A. FINAL REPORT 2005 - RAPID ASSESSMENT STUDY - TOWARDS INTEGRATED PLANNING OF IRRIGATION AND DRAINAGE IN EGYPT IN SUPPORT OF THE INTEGRATED IRRIGATION IMPROVEMENT AND MANAGEMENT PROJECT (IIIMP)

INTRODUCTION

In early 2004, the Agriculture and Rural Development Department (ARD) of the World Bank invited IPTRID to conduct a Rapid Assessment of Egypt's current irrigation and drainage development. The purpose of the study was to assist the Government of Egypt (GOE) to plan and design the Integrated Irrigation Improvement and Management Project (IIIMP), which follows on from the Irrigation Improvement Project (IIP). The study took advantage of an innovative method known as the DrainFrame approach (Abdel-Dayem et al,. 2004), which was conceived as a complement to ongoing preparations for the Integrated Irrigation Improvement and Management Project. IPTRID agreed to divide the study into two phases, of which Phase I involved intensive field and consultation work. Phase II was to present the findings to a larger audience of stakeholders. Their comments and suggestions formed the basis of an improved main report, which was completed in November 2004.

The present report contains the findings of the Phase I Rapid Assessment Study conducted during May and June 2004 in Egypt. Following this brief introduction the report provides information on the study background and the national context of the water sector. A description of the study follows and the principles of the DrainFrame approach explained. This includes definitions of landscapes, their functions, values, and the diagnostic steps that lead to the improved selection of interventions for the water management system. Stakeholders are described and attention is drawn to existing problem areas, the causes and effects of the various elements of the land and water management systems and the socio-economic conditions of the people involved. Opportunities for improvement are elaborated on for each problem area. The anticipated effects are described in technical terms and the possible economic, social and environmental impacts discussed. Conclusions and recommendation are made focusing on the six key areas of improvement: the use of DrainFrame in IIIMP, the integration of irrigation and drainage systems, the integration of water resource management, institutional capacity building and environmental management, and future research needs.

The rapid assessment study was conducted over a period of less than one month and the result is rather qualitative, further quantification of anticipated effects and impacts would be worthwhile in the proposed follow-up study.

STUDY BACKGROUND

Context of the water sector

In 2000, Egypt ranked at the top of the list of water-scarce countries with its internal renewable water resources of about 900 m³ per person (FAO Aquastat, 1995). Over 90 percent of this resource was withdrawn by irrigated systems (92 percent). The Government increasingly recognizes that the rapidly growing population's high demand for freshwater resources requires rational improvement, effective planning and management (MWRI, 2002). The expected increase in population will cause the internal renewable water resource situation to fall to 350 m³ per person by 2025. Demand for irrigation water will continue to increase over the next decade and beyond and, in the future, the agricultural sector will have to adjust to a smaller share of the Nile's water (MALR/FAO, 2003).

In its contribution to the Third World Water Forum, held in Kyoto in 2003, Egypt's Ministry of Water Resources and Irrigation identified fragmentation among different governmental institutions as a key constraint to more effective water management. An overwhelming number of governmental and non-governmental institutions are involved in water resource management at all levels, including the national, governorate and district level. Linkages between the Ministry of Water Resources Management and Irrigation with other stakeholders do not appear strong enough to eliminate conflicts (MWRI, 2002). Accordingly, the fragmented responsibility for the planning and design of irrigation and drainage is perceived as a major constraint for improved water use efficiency in Egypt.

The Kyoto Report identified nine determining factors for water resource development in Egypt that may affect the sector either positively or negatively. These factors include population growth, current irrigation practices, functional local water user institutions, youth employment, land availability and tenure, cropping patterns (rice), food security policies and water pricing. In response to these constraints, the Government identified the challenges for future water-resource management in Egypt. These include meeting basic needs, securing food security, protecting ecosystems, shared water resources, valuing water, managing risk and governing water wisely.

The Government of Egypt is firmly committed to face the compelling challenges imposed by water scarcity and competition for water use, while demands for crop production are increasing. A number of important projects have been initiated and implemented, among which the National Drainage Project and the Irrigation Improvement Project were significant, both projects were co-financed by the World Bank. Over the next two decades, substantial investments will be required in the water sector to achieve improved water use efficiency and water quality and to protect water bodies from pollution.

Irrigated agriculture

With some 16.8 percent of added value Egypt's agricultural sector made a significant contribution to the 2002 gross domestic product (GDP) of US$1 470 per capita. With a growing population, and limited opportunities for increasing water supplies, there is enormous pressure on the Government to meet future food demand. In its 2017 agricultural strategy (MALR/FAO, 2003) the Government emphasized the need to considerably increase water use efficiency. Accordingly, total water consumption is to rise from about 49.2 billion m³ in 1995 to about 67.0 billion m³ in 2017. The estimate assumes a cropped area of about 21.9 million feddans in 2017. Efficiency gains are expected through:

• decreasing the area under rice to one million feddans, in addition to planting a short duration and low water consuming rice variety;

• savings in sugar cane water consumption resulting from improved water use efficiency and land levelling; and

• further savings are expected from the enhanced use of drainage water and non-conventional water resources.

It is estimated that newly developed short duration rice varieties, coupled with water management changes, could reduce applied water by about 2 000 m³ and "save" as much as 1 000 m³ of consumptive use per feddan. However, early planting of the winter crop (principally clover) would somewhat reduce these savings. Today, rice has become a very attractive crop to farmers, particularly short season varieties that permit the establishment of early clover. The Ministry of Water Resources and Irrigation (MWRI) personnel unofficially estimated that rice cultivation expanded to about 2 million feddans in 1999, which would have eliminated most of the calculated water savings (MALR, 2003)^[1].

Productivity and efficiency of water use

Ninety-seven percent of fresh renewable water resources are controlled at a single location (Aswan High Dam). Flows target a single sink, which is the Mediterranean Sea. Between the two points, no significant storage or disposal facilities can be found including numerous canals, pumping stations, culverts, control points, water reuse stations, drainage facilities and disposal points. The travel time from Aswan to the sea takes about 12 days. The current complexity of the water control system has been evolving over the centuries, and has been governed by different technological ages, political systems and social values.

In 1995, Keller et al., published the results of a strategic research programme on the effective irrigation efficiency of Egypt's Nile system. Their main findings suggest that the effective irrigation efficiency of the Nile basin between the High Aswan Dam and the Mediterranean Sea is 94.8 percent, which is nearly twice as much as classical irrigation efficiency (57 percent)^[2].Their conclusion is that the concept of classical irrigation efficiency is not well suited to making water allocations and policy decisions. Accordingly, Egypt's planners are advised not to make the mistake of justifying and authorizing irrigation improvement projects designed to raise classical irrigation efficiency.

