Surface Irrigation for Cotton and Wheat at Ras El Ain Assessment and Issues for Improvement

Master Theses in

“Land and Water Resources Management: Irrigated Agriculture”

Surface irrigation for cotton and wheat at Ras El Ain: assessment and issues

for improvement

Hanaa Darouich

(Syria)

Supervisors: Prof. Luis Santos Pereira Prof. Nicola Lamaddalena

Co-supervisor:

Prof. José Manuel Gonçalves

October, 2006

Acknowledgements First my deepest thanks for my supervisors, professors Luis Santos Pereira and Nicola Lamaddalena for the splendid guidance during this year, and to Professor Jose Manuel Gonçalves for the great help and useful advices during the year. No-words can express my gratitude for him and his family, and his colleagues Fatima and Kiril, for receiving me in Coimbra, that made me to deeply appreciate the time I spent there, and his interest in guiding the theses. Thanks go to the Ministry of the Agriculture of Syria for giving me the opportunity to study in Bari; particular thanks go to Dr. Majad Jamal. I also deeply thank the directors of the Cooperative Italian Project in Syria Mr. Biago De Terlizzi, Ms Chiara Morini and Professor Lamddalena for giving me the chance to study in this interesting project, and main thanks to the Project Director Mr Luigi Cavestro for the help and the facilities that he offered to me; thanks also go for Mr. Farid, head of the Syrian counterpart. I would like to thank the PMU staff of the Project, namely those going with me to the field and providing me the required information: Jak, Toudour, Dr. Saleh, Linda, Jan, Basel, Eissa, Mohamad, Gorag, Daniele, Jan Luka, Mais, Shamsa.Ruba, Mieada. Many thanks to the farmer Razuk and his brother to help me in Bab El Faraj. And especial thanks for Mr.Amin Arafat for every things, and for Prof. Lebdi Fethi for the useful advices. Thanks a lot to Silvia sharing with me the good and the bad things in Al Hassakeh. I would like also to thank the wonderful staff in Lisbon, that gave me help for the thesis: Paula, João Rolim, Vanda, Maria, Teodoro. Many thanks to the IAM administration staff, Mrs Olyimpia Antonelli, Mrs Mimosa Myrta, and Ms Maria Carla Martelli. My great gratefulness goes to Director, Mr. Cosimo Lacirignola, and Vice-Director, Mr. Maurizio Raeli for giving a unique pleasure and opportunity to study in this institute. Hanaa

I

II

Abstract: Ras Al Ein irrigation district is located in north of Syria, an area with serious problems of water scarcity and agriculture sustainability. The actual irrigation practices have a very low application efficiency and high labour consume. Farmers apply less appropriate irrigation schedules, creating water losses and crop water stress. To contribute to improve irrigated agriculture, field experimentation and modelling was carried out. Field evaluation of furrow irrigation with different conditions and irrigation scheduling modelling proved that surface irrigation has a high water saving potential and allows farmer income increasing. The application of a decision support system to build up and select improved solutions shows that, for cotton irrigation, furrow and border methods have the best results with the combinations slope - length of 0.5% - 50m and 0.8% - 100m, for flat area, or 150-200m, for sloped areas. For wheat irrigation, several options are feasible, being the level furrowed basin the best performing system for flat fields, being also the flat level basin an acceptable solution, and the graded borders, with 0.5 and 0.8% slope, for sloping fields. Key words: Ras Al Ein, irrigation schedule strategies, furrow irrigation, multicriteria analysis, decision support system (DSS) Résume Ras AI Ein arrosage district est localisé au Nord de la Syrie, une area avec des grands problèmes au niveau de la sècheresse et de la sustentabilité de l’agriculture. Les techniques actuelles d’arrosage ont une efficience très baisse et utilisent un grand nombre de heures de travail. Les agriculteurs utilisent des techniques d’arrosage qui conduisent à perte d’eau et au stress des cultures. Pour montrer l’importance de l’arrosage en agriculture, des essais sur terrain ont été mis en place. Les observations faites en différents conditions d’arrosage ainsi que les techniques d’irrigation testées, ont permit montrer que la irrigation de surface a un grand potentiel pour économiser de l’eau et pour la production agricole. L’application d’un system de l’aide à décision pour choisir les meilleurs solutions a montré que pour l’arrosage du coton avec sillon et bandage, les meilleurs résultats ont été obtenus avec la combinaison déclive – longueur de 0.5% - 50m et 0.8% - 100m sur les areas plaines et 150-200m, sur les areas de pendent accentuée. Pour l’arrosage du blé plusieurs options sont possibles. La meilleure solution pour les situations plaines est le bassin avec sillon et de bandage avec a déclive 0.5% - 0.8% pour les zones à pendent. Clé mot: Ras Al Ein, programmation des arrosages, irrigation par sillon, multiple critère analyse, system de l’aide à décision

Table of the contents

Acknowledgements ……………………………………………………………………I

Abstract ……………………………………… ……………………………………….II

Table of the contents ………………………………………………………………...III

List of figures ……………………………………………………………………….....V

List of tables …………………………………………………………………......VIII

1- Introduction……………………………………………………………………1

1.1. Statement of the problem…………………………… ……………….1

1.2. Objective……………………………………………………… ……….2

2- Ras-El-Ain Irrigation district……………………………………………….……...3

2.1. General Information………………………………………………...…3

2.2. Water resources……………………………………………………….4

2.3. Climatic and soil characteristics……………………………………..4

2.4. Agriculture and land use……………………………………………...6

2.5. Irrigation system ………………………………………………………7

3- Material and Methods …………………………………………………………....9

3.1. Experimental sites: Arnan and Bab Al Faraj………………………..9

3.2. Irrigation scheduling: the ISAREG model ……………….......... …12

3.2.1. Model description …………………………………………...12

3.2.2 Input data………………………………………………………14

3.2.2.1 Climatic and Soil data ………………………………..14

3.2.2.2 Crop data for wheat and cotton ……………………..15

3.3. Surface Irrigation assessments and modeling……………………..16

3.3.1 Field evaluation ……………………………………………….16

3.3.1.1 Infiltration tests………………………………………...16

III

3.3.1.2. Field evaluation of surface irrigation……………….18

3.3.2. Performance indicators ……………………………………...22

3.3.3. The SIRMOD model for surface irrigation simulation…….23

3.4. The Decision Support System (DSS) SADREG ……………………24

3.4.1. Model description …………………………………………….24

3.4.2. Performance Indicators, criteria and ranking with

SADREG…………………………………………………….31

4- Results and Discussion ………………………………………………………….33

4.1. Irrigation scheduling of wheat and cotton……………………. ……33

4.2. Field Irrigation Assessments…………………………………………37

4.2.1. Irrigation and hydraulic parameter …………………………37

4.2.2. Irrigation performance ……………………………………….39

4.3. Alternatives for improving surface irrigation. SADREG

application…………………………………………………………..47

4.3.1. Projects: build up improved irrigation scenarios…………..47

4.3.2. Alternatives, selection and ranking ………………………...51

4.3.3. Projects comparison and discussion……………………….53

4.3.4. Conclusive remarks………………………………………….60

5- Conclusions ……………………………………………………………………….61

6- References ………………………………………………………………………..65

IV

List of the Figures Figure 2.1. Location map of Ras Al Ein Irrigation District………………………...3 Figure 2.2. Average precipitation, reference evapotranspiration and minimum and maximum temperatures ………………………………………………………. .5 Figure 2.3. Average and minimum monthly humidity with wind speed …………5 Figure.3.1.Arnan experimental farm………………………………………………...9 Figure.3.2. The soil water content in 90 cm soil depth of Arnan experimental farm…………………………………………………………………………………….11 Figure.3.3. Simplified flow-chart of ISAREG model (Teixeira & Pereira,1992)………………………………………………………………………...12 Figure 3.4. Infiltration curves obtained with the double ring test at Arnan farm…………………………………………………………………………………....17 Figure 3.5. The inflow rate in 0/4 and 2/4 station (Bab Al Faraj)…….…………20 Figure 3.6. Typical furrow cross sections before irrigation……..……………….22 Figure 3.7. Conceptual structure of SADREG (Gonçalves et al., 2005a, b)…. 25 Figure 3.8. Modular components of SADREG (Gonçalves et al., 2005a, b) ….26 Figure 3.9. Land levelling module flowchart (Gonçalves et al., 2005a, b)……..27 Figure 3.10. Flowchart relative to the execution of the SIRMOD application procedure (Gonçalves et al., 2005a, b)……… ……………..…………………….27 Figure 3.11. SADREG data structure elements………..…………………………29 Figure 3.12. Design variables for the alternatives generation procedure……...29 Figure 3.13. Flowchart of the alternatives generator module (Gonçalves et al., 2005a, b)……………………………………………………………………………...30 Figure 3.14. Irrigation water-yield function (Gonçalves et al., 2005a, b)……....32 Figure 4.1.Simulated soil water dynamics for Alternative A and the current water saving schedule ……………………………………………………………………..34 Figure 4.2. Comparing ETa and the ETa/ETm ratio for the strategy A (a) and for the rainfed crop (b). …………………………………………………………………35

V

Figure 4.3. Comparing the total net water applications and relative yield reductions for the simulated strategies with the rainfed crop……………………………………………………………………..……………..35 Figure 4.4. The simulated soil water dynamics when MAD=100% Өp and variable irrigation depths are adopted (MaxY)………………………..…. …...…36 Figure 4.5. Time evolution of daily Eta, and Eta/ETm ratio during the cotton season for the case MaxY. …………….………………………………………...…36 Figure 4.6. Comparing consumptive water use and relative yield losses for cotton relative to several simulated strategies………………….…………………37 Figure 4.7. Infiltration curves observed at Arnan (2005-2006) and those from the SADREG database ………………………………………………………………….38 Figure 4.8. Infiltration curves observed at Bab Al Faraj (2005-2006) and those from the SADREG database………………………………………..………………39 Figure 4.9. Observed & simulated advance and recession times for treatments B1.1 (a) and B3.1 (b)………………………………………………………………...39 Figure 4.10. Observed and simulated advance and recession time for treatment B1.2 and B3.2………………………………………………………………………...40 Figure 4.11.The characteristics of the first irrigation for furrows 100 m long…..41 Figure 4.12.The performance of the first irrigation for furrows 100 m long…….42 Figure 4.13.the characteristics and the performance indicators for B3.2, C3, D3. …………………………………………………………………………………………42 Figure 4.14 Furrow E1, second irrigation: a) Observed advance and recession time, and required and infiltrated depths; b) simulated advance and recession time, and infiltrated depth considering advance and recession or only advance data……………………………………………………………………………………44 Figure 4.15. The advance and recession time, and required and infiltrated depths for the zigzag furrow………………………………………………………...44 Figure 4.16. Characteristics and performance for Bab Al Faraj irrigations…….45 Figure 4.17. Simulated characteristics and application efficiency for 100 m in Bab Al Faraj for 0.01 m/m and zero slope as a function of the inflow rate…….46 Figure 4.18 – SADREG window for field size, slopes (a) and infiltration curves (b)……………………………………………………………………………………...49 Figure 4.19 – SADREG window for input of used crop data on cotton and wheat……………………………………………………………………………….….50

