systemsis to locate the ridge as precisely as possi-ble along the contour. Otherwise water willflow along the ridge, accumulate at thelowest point, eventually break through anddestroy the whole downslope system.Surveying instruments, or an A-Frame andhand tools, can be used for contouring, butthese methods are too sophisticated andtime-consuming for most small-scale farm-

7

Contourridges con-structed atthe ICARDAresearch farmat Tel Hadya,northernSyria.

ers. The simplest method is a transparent,flexible tube 10–20 m long, fixed on twoscaled poles. The tube is filled with waterso that the two water levels are clear on thescale. Two people can trace the contour byadjusting the position of one of the poles sothat their water levels are the same.

If precise contouring is not feasible,small cross-bunds (ties) may be added atsuitable spacing along the ridge to stop theflow of water along the ridge. Contourridges are one of the most important tech-niques for supporting the regeneration andnew plantations of forages, grasses andhardy trees on gentle to steep slopes in the

steppe. In the semi-arid tropics, they areused for arable crops such as sorghum,millet, cowpeas and beans.

A special form of contour ridge may beconstructed for use with stone bunds ongentle slopes. Stone bunds are permeablestructures working only to slow downsheet flow and promote infiltration. Earthcan be excavated and added to theupstream side of the bund to turn it into animpermeable contour ridge. In the semi-arid tropics, this system is sometimes com-bined with other techniques, such as thezay system or the tied-ridge system. Thesestone bund systems can be used only ifsuitable large stones are available in thevicinity.

2. Semi-circular and trapezoidal bunds

These are usually earthen bunds in theshape of a semi-circle, a crescent, or atrapezoid facing directly upslope. They arecreated at a spacing that allows sufficientcatchment to provide the required runoffwater, which accumulates in front of thebund, where plants are grown. Usuallythey are placed in staggered rows.

The diameter or the distance betweenthe two ends of each bund varies between1 and 8 m and the bunds are 30–50 cmhigh. Cutting the soil to form the bund

8


The tube level is used tolocate the contour precise-ly before ridge construc-tion. Low in cost and easyto use by farmers.

Contour ridges supporting young shrubs with sufficientamount of water in the dry environment of the badia ofJordan.

Layout of semi-circular bunds in staggered rows in thefield. Photo from the Syrian badia.

immediately upstream creates a slightdepression. Runoff is intercepted here andstored in the plant root zone. If soil is cutupstream the slope is increased and thisraises the runoff coefficient; in this way thetechnique can be used on level land, but itcan also be used on slopes up to 15%.These bunds are used mainly for the reha-bilitation of rangeland or for fodder pro-duction, but may also be used for growingtrees, shrubs and in some cases field crops(e.g. sorghum) and vegetables (e.g. water-melons).

An eyebrow terrace is a form of semi-circular bund supported by stones on thedownstream side. The greater the slope,the more the bunds have to be strength-

ened with stones. The establishment andmaintenance of this system is labor-inten-sive.

3. Small pits

Pitting is a very old technique used mainlyin Western and Eastern Africa, but adoptedin some WANA areas. It is excellent forrehabilitating degraded agricultural lands.The pits are 0.3–2 m in diameter. The mostfamous pitting system is the zay systemused in Burkina Faso. This consists of dig-ging holes to a depth of 5–15 cm. Manureand different kinds of grass are mixed withsome of the soil and put into the zay. Therest of the soil is used to form a small dike

9

The zay system concentrates substantial runoff water inthe pit, where the crop is grown. (Photo by Prinz).

Semi-circularbunds plantedwith Atriplexhalimus, col-lected addition-al water after amoderate storm,in northernSyria.

An eyebrow terrace supporting fig trees with additionalrunoff water near Salloum in northwestern Egypt.

down the slope of the pit. Pits areapplied in combination withbunds to conserve runoff, which isslowed down by the bunds. Thissystem allows much degradedagricultural land to be put backinto use.

Pitting systems are used mainly for thecultivation of annual crops, especially cere-als such as millet, maize and sorghum.However, if the pits are dug on flat insteadof sloping ground, they may be regardedmore as an in situ moisture-conservationtechnique than as a water harvesting one.The labor requirement for digging a zaysystem is high and may constitute a con-

siderable investment in thefirst and even in subse-quent years since, aftereach tillage, the pits haveto be restored. A specialdisk-plow may be adaptedto create small pits forrange rehabilitation.

4. Small runoff basins

Sometimes called negarim,small runoff basins consist

of small diamond- or rectangular-shapedstructures surrounded by low earth bunds.They are oriented to have the maximumland slope parallel to the long diagonal ofthe diamond, so that runoff flows to thelowest corner, where the plant is placed.The negarim is best used on even ground.The usual dimensions are 5–10 m in widthand 10–25 m in length. Small runoff basinscan be constructed on almost any gradient,

including plainswith 1–2 %slopes; but onslopes above 5 %,soil erosion mayoccur and thebund heightshould beincreased. Theyare most suitablefor growing treecrops such as pistachio, apricot,olive, almond,and pomegran-ate, but may beused for othercrops. When they are used

10


Left: The Kimseed pitting machine adapted by ICARDAfor use in West Asian environments

Below: Pits created by a pitting machine ensure a suitableenvironment for reseeding steppe areas.

The negarim allows runoff water to concentrate at the lowest corner of the basin where theplant needs it.

for trees, the soil should be deepenough to hold sufficient waterfor the whole dry season.

If the catchment is well main-tained, 30–80 % of the rain canbe harvested and used by thecrop. Soil conservation is a posi-tive side-effect of negarims. Oncethe negarim system is construct-ed, it lasts for years with littlemaintenance. Plowing to controlweeds may not be practicalwithin the small space of eachbasin, so weeding may have tobe done by hand or with chemi-cals. If the negarim is built onheavy or crusting soils, a highrunoff coefficient may be achieved.However, since the system supports high-value crops, it can be economical to takemeasures to induce additional runoff.

5. Runoff strips

The technique of runoff strips is suitablefor gentle slopes. The strips are used tosupport field crops in the drier environ-ments (such as barley in the badia), whereproduction is risky yields are low. Thefarm is divided into strips along the con-tour. An upstream strip is used as a catch-ment, while a downstream strip supportscrops. The downstream strip should not be

too wide (1–3 m), while the catchmentwidth is determined in accordance with theamount of runoff water required. Runoffstrip-cropping can be fully mechanizedand needs only a relatively low input oflabor. The same cropped strips are cultivat-ed every year. Clearing and compactionmay be needed to improve runoff.

Agricultural inputs such as fertilizersand pesticides are applied to the cultivatedarea, as well as water. Under good man-agement, continuous cultivation of thecropped strip can build up soil fertility andimprove soil structure, making the landmore productive. This technique is highlyrecommended for barley cultivation andother field crops in large steppe areas of

WANA, where it can reducerisk and substantially improveproduction. The catchmentarea can be used for grazingafter the crop has been har-vested.

