Home-immediately access 800+ free online publications. Download CD3WD (680 Megabytes) and distribute it to the 3rd World. CD3WD is a 3rd World Development private-sector initiative, mastered by Software Developer Alex Weir and hosted by GNUveau_Networks (From globally distributed organizations, to supercomputers, to a small home server, if it's Linux, we know it.)ar.cn.de.en.es.fr.id.it.ph.po.ru.sw

CLOSE THIS BOOKDesign and Operation of Smallholder Irrigation in South Asia (WB, 1995, 134 p.)
Chapter 6 - Irrigability
VIEW THE DOCUMENTSoil surveys and land classification
VIEW THE DOCUMENTSoil constituents
Soils problems on irrigation
VIEW THE DOCUMENT(introduction...)
VIEW THE DOCUMENTSaline and alkaline soils
VIEW THE DOCUMENTGypsiferous soils
VIEW THE DOCUMENTAcid sulphate soils (cat clays)
VIEW THE DOCUMENTLateritic soils

Design and Operation of Smallholder Irrigation in South Asia (WB, 1995, 134 p.)

Chapter 6 - Irrigability

Soil surveys and land classification

Soil surveys classify the physical and chemical characteristics of the soils of an area. Irrigability surveys (land classification) add a further dimension, i.e. the potential economic productivity of the lands of the area in question. Land classification came into vogue during the major campaign of irrigation development in the Western United States, carried out by the Bureau of Reclamation in the early 1900's. At one point Congress decided that in some cases public funds were being spent on bringing irrigation to lands which, for reasons of soil deficiency or other factors, could not provide a reasonable living to the irrigator nor an economic return on the capital expenditures involved. It was then decreed that future project proposals should include information on the economic irrigability of the lands concerned. The system of irrigability classification evolved by the Bureau at that time has remained a key feature of irrigation planning, and is still widely used. The term "land classification" was -employed, rather than "soil classification. as other factors aside from the type of soil (pedology) were involved, including the cost of bringing water to the particular lands concerned. Criteria for a number of factors including soil depth, infiltration rate etc. were established, also specifications for the field surveys. The soil surveyor who previously had confined himself to soils now became a member of a multi-disciplinary team mapping economic land capability. When the Bureau extended its activities to the investigation of overseas projects the same approach was employed, although irrigability criteria were modified to suit the local situations.

Irrigation of smallholder areas involves the same soil factors (pedology) as does irrigation of larger holdings. However, with respect to irrigability classification, there are substantial differences, the principal one being the prospective input of the cultivator to the development of his holding. This may far exceed conventional economic limits. The wellbeing of the smallholder and his family is irrevocably determined by the productivity of his holding, and he has little other opportunity for bettering his situation than improvement of his land. The low opportunity cost of labor of the farmer and his family, given time, can accomplish wonders. An apparently barren boulder-strewn area with unproductive subsoil of a few inches depth can be changed, with patience and labor, into a fertile field. Steep slopes can be converted to small terraces, each with carefully constructed stone pitching.

It is not intended to convey that conventional economic irrigability classification is irrelevant in smallholder areas. It is very relevant in determining whether or not to bring irrigation into a particular project area. However, once a decision has been made to proceed with a project, irrigability classification is less important in determining whether to provide service to local areas with particular deficiencies. A small cultivator within the project perimeter but unfortunate enough to have relatively poor soil would be doubly unfortunate if he were to be excluded from supply of water. Provided that there is the technical possibility of substantially improving his holding, particularly under irrigation, it can be argued that equity demands that the cultivator in question be supplied with water and given the opportunity to make that improvement.

Irrigability classification involves making certain assumptions regarding the irrigation practices which will be followed. A case in point is the classification of lands as suitable, or unsuitable, for cultivation of wet-land rice (paddy). This involves consideration of infiltration rates. Rates of more than 2-3 cm per day are usually considered excessive for wet-land (flooded) paddy. Such lands would normally be classified as unsuited to that crop. However, smallholders in traditionally rice-eating areas may irrigate paddy in soils with infiltration rates ten times that amount, using semi-wetland techniques, i.e. without continuous flooding, rather than growing a more appropriate crop requiring less water.

