Types of Irrigation Systems

In engineering terms, there are many types of irrigation systems, but the most common distinctions are full versus supplemental systems, modern versus traditional (informal) schemes, and large versus small schemes. More than one type of system may have a place in a national water strategy; accordingly, all types should be reviewed in a national water assessment.

The word informal refers to traditional practices such as recessional (décrue) irrigation following the receding of annual flood waters, sometimes enhanced by simple structures such as polders and river intakes, or small rainfall catchment structures. Spate irrigation, assisted by simple earthen diversion structures, is common in places like Yemen. In spite of the pace of construction of irrigation systems in the past, traditional systems still represent the dominant form of irrigation in some areas. For example, of all irrigated rice grown in the 19 principal rice-producing nations of Africa, as much as 72% has been produced via traditional practices.54 In general terms, it cannot be said that informal or traditional irrigation systems55 are necessarily preferred to modern, engineered systems, or vice versa. An evaluation has to be made of the circumstances surrounding each case. What is needed is to study more widely the experiences with traditional irrigation, and to take into account objectively the option of expanding it for some areas when national irrigation strategies are designed. To date, international agencies have tended to ignore the potential which could be offered by modest improvements of the traditional systems, and this oversight should be corrected. Moris and Thom have explained the situation in the following terms:

. . . in many African countries, irrigation systems are polarized between a few, larger-scale government schemes and a number of very small independent irrigators. That latter practice various 'traditional' techniques, employed with hardly any outside assistance. These days they are incorporating some modern (or 'introduced') equipment, notably small pumps, but their whole mode of financing and operation is very different from that employed on the large-scale, official schemes. . . . The project documentation which one finds in donors' files tends to represent these official schemes.. . . Farmers' own efforts to control water are usually on a very small scale. The purchase of a single pump may represent the culmination of a major effort among subsistence farmers At this extreme, few expatriate engineers would consider such water management practices as 'irrigation'. Nonetheless, they achieve the same objective as the much more expensive, imported technologies used on official schemes. .. . The tremendous differences between the two main types of irrigation . . . has inhibited any sharing of experience or assistance between them. It has proven very difficult for government agencies and

54. Calculated from figures presented in J. R. Moris and D. J. Thom, 1991, p. 41.

55. In some taxonomies, traditional irrigation systems fall under the rubric of small-scale systems. The FAO, for example, says 'Small-scale programs include a diversity of technologies such as water harvesting, well development, river offtakes and use of wetlands' (FAO, 1993, p. 287).

external donors to work with Africa's small-scale systems, though there are a few instances of partial success (in Senegal and Tanzania). The extreme duality which characterizes the irrigation sector in most sub-Saharan countries is unfortunate. It makes it unlikely that successful smaller projects will evolve into medium-scale operations which might combine high farmer involvement with economies of scale in water management.56

It is not necessary to travel to remote areas to see traditional irrigation practiced. Where the Niger River passes through Bamako and its environs, for example, many small farmers can be seen taking water from the river in gourds and plastic buckets to water vegetables planted a few feet from the river's edge. A small number of them have invested in pumps and hoses.

A process of development of an irrigation strategy which incorporates greater participation on the part of farmers may lead, in some cases, to greater emphasis on expansion and improvement of traditional irrigation systems. For strategic planning, the significance of traditional irrigation resides in the following: (a) its value should be recognized, particularly when designing projects (usually dams) that may result in the reduction or elimination of existing opportunities for its use; (b) technical possibilities of incremental improvements in traditional irrigation, for example, by supplying more pumps or building small polders, may exist and may be considered where land tenure and agronomic, economic and social conditions are appropriate; (c) in considering expansion or improvement of traditional irrigation, priority should be given to involving in the discussions the farmers who are presently practicing it, soliciting their ideas for how to increase its effectiveness and returns. The objective in this case is to improve traditional systems of irrigation without undermining the inherent strengths that caused them to be developed in the first place.

Supplemental irrigation is used to compensate for dry spells during the rainy season, or to prolong the season of water availability for crops. It usually is based on pumping, whether from surface or ground water. Its desirability is dictated

In some of the literature there is confusion between irrigation systems or schemes, on the one hand, and irrigation methods, on the other. In a very simple classification, irrigation methods can be put into two categories: surface flow and pressurized irrigation. Surface flow irrigation can be provided in many ways (basin, furrow, borders, etc.) whose common characteristic is that water is applied at a certain point on a parcel of land and from there it moves on the surface over the rest of the parcel. Until the development in the 20th century of pressurized techniques, surface flow was the only method used in history, and it still is the most widely used. Although it has marked disadvantages such as low water application efficiency, the needfor land leveling, difficulties in applying the correct volumes of water in the right frequency, and a high demand for labor, it is expected that it will continue to be the most commonly used method by far.