One problem the farmer mayface, however, is that the dis-tribution of water across thestrip may not be uniform. Thishappens especially on gentleslopes when the cropped stripis too wide, as with the khushk-aba system in Pakistan, or if a

11

Negarim plots are best for trees, particularly when runoff inducementmeasures are economical, such as this system, for almonds in theJordanian badia.

Runoff strips under preparation at the ICARDA research station, TelHadya, Syria.

small ridge is formed during cultivationalong the upstream edge of the croppedstrip. To overcome this problem, it is rec-ommended that the cropped strip shouldnot exceed 2 m in width, and that waterdistribution should be helped by goodpreparation of the strip surface.

To increase the uniformity of water dis-tribution across the cropped strip,ICARDA has developed a simple imple-ment towed behind the sowing machine toform small corrugations. These corruga-

tions enhance the flow of surface runoffinside the cropped area.

6. Inter-row systems

Inter-row systems, also called “roadedcatchments,” may be the best technique toapply on flat lands. Triangular cross sec-tional bunds or levees are constructedalong the main slope of the land as shownin photo 26. When high-value crops suchas fruit trees and vegetables are involved,the bunds may be compacted or possiblycovered with plastic sheets or treated withwater repellent materials to induce morerunoff. The bunds, with a height rangingfrom 40 to 100 cm, are built at distances of2–10 m.


Right: A runoff strip planted with fieldcrops receiving additional runoff waterafter a rainstorm in northern Syria. Waterdistribution uniformly needs enhancement.Below: Corrugations created by a specialimplement developed at ICARDA to over-come uneven water distribution across thecropped strip.

Runoff stripsare best forfield cropssuch as cere-als andlegumes, inareas whererainfall islow andslopes aregentle.

12

Runoff flowing down the slope is col-lected between the ridges and either direct-ed to a reservoir at the end of a feed canalor to a crop cultivated between the ridges.The catchment area has to be weeded andcompacted on a regular basis to maintainhigh runoff output.

7. Meskat

Meskat is a term used in Tunisia for anindigenous water-harvesting system sup-porting mainly olives and figs. This systemconsists of a catchment, or meskat, occupy-ing the slope adjacent to a flat cultivatedarea called manqa. Sometimes, the catch-

ment areas are surrounded by a bund andmay be provided with spillways to letrunoff flow between plots without causingerosion.

Khushkaba is a similar technique used forgrowing field crops in Balochistan,Pakistan. Large plots of 1000–5000 m2 aredivided into two parts; the higher oneserves as a catchment, supplying runoffwater, and the lower one as a croppingarea. A drawback of this system is the lackof uniformity in water distribution acrossthe cropping area. If the width of the crop-ping area is reduced to improve uniformi-ty, the system comes to resemble the runoffstrip system described earlier. The khushka-ba system is used mainly for improvingwheat and barley production in very dryenvironments (< 250 mm rainfall).

8. Contour-bench terraces

Contour-bench terraces are constructed onvery steep slopes to combine soil andwater conservation with water-harvestingtechniques. Cropping terraces are usuallybuilt level and supported by stone walls toslow down the flow of water and controlerosion. They are supplied with additionalrunoff water from steeper, non-croppedareas between the terraces. The terraces areusually provided with drains to releaseexcess water safely. They are frequently

13

Inter-row system concentrating runoff water in a reser-voir in Australia. Photo from booklet on water harvest-ing by the State Department of Western Australia.

A Tunisian meskatsystem.

A meskat system supporting olive plantation with har-vested water in middle Tunisia, with rainfall less than250 mm.

used to grow trees and bushes,but rarely used for field crops in theWANA region. The historic bench terracesin Yemen are a good example of this sys-tem. Since they are constructed on steepmountainsides, most of the work is doneby manual labor. The drawback of this sys-tem is that construction and maintenancecosts are high.

Rooftop SystemsRooftop systems collect and store rainwa-ter from the roofs of houses or large build-ings, greenhouses, courtyards, and similarimpermeable surfaces, including roads.Most of the rain can be collected andstored. How the harvested water will be

used depends on the type of surface usedand it’s cleanliness as well as users’ needs.Modern roofing materials and gutters, forexample, allow the collection of cleanwater suitable for drinking and otherdomestic uses, especially in rural areaswithout tap water. However, farmers usu-ally avoid storing the runoff from the firstrain as this may not be clean enough todrink. If water is collected from a surfacethat may have soil or plant debris on it, therunoff has to be passed through a settlingbasin before storage. Such systems providea low-cost water supply for humans andanimals in remote areas. Although mainlyused for domestic purposes, this techniquealso has agricultural uses. Water not suit-able for drinking may be used to supporthome gardens. Rainwater harvested from agreenhouse roof may be used in the greenhouse for irrigation.

Macro-catchments andFloodwater SystemsMacro-catchment and floodwater-harvest-ing systems are characterized by havingrunoff water collected from a relativelylarge catchment. Often the catchment is anatural rangeland, the steppe, or a moun-tainous area. Catchments for these systemsare mostly located outside farm bound-aries, where individual farmers have littleor no control over them.

14


Ancient contour-bench terraces support-ing coffee and qat trees in the mountainsof Yemen.

Rooftop water-harvesting system providing drinking andsupplemental irrigation water in remote areas.

Macro-catchment systems are sometimesreferred to as “water harvesting from longslopes” or as “harvesting from an externalcatchment.” The predominance of turbu-lent runoff and channel flow of the catch-ment water contrasts with the sheet or rillflow typical of micro-catchments.

Generally, runoff capture is much lowerthan for micro-catchments, ranging from alow percentage to 50% of annual rainfall.Water is often stored in surface or subsur-face reservoirs, but may also be stored inthe soil profile for direct use by crops.Sometimes water is stored in aquifers as arecharge system. The cropping area iseither terraced on gentle slopes or locatedon flat terrain.

Water rights, affecting the distributionof water between the catchment and thecultivated areas and to the various users in

the upstream and downstream areas of thewatershed, are among the most importantproblems associated with these systems.The best solution to these problems is toplan interventions using an integratedwatershed development approach in whichall stakeholders are involved.

Large macro-catchment systems insteppe areas are often called “floodwater-harvesting systems.” According to the loca-tion of the target area, two types of macro-catchment and floodwater system exist:wadi-bed systems and off-wadi systems.

Wadi-bed SystemsIn this system, the wadi bed is used to storethe water, either on the surface by blockingthe water flow, or in the soil profile byslowing down the flow and allowing it toinfiltrate the soil. The following wadi-bedsystems have been found to be the mostsuitable for the steppe areas of WANA:

1. Small farm reservoirs

Farmers who have a wadi passing throughtheir land can, if a suitable location exists,build a small dam to store some or all ofthe runoff water flowing down the wadi.

