Previous Page Table of Contents Next Page


Options for effective rice water management - W. Hundertmark a and T. Facon b

a Consultant, Water Resources, Development and Management Service and
b Water Management Officer, Regional Office for Asia and the Pacific, FAO, Bangkok, Thailand

INTRODUCTION

As the major contributor of staple food in the world, the rice sector is also the biggest water user. Irrigated systems alone use over 70 percent of freshwater depletions, and in Asian countries over 80 percent, half of which are used in irrigated rice production (Guerra et al., 1998). Over the past 25 years, the irrigated land area per person in Asia has dropped by between 10 and 15 percent (Figure 1).[34] As population growth rates remain high, this trend is likely to continue for two main reasons:

FIGURE 1
Rice cultivated area in Asia and Africa, grouped by ecology

Source: IRRI website, 2002.

Despite the importance of irrigated systems, nearly half the world’s rice area is cultivated under rainfed conditions with only partial control over the water. In 1995, about 31.4 percent of the total harvested area in Asia was in rainfed lowland, 7.7 percent in upland and 4.9 percent in other rainfed ecologies. In Africa, the portion of the rainfed (both lowland and upland) system is much higher than in Asia. Notably, the per caput irrigated riceland in Africa - 2.45 ha per 1 000 persons in 1980 and nearly 3.0 ha per 1 000 persons in 1995 - is almost ten times less than in Asia. Productivity measures of rice systems are normally given in terms of land productivity and less frequently in terms of water productivity. Molden et al. (1998) examined the performance of 18 mainly rice-based irrigation systems around the world and found significant variation in terms of standardized gross values per unit of water consumed: between US$0.1 and 0.9 per m3. In both Asia and Africa, rice yields are much higher in irrigated ecologies.

The System-Wide Initiative on Water Management (SWIM) presented an important paper entitled “Producing More Rice with Less Water” (Guerra et al., 1998); it elaborates management strategies capable of significantly increasing water productivity in irrigated rice production, mainly in Asia.

This article extends the focus in three directions:

DESCRIPTION OF RICE-BASED SYSTEMS

Ecological boundary conditions for water management

Asia’s and Africa’s water problems are caused partly by the uneven distribution of rainfall. About half of China receives less than 400 mm of rainfall a year, and extensive areas of northwest, central and south Asia are drought-prone. In other areas, such as the northern Philippines, average annual rainfall is 3 530 mm. As a further complication, rainfall in Asia usually arrives at high intensities in short periods of time. The monsoon is often erratic, with the result that in many countries, floods and seasonal water shortages occur (Spurgeon, 1999). The hydrological situation seems destined to deteriorate, with lowland and deep-water ecologies particularly subject to change. Hayese (1999) carried out a hydrological assessment of the Chao Praya Basin, Thailand, describing the succession of rice ecologies and demonstrating various rainfall-runoff characteristics.

Evolution of the rice-based system

Ecological adaptation

For hundreds of years, natural selection pressures (e.g. drought, submergence, flooding, nutrient stresses and biotic stresses) have contributed to diversity in rice ecosystems. The plant’s adaptation strategies include the ability to: survive submerged conditions without damage; elongate stems by several decimetres in just 1 day in order to escape oxygen deficiency due to rising watertables; and withstand severe drought periods. On the basis of this agro-ecological diversity, ecologists have proposed several classification systems, of which the most widely used distinguishes five categories (Khush, 1984): rainfed lowland, deep water, tidal wetlands, rainfed upland and irrigated rice.

BOX 1
Hydrological characteristics of the Chao Phraya basin, Thailand.

The Chao Phraya basin is divided into ten upstream sections with mountainous topography, a mid-stream section where inland flood plain are found and a downstream basin with its delta-shaped estuary plains. Both rain-fed and irrigated rice ecosystems are found in smaller valleys and on sloped and terraced land. The mid stream basin is composed of terraced land, gently sloped, extending along the margin of central plains. In this area water scarcity is a common problem. Locally the relief is undulating and level differences of 5m and more can be found. Inundated lowlands water height reach easily 4m. These areas are used by deepwater rice. Floods are passing through the inundated zone of the midstream and subsequently reach the downstream basin, which forms a large retardation zone, enclosed by polders to protect urban areas of Bangkok. Eventually floods enter into the vast areas of the delta, where water heights vary between 0.5-1.0m. The river discharge is largely controlled by the tidal movement of the Gulf of Thailand. The average specific discharge rate of the midstream basin is given as 2.3m3/sec per 100km2, which is regarded as relatively low.

Source: Hayese (1999), cited in Mizutani et al. (1999).

Change of management conditions

The development from rainfed to irrigated systems is characterized by a process of gradual change in the management conditions. Improving water control conditions offers the highest likelihood of success in farmers’ eyes. The process typically begins with the division of rice fields into small basins enclosed by small earth dykes (bunds), known as “paddies”. Once paddies are developed, more rainfall can be stored, water availability increases and water is therefore used more effectively. Farmers then dig irrigation canals and divert water from natural streams or catchment areas to their rice fields. This is followed by a third step: the development of an artificial drainage system with canals linking the field to the main drainage exit, which often happens to be the natural stream.

Irrigation depends on the ability of the rice plant to adapt to natural conditions combined with the ability of farmers to change management conditions within the boundaries set by the ecological conditions.[36]

Upland systems

In upland rice systems, crops are grown within small plots enclosed by bunds, or in microbasins of less than 0.15 m2 in size (mainly in Africa). In Asia, the mountainous landscape is terraced and divided into several elevated, slightly sloping layers. Rainfall is the primary source of water and it is captured and conserved within paddy plots and microbasins. A secondary source of water in terraced systems is deep seepage from the next plot upstream. Although the level of control over seepage utilization is often very poor, Seckler (1999) notes that “if the upstream farmer uses water more efficiently, the outflow from his fields will be reduced, and the downstream farmer will suffer water shortage. A separate delivery system would have to be constructed to serve the downstream farmer, at considerable cost, little economic gain, and no increase in the overall irrigation efficiency of the system as a whole”. Other factors, such as requirements in water control and service quality, may also need to be considered.

Lowland rice

Rainfed lowland rice is the predominant ecosystem in the world’s most densely populated rural regions and home to some of the world’s poorest rural and urban populations. In lowland systems, rice is transplanted or direct seeded into puddled soil on level to slightly sloping, bunded or diked fields with variable depth and duration of flooding, depending on rainfall. Soils alternate from flooded to non-flooded. Box 2 describes a lowland cropping situation in Cambodia and illustrates the diversity of systems and cultivation strategies adopted by farmers.

BOX 2
Lowland rice systems in Cambodia

Rain-fed lowland rice is generally classified according to three characteristics: topography, water depth in the field and variety type. Field levels of water are categorised as high (30-80cm), middle (10-30cm) and low (0-15cm) which are associated as a continuum. According to water depths, there are varieties that fit to each environment. Early-duration varieties are suited to high fields, medium varieties grow well in the middle fields and late varieties are best for the low lands where water levels subsided late. All fields are exposed to drought and floods, whereby high fields are generally more drought-prone and low fields more flood-prone. Middle fields are more drought-prone than the low fields and more flood-prone than the high fields. Deepwater fields are subject to prolonged periods of floods in certain years.

The uncertainty of occurrence, duration, and amount of rainfall affects substantially the productivity of the rain-fed lowland rice ecosystem. Inadequate rain at the onset of the rainy season in May and June delay seedbed establishment, lead to poor land preparation, favour the build-up of trips, army worms, and mole croquets in the nursery, and produce less vigorous seedlings. Inadequate rains in July and August delay transplanting, lead to poor transplant recovery, and poor vegetative growth. The dry soils also encourage weed growth. In general the unfavourable effects of insufficient rainfall from May to August are mostly felt in the high fields, where early maturity varieties are grown. The least effect is experienced in fields planted to late duration varieties.

Excessive rains from September to October coupled with flood water levels in the Tonle Bassac and Mekong rivers, cause flooding in the rainfed lowlands. This effect is greatest in low fields and least in high fields. The strong current of flash floods laden with silt can damage leaves and submerge the crop for a number of days. Flash floods can reduce the plant density. Reduced rains in October and early cessation of rains in November can adversely affect the yield of the crop during its reproductive phase. Most vulnerable are late duration varieties followed by medium to short varieties.

Source: Nesbitt, 1997.

Deep-water/flood-prone rice

Deep-water rice is grown on medium to very deep flooded depressions (50-300 cm), typically located within river mouth deltas. The rice crop grows as floodwater rises, and is harvested after the water recedes. More than 15 million ha (Mha) in South and Southeast Asia are subject to various types of uncontrolled flooding. Flood-prone riceland is also found in the inner delta of the Niger river and in Nigeria, West Africa. Rice is often the only crop that can be grown in flood-prone areas. Yields are low because of problem soils and unpredictable combinations of drought and flood, and crop failures are common. The flooding pattern is characterized by the onset time of inundation, the rate of water rise, the maximum water depths and the duration, time and rate of recession. A drastic change in any one factor can result in complete crop failure. Unpredictability is the main reason why farmers employ minimum inputs in deep-water systems. In many parts of Southeast Asia, a late-dry-season/ flood-recession rice is cultivated. Flood-recession land is found on deep depressions alongside rivers and lakes where rapidly rising watertables do not permit deep-water rice. Flood recession areas are level paddies, situated on sloping land, from which water recedes slowly. Cultivation of land follows the receding water in short sequences of 10 to 14 days each.