Incorporation of soils data in an irrigability classification, without also reporting on the soils data separately, may be quite appropriate in a feasibility report prepared by a major organization which is also responsible for detailed design and execution of a project, as well as its investigation. However, where project design is subject to review and possible modification by agencies other than the one which carried out the original field investigation, such as prospective financing institutions, a separate soils survey report should also be provided. It is not readily possible to extract basic pedological data from an irrigability report (to "unscramble the omelette"). Pedology is basic, while irrigability classification involves judgement on many factors other than those related to soils, judgements on which other agencies subsequently involved may not always concur.

Irrigation of many soils, including the commonly-occurring silty or sandy loams, is relatively straight forward and does not call for extensive knowledge of pedology on the part of the irrigation engineer. Problem soils may be encountered, however, and these present the engineer with the difficulty that soils science is a complex subject, obscured by an esoteric nomenclature ("taxonomy") which is intimidating to other than a soils scientist. There is no middle ground in the literature, which either stops at simple soils water relationships, or requires a depth of background in soil chemistry and physics which only a soils scientist would have time or inclination to acquire.

Soil constituents

The following brief description of the principal factors influencing the behavior of soils under irrigation is given as background to discussion of particular soils problems. For more detailed treatment, reference should be made to Richards (1954), the classic original text on salinity and alkalinity, and to Tanji (1990).

Formation of soils from the parent material produces an array of constituents ranging from relatively unweathered resistant components (notably silica) to fully weathered material, part of the latter being in the form of clays. Organic material is usually also present. From the agronomic viewpoint, the soil may be grouped into relatively inert components, material still in the process of breaking down (a source of nutrients), days, and organic material. Soil moisture is also an essential ingredient.

The clay fraction plays a very important role, due to its ability to absorb ions on its surface. Positively charged ions (cations) of principal significance are calcium, magnesium, sodium, and to a less extent potassium. Although tightly bonded to the clay mineral by electrostatic forces, they may be exchanged with other cations in the soil solutions and thus constitute a source of plant nutrients. The adsorption sites not occupied by these cations may be occupied by hydrogen ions. The ability of a soil to absorb cations is referred to as its Cation Exchange Capacity (C.E.C). The extent to which that capacity is occupied by calcium, magnesium, sodium and potassium is termed the percentage of base saturation.

As cation absorption is a surface phenomenon it is primarily associated with clays, which due to the lamellar nature of the clay mineral have very high specific surface. The "2:1" clays such as mon-tmorillouite and illite which have both "internal" and "external" surfaces have C.E.C of some 100 milk equivalents per 100 g. The "1:1" clays such as kaolinite have C.E.C of 10 to 15 meq/100 g. Colloidal organic matter (humus) has C.E.C of up to 200 meq/100 g. Fine textured non-laminar minerals (e.g. fine silts) also have adsorptive capacity, but to a much lesser degree than clays.

Values of Cation Exchange Capacity for composite soils commonly range up to 30 milliequivalents per 100 g, the actual figure depending upon the clay content. A relatively high value of C.E.C., particularly with a high degree of base saturation, usually signifies high fertility. However, soils with C.E.C as low as 4 or 5 meq/100 g can grow irrigated crops provided that sufficient fertilizer is applied and that the interval between irrigations is short.

The undesirable cation to have on the exchange complex, if in excess, is sodium. Particularly at low levels of soil moisture salinity, sodium on the adsorption complex above a certain limit may hydrolyze, resulting in an alkaline condition. This can cause deflocculation and dispersion of the clay, with drastic reduction in soil permeability, hence the interest in the percentage of sodium on the exchange complex and in means of controlling it. The concentration of a particular cation on the exchange complex is influenced by the concentration of the same, and other, cations in the soil moisture with which it is in contact. In the long-term an equilibrium is achieved. The equilibrium concentration of sodium on the complex, corresponding to prolonged irrigation with water of a particular chemical make-up, is obviously a matter of primary importance. The relationship is an empirical one, which has been determined by study of a wide range of soils and irrigation waters. It relates the value of a function referred to as the Sodium Absorption Ration (S.A.R.) of the saturation extract of the soil moisture, to the Exchangeable Sodium Percentage (E.S.P.) on the soil exchange complex. It is noted that soil moisture is referred to, rather than irrigation water, as the exchange complex is in contact with soil moisture, not directly with irrigation water (other than at ground surface). The concentration of cations in the soil moisture at plant root level is two to three times that of the incoming irrigation water (averaged over a period of time), due to extraction of water by the plant. In determining the S.A.R. value of the soil moisture from data on the chemistry of the irrigation water, this increase in cation concentration is taken into account.