The water can then be used to irrigatecrops or for domestic and animal con-sumption. These reservoirs are usually

15

A typical macro-catchment water-harvesting system.

Small farm water-harvesting reser-voir for supple-mental irrigationof field crops atICARDAresearch stationin northern Syria.

small but may range in capacity from 1000to 500,000 m3. Assistance may be neededfrom an engineer to plan, design and buildthe dam.

The most important feature is to have aspillway with sufficient capacity to allowfor the excessive peak flows that may passthrough the wadi. Many of the small farmreservoirs constructed in the West Asianbadia, have been washed away throughlack of, or insufficient, spillway.

Small farm reservoirs are very effectivein the badia environment. They can supplywater to all crops, increasing and stabiliz-ing production. Moreover, the benefits tothe environment are substantial. To maxi-mize water-use efficiency and reservoircapacity and mini-mize losses in evapo-ration and seepage, itis advisable that thecollected water ispumped as soon aspossible and stored inthe crops’ root zone(except for the waterneeded for drinkingand for consumption

by animals). This implies that, for greaterefficiency, water should be used for thesupplemental irrigation of winter crops,during the winter rainfall period, ratherthan left for the full irrigation of summercrops.

2. Wadi-bed cultivation

This technique is very common in wadibeds with gentle slopes. As a result of theslow water velocity, eroded sediments usu-ally settle in the wadi bed and create goodagricultural land. This may occur naturallyor can be achieved by the construction of asmall dam or dike across the wadi toreduce the flow speed and allow soil sedi-ments to settle. Wadi cross-walls, usually

16


Small earth dam, broken due to lack of a sufficient spill-way. Photo from booklet on water harvesting by theState Department in Western Australia.

Stonewalls, if properly constructed across the wadi bed,can help to store and distribute runoff water evenly andto build up soil behind them.

Wadi-bed stonewalls sup-porting fig plantations inMatrouh, north-western

Egypt.

no higher than 1 m, are preferably made ofpermeable stone and may be reinforcedwith gabions. The top of the wall should beall at the same level, so as to create uni-form land behind it, allow excess water tooverflow along its entire length. Distancesbetween walls along the wadi bed aredetermined according to the slope of thewadi bed and the height of the wall. Thistechnique is very common for fruit treessuch as fig, olive, date palm and otherhigh-value crops, since the soil in the wadibed is usually fertile and the availability ofwater can be relied on with reasonable cer-tainty. The walls increase the range ofcrops that can be grown in these marginalareas.

The main problems associated with thistype of water harvesting are the costs andthe maintenance of the walls. Anotherproblem that has arisen recently in someWANA areas, where catchments are beingdeveloped for housing, is that less runoffwater reaches the wadi beds, with the resultthat downstream crops become increasing-ly water-stressed. In such circumstances,an integrated watershed developmentapproach is needed to decide on a fair allo-cation of the water supply.

3. Jessour

Jessour is an Arabic term describing thewidespread indigenous wall structuresbuilt across relatively steep wadis in south-ern Tunisia. The walls are usually high

because the slope is steep. They are madeof earth, stones or both, but always have aspillway, usually of stone. Over the years,as water is stopped behind these walls,sediments settle and accumulate, creatingnew land for planting, mainly with figsand olives but also with other crops.

This system is similar to wadi-bed culti-vation except that it is used on steep wadibeds and always includes a spillway torelease the excessive water. Usually there isa series of jessours along a wadi originatingfrom a mountainous catchment. These sys-tems require maintenance to keep them ingood shape. Since the importance of thesesystems for food production has declinedrecently, vigilance and maintenance havealso reduced, and many systems are break-ing down.

17

The ancient jessours of Tunisia have supported oliveand fig trees for centuries.

The jessour, built along steep wadi beds.

Group of ancient jessours in a wadi system in southernTunisia.

Off-wadi SystemsThe rainwater harvested in off-wadi sys-tems is applied outside the wadi bed.Structures may be used to force the wadiwater to leave its natural course and flowto nearby areas suitable for agriculture.Similar structures may also be used to col-lect rainwater from catchments outside thewadi bed. The following are the mostimportant off-wadi techniques:

1. Water-spreading systems

In this technique, also called “floodwaterdiversion,” part of the wadi flow is forcedto leave its natural course and is then con-veyed to nearby areas and applied togrowing crops. Water is stored only in theroot zone of the crops, i.e. it supplementsrainfall. Diversion is usually accomplishedby means of a structure that raises thewater table in the wadi bed, allowing theflow to spread by gravity, on one or bothsides of the wadi. The flow is directed by alevee, which should run a little off the con-

tour, away from the wadi path. Water spreading requires relatively uni-

form land with a gentle slope. Agriculturalland may be graded and divided intobasins by levees to allow enough water tobe stored for the season. Soils should bedeep, with sufficient water-holding capaci-ty.

As this system requires proper siting,design and construction of the structureand the conveying canal, the expertise ofan engineer may be required. The structureshould be strong enough to resist the flowand be at an elevation sufficient to divertthe required portion of the flow. Different

18


Water-spreading system inoperation in southern Tunisia.

materials have been used to build diver-sion structures, including stone and con-crete. The most durable structures aremade of stone-filled gabions.

One important point to consider is thatthe slope of the conveyance canal shouldallow a flow velocity sufficient to preventthe accumulation of sediments near thestructure, otherwise these will block theflow and entail high maintenance costs.

2. Large bunds

Also called tabia in Tunisia, this systemconsists of large semi-circular, trapezoidalor open V-shape earthen bunds with alength (the distance between the tips ofeach bund) of about 10–100 m and a heightof 1–2 m. Often, they are aligned in long,staggered rows facing up the slope. Thespace along the contour between adjacentbunds is usually about half the length ofthe bund. The tips of the bund should beprotected against erosion, as water oftenoverflows round them. Large bunds areusually constructed by machinery and onlyrarely by hand. They are used to supporttrees, shrubs, and annual crops in WANA,but also sorghum and pearl millet in sub-Saharan Africa.

Large semi-circular bunds can storelarge quantities of water, but can breakunder extreme rainstorms, so a control

overflow has to be planned for. The mostcritical period is immediately after con-struction, before they are fully consolidat-ed. Any breakage must be repaired imme-diately. As these systems are not tradition-al, adoption can pose problems.

3. Tanks and hafair

Tanks are usually earthen reservoirs dug inthe ground in gently sloping areas thatreceive runoff water either by diversionfrom wadis or from a large catchment area.They are known as “Roman ponds” inparts of North Africa, where they are usu-ally built with stone walls. The capacity ofthese ponds ranges from a few thousandcubic meters, in which case they are called

hafair, to tens of thousands of cubic meters.Tanks are very common in India, wherethey support over 3 million ha of cultivat-ed land. In the WANA region, especiallySudan, Jordan and Syria, the smaller tankis more common and is mainly used forwater consumption by humans and animal.