Irrigated rice

Irrigated lowland rice is cultivated on levelled and bunded fields with water control, in both dry and wet seasons. Early-maturing varieties dominate dry-season rice production. In general, the productivity of this rice ecosystem is higher than in rainfed lowland rice, due to better water control and higher solar radiation. Where modern technology is used, yields can reach 5 t/ha in the wet season and more than 10 t/ha in the dry season. Some of the world’s most productive rice-based irrigation systems are found in arid and semi-arid regions, such as the Indus Valley in Pakistan, the Nile Delta and the Sahel. On the African continent the most impressive productivity gains are reported from the Office du Niger - a 65 000-ha rice irrigation scheme where yield levels doubled over a period of 10 years from 3 to 6 t/ha following modernization of the system and changes in institutional structures allowing farmers to participate in the management of the scheme. However, water applications of dry-season rice frequently reach 50 000 m3/ha.

Water-use in rice systems

Water takes on a prominent role in rice production. Unlike many other cropping systems, where water is mainly used for productive purposes (evapotranspiration), the rice cropping system uses water in numerous ways, both beneficially and non-beneficially. Conventionally, the water requirement for productive and beneficial use in rice systems is divided into three categories:

TABLE 1
Estimated consumptive use of water for evapotranspiration and other uses

Purpose of water use

Consumptive use
(mm)

Remarks

Low

High

Land preparation

150

250

Refilling soil moisture, ploughing, and puddling

Evapotranspiration

500

1 200


Seepage and percolation

200

700

To maintain water layer

Mid-season drainage

50

100

Refill of water basin after drainage

Total

900

2 250


The range of daily dry-season evapotranspiration rates is given as between 5 and 12 mm/day (Guerra et al., 1998; Watanabe, 1999), adding 500 to 1 200 mm on the basis of 100 irrigation days. Evapotranspiration needs include two components: water for maintaining physiological processes that lead to plant development and growth, and water to compensate for evaporation from the soil. Land preparation needs are in the range of 150 to 250 mm.

Seepage and percolation needs of 2 to 7 mm/day are required if moisture is to be kept at saturation levels (200-700 mm), or if submerged conditions are to be maintained. If drainage of water prior to the tillering stage (panicle initiation) is practised, another 50-100 mm must be added. On the basis of this calculation, the total water requirement of irrigated rice ranges between 900 and 2 250 mm. The actual water demand of farmers is often much higher to account for conventional application efficiencies of less than 50 percent (taking into consideration seepage and percolation losses) or for special water management practices.

Figure 2 gives the typical breakdown of rice crop water requirements for irrigated rice in Côte d’Ivoire, West Africa. Typically, water needs peak at the beginning of the season, when land preparation and transplanting require high water supply rates. Roughly, it can be assumed that 50 percent of the net crop water requirement is needed for evapotranspiration. The remainder goes to fill the soil moisture deficit, is used for submerging soils or is lost through seepage.

FIGURE 2
Rice water requirement (West Africa)

Note: Soil moisture = water needs for initial soil saturation; ETC = evapotranspiration; CWR = total crop water requirement; NET = net irrigation requirement.

Not all authors agree that seepage and deep percolation should be included under “productive or beneficial use” alongside evapotranspiration and special water management practices. It is difficult to separate evapotranspiration (which cannot be recirculated or reduced) from seepage water (which is not a consumptive use and is often recirculated via wells or drains or reduced - as described below). Furthermore, deep percolation losses due to non-uniformity of water application and poor timing are unavoidable in all crops (not only paddy rice), and such a convention should therefore apply to all crops. It is not applied in the Rapid Appraisal Procedure for evaluation of irrigation system performance or by the American Society of Civil Engineers (FAO, 1999; Burt et al., 1997).

Productive and beneficial effects

The underlying assumption made in the above calculations is that the soil is submerged, in order to control weeds. The weed-control effect of submerged water is very effective but highly water consumptive (unless there is excellent levelling). Alternatives to this practice include labour-intensive hand-weeding or the use of chemical herbicides. Neither of these alternatives is suitable for the majority of rice-producing smallholdings. Labour is increasingly scarce and herbicides are expensive, but a valid alternative to water-based weed control remains to be found. Scientists are now turning to the development of integrated weed management practices, in order to propose ways to control weeds with reduced use of chemicals (Solaimalai et al., 2000; Diallo and Johnson, 1995).

Findings reported from Japan suggest that a constant water layer has a stabilizing effect on the microclimate in a rice field, particularly on the temperature regime between day and night. According to the energy conditions, day temperature in the rice field is lower than the air temperature of the surrounding environment. At night, temperatures in the fields tend to be higher than outside the field. Hence, during the day mild temperatures may be favourable for crop growth, and at night cooling stress can be prevented (Takese, 1999).

Land preparation techniques can be divided into: soaking of the soil (in order to facilitate subsequent activities); ploughing; and puddling. Puddling is a common practice and it facilitates transplanting or direct seeding in paddy soils. Another effect of puddling is levelling, which ensures more even distribution of water under submerged conditions, allowing fertilizer to be well mixed. The technique is important for controlling seepage. In many Asian soils, paddy layers with a bulk density of between 1.2 and 1.6 g/cm were found at depths of 10 to 20 cm below ground. The hydraulic conductivity of such high bulk densities is extremely reduced, which is a precondition for reduced downward flow.

Non-beneficial effects

Under submerged conditions water can indirectly affect the rice ecology in negative, non-beneficial ways. Examples include: changes of the soil’s physical and chemical properties, which may lead to irreversible degradation of resources; environmentally sensitive emissions of greenhouse gases (carbon dioxide and methane); and adverse effects of submerged conditions on the physiology of the plant itself.

The effects of submerged conditions on the physical and chemical conditions of the soil are complex and go beyond the scope of this paper. As soon as soils are exposed to anaerobic conditions, their redox potential changes from an oxidized to a reduced state, and the soil milieu (pH) changes from alkaline to acid. Under reduced conditions, a characteristic layer develops which can act as a permeability barrier for water uptake by the roots. In acid soils in West Africa, iron toxicity has been identified as a major problem in rice-based systems (K. Sahrawat, personal communication) causing several million dollars worth of damage every year.

It is estimated that flooded rice fields contribute some 30 percent of the total annual methane emissions of 500 Mt. The rice plant acts as a chimney for up to 90 percent of the methane that escapes from flooded fields (Lantin et al., 1995). Asignificant reduction in the methane efflux was observed where water management practices had been changed (Mishra et al., 1997).

There is concern that many of the irrigated rice-based systems in the Sahel may become severely affected by increased accumulation of soluble salts in soils. Under the prevailing conditions, salt effects may lead to soil degradation and a loss of potential productivity, which is normally associated with reduced crop yields, or even complete crop failure. Estimates suggest that about 45 000 ha of irrigated land in West Africa are severely affected by salts in the form of salinity or sodicity (Bertrand et al., 1995; Boivin, 1995), a condition resulting from the region’s high evaporative demand of over 2 000 mm per year. Salinity problems are mainly confined to areas such as the Senegal River Delta, where land is exposed to marine conditions; sodic and alkaline conditions are observed in the basin of Senegal, Niger and Lake Chad. Some 13 years ago it was found that problems at the Office du Niger, Mali, coincided with a sharp rise in watertables from about 30 m to less than 1 m below the surface in under 40 years (Bertrand et al., 1995).

It is estimated that in China alone, an area of 24.58 Mha is subject to waterlogging and salinity and 35 percent of the Indus Basin is subject to the same problem (Smedema, 2000).

DESCRIPTION OF OPTIONS

Conceptual framework

The proposed conceptual framework for the development and the adoption of rice water management options is based on the concepts of effective efficiency (Keller and Keller, 1995) and integrated water resource management systems (Keller et al., 1996) and follows three important principles: rice ecology; integrated resource water management; and sustainability.

Rice ecology principles

Effective water management strategies and options must recognize that the ecology in which rice is cultivated determines the level, the scope and the feasibility of the intervention. Therefore, all strategies to improve the system’s performance require sufficient understanding of the rice ecological system, especially of the system’s hydrology, and they fall within specific boundaries. For example, if the intention is to increase the drainage discharge rate of a rice-based lowland system, there may appear to be a given physical limit set by the available gradient that drives the water flows, and by the distance between the system and the main drainage exit (river bed). Unless pumping is an option, a drainage system may not be sufficient for evacuating excess water from the paddy fields. Pumping is of course always an option but it may not be an economical one, so economic factors do play an important role in determining the level, scope and feasibility of the intervention, the cost of which may increase if it aims to offset the boundaries set by the rice ecology.

Integrated water resource management

Options must recognize that the allocation of water resources to the rice sector is firmly inserted in an integrated water resource management framework that gives equal opportunity to sectors other than rice. Water allocation decisions at basin, system and farm level are made on economic, technical, social and legal grounds, and investment into water management must adhere to a set of national policies concerning food, poverty and environmental issues. Careful consideration is to be given to linkages between the levels concerned, in order to meet water demand and supply the needs of rice-based systems. Linkages exist between the various levels: a higher level supplies water to the lower level, which demands water from the higher level. Thus, an intervention that changes demand at one level should be matched by a corresponding change in supply at the upper level. The notion of service may be introduced to describe the linkages between the various levels, each level providing a water delivery service to the next, lower level, from basin down to farm level. Therefore a guiding principle of the conceptual framework is the integration of supply and demand management options at all levels including basin, system and field, and application of a consistent service-oriented approach.

Sustainability (integration of time)

An important strategic element in the development of effective water management options is the integration of time. Options, such as the shift of field level irrigation scheduling, must be accompanied and may need to be preceded by medium- to long-term changes in the system’s water supply system. The term “sustainability” describes the time aspect of the system’s performance. Accordingly, sustainability means that the performance of a system is maintained over time. The term includes technical, economic, social and institutional aspects (Merry, 1996).