Pressurized irrigation, sometimes called micro-irrigation, can be divided into sprinkler and localized irrigation techniques, with the latter referring mainly to drip and microsprinkler options. When they are well designed and managed, both pressurized techniques allow higher water application efficiency than surface flow methods do. Localized irrigation can apply water and fertilizers on a daily basis according to crop needs, thus promoting higher crop yields and quality, and labor is saved as well. The disadvantages of pressurized methods include high investment costs, the necessity of energy, and the use of sophisticated components that are not always available. For these reasons, the use of pressurized irrigation normally is limited to high-value crops such as fruit trees and vegetables.

by climatic conditions; in regions where the rainy season often is irregular, it can play a vital role in preventing severe damage to crops. Worldwide, most irrigation is supplemental to one degree or another, except in very arid climates and in greenhouses. Supplemental irrigation can be critical not only for increasing the volumes of production but also for ensuring the quality of products like fruit and vegetables, for it enables farmers to control the timing of water deliveries to the plants. In view of the increasing importance of product quality for export, and even for domestic markets in developing countries, this contribution of supplemental irrigation can be significant for increasing the incomes of farm households.

Commercial farmers have played an important role in developing private systems of supplemental irrigation. A noteworthy example is that of coffee farmers in Kenya. The case for supplemental irrigation has been stated in the following terms:

In any farming system dependent on rainfall where the mean total [precipitation] lies near the boundary required for successful cultivation, minor deficits. . . can have a dramatic impact on crop yields. ... In policy terms, what is important is to realize that the potential benefits from irrigation may be just as great in supplementing rainfed cultivation as in supplying all plant water requirements under 'total' irrigation in a semiarid environment.57


Commercial farmers in East and Southern Africa have usually found it necessary to develop supplemental irrigation in order to achieve reliable crop yields. If so, the same need probably exists within smallholder farming. Regularization of rainfed crop returns by stabilizing planting dates and eliminating the within season dry spells might represent a more desirable (and water conserving) objective than

'full' irrigation with its heavy water demands. The main problem is, of course, the high cost of present technologies for achieving this objective.. . . We do not yet have answers, but the need to pay more attention to partial irrigation seems obvious. . . .58

The experience of the Machakos District in Kenya provides an example in which farmers led the way in developing traditional irrigation:

As a technique for increasing water supply, water harvesting methods work in the Machakos District in Kenya [via] the building of terraces to control soil erosion and to slow runoff, thus increasing the supply of water in the crop root zone. These measures have increased crop yields and production. Perhaps the most significant feature of this achievement was that it resulted almost entirely from farmers' decisions to invest their own resources in improving the natural resource base and other aspects of farm enterprise. The government contribution -by no means trivial - was to improve farm-to-market roads, both within the region and between the region and Nairobi (N. P. Sharma, et al., 1996, p. 47).

Within the category of modern irrigation systems, the wisdom of large-scale versus small-scale irrigation projects has been the subject of considerable discussion in recent years, with the consensus opinion inclined in favor of smaller projects but not excluding the larger ones under favorable circumstances. Moris and Thom reason as follows:

If in Africa small-scale projects are not necessarily cheaper to build, they are nevertheless easier to withdraw from; managerial assistance by an NGO rather than the government is more feasible; field layouts can be more adapted to farmers' needs; and there is at least a theoreti-

5?. J. R. Moris and D. J. Thom, 1991, pp. 16-1?. 58. Op. cit., p. 572.

cal possibility farmers will be more involved and consequently more committed. We recommend, therefore, a bias towards assisting small-scale projects and technologies. .. . This recommendation. . .. ignores the fact that schemes requiring large reservoirs or major canals are bound to be large-scale in nature [and] a pervasive opinion . .. that small projects are just as demanding of supervision and management as are large ones. While this may be true, the consultants drawn upon in this study were nearly unanimous that in Africa smaller, flexible projects on average outperform the large ones. . . . these arguments do not rule out experimentation within large systems to decentralize scheme functions and increase farmer participation, e.g. as the Dutch have attempted in the Office du Niger.59