Several problems are associated withhafairs: stagnant water may be polluted,attract insects and become a source of dis-ease. As they usually have no protectionaround them, there is a risk of drowningaccidents, which may involve people oranimals. Large losses through seepage and

19

Typical tabia from the Matmata Mountains in southernTunisia.

Hafair in western Sudan provide water for people andtheir livestock during the dry season.

20

Guidelines for Selecting Water-Harvesti

Technique Crop Soil Land Land Depth (1) Texture Slope (2) Vegetation (

Micro-catchmentsOn-farm systems

Contour ridges range variable variable med.,steep poor,med.field med.,deep " medium poortrees deep med.,heavy low,med. poor,med.vegetable med.,deep " " "

Semi-circular bunds range " variable " poor(trapezoidal field " med.,heavy " "triangular) trees deep " " poor,med.

vegetable " " " "

Small pits field " " " poorrange shallow,med. " " poor,med.

Small basins range med.,deep " " poor(Negarim) trees deep " low poor,med.

Runoff strips range variable " low,med. poorfield med.,deep " " poor,med.

Inter- row system trees deep " Low poor(roaded catchment) vegetable medium variable " "

field " " " "

Meskat (Khushkaba) trees deep med., low,med. poor,med.field medium heavy " poor

"

Contour bench trees deep " Steep "terraces field medium " " "

Rooftops drinking na na na navegetable variable variable " "

Macro-catchmentsWadi-bed systems

Small farm reservoirs all crops variable " low,med. variable

Wadi-bed cultivation trees/ med.,deep med., " poorvegetable heavy

Jessour trees " " med.,steep variable

Off-wadi systems Water spreading field/trees " " low,med. poor

Large bunds trees deep " " poor,med.field medium " " poorrange shallow,med. variable " med.,dense

Tanks and hafair all crops variable med., heavy low variable

Cisterns drinking/ deep rock all slopes poor,med.trees/vegetable

Hillside runoff field/trees med.,deep med.,heavy low,med. poor,med.systems

(1) shallow < 50 cm, medium: 50-100 cm, deep> 100 cm; (2) low < 4%, medium: 4-12%, steep > 12%; (3) poor < 15%, medium: 15-30%, dense > 30%(7) low < 5 man-day/ ha, medium: 5-20 man-day/ ha, high > 20 man-day/ ha; na : not applicable.

21

g Techniques in the Drier Environments

Cover Socio-economics Storage typeStoniness (4) Farm size (5) Capital (6) Labor (7) Skill

low-med. variable low medium local/ soil profilelow small,med. " " training "" " " " " "" small " " " "

low,med. variable " high local/no "" small,med. " " training "low " " " " "" small " " " "

" variable " medium " "low,med. " " " " "

" small " high " "" small,med. " " " "

" " " low " "" " " " " "

low large high medium local/training "" " " " " "" med.,large " " " "

" variable low " local/no "" small,med. " low training "

"

" " high medium external skill "" " " " " "

na small " medium local/training surface/" " " " " subsurface

"

variable med.,large high high external skill surface/subsurface

Low small,med. medium med.,high local surface/soilprofile

variable small " high local/training "

low,med. variable " medium external skill soil profile

" med.,large " " local/training "low medium " " " "variable med.,large " " " "

" " med.,high " external skill surface/subsurface

variable small,med. medium high local/training subsurface

low small,med. " high local/training soil profile

ow < 10%, medium: 10-25%, high> 25%; (5) small<5 ha, medium:5-25 ha, large> 25 ha; (6) low < $ 25/ ha, medium: $ 25-100/ ha, high > $ 100/ ha;

evaporation are further disadvantages.Several improvements, including fencing,lining and settling basins, have been pro-posed and, occasionally, introduced toovercome these limitations.

4. Cisterns

Cisterns are indigenous, subsurface reser-voirs with a capacity ranging from 10 to500 m3. They store water for human andanimal consumption. In many areas, as inJordan and Syria, they are dug in the rock,in which case they usually have a smallcapacity. In northwestern Egypt, farmersdig large cisterns (200–300 m3) in the earthdeposits underneath a layer of solid rock.The rock layer forms the ceiling of the cis-tern, whereas the walls are covered with

impermeable plaster. Modern concrete cis-terns are being constructed in places wherethere is no rocky layer.

Runoff water is collected from an adja-cent catchment or channeled from a remoteone. The first rainwater runoff of the sea-son is usually diverted away from the cis-tern to reduce the likelihood of pollution.Settling basins are sometimes constructedto reduce the amount of sediments, butfarmers usually clean the cisterns once ayear or every other year. Water is typicallylifted by a bucket and a rope.

Cisterns are still the only source ofdrinking water for humans and animals inmany dry areas of WANA, and their rolein maintaining rural populations in theseareas is vital. They are now often used tosupport home gardens, in addition tomeeting domestic needs. Problems includethe cost of construction, limited capacity,sediments, and sources of pollutants fromthe catchment.

5. Hillside-runoff systems

In Pakistan, this technique is also calledsylaba or, as previously mentioned, sailaba.Runoff water is directed by small conduitsto flat fields at the foot of the slope. Fieldsare leveled and surrounded by levees witha spillway to drain the excess water toanother field downstream. When all fieldsin a series are filled, water is allowed torejoin the wadi. When several feeder canalsare planned, distribution basins are useful.This is an ideal system for using the runofffrom bare or sparsely vegetated hilly ormountainous areas.

Hillside conduit schemes require properdesign, a high labor input, and, probably,the assistance of an engineer. The conduitsshould have sufficient slope to preventsedimentation, otherwise they have to becleared after heavy rainstorms. Fields needto be leveled and spillways constructed at

22


Ancient roman tanks are still functioning in manyWANA areas.

Cisterns are vital to people in remote areas where noother source of water is available. This one in Egypt stillsupports people, livestock and gardens in rural areas.

a proper elevation to ensure uniformdistribution. This technique can beapplied to almost any crop.

Socioeconomic AspectsWater-harvesting projects do not dependsolely on good engineering and suitableagronomy for their success. Socioeconomicconsiderations are just as important. In thedrier environments people have been liv-ing at subsistence level for centuries andhave developed their own priorities fortheir way of life and survival. It is of theutmost importance, therefore, to take theirvalues, perceptions, attitudes and prefer-ences into consideration rather than try toimpose solutions on them.

One effective way of introducing anddeveloping water-harvesting projects isfirst to go to the prospective beneficiaries,talk to them, learn from them and showwillingness and potential to serve them. Ifthe scheme really meets their needs, theninterventions can be planned with them,starting with their knowledge and buildingon it, using what they have. It is importantto make them feel that the project is theirown and that it will be of real use to them.Although planners may often ignore indi-rect benefits when conducting feasibilitystudies of water-harvesting projects, it isessential to understand the importance of

these too. They include halting land degra-dation, combating desertification, supply-ing drinking water for animals, slowingmigration to the cities, minimizing socialproblems, improving the standard of livingof the farmers’ families and enhancing thestability and security of village life.Farmers who implement water-harvestingprojects in the drier environments will becontributing to these benefits for the popu-lation in general.