The water management options described are based on three levels of analysis: field, system and basin. Rice systems are portrayed in the context of a multi-layered basin. In each layer, the proposed technical and managerial options are further broken down into three strategic areas of intervention:

The effectiveness of each technical and managerial option should be assessed against the intended and measurable effect on the water-use efficiency and productivity indicator of a given rice ecosystem - one at a time. For example, an option can be very effective in increasing crop productivity in an upland system. Applied to deep-water rice, the same option has no effect. The time horizon given in order to make the option effective may vary greatly from option to option.

Field-level options

Demand-side options

Field-level rice water management options are the most covered in literature for two reasons: first, as great losses occur at field level, it is naturally thought that there is ample scope for “saving” water; and second, field-level research is the domain of international and national agriculture commodity centres (IRRI, WARDA and NARS) where most water-related rice research is carried out. Guerra et al. (1998) propose four field-level strategies for increasing water-use efficiency in irrigated rice:

As the field level is the lowest level in the water resources system, all options are demand-side options.

This paper highlights field irrigation regimes - also known as “alternate wet/dry irrigation” (AWDI) or “water-efficient irrigation” (WEI) - which have attracted much interest among the irrigation community. The regimes were developed in China and elsewhere in Asia, and by 1997 the area under the new irrigation regimes had reached 5.7 Mha, i.e. more than 18 percent of the total paddy rice cultivation.

Water conservation through technical redesign and management: There is a general assumption that seepage and percolation at field level are the result of vertical discharge processes. Walker (1999) observed that in many rice paddy soils in Asia direct vertical flux from the field is reduced by a semi-permeable hard pan. Alarge proportion of the water first moves laterally into bunds, where it changes direction flowing vertically towards the watertable. The practical implication of this observed flow regime is that substantial water could be conserved at field level by preventing water from entering bunds. Technically this could be achieved by redesigning and lining the bunds. Improved surface storage within paddy plots has an increasing effect on rainfall efficiency; increased storage may also contribute to reducing the risk of drought. On the other hand, improving plot-level storage may have adverse effects on downstream opportunities for seepage recycling at system level. It is therefore important to study the effect that redesign at plot level has on water availability further downstream.

Improving irrigation scheduling: An important element in effective water management in rice production is an adequate knowledge of crop water needs and irrigation requirements for various crops in the given climatic conditions. Standardized procedures with FAO software and an extensive climatic and crop database allow routine calculations and the latest available information on crop water climate conditions in the proposed project areas (FAO, 1977).[37]

Based on crop water requirement and the concept of yield response to water, improvements in irrigation scheduling can be planned easily and precisely. Monitoring the soil moisture can also improve irrigation scheduling. There are numerous methods for measuring the moisture content of the soil and scheduling irrigation, ranging from simple penetration sticks to tensiometers and more sophisticated computerized methods of water balancing (FAO, 2001). Rational irrigation scheduling can also optimize agricultural production under drought conditions. Depending on the specific water demand of the crop, the length of the growing cycle, yield response and the value of the crop, decisions have to be made as to which crop to irrigate and how much it should be irrigated in order to minimize crop losses.

Figure 3 presents a graphical description of three regimes (a, b and c). The overriding principle of these regimes is the combination of submerged conditions (mainly during transplanting and the middle of tillering) and wet (saturated) to soil moisture conditions near field capacity. In semi-dry cultivation (SDC), a shallow water layer during the vegetative growing phase is combined with saturated conditions during the generative phase. In alternate wetting and drying (AWD), a shallow watertable during the vegetative phase is combined with a regime of variable irrigation depths and irrigation intervals depending on a soil moisture content of 80 percent. The third regime is a shallow watertable with wetting and drying (SWD): a shallow watertable in the early development stage of the plant, followed by a wet period (soil saturated) and a drying-off period at tillering. During the generative phase, a shallow watertable is maintained until the milk ripening stage; during the yellow ripening stage that follows, the soil is allowed to dry off in order to facilitate harvesting.

FIGURE 3
Rice irrigation schedules adopted to irrigated rice systems in China distinguishing submerged, wet and dry irrigation (SWD), alternated wet and dry (AWD) and semi-dry cultivation irrigation (SDC)

Source: Mao Zhi, 2001.

According to Mao Zhi (2001), up to 50 percent of the water consumption of traditional irrigation under submergence could be saved by utilizing the three regimes. The regimes had only a mild effect on yield per hectare (9% increase), but a significant effect on water productivity. This effect is attributed to three phenomena:

1. seepage and percolation are reduced by 20 to 65 percent;

2. evaporation from the soil is reduced by between 3 and 15 percent through elongated periods of less hydraulic head and prevalence of unsaturated soil conditions; and

3. effective rainfall is increased by 5 to 15 percent through the increased storage capacity of the soil.

In all regimes the water requirement for weed control is much reduced, and to compensate, the use of chemical herbicides is obligatory. The beneficial productive effects are attributed to the improved soil aeration, which has positive effects on biological soil activity and the rooting system and improves the rate of fertilizer uptake. Also, a change in the microclimatic conditions is believed to reduce the incidence of pest and insect populations, especially of malaria carriers (Van der Hoek et al., 2001).

Solaimalai et al. (2000), in a review of the literature on the Indian subcontinent, confirmed the positive effects of various irrigation regimes on water consumption and yield. In some cases, however, yield levels were much reduced.

Bouman and Toung (2000) established a relationship between water stress and yield reduction in the irrigated lowland system, based on a large number of field experiments in Asia. The study confirms that irrigated rice yields decline as soon as the water content in the field drops below saturation level. Even when the soil moisture content remained at saturation, a yield reduction of up to 12 percent was observed. All tested varieties responded sensitively to the severity of drought conditions with the period of anthesis being the most vulnerable. Water productivity was found to be highest when there was no submergence and the soils were saturated with water. The study concluded that soil moisture kept at saturation was the most promising option for water saving, high water productivity and acceptable land productivity reductions. Furthermore, the implementation of such field water management regimes requires very precise and timely water deliveries, i.e. high flexibility of the water delivery service to the field.

Reducing water use for land preparation and crop establishment: A shift from transplanting (involving a nursery period and the subsequent setting of plants into puddled water) to direct seeding offers considerable water savings at field level (Guerra et al., 1998). In heavy, cracked soils, 45 percent of the water used for soil soaking can bypass the root zone and be ineffective. The reduction of water needs for land preparation through the utilization of dry shallow tillage after harvest helps increase the water-use efficiency of the soil. This technique is suitable for upland, lowland and irrigated systems, but unsuitable for deep-water systems.

Crop intensification and diversification: Crop intensification implies the utilization of improved cultural practices, such as improved varieties, integrated fertility and pest management and improved postharvest technologies. Improved short- medium- and long-duration varieties are available for upland, lowland, deep-water and irrigated rice, which differ considerably in their net crop water requirement. A change from long- to short-duration varieties offers significant opportunities for increasing the productivity of water. With the arrival of photo-thermal-sensitive rice varieties, the potential of elongated rice cultivation has been significantly increased. Dingkuhn (1995) reports on the cultivation of photo-thermal-sensitive varieties in West Africa avoiding low temperatures during spikelet formation.

Diversification at field level means the utilization of land resources by crops other than rice. Crop intensification and diversification are suitable options for increasing crop water productivity. Special attention is given to constraints associated with the existing water management regime. For example, if water is supplied in rotation, the interval may not be sufficient to support this option. Molle et al. (1999) studied the implications of a changed cropping system for the irrigation supply; it is concluded that daily care and attention is required for good management, which demands numerous minor adjustments and operations. Flexibility in system operations is therefore very important. This is a good illustration of the links between the different levels of water management: demand depends on field level changes and supply depends on the system level; the supply must meet the new demand by providing a more flexible water supply service. Changes in the design and operation of control structures may be required, as well as an upgrading of the system water management processes.

More generally, it has been argued (FAO, 2002) that an important reason for field-level water management options remaining mostly in the research stations and not being adopted in the field in Asia is the lack of attention to the level of system operations. Table 2 illustrates the compatibility between design and operation strategies and various distribution schedules.

TABLE 2
Linkage between level of service, operational parameters, flow control system and human resources requirements

Level of service

Operation parameters

Flow control system

Human resources

Water delivery

Class

Rate

Dur.

Freq.

Prop.

UC man.

UC aut. loc.

UC aut. cent.

DC aut.

HR no.

HR skill

On demand

Unrestricted

Ia

V

V

V

n.a.

-

-

o

++

++

o

Limited rate

Ibr

C

V

V

n.a.

-

-

o

++

++

o

Limited duration

Ibd

V

C

V

n.a.

-

-

o

++

+

o

Limited frequency

Ibf

V

V

C

n.a.

-

-

+

++

+

o

On request

Unrestricted

IIa

V

V

V

n.a.

-

-

+

++

-

-

Fixed rate

IIbr

C

V

V

n.a.

-

o

++

++

o

o

Fixed duration

IIbd

V

C

V

+

-

o

++

++

-

o

Fixed frequency

IIbf

V

V

C

+

-

o

++

++

o

o

Fixed rate+duration

IIcrd

C

C

V

n.a.

o

+

++

++

o

o

Fixed rate+frequency

IIcfc

C

V

C

n.a.

o

+

++

++

o

+

Fixed duration+frequency

IIcdf

V

C

C

+

o

+

++

++

o

o

Imposed

Fixed rate+duration

IIIcrd

C

C

V

n.a.

o

+

++

++

+

o

Fixed rate+frequency

IIIcrf

C

V

C

n.a.

o

+

++

++

+

o

Fixed duration+frequency

IIIcdf

V

C

C

+

o

+

++

++

+

o

All fixed

IIId

C

C

C

n.a.