The international action plan for irrigation that is guided by the FAO (IAP-WASAD) also has identified small-scale irrigation as one of its priority areas. The requirements for successful small-scale irrigation, as put forth in that plan, include adequate technical advice, more participatory approaches to management of systems, and strengthened and more accountable public sector institutions.60

Sharma et al. acknowledge the greater success rate of the smaller projects but point out that larger ones can be successful also:

While the majority of irrigation success has come with small and medium-scale schemes, this does not mean governments should completely ignore the value of large-scale irrigation projects. Through a coordinated effort, Nigeria has irrigated 70 000 hectares of land. . . and Sudan, through careful upstream management to control sedimentation, has operated its Sennar dam for seventy-six years with only a 56 percent loss in total capacity. . .. Factors which hold true for other development efforts also hold true for irrigation projects of such magnitude: the availability and extension of a comprehensive package of improved technological messages, liberalization of crop marketing and processing, land tenure security, improved roads, bureaucratic reform, government commitment, focus on well-defined goals (in this case, water management), partnership with the individual farmer-producers, and donor coordination.61

Adams' review of the issue of project size in Kenya concludes that farmer participation and control is more important to the success of the scheme than size itself, that schemes run by bureaucracies tend to perform poorly independently of their size.62 This lesson would appear to be valid for other countries as well. The challenge of ensuring farmer control at the level of tertiary canals, and maintaining adequately their communications with upstream managers, may be greater in larger schemes but, with adequate definition of responsibilities at all levels, it is not insuperable. The technical and managerial problems of ensuring stability of water levels in secondary and tertiary canals also tend to be greater in the larger systems.63

Provided that the institutional, engineering and policy requirements for proper operation of a modern irrigation system can be satisfied, size should not be a barrier. Irrigation systems in several countries, including Mexico, Pakistan and

63. For those involved in the design of small-scale irrigation schemes in developing countries, detailed and practical guidelines can be found in: (a) F. M. Chancellor and J. M. Hide, Smallholder Irrigation: Ways Forward, Guidelines for Achieving Appropriate System Design, H. R. Wallingford Ltd, Report OD 136, Department for International Development (DFID), Wallingford, Oxon, UK, August 1997; (b) G. Cornish, Modern Irrigation Technologies for Smallholders in Developing Countries, ITDG Publishing, London, 1998.

India, as well as China, Nigeria and Sudan, confirm this conclusion. However, where management institutions for irrigation are in their infancy or are not well structured, and traditions of farmer participation are not well developed, experience would suggest smaller schemes are more likely to be successful, in light of the diverse kinds of challenges that have to be met in making irrigation work.

Hervé Plusquellec, Charles Burt and Hans Wolter have introduced an additional and important distinction in the irrigation typology, and in irrigation strategies as well, by suggesting different engineering approaches to the design of surface systems. They point out that actual efficiencies tend to run at 50-85% of design efficiencies, and propose that efficiency levels can be raised significantly through use of modern design concepts. Many irrigation systems routinely fail to achieve the basic goal of delivering water to farmers in the requested quantities and with the specified timing. They emphasize the importance of reliability of water deliveries, and one precondition for this is maintaining stable water levels in main canals. The statement of the problem by Plusquellec, Burt and Wolter is as follows:

Many designs are difficult to manage under real conditions. Operating instructions are often conflicting and sometimes meaningless. Murray-Rust and Snellen,64 studying the Maneungteung Irrigation Project in Indonesia, observe:

The system calls for bi-weekly assessment of demand for every tertiary block, and a readjustment of every gate in the system to meet the changed water distribution plan. This requires a very intensive data collection program and an efficient and effective information management system. Because it is carried out in an environment of unpredictable water availability it becomes almost impossible to achieve even if there were a huge increase in number and skills of field staff.

Another example is the Kirindi Oya Irrigation Project in Sri Lanka it takes up to four days to reach a new steady state after changing the flow at the headworks. In the upper reaches of the canal the steady state is reached soon and the water level fluctuations are low. But in the lower reaches of the canal the water level fluctuations of about one meter occur for up to four days. ... If the discharge at the headworks is only changed once in a week, steady flow conditions are rarely achieved. .. .