23

The hillside runoff system captures runoff water as it flows down the hill. (Photo by Prinz.)

Yemeni farmer holding the controversial crop, qat,grown in the famous water-harvesting terraces of theYemen mountains. High returns insure maintenance ofthe system.

Planning, Design and Implementation

Another important issue is land owner-ship. This can often be a problem, sincemuch of the land requiring structural inter-ventions may be communally used orowned by the state. In these circumstances,individual farmers have little incentive toinitiate costly development.

Designing Water-HarvestingSystems

Selecting the Site and Technique

The suitability of an area for water-harvest-ing depends on its being able to meet thebasic technical requirements of the system.In addition, whatever technique is selectedmust be compatible with local social condi-tions and farming practices. In planningthe systems, appropriate data must beavailable on the climate, soil, crops, topog-raphy and the socioeconomics of the pro-ject area. Among the tools and methods ofdata acquisition for planning, designingand implementing water-harvesting sys-tems are field visits, site inspection, topo-graphic and thematic maps, aerial photosand satellite images (remote sensing) andGeographic Information System (GIS). Asan aid to selecting the most suitable

method, Table 1 provides general guide-lines on the requirements of the mostimportant water-harvesting techniques.

The site and methods are determined bytaking into account the use to which theharvested water will be put. Proximity tobuildings and purity of water will be issuesto consider if the water is for domestic useor for livestock, while different demandsmust be met for water needed for agricul-ture or for multi-purpose use.

Although water-harvesting systems maybe implemented on a wide range of slopes,topography is still a major factor in theselection of an appropriate technique.Generally, but not invariably, steeperslopes with shallower soils are used ascatchments, and cropping is allocated togentler slopes, where soil is deeper. Thisallows the less productive, shallow soil tocontribute its share of rain to the deeper,more productive soil.

Soils with high infiltration rates, such assandy soils, are not favorable as catchmentareas for water harvesting without somemeasures to induce runoff. Generally, thisis only feasible when using micro-catch-ments to harvest water for human and ani-mal consumption or for the production ofhigh value crops. Soil texture should beconsidered because it affects soil erosion in

24


Remotely sensed data combined with ground data in aGIS environment can result in substantial saving ofeffort in investigating the suitability of sites for waterharvesting. (Photo by Oberli.)

Classification of areas according to suitability for variouswater-harvesting methods. (Photo by Prinz.)

the catchment. Both soil texture and depthinfluence the total water-storage capacityof the soil profile, and this, in turn, controlsthe amount of water that can be madeavailable for crops during the dry periods.

Water rights, land tenure and use areamong the problems that sometimes hin-der the selection of proper sites and tech-niques. In the past many water-harvestingprojects have failed simply because theseissues were not taken fully into account.Collective land ownership increases thenumber of options available, including thechoice of a macro-catchment. Large-scalesystems may be more economical becausethey require less work to install and lessmaintenance per unit area.

As a result of changes in social attitudestowards private ownership, economicincentives and personal aspirations, small-scale farmers are now more open to theidea of introducing a micro-catchment ontheir land. The farmer’s ability to operateand maintain the system, however,remains a constraint to using more sophis-ticated systems. Construction requirementssuch as availability of materials and skilledlabor must also be taken into account inthe selection of systems.

Selecting the CropsIn general local crop or tree species are bestadapted to the environment and shouldhave priority over introduced species.Water harvesting may, however, allowfarmers to grow species previously consid-ered too risky. Improved varieties may besuitable, provided their introduction fol-lows research and adaptation programsthat prove their viability. The crops andtrees selected should be capable of integra-tion with the local farming system and ableto withstand the 2–3 days of waterloggingthat are typical in most water-harvestingsystems after heavy storms. Maize, forexample, is not suitable.

As water-harvesting systems can onlypartially compensate for low and erraticrainfall, it is strongly recommended thatdrought-tolerant trees, shrubs and crops beselected. These can often survive even inextreme drought, when the system fails toprovide sufficient moisture. In drier envi-ronments, fodder shrubs and trees are gen-erally able to recover more quickly afterheavy browsing. To ensure the most effi-cient use of water and to obtain a harvestrapidly, winter crops should be given pri-ority over summer ones. When trees areselected, it is essential that deep soil withsufficient water-storage capacity is avail-able in order to provide the necessarymoisture over the entire dry period of theyear. In areas where there is a possibility ofwater standing on the surface for long peri-ods, crops tolerant to waterlogging must beselected.

Designing the SystemThe design of the water-harvesting systemshould ensure, with reasonable probability,that a specific amount of water is availablefor the designated use. It is important toemphasize that in the drier environments,it is not always necessary for the potentialwater demand to be fully met, as crops

25

Almonds, olives, palms and figs are among the mostsuitable trees to be grown in the drier environmentsusing water-harvesting systems.

may be grown economically, and produce,without their water needs being complete-ly satisfied. The design water amount, i.e.the amount of water the water-harvestingsystem is designed to provide, shouldallow maximum economic, social and envi-ronmental return.

With micro-catchment systems thecatchment area must be able to supply thedesign water amount to the target area.The size of the catchment can be decidedaccording to rainfall characteristics, landslope, soil characteristics, land cover, crops,and economic considerations. Similar careis needed in the design and provision ofthe water-harvesting structures used forconveying and then storing or distributingthe water. The design of the entire systemmust be flexible enough to allow for anynecessary changes in the cropped area orcrop type during implementation andfuture operation.

The main design steps for micro-catch-ment water-harvesting systems are:1. Determine the design annual runoff

coefficient of the selected site. This coef-ficient is the ratio of the amount ofannual runoff to the amount of annualrainfall. Its value depends on rainfallamount and intensity, soil, topography,land surface, and catchment size.

Coefficients derived by techniques validfor macro-catchments are not applicableto micro-catchments.

There is a notable lack in the litera-ture on reliable and practical methods ortechniques for determining runoff fromsmaller catchments (those ranging insize from a few to several hundredsquare meters). One way to determinerunoff is by experimentation using fieldplots or by using field rain simulators.

Since the design annual runoff coeffi-cient depends on rainfall characteristics,the value adopted must be at an accept-able level of probability. Treating thesoil surface physically or chemically (orboth) can significantly increase therunoff coefficient, but also increasescosts. Such treatment for inducingrunoff may be economically justified,depending on the purpose of the water-harvesting system.

2. Determine design crop water require-ments (under water-harvesting condi-tions). These requirements are usuallyset at only a fraction of the stress-freewater requirements usually associatedwith the use of inputs and proper sys-tem management. Normal methods ofestimating crop evapotranspiration maybe used with an estimated stress coeffi-cient reflecting the level of stress thecrop can be expected to tolerate duringthe dry period.