+

+

++

++

+

+

Note: Dur. = duration; Freq. = frequency; Prop. = proportional; UC = upstream control; man. = manual; aut. = automatic; loc. = local; cent. = central; DC = downstream control; HR = human resources; no. = number; V = variable; C = fixed; n.a. = not applicable; - = very unfavourable; - = unfavourable; o = neutral; + = favourable; ++ = very favourable.

Source: Malano and Hofwegen, 1999.

Sustainability options

Salinity control “puddle and flush”: In the Senegal River Delta a special technique is used to wash out salts from the top layer prior to transplanting or direct seeding of rice. This technique is based on puddling the top soil with the subsequent removal of excess water from the soil surface: “puddle and flush”. Raes et al. (1995a, 1995b) estimate that about 400 mm of water are required if 4 tonnes of salt are to be removed by lateral drainage from the top soil.[38] With this technique the electrical conductivity of soils could be maintained at acceptable levels.

Use of salt-tolerant varieties: In a location where soil salinity levels cannot be controlled through cultural practices, rice varieties that tolerate moderate levels of salinity are a possible option. A number of varieties are available that perform well under both Asian and African conditions (Asch et al., 1997).

System-level options

Water balance of the system

Before options for improved water management at system level can be proposed, it is important to obtain a general understanding of the water balance of the system. Figure 4 shows the water balance in a rice-based irrigated system in Côte d’Ivoire (Hundertmark and Abdourahmane, 2000) using the IIMI Water Balance Framework (Perry, 1996). The water balance components are calculated for an irrigated and a non-irrigated area. At the outflow, estimates are given for crop evapotranspiration and non-beneficial evaporation. Drain outfall and seepage are calculated separately for the three levels, including main canal, water (secondary and tertiary) and field level entry supplies. The water balance also accounts for pumping from groundwater.

FIGURE 4
Estimated water balance (million m3) using the IWMI water balance framework applied to an irrigated system in Côte d’Ivoire

Source: Perry, 1996; Hundertmark and Abdourahmene, 1999.

The Rapid Appraisal Procedure developed by the International Rice Consortium (IRC), International Program on Technology Research in Irrigation and Drainage (IPTRID), World Bank and FAO for the appraisal of external performance indicators and internal process indicators of an irrigation system permits the rapid and rigorous appraisal, using the available data, of a system’s water balance and related performance indicators.

Supply management

Improving service quality through modernization and improved management: The objective of this intervention is to modernize a system so that it is capable of supplying a delivery service of better quality water to farmers’ fields. Demand-level options at field level usually require changes in system supply in order to be adopted. Typical changes aim at increasing the reliability, equity and, in many cases, flexibility of the water delivery service. Attempts to improve the performance of large rice-based irrigation systems have proved disappointing in the past. The recent World Bank/IPTRID research on the impact on performance of modernization projects (FAO, 1999) included several rice-based projects in Asia, Latin America and the Near East. The study showed that none of the projects could be considered fully modernized and that the modernization approach had in many cases been inadequate. However, partially modernized projects did result in improvements in service quality and did not show the level of chaos and anarchy of traditional irrigation systems. FAO concepts on the modernization of irrigation systems are based on the analysis of successes and failures in previous modernization projects.

An FAO publication (FAO, 2002) makes a major effort to bring more structure into this important field. It begins with a historical review of irrigation design and goes on to draw attention to modern design principles based on a close interaction with system management and technology and on a service concept.

A modern design is the result of a thought process that selects the configuration and physical components in light of a well-defined and realistic operational plan that is based on the service concept. A modern design is not defined by specific hardware components and control logic, but use of advanced concepts of hydraulic engineering, irrigation, agronomy and social science should be made to arrive at the most simple and workable solution”.

Appendix 1 is an example of a priority action plan for the Makhamtao-Uthong system in Thailand, based on an initial appraisal of the system with a rapid appraisal procedure and the application of modernization concepts. These concepts include the recirculation of drainage water and conjunctive use.

With regard to the definition of design criteria, Horst (1998) made an important contribution by advocating greater attention to design issues. He argued for a strict simplification strategy of the hydraulic infrastructure, i.e. simplified technologies for proportional diversion of water deliveries: proportional outlets, division, weir-type structures, on-off gates and step-wise distributors. The results of an assessment of three irrigated rice-based cases suggest that simplified technology would be applicable for water service within secondary and tertiary blocks, even if the rice system is extremely staggered.

FAO, together with Horst, estimates that these design principles are suitable for farmer-managed run-of-river systems. It should be noted, however, that proportional flow division does not provide flexibility or control. The problem (addressed by Horst) of the excessive complexity of fully-gated systems can also be overcome by the application of modern water control design principles to simplify canal operations. Objections to automation become less justified with the evolution of technology and the costs entailed, and with the adoption of new approaches to progressive, distributed automation, as opposed to previous attempts at centralized automation of irrigation systems.

Apublication on smallholder irrigation (Albinson and Perry, 2002) deals with the design of structured systems including paddy. The paper contains useful technical notes on the design principles of structured paddy systems (see Box 3). By definition, a structured system includes a regulated upper canal network feeding groups of Service Areas (SA). In the SA, the flow is distributed proportionally to individual watercourses; flow in the upper regulated portion is variable and canals usually run at partial capacity. Within the SA, the canals always run more or less at full design discharge or are completely closed.

BOX 3
Structured irrigation system design principles: The Special Case of Paddy. Watercourse layout for paddy areas in systems structured at the distributary level

For cultivation of a rice paddy, field to field irrigation rather than rotation through field channels will probably remain the preferred method in many areas. Tabal and Wickham (1981)a showed that providing equitable and reliable flow at the head of the tertiary canal is more important than on-farm distribution. A reasonable interpretation of Wickham’s data would lead to the conclusion that the number of fields crossed is not a significant factor, although the distance to the nearest farm channel should not be more than 300m. To satisfy these criteria, the watercourse area can be about 40 hectares, divided into four sub-areas. The peak duty is usually about 1.5l/s/ha for pre-saturation and land preparation and 1.0l/s/ha for maintenance flows during crop growth. The peak flow will be 60 l/s and this can be rotated round the four sub-areas in turn, starting at the highest level. When preparation is complete, the flow can be reduced to 40l/s divided proportionally among the four sub-areas. The distributary system will have to supply any flow between 1.5 and 1.0 l/s/ha proportionally to all watercourses in the Service Area, with adequate command at all design flows. This is possible using correctly sized pipe outlets. The main canal will have to be regulated to supply the distributary channels at the flow rates necessary to meet the operational plan. In the case of systems where paddy is grown during the wet season and upland crops are irrigated in the dry season, temporary field channels are required for the upland crops. These temporary field channels will have to be constructed each year, because they will be lost during the paddy season. With the layout described the average length of temporary field channel will not be much more than 200m.

a Tabal, D.F. & Wickham, T.H. 1981. Effects of location and water shortages in an irrigated area. Philippines, IRRI.

Source: Albinson and Perry, 2002.

According to the authors, the domain of applicability of structured design for smallholder irrigation (up to 10 ha) is where: water is scarce; a clear definition of water entitlements is needed; management capacity is limited; and investment resources are limited. This design may not be the best option where: water is abundant; water rights are already well established; management capacity is adequate; or investment resources are available.

Optimizing water re-use systems through recycling and conjunctive use of groundwater: This option (already included in irrigation system modernization) is of major importance in rice-based systems in Asia and in Africa. Field-to-field irrigation is regarded as a prime example for a re-use system, although it may also be considered that the final distribution stage consists of several fields. Another example is conjunctive use of groundwater resources to supplement irrigation supplies for dry-season crops including rice. Assuming an efficiency of 40 percent, the system efficiency could be raised by 10 percent if only one-quarter of the seepage and percolation is recyclable. Water re-use systems could be introduced to all main rice ecologies.

Technically, it is feasible to install interceptor drainage canals to capture seepage and percolation from sloped land. Satoh and Goto (1999) suggest building a diversion into a drainage canal and redirecting drainage water to the next irrigation canal, providing that the water quality is no concern. Another technique utilized in Japan is “pipe network rotation” (PINETRON). The system is based on a mixture of gravity irrigation and pumping of drainage water to a rotation tower from which it is fed into the network. The MUDA scheme (Malaysia) is a good example of a modernized irrigation system making use of recycled drainage water.

In Asia, over 40 percent of the irrigated area is supplied by groundwater, most of which is found in India where it is used year-round to satisfy intensified rice-wheat systems. It is estimated that aquifers support 60 percent or more of the food grown on irrigated land in India, which is about 50 percent of India’s total food production (Seckler, 1999). The positive effects of groundwater exploitation are that it is an easy means to get access to a large extra resource and, when developed privately, it provides the flexible and reliable water delivery service farmers require (FAO, 2002). Pumping costs for groundwater irrigation account for about two-thirds of the total per-hectare irrigation costs (Murray-Rust, 2000). Shallow tubewell irrigation, on the other hand, is generally highly profitable. However, many of the aquifers in India are being depleted and in some cases the draw-down is over 1 m per year, and there is concern about the degradation of resources due to salt. Where groundwater is a water sink, conjunctive use cannot be applied for water quality reasons.

In Nigeria, the total area under formal and informal irrigation reached about 1 Mha in 1990 with about 220 000 ha developed on floodplains. There were projects directed at the development of the smallholder system on fadama land. Projects provided services to smallholders, including drilling of shallow wells, selling of gasoline pumps at subsidized prices and promotion of group ownership and use of facilities. The impact of these technologies has been considerable. The introduction of pumped irrigation systems opened up the use of surface floods in conjunction with groundwater resources. This conjunctive use considerably increased the production of dry-season crops, which contributed to improved food security and a wider choice of food. Furthermore, a thriving service industry developed for the maintenance of pumps (Purkey and Vermillion 1995).