VI

Figure 4.20 – SADREG window showing the used unitary cost and financial data…………………………………………………………………………………….50 Figure 4.21 - Tasks duration data; a) irrigation; b) land leveling maintenance… ……………..…………………………………………………………………………..51 Figure 4.22 – SADREG Window showing the tree to explore alternatives…….53 Figure 4.23 Utilities of cotton irrigation projects, in Arnan……………………….55 Figure 4.24 Utilities of cotton irrigation projects, in Bab Al Faraj……………….55 Figure.4.25 Utilities of wheat irrigation projects, in Arnan……………………….56 Figure 4.26 – Performance indicators for graded furrows and borders for cotton irrigation in Arnan…………………………………………………………………….57 Figure 4.27 - Performance indicators for graded furrows for cotton irrigation in Bab Al Faraj…………………………………………………………………………..58 Figure 4.28 – Performances of basin and borders for wheat irrigation, Arnan ………………………………………………………………………………………….59

VII

List of Tables Table 2.1 Cultivated area and the yields for wheat and cotton in Al Hassakeh……………………………………………………………………………….6 Table 2.2 Characteristics of irrigation systems applied for the cotton crop (source: Hassakeh research center, Ministry of Agriculture)……………………..7 Table.3.1. The cropped fields in the Arnan experimental farm, 2005/2006 …...10 Table.3.2. Physical soil properties of Arnan experimental farm………………...10 Table 3.3. Chemical soil properties of Arnan experimental farm…………….....10 Table.3.4. Physical soil properties of Arnan experimental farm as measured in the ICARDA Lab ……………………………………………………………………..11 Table.3.5. Chemical water properties (Arnan) ……………………………………11 Table 3.6, Meteorological data from Arnan station (completed with those for Ras El Ain), 2005-2006…………………………………………………………………...15 Table.3.7. Average crop parameters for wheat and cotton……………………...16 Table 3.8. Field treatments in Arnan and Bab Al Faraj…………………………..19 Table 3.9. Topographic characteristics of field E2, Arnan………………………19

Table 3.10. Design variables………………………………………………………..31

Table 4.1. Wheat irrigation scheduling simulations compared with the current wheat irrigation (assuming Ea=60%)……………………………………………....33 Table 4.2. Cotton irrigation schedule simulation for surface irrigation………….35 Table 4.3. Infiltration and hydraulic roughness parameters observed at Arnan and Bab Al Faraj……………………………………………………………………...38 Table 4.4.a). Characteristics and performance of the first irrigation for long furrows (100-150m)…………………………………………………………………..40 Table 4.4.b. Characteristics and performance of the first irrigation for short furrows, 50-36m………………………………………………………………………41 Table.4.5.Characteristics and performance of the second irrigation and traditional irrigation (zigzag basin) for the first irrigation…………………………43 Table 4.6.Characteristics and performance in Bab Al Faraj……………………..45

VIII

Table 4.7 - Project characteristics for improved systems………………………..47 Table 4.8 - Workspaces characteristics and applied projects…………………...48 Table 4.9 - Example of a project alternative layout……………………………….52 Table 4.10 - Criteria and weights for alternative selection……………………….53 Table 4.11 - Indicators applied for projects’ comparison………………………...54

IX

1

Charter 1 1- Introduction 1.1 Statement of the problem Water security is an important issue in several countries that are suffering from gradual but important reduction in water resources availability, particularly the ground water. Syria is one of these countries, where some regions are semi-arid and respective water resources became very scarce in a short period of time. Therefore research is required and several projects are now running in the country intending to study the reasons of this phenomenon and to find appropriate issues to control related problems. The Ras El Ain area is considered one of the most critical regions in Syria because it is heavily affected by water scarcity and the drawdown of the ground water table. This phenomenon is influenced by several factors related to climate aridity, heavy use of available resources and decrease of flow in most of the rivers due to the construction of dams in the upper parts of the basins inside Turkey. This situation led to set up a group of wells to fed the Al Kabour river in the Ras El Ain area, high ground water use for agriculture, irrigation in particular, Traditional irrigation methods are applied, which have relatively low performance, while population and farmers have not high awareness about the value of water and the need for water conservation and saving. Ras El Ain area is an ancient arable and fertile land where the farmers plant cotton and wheat since long time. The Syrian Government considers these crops as strategic ones, and supports the farmers with soft loans and various facilities in order to keep planting these crops. However, cotton requires quite large amounts of water, that was plenty available until the last five or seven years ago. Surface irrigation prevails there; Basin and border irrigation are adopted using earthen ditches with poor distribution management. This caused water wasting and salinity is building up in the water table. Farmers start to feel the problem created by lowering the water table because costs to pump the water from deeper depths increase. Farmers that depend upon the river to irrigate start to use conveyance pipes to achieve more efficient water use and reducing wastages. Water management to cope with water scarcity requires measures and policies aimed at reducing the demand and make water use more efficient (Pereira et al., 2002a, b) and demand management for irrigation under water scarcity includes practices and management decisions of agronomic, economic, and technical nature. Therefore, this study concentrates on demand management in order to save water through better irrigation scheduling and improved surface irrigation systems. To assist the development of irrigation scheduling strategies for coping with water scarcity, the simulation model ISAREG is used to evaluate the current irrigation scheduling for wheat and cotton, and to produce appropriate irrigation schedules for both crops. In fact that irrigation scheduling plays an

2

important role in achieving water saving, higher irrigation performances, and controlling the percolation (Smith et al., 1996; Pereira et al., 2002a, b). By the time the farmer is looking the highest and the maximum yields, for several crops maximizing yield is at the account of the water productivity, i.e. the harvestable yield per unit volume of water used. For small grain cereals, water productivity drops at high yield levels under full irrigation (Zhang and Oweis, 1999). Maximizing water productivity may be more profitable to the farmer than maximizing crop yield. This occurs when the water saved by adopting deficit irrigation can be used to irrigate additional land. To achieve effective water conservation and saving, improving water productivity and increasing farmers Incomes are also required. Formulating improved irrigation scheduling scenarios needs to be combined with higher performance of surface irrigation systems, which highly depends upon the design process, including decisions on land levelling, field shape and dimensions, and inflow discharge. In addition, it depends on farmer decisions such as for land levelling maintenance and timeliness and duration of irrigation events. Studies aimed at improving furrowed basin and border irrigation were extensively used to build the demand scenarios for selected years using an irrigation demand and delivery decision support system (DSS) focusing on the farm scale, the SADREG model, where scenarios for the farm application and distribution systems are evaluated using multicriteria analysis (Gonçalves et al., 2001, 2004, 2005a, 2005b, 2006a, 2006b). The model helps the selection of the best decisions to achieve the sustainable use of the water in irrigated agriculture taking in account the environmental, economical and technical impacts relative to reducing the water use. The study is part of the Project “Rationalization of Ras EL Ain Irrigation System Project” and data were collected in the framework of this project. 1.2. Objectives: The general thesis objective is to contribute to improve the water management on irrigated agriculture in Ras-al-Ain region, Syria, aimed at to solve the actual problems related with water scarcity and agriculture sustainability. The specific objectives are:

1) The field assessment of actual irrigation systems and scheduling, to better understanding the problems, making a diagnostic analysis, and recognizing main constraints;

2) Determination of representative parameters for modelling the irrigation process, based on field observation and trials; it includes crop and soil observations and the assessment of surface irrigation practices;

3) To carry out field experimentation of improved surface irrigation systems, and to test the effectiveness of improved techniques;

4) To analyse the irrigation scheduling strategies for wheat and cotton in combination with improvements of surface irrigation systems;

5) To apply a decision support system to select improved solutions for on-farm surface irrigation and ranking alternative approaches using a multicritera methodology.

3

Chapter 2 2- Ras El Ain Irrigation District 2.1- General information: The Ras Al Ein Irrigation District is located in the northeast of Syria, and has an irrigated area of 42,220 ha, which represents about 20% of the total farming area. The Ras Al Ein area is within the governorate of Al-Hassakeh that border Iraq from the east and south, Turkey from the north, the Der El-Zor governorate from the southwest and the district of Rakka from the west. Ras El Ain area is classified as the second settlement in the Governorate. The Al-Hassakeh governorate is located on 35 – 37° latitude and 39 – 42° longitude east. The total area is 23,359 square kilometers and is crossed by the Khabour and jakjak rivers, Ras Al Ein Irrigation District is located in the Euphrates basin, with an elevation range from 165 to 325 m over the sea level, with an annual rainfall of 200-250mm and an annual potential evaporation of 1,600-2,800 mm

Figure 2.1. Location map of Ras Al Ein Irrigation District

4

2.2. Water resources The springs of Ras el Ain Irrigation District are situated near the border with Turkey. They still represent the main source of water supply for irrigation. The main rivers cross in the region are: (i) Al-Khabour River and its branches: it is 402 km long, the actual discharge is around 6.6 m3/s, and in the previous years (until 1995) it was around 21.5 m3/s. In summer, its discharge might become around 3 m3/s; (ii) Al-Gakgak River: it is 100 km long; the actual discharge is equal to 2.1 m3/s while the previous discharge was around 18 m3/s. It is possible to utilize this river just during winter and spring because during summer it is blocked from the Turkish side. Note that data above provided by the Irrigation Ministry for 2004 is overestimated because recently much less discharges are observed. The artificial reservoirs are: (i) Al-Hassakeh western dam, with a maximum storage capacity of 91 million m3; (ii) Al-Hassakeh eastern dam, with a maximum storage capacity of 234 million m3, with a channel linking them. However, the farmers in the project area depend upon the ground water table and the Khabour River. A channel feeds both dams from pumping from the Ras-El-Ain groundwater since the Khabour River has not enough water after the construction of dams in the Turkish upper basin. This is one of the reasons for the progressive depletion of the ground water table. Pumping amounts now 20 m3/s, and the ground water table is decreasing by 1.02 m/year. In addition there are numerous illegal wells explored by the farmers because they have no more access to the river water. Considering the situation in the Ras_Al_Ein area, the cooperative Italian project “Rationalization of Ras El Ain Irrigation System” was established to help solving the problem of the scarce water resources in the project area of Ras el Ain, around the springs of Al Khabour, associated with the improper use of traditional irrigation techniques. The main objective of this project is Food and income security through (i) increase in yield of strategic crops in the province; and (ii) increase in farmers’ income. The specific purpose is to reduce the use of groundwater resources in the project area through development and adoption of more efficient irrigation methods by farmers. 2.3. Climate and Soils The annual rainfall range from 250 to 350 mm Based on the average of nine year of climatic data (1993-2002) from Al Hassakeh research station, the main characteristics of the climate are presented in Figs 2.2 and 2.3, the coldest month is January and the hottest one is July or August. The air humidity is highest during the winter months and lowest during the summer time. The predominant wind blows from the west, and the wind velocity rises during the summer. The Fig. 2.2. shows that arid conditions prevail after mid

5

April, thus increasing the crop water demand during the critical growth stage of wheat and the full growth season of cotton. Rainfall is erratic in both volume and distribution; the maximum rainfall occurs during January and is insufficient during the wheat and cotton seasons (Fig. 2.2). A significant water deficit occurs then due to the high evapotranspiration and low precipitation.