Some irrigation designs guarantee anarchy at the turnouts. When water delivery is erratic, water users lose respect for the rules and regulations governing water usage. These conditions lead to passive water user associations (WUAs) and tremendous damage to distributaries and turnouts. Reports judge the incidence of such damage to be as high as 80 percent, even in some Asian countries with long traditions of irrigation. Such anarchy is not a necessary part of an irrigation project, nor is it inherent to any culture.. . . The authors contend that inadequate design and operation is a much more significant factor in creating conflicts and disorder than the absence of irrigation tradition or social and legal norms.65

These authors propose systematic investigations as to why an irrigation system is not fulfilling its operational potential and, when appropriate, modification of the design to make the system simpler to operate and more effective. Their approach to system design is worth review

64. D. H. Murray-Rust and W. B. Snellen, Performance Assessment Diagnosis (draft), International Irrigation Management Institute (IIMI), International Livestock Research Institute (ILRI) and Institute for Water Education (IHE), 1991.

65. Hervé Plusquellec, Charles Burt and Hans W. Wolter, Modern Water Control in Irrigation: Concepts, Issues and Applications, World Bank Technical Paper No. 246, Irrigation and Drainage Series, The World Bank, Washington, DC, USA, 1994, pp. 2-4.

ing at a policy level, in the course of developing irrigation strategies. It includes the following elements:

A good design increases reliability, equity and flexibility of water delivery to farmers. It reduces conflicts among water users and between water users and the irrigation agency. It leads to lower operational and maintenance costs.

Extended gravity irrigation schemes with manually operated gates and control structures rarely work, despite all efforts to improve irrigation management and the capacity of staff. The performance is sometimes inferior to systems without adjustable structures. Basically there are two options to improve irrigation performance: (a) simplification through proportional dividers, unadjustable gates and rigid scheduling, or (b) modernization through the application of hydraulic principles, automation, improved communication and decentralization. . . .

A good design [produces] the simplest and most workable solution. A good design is user friendly and not necessarily synonymous with 'high costs', 'high maintenance', or 'complexity of operation'. . .. Some modernized irrigation projects have failed because of improper choice of control structures, incompatible components, and a design that was not based on realistic operation and maintenance plans. This has created the wrong impression that modern design concepts are not suitable for the environment of developing countries.66

According to these authors, the characteristics of modern irrigation design include robustness, good communications systems and 'social capital' in the sense that the users have a sufficient degree of mutual trust and participate in designing and supervising the water allocations:

Each level is technically able to provide reliable, timely and equitable water delivery services to the next lower level.. . . An enforceable system is in place that defines the mutual obligations and creates confidence at each level that the next higher level will provide reliable, equitable and timely water delivery service. . .. Good communications systems exist to provide the necessary information, control and feedback on system status. .. .

The hydraulic design is robust, in the sense that it will function well in spite of changing channel dimensions, siltation, and communications breakdowns. Automatic devices are used where appropriate to stabilize water levels in unsteady flow conditions. . . .

Engineers do not dictate the terms of water delivery; rather, agricultural and social requirements are understood and satisfied at all levels and at all stages of the design and operation process within overall resource availability.67

The FAO properly cautions against overemphasis of the engineering aspects of irrigation systems,68 but nonetheless the modern design approach appears to offer significant practical advantages. The basic point here is that improved system design, generally in the direction of simplifications, can make the system operations both more efficient in water use and more equitable among irrigators. There would appear to be a need for further training of engineers who are involved in the design of irrigation systems and scope for regularly seeking additional professional opinions on designs before they are implemented.

68. 'Another important influence on water resource policy is societies' partiality for technical solutions. In most countries, water management is typically relegated to the engineering domain. Indeed, most water managers are engineers, who are trained to solve technical problems. As inadequate public policies are increasingly blamed for water-related problems, a strong case is emerging for emphasizing human behavior as an additional component of water systems' (FAO, 1993, p. 257).