The management of conjunctive-use systems may represent a feasible alternative to improving the performance of the surface systems, but it entails difficulties:

The dynamics and policy implications of such systems are not fully understood. Further research is necessary in order to propose water management guidelines based on optimization and sustainability aspects.

Demand management

Optimization of cropping pattern: Experiences gained during the green revolution in Asia and Latin America shows that optimizing a system’s overall cropping pattern is a highly effective demand management intervention for enhancing the water-use efficiency in rice production.

In general, there are two strategic directions: cropping intensification and cropping diversification. At system level, intensification implies an increase in the cultivated area over the area available (irrigation intensity). This can be achieved through multiple cropping during the wet and dry seasons. Diversification has the objective of changing from high (rice) to low water-consumptive crops or - in economic terms - shifting from low-value (rice) to high-value crops.

To fully explore the opportunities and constraints of a shift in cropping pattern, recommendations must be based on a comprehensive understanding of the system as a whole. Careful attention is to be given to the understanding of the farming system, especially in terms of the farmer’s preferences. Given the diverse viability of the rice-based system, it may be assumed that a shift in cropping patterns is made for financial reasons. However, attention must be given to food security and the socio-economic and management conditions, implying an analysis of constraints in the availability of market opportunities, agricultural inputs, labour, mechanization and finance.

Hundertmark and Abdourahmane (2000) developed and utilized a model-based Diagnostic Framework for System-Level Water Use in a rice-based irrigation system in Côte d’Ivoire, West Africa. Cropping options are based on irrigated rice ecological considerations recommended by the West Africa Rice Development Association (WARDA) (Becker and Johnson, 1999). A set of short-, medium- and long-term strategies is proposed that would allow to double water productivity in the long term from 0.4 to 0.8 kg/m3. Recommendations are checked against the priorities of the farmers, taking into account water availability, management constraints of the hydraulic system, marketing opportunities and production costs.

The understanding of the system as a whole also includes an appraisal of the current performance and an assessment of changes required to enable intensification and diversification. The Rapid Appraisal Procedure means that this can be done rapidly in medium- to large-scale irrigation systems.

Sustainability options

Improving the drainage system: Most of the irrigated systems in northern China, India and Mongolia have access to drainage. In India, drainage systems cover about 5.8 Mha. Current drainage systems are in a poor state due to lack of maintenance. There has been a cut in investment in drainage works on irrigated systems (Facon, 2001).

Improving the discharge capacity of the drainage system - especially in smallholder rice-based systems - is of great importance, as the excess water can cause considerable damage to the plants. A functional drainage system takes care of excess floods in lowland and irrigated systems. Specific design discharge rates are given by Smedema and Rycroft (1983), based on the permitted temporary excess depth of water on bunded paddies.

Salinity control: Kijne (1996) studied two irrigated systems in Pakistan where over-irrigation raised the water level closer to the surface, with salt subsequently accumulating in the root zone. Proposed salinity control options include: a marked reduction in water supplies to farmers; a substantial reduction in the area cultivated to rice; and the introduction of a drainage system capable of controlling the watertable at a depth where the capillary fringe remains below the root zone.

Basin level options

At basin level, a total of five water management options are considered relevant, two of which focus on increasing the availability of water and one on optimizing the productive and beneficial use of water in rice. Another two options would contribute to reducing non-beneficial effects at basin level. As the basin level is the highest, all options are on the supply side.

Supply management

Upstream water conservation: Upstream water conservation is an effective but costly means of increasing water availability in both upstream and downstream locations. The strategy behind this landscape-level intervention aims to: restore vegetation in deforested or degraded uplands; and construct ridges and small grass strips, thereby diverting rapid surface run-off into subsurface flow. The intended effect of this intervention is the transformation of the basin discharge curve from a rapid response to a more prolonged type, i.e. elongation of the concentration time and smoothing of the recession tail, thereby facilitating groundwater recharge and reducing soil erosion sediment loads and flood damage.

Terraced upland systems have an important role to play in preventing lowlands from flooding. Hydrological observation from Japan suggests that the discharge curve from a rice-terrace watershed is comparable to that of a forest-based system (Mitzutani et al., 1999). Storage of water in paddy systems has a considerable effect on the hydrology of a river. Hayese (1994) estimated dramatic changes in the discharge characteristics of a former rice-based area of 52 km2 subject to heavy urbanization. The depth of inundating floodwater would rise from 103 to 135 cm and the peak discharge would almost double (from 82 to 158 m3/s). The period of inundation would be extended from 96 to 127 hours.

According to Facon (2000a), the total water storage capacity of paddy fields in Japan is estimated at 4.4 billion m3, which is more than the storage capacity of dams altogether. The economic benefit attributable to paddy fields for flood control is estimated at nearly 2 trillion yen a year. A further 800 trillion can be added in the form of additional groundwater storage.

A 2000 e-mail conference on land-water linkages organized by the Land and Water Development Division of FAO brought to light major concerns as to the effectiveness of water conservation in rural watersheds (FAO, 2000; Kirsch, 2002). Land-use-induced changes in the hydrological regime are measurable only in basins of up to 100 km2. On a larger scale, these effects seem to disappear.

Facilitating water re-use: Facilitating water re-use at basin level through the construction of interception systems is another supply-oriented option for increasing the availability of water to be used in downstream systems. The intended effect of this option is to increase the number of cycles of water use before it is discharged into the sea. There has been broad discussion about water re-use systems at basin level (Seckler, 1999).

There are only a few examples of the quantification of basin-level return flow and re-use. Satoh and Goto (1999) present a hydrological modelling exercise of the Tone River Basin, one of the largest basins in Japan. The assessment took into consideration natural return flow to the river and artificial flow to the drainage canal system. With the help of a two-dimensional model, the actual return flow was estimated to be 50 percent at the beginning and 60 to 80 percent at end of the season. The overall water-use efficiency was estimated to be 75 percent.

In order to improve the planning, targeting and effectiveness of water conservation measures, Molden et al. (2001) proposed the concept of Hydronomic Zones, with the basin divided into six zones: water source zone, natural recapture zone, regulated recapture zone, stagnation zone, final use zone and environmentally sensitive zone. This very useful concept assumes that management strategies in the water source zone can affect basin-wide water use.

Upgrading from flood-prone to irrigated systems:

Upgrading flood-prone lowland systems to irrigated lowland systems through improved flood control and drainage is one of the reasons for the increase of irrigated rice in Viet Nam from 40 percent in the 1980s to nearly 55 percent in 1997 (FAO-WAICENT, 2002). In the same period, the harvested area increased from 5.6 to over 7 Mha. According to Mahabub-Hossain et al. (1995), provincial and district governments in Viet Nam made substantial investments in the development or reconstruction of: 75 large-scale hydro-agricultural systems; 750 big to medium-sized reservoirs; and 7 000 km of canals and embankments for flood control, drainage and irrigation in the Mekong River Delta (Le Huu Ti and Facon, 2001), and as a result the rice ecosystem has been converted from deep-water to irrigated lowlands. The area under deep-water rice declined from two-thirds to less than one-third of the rice area by the end of the 1980s. The change in cropping system had a positive effect on rice production, but also led to additional costs for agrochemicals and labour and a reduction in income from the upland crops that used to be grown during the dry season after harvesting of deep-water rice. Overall, however, “the average household income from non-farm and off-farm activities was about 40 percent higher in the irrigated ecosystem compared with the deep-water ecosystem. Since these incomes accrue more to the lower income households, the change in the ecosystem has had a positive impact on socio-economic equity and the alleviation of poverty. The investment on the canal systems had high rates of return.” (Duong-Ngoc-Thanh, 1994)

Sustainability options

Improving flood control and drainage: In Bangladesh, an average 22 percent of the country is flooded every year and 50 percent of water development expenditures are spent on flood control and drainage. In Myanmar, in the Ayeyarwady Delta, drainage and flood control structures are also linked. Drainage covers 1 Mha in northern and central Viet Nam, mostly in the Red River Delta. Flood protected areas in China represent 32.69 Mha (Facon, 2001).

River and non-river salt disposal: Above certain salinity level thresholds, drainage return flows cannot be re-used for irrigation of rice. Smedema (2000) examines options for the disposal of salt and suggests exploiting river disposal in basins, for example, in the Indus Basin, where downstream salt concentrations are still relatively low. Non-river disposal options include evaporation pans - isolated areas from which water evaporates leaving behind substantial salt loads.

OPTIONS FOR TECHNICALAND MANAGERIAL CAPACITY-BUILDING

Effective rice water management cannot ignore the constraints and the management capacity of farmers, farmer organizations, irrigation service providers, other services providers and planners. Therefore, it is important that all interventions are based on participatory principles in water resource management and irrigation (FAO, 2002; FAO, 2001; Facon, 2001).

Rice water management options can only be effective if technology transfer needs are identified and a solution found which meets the socio-economic and cultural environment of the farmers. It is, therefore, important that support services be established to ensure the successful and sustainable introduction of new options in farmers’ fields. A cornerstone of the technology transfer process is a strong knowledge-based planning capacity at central level or basin level, with training both for staff at the various institutions and agencies and for farmers. A consistent approach is required at all levels, especially at system level, so as to support changes at field and farm level.

Strong institutional arrangements at basin and national level

It is widely recognized that an integrated water resource management approach requires strong institutional arrangements at all levels, including basin, system and community level. In irrigated systems, a general water-service orientation of semi-governmental agencies and water users’ associations has become a key element of participatory irrigation management and irrigation management transfer.