0

50

100

150

200

250

300

Jan

Feb

Mar Apr

May Ju

nJu

lAug Sep Oct Nov

Dec

months

ET0,

P m

m

0

5

10

15

20

25

30

35

40

45

T C

°

Rainfall mm Eto mm Max T C° Min T C°

Figure 2.2. Average precipitation, reference evapotranspiration and minimum and maximum temperatures

30

40

50

60

70

80

90

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

hum

%

0

0.5

1

1.5

2

2.5

3

3.5

4

win

d sp

eed

m/s

min hum % Humidity % wind speed m/s

Figure 2.3. Average and minimum monthly humidity with wind speed

6

The land in Al Hassakeh area is almost flat with smooth topography, except for the mountains around the governorate. The soil classification in Al Hassakeh area has the following distribution (i) 45% Cennamonic soil; (ii) 31% Gypsiferous soil; (iii) 19% Ggrumusol ; (iv) 5% alluvial (Irrigation Ministry). The main soil classification according the survey carried out in 1984 by Agrocomplect, an Engineering Economic Organization, is as the following: (i) Brown soil (thick and medium thick); (ii) Light brown soil (thick and medium thick and shallow); (iii) Meadow dark brown soil (thick); (iv) Meadow brown soil (thick);(v) Undeveloped soil with gypsum bulges; (vi) Complex of brown and light brown soil. The soil in the study area is classified as thick brown soil. 2.4. Agriculture and Land Use The most important crops cultivated in Hassakeh province are wheat and cotton, followed by vegetables, fruits and fodder crops. Out of the total cropped area, wheat accounts for 31% (of which 40% under irrigation), and 28 % barley, that is mostly rainfed. Cotton is the most profitable crop, but also the most water demanding and is only cultivated under irrigated conditions. Table 2.1 shows the wheat and cotton yields (1999) for the Al Hassakeh province, and how important these crops are compared with the production of the whole country. The wheat is cultivated in winter-spring seasons, it is drought tolerant and needs a minimum of 300 mm of rain (under rain fed conditions). It is sown in December and harvested by June. Local improved varieties produced by MAAR or in collaboration with ICARDA are usually utilised. With the current irrigation system, about 6000-7000 m3/ha of water per season are required, that is supplied in 4-5 water applications. The wheat crop is fairly resistant to soil salinity, and the yields can reach as much as 4000-5000 kg/ha under traditional irrigation systems. Table 2.1 Cultivated area and the yields for wheat and cotton in Al Hassakeh.

The cotton crop performs best in hot and dry climate, with a quite high water requirement. It is one of the most salt resistant crops in the area. Cotton is grown in summer, sown in April and harvested by October - November, with a growing season of about 180-190 days. The traditional water requirement

CULTIVATED AREA (ha)

PRODUCTION (t)

Irrigated wheat 282,518 833,750 Rain-fed wheat 400,410 110,778

wheat (total) 682,928 944,538 (35% of the total national production)

cotton (irrigated) 85,536 357,808 (39% of the total national production)

7

with the current furrow irrigation system is about 16-18,000 m3/ha in 10-12 water applications; the yield is very good, reaching 4000-4500 kg of cotton seed. The cotton is very sensitive to the water stress and the soil water is the most important environmental factor affecting water content of cotton leaves (Namken, 1964). The most sensitive period for the cotton is from the appearance of the first flowers to peak flowering (Bruce and Ship,1962). when should be supplied about 30-35% from the total water used (Bielorai and Shimshi,1963), Otherwise the water deficit will cause yield reduction greater than that imposed at any others stages Miller and Dickens (1970). 2.5. Irrigation Systems The total irrigated area in Hassakeh according to estimation of Agriculture Ministry is 420,872 ha, of which only 1030 Ha regard fruit trees and grapes. This estimation probably is under evaluated because only 65% of wells are legally built. Most of land is irrigated through surface irrigation while only 23,301 ha are sprinkler irrigated and 1,690 ha are irrigated by drip systems. About 80% from the farmers use traditional irrigation systems and 20% adopt some improved systems (project Baseline Survey 2005). Table 2.2 shows the main characteristics of different irrigation systems applied for cotton. Table 2.2 Characteristics of irrigation systems applied for the cotton crop (source: Hassakeh research center, Ministry of Agriculture)

The wheat that is irrigated from wells requires usually five irrigations with a total water use of 7,000 – 8,000 m3/ha, while the wheat irrigated by the government water resources are given 2-3 irrigations according to the available water in the dams. Often the last irrigation is made in the second half of May. Sometimes the farmers make irrigation before the tillage to reach the field capacity of the soil and to better prepare the land. The land is divided as furrowed basins and small borders. The size of the basins or borders relate to the topography of the land. The water is conveyed through a main earth canal and divided into distributors in the field. These systems produce a loss of the cultivated surface. For cotton irrigated by wells around 10 - 12 irrigations are given totalizing up to 16,000 – 18,000 m3/ha with low efficiency. The first cotton irrigation usually is up to 200 mm at planting; the second is made after one month. Than all

Irrigation System Trickle Sprinkler Traditional

7800-8700 10000-10600 14500-16000 Water use (m3/ha) 46 32 0 Saving (%) 4650 4450 3750 Production (kg/ha) 0.563 0.42-0.44 0.23-0.25 Yield per water unit (kg/m3) 80-85 70-75 45-55 Application efficiency (%)

8

others irrigations are applied regularly every 7 - 10 days with 130 – 150 mm each. The most common system is zigzag furrow with basin spacing usually of 90 cm. sowing is done on the both sides of the furrow. The irrigation efficiency is estimated to be very low: for cotton (furrow irrigation), near 50% (including conveyance and effective distribution losses), whereas for wheat (basin and furrow irrigation) it can reach 60%. Recently, to avoid conveyance losses, wells are provided with mobile polyethylene pipes (diameter of 5-6”), long enough to convey water directly to the borders of the plot under irrigation (Project repot 2003). Sprinklers and drip irrigation are not common because pf their costs. Sprinklers could be used for both wheat and cotton while adopting drip for cotton the farmer should keep surface irrigation for wheat; by the time the farmers have two crops in rotation, they should adopt only one irrigation system as for the improved surface that is applied by some farmers, as in Bab Al Faraj, but this is not common (Pereira, 2006).

9

Chapter 3 3- Material and Methods 3.1 Experimental sites: Arnan and Bab Al Faraj The field trials were developed in two sites, Arnan and Bab Al Faraj, located in representative parts of Ras El Ain irrigation District (Figure 2.1). The experimental farm of Arnan is located in the southern west of Ras El Ain town, and is managed by the Syrian Ministry of Agriculture and the cooperative Italian project. The total cultivated surface area is 32, 62 ha and is divided into many fields for the 2005-2006 season (Fig. 3.1, Table 3.1).

Figure.3.1.Arnan experimental farm. Fields D1 and part of D2, about 1, 9222 ha, were planted with: (i) winter crops (fababeans, chick pea and lentil); and (ii) summer crops (maize, soybean, and peanut). Both traditional and sprinklers irrigation systems are used.

10

Table.3.1. The cropped fields in the Arnan experimental farm, 2005/2006 Field Crop Irrigation System area (ha)

A wheat sprinklers 2,0869 B wheat Improved surface irrigation 5,1243 C1 wheat Improved surface irrigation 3,9595

C 2 + C 3+ C 4 wheat Traditional 11,0461 *D1 wheat sprinklers 1,8324 *D2 wheat Traditional 2,0986 D1 cotton sprinklers 1,2000 D2 cotton Traditional 1,5000 E1 cotton Drip 1,5000 E2 cotton Improved surface irrigation 1,5500 E3 cotton Traditional 1,5000

The map of Arnan (Fig. 3.1) shows the variation in elevation for each field. The fields irrigated by traditional and improved irrigation system have the following slopes S: (i) field C1 with improved S = 0.007 m/m; (ii) field B, that was divided to five parts, with S ranging from 0.007 to 0.009 m/m; (iii) field D2 with slope = 0.009 m/m; (iv) field E2 and E3, with improved surface irrigation, were divided into three parts with S from 0.0065 to 0.01 m/m. In field E land levelling has been performed but keeping the slope, thus just smoothing the land topography. The physical and chemical soil characteristics are shown in Table 3.2 and 3.3. Analyses were performed by the Soil and Water Laboratory of the Al Hassakeh Research Center. Table.3.2. Physical soil properties of Arnan experimental farm

Table 3.3. Chemical soil properties of Arnan experimental farm

Bulk

Density

FC weight

%

WP weight

%

FC volume

%

WP volume

%

TWA mm/ m

Sand %

Silt %

Clay %

soil texture classification

A 1.37 26.74 17.88 36.78 24.55 122.35 30 27 43 Clay

B 1.37 28.23 17.96 38.76 24.66 141.09 30 29 41 Clay

C 1.33 27.97 17.46 37.13 23.21 139.27 31 31 39 Clay Loam

D 1.36 29.12 18.86 39.46 25.56 139.02 23 31 46 Clay

field pH Ec m

mhos/cm ORGANIC

% CACO3

% P p.p.m K p.p.m

A 7,14 2,3 1,47 23,5 14 100

B 7,59 0,88 1,34 25,9 10,7 125

C 7,62 1,39 1,32 24,4 5,65 87,9

11

The physical soil properties for the fields C and D, whose samples were taken at three depths in each 30 cm, were also analysed at the ICARDA Lab. Results averaged for 90 cm soil depth are shown in Table 3.4 and Fig. 3.2.

Table.3.4. Physical soil properties of Arnan experimental farm as measured in the ICARDA Lab

-100

-90

-80

-70

-60

-50

-40

-30

-20

-10

00 10 20 30 40

water soil content volum %

soil

dep

th c

m

PWP FC Figure.3.2. the soil water content in 90 cm soil depth of Arnan experimental farm. Water is obtained from two wells (W45 and W2) that have an average discharge of 45 l/s and 25 l/s, respectively; discharges have a little variation during the year. The chemical water properties were measured at the Soil and Water Laboratory of the Al Hassakeh Research Center and are shown in Table 3.5.

Table.3.5. Chemical water properties (Arnan) mg /L Well n pH Ec m

mhos/cm Ca Mg Na K Cl HCO3 NO3

p.p.m

Well45 7,7 0,7 8,25 1,65 2 0,019 1,92 1,8 5,59 Well2 7,6 1,37 11,52 6,05 2,6 0,025 3,84 3,6 9,15

The Bab Al Faraj experimental site Is located in the eastern side of Ras El Ain town, about 35 km faraway. One farmer there, Ammar Jebraiel, has a large farm and adopted an improved surface irrigation system for cotton designed as graded furrows, with long and short lengths, with medium and quite large slope. The water is distributed through a ditch with small pipes across the ditch wall; the discharge is controlled by the water head in the supply ditch and by the diameter of the pipes (Pereira and Gonçalves, 2005).