26


Where thereis no localinformationavailable,runoff coeffi-cient plotsare essentialto determinedesign para-meters formicro-catch-ment sys-tems.

Low cost weirs may be installed on wadis to measurerunoff rate from macro-catchments. This one is inMouaqar, Jordan.

3. The design of the water-harvesting sys-tem should be based not on average val-ues of rainfall in the area, but on lowervalues, so as to ensure greater depend-ability of the system. Water-harvestingsystems for trees, shrubs and perennialcrops should be based on rainfall data.Generally, the design should be able toprovide water to the plant economicallyat least three years out of four; i.e. adependability level of 75%.

4. Based on the outcome of Steps 1, 2 and

3, the ratio of the catchment area to thecropped area is determined. Anallowance for non-uniformity of waterdistribution and deep percolation in thecropped area should be providedthrough a storage-efficiency factor. Thisis represented by the ratio of the volumeof harvested water stored in the effectivedepth of the root zone to the total vol-ume of water harvested.

Typical values for this storage effi-ciency range from 50 to 75%. If the ratioof the catchment area to the croppedarea is underestimated, the actual valueof this factor may approach 100%.Conversely, if this ratio is overestimated,the actual storage efficiency may be verylow, resulting in water being wastedand more land being unnecessarily putout of production as catchments.

5. If the size of the cropped and catchmentareas is known, their shape and dimen-sions can be decided depending on the

27

Table 1. Estimated runoff coefficients and cost of typi-cal inducement techniques

Treatment Runoff Estimated Cost coefficient life (US$/

(%) (years) 100 m2)

Catchment clearing 20-35 1-3 1-4Surface smoothing 25-40 2-4 2-4Soil compacting 40-60 2-3 6-10Surface modification 70-90 3-5 10-20Surface sealing 60-80 5-10 4-10Impermeable cover 95-100 10-20 20-100

Layout of a small earth dam and a spillway designed and built in the Muaqqar area of the Jordanian badia.

type of system, the kind of crop and thetopography. The engineering worksrequired can then be planned. Theseinclude system layout, details and quan-tities of the earth or stone works and ofother water-control structures.

Macro-catchment and floodwater sys-tems usually involve the design of asmall dam, diversion structures, and therequisite water conveyance and distribu-tion systems. These systems may alsoinclude facilities to store water for sub-sequent use. The size of a macro-catch-ment or floodwater catchment area isnot under the designer’s control. Thedesigner merely determines the extent ofthe cropped area to be served by theexpected runoff.

Runoff may be determined fromrunoff measurements, gauges at thewadi, and, more rarely, from flow cross-sectional area and velocity measure-ments. Alternatively, it may be estimat-ed from simulation models. As suchlarge-scale systems are beyond the abili-ty of individual farmers, engineers areneeded to do the design.

ImplementationWater-harvesting systems may be imple-mented by:

• The farmer As mentioned earlier, micro-catchment sys-tems are usually within individual farms.This is a simple and low-cost approach,although farmers may experience some dif-ficulty with elements requiring precision,such as following the contour lines ordetermining maximum slope.

• The communityThe community may be involved in micro-and macro-catchment or floodwater-har-vesting systems, typically through a projectplanned locally with the help and guidanceof the government.

• Public agencies These are usually needed for large-scalemacro-catchment and floodwater harvest-ing schemes. Government services or con-tractors are used in this approach. Usuallymachinery is used or paid local laboremployed. Initial cost is relatively high.This is a top-down approach, however, andthere is a risk that the scheme will not beaccepted or maintained by farmers.

The first two approaches have provedmore successful than the third, but do needgovernment support with simple demon-strations, training and extension services.A round-table discussion involving all par-ties, with the aim of finding the best techni-cal approaches for the locality, is helpful.The selected plan of action should be sim-ple enough for the local people to imple-ment themselves. The planner should be

28


A water-harvesting system implemented by thefarmer has a higher chance of success. Photo fromBalochistan, Pakistan.

Proper con-struction ofwater harvest-ing systemsrequires skilledlabor.Enhancing thecapacity ofpeopleinvolved isusually needed.

ready to listen and learn from the farmersso as to respond effectively to their needs.Putting the farmer in a managerial role willalso contribute to the success of the project.

Operation and MaintenancePoor management and lack of maintenanceare the main reasons for the failure ofwater-harvesting projects. Large-scale sys-tems require the creation of a local associa-tion to manage the facility and liaise withthe appropriate government agencies.Guidelines and procedures for the opera-tion and maintenance of all components ofthe water-harvesting system are needed atthe outset of the project.

New systems should be inspected often,especially during the first one or two rainyseasons after construction. Micro-catch-ment systems should be inspected afterevery runoff-producing rainstorm so thatany minor break in bunds can be promptlyrepaired. Special attention should be paidto earthen dikes and bunds, water-storagefacilities and their spillway and the diver-sion structures. Treated catchments, inaddition to target structures, should beprotected against damage by grazing ani-mals. Silt and rubbish should be removedfrom the water conveyance and distribu-tion systems and from storage facilities.

Farmers unfamiliar with irrigation mayneed advice about irrigation techniques

and related activities. This should includemethods for improving soil fertility anderosion control. Systems providing drink-ing water should be protected against pol-lution and should be cleaned regularly.Cleaning the catchment annually, main-taining silt traps and settling basins, andde-silting cisterns are further necessarymeasures.

Prerequisites for Success• People’s participation

The people directly concerned should beincluded in all phases of the project:planning, development, implementation,operation and maintenance.

• Acceptability to the beneficiariesThe project should be simple and appro-priate to their existing farming opera-tions.

• Demonstrations and trainingThese are needed to support newlyintroduced systems or practices.

• Attractive benefits On account of the risk and uncertaintiesinvolved, the farmer may need to beconvinced of the benefits of water har-vesting. The most important form ofbenefit is increased income from the saleof additional produce. It is thereforeessential to check that there is a marketfor this. Micro-credit schemes or otherforms of subsidy may still be needed tofinance the construction, operation andmaintenance of systems before the bene-fits from additional crop production arefelt. With appropriate policies and regu-lations, micro-credit may also be used tosupport the creation of local groups, andhuman-capacity building throughdemonstrations, training, extension ser-vices and dissemination of information.

Subsidies may also be needed to com-pensate farmers for their contribution to

29

Planning water-harvesting systems at large scalerequires appropriate information in map form.

improvements in the national environmentand socioeconomics resulting from waterharvesting projects.

• Clear land tenure and property rightsThese concerns have been overlooked inmany projects in the past leading to con-flicts over both land and water. Theseoften severe constraints to developmentneed to be addressed early in the plan-ning process.