Basin and national level coordination

In many rice-based basins, national and international river authorities have taken on an important coordinating and advisory role in the allocation and management of water resources. Water for rice production is an important indicator for efficient use of this increasingly scare resource, for which the basin is regarded as the most appropriate unit. The harmonization of national water and food security policies and legislation is an important task to be undertaken by regional organizations and basin authorities. Asian countries, including Malaysia, Philippines, Thailand and Viet Nam, have already engaged in the development of national water visions and strategies towards integrated water resources management, as part of a coordinated regional water vision and strategy dialogue (Le Huu Ti and Facon, 2001). Although the rice sector has not been particularly included in the subregional/regional visions and strategies for the preparation of the Second World Water Forum (The Hague, March 2001), the sector nevertheless has an important place in national strategies in economic, food and social security terms. Recently, however, there has been more focus on water management for rice paddy fields in Asia, notably: the JICID Tokyo Regional Conference on Water Management for Rice Paddies 2000 and the ICID Conference of Seoul 2001 (a preparatory conference for the Third World Water Forum [Kyoto 2003] on Multi-Functionality of Rice Paddy Fields 2002).

Strengthening planning capacity at national and provincial level

The complexity of water management options calls for improved planning capacity at central and basin level. Sophisticated tools for improving the strategic planning of rice-based water resource systems are leaving the research laboratory and are more user-friendly than a decade ago. Successful examples include hydrological models, crop growth models and geographical information systems (GIS).

Hydrological modelling is the application of a large number of hydrological models that estimate the hydrological processes of a basin in time and space. Of numerous models, Hayese (1999) reviewed those of specific relevance in rice-dominant basins, particularly in Japan and elsewhere in Asia. Examples of model applications in Southeast Asia include the Mekong and the Chao Phraya river.

Following intensive field testing and verification, the rice phonology model RIDEV(Dingkuhn, 1995), which is a much reduced version of the ORIZA model series, was transferred to national research and planning authorities in West Africa. The model is successfully used by extension staff to recommend best suitable rice varieties under given climatic conditions.

Similarly, Ceuppens (2000) developed a regional model for estimating rice water requirements for locations in the Senegal River Basin. The model was subsequently coupled with an in-house geographical information system of SAED (Senegal’s irrigation research and extension body).

Strong and autonomous farmer and water user organizations at system level

Farmers and farmer organizations are considered to bear ultimate responsibility for rice system management. Where appropriate, the formation of Water Users Associations (WUA) needs to be facilitated, with support provided in the technical and financial management and operation of systems. Several FAO publications (FAO, 2001; FAO, 2002) provide excellent guidelines as to how participatory irrigation management systems should be encouraged and supported.

Following a more strict service-oriented approach in rice-based irrigation systems, it is evident that a complex network of hydraulic infrastructure of main, secondary and tertiary levels requires an appropriate technical, organizational and institutional set-up, which is coherent, sufficiently flexible and adjustable, in order to adequately respond to the needs of a diversified rice-based smallholder system.

One of the most important issues in participatory irrigation is that farmers should have or share authority at system level; the system will then adapt to accommodate the service requirements of the farmers. It does not mean that they are responsible for O&M at all levels (particularly upper levels), as this may be contracted to various kinds of irrigation service providers. The responsibility of the WUAs may be limited to low-level operations, as is the case in Viet Nam and China. The relationship between upper-level irrigation service providers and their client WUAs can be defined contractually though a service agreement, specifying transactions, service level and accountability mechanisms.

Farmers and extension staff training at field level

Within its Special Programme For Food Security (SPFS), FAO established guidelines for the development of the water management and irrigation development component (FAO, 2001). A phased approach is applied comprising a preparation phase, demonstration phase and extension phase. The result of the preparation phase is a framework for irrigation development and a plan of action for the demonstration phase. The demonstration phase is designed to demonstrate, verify and adapt suitable technologies for improving water management in existing irrigation schemes and new approaches to irrigation development. The results of the demonstration phase will be used in the formulation of an irrigation expansion programme.

An essential element of the programme is the training of national staff and farmers. Staff training consists of a series of training sessions for technical and extension staff. Farmer training is implemented as an integrated part of the extension programme and is scheduled in line with the various activities of the water control component and the staff training programme. The training aims to ensure sustained support to farmers for successful adoption of the new techniques and technologies and the creation of a support structure to guarantee adoption on a national scale.

The transfer of the new technologies is done with the full participation of the farmers. Through participatory appraisal and learning, farmers determine which technologies would best fit into their farming system. Associated problems and constraints are jointly analysed.

The case for a massive retraining of engineers and managers in irrigation agencies, consulting firms and irrigation service providers

Intensified and ongoing training programmes for professionals in the reformed irrigation agencies and consulting firms (providing advisory services to WUAs and their managers) and for technical staff employed for the operation and maintenance of irrigation schemes are essential to the sustained success of the transfer programmes.

It is therefore essential that these programmes introduce and provide knowledge on ways and means to design and operate irrigation systems cheaply for good performance and adequate service to farmers as they evolve towards more commercial forms of agriculture. An appraisal of the initial conditions and of the performance of the systems to be transferred would allow both better design and strategic planning of physical improvements, together with a definition of the service to be provided - by the irrigation service provider to WUAs and by WUAs to their members - and indication of ways and means to achieve these service goals and improve them in the future.

It is suggested that the rapid appraisal process developed and used in the evaluation of modernization programmes and introduced earlier in this paper could be used for this purpose at programme appraisal stage and for individual irrigation systems.

The use of internal process indicators would be useful in monitoring and evaluating systems. A pilot training programme on modernization concepts and application of the rapid appraisal procedure has shown promising results; it builds on the knowledge synthesis acquired in recent years of modern design principles and participatory irrigation management. Its application to a system in Thailand by staff in the Royal Irrigation Department showed great potential (see Appendix 1). Similar training courses were held in Viet Nam in March and May 2002, in Indonesia in July 2002 and in the Philippines in September-October 2002. A concept for a more ambitious retraining programme based on the same concepts and tools has been developed by FAO and could be supported in the context of efforts to improve the performance of programmes to transfer the management of irrigation systems to the users.

CONCLUSIONS

The literature available on rice water management and irrigation matters provides recommendations for increasing water-use efficiency in irrigated systems at field and farm level. An impressive contribution is made by rice researchers from China, who questioned the traditional method of submerged irrigation and systematically developed alternatives based on a combination of submerged and unsaturated soil moisture conditions. These water management regimes have clearly stimulated broad discussion. However, the proposals may not be a feasible option for rice farmers who do not have access to full or partial water control facilities.

Therefore, the research community and development institutions must propose, on the one hand, options for improving irrigation systems in order to improve water control at farm level to enable farmers to adopt these new technologies and, on the other, options for improving water management but which do not require full or partial water control at farm level.

A hydrology-driven and rice-ecology-based approach in rice water management appears to be the appropriate way to address the diversity of constraints in upland and lowland systems, where elongated droughts and increased floods alternately threaten large areas of paddy fields. The rice ecology of a given location and its economic circumstances determine the degree of freedom for effective intervention in the water resource system. Moreover, field- or farm-level options are complemented by a set of system- and basin-level options upstream.

An integrated resource management approach in rice water management is considered the appropriate answer to the coexistence of several rice ecosystems that cannot be seen in isolation. By linking upland and downstream systems through: water conservation; more efficient use of water in the lowland systems; and reduction of the non-beneficial effects of water, such as flood damage, waterlogging and salinity, it is believed that the beneficial and productive overall water-use efficiency can be increased effectively. Effective rice water management should combine: improved assessing and planning of water resources at all levels; utilization of improved technologies; and increased management capacity of the rice-growing communities, who bear the ultimate responsibility for the management of their resource base.

In irrigated systems, particular attention is to be given to improving quality water services at field level, which includes much improved water supplies in terms of flexibility and reliability, as well as access to sufficient drainage when required. Otherwise, none of the field level options would be effective. The modernization tools and concepts have been successfully introduced in rice irrigated systems in Southeast Asia by FAO through training courses, and it is believed that they would greatly facilitate the adoption of feasible strategies for system modernization.

Such ambitious approaches require the full attention of all stakeholders, including researchers, donors and development organizations, irrigation agencies, irrigation service providers, communities, water users and farmers. It is essential to create an enabling technical, social and institutional environment that is conducive for the assessment, development and utilization of intelligent water management practices.

REFERENCES

Albinson, B. & Perry, C.J. 2002. Fundamentals of smallholder irrigation: The structured system concept. Research Report 58. Sri Lanka, International Water Management Institute.

Asch, F., Dingkuhn, M. & Dörfling, K. 1997. In Miéezan, K.M., Wopereis, M.C.S., Dingkuhn, M., Deckers, J. & Randolph, T.F. Irrigated rice in the Sahel: prospects and sustainable development. Boauke, Côte d’Ivoire,WARDA.

Becker, M. & Johnson, D.E. 1999. Rice yield and productivity gaps in irrigated systems in the forest zone of Côte d’Ivoire. Field Crops Research, 60: 201-208.

Bertrand, R., Keita, B., N’Diaye, M.K. & Veldkamp, V.J. 1995. La dégradation des sols irrigués en Afrique de l’Ouest. In Miéezan, K.M., Wopereis, M.C.S., Dingkuhn, M., Deckers, J. & Randolph, T.F. Irrigated rice in the Sahel: prospects and sustainable development. Boauke, Côte d’Ivoire,WARDA.

Boivin, P. 1995. Soil degradation in irrigation schemes in the Senegal river middle valley: mechanism characterization and actual situation. In Miéezan, K.M., Wopereis, M.C.S., Dingkuhn, M., Deckers, J. & Randolph, T.F. Irrigated rice in the Sahel: prospects and sustainable development. Boauke, Côte d’Ivoire,WARDA.