FC weight

%

WP weight

%

FC volume

%

WP volume

%

TWA mm/ m

Sand %

Silt %

Clay %

soil texture classification

C 26,2 16,2 34,8 21,5 132,6 27,9 30,8 41,3 Clay

D 26,4 16,7 35,9 22,7 131,9 23,4 31,5 45,1 Clay

12

The sowing is done by a seeding machine The preparation of the land for the cotton crop is made by creating small furrows that supply the water in the first irrigation with a very small discharge; the furrow spacing is 0.7 m and the planting is on the top of the bed, Then, before the second irrigation, they enlarge the furrow using another machine. A topography survey was done for three different field parts: (i) R1.1 representing the long field, with length L = 224 m and slope S = 0.019 m/m; (ii) R1.2 represents the short one, with L = 90 m and S = 0.0096 m/m; and (iii) R1.3 represent another field with L = 100 m and S = 0.007 m/m. 3.2. Irrigation scheduling: the ISAREG model 3.2.1 Model description An appropriated irrigation scheduling plays an important role in achieving water saving, higher irrigation performances, and controlling the percolation (Smith et al., 1996; Pereira et al., 2002b). Computer models are an easy approach for developing and evaluating alternative strategies for irrigation, and a large number of models are available for computing the soil water balance and generating improved irrigation schedules. This is the case of the ISAREG model (Pereira et al., 2003) that was widely successfully applied in several countries. In Central Asia (Сholpankulov et al., 2005) it contributed to control soil salinity and to decrease water use, so to combat desertification problems in the Aral Sea basin, where it is also operational in a GIS environment (Fortes et al., 2005), In the north of Syria it was applied in the ICARDA’s Tel Hadya research farm, to develop supplemental irrigation strategies for coping with droughts and water scarcity (Oweis et al., 2003). The ISAREG model performs the soil water balance for a multilayered soil using different options to define and evaluate the irrigation schedules, as described by Pereira et al. (2003). Input data (Fig. 3.3) include precipitation, reference evapotranspiration, total and readily available soil water, soil water content at planting, parameters characterizing conditions for groundwater contribution, crop coefficients and soil water depletion fractions p for no stress relative to crop growth stages, root depths and the water-yield response factor. Depending on weather data availability, various time step computations can be used, daily, decennial or monthly. The reference evapotranspiration ETo (mm) is computed with the FAO-Penman-Monteith method (Allen et al., 1998) using the program EVAP56. The crop data parameters are determined with the KCISA program (Rodrigues et al., 2000), using the methodology proposed by FAO (Allen et al., 1998). The crop evapotranspiration ETc (mm) is computed as the product of ETo by the crop coefficient Kc.

13

IRRIGATION MODES

Irrigation Options

Water Supply Restrictions

Potential Ground Water Contribution

- type of soil - water table depth

SOIL WATER BALANCE

IRRIGATION REQUIREMENTS - decaday - month - season

EVALUATION OF A GIVEN IRRIGATION SCHEDULE

Yield reduction

OPTIMIZED IRRIGATION SCHEDULING

METEOROLOGICAL DATA

Reference Evapotranspiration

Effective Precipitation

AGRICULTURAL DATA

Crops: - crop stages - crop coefficients - root depth - yield response factor - soil water depletion factor

Soils: - soil layers depth - field capacity - wilting point

Figure.3.3. Simplified flow-chart of ISAREG model (Teixeira & Pereira, 1992) The crop coefficients (Kc) and the soil water depletion fraction for no stress (p) were adjusted to the Ras El Ain climatic conditions. Kcmid and Kcend were

adjusted with equation [1] and p with equation [2]:

302 3

4500402040 ..min)( ))]((.)(.[

hRHUKcKc tabmidmid −−−+= [1]

)(. ETcpp tableadj −+= 5040 [2]

where Kcmid(tab) is from the FAO table, U2 is the wind speed (m/s), RHmin is the minimum relative humidity (%) and h is the height of the plants (m) at the mid season stage. The ISAREG model computes the irrigation water requirements through the soil water balance that is calculated for the effective root depth as:

( )ir

iiiaiwiiii z

GWDPETIROP

10001

+−−+−+= −θθ [3]

14

where θi and θi-1 are soil water content in the root zone (mm mm-1), in the days i and i-1, Pi is the precipitation (mm), RO i is the runoff (mm), Iri is the net irrigation depth (mm) that infiltrates in the soil, ETci (mm) is the crop evapotranspiration (mm), DPi is deep percolation (mm), GWi is the groundwater contribution (mm), and zr i is the rooting depth (m) in day i. GW and DP are estimated from soil hydraulic properties and the water table depth as described by Liu et al. (2006) after appropriate parameterization of the respective functions as referred by Сholpankulov et al. (2005). When water availability is non-limiting, the frequency of irrigation is not restricted and varies along the crop season according to the crop demand, the irrigation depth and the dates due to different objectives are computed according to the water depth limits and water soil threshold defined by the user. If the irrigation threshold is defined by the depletion fraction for no stress the irrigation is scheduled when the soil water is

( ) ( ) WPWPFCp θ θθ p1θ +−−= [4]

The corresponding net irrigation depth is

( )pFCirzI in θθ −=1000 [5]

In case of allowed stress, where the management allowed deficit MAD<p (Martin et al, 1990), the irrigation are scheduled for :

( )( ) WPWPFCi θθθθθ +−−== MAD1MAD [6] The applied depth is either a user selected fixed quantity D (mm), or a variable quantity D = θFC - θi. The water stress impact on the crop yield is evaluated through the model proposed by Stewart et al. (1977) where relative yield losses depend upon the relative evapotranspiration deficit

−=

−

c

dy

c

d

ET

ETK

Y

Y11 [7]

where ETd and ETc are respectively the seasonal actual and potential crop evapotranspiration (mm) and Yd and Yc are the corresponding achieved yields and Ky is the water-yield response factor. 3.2.2. Input data: 3.2.2.1 Climatic and soil data Climatic data was observed by the meteorological stations in Arnan and Ras El Ain, the last one located about 8 km far from the experimental farm. Both

15

meteorological stations are installed and managed by the Italian Project and the Agriculture Ministry, In case of missing data from one station due to any technical problem, they could be replaced from those of the other one. The Arnan meteorological station is located at 36.404° latitude and 330m altitude. Table 3.6 shows the climatic data used with ISAREG for both crops wheat and cotton. Data for December until July relate to the year 2006 and from August until October are for 2005. The ETo data were calculated with the available climatic data, with the vapour pressure deficit computed from air temperature. Table 3.6, Meteorological data from Arnan station (completed with those for Ras El Ain), 2005-2006

The hydraulic proprieties of the soil were averaged for one single layer with FC = 35.5% and WP = 22% (volume ratio). The total available soil water is

( ) zWPFCTAW 10×−= [8] Thus, for a homogeneous soil and 1 m depth it results TAW = 135 mm/m. A soil electrical conductivity Ec =1.4 mmhos/cm was considered. 3.2.2.2 Crop data for wheat and cotton The crop data include the phenological plant stages, crop coefficient Kc, root depth during the season of the crop, soil water depletion fraction p, and yield response factor Ky. The phenological stages of the wheat and cotton for Arnan experimental farm were observed during 2005-2006 season (Table 3.7). The Kc and p in Allen et al. (1999) and Pereira et al. (1999) were adjusted with equations 1 and 2. The minimum and maximum effective root depths were taken from Allen et al. (1999) and Pereira et al. (1999): for wheat the

Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov

Temperature

[ºC] Max 16.1 9.2 14.0 20.4 23.9 32.9 40.5 41.4 39.5 35.6 28.6 19.6

Temperature

[ºC] Min 3.7 0.8 2.6 4.1 10.1 11.9 17.8 23.7 23.8 14.4 8.0 3.1

Relative humidity [%]

75.1 85.0 78.5 68.6 78.0 54.9 31.4 33.7 37.8 42.7 46.9 60

Wind speed

[m s-1] 0.90 1.64 1.66 1.69 1.51 1.20 2.39 2.64 2.14 2.70 1.18 1.06

ETo[mm d-1]

Tmin 2.24 0.95 1.63 3.15 3.92 6.19 8.98 8.19 6.67 6.54 2.34 2.1

Precipitation [mm]

45.0 73.7 124.2 5.9 82.3 3.7 0.0 0.0 11.9 7.7 4.1 19

16

depth of 75 cm was considered for the whole season, whereas for cotton 100 cm were estimated. Table.3.7. Average crop parameters for wheat and cotton

The yield response factors Ky proposed by Stewart et al. (1977) were adopted: for cotton is Ky = 0.85 and for wheat is Ky =1.05. The initial soil water content for wheat was estimated for the seeding layers as 40% from the TAW and 30% for the layers below it, and for cotton 20% and 40% respectively. 3.3. Surface Irrigation assessment and modelling

3.3.1 Field evaluation 3.3.1.1 Infiltration tests Infiltration is one of most crucial factors affecting surface irrigation. It controls not only the amount of the water entering the soil, but also the advance rate of the overland flow. The empirical Kostiakov function is usually applied to model the infiltration process in surface irrigation conditions:

01 faKI a += −τ [9]

ττ ⋅+⋅= oa fkZ [10]

where I is the infiltration rate (mm/h), Z is the cumulative infiltrated depth (m3/m), a and K are empirical parameters and f0 is the final infiltration rate and τ is the intake opportunity time (h).

Parameters Crop development stages crop

Initial Development Mid

season End

season Period length

(dates) 1/1 - 1/2 2/2 - 30/4 30/4- 26/5 27/5 - 15/6

Crop coefficients, Kc

0.54 0.54 – 1.10 1.10 1.10 - 0.32

Depletion fraction, p

0.72 - 0.70 0.70 - 0.60 0.60 - 0.59 0.59 - 0.62

Wheat

Root depth cm 75 75 75 75 Period length

(dates) 07/5 - 6/6 7/6 - 7/7 8/7 - 8/9 8/9 - 11/10

Crop coefficients, Kc

0.30 0.30 – 1.13 1.13 1.13 - 0.61

Depletion fraction, p

0.75 - 0.59 0.59 - 0.50 0.50 - 0.62 0.62 - 0.74

Cotton

Root depth cm 100 100 100 100

17

The double ring infiltrometer is one of the methods used to determine the permeability of the soil (infiltration rate), but is affected by three main factors (i) presence of a compacted layer; (ii) the bulk density; (iii) the porosity that refer to tillage, crop residue and root development, for example. This instrument consists of two metal rings with diameter 24” and 12”, designed to prevent divergent flow in the soil layers; the outer ring acts as a barrier to encourage only vertical flow from the inner one, when both of them are filled partially of the water that is controlled to the limited level by a Mariotte tube, depending on the type of the soil and the permeability level (Walker and Skogerboe, 1987). The procedure utilized is the following: (i) Place the ring (outer and inner) onto the soil, the inner is in the middle of the outer one; (ii) Forcing the rings to the ground by the hummer, until 3” of the ring stick in of the ground with keeping the ring level horizontal; (iii) put a plastic sheet inside both of the rings in order not to allowed any leaking through it into the soil; (iv) fill slowly the cylinder infiltrometer with a known volume of the water ( to a depth of about 10 cm) so that the initial depth at the time can be calculated as a zero; (iv) set the time, and gently pull the sheet out; as the water hit the soil start the time; (v) record the time with level of the water in the inner ring every second for the first inch then increase the interval observation time recorded during the test until the stable infiltration rate, keeping refilling the water with correcting the reading of the gage. The measurement should usually be continued until 4 hours elapsed time. Data as recorded in the field are presented in Annex (1). Fig. 3.4 shows an example of observed data and respective fitting with eq. 9 and 10, i.e. the infiltration rate I (mm/h) curve and the cumulative infiltrated depth Z (m3/m) curve, the latter corresponding to the equation

3227.0797.11 τ=Z [11]

y = 11,797x-0,6773

R2 = 0,9906

y = 11,797x0,3227

R2 = 0,9601

0

10

20

30

40

50

60

70

80

90

100

0 50 100 150 200 250 300 350time (min)

infil

tratio

n (m

m)

0,000

2,000

4,000

6,000

8,000

10,000

12,000

14,000

Fig. 3.4. Infiltration curves obtained with the double ring test at Arnan farm.