It is noted that the term "Sodium Absorption Ratio" causes some confusion as "absorption" occurs on the soil complex, not in the solution. However, it is the term customarily applied to the above-defined function of the soil solution.

Determination of the Exchangeable Sodium Percentage on the exchange complex, the associated S.A.R. of the soil moisture, and the S.A.R. of the proposed irrigation supply, is of interest for three reasons. First for classification of the soil in terms of its alkalinity hazard, second for assessment of the effect on the soil of long-term application of the particular irrigation water proposed to be used, and thirdly for design of remedial treatment if needed.

The adjustment of the E.S.P. of a soil to come into equilibrium with the S.A.R. of an irrigation supply can be a very slow process due to the large quantity of cations held on the exchange complex compared with the relatively small concentration in the irrigation water. In fact, amelioration of an alkaline condition (as distinct from saline) in the course of normal irrigation, or by leaching with irrigation water, is unlikely to be rapid enough to be of practical significance, except under special conditions (e.g. presence of gypsum or lime in the soil). However, in the opposite circumstances in which the nature of the irrigation water is such that it slowly increases the amount of sodium on the complex this would cause serious alkalinity over a long-term period, and historically has done so in some areas. Chemical and other remedial treatment of alkaline soils is discussed in the next section.

Soils problems on irrigation

In the following discussion a number of soils which present particular features or problems in irrigation development are described.

Saline and alkaline soils

Soils are classified with regard to salinity and alkalinity in accordance with the conductivity of the saturation soil moisture extract and the Exchangeable Sodium Percentage on the exchange complex. The classification is nominal only, as the performance of a soil from the agricultural viewpoint is influenced by other factors in addition to conductivity and E.S.P, including soil texture and the sensitivity of the particular crops to be grown. The classification is as follows:


Exchangeable Sodium


Percentage at 25°C

Non-saline, Non-alkaline

Less than 4

Less than 15

Saline, Non-alkaline

More than 4

Less than 15

Saline, Alkaline

More than 4

More than 15

Non-saline, Alkaline

Less than 4

More than 15

The continued application of even relatively high-quality irrigation water in the absence of internal drainage sufficient to remove the incoming salts will inevitably result, eventually, in a saline soil condition. This has occurred in a number of historic areas in which agriculture ultimately has had to be abandoned. A number of such areas still await reclamation.

A purely saline condition requires leaching only. If natural drainage conditions are inadequate to remove the leachate, internal drainage must be provided. Attempts to remove salt from the cultivated depth of soil by leaching laterally into furrows have not proved successful, as capillary action brings salt to the surface from the un-leached subsoil. The depth of leached soil must be greater than the height of capillary rise, which usually means more than a meter. In the simplest situation leaching is down to a watertable which is at considerable depth, (and which will remain so). Otherwise, internal drainage may have to be provided, either by tubewells which hold the watertable at sufficient depth below the surface ("vertical drainage"), by a perforate dpipe drainage network about at 2 m depth ("tube-drainage"), or by open drains of sufficient depth. Each solution has its limits and its problems. The tubewell system requires the existence of a sub-surface horizon, virtually an aquifer, of sufficient transmissivity to permit extraction of water at reasonable cost. Tube-drainage requires sufficient permeability of the soil horizon being drained to permit economically practical spacing of the tube network. Open drainage, which might appear at first sight to be the most straight-forward solution, particularly for developing economics, requires a depth and spacing of drains consistent with the permeability of the soils being drained (the open drains are required to function as internal drains, not as surface drains). In many cases such depth and spacing would be impractical, e.g. 3 m depth at 100 m spacing, in view of the often major difficulties of maintaining deep open drains, and the substantial surface area which they would occupy.

Tube-drains, open drains and tubewells may require pumping if gravity out-fall is not available. The principal problem of all three systems may be the disposal of the saline effluent. Where the salinity problem is local, the effluent may be returned to the irrigation system, or to a stream, where there will be adequate dilution. However, where the salinity problem is regional it may not be acceptable to discharge saline effluent into a river, or to re-cycle it into irrigation. Other means of disposal are by construction of an out-flow canal to the ocean, or by evaporation ponds. An ocean outflow canal is being constructed on a heroic scale in the Indus basin, and could eventually be required on an even more heroic scale, including major pumped-lift, in an adjacent area. Evaporation ponds are technically feasible but the surface area required can be considerable, as the evaporation rate reduces considerably (compared with fresh water) as the salinity in the pond rises. Desalination, now being adopted in the western United States, is not considered an economically practical solution for the foreseeable future in developing countries.