Bouman, B.A.M. & Tuong, T.P. 2000. Field water management to save water and increase its productivity in irrigated lowland rice. Agricultural Water Management, 1615: 1-20.

Burt, C.M., Clemmens, A.J., Strelkoff, T.S., Solomon, K.H., Bliesner, R.D., Hardy, L.A., Howell, T.A. & Eisenhauer, D.E. 1997. Irrigation performance measures - efficiency and uniformity. J. of Irrigation and Drainage Engineering, ASCE 123(6): 423-442.

Ceuppens, J. 2000. Water and salinity management for irrigated rice in the Senegal River Delta. Catholic University of Leuven, Belgium. (Dissertation)

Dawe, D., Seckler, D. & Barker, R. 1998. Water supply and research for food security in Asia. In Proceedings of the Workshop on Increasing Water Productivity and Efficiency in Rice-Based Systems, July 1998, IRRI, Los Baños, Philippines.

Diallo, S. & Johnson, D. 1995. Les adeventices du riz irrigué au Sahel et leur contrôle. In Miezan, K.M., Wopereis, M.C.S., Dingkuhn, M. & Randolph, T.F. Irrigated rice in the Sahel: prospects for sustainable development. Boauke, Côte d’Ivoire, WARDA.

Dingkuhn, M. 1995. Characterising irrigated rice environments using the rice phonology model RIDEV. In Miezan, K.M., Wopereis, M.C.S., Dingkuhn, M. & Randolph, T.F. 1995. Irrigated rice in the Sahel: prospects for sustainable development. Boauke, Côte d’Ivoire, WARDA.

Duong-Ngoc-Thanh. 1994. Productivity and profitability of cropping systems in deepwater and irrigated lowland rice ecosystems in the Mekong Delta of Vietnam. Laguna, Philippines, College. 165 pp.

Facon, T. 2000a. Intervention 14 of the FAO electronic workshop “Land-water linkages in rural watersheds”, FAO, Rome.

Facon, T. 2000b. Performance evaluation of Makhamtao-Uthong project with a rapid appraisal procedure. In Proceedings of Asian Regional workshop on Sustainable Development of Irrigation and Drainage for Rice Paddy Fields, p. 312-316, 24-28 July, Tokyo, Japan. Japanese National Committee of ICID.

Facon, T. 2001. Rice production and water management in Asia: some issues for the future. Bangkok, FAO.

FAO. 1977. Crop water requirements. Irrigation and Drainage Paper No. 24. Rome.

FAO. 1999. Modern water control and management practices in irrigation - Impact on performance. Water Series Paper No. 19. Rome.

FAO. 2001. Irrigation guidelines on CD-Rom. Land and Water Digital Media Series No. 12. Rome.

FAO. 2002. Design, policy management and performance of irrigation projects. Bangkok.

FAO-WAICENT. 2002. Available online at www.fao.org/WAICENT/AGRICULT/AGP/AGPC/doc/riceinfo/Asia/VIEBODY.htm.

Guerra, L.C, Bhuiyan, S.I., Tuong, T.P. & Barker, R. 1998. Producing more rice with less water from irrigated systems. SWIM Paper No. 5, Colombo, International Water Management Institute.

Hayese, Y. 1999. Runoff analysis of paddy field based watersheds. In Mizutani, M., Hasegawa, S., Koga, S., Goto, A. & Murty, V.V.N. 1999. Advanced paddy field engineering. Tokyo, Japan, The Japanese Society of Irrigation, Drainage and Reclamation Engineering.

Hayese, Y. 1994. Evaluation of paddy field’s function for flood control and a proposal of its enhancing project. Cited in Mizutani, M., Hasegawa, S., Koga, S., Goto, A. & Murty, V.V.N. 1999. Advanced paddy field engineering. Tokyo, Japan, The Japanese Society of Irrigation, Drainage and Reclamation Engineering.

Horst, L. 1998. The dilemmas of water division. Consideration and criteria for irrigation system design. Sri Lanka, International Water Management Institute.

Hundertmark, W. & Abdourahmane, A.T. 2000. Model analysis of rice-based irrigation system performance - a case study from Côte d’Ivoire. Discussion paper, IWMI program planning meeting, Colombo.

Keller, A. & Keller, J. 1995. Effective efficiency: A water use concept for allocating freshwater resources. Water Resources and Irrigation Division Discussion Paper 22. Arlington, Virginia, USA, Winrock International.

Keller, A., Keller, J. & Seckler, D. 1996. Integrated water resource systems: Theory and policy implications. IWMI Research Report No. 3. Sri Lanka, International Water Management Institute.

Kirsch, B. 2002. Land use impacts on water resources: a literature review. Discussion paper, Rome, FAO. Khush, G.S. 1984. Terminology for rice growing environments. In IRRI. Terminology of rice growing environments. Los Baños, Philippines, IRRI.

Lantin, R.S., Wassmann, R. & Neue, H.U. 1995. Climate change and wetland rice ecosystem. (IRRI Call no. QD49 A7A85 1995), Philippines, IRRI.

Le Huu Ti & Facon, T. 2001. From vision to action. A synthesis of experineces in Southeast Asia. The FAO-ESCAP Pilot Project on National Water Visions, Bangkok.

Hossain, M., Duong Ngoc Thanh & Gascon, F.B. 1995. Change from deepwater to irrigated rice ecosystem in the Mekong River Delta, Vietnam: impact on productivity and on farmers’ income. In Denning, G.L. & Vo-Tong-Xuan eds. Vietnam and IRRI: A partnership in rice research, p. 263-273. Proceedings of a conference. Los Baños, Laguna, Philippines, IRRI.

Malano, H. & van Hofwegen, P. 1999. Management of irrigation and drainage systems - a service approach. IHE monograph 3, A.a. Rotterdam, Balkema/Brookfield.

Merrey, D.J. 1996. Institutional design principles for accountability in large irrigation systems. Research Report 8. Colombo, Sri Lanka, International Irrigation Management Institute.

Mao Zhi. 2001. Water-efficient irrigation and environmentally sustainable irrigated rice production in China. Wuhan, China, Wuhan University.

Mishra, S., Rath, A.K., Adhya, T.K., Rao, V.R. & Sethunathan, N. 1997. Effect of continuous and alternate water regimes on methane efflux from rice under greenhouse conditions. Biol. Fertil. Soils, 24(4): 399-405. Berlin, Germany, Springer-Verlag.

Molden, D.J., Keller, J. & Sakthivadivel, R. 2001. Hydronomic zones for developing basin water conservation strategies. Research Report 56. Colombo, Sri Lanka, International Water Management Institute.

Molle, F., Sutthi, Ch., Keawkilya, J. & Kkorpraditskul, R. 1999. Water management in raised bed systems: a case study from the Chao Phraya delta, Thailand. Agricultural Water Management, 39: 1-17.

Nesbitt, H.J. 1997. Rice Production in Cambodia. Los Baños, Philippines, IRRI.

Perry, C.J. 1996. The IIMI Water balance framework: a model for project level analysis. Research Report 5. Colombo, International Water Management Institute.

Purkey, D. & Vermillion, D. 1995. Lift irrigation in West Africa: Challenges for sustainable local management. IIMI Working Paper, Sri Lanka, International Irrigation Management Institute.

Raes, D., Deckers, J. & Diallo, A.M. 1995b. Water requirement for salt control in rice schemes in the Senegal river delta and valley. Irrigation and Drainage Systems, 9: 129-141.

Raes, D., Sy, & Deckers, J. 1995a. Simulation of irrigation water requirements of rice schemes in the Senegal river delta and valley. In Miezan, K.M., Wopereis, M.C.S., Dingkuhn, M. & Randolph, T.F. 1995. Irrigated rice in the Sahel: prospects for sustainable development. Boauke, Côte d’Ivoire, WARDA.

Satoh, M. & Goto, A. 1999. Water Reuse Systems. In Mizutani, M., Hasegawa, S., Koga, S., Goto, A. & Murty, V.V.N. Advanced paddy field engineering. Tokyo, Japan, The Japanese Society of Irrigation, Drainage and Reclamation Engineering.

Seckler, D. 1999. Revisiting the “IWMI Paradigm:” Increasing the efficiency and productivity of water use. Water Brief 2. Colombo, Sri Lanka, International Water Management Institute.

Smedema, L. 2000. Irrigation-induced river salinization: Five major irrigated basins in the arid zone. Colombo, Sri Lanka, International Water Management Institute.

Smedema, L.K. & Rycroft, D.W. 1983 Land drainage: planning and design of agricultural drainage systems. Ithaca, New York, Cornell University Press.

Solaimalai, A., Sivakumar, C., Chandrasekaran, R., Sankaranarayanan, K. & Sdhakar 2000. Water management practices for rice a review. Agric. Rev., 21(1): 53-59.

Spurgeon, D. 1999. Water: A looming crisis. Los Baños, Philippines, IRRI.

Takese, K. 1999. Environmental issues. In Mizutani, M., Hasegawa, S., Koga, S., Goto, A. & Murty, V.V.N. 1999. Advanced paddy field engineering. Tokyo, Japan, The Japanese Society of Irrigation, Drainage and Reclamation Engineering.

Van der Hoek, W., Sakthivadivel, R., Renshaw, M., Silver, J.B., Birley, M.H. & Konradsen, F. 2001. Alternate wet/dry irrigation in rice cultivation: A practical way to save water and control malaria and Japanese encephalitis? Research Report 47. Colombo, Sri Lanka, International Water Management Institute.