18

Another infiltration field method used in this study uses furrow advance data observed in furrow evaluation to predict the infiltration parameters. In this field evaluation procedure: (i) divide the furrow in four parts limited by stations at 0/4, 1/4, 2/4, 3/4, 4/4 of field length; (ii) place a steak with the number of the station along the furrow; (iii) place a flume in the station that wanted to measure the inflow rate through it; (iv) numbering the treatments in the field with target discharge in order to adjust it through the valves in the pipes for each treatment except the present one; (v) make three repetitions for each treatment to get the average. In this study some of the treatments had a repetition especially in Bab Al Faraj. The final infiltration rate fo is calculated from inflow & outflow hydrographs analysis, after the stabilization of discharge:

L

QQf outin

o

−= [12]

The parameters a and k were determined using the ‘‘two-point’’ method lying an advance phase water balance proposed by Elliott and Walker (Walker and Skogerboe, 1987), with making small adjustments for k parameter for some of the treatments in order to have right water balance.

aLz

L

rxad

z

Ly

LL

Ly

LL

LL

LL

t

Vk

tpx

ra

arar

tfA

L

tQV

r

tfA

L

tQV

tt

VVa

σ

σ

σ

σ

=

⋅=++

+−+=

+−−=

+−−=

=

)(

)1)(1(

1)1(1

21

)/ln(

)/ln(

5.000

5.005.0

00

0

5.0

5.0

3.3.1.2 Field evaluation of surface irrigation The methodology used for the evaluation of the furrow irrigation follow that proposed by Merrian and Keller (1978) as adapted by Calejo et al. (1998). Measurements were carried out in Arnan (field E2) and Bab El Faraj; they included observations of the land level condition, furrowed cross section, advance time, recession time, roughness and the applied water depth. Table 3.8 shows the treatments adopted for the 1st and 2nd irrigations, with different length, slope and discharge.

[13] [14] [15] [16] [17] [18]

19

Table 3.8. Field treatments in Arnan and Bab Al Faraj.

ET - Irrigation every furrow AF - Irrigation of alternate furrows In the field E, a topography survey was made by a team of the Irrigation Ministry, and the spacing between the contour lines in the field corresponding to the map is 0.1 m/m (Annex 1). The field size and slopes are in Tab.3.9. The field width is W = 50 m. Table 3.9. Topographic characteristics of field E2, Arnan. Field E (300x50) Length m Longitude slope m/m Cross slope m/m First part 0-100 0.01 0.007 Second part 100-200 0.0064 0.005 Third part 200-300 0.0076 0.005

The irrigation schedule in the field was analyzed with ISAREG model, assuming MAD = p. The net irrigation depth per irrigation was In = 80mm. The discharge in each furrow was measured at the upstream, middle and downstream sections by volumetric methods with a 2.7 liter bucket and two kinds of portable flumes. With a small 60° V-notch flume the discharge was calculated according to the following equation (Trout, 1983):

61.2)15.0(*249.0 −= huQ [19] where Q is the discharge (l/min), hu is the gage reading (cm) and the 0.15 cm is the correction for meniscus distortion.

Location Irrigat. Nr.

Irrigation treatment

Length (m)

Slope (%)

Inflow (l/s)

Furrow Spacing (m)

Irrigation Management

Arnan 1st A0 50 0.9 4.78 0.9 Zigzag Basin

A 150 0.95 0.53 0.7 EF - Open Furrow B1.1 100 0.97 0.71 0.7 EF - Open Furrow B1.2 100 0.97 0.77 0.7 EF - Diked Furrow B2 100 0.97 0.52 0.7 EF - Diked Furrow B3 100 0.97 0.35 0.7 EF - Diked Furrow C1 50 0.8 0.78 0.7 EF - Diked Furrow C2 50 0.8 0.52 0.7 EF - Diked Furrow C3 50 0.8 0.35 0.7 EF - Diked Furrow D1 36 0.8 0.75 0.7 EF - Diked Furrow D2 36 0.8 0.50 0.7 EF - Diked Furrow

1st

D3 36 0.8 0.32 0.7 EF - Diked Furrow E1 100 0.97 0.74 0.7 EF - Diked Furrow E3 100 0.97 0.37 0.7 EF - Diked Furrow

Arnan

2sd

F 100 0.97 0.36 1.4 AF - Diked Furrow G 224 1.9 1.3 0.7 EF - Diked Furrow H 90 0.95 0.52 0.7 EF - Diked Furrow

Bab-Al- Faraj

1st

I 100 0.7 0.36 0.7 EF - Diked Furrow

20

The second fume is the modified trapezoidal broader-crested weir (Walker and Skogerboe, 1987). The average discharge is computed by the time weight Qavg (l/min):

ap

n

i qi

avg t

AQ

∑ == 1 [20]

where Aqi are the inflow volumes (l) during the time intervals from ti to ti-1 computed as

))((30 11 −− −−= iiiiqi ttQQA [21]

and ti and ti-1 are the times of two successive inflow rate measurements (min), which start at the moment when the irrigation has started; Qi and Qi-1 are the recorded furrow flow rates (l/s) at those times ti and ti-1; n is the number of flow rate measurements; and tap is the total time of water application (min) For the first treatment A0 (Table 3.8, traditional irrigation system where the water is supplied to a zigzag furrowed basin) the width of the basin varies from 3.5 m to 6 m according to the machine that is used and size of the farm, while the length depends on the field and influences the water volume in relation to the mean discharge, slope and the application time. In this treatment A0 the length of the basin was L = 50 m, the width of the basin was W = 5 m, and the furrow spacing was FS = 0.9 m. Planting was done on both sides of the furrow. The observation stations along the basin were set up on the 0/4, 1/4, 2/4, 3/4, 4/4 of basin length as in the furrow. These stations were used to record the advance and recession times, which identify the opportunity time for the infiltration τ = trec - tadv, The inflow rate was measured upstream of the furrow by a portable flume. Recession times were recorded at the times when water fully infiltrated the soil at the observation sections; however, when unevenness of the furrow bed caused the water to pond for long time, trec were recorded when water disappeared from the furrow bed in the areas nearby the measurement section. It resulted that advance time measurements were more accurate than recession ones. The inflow rates in the 0/4 and 2/4 stations are shown in Fig. 3.5. The example refers to a 90 m long furrow.

21

inflow rate l/s

0,00

0,10

0,20

0,30

0,40

0,50

0,60

0,70

0 30 60 90 120 150 180 210 240 270 300 330 360 390time min

dis

char

ge

l/s

upstream inflow rate l/s infow rate in the 2nd sta l/s

Fig.3.5. The inflow rate in 0/4 and 2/4 station (Bab Al Faraj) The Manning’s roughness coefficient n (m-1/3 s) was calculated from observations of the furrow cross-sectional area, flow rates, flow water depths and water surface width with the equation

QSAR

n2132 //

= [22]

where Q is the inflow rate to the furrow (m3/s); A is the cross-sectional area of the furrow flow or wetted area (m2) eq (5); S is the hydraulic gradient, which was assumed to equal the furrow slope (m/m), R is the hydraulic radius (m) defined as WAR /= ; and W is the wetted perimeter (m). The Kostiakov infiltration equation, which is adopted in the model SIRMOD (ISED,1989), was used in this study as mention above:

τττ ⋅+⋅= oa

i fkZ )( [23]

The infiltration parameters were estimated using the inverse method (Katopodes et al., 1990) in which observed advance and recession data are compared with those computed with the simulation model SIRMOD. The best parameter values were obtained after several iterations aiming at minimizing the sum of the squares of the deviations between observed and simulated advance and recession times. The need for using both advance and recession observations when searching the infiltration parameters (Calejo et al., 1998) was confirmed in this study. The furrow shape parameters W and A may be described with equations 19 and 20:

0202 ))

2((2 T

TwhW +

−+= [24]

hTw

A )2

( 0+= [25]

where h is the observed water depth in the furrow, To is the width of the furrow bottom, w is the observed water width.

22

The total infiltration volume was computed from the equation

)22(2 210 nZ ZZZZ

n

LV ++++= L [26]

where L = the total length of the furrow, n is the number of stations along the furrow and Z is the cumulative infiltrated depth in each station. The furrow cross sections before and after irrigation were measured using a piece of wood with 16 holes 5 cm apart and 16 recording wood steaks. An example is in Fig. 3.6. Observation after three irrigations for the cross section show that shapes change from triangular to near rectangular as a consequence of in-furrow erosion and deposition.