• Downstream water rights ensuredDownstream water users and environ-mental water needs should be taken intoconsideration when adopting macro-catchment and floodwater-harvestingsystems. It must never be forgotten thatthe development of water resources

demands an integrated watershed man-agement approach.

30


Farmers in Matrouh Natural Resource ManagementProject in north Egypt participate in the planning andimplementation of water-harvesting works at theirfarms.

ICARDA’s Research ActivitiesThe ecoregional project “On-farm WaterHusbandry in WANA” was launched byICARDA in 1996 to promote the integra-tion of water harvesting in the farming sys-tems of the drier environments. Since then,national teams in eight countries of WANAhave been implementing the project inpartnership with ICARDA. The countriesinvolved are: Egypt, Libya, Tunisia,Morocco, Syria, Jordan, Iraq and Pakistan.Then, in 1999, Algeria, Yemen and Iranjoined the project.

Research themes have been identified inwhich the complementary strengths of allpartners are combined in strategic andapplied research throughout the region.The project research themes are:1. Water in present land-use systems:

indigenous knowledge, and end-userperceptions and participation. Thistheme involves documentation andanalysis of indigenous water manage-ment and recognition of what can belearnt from it.

2. Water resources and capture potential:This theme involves analyzing rainfall-runoff relations and developing method-ologies for investigating water-harvest-ing potential under different conditions.

3. Options for water use: Water-harvest-ing methods and techniques suitable forthe prevailing land, crops and socioeco-nomic conditions are studied and adapt-ed and ways of integrating these tech-niques into farming systems areresearched.

4. Dissemination, development andimpact: This theme involves buildingpeople’s capacity through training anddemonstration. It also includes studiesand evaluation of how the social, eco-nomic and policy issues governing theadoption of improved water-harvestingsystems will affect the farmers, theirlivelihoods and the environment.

The following are examples of theproject’s major findings:

Planning WaterHarvesting using RemoteSensing and GIS

Conventional investigationsfor planning water harvestingare intensive. If water-harvest-ing is needed on a large scale,such investigations becomestedious and very costly.ICARDA has used remotesensing and GIS to develop amethodology for classifyinglarge areas according to theirsuitability for water harvestingusing different techniques andsystems. Data from a full LANDSAT scenefor central Syria were first used to providevaluable information on vegetative cover,surface conditions, and drainage systems.

Additional information including soiltypes, topography, rainfall and expert cri-teria on water-harvesting techniques werethen entered in a GIS environment and

analyzed together with satellite data toproduce the final classification. Thismethodology gave very promising resultsat low cost. Work is continuing to includeprovision for hydrology so as to estimatethe amount of runoff to target areas andfor soil depth relevant to aid crop selectionin specific areas.

31

Water-harvesting-suitability map for central Syria, produced at ICARDAthrough research using remotely sensed data, ground data and GIS techniques.

The efficien-cy of indige-nous waterharvestingsystems isbeingresearched atICARDAwith resultsthat shouldprove applic-able in manyareas ofWANA.

Indigenous Water-HarvestingSystems

Indigenous water-harvesting systems inEgypt, Iraq, Jordan, Libya, Morocco,Pakistan, Syria, and Tunisia were docu-mented and analyzed, and lessons on theirrelevance to future development wereidentified. A national team supported byICARDA scientists produced the documentin each country.

Problems with these systems include thecost of construction and maintenance, lim-ited capacity, sediments, and sources ofpollutants from the catchment. ICARDA isconducting research in many areas to findsolutions to these problems. Of particularimportance is research being conducted innorthwestern Egypt to improve water col-lection, storage and use in cistern systems.

Micro-catchment Success Stories

Research results in Syria, Jordan, Moroccoand Egypt have shown great success in

using micro-catchments. In the Syrian badia(annual rainfall 100-200 mm), a specialimplement mounted on a tractor was usedto construct semi-circular bunds mechani-cally. The mechanized tool can be used toimplement this technique on a large scale.

In an ICARDA-supported research pro-ject in the Syrian badia, the implement wasable to provide over 20 ha with manybunds per day, creating over 5,000 bundsof varying size and spacing. It is idealwhen labor is not available or when it iscostly. Operating cost is around US$100/ha, including planting. Shrubs plant-ed in the semicircular bunds showed a sur-vival rate of over 90%, compared to 10%without water harvesting.

In Jordan and Egypt, small basins andsemicircular bunds provided almond andolive trees with sufficient water to grownormally in areas with 120–150 mm annualrainfall. In Morocco, a system combiningtrees and shrubs with the use of contourridges proved very successful in areas withrainfall of 100–200 mm.

Water Harvesting for SupplementalIrrigationApplying limited amounts of water duringstress periods as a supplement to rain sub-

32


A mechanized implement for making micro-catchmentssaved labor costs and proved suitable for improving pro-duction and reducing degradation in the Syrian badia.

stantially increases and stabilizes produc-tion. In areas where full-scale irrigation isnot available, water harvesting can providethe essential moisture. In Tunisia smallwater-harvesting reservoirs have been builtin the mountainous areas and are beingassessed using GIS for their use in supple-mental irrigation.

The research work showed great poten-tial in improving water-use efficiency inthe production of winter and summercrops such as wheat and vegetables.Similar research conducted by thePakistani team in Balochistan provincedeveloped some promising options for useof ground water in conjunction with waterharvesting in supplemental irrigation.

Water Harvesting from GreenhouseRoofs

A system for harvesting rainwater fallingon the roofs of plastic greenhouses wasintroduced and monitored by the Egyptianteam of the on-farm, water-husbandry pro-ject near Alexandria, where annual rainfallis less than 100 mm. The harvested waterwas conveyed to the vegetable crop in thegreenhouse, where a technique known as“hydroponics” was used to provide waterand nutrients at high use efficiency. The

33

The sustain-ability ofmountainreservoirs andthe efficiencyof using har-vested waterin supplemen-tal irrigationare beingresearched inTunisia.

Greenhouse roofs provide a sizeable share ofcrop water requirements in the Al Busailiarea near Alexandria in northern Egypt.

greenhouse roof was able to provide 50%of the water requirements of the vegetablecrop allowing about 15 tonnes of melons tobe produced. In areas where water is veryscarce, and farmers otherwise have to buyin costly water to produce good crops, thismethod can make all the difference.

Runoff Inducement

Light soils with high infiltration rate do notproduce much runoff. This is a major prob-lem in many sandy areas of WANA wherewater harvesting is very much needed.Although the soil surface is sometimesmodified to induce runoff, the search con-tinues for practical, low-cost and environ-mentally friendly material that can be usedfor this purpose.

The project team in Iraq has patented aprocess to overcome the difficulties ofusing paraffin wax in inducing runoff.They have succeeded in emulsifying thewax by using low-cost additives and amachine they developed specially. Theemulsified wax, which looks like milkcould be applied easily to research plots bymeans of a small sprayer. The use of thewax p a tripled the amount of runoff fromsmall plots.