In a study on the use and productivity of Egypt's Nile water, Molden et al., (1998) used the 1993/94 Nile water balance to prepare a water account, which is a methodology for revealing water use information at different levels. It was found that most of the water released from the High Aswan Dam is depleted (84 percent) and being put to productive use (process fraction 0.82). In financial terms, each cubic metre of water delivered (gross) in agriculture returned US$0.16, which is classified as "quite good". The opportunities for water savings are restricted to capturing more drainage water and the potential for reuse is determined by the quantity of water (and salt) that is to be released into the northern lakes and the Mediterranean Sea.

The status of project development

The Irrigation Improvement Project (IIP)

Parallel to the implementation of drainage facilities, the Government launched several pilot projects that aimed to improve irrigation facilities, productivity of irrigated land and irrigation management. The Irrigation Improvement Project (IIP), jointly financed by the International Bank for Reconstruction and Development (IBRD) and Kreditanstalt fur Wiederaufbau (the German Bank for Reconstruction, KfW), has been a milestone project in this respect. Near completion, (December 2005) its main interventions include improvement of mesqa systems, a change from rotational to continuous flow irrigation at the branch canal and tertiary (mesqa) levels and promotion of water user participation in irrigation system operation, maintenance and management through water user associations.

Equally important, the Government enacted several policies that are of paramount importance to the development of improved water resource management arrangements based on considerations of integration and sustainability.

The National Drainage Programme (NDP)

Soon after the completion of the Aswan High Dam (1968), a sizeable programme of drainage implementation began with funding from the national budget and international organizations. In 1973, a central organization was established within the Ministry of Irrigation for the planning, design, installation and maintenance of drainage works, named the Egyptian Public Authority for Drainage Projects (EPADP).

EPADP's objective of designing and implementing subsurface drains on 2.7 million hectares and open drains on 3.1 million hectares has led organizational development for the past 30 years. Today, more than 75 percent of the horizontal pipe drains in the planned for areas, and more than 95 percent of the open drain areas have been completed.

Incremental agricultural benefits that can be attributable to this development are estimated at US$200 and US$375 per hectare, depending on the level of salinity in the area (van Achthoven et al., 2003). Identified positive drainage effects on the natural resource system include improved soil fertility, workability and the removal of accumulating salts from the root zone. Management responsibility of the drainage facilities remains with the implementing agency, EPADP, which introduced a cost recovery system for the investment including crop compensation to farmers who lost land or were affected by construction work. The collection mechanism is part of the general land tax that is payable by all farmers.

Integrated Irrigation Improvement and Management Project (IIIMP)

The development objective of the new Integrated Irrigation Improvement and Management Project (IIIMP) is to assist implementation of the measures to ensure efficient and sustainable use of the country's water and land resources. The proposed project would provide for the introduction of a framework for integrated water resources management (IWRM) in the irrigation and drainage sector. This will support planned institutional reforms including rehabilitation of irrigation, drainage and pumping stations and removal of system bottlenecks in the major canals of the three command areas in the Nile Delta. The main objectives of the project are to:

a) develop a framework for IWRM plans and programmes in selected command areas, considering management of water quantity and quality by inter-agency and stakeholder consensus;

b) improve institutional, financial and environmental sustainability of water services through intensive user and private sector participation in the investment and operation and maintenance at the district/branch canal levels and below, and to improve water quality management practices; and

c) increase farm incomes through improved agricultural production based on efficient, more equitable and sustainable use and management of water and land resources.

The priority for IIIMP is to address water management improvement in each of the selected command areas as a single integrated venture. Increased stakeholder involvement at all stages will become a key feature of this new project. Coordinated planning and management of interventions will require related institutional development activities within and among responsible agencies. Multiple stakeholder involvement would imply the development and evolution of water user organizations within the command areas. Furthermore, the creation of effective planning and management arrangements at the command area level is an important objective of the project.

STUDY APPROACH

DrainFrame concept

DrainFrame (Drainage Integrated Analytical Framework) is a tool for analysing and assessing functions and values embedded in a participatory planning process. The analytical component of the tool comprises systematic mapping of the functions of goods and services provided by natural resources systems and the values attributed to these functions by people. In addition, the implications (effects and impacts) of particular drainage interventions are explored. The tool includes a communication, planning, and decision-making support component to facilitate discussion and negotiation of tradeoffs related to the different functions and values directly related to and influenced by drainage (Abdel-Dayem et al., 2004; Abdel-Dayem, 2004).

DrainFrame stresses the importance of landscapes and their functions. Landscapes are defined as units of land with homogeneous natural resources such as soil, water, climate, vegetation, etc. that perform a homogeneous set of functions^[3], which subsume the goods and services provided and performed by natural resource systems. A natural resource system, such as soil, is not a service provider in a classical business sense; however, it is potentially productive. Those using the system will regard it as more valuable the more productive it is. Nonetheless, use of a natural resource system may generate externalities that other members of a society, such as domestic households and fishermen may perceive as disturbing, damaging or as counter productive. Both internal and external effects on functions should be taken into account to evaluate the costs and benefits of a project seeking to intervene in the systems and to remove a constraint or problem in that sector (Abdel-Dayem et al. 2004).

Figure 1 Revised iterative analysis of the DrainFrame approach applied in the study

Applied DrainFrame approach

DrainFrame provides a coherent framework of thinking and analysis that can assist in the formulation of adequate water resources management interventions. It implies an iterative process of descriptive and diagnostic steps at different landscape levels that address issues at the various levels of organization (from local to national). The DrainFrame approach is based on the active participation of stakeholders (people and institutions) and appreciates the diversity of existing water resource use and management situations.

It is believed that using the integrated approach both potential primary and secondary benefits as well as social and environmental hazards and risks may be more clearly identified and earlier. As a result, ensuring their inclusion in the feasibility study, would ultimately lead to a better choice of physical intervention, ownership and broader support from an increasing number of stakeholders. The team realized at the beginning of the study that the DrainFrame approach should be geared towards a "real life situation". This implied a translation of the theoretical approach into a practical and tangible team-oriented methodology. Most important, the methodology needed adjustment to accommodate the institutional aspects in the analytical process more visibly. In order to gear the approach towards a "real life situation", the team realized there was a need to conduct a "problem assessment" prior to the impact assessment and support to decision-making. The revised iterative analysis of the DrainFrame approach is shown in Figure 1 while the sequence for the applied analytical study approach is summarized in Table 1.