0

20

40

60

80

100

120

140

160

180

-400 -300 -200 -100 0 100 200 300 400

X(mm)

Y(m

m)

in 1 st irri

Fig. 3.6. Typical furrow cross sections before irrigation 3.3.2. Performance indicators The performance indicators considered in this study are the application efficiency, Ea (%), and the distribution uniformity, DU (%). DU characterizes the irrigation system and Ea is a management performance indicator (Pereira and Trout, 1999; Pereira et al., 2002b). They are described by the following relationships:

100×=D

ZEa req If reqlq ZZ > [26]

100×=D

ZEa lq If reqlq ZZ < [27]

100×=avg

lq

Z

ZDU [28]

where Zreq is the average depth (mm) required to refill the root zone in the quarter of the field having higher soil water deficit; D is the average water

23

depth (mm) applied to the irrigated area; Zlq is the average low quarter depth of water infiltrated in the field (mm); Zavg is the average depth of water infiltrated in the whole irrigated area (mm). Zreq were estimated from field measurements of the soil water content before the irrigation, which were used to compute the soil moisture deficit, SMD (mm), in the root zone. Measurements were carried out for the fifth stations along the furrow. The maximum SMD observed were assumed as the best estimates of Zreq. For all irrigation events, the root zone depth was assumed equal to 0.1 m based on the maximum development of cotton root masses. Zavg was estimated from computing the depth of water infiltrated during the intake opportunity time relative to each location i, at each 10 or 20 m for short and long furrows, respectively. The Kostiakov equation was used with this purpose with the estimated infiltration parameters as referred above:

( ) ( )[ ] ( ) ( )[ ]iaira

iair ttfttkZ −+−= 0 [30]

where k, a and f0 are the infiltration parameters characterizing each irrigation; (tr)i and (ta)i are the, respectively, times of advance and recession relative to the location i (min). Zlq was estimated from the average relative to the quarter of the furrow where infiltration was smaller. The average depth of the water applied D (mm), was computed from:

sL

tqD coavf

×××

=60

[31]

where qavf is the average furrow inflow rate (l/s) during an irrigation event, tco the cut-off time or duration of the inflow (min), and s is the spacing between furrows (m). Similarly, the average outflow depth at the tail end of the furrow for some of treatments, Vout (mm), was calculated from:

sL

tqV outout

out ×××

=60

[32]

where qout is the average discharge rate at the end of the trail of the furrow (l/s) during the runoff time tout (min). 3.3.3. The SIRMOD model for surface irrigation simulation The process of surface irrigation combines the hydraulics of surface flow over the irrigated land with the infiltration of water into the soil profile. The equations describing the hydraulics of surface irrigation are the continuity and momentum equations (Walker and Skogerboe, 1987; Pereira and Trout, 1999). The SRFR (Strelkoff, 1993) and SIRMOD (ISED, 1989) are the most worldwide used software packages incorporating the solutions for those

24

equations. Both models can be used for irrigation design and evaluation and adopt the Kostiakov infiltration equation (9). The model SRFR solves the non-linear algebraic equations adopting time-space cells with variable time and space steps and also includes a full hydrodynamic model. The model requires as inputs the following data: (a) the longitudinal and cross sectional characteristics of the fields, like field geometry and slope, (b) the infiltration parameters of the Kostiakov equation (obtained from field observations and optimized using the inverse solution of the model where advance and recession are input data), (c) the hydraulic roughness coefficient, (d) the inflow rate or the inflow hydrograph when that is varied in time, (e) the target infiltration depth, and (f) the time to cut off. The model allows that slope; roughness coefficient and infiltration parameters vary along the field. The SIRMOD model requires an input similar to the SRFR and solves the continuity and momentum equations using the Eulerian integration method (Walker and Skogerboe, 1987). As output, the model produces several performance indicators such as the application efficiency (Ea) based on the stated target infiltration depth, and the distribution uniformity based on the low quarter depth infiltration (DUlq) as referred above (eq. 26 through 28). Once validated, the models may be used to generate alternative improvements concerning inflow rates, basin sizes, and land leveling impacts.

3.4 The Decision Support System (DSS) SADREG

3.4.1 Model description

The performance of surface irrigation systems highly depends upon the design process, including decisions on land levelling, field shape and dimensions, and inflow discharge. In addition, it depends on farmer decisions such as land levelling maintenance and timeliness and duration of irrigation events. The variety of aspects influencing irrigation performances makes the decision process quite complex and often out of the farmers experience and empirical knowledge. The improvement of farm irrigation systems in large surface irrigation projects can be well supported by DSS tools such as SADREG (Gonçalves and Pereira, 1999, Gonçalves et al., 2001, 2004). Its application may be performed at field level or at sector level when linked with a GIS (Gonçalves et al., 2005 a and b; 2006a and b). SADREG is a DSS aimed to assist designers and managers in the process of design and planning improvements in farm surface irrigation systems. SADREG includes a database, simulation models and user-friendly interfaces; and allows for ranking and selection of design alternatives through a multicriteria decision process, to aid selecting the best decisions. The SADREG application scope comprises: (a) a single field analysis relative to alternative design options for furrow, basin or border irrigation considering several decision variables such as field slopes, water delivery methods and

25

equipments, as well as reuse options; (b) an irrigation sector analysis, when a spatially distributed database relative to the farm systems is available through GIS, and where improvement alternatives are assessed jointly with modernization options relative to the conveyance and distribution network (Gonçalves et al., 2005b). SADREG is helpful to search and analyze modernization solutions for surface irrigation because designing surface irrigation systems imply the selection among a large number of combinations of main factors such as soil infiltration and water holding capacity; field sizes, slopes and topography; crop irrigation requirements, and inflow rates, which become easier to manipulate and ranking through a DSS tool. When several fields within an irrigation district are considered, then the task becomes only feasible if a spatially distributed database is also available. In addition, SADREG is conceived in such a manner that the user may learn through the application process. SADREG comprises two components: design and selection (Fig.3.7). The first one applies database information and produces a set of alternative designs, which characterization data is used for ranking and selection. The selection component is based on a multicriteria analysis in which the project alternatives are ranking allowing the decision-maker to select the best alternative. The decision-maker participates in all decision process through interface dialog structures, expressing its preferences and priorities required for ranking and selection of alternatives.

Alternatives

Models of multicriteria

analysis

Interface Dialog structure

Basic Data

Design Models: - Irrigation simulation - Impacts analysis

Interface

Dialog structure

Project Selected

DSS

Selection

component

Design

component

(Planner) User

(Manager)

Fig.3.7. Conceptual structure of SADREG (Gonçalves et al., 2005a, b).

26

The modular components of SADREG include a database, simulation models and the multicriteria analysis model (Fig.3.8). The database concerns field sizes and topography, soil intake rates, soil water holding capacity, economic data, crop data, and irrigation management data created through interactive simulations with the ISAREG model (described in Section 3.2).

Fig.3.8. Modular components of SADREG (Gonçalves et al., 2005a, b). SADREG is applied to a field assumed with rectangular shape, uniform soil intake characteristics and cultivated with a single crop. The water is supplied from a collective conveyance system or any other source that delivers the water from a given hydrant, which has specific hydraulic characteristics, like the maximum discharge and head. These data may be referring to an existing system, or user may select values. The surface irrigation models include a land levelling module, that applies an iterative optimization of landforms with minimal soil movement (Fig.3.9), and the SIRMOD simulation model (ISED, 1989) for surface irrigation design (Fig.3.10). The farm surface irrigation systems refer to basin, border and furrow irrigation. The later concerns continuous and surge-flow, automatic or manually controlled. Farm distribution systems refer to layflat tubing with gates, gated pipes, concrete canal with lateral holes, and unlined canals with or without siphons. The user may consider several design options, including relative to runoff water reuse and field length adjustments. The option of length adjustment could be interesting for long fields by comparing gains in the application efficiency against increased labour and operation costs. The evaluation analysis refers to cost and benefits as well as to environmental and performance indicators.

27

Windows interface

Land LevellingList of land levellingoptions

TopoSurvey file

Field file

irrigation methodoption

level basins gradedbasins/borders

gradedfurrows

select slopes

execute

output

update slopes

if user accept

insert newoption

Windows interface

Land LevellingList of land levellingoptions

TopoSurvey file

Field file

irrigation methodoption

level basins gradedbasins/borders

gradedfurrows

select slopes

execute

output

update slopes

if user accept

insert newoption

Fig. 3.9. Land levelling module flowchart (Gonçalves et al., 2005a, b).

input data

q=qmin

q ≤ qmaxexit No

Lengthinfiltrationroughness

Zreqtail end

inflow manag.

SIRMOD

control by dose

diverges? mod=modmax?Yes

mod=mod+1

recover(mod)

simulation notsuccefull

q=q+∆q

No

ez=100%?

tap=tap+ ∆t

SIRMOD

control by timeYes simulation

succefull

save resultson SIMDB

file

Yes

applied depth adjustment

irrigation simulation(control by dose)

input data

q=qmin

q ≤ qmaxexit No

Lengthinfiltrationroughness

Zreqtail end

inflow manag.

SIRMOD

control by dose

diverges? mod=modmax?Yes

mod=mod+1

recover(mod)

simulation notsuccefull

q=q+∆q

No

ez=100%?

tap=tap+ ∆t

SIRMOD

control by timeYes simulation

succefull

save resultson SIMDB

file

Yes

applied depth adjustment

irrigation simulation(control by dose)

Fig.3.10. Flowchart relative to the execution of the SIRMOD application procedure (Gonçalves et al., 2005a, b).

28

The main steps on a SADREG application are: (i) Identification of field characteristics; (ii) Scenario development relative to decision variables such as field

water supply, crop irrigation, furrow spacing, management allowed depletion (MAD), and furrows inflow regime (continuous vs. surge irrigation);

(iii) Data input referring to soil water data, infiltration and roughness parameters based on field experiments and/or databases, crop data, operation and equipment costs, labour and machine time durations, and water supply characteristics, such as the hydraulic head and number and discharge of field outlets;

(iv) Design procedure to create alternatives using both design models referred above (Fig.3.8) relative to the considered scenarios (item ii above); and

(v) Ranking and selection of alternative designs using multicriteria analysis where weights are defined according the user priorities.

To carry out this sequence of operations it is necessary to understand the main concepts and the hierarchy of the elements that compose the SADREG data structure:

(i) Field - is an rectangular shape on-farm land parcel, with a well known geographical location, with an uniform soil intake characteristics and a water supply hydrant; it is an element of a farm enterprise and belonging to a Water Use Association area;

(ii) Hydrant - is a gate on the network delivery system that supply the field; and

(iii) Outlet - is a discharge point, inside the field, connected to the field distribution system (a field can have one or more outlets).

The SADREG data structure can be described as follows (Fig.3.11):

(i) Workspace - is the basic element of SADREG data structure; corresponds to an individual Field and include all its data files. The information for each Field include: location; dimension; agronomic data; topographic survey, etc;

(ii) Project - each Project is a Field Scenario to develop a design for the selected field. Several projects can be created for each Workspace receiving different names;

(iii) Alternative - is a complete design solution for the selected field; (iv) Group of alternatives - a cluster that are differentiated by structural

decision variables (e.g. land levelling, irrigation method, equipments); within a group, the alternatives are differentiated by the operative values (unitary inflow rate and application time) and the number of sub-units;

(v) Unit – it is a field subdivision irrigated by a single outlet; it is assumed that all units of a field are similar; and

(vi) Sub-unit – it is the fraction of a unit that is irrigated at the same time.

29

GROUPirrigation method

distribution system

tail end manag .ALTERNATIVEunitary inflow rate

number of sub -units

performance attributes

operative variables

PROJECTwater supplySoil roughnesscrop

GROUPirrigation methoddistributionsystemtail end manag

.