Where groundwater salinity has not yet risen to unacceptable levels watertable elevation may be controlled and natural leaching promoted by tubewell irrigation, including cultivator-owned shallow tubewells. However, in an area where soil salinity is already depressing crop yields, cultivators are unlikely to take up tubewell installation on a significant scale.

The most practical solution to a salinity problem in some areas (as also waterlogging) may be to reduce the rate of inflow of irrigation water into the area to an amount such that the watertable remains low enough for natural leaching to occur. Watertable elevation is determined by watertable gradient, which is influenced by rate of seepage from irrigation. Seepage from canals is also a contributing factor, and this may be a reason for canal lining in some situations.

In some areas, an alternative solution to removing the saline condition is adopting appropriate cultivation practices and selection of crops. Frequently this is the only immediate course available to the small cultivator, in some cases with considerable success. The balancing act performed by cultivators in the lower Nile delta, depressing the salinity level with a paddy crop and taking a follow-on cotton crop before the salinity rises again is an example, but probably applicable only to their particular soil situation. The ultimate example of living with salinity is conversion from agriculture to pond fish culture, or prawn-culture, in such areas.

Reclamation of purely saline soils has its problems, but the treatment of saline-alkaline soils is complicated by two further factors. First, the permeability of the soil may already be very low, due to de-flocculation of the clay mineral, or it may become so as soon as leaching lowers the salinity. Second, as discussed earlier, alkalinity cannot generally be removed by simple leaching. Exchange of sodium on the exchange complex, by calcium, must also be achieved.

Reclamation of alkaline or saline-alkaline soils can range from the relatively simple to the virtually impractical. Drainability, with a sufficient degree of permeability and sufficient watertable depth to permit downward leaching of salts, including displaced sodium, greatly facilitates the process. Progressive application of gypsum (calcium sulphate) and leaching, or cultivation of paddy to provide leaching, may be all that is required.

However, where permeability is already very low, either because of dispersion of the soil due to alkalinity or due to the inherently fine texture of the soil, getting water into the soil and getting the leachate out can present considerable difficulty. If dispersion is the problem the classic solution is to begin leaching with water of sufficiently high salinity to de-flocculate the soil (if such water is available), at the same time providing gypsum for displacement of sodium. After the latter process has proceeded far enough salinity may be reduced by leaching with water of low salinity. However, if texture is the problem (e.g. a heavy clay soil), rather than dispersion, leaching with high salinity water will not improve permeability, and a difficult drainage problem is presented.

In such circumstances solutions may be available to the small cultivator, with his low opportunity-cost labor, which would not be economically viable on a mechanized scale. An example is the practice of some cultivators in portions of the lower Nile delta. The surface soils are saline-alkaline, underlain by unripe near-impermeable silty clays. Watertable is high. As there is virtually no downward movement of water any leaching has to be laterally. The cultivators consequently dig ditches of 1.5 to 2 m depth at as close as 25 to 30 m spacing. There is sufficient lateral gradient into the ditch to provide appreciable water movement. Gypsum, brought from supply depots by pannier-laden donkeys, is ploughed in each year, and each year the ditches are in-filled with reclaimed surface soil and dug again in a new position. Progressively, and at great labor, the entire area of the holding is eventually trenched and reclaimed in this manner. This is obviously not a procedure which would appeal to cultivators everywhere. The equivalent mechanized approach, also practiced in the area referred to, and with limited success, is to employ a heavy tractor-drawn chisel plough and to sprinkle gypsum into the temporary slot opened behind the chisel. The slots, at some 2 m spacing, extend between open collector ditches at about 150 to 200 m. The hope is that each slot, in-Sued with gypsum improved soil, will provide a permanent conduit for leaching from the area. Gypsum is ploughed in across the area as a whole. Mole plowing has also been tried in this area, as a means of providing temporary drainage, rather than the chisel plough slots. However, the drainage holes formed by the mole plough have very short life except in very special soil conditions. It is noted that the spacing required for permanent tube drains in the particular conditions referred to would be 10 m to 20 m, which would be economically unattractive.