DESCRIPTION OF THE STUDY AREA

General

The study area is located in the northwest area of the Nile Delta, in Beheira governorate, about 150 km southwest of the city of Alexandria, (see Figure 2).

Landscapes

A distinction is made between landscapes identified within the Mahmoudia command area (MCA) and those in neighbouring areas upstream and downstream of the MCA. This distinction is relevant as the DrainFrame methodology differentiates between on-site and off-site effects of water management interventions. Off-site effects take place on neighbouring ground and on-site effects can be identified within the command area. Overall, the study identified the following main types of landscapes:

Mahmoudia command area

• Alluvial land

• Reclaimed lake bottom

• Irrigation canal network (main, branch, mesqa)

• Drainage canal network (main drain, collector drain, subsurface drains)

Upstream

• Rosetta branch and river levees

• Upstream old land command areas

• Nuberia old-new land

Downstream

• Coastal lakes: open water and reed land

• Coastal lakes: fishponds

• Coastal lakes: reclaimed/abandoned land

• Coastal ridges and dunes

• The Mediterranean

Table 1 Applied study approach

Figure 2 The Mahmoudia study area

Source: FAO country profiles

In the proposed classification of landscapes, it is assumed the hydraulic water control systems (irrigation and drainage canals) to form landscapes of their own. This is because these water control systems can be characterized using a specific set of functions that are difficult to attribute to land and the boundaries of the command area are greatly determined by the reaches of those control systems. It is important to note that this command area boundary is not congruent with the catchment area for the drainage network. For example, the Umum and Edku main drains take on water from both the MCA and other command areas. The secondary drains normally withdraw water from both sides of the branch canal. Similarly, the area served by the sub-surface drainage systems is not necessarily equivalent to the area commanded by the tertiary irrigation canals (mesqas).^[5] Other important landscape features include settlements, such as urban centres, villages as well as coastal lakes (Lake Edku) and its surrounding areas, dunes and the Rosetta branch of the River Nile. Land located in the west of the old Nile Delta is called new land, indicating that the area has been subject to recent land development see Figure 3 and Table 2).

Figure 3 Sketch of the study area showing hydro-ecological zones and landscapes

Note: not to scale - Source: Working paper, Natural Resources Perspective

Mahmoudia command area

Alluvial land and reclaimed lake bottoms

The MCA was largely developed on the old alluvial land of the River Nile Delta. The total area is about 122 000 hectares (305 000 feddans). Soils are heavy clays, mainly of montmorillonite type. The topography is very flat generally sloping from 2 metres (m) above sea level in the south to 1-2 m below sea level in the lowest northwestern parts. Isolated higher spots (up to 5 m above sea level) most likely form remnants of old river levees and dunes and are found scattered throughout the command area (EGSA, 1996). These are often used for settlements. Groundwater is found at shallow depths.

There is significant rainfall only during the winter season reaching an average of 200 mm per year near the coast and only 175 mm per year in the south of the MCA (RIGW, 1986). Waterlogging is frequently observed where the drainage of land is limited, especially along continuously flowing irrigation canal sections and is caused by seepage from the canal. Nearly the entire study area is covered with land drainage systems (tile drainage), implemented under the responsibility of the Egyptian Public Authority for Drainage Projects (EPADP). Groundwater tables on drained land are found at depths of 1-1.6 m below the surface.

A large portion of the landscape is probably subject to seawater intrusion. Saline water entering from the Mediterranean Sea underlies a layer of fresh water. Towards the north, this layer of fresh water becomes thinner and more saline. Part of the area is identified as reclaimed lake bottoms. These sublandscapes are less fertile lands, very flat, at an elevation of around sea level. Groundwater tables are usually shallow. Farmers often add topsoil to improve the chemical and physical properties of the soil. A larger stretch of this landscape is found north of Damanhour.

Table 2 Overview of landscapes in the study area

Mahmoudia irrigation canal

The Mahmoudia canal runs for a distance of 77 km from the Rosetta branch of the Nile down to the Mediterranean Sea at Alexandria and serves a total command area of about 305 000 feddan through 70 branch (distributor) canals. The Mahmoudia canal receives water from three different sources. The main source is the El-Atf pumping station on the Nile, which lifts an average of 6 million m³/day (about 80 percent of the total annual supply to the canal). The canal is also fed by two subsidiary sources, namely Edku pumping station that lifts an average of about 0.6 million m³/day from the Edku drain, which passes below the Mahmoudia canal at km 8 850. Excess flow amounts to an average of 2.3 million m³/day from El-Khandak El-Sharki canal at km 15.270. In addition, the Mahmoudia canal is navigable and is the main source of the municipal and industrial water supply to Alexandria and many other towns and villages. The maximum water demand from the Mahmoudia canal is estimated at 15.56 million m³/day, which occurs during June.

Figure 4 Mahmoudia command area

Drainage network

The drainage system in the Mahmoudia canal command area consists of open and tile drains. Drainage water is collected from the fields by subsurface drains, flows into secondary open drains and reaches the Mediterranean Sea through open drains. Drainage water is pumped either back into the irrigation canals for reuse or into the Mediterranean Sea.

On the right bank of the Mahmoudia canal, secondary drains discharge into the Moheet Edku, which forms the northeast boundary of the command area bordering Lake Edku. The Moheet Edku drain in turn discharges into the Edku drain and then via Edku pumping station into Abu Qeir bay. Drainage water from the upper part of Moheet Edku catchment is lifted by Beseeq and Halq El-Gamal pump stations.

On the left bank of the Mahmoudia canal, secondary drains discharge into either Umum drain or, to the south, the Shereshra drain, which together form the southwest boundary of the command area. The Shereshra drain is pumped into the Umum drain while water from the upper reaches of the Umum drain is lifted by the Abu Hummus pumping station. The Umum drain discharges into the sea through the El-Max pumping station west of Alexandria.

Upstream landscapes

The Meet Yazeed command area, located south of the MCA, delivers surface drainage water, salts, and groundwater to the MCA landscapes. The amount of reusable surface drainage water is especially important for the agricultural production in the command area and other landscapes downstream.

The neighbouring landscape to the west and southwest include old new land, currently developed as the Nuberia irrigation command. Most of the newly developed land was based on rather sandy, calcareous soils. The divide is formed by the Umum and Shereshera drainage system. Initially the irrigation of these old-new lands created problems of upward seepage, waterlogging and salinity in the old land. The problem was solved through the construction of interceptor drains and the installation of subsurface drainage.

Downstream landscapes

The Rosetta Nile branch (east of Alexandria) discharges into the Mediterranean Sea through four coastal lakes: Manzala, Burullus, Maryut and Edku. These lakes may be considered the transitional sinks for most of Egypt's anthropogenic waste. The budget of the pollutant cocktail in these lakes is exposed to significant alteration in quantity and quality before reaching the sea.