An ironic illustration of the importance of appropriate system design is provided by the contrast between the functioning of the older and newer irrigation works in Egypt:

Despite a minimum of management [the] traditional irrigation system has a high overall efficiency.. . . The irrigation infrastructure in the newly reclaimed lands from the desert ('new lands') is designed along the same lines as the 'old lands', except that canals are lined. In the 'new lands', however, the lack of water management devices, night or buffer storage and the inability to recycle spills has led to very low efficiencies. The heavy losses have caused water logging in adjacent 'old lands'.69

The option of the modern design approach should be considered not only for new systems but also when reviewing systems that need rehabilitation. 'Whether to rehabilitate current projects to existing standards only or upgrade them to standards for (future) adoption of improved irrigation technologies on the farm is an issue that has not been sufficiently addressed in rehabilitation projects'.70 Van Tuijl had in mind the alternative of pressurized irrigation, but the same statement could be made with respect to modern design improvements in canal and control infrastructure. A call for design modification by simplifying the infrastructure and operations requirements by converting to fixed and automatic controls that need less discretionary intervention was issued by E. B. Rice, for the systems he studied in south east Asia.71

Moris and Thom have underscored the need to adapt the engineering to the local agronomic and socio-economic conditions, which may place severe limits on system potentials: 'Why are so many irrigation projects in Africa designed and justified for double-cycle cropping when it is common knowledge that few projects can attain such crop ping intensities? Why do field specifications continue to call for wire gabions in surroundings where people have a high incentive to steal the wire?... By parceling out specialized tasks within the project cycle, donors have insulated the specialists involved at the design phase from the necessity of learning from past mistakes'.72

While improved design of the infrastructure for water delivery and more effective system management can contribute to significantly better performance of irrigation, there are two problems that have proven rather intractable, worldwide: the tendency toward waterlogging and saliniza-tion, and the difficulty of maintaining level plots. When land is not level, irrigation efficiencies drop sharply, and problems of waterlogging can be exacerbated. It is recommended that special attention be paid to these two issues, in preparing both new investments and rehabilitation projects, and as a matter of basic orientations of irrigation strategies.

Appropriate system design is also critical from a gender perspective. To ensure gender issues are addressed in the design, it is important to carry out a gender analysis of the involved communities first, with special emphasis on identifying the agricultural and water-related tasks that women carry out. The design process should be participatory and women's groups should be consulted without the presence of men. Demonstration visits to functioning irrigation sites should include women, and it is important that participants in the scheme understand what will be the workload implications for both men and


In closing this brief review of issues related to types of irrigation, it is worth reiterating the importance of adequate planning of use of groundwater and the option of conjunctive use of surface and groundwater in some locations, since in effect groundwater represents stored water that can be used during droughts. At the same time, it

73. Useful recommendations on dealing with gender issues in irrigation design are found in the 1999 study by Chancellor, Hasnip and O'Neill, mentioned in the box on p. 223.

is important to be aware of the hazards of using systems based on pumping in environments that cannot readily support them. In the words of Moris and Thom:

In Africa today there are probably more pumps that do not pump than there are those that do. .. . Pumps, when employed by unskilled operators without benefit of adequate mechanics and parts, may not last through their second season - the actual experience of many. . . during the first phase of a USAID-assisted Mali Project. . ..

Pumps appear to have been very problematic when introduced into the more remote setting where small schemes are often located. The sources of the difficulty are the following:

— The great vulnerability of 'orphan' equipment which cannot be supported within the immediate commercial environment;

— The fact that a pump breakdown may immediately threaten the associated production system;

— Poor maintenance, which leads to a high rate of breakdowns and rapid deterioration of equipment;

— Frequent problems associated with obtaining fuel or power to keep pumps operating;

— Farmers' inability to pay operating costs at the times needed;

— Difficulties caused by fluctuating water levels, which may exceed a pump's lift capacity; and

— The generally poor quality of backup services (mechanics, parts, assistance, etc.).74

A solution to this problem is for international agencies to support the development of pumps based on local materials, as has been done in Nicaragua with the development of the award-winning 'mecate' (jute rope) pumps that are now widely used throughout the country and other parts of Central America.

Longdale irrigation scheme [in Zimbabwe] received an electric pump from DANIDA in 1993. Because no information was given on service requirements, the pump received little attention and has recently begun to give problems. In 1998, the pump was out of action for a number of months due to failure of a rubber seal, which led to the loss of one season's crops. A replacement rubber seal could not be found in Zimbabwe or South Africa. Finally the seal was reconditioned in Masvingo, but this is only a temporary measure and a new seal will be needed in the near future (F. Chancellor, N. Hasnip and D. O'Neill, Gender-Sensitive Irrigation Design, Guidance for Smallholder Irrigation Development, H. R. Wallingford Ltd, Report OD 142 (Part 1), Department for International Development (DFID), Wallingford, Oxon, UK, December 1999, p. 32).

0 0

Post a comment