ALTERNATIVEunitary inflow rate

number of sub

-

units


operative variables

WORKSPACEField data:

soil TAW

soil infiltration

length

width

area

topography


distribution system




operative variables

PROJECTwater supplysoil roughnesscrop

GROUPirrigation methoddistribution systemtail end manag

.


number of sub

-

units


operative variables

WORKSPACEField Data:

soil TAW

soil infiltration

length

width

area

topography


distribution system




operative variables

PROJECTwater supplySoil roughnesscrop

GROUPirrigation methoddistributionsystemtail end manag

.


number of sub

-

units


operative variables

WORKSPACEField data:

soil TAW

soil infiltration

length

width

area

topography


distribution system




operative variables

PROJECTwater supplysoil roughnesscrop

GROUPirrigation methoddistribution systemtail end manag

.


number of sub

-

units


operative variables

WORKSPACEField Data:

soil TAW

soil infiltration

length

width

area

topography

Fig.3.11. SADREG data structure elements. To construct the different alternatives of one project is necessary to have in mind the existing relations between the several irrigation design options (Fig.3.12.).

Land levelling option

irrigation method level basins graded basins/borders graded furrows

continuous continuous continuous surgeinflow regime

diked open without reuse gravity reuse pumping reuseirrigation method

side X side Yupstream head side

irrigation method full length half length 1/3 length

distribution system unlined canal lined canal layflat tubing PVC gated pipe

sub-units Group of alternative generation 0,5 lps < unitary inflowrate < 4 lps

Land levelling option

irrigation method level basins graded basins/borders graded furrows

continuous continuous continuous surgeinflow regime

diked open without reuse gravity reuse pumping reuseirrigation method

side X side Yupstream head side

irrigation method full length half length 1/3 length

distribution system unlined canal lined canal layflat tubing PVC gated pipe

sub-units Group of alternative generation 0,5 lps < unitary inflowrate < 4 lps

Fig.3.12. Design variables for the alternatives generation procedure. The generation of alternatives has to be made as described in the following:

(i) To select the irrigation method: flat level basin, graded basins or border, or graded furrows;

30

(ii) To choose the inflow supply regime, that for basins and borders is continuous constant flow, while for graded furrows it can be surge-flow;

(iii) To select the water distribution system among rigid pipe, lay-flat pipe, earth canal, or lined canal; if the surge-flow is chosen, then the control system is also to be selected between manual or automated control; and

(iv) To select the tail water management: for basins the option is diked, while for borders and furrows it may be diked, open without reuse, reuse with pumping, and gravity reuse.

Once the design options are selected, the programme generates the design alternatives as indicated in Fig.3.13.

Windows interface

Generate alternativesList of land levellingoptions

(file levelling.txt)

Field data

Design options

Execute

Savealternatives

if SIMDB is not complete

SIMDB

List of groupsand alternatives

Execute sirmod

Return

Database

Load simdb

Insert

irrigation methoddistribution systeminflow nad tailwater

Windows interface

Generate alternativesList of land levellingoptions

(file levelling.txt)

Field data

Design options

Execute

Savealternatives

if SIMDB is not complete

SIMDB

List of groupsand alternatives

Execute sirmod

Return

Database

Load simdb

Insert

irrigation methoddistribution systeminflow nad tailwater

Fig. 3.13. Flowchart of the alternatives generator module (Gonçalves et al., 2005a, b).

The user design options to generate alternatives for furrow irrigation are described as follows:

(i) Field outlets or hydrants: number per field and respective discharge and head; it is assumed that all outlets are identical and each one irrigates the same area, named unit;

(ii) Upstream supply side: side X or Y or both; (iii) Land levelling: cross and longitudinal field slopes; (iv) Length adjustment: full, 1/2 or 1/3, i.e. not adjusting or reducing the

length to the half or the third of the actual length (not used in this application);

(v) Distribution system: rigid pipe, lay-flat pipe, earth canal, or lined canal; (vi) Inflow supply regime: continuous or surge-flow; operated by an

automatic or a manual valve;

31

(vii) Tail-end flow management: diked, free drainage, or water reuse by pumping or gravity to downstream fields; and

(viii) Crops irrigation scheduling, with every furrow or alternate furrow irrigation.

The design variables are described in Table 3.10.

Table 3.10. Design variables

- distribution side (OX or OY) - field length (L) - transversal slope (SoT)

Topography (land levelling options)

- longitudinal slope (SoL) - number of outlets (No) - number of units (Nu) - total field supply discharge (QF) Supply system - outlet discharge (Qo) and head (Ho) - field delivery time (tF)

Distribution system - continuous or surge flow - canal, layflat tubing or rigid gated pipe - automatic or manual surge valves - no reuse - reuse by pumping Tail water management - gravity reuse on other fields

Operative - application time, per sub-unit (tap) - unit inflow rate (q) - number of sub-units per unit

3.4.2 Performance indicators, criteria and ranking with SADREG To analyze the results the following performance indicators are considered (Gonçalves et al., 2005a, b):

(i) Land productivity (LP) - it is the amount of yield (cotton crop) per unit area (kg ha-1);

(ii) Land economic productivity (LEP) (€/ha); (iii) Water productivity (WP) - expresses the yield per unit volume of

irrigation water (kg m-3) to qualify the irrigation performance relative to water when this factor is scarce and expansive;

(iv) Water economic productivity (WEP) (€/m3); (v) Total water cost (TWC) - is the total cost relative to irrigation (€ ha-1); (vi) Total water use (TWU) - it is the annual amount of irrigation water

applied per unit area (mm; m3 ha-1year-1); (vii) Salinization risk (SR) - it is the volume of water that deep percolates

(mm), meaning the potential for transporting salts to the groundwater; (viii) Beneficial water use ratio (BWUR) – ratio between the beneficial

water used and the total water use; (ix) Yield to cost ratio (kg/€); (x) Total cost to water use ratio (€/m3); (xi) Fixed cost to water use ratio (€/m3);

32

(xii) Variable cost to water use ratio (€/m3); (xiii) Runoff ratio; (xiv) Soil impacts of land levelling (excavation average depth, cm); (xv) Soil erosion index; (xvi) Global Utility (U) - it is the aggregated utility characteristic of

alternatives, being dependent of irrigation performance, criteria weights and utility functions;

(xvii) Distribution uniformity (DU) – defined in eq. (28); and (xviii) Application efficiency (Ea) – defined in eq. (26 and 27).

The impact analysis includes the crop yield estimation based upon the total water use during the irrigation season and adopting an user selected yield function relating the relative yield with the relative water application (Fig.3.14). Three functions are available: the quadratic one, with an adjusted decreasing branch where the user selects the parameter dw relative to the deviation relative to the quadratic function, and a fitting function where the user provides the decreasing branch through a table.

0

0.2

0.4

0.6

0.8

1

0 0.5 1 1.5 2 2.5 3 3.5 4relative supplied water (W/Wopt)

rela

tive

yie

ld (Y

/Yo

pt)

Quadratic function Adjusted decreasing branch Table function

Fig.3.14. Irrigation water-yield function (Gonçalves et al., 2005a, b). The multicriteria analysis applies linear utility functions for benefits, costs and environmental criteria; the weights for every criterion are user defined and the global utility value to rank alternatives is computed by a linear weighing method. The programme generates a large number of alternatives in consequence of combination of design variables; however, it is very difficult for user to view and analyzes, one by one, the existing alternatives on database. Thus, the multicriteria analysis module has a very important role on automatic management of large amount of data. It screens the alternatives, removing the not satisfactory and dominated alternatives, and selecting the most adequate one, by groups and by projects.

+dw

33

Chapter 4 4. Results and Discussion 4.1. Irrigation scheduling for wheat and cotton Two irrigation strategies have been simulated using the ISAREG model:

(i) Irrigation to maximize yields avoiding water stress except for the end crop period: the irrigation are applied when the average soil water content equals pθ (eq. 3.4) and the irrigation depths are

those required to refill the soil moisture to the field capacity; (ii) deficit irrigation: irrigation schedules were simulated by

progressively lowering the threshold θMAD

< θp

as referred by

Pereira et al. (2003). (iii) Computations were done by imposing restrictions on the irrigation

period and the available water for irrigation (net season irrigation depth) and allowing a limited water stress, i.e.

The results for the simulation of wheat irrigation scheduling are presented in Table 4.1. It was considered an application efficiency of Ea = 60%, a typical value for Arnan experimental farm under traditional irrigation. Table 4.1. Wheat irrigation scheduling simulations compared with the current wheat irrigation (assuming Ea=60%)

Available Water for Irrigation mm

Strategies Jan - Mar April - June

Total Irrigation water (mm)

Total rainfall (mm)

Not used rainfall (mm)

ETa (mm)*

RYD** %

A Variable depths & avoid stress

71.7 200.6 272.3 289.8 134 466.6 11.4

B*** 100% Өp 60 240 300 289.8 122 499 4.9

C*** 70% Өp 60 180 240 289.8 124.6 444.8 15.8

D*** 45% Өp 0 180 180 289.8 90.1 408.6 23.1

Curr current irrigation

65 160 225 289.8 111 409 23

NI Non irrigated 0 0 0 289.8 90 232 58.5

* ETm=523.6 mm; ** RYD = relative yield decrease (%) corresponding to Ky = 1.05 *** Constant net water depth In= 60 mm

34

In this table, for 2005-2006 weather conditions, the wheat water requirement is 524 mm for no stress and maximum yield in the Ras Al Ein. Adopting supplemental irrigation this is easy to achieve with a total rainfall 289.8 mm. The soil water dynamics relative to alternative strategies simulated with ISAREG and referred in Table 4.1 model are in Figs.4.1, with the current water saving schedule in Fig. 4.1 b.

15

17

19

21

23

25

27

29

31

WP(%) Soil Moisture OYT(%) FC(%)

15

17

19

21

23

25

27

29

1 11 21 31 41 51 61 71 81 91 101 111 121 131 141 151 161 171 181

WP(%) Soil Moisture OYT(%) FC(%) Figure 4.1.Simulated soil water dynamics for Alternative A and the current water saving schedule For the first strategy (Fig.4.1 a), not allowing stress except by the final stage of the crop, requires 4 applications with a total net depth of 272 mm but the relative yield decrease (RYD) is about 11.4%. Better results could be achieved with 5 irrigations (strategy B) with constant In=60 mm, totalling 300 mm, which reduces stress by the end stage and causes only about 5% RYD. Comparing with the current irrigation for water saving at Arnan it used 225 mm (net) producing a RYD of 23%. The difference in RYD highly relates to the fact that applications for the latter are delayed, so inducing stress. For the strategies with imposing restriction on the irrigation period, if 4 irrigations are applied (alternative C) it results a RYD of 15.8%, whereas with larger stress (alternative D) RYD = 23%, i.e. the same as for the current schedule that uses 225 mm. Large reduction of the yield, about 59%, was obtained for the rainfed crop. A comparison of ETa and the ETa/Etm ratio for the strategy A and for the rainfed conditions is presented in Fig.4.2. it shows that the crop length period is shorter for the rainfed one due to adopting early harvesting since the water stress induces early maturation. A summary of water consumptive use and relative yield deficits is shown in Fig. 4.3, shoeing that yields are reducing when consumptive use decreases.

35

0

1

2

3

4

5

6

7

8

9

10

1 12 23 34 45 56 67 78 89 100 111 122 133 144 155 166 177 188

Eta Eta/Etm (a)

0

1

2

3

4

5

6

1 11 21 31 41 51 61 71 81 91 101 111 121 131 141 151 161 171

Eta Eta/Etm

(b) Figure 4.2. Comparing ETa and the ETa/ETm ratio for the strategy A (a) and for the rainfed crop (b).