Lake Edku is a coastal lagoon covering an area of about 126 km². This coastal lagoon is considered an "agricultural drain" and covers an area of 126 km² with a mean depth of one metre. About 389 million m³ of agricultural wastewater is discharged annually into Abu Qir through its connection with El Boghaz bay (Awad and Youssouf, 2002). It was estimated that about 730 million m³ of wastewater discharges annually through the El Tabia pumping station in the southeastern part of the adjacent bay. Waste is mainly industrial, with some agricultural and domestic contributions.

Lake Maryut is another coastal lake in the study area. It has lost 48 percent of its surface to reclamation for agriculture, from 32 160 feddan in 1950 (1955: 31 370 feddan, 1973: 16 280 feddan) to 16 240 feddan in 1981. The lake's water sources are the Umum drain, 74 m³/s (agriculture) and Qalaa drain (mixed wastewater 0.7 million m³/day). The lake has no sea connection, which may cause large areas of the lake to exhibit anoxic conditions, producing badly smelling hydrogen-sulphuric gases making the area unsuitable for residential settlement. Because of the lack of exchange with the open sea, the lake acts as an evaporation pan where organic and chemical pollutants concentrate.

In a complete analysis, the influence of water resources management on the Mediterranean should be incorporated in the study. Deterioration of the marine environment in some areas, especially off the Alexandria metropolis, is a result of the direct discharge of industrial pollutants and untreated domestic wastes into the coastal waters. The Mediterranean coastal waters off Alexandria are considered as one of the region's "hot spots".

Land and water control systems

Irrigation system

The traditional irrigation system in the old lands of the Nile valley is a combined gravity and water lifting system. The main canal system (first level) takes its water from El-Atf pumping station, located upstream of the Nile dams. Water is distributed along branches (second level) where the flow is noncontinuous. At the third level, low-level tertiary canals called mesqa receive water according to a rotation schedule and provide water to the fields through quaternary channels known as marwa. The area served by a mesqa is in the range of 50 to 200 feddan while the marwa typically serves around 10 to 20 feddan. Under the new improved system, the level of tertiary canals (mesqa) is raised to above ground level or set under pressure to allow for improved distribution along the canal. The rotational systems along branch canals are being replaced by continuous flow.

The Mahmoudia canal essentially operates under a system of upstream control, as it is part of the main system fed from Aswan as a feeder canal. It receives a seasonally varied supply of water, which is centrally determined by the Irrigation Sector in accordance with the expected cropping pattern. Control of discharges in the feeder canals occurs at the main head regulators and at cross regulators located at the boundaries between irrigation directorates.

Branch canal head regulators are generally equipped with lifting gates. Regulation of flow is achieved by adjusting gates to maintain target downstream water levels. These target levels vary seasonally and serve as an indirect and rather approximate means of controlling discharges; in practice, they are largely determined based on accumulated experience. Downstream branch canal offtakes permit only limited hydraulic regulation. Improved branch canals have automatic downstream controlled head regulators and are operated under continuous flow.

The particular feature of the Egyptian traditional irrigation system was that mesqas deliver water about 0.5-1.0 m below ground level. This requires farmers to lift the water onto their land. In the past, lifting was mainly done by animal-driven water wheels (sakias), which were licensed by the irrigation districts. The sakia was a fixed installation whose sump was connected to the canal or mesqa by an intake pipe of specified diameter. The farmers' capacity to abstract water from the delivery system was thus restricted in terms of both number and location of lifting points and of discharge. In particular, the need to share the use of the sakia with several other farmers in the same sakia "ring", and the limited discharge of the sakia, combined with the restrictions of the rotation system, meant that farmers were considerably constrained in terms of when and for how long they could irrigate. Further restrictions were applied at the mesqa offtake, which takes the form of a pipe whose diameter was originally related to the area served because of a defined head loss at the design discharge. In effect, the particular characteristics of the sakia introduced a degree of self-compensation into the operation of the system, which helped to ensure equitable distribution.

Over the last 20 or 30 years, mobile diesel pumps, owned by individual farmers, have progressively replaced the collectively owned sakias. Today, farmers may take turns to irrigate using different pumps at a particular lifting point. Farmers with fragmented holdings may use a single pump that is moved between their different plots. The widespread introduction of diesel-driven pumps has largely removed the various constraints imposed by the sakia-based system. The larger discharge provided by these pumps means that farmers complete their irrigation in less time. In addition, many farmers, whose fields are adjacent to canals or mesqa, have established additional lifting points. Even where lifting takes place at former sakia sites, often the pump suction is placed directly into the canal or mesqa rather than in the old sakia sump because the sakia inlet pipe would not be large enough to supply the pump discharge. In some cases, farmers have replaced the original mesqa offtakes with pipes of larger diameter. Many of these changes, particularly direct irrigation from canals and enlargement of mesqa offtakes are, strictly speaking, illegal. One result of the removal of the time constraint is that farmers can concentrate their irrigation mainly during the daytime, taking advantage of the considerable natural storage capacity of the canals and mesqa, which has been enhanced by the progressive widening of cross-sections resulting from bank degradation and over-excavation during maintenance. This informal night storage has become a significant feature of the existing system.

Drainage system

The principal design of the horizontal drainage system in the study area comprises a network of laterals and collector drains. The laterals are corrugated 72 mm wide PVC pipes, clay pipes (diameter 50 mm) then concrete 50 cm pipes were used prior to the use of PVC. The laterals, on average 200 m long with an average spacing of 50 m, discharge into collectors. The collectors are plain concrete pipes with inside diameters of 150 mm to 400 mm set at varying slopes that evacuate the drainage water into the main and branch open drains. The average length of the collectors is 1 500 m with a spacing of 400 m (Van Achthoven et al., 2003).

Water reuse system

Drainage water reuse in the Nile Delta increases the amount of water available for use, particularly for irrigation. Reuse is centrally organized by the Ministry of Water Resources and Irrigation (MWRI) and involves the pumping of water from main drains into main canals taking advantage of situations where the two types of channel cross each other at a few strategic points in the delta region. Of the strategic water budget of 55.5 billion m³ reaching Egypt via the River Nile, most is used for irrigation along the Nile valley and the Nile Delta, leaving approximately 14 billion m³ flowing out to the sea. This so-called "official" reuse rose from 2.6 billion m³ in 1988/89 to 5 billion m³ in 1998/99. The present aim of the Government of Egypt is to reuse up to 8 billion m³.