A degree of reclamation of otherwise intractable heavy alluvial/marine clays has been provided in an area near Bangkok, by construction of raised beds. The beds, about 10 m in width, are separated by excavated water-ways, which are the source of the soil for constructing the beds. The water-ways serve both as drains and as irrigation supply, water being applied to the beds in this case by spray from a gasoline-driven pump mounted on a small boat which travels up and down the water-ways, propelled by jet reaction from the pump nozzle. After forming the raised beds the soil, initially in heavy clods, is left to mature by weathering for one or more seasons before cultivation. The raised-bed system, with bed surface about one meter above adjacent water-level, provides internal drainage, although to limited depth. The beds are devoted to raising of vegetables for the adjacent Bangkok market. The system is quite effective, but at the cost of heavy manual labor in forming the beds, and requires considerable skill in subsequent soil management. It is not a generally-applicable solution to that type of soil situation. To summarize, reclamation of alkaline soils can be a difficult problem, and may not always be practical. Diagnosis and development of appropriate treatment calls for special expertise in soils chemistry and drainage. Cultivator ingenuity and labor has been an essential part of the solution in some situations.

Expansive days

These are variously termed Black Cotton soils, cracking clays, or vertisols. They occur widely in areas of relatively limited monsoonal rainfall. They range from heavy clay to silty clay, the clay mineral being principally montmorillonite (clay content is commonly 20 to 40%). Calcium usually predominates on the exchange complex, and calcium carbonate nodules (kankar) often occur throughout the profile.

A notable feature of these clays is the high degree of expansion and shrinkage on wetting and drying, causing conspicuous cracking in the dry season. Cracks may be as wide as two centimeters, and up to one meter in depth. The soils, under irrigation, produce a variety of crops including food-grains, cotton, and sugar-cane. Problems have been encountered, however, with wet-land paddy due to deficiency in available phosphorus under saturated (anaerobic) conditions. The soil management problem sterns from the very sticky unworkable nature of the soil when wet, and its very hard intractable nature when dry. The range of soil moisture content under which conditions are suited to cultivation is narrow. In low-intensity rainfed agriculture the problem of cultivation is minimized, but it may be a limiting factor in introduction of intensive multiple cropping under irrigation. It is noted that certain of these clay soils, but by no means all, have the very beneficial characteristic of "self-mulching", i.e. shrinkage near the surface produces a network of very fine cracks and pea-size particles, a fine natural filth. In other areas the cracking is massive.

From the irrigation engineering viewpoint, there are two problems with these soils. One stems from the effect of expansion/contraction on structures. The other, in some areas, is their extreme erosibility. The pressure which is exerted by such clays, if expansion is restrained, can be destructive. If lined canals are to be built, over-excavation and replacement with granular non-expansive material (if available) in the vicinity of the lining may be necessary. In some areas a horizon of suitable semi-granular non-expansive material (termed "murrum in India) occurs between the weathered parent rock and the over-lying day soil, but this is not the case everywhere, and haulage of non-expansive fill may be a major item of cost. Use of a reinforced concrete flume with free-standing walls, or supported on pedestals, may be resorted to, in order to avoid the expansion problem. In addition to expansive pressure, shrinkage cracking can present a threat to small and medium-sized hydraulic control structures, which may be completely by-passed by flow through massive shrinkage cracks. To avoid this problem unusually extensive cut-off walls may be required, or provision of a length of upstream lining (duly protected against expansive pressures).

The susceptibility of expansive clay soils to erosion varies, for no immediately apparent reason, from moderate to very great. With the most erosive clays, runoff from rainfall on a 1-2 % slope can cause heavy sheet erosion, the outflow being a dense slurry which can block drainage channels overnight. Lateral erosion due to runoff down the banks of an unlined canal can change its original trapezoidal section into a shallow saucer-shaped depression in one season. The only solution is to line the canal, posing of course the problem of expansive pressure on the lining. Roads in such clays are quite impassable in the wet season, unless given substantial surfacing with granular material. All soil conservation works, particularly contour bunding, must be fully protected against erosion, preferably with vegetative cover.

Gypsiferous soils

While gypsum is highly beneficial in the reclamation of alkaline soils, it can become a problem where it occurs in excess. The problem can be either agronomic, or technical relating to irrigation distribution.