34


Emulsified paraffin wax developed in Iraq being sprayedon a micro-catchment to induce runoff near Aleppo,Syria.

Plastic sheets ensures very high runoff coefficient; it ishowever, costly and mostly used for harvesting water tocash crops.

Future ChallengesConsiderable progress has been made inidentifying the most important compo-nents of efficient schemes for harvestingand using water for crop production.Constraints to the implementation andadaptation of these schemes include: diffi-culties arising from farmers’ unfamiliaritywith the technology; conflicts and disputesover water rights, land ownership and use;and lack of adequate characterization ofrainfall, evapotranspiration and soil prop-erties. These areas need more research,based on sound hydrology and engineer-ing. Recommendations and future chal-

lenges for research and developmentinclude:• Interdisciplinary research aimed at con-

straints to the adoption and manage-ment of water harvesting at the farmlevel. This would require better charac-terization of climate, soils, adaptablecrops and socioeconomic factors.

• Definitions of optimum catchment sizerelative to cropped area under variousfield conditions. Modeling and simula-tion, backed up by carefully designedfield experiments, play a major rolehere.

• Detailed cost-benefit analysis in relationto additional crop yield and cost of thesystem. This should take into accountother benefits such as controlling soilerosion and preventing migration.

• Further work on runoff-inducementtechniques and their economical feasibil-ity. Cost of materials is a major con-straint. Research is therefore needed toidentify low-cost, more durable materi-als that are easy to use.

• Identification of water-harvesting siteswith high potential, taking into accountall hydrological and environmental fac-tors. WANA countries should preparemaps indicating the location of thesesites some of which may be used forpilot projects.

• Study of the patterns of water consump-tion of recommended crops and treesunder water-harvesting regimes. Cropsunder water harvesting undergounavoidable periods of water shortageand stress, which inevitably reduceyields. This, of course, affects yield too.Information regarding crop yield underdifferent levels of soil-water availabilityis notably lacking in the literature. Thisinformation is essential for designingsound systems and in modeling and

simulation studies of water harvesting.Figures for stress-free crop consumptionare usually not valid in the design ofwater-harvesting systems for the drierareas.

• Water harvesting is a complex interdisci-plinary technology and its various tech-niques are mainly site-specific. To betechnically and economically feasible,strong commitment and cooperationfrom the beneficiaries are required.

• Development of new techniques forminimizing evaporation and seepagelosses from storage facilities.

Finally, it should be emphasized that theultimate goal of on-farm water harvestingis a sustainable and environmentallyfriendly system of agricultural production.The aim is to complement rather thanreplace the existing water-use system.Improved systems must be socially accept-able as well as more productive.

It is highly recommended that water-harvesting interventions form part of aplan for integrated land and waterresources development and that such aplan that takes into consideration all thenecessary technical, agricultural, socioeco-nomic and institutional aspects and inputs.

35

Providing sufficient water for drinking and agricultural production in remote dry areas is one of the most seriouschallenges for water harvesting research and development, related directly to the poor.

Boers, Th. M. and J. Ben-Asher. 1982. A reviewof rainwater harvesting. Agricultural WaterManagement 5: 145-158.

Critchley, W. and K. Siegert. 1991. WaterHarvesting. FAO, Rome, Italy.

FAO (Food and Agricultural Organization ofthe UN), 1994. Water Harvesting forImproved Agricultural Production. WaterReports 3, Proceedings of the FAO Expert

Consultation, Cairo, Egypt, November 1993,FAO, Rome, Italy.

Department of Agriculture, State Development,western Australia, 1992. Water Harvesting,Improving the reliability of farm dams inwestern Australian wheatbelt. A booklet.pp. 22.

Frasier, G.W. 1990. Runoff Farming: IrrigationTechnology of the Future. Pages 404-409 inVisions of the Future. ASAE Publication 04-90. American Society of AgriculturalEngineers. St. Joseph, Michigan.

Frasier, G.W. and L.E. Myer. 1983. Handbookof Water Harvesting. AgriculturalHandbook No.600. United StatesDepartment of Agriculture, AgriculturalResearch Service, USA.

Hachum, A.Y. and J.F. Alfaro. 1980. APhysically-Based Model of Water Infiltrationin Soil. Utah Agricultural ExperimentStation Bulletin 505. Logan, Utah, USA.

Hudson, N.W. 1987. Soil and WaterConservation in Semi-Arid Areas. FAO SoilsBulletin 57. FAO, Rome, Italy.

Hutchinson, G.R., and M.A. Garduno (eds.).1981. Rainfall Collection for Agriculture inArid and Semiarid Regions. Proceedings ofa Workshop, University of Arizona, USA.Commonwealth Agricultural Bureaux, UK.

Nasr, M. 1999. Assessing Desertification andWater Harvesting in the Middle East andNorth Africa. Policy Implication Discussion

Paper No. 10. Center for DevelopmentResearch (ZEF), Bonn, Germany.

Oweis, T.Y. and A.Y. Taimeh. 1996. Evaluationof a small basin water-harvesting system inthe arid region of Jordan. Water ResourcesManagement 10: 21–34.

Oweis, T.Y., A. Oberle and D. Prinz. 1998.Determination of potential sites and meth-ods for water harvesting in central Syria.Advances in GeoEcology 31: 83–88.

Oweis, T, A. Hachum, and J. Kijne. 1999. WaterHarvesting and Supplemental Irrigation forImproved Water Use Efficiency. SystemWide Initiative on Water ManagementPaper No. 7. International Water.Management Institute, Colombo, Sri Lanka.

Prinz, D. 1994. Water harvesting - past andfuture. In: Pereira, L.S. (ed.). Sustainabilityof Irrigated Agriculture. Proceedings of theNATO Advanced Research Workshop.Vimeiro, March 1994. Balkema, Rotterdam.

Reij, C., P. Mulder and L. Begemann. 1988.Water Harvesting for PlantProduction.World Bank Technical Paper No.91, Washington D.C.

Sharma, K.D. 1986. Runoff behavior of water-harvesting microcatchment. AgriculturalWater Management 11: 137–144.

Stewart, B.A. and J.T. Musick. 1982. Conjun-ctive use of rainfall and irrigation in semi-arid regions. Advances in Agronomy 1: 1– 24.

Tauer, W. and G. Humborg. 1992. RunoffIrrigation in the Sahel Zone. Verlag JosefMargraf Scientific Books, FR, Germany.

UNDP (United Nations EnvironmentProgram). 1983. Rain and Storm WaterHarvesting in Rural Areas. TycoolyInternational Publishing Limited, Dublin.

Vorhauer, C.F. and J.M. Hamlett.1996. GIS: atool for siting farm ponds. Journal of Soil andWater Conservation 5(5): 434–438.

36


Selected Bibliography

water harvesting - doc-developpement-durable.org€¦ · key words:water harvesting, drier...

Documents