In addition to the "official" reuse of drain water, there is significant "unofficial" irrigation carried out by individual farmers throughout the region. The estimated "unofficial" drain water reuse is between 2.8 and 4 billion m³. There is no control of the amount of unofficial reuse by farmers. The concern is that this could increase significantly and threaten flows into the main drains used for official reuse. The danger is that the amount of water available for reuse, and needed for major projects such as the Salaam Canal Land Development Project, could be reduced.

The drainage canal and water reuse systems of the Mahmoudia are a complex of pumping and mixing stations that operate concurrently. The main mixing station relevant to the study area is the Edku pumping station. There is no hydraulic connection, as the canal water passes underneath the drain by means of a large siphon structure and the drainage water is lifted into the canal by means of the Edku pumping station. This diverts a certain amount of the drain flow to provide a fixed proportion into the canal, so that there is sufficient dilution of salinity levels for it to remain within guidelines.

Principal functions of landscape and water control systems

The principal functions of the landscapes include irrigation drainage and water reuse systems and are listed in the following sections. Functions are grouped into production, carrying, processing, regulation and significance functions. Where appropriate a specific landscape type is mentioned.

Landscape functions

Production functions

• Production of a rice-based irrigated crop mix; food supplies, farm income generation and employment (alluvial and lake bottom).

• Inland fisheries from lakes and canals for food supply and income generation.

• Commercial lake-bottom land; income generated by aquaculture (fish ponds).

• Raising of domestic geese and ducks.

• Production of reed grass for small enterprises (coastal lakes and wetlands).

• Commercial production of fruit and vegetables using un/treated drainage water (on sand culture of lake bottom land).

Carrying and supply functions

• Space for rural and urban settlements accommodating domestic and non-domestic housing.

• Space for industrial enterprises (alluvial land).

• Capture, conveyance and distribution of Nile water to and within the command area according to needs.

• Space for weed growth and sedimentation of gravel and sands (adverse functions).

Processing, regulating and control functions

• Maintenance of the fresh/saline groundwater balance in coastal areas that are subject to sea water intrusion and lateral seepage (lakes and rice belt).

• Capture, conveyance and disposal of sewage and solid wastes mainly within settlements.

• Domestic water supply for washing.

• Flood control (coastal lakes).

Significance functions

• Disposal of drainage water (coastal lakes).

• Disposal of industrial and urban liquid waste (coastal lakes).

• Breeding ground of migrating birds (coastal lakes and wetlands).

Irrigation system

Production functions

• Inland fisheries from canals for food supply (irrigation canals).

• Raising of domestic geese and ducks (open drains and canals).

Carrying and supply functions

• Capture, conveyance and distribution of Nile water to and within the command area according to needs (main and branch canal and mesqa level).

• Capture, conveyance and disposal of sewage and solid waste mainly within settlements (open canals and drains).

• Domestic drinking water supply for Alexandria.

• Irrigation water supply of downstream users.

• Space for weed growth and sedimentation of gravel and sands (adverse functions).

• Navigation.

Processing, regulating and control function

• Downstream-based control of water levels (control gates).

• Domestic water supply for washing (open canals).

• Facilitation of official water reuse system (open main canals, drains and pumping stations).

• Facilitation of unofficial water reuse (collector drains).

Significance functions

• Temperature regulation of domestic livestock (irrigation canals and open drains).

• Breeding of bilharzia vector snail (open canals, mesqas).

Drainage system

Production functions

• Aeration of soils (land drainage system).

• Raising of domestic geese and ducks (open drains and canals).

• Removal of nutrients (adverse function).

Carrying and supply functions

• Capture, conveyance and disposal of drainage from land to sea (from piped to collector and main drains).

• Space for weed growth and sedimentation of gravel and sands (adverse functions).

• Facilitation of inland fish migration (main canals).

Processing, regulating and control functions

• Capture, conveyance and disposal of sewage and solid wastes mainly within settlements (open drains).

• Facilitation of official water reuse system (open drains).

• Facilitation of unofficial water reuse (collector drains).

• Water table control (piped land drainage systems).

• Soil salinity control (land drainage systems).

• Natural cleansing of water (open drains).

Significance functions

• Breeding of bilharzia vector snail (open drains).

Water reuse system

Production functions

• To overcome crop water stress and failure due to insufficient water deliveries (unofficial reuse).

Carrying and supply functions

• Increasing the availability of suitable irrigation water supplies.

• Capture, pumping and conveyance of drainage water from main drains into main canals at specified water quality targets.

• Capture, pumping and conveyance of drainage water from collector drains onto field to overcome water supply shortages.

Processing, regulating and control functions

• Mixing of drainage and Nile water to acceptable quality targets.

Principal stakeholders and values

Farmers

The total number of people benefiting from the Mahmoudia command area is estimated at nearly 139 000 farm households, equivalent to a total of 1.4 million beneficiaries^[6]. The majority of the farmers in the command area cultivate five feddan or less. It was reported that 70-80 percent of the farmers in the Mahmoudia area are small-scale landholders having about one feddan or less. According to the 1999, Agricultural Census the number of villages in the three districts is 37, 19 and 40 respectively^[7]. The total number of agricultural holdings in Beheira governorate amounts to about 339 thousand and their area estimated at 1.4 million feddan. The total number of holdings in the three districts under the study was estimated at 82 600 in the 1999-2000 agricultural census, while their area was estimated at 251 100 feddan.

Table 3 Number and size of agricultural holdings in districts of study area

Farmers in the study area produce winter and summer crops. Berseem (long clover), beseem (short clover); wheat and broad beans are the main crops cultivated in winter besides the winter vegetables and some perennial crops. Cotton, rice and maize are the main crops in the summer season as well as watermelon seeds, summer vegetables and some perennial crops. Rice is the most water demanding crop of all the summer crops.

The overall estimated net revenues produced by the irrigated sector in the study area amount to approximately LE 520 000 000, which is equivalent to US$84 000 000. Each hectare of land returns LE 4 852 per year (US$783) representing a cropping intensity of 178 percent. The crops generate a net return of LE 27, which is about US$0.04 per cubic metre of water consumed (Table 4).

Farmers practice traditional irrigation and receive information and advice from agricultural extension agents in the area. Farmers' seem to have a limited understanding of cultivation techniques, e.g. long furrows, dry planting of berseem, using machines to plant rice, etc. Investigations show that farmers are aware of laser levelling and its benefits and are willing to apply it. However, its practice is not common because of its high cost and limiting small size of the fields.

Table 4 Estimated economic value of agricultural production

Source: MALR, Agricultural Affairs Sector, Agricultural Statistics, crop data 2003

Fishermen

Downstream of the project area there are an estimated 12 000 fishermen who are directly involved in inland fisheries. There are also intermediary traders and the local population benefits from an affordable source of protein. It is estimated that fish consumption is 10 kg per person per year.