Gypsum (hydrous calcium sulphate, CaSO4 2H2O) occurs in arid or semi-arid situations. In concentrations of up to 25% in the soil, and if in finely divided form, gypsum has little adverse effect on crop yields. Soils with up to 50% gypsum are in fact cultivated, although at reduced yields, in part due to fixation of phosphorus in such soils.

The problem in irrigation of gypsiferous soils stems from the solubility of gypsum. This can cause irregular settlement of irrigated fields, heavy leakage from unlined channels due to formation of solution paths down to the watertable, and disruption of lined channels. The smallest seepage from a lining, at a joint or crack, may cause a solution cavity to form behind the lining, resulting in eventual collapse of the lining into the cavity. The failure is progressive, differential settlement behind the lining causing cracking, with increased leakage and further solution settlement. The problem may be aggravated through attack on the lining by the sulphatebearing ground water.

An expedient adopted in some gypsiferous areas is to construct the channel as a reinforced flume, elevated on pedestals. The pedestals are located mid-way between the joints in the flume so as to be unaffected by solution due to joint leakage. Another system incorporates a heavy-duty plastic or composite lining behind the concrete inner lining, the plastic sheet forming the primary water barrier. The latter must be proof against rodent and termite attack and must be sufficiently flexible to accommodate deformation due to minor settlement.

Acid sulphate soils (cat clays)

This is one of the most difficult types of soil from the water management viewpoint. It commonly occurs in tidal mangrove areas. The notable feature of the soil is the occurrence of pyrites (iron sulphide, FeS), formed by bacterial reduction of sulphates from sea-water or brackish water, under saturated anaerobic conditions. In reclamation of these soils, usually for rice cultivation, lowering the watertable can result in oxidation of the pyrites to form sulfuric acid, also consequential release of free aluminum ions which are toxic to the crop.

Water management under these conditions involves very careful control of watertable elevation, through regulation of irrigation and drainage, to avoid such oxygenation. Management of water levels to the required closed tolerances can be made the more difficult in such areas due to settlement of such organic soils when drained and also to tidal variation in drainage outfall levels.


Podzols are a classic case of a soil which may be supporting a stable vegetative cover (commonly forest) in the natural state, but which is poorly productive when the natural cover is disturbed. The characteristic feature of podzols is leaching of the top-soil by humic acids, generated from rotting vegetative material on the forest floor. The acid leaching process breaks down all susceptible minerals, eventually including clays. The leachate transports the product down into the subsoil, leaving essentially a washed fine silica sand topsoil, with humus cover.

The system is in fine-tuned ecological balance. Disturbance, particularly removal of the vegetative cover, can invite disaster. The virtual absence of clay mineral and the fine, often powdery, texture of the topsoil makes it highly susceptible to erosion. It is also of low fertility with very restricted potential for agricultural development.

Lateritic soils

Lateritic soils are also the product of leaching, but differ from podzols in that humic acids are not involved in the leaching process, and the break-down of soil minerals is not as complete. Lateritic soils occur extensively in upland tropical areas on acidic parent rock, including granite. They are of characteristic red-yellow color and granular in texture, with a small amount of clay mineral. They are readily erodible, and consequently may be of shallow depth (10 to 20 cm), particularly in undulating terrain. However, depth may be a meter or more in higher rainfall areas where weathering has proceeded more rapidly. The soil transitions into less weathered, fragmented, parent rock.

Lateritic soils have moderately low fertility and low moisture retention capacity. Particularly where of shallow depth they would be classified, by most criteria, as poorly suited to irrigation. However, such soils occur extensively in some areas, including the granitic portion of the Deccan, in central India Under rainfed conditions cultivation is precarious, usually limited to the monsoon season, and generally at subsistence level. Where water can be made available multiple cropping becomes possible and productivity is substantially increased. The shallow depth and low moisture retention capacity remain a problem, however, necessitating a short irrigation interval. Land shaping for irrigation in such conditions requires considerable care, in order to avoid removal of topsoil.

An important factor in management of such soils under irrigation is the role of the subsoil, both in storage of soil moisture, which is available to plants by capillary rise or through root penetration, and as a potential contributor to deepening of the topsoil by mixing in the course of cultivation over the years. Work on agricultural research stations in areas of these soils has demonstrated the effectiveness of this process.