Walker, S.H. 1999. Causes of high water losses from irrigated rice fields: field measurements and results from analogue and digital models. Agricultural Water Management, 40: 123-127.

Watanabe, T. 1999. Irrigation water requirement. In Mizutani, M., Hasegawa, S., Koga, S., Goto, A. & Murty, V.V.N. Advanced paddy field engineering. Tokyo, Japan, The Japanese Society of Irrigation, Drainage and Reclamation Engineering.

Appendix 1
Proposals for the modernization of the Makhamtao-Uthong system
[39]

The Makhamtao-Uthong irrigation project in Thailand was appraised with a Rapid Appraisal Procedur (RAP), to develop a Modernization Priority Action Plan. The RAP evaluates a project to assess what type of modernization is needed, through external performance indicators characterizing the inputs/outputs of irrigation projects and internal process indicators that rate hardware, management, and service throughout the system, to visualize where changes are needed, and what impact the changes would have. Simple and inexpensive hardware and operational changes could give immediate benefits to most projects. The project is performing well in terms of productivity and water use despite the low capacity delivery of the canal system. Farmers have invested in individual farm pumps which have allowed secondary water sources to be tapped, the development of conjunctive use, increased reliability in water supply and to some extent in crop diversification and fishponds. The importance of individual pumping is overwhelming but has an economic impact on farmers’ income. Evaluations of the project with external indicators would have concluded that the project performs well in productivity and water use and recommended only institutional measures. The internal indicators provided the basis for an improvement program to enhance operation, management and outputs.

Internal process indicators computed for the Makhamtao-Uthong project were the following:

I -1

Actual service to individual fields, based on traditional on-farm irrigation methods

4.1

I- 2

Actual Service to avg. point of Effective Differentiation based on Traditional On-Farm Methods

2.6

I -4

Actual Service by main canal to its subcanals

3.7

I- 5

STATED service to fields.

3.4

I- 6

STATED service to avg. point of EFFECTIVE differentiation.

4.1

I- 8

Stated Service by main canals

6.9

I- 9

Evidence of Lack of Anarchy in Canal System u/s of ownership change

4.8

I- 10

Cross-Regulator Hardware (Main Canal)

2.3

I -11

Capacities (Main Canal)

5.7

I- 12

Turnouts (from Main Canals)

5.1

I- 13

Regulating Reservoirs

0

I- 14

Communications (Main Canal)

4.7

I -15

General Conditions (Main Canal)

7

I -16

Operation (Main Canal)

0.5

I -28

Number of Turnouts/(operator, gate oper., supervisor)

0.5

I -29

Feedback Information

0.5

I -30

Computers for billing/record management

2

I -31

Computers for Canal Control

0

I-32

Effectiveness of water supply releases from reservoir

3

I-33

Effectiveness of main system operation

5

I-34

How closely are instructions followed?

10

The analysis of the internal indicators and sub-indicators reveals that:

- the main canal provides a very poor and inequitable service to the secondary canals and sub-projects

- the reason lies not in poor maintenance nor in the cross-regulators but how they are operated

- secondaries provide a very poor service to the points of effective differentiation (irrigation blocks

- poor performance at the farm level is compensated by pumping and conjunctive use

- there is a water supply problem which cannot be solved overnight (linked to water level at the Chao Phraya dam on the Chao Phraya River)

- as instructions are followed, the solution lies in changing the instructions to gate operators. Some minor adaptation of cross-regulators would provide much more flexible and equitable distribution

- communications and procedures can be drastically improved

- staff density is very high and can be reduced

- there is no reliable measurement at any level

A transfer of the canal as it is presently operated would create problems of inequity between secondaries and sub-projects which cannot be solved by institutional measures alone. Rules established by WUAs equivalent to present rules would be subverted by farmers as they are at present, for the same reasons. Establishment of WUAs at the level of secondaries would allow to control the problem of illegal turnouts and implement a different operation strategy negotiated with the main canal ISP. Upstream and downstream areas have different cropping patterns which could be accommodated by a different service. Problems of inequity between upstream and downstream also require a re-centralization of operational responsibilities for water dispatching in the main canal.

The recommendations derived from the assessment of the external and internal indicators included:

Priority

Cost (Million Baht)

Action

1

0

Change instructions: water level control and empower operator to make adjustments

2

0

Establish a single operation unit for the main canal

3

5

Flow measurement at the head of the canal and each project

4

1.2

Flow measurement at the head of each secondary (flumes)

5

5

Long-crested weirs at cross regulators

6a

0

Better transfer of data on turnouts

6b

0.1

Walkie talkies for zonemen and gate operators

6c

0.4/year

Improved mobility/transport

7

6

Motorize cross-regulator gates

8

0

Control or eliminate illegal turnouts

Longer-term measures identified were:

1. Management of water recirculation within the project

2. Management of conjunctive use

3. Development of tertiary network

4. Improve water supply to the main canal

Benefits expected from a modernization program are expected to be a reduction in pumping costs and a reduction in operation & maintenance (O&M) staff. Given the present efficiency of the project and of the Lower Chao Phraya Project it is doubtful that a modernization program could generate some water savings. However the improved reliability associated with a better discipline in water allocation should have a positive impact on crop productivity. Proposed changes would enable a transfer of water management to users in good conditions of equity between different areas with the possibility to implement rules at the level of the secondaries, and to apply and enforce a water allocation and water charging system. Most of the actions identified are of the software type and require training only. The physical upgrading identified is minimal and costs less than the upgrading budgets currently spent on the system.

Appendix 2
Strategies for effective water control at basin, system, field and crop level


EFFECTIVE EFFICIENCY

REMARKS


Rainfed Upland

Rainfed Lowland

Deep-water

Irrigated Rice


1 Field-level options






1.1 Demand management options






(i) Water conservation through technical re-design

Rainfall efficiency (+)

Rainfall efficiency (+)

No effect

Rainfall & field application efficiency (+)

Short-term

(ii) Practising irrigation scheduling






Submerged/wet&dry (SWD)

With partial water
Application eff. (+)
Rainfall eff. (+)
Soil water-use eff.

No effect

Application eff.(+)
Rainfall eff. (+)
Soil water-use eff. (-)

Effective only with full or partial water control and flexibility of supply

Alternating wet & dry (AWD)

Semi-dry cultivation (SDC)

Keeping soil moisture at saturation

(iii) Reducing water use for land preparation and crop establishment

Soil water-use efficiency

Soil water-use efficiency

No effect

Soil water-use efficiency


(iv) Crop intensification/diversification






Variety change

Crop water productivity (+)

Crop water productivity (+)

Crop water productivity (+)

Crop water productivity (+)

Medium- to long-term

Chemical weed control

Integrated nutrient management

1.2 Reducing non-beneficial effects of water use (sustainability options)






(i) Improving drainage control

Crop water productivity

Crop water productivity

Crop water productivity

Crop water productivity


(ii) Salinity control (“puddling and flushing” and leaching)

Crop water productivity (+);
Leaching efficiency (+)

No effect

Crop water productivity (+), leaching efficiency


(iii) Use salt-tolerant varieties

Crop water productivity (+)


2 System level options






2.1 Improving the availability of water(supply management options)






(i) Improving service provision through redesign, rehabilitation and modernization

Rainfall efficiency (+)

No effect

Irrigation efficiency (+), Supply equity (+) Supply reliability (+) and flexibility (+);


(ii) Optimizing water re-use systems and conjunctive use of groundwater

Recycling efficiency (+); System water productivity (+)

System recycling eff. (+); Dry season effect: on irrigation intensity and crop water productivity

Sustainability risk through resource degradation

2.2 Optimizing the efficiency of productive and beneficial water use (demand management options)






(i) Optimizing cropping pattern through intensification & diversification

Cropping intensity (+); Crop water productivity (+)

Not an option

Cropping intensity;


2.3 Reducing the non-beneficial effects of water (sustainability options)






(i) Improving the drainage system

System-level crop water productivity

System-level crop water productivity

System-level crop water productivity

System-level crop water productivity


(ii) Salinity control




System level drainage efficiency (+)


3 Basin level






3.1 Supply management






(i) Upstream water conservation

Basin-level rainfall efficiency (+)

Effective only within basin of up 100 km2

(ii) Facilitating water re-use systems through improved recapturing and conjunctive use of groundwater

Recycling efficiency (+), Basin-level efficiency (+)

Basin recycling eff. (+); Dry season effect: on irrigation intensity and crop water productivity

Sustainability risk through resource degradation

(iii) Conversion of rainfed lowland and deepwater systems to irrigated systems


Crop water productivity (+)



3.2 Reducing the non-beneficial effects of water(sustainability options)






(i) Improving flood protection and prevention

Rainfall eff. (+), Crop water productivity (+),


(ii) Improve main drainage discharge capacity through the installation of drainage systems

Rainfall eff. (+), Crop water productivity (+),


(iii) River- and non river salt disposal




Basin leaching efficiency (+)


a Photosynthetic productivity, repiration efficiency and harvest index.


[34] Estimates by Guerra et al. (1998) point to a drop in irrigated land per 1000 people from 19.4 to 18.4
[35] Irrigated area expansion has continued, however, mostly through largely private groundwater development.
[36] Irrigation development in the last 50 years, i.e. most irrigated areas in Asia, was not initiated by farmers but by irrigation agencies and therefore did not follow this evolution process.
[37] Rice is a special case due to the specific complexity of the rice ecology and cultural practices. Estimates of the water needs require additional calculations on the part of planners.
[38] Reas and Deckers (1995) estimated that in one season about 4 tonnes of salt are taken in by both irrigation water and capillary rise from the groundwater.
[39] Source: Facon, 2000b.

Previous Page Top of Page Next Page