According to FAO (2003) Egypt has about 8 716 km² of inland waters, including rivers, lakes, reservoirs and brackish water lagoons. These inland water resources represent various fisheries ecosystems such as fresh and brackish water. The former include the River Nile, irrigation canals, the Aswan High Dam and Lake Nasser and some western desert water bodies. During the last ten years, the recorded catch from the Nile River basin in Egypt increased each year from 40 000 tonnes in 1992 to 110 000 tonnes in 2001. The catch comprised mainly tilapia and catfish from the artisan-fishing sector.

Both commercial and sport fishing take place on these waters. Some inland waters are regularly restocked with both marine and freshwater fish fry. The inland fishing fleet is made up of over 38 500 small wooden boats (4-6 m in length) catching about 295 500 tonnes, or 69 percent of Egyptian landings. Most of those who fish are unregistered.

The brackish water lagoons are in the northern delta (Lakes Maryut, Edku, Burollus, Manzala, and the almost dry Wadi Al Raiyan). Delta lakes are eutrophic lagoons where trammel nets and various traditional fishing methods are used including by hand and collecting fish under vegetation using a coneshaped net. Manzala is the largest lake, followed by Borollus, Edku and Maryut. The catch is mainly tilapia, catfish and mullet. Pollution, reclamation, fragmentation, over-fishing and illegal harvesting of fish are the major environmental issues threatening the fragile ecosystem of the northern lagoons.

Fish consumption in Egypt is characterized by a longstanding traditional preference for fresh fish. National fish production in Egypt was calculated in 1998 to be 545 593 tonnes. Fisheries represented 406 204 tonnes, and fish culture 139 389 tonnes. The northern lakes combined represent some 30 percent of national fish production. Lakes Edku and Maryut represent respectively 1.9 percent and 0.8 percent of this total. The total fish catch in Lake Maryut fluctuates strongly from year to year with a high of 7 700 tonnes in 1985 and a low of 1 900 tonnes in 1995. The latest available data, from 1998, show that fisheries have partially been restored to 4 521 tonnes. The serious environmental degradation of the lake is caused by industrial and urban pollution. Fluctuations are also caused by the closed nature of the lake that prevents incidental flushing.

Government institutions (MWRI, MALR and MLD)

A number of government institutions are engaged in the development of the irrigation and drainage sector and the Ministry of Water Resources and Irrigation (MWRI) plays a leading role. Other ministries include the Ministry of Agriculture and Land Reclamation and its Fishery Authority and the Ministry of Local Development.

The MWRI is organized into two main technical departments the Irrigation Department (ID) and the Mechanical and Technical Department (MED). The Irrigation Department is subdivided into five sectors including the Irrigation Improvement Sector, the Irrigation Sector, the Horizontal Improvement Sector, the Dams and Grand Barrage Sector and the Groundwater Sector. All the above-mentioned departments are organized into administrative units at the national, provincial and district level (see Figure 4).

A number of public authorities are associated with the MWRI and include the High Aswan Dam Authority, the Coastal Protection Authority and the Egyptian Public Authority for Drainage Projects (EPADP). EPADP is in charge of the design, planning and implementation of rehabilitation and maintenance of the national drainage systems of open and subsurface drains. EPADP employs a permanent staff of about 4 000 at Cairo headquarters and directorates. About 3 000 casual labourers work in the field of drainage maintenance. At field level EPADP has 30 field directorates and 168 drainage centres, which are charged with the cleaning and flushing of subsurface drains and the cleaning and desilting of open main drains (van Achthoven et al., 2004).

Water user organizations

The Irrigation Improvement Project established a total of 1 309 water user associations (WUAs) within the Mahmoudia command area. In addition, ten branch canal water user associations (BC-WUAs) are being formed.

Urban and rural communities

With a population of some five million Alexandria is the largest urban community near the project area. A large portion of the city's water supply originates in the Mahmoudia canal. Urban centres and numerous rural communities are scattered over the entire project area. The smallest unit, a hamlet, is composed of a few homesteads.

Industries

Principal industries along the Mahmoudia canal include brick manufacturers, textile and leather industries. Although small in quantity, industrial water usage is relevant as it is loaded with pollutants and waste as it travels through industrial processes. Brick manufacturers have also left behind large quarries that can serve as entry points for significant groundwater pollution.

SUMMARY
Table 5 List of functions/values related to the Mahmoudia command area (on-site) landscapes

Table 6 List of functions/values related to neighbouring (off-site) landscapes

ANALYSIS

Water resource planning and management

Problem areas

Fragmentation of water management responsibility

Egypt's water resource management and planning is carried out at the central administrative level with little responsibility being given to lower administrative units. Decentralization has been identified by the Government as a possible strategy for more effective water resource management based on water user participation. Institutionally, one irrigation directorate (Western Delta Irrigation Directorate) and three irrigation districts cover the Mahmoudia command area. The same area is served by two drainage directorates: North Beheira Drainage Directorate and South Beheira Drainage Directorate.

Fragmentation of water management responsibility is recognized as a serious constraint to effective planning, implementation and management of projects. Within the scope of the Water Board Project a mapping exercise identified numerous governmental and non-governmental institutions as being responsible for irrigation management (refer to Figure 5). The project identified a similar map of institutional responsibilities for the drainage, and groundwater sector. It is assumed that altogether this level of fragmentation has created a management landscape based on the line administration and hierarchical structures of the mother organization. Nonetheless, water user participation in management is increasingly complex and hard to overlook.

Increasing competition over scarce water resources at the local level

Egypt's annual population growth rate is estimated to be about two percent. In rural areas, there is increasing competition over freshwater supplies for domestic purposes and livestock. Although small in quantity, the need for good quality water is critical, especially where irrigation water has gone through a number of use cycles. This is particularly the case at the end of branch canals and requires the careful attention of local water resource managers.

Scattered responsibility for water quality control at the local level

It is increasingly recognized that water quality is as much an issue of water resource management as water scarcity. Although under the overall responsibility of the MWRI, the management of water quality and quantity is still organized under separate units within the ministry, which may not be a problem. A problem may exist in that the central organization of responsibilities is not reflected in the organization at the lower administrative levels, where district officers still play an important role in the day-to-day management of the hydraulic infrastructure including water quality control (Figure 6).

A study conducted by the USAID-funded Agricultural Policy Reform Programme (APRP) on the water quality of the Nile River concluded that 25 agencies under seven ministries are involved in water quality monitoring programmes. The study identified considerable gaps that need to be filled in order to provide a full picture of water quality in Egypt (APRO, 2002).