Dune sands

Due to the process of wind-loom movement of dunes ("saltation"), dune sands fall within a narrow range of particle size. Some 90% or more of a typical dune is sand, in the range of 0.1 to 0.5 mm. The remainder is fine silt, with a very small proportion of clay. Fertility and water retention capacity are low. Infiltration rate is very high, with consequent problems in irrigation application, also in leaching of fertilizer. This is obviously not a soil which would be given priority in selection of areas for irrigation development, if there were other options. In some situations, however, other options are not available, an example being the lower end of the Rajasthan Canal system in India. Fortunately the fine silt and colloidal material brought in with water from the supply canal, in the Rajasthan case, results in substantial improvement in the character of the virgin dune sand, and viable levels of productivity can be obtained. However, several factors remain, which make irrigation management in such an area difficult.

Sand is obviously highly susceptible to wind erosion. The same forces which formed the dunes will immediately begin to re-form dunes on levelled areas, if left unprotected. The most effective protection is a crop, or crop residue from a harvested crop. This implies that dune areas should be levelled only at a rate with which the cultivator can keep pace, i.e. keep under active cultivation. It also implies that, for an extended period, an area newly coming under irrigation will still have undeveloped dunes as islands within the already levelled and cropped portion of the area. From these dunes and dunes around the permanent perimeter of the irrigation area, hot wind-blown sand can destructively erode adjacent crops, particularly at the seedling stage. There is no simple solution to this problem in an area newly under development, as the only remedy involves planting of wind-breaks or secondary cover-crops on the dunes, and this may require a limited amount of irrigation by pumping and delivery by hose or movable sprinkler system. Such facilities are unlikely to be available to a small cultivator newly arrived in the area.

The very high infiltration rate of sand poses a problem in ensuring uniform distribution of water on the field. The solution adopted by some experienced cultivators is to divide a field (nominally a 50 m x 50 m basin) into narrow strips 2 m wide, by temporary ridges, and to direct the whole flow of some 2 ft³/sec (56 liters/sec) into each such strip in turn, completing the delivery in a matter of three or four minutes. The imponded water then infiltrates uniformly. Field application efficiencies as high as 75% can be obtained on dune sands which have been under canal irrigation for several years.

Sprinkler irrigation is a classical method of water application in such high infiltration rate soils. It is employed in large scale irrigation of desert areas. It is not, however, a solution readily available to the small cultivator, nor is it particularly efficient in areas of frequent high winds.

In the tertiary distribution system in dune-sand areas the problem is again wind-blown sand. An unlined tertiary, or water course, has two disabilities in such circumstances. First seepage losses are very high, compounding problems of rising watertable. Second, the channel can be filled and obliterated overnight in a sand storm. A lined channel can also be filled, but the in-filling sand can be removed and the channel restored to use. Clearing an unlined channel constructed in sand poses a problem as there is no evidence of when clearing has reached the original floor or sides of the channel; the original geometry of the channel is lost.

The use of covered lined channel, or pipe, can nominally avoid the problem of windblown sand. However, water from the supply canal carries sand and silt, which can deposit in the covered lined channel or pipe, as the flow velocities are necessarily small in this situation, due to low head and relatively flat gradient. A sectionalized removable cover on a lined channel would permit clearing of such a channel, but this procedure is not applicable with a pipe. A covered desilting cistern at the intake could be a solution, but its maintenance would need to be assured.

With regard to lining of small channels in dune sand areas, the lowest-cost solution, a trapezoidal section with sides supported by the fill, has been found troublesome, as the supporting sand may be eroded by wind, resulting in collapse of the linings. The alternative of a rectangular channel with structurally self-supporting sides is more satisfactory in this respect. Responsibility for cleaning of sand from small channels (water courses and minor canals) is a critical question. In view of the very considerable length of channel involved, particularly wafer courses, it is highly desirable that such maintenance be carried out by the beneficiary cultivators. This does not present a problem in a well developed area with well organized cultivator groups.- However, it can be a considerable problem in the early stages of settlement when some cultivators have not yet taken up their allotments, and a cultivator at the downstream end of a channel kilometers in length may find that the channel is blocked upstream where it traverses as yet unoccupied holdings. A dune sand area can be a very hostile environment for a cultivator, particularly in the early stages of its development. The farmer deserves particularly close institutional support.