Opportunities, effects and possible impact

Irrigation integration versus integrated irrigation

It is recognized that the physical interventions of IIIMP need to be viewed and handled from an integrated perspective. The question is: Which approach should be adopted "Integrated Irrigation" or "Irrigation Integration"? The first approach would make irrigation improvement the core of integrated water resources management (IWRM) and other elements satellites in its orbit. The second, irrigation integration would aim to place irrigation in the overall IWRM framework.

Figure 5 Map of institutions involved in irrigation development and management grouped by administrative level

Source: MWRI 2004: Water boards project

Figure 6 Organization of the General Directorate Irrigation Beheira

Source: MWRI, Water Board Project, 2004

With its Water Board Project, the MWRI has embarked upon the concept of integrated Water Resource Management at the irrigation district level. Within this concept, stakeholders from multiple sectors are encouraged to engage in the management of water resources at their respective levels. There is a risk that these initiatives tackle the integration problem from the technical or physical viewpoint rather than considering its socio-economic and environmental dimensions. Integration may be confused with coordination of irrigation, drainage, groundwater and water quality activities. Typically, one of these technical activities may be considered central to the process of integration and should take the lead or dominate. This, however, is against the principle of integration, which means identification of subsystems and their interdependence. Therefore, natural resources and equilibrium of the societal subsystems should be analysed and addressed together. Thus, proceeding with regular technical activities and physical improvements by sector and discipline, or handling any non-technical dimension as a boundary, is a major drawback.

Increasing water reuse efficiency

The opportunity for IIIMP to overcome the restricted scope for water savings is to capture more of the drainage outflow and increase the efficiency of water reuse^[8]. As the level of official and unofficial reuse is already very high, it is important to take into account the potential for increased reuse that includes all levels, main, branch and mesqa as well as field and crop levels. Official water reuse is already practised at the main canal level with considerable success. Intermediate reuse is being discussed as an option for drainage reuse at the branch canal level. Unofficial drainage reuse is widespread at the mesqa level. At the field level, conjunctive use of groundwater may become an option, and reuse may even be considered at the crop level if roots reach deep enough to take up water from the capillary fringe.

From the viewpoint of water resource management there is a limit to reuse, determined by two factors: first, by the hydraulic conveyance of the drain, which must remain effective; and second, by the water salinity thresholds for irrigation water. Thresholds need to be maintained at acceptable levels to prevent crops being affected by increased salinity in the irrigation water. Other constraints to reuse are prevention of saltwater intrusion and the maintenance of freshwater ecosystems.

Improved water reuse management may affect the overall level of water depletions, at given primary diversions from the source. Thereby enhancing the overall available water for processing purposes; although an adverse effect on water quality can be expected, as reuse is naturally associated with an accumulation of solutes.

Conclusions

The opportunities for reducing the complexity of water control systems are limited to a system evaluation review of design criteria, standardization and system modernization and simplification. The opportunities for reducing the fragmentation of management responsibilities are provided by the encouraging results of the Water Board Project. Water user participation at various levels implies that responsibilities are transferred to water user institutions at lower administrative levels and that their executive committees are fully accountable to their members.

If IIIMP wants to achieve its overall goal of increasing the efficiency and sustainability of water use, more attention should be given to demand management of the agricultural water use system. Only if supply management interventions are accompanied by demand management measures will real water savings be accomplished. Otherwise, savings at one level of the system may be outbalanced by efficiency losses elsewhere. In order to make irrigation improvement investments an effective and feasible concept, water savings sought should be real savings rather than paper savings.

If current unofficial drainage water reuse at the mesqa level is eliminated, there is a risk thamesqa that overall reuse efficiency may decline unless the reuse gap is compensated at levels other than the mesqa.

LAND AND WATER MANAGEMENT

Problem areas

Productivity of water use (water use efficiency)

High water productivity and livelihood security are intricately connected because farming in Egypt depends entirely on irrigation. As there are few new major water resources to develop, expansion of land is sought through increased use of groundwater and from intensified reuse of drainage water; both have an upper limit. At the same time, urbanization progresses in large cities and small rural centres. This translates into additional demand for domestic and industrial water but, more important, into a larger pollution load and an increased need to address water quality.

Crop productivity levels in Egypt are relatively high when compared to world standards. This is thanks to farmers' skills, the highly productive soils, and good climatic conditions, perennial source of irrigation water and relatively favourable irrigation and drainage services. The crop productivity of rice, corn and wheat is 9.4, 8.1 and 6.5 tonnes per hectare respectively^[9]. On the other hand, the productivity of water in agriculture is relatively good with a very high percentage of inflowing water being depleted and most of this depleted water being put to productive use (Molden and Sakthivadivel, 1999).

The income level of many farmers is still low and many in the population can no longer subsist on agricultural production. The main cause is the very small land area per household and, therefore, low farm revenues in terms of food and cash. Many households need additional off-farm employment to generate a minimum income. A farm with an area of one hectare, cultivating the average cropping pattern at average variable cost and rent would have a net revenue of about LE 5 000 or US$800 per year^[10].

Various interconnecting paths lead to increased agricultural productivity. Among those having agronomic as well as water and land engineering characteristics are: improving irrigation system efficiency, crop development, reducing land degradation, introducing better irrigation techniques and improving irrigation management and field practices, as well as integrating drainage water reuse into a broader vision of irrigation water management. The question here is to what extent can each of these paths improve agricultural productivity per unit of water?

Opportunities, effects and possible impact

Improving irrigation system efficiency

In recent years, the term "irrigation efficiency" has evolved from its classical definition, accounting for farm level and delivery system efficiencies, to a broader definition that includes the expanded physical boundary considered when evaluating water use such as the terms "effective efficiency" or "global efficiency". Egypt is a unique case, where these new terms can be perfectly applied. The various efficiencies aggregate up to the basin-wide level where a real difference can be made. In Egypt, the classical overall irrigation efficiency is only 40-50 percent. In the Egyptian irrigation sector as a whole, this is close to 80 percent because of the reuse of drainage water. Much of the remaining 20 percent is used beneficially in other sectors, including the environmental benefits of discharges to wetlands and the Mediterranean (Molden and de Fraiture, 2000).

In their Interim Report, the consultant for the IIIMP preparation study, SOGREAH, estimates the classical overall irrigation efficiency at 51 percent (Table 7) and recommends a set of preliminary actions that includes strengthening the on-farm development programme as this will lead to a significant enhancement of irrigation efficiency. It is not known if there is potential scope to save water through improvement of water management to achieve a higher level of classical efficiency in the Egyptian irrigation sector, where the basin wide efficiency exceeds 80 percent. It can be argued that such efforts in achieving a higher level of classical irrigation efficiency would largely reduce the volume of surface runoff and deep percolation used by downstream farmers in the basin, while generating little or no gain for water available for irrigation.

The IIIMP is perceived to be the natural follow-up to several projects in Egypt, notably the Irrigation Improvement Project (IIP), National Drainage Programme (NDP) and National Pumping Station Rehabilitation Programme (NPSRP). IIIMP will strive to combine these various sectoral projects into one integrated programme. It is expected that the technical concepts developed under IIP to increase irrigation efficiency and agricultural productivity will be considered and reviewed in conjunction with similar technical improvements for drainage, mechanical-electrical and groundwater development.

Table 7 Irrigation efficiencies assumed by IIIMP

It should be borne in mind that, in general, any increase in an irrigation system's classical efficiency reduces the scope for drainage reuse by a corresponding amount. More importantly, the priority use for any savings due to localized increases in water use efficiency will be to improve canal supplies to watershort tail areas that rely on direct irrigation from drains for all or part of their supplies. This substitution of water previously lost to the drains, for water previously taken from the drains, will be neutral in terms of overall water savings. Thus, it is the balance between water availability/sufficiency and efficiency benefits that should be taken into careful consideration.

Cropping system intensification/crop development

Economic analyses show that there are differences in the total economic returns made on different crops grown in Egypt. To illustrate this, one may compare the net revenue per hectare of irrigated rice and maize in the Mahmoudia canal area with the net revenue per cubic metre of these two summer crops, bearing in mind that rice normally requires a water application of about 1 900 mm and maize about 1 000 mm (Table 8). The adage "more crop per drop" is easier said than done and should be replaced by the adage "more value per drop".

Table 8 Revenues per hectare and unit of water

* Provisional figures

Reducing agricultural water consumption is effective in increasing water productivity. This may include the gradual replacement of sugar cane with sugar beet; the reduction of rice-cultivated areas; replacement of currently used rice varieties with shorter-life varieties that have higher productivity and require less water; development of new crop varieties using genetic engineering resulting in higher productivity and lower water consumption, and the design of indicative cropping patterns.

The value of certain agricultural products cannot only be expressed in economic terms. For example, wheat production has a strong social dimension for small-scale farmers on Egypt's old lands as it provides them with food security. Changes in cropping pattern are therefore sometimes difficult to realize. Nonetheless, there is still potential scope to improve the productivity of water.

Improving on-farm water management

Research and demonstration activities in Egypt, including those of the IIP, have identified a wide array of interventions related to on-farm water management and best agricultural practices that can positively affect economic returns from the crops grown. Extensively applied but relatively minor incremental investments in on-farm water management improvements could result in substantial benefits. Items include: laser land levelling; modernizing quaternary canals (marwas) through installation of low pressure pipelines and flexible hoses; training of farmers in on-farm water management techniques; and supporting the strengthening of irrigation and agricultural advisory and extension services.

On-farm water management conforms to a key component under the ongoing Irrigation Improvement Project and it is complementary to mesqa improvement activities. The key objective of this component is to introduce and demonstrate improved on-farm irrigation/cultivation techniques to farmers that would reduce water losses and increase crop yields and returns. The programme included laser land levelling and using long furrows and was implemented by the Ministry of Agriculture and Land Reclamation (MALR). The activities focussed on demonstration plots at the head, middle and tail ends of improved mesqas. According to the consultant's Interim Report for the IIIMP preparatory study, SOGREAH, the results so far are encouraging. Yields and water use efficiency have increased by about 10 percent and 10 to 30 percent respectively.

It seems that knowledge of cultivation techniques is limited, despite the successful verification of the feasibility of the introduced techniques. Thus, the effort under IIP remains limited in scope, particularly from the perspective of technology transfer. According to the current IIP monitoring and evaluation programme (2002-2005), the majority of surveyed farmers in the Mahmoudia command area do not know about irrigation using relatively long furrows and the percentage of farmers using machines to plant rice is less than one percent. Most farmers in the command area still use traditional agriculture techniques in spite of the implementation of the on-farm IIP demonstration programme.

The above argument raises a question as to whether the on-farm irrigation management component of the IIP field demonstration programme has been effective or its impact limited. The answer to such a question is crucial and needs further investigation since the new project (IIIMP) seems to be considering the same measure (field demonstration programme). The report of the project's pre-appraisal mission states that "the on-farm water management component will continue and be implemented by MALR through its extension and research centre headquarters and governorates, in close cooperation with the MWRI".

The following measures provided by the consultant of the IIIMP preparatory study, SOGREAH, in the Interim Report could be of value in creating a more effective on-farm management programme:

• more mesqas be included in any future programme and selected mesqas be more dispersed within the project area to allow for more effective dissemination to farmers of neighbouring mesqas;

• data collection to be standardized across crops and mesqas, and differences between head, middle and tail sites should be made more explicit;

• returns on improved techniques to be computed on all demonstration plots;

• lead farmers to be selected with the help of the WUA to assist in the programme and ensure continuity after project completion; and

• strong linkages with WUAs to be established to ensure active participation of WUAs in the technology transfer effort under the IIIMP.

Measures such as the modernization of quaternary canals (marwas), through the installation of low-pressure pipelines and flexible hoses, should be carefully studied for their efficiency/economic benefits against investment. Greater attention should be given to its social dimension, which is acceptance by farmers.

Agricultural marketing

To attain higher net returns and income from the same land and water resources, farmers would need to diversify their cropping patterns into high value crops (vegetables and fruit) for the export market. Farmers of the project areas are traditional, subsistent by nature, and thus risk avoidant. At present, market system information in the project area is weak. Farmers have little experience in growing high value crops in a way that is responsive to market requirements. This can be mitigated through introduction of a marketing component into the IIIMP. The inclusion of this sub-component is regarded as crucial to allowing farmers to benefit from improved water delivery and thus shift to high value crops. Another key requisite is that this component should be implemented through MALR for effectiveness; the appropriate coordination between MWRI and MALR must be developed at central and local levels.

Conclusions

If IIIMP wants to achieve its overall goal of increasing the efficiency and sustainability of water use, much more attention should be given to demand management of the agricultural water use system. Only if supply management interventions are accompanied by demand management measures will real water savings be accomplished. Otherwise, savings at one level of the system can be outbalanced by efficiency losses elsewhere. In order to make irrigation improvement investments an effective and feasible concept water savings sought should be real savings, rather than paper savings, and the alternative allocation of the water should be made explicit.