0

50

100

150

200

250

300

350

Max Y 100%Өp 70%Өp 45%Өp Present No irri

irriga

tion d

ep

th m

m

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

Yie

ld lo

ss

TW yield loss

Fig 4.3. Comparing the total net water applications and relative yield reductions for the simulated strategies with the rainfed crop Main results for the simulation of cotton irrigation scheduling are given in Table 4.2 concerning irrigations for maximum and deficit irrigation for ӨMAD = 0.7 Өp. A fixed net irrigation depth In =80mm is adopted. Table 4.2. Cotton irrigation schedule simulation for surface irrigation

Surface Irrigation for the cotton In= 80 mm , ETm= 922.3 mm

Irrigation water amount mm

May June July Aug Sep

Total Irrigation

water

ETa mm

RYL %

N of the irrigatio

n

Max Y 109.1 163.8 299.5 243.7 94.5 910.6 922 0 11 100% Өp

80 280 240 240 160 960 921.9 0 12

70% Өp 80 160 320 160 160 880 881.3 3.8 11

36

The total water requirement for cotton is 922 mm. In the strategy MaxY variable depths are adopted which are close to 80 mm except for the first and the last irrigation, with a time interval between irrigations averaging 7.6 days (Fig.4.4). The actual evapotranspiration attains its maximum by the end of July (Fig .4.5).

15

17

19

21

23

25

27

29

31

1 10 19 28 37 46 55 64 73 82 91 100 109 118 127 136 145 154 163 172

WP(%) Soil Moisture OYT(%) FC(%)

Fig 4.4. The simulated soil water dynamics when MAD=100% Өp and variable irrigation depths are adopted (MaxY)

0

2

4

6

8

10

12

14

1 11 21 31 41 51 61 71 81 91 101 111 121 131 141 151 161 171

Eta Eta/Etm

Fig 4.5. Time evolution of daily Eta, and Eta/ETm ratio during the cotton season for the case MaxY. For cotton the farmer applied 10 irrigation with more than 200 mm per each one, whereas in the schedules analysed much less are required (Fig. 4.6). In the simulations the net depth was 80 mm which corresponds to a gross depth of 115 mm assuming an application efficiency of 70%, which is achievable as indicated by results in Section 4.3, then avoiding water wastes as deep percolation. Improved schedules could be selected early in the crop season, proposed to farmers planning irrigation in advance, and then periodically

37

adjusted to the actual climate conditions. This option could lead to appreciable water savings.

0

200

400

600

800

1000

1200

Max Y 100%Өp_80 70%Өp_80 100%Өp_25 70%Өp_25

Irri d

epth

mm

0

0,02

0,04

0,06

0,08

0,1

0,12

0,14

0,16

0,18

yield

loss

TW yield loss

Fig.4.6. Comparing consumptive water use and relative yield losses for cotton relative to several simulated strategies 4.2. Field Irrigation Evaluations 4.2.1 Infiltration and hydraulic parameters Main results or the infiltration parameters and the hydraulics roughness parameter n are presented in Table 4.3. For Arnan, n varies from 0.033 m-1/3s as an average in the first irrigation to 0.073 m-1/3 s for the second irrigation. This variation is related to the weeding operation performed one day before the second irrigation. For Bab Al Faraj it ranged from 0.011 m-1/3s (curve B1.3) for long graded furrows to increase to 0.022 m-1/3s for less slope and shorter furrows (curve B1.3). A great variability of infiltration parameters have been observed from field measurements based on the advance and recession times. Groups (families) of infiltration cumulative curves (Table.4.3) were created relative to each of the two irrigations for different sites representing low, medium and high soil infiltration rates. The corresponding infiltration parameters were later used for the simulations with SIRMOD to evaluate the irrigation systems performance for each treatment, and then to use the average of each irrigation event for design improved solutions. The variability of the average values of the parameters k and a is small for the 1st and 2nd irrigation in Arnan 2006. Taking curves R3.1 and R3.2 in Table 4.3 it shows that the permeability is higher for the first irrigation than for the second one, but both curves represent medium infiltration rate soils. The cumulated infiltration curves obtained for Arnan and Bab el Faraj are shown in Fig.4.7 and 4.8 respectively together with others existing in the database of SADREG.

38

Table 4.3. Infiltration and hydraulic roughness parameters observed at Arnan and Bab Al Faraj Area Curve n° a K m/min^a f 0 m/min n

R1.1 0.191 0.0143 0.00026 0.059 R1.2 0.071 0.028 0.000226 0.0329 R1.3 0.081 0.0395 0.000162 0.038 R1.4 0.279 0.0059 0.0002 0.0023 R2.1 0.1556 0.011 0.000336 0.0338 R2.2 0.1637 0.02285 0.000103 0.0875 R2.3 0.3078 0.0046 0.00024 0.097 R1 average 2005

0.3302 0.017 0.00015 non

R3.1 average 1st irrigation 2006

0.1175 0.0299 0.00021 0.033

R3.2 average 2nd irrigation 2006

0.1588 0.0132 0.000225 0.0728

Arnan Experimental farm (field E)

R double ring 2006

0.3223 0.011797 0 non

B1 0.3178 0.01015 0.00037 0.011 B2 0.2352 0.0163 0.00028 0.0233 B3 0.2411 0.0136 0.00013 0.0222 R2 average 2005

0.4304 0.0071 0.0001 non

Bab Al Faraj

R4 average 2006

0.273 0.015 0.00026 0.0188

Arnan infiltration curves 2005-2006

0

0,05

0,1

0,15

0,2

0,25

0,3

0,35

0,4

0 50 100 150 200 250 300 350 400time (min)

infil

tra

tion

(m

)

SCS-CN field B (double ring)2006 avarage 1(R3) doubl ring 2005 avarage 2 mean 2005(R1)

Fig 4.7. Infiltration curves observed at Arnan (2005-2006) and those from the SADREG database

39

Bab_Al_Faraj infiltration curves 2005-2006

0

0,05

0,1

0,15

0,2

0,25

0,3

0,35

0,4

0 50 100 150 200 250 300 350 400

time (min)

infil

tratio

n (m

)

SCS-CN Avarage 2005(R2) B1.1 B1.2 B1.3 Avarage 2006

Fig 4.8. Infiltration curves observed at Bab Al Faraj (2005-2006) and those from the SADREG database 4.2.1 Irrigation performances Observed advance and recession times differ among treatments as a function of discharge, furrow length and slope; these treatments were simulated with SIRMOD model to match simulations with the observed values in order to evaluate the application of the SIRMOD model in the field to improve the irrigation performance and to select the appropriated solution. For example, for the treatments B1.1 and B3.1 having the same length and slope but different discharges 0.69 and 0.35 l/s respectively, the advance time for the first irrigation is 130 min for B1.1 and 274 min for B3.1. The recession time for both is nearly liner (Fig.4.9 a, b).

th adv & rec time observed and simulated one treatment B1,1

0

50

100

150

200

250

300

0 20 40 60 80 100 120Distance along the furrow (m)

Tim

e (m

inut

es)

obs adv time simulted adv time t res obs siumlated res time

(a)

th adv & rec time observed and simulated one

0

50

100

150

200

250

300

350

0 20 40 60 80 100 120Distance along the furrow (m)

time

min

simulted data observed data res observed rec simulited

(b)

Fig. 4.9. Observed & simulated advance and recession times for treatments B1.1 (a) and B3.1 (b)

40

Results in Fig. 4.9. show therefore a better performance for B1.1 than for B3.1. because the infiltration opportunity time for the second is much small by the end of the second than for the first. Worst results are shown in Fig 4.10 where for B1.2 recession is affected by ponding and in B3.2 where advance is not completed.

Adv & Rec time B1.2

0

50

100

150

200

250

300

350

0 20 40 60 80 100

time

min

adv time min rec time min simulated Adv time simulted rec time

(a)

th adv & rec time observed and simulated one B3.2

0

50

100

150

200

250

300

350

400

450

0 20 40 60 80 100 120

Tim

e (m

in

simulted data observed data res observed rec simulited (b)

Fig 4.10. Observed and simulated advance and recession time for treatment B1.2 and B3.2 The characteristics and performances of the first irrigation for are presented in Table.4.4 a) for long furrows (100 – 150 m) and Table 4.4 b) for short furrows (50, 36 m) with different discharges for each length. The infiltration depths were calculated with an infiltration function whose parameters were obtained from advance data. To allow that the measured inflow volume be equal to intake and runoff, the coefficient K was adjusted by a trial and error procedure. Table 4.4.a). Characteristics and performance of the first irrigation for long furrows (100-150m)

F.L 150mFurrow length 100 m

Inflow rate (l/s) 0.53 0.3-0.4 0.5-0.6 ~ 0.7

A B3.2 B3.1 B2.1 B2.2 B1.2 B1.1 T. IP

0.53 0.32 0.35 0.52 0.58 0.75 0.77 0.7 qin (l/s)

0.06 0 0.1 0.134 0 0 0 0.3 qout (l/s)

381.2 400 293 294 185 180 180 212 tco (min)

11.73 7.7 6.15 8.59 6.4 8.06 8.32 8.9 Tot V (m3)

11.5 7.7 5.96 8.64 6.4 8.06 8.32 7.62 Vin (m3)

80 80 80 80 80 80 80 80 Zreq (mm)

114.8 107 87.9 131 91 115 118 126.7 D (mm)

45 18.7 43 74.4 65.7 89 95.5 76.4 Zlq(mm)

106.3 107 59 121.2 91 115 118 106.9 Zavg (mm)

39 17.5 55.5 57 72.2 77.4 80.9 60 Ea (%) 42.21 17.5 58.1 61.3 72.2 77.4 80.9 71 DU (%)

R1.4 R1.4 adopt R1.3 R1.2 Average1 Average1 Average1 R1.1 Inf. Curve

41

Table 4.4.b. Characteristics and performance of the first irrigation for short furrows, 50-36m

In the first table are included open and diked furrow for 100m length, with some of of them having deficit irrigation in the last quarter of the field, i.e. Zlq < Zreq. The indicators show that diked furrows have better irrigation performance than the open ones, with Ea and DU > 77% (Fig.4.12). Open furrows also have runoff 18% as B1.1, and in the B2.2, B3.2 with different discharge 0.58 and 0.32 l/s respectively, by applying less amount of the water B2.2 than B3.2, the result indict for extremely different in the Ea and DU 72% for B2.2 to be about 17 % in D3.2 (Fig.4.12) these Ea and DU would be higher by increase the cutoff time but the different in the performance between the treatment will be the same.

0

20

40

60

80

100

120

140

B1.2 B1.2 B1.1 B2.1 B2.2 B3.2 B3.1

dept

h m

m

0

1

2

3

4

5

6

7

8

9

10

volu

me

m3/

furr

ow

Zr mm D mm Zlq Zavg Total V

Figure 4.11.The characteristics of the first irrigation for furrows 100 m long.

36 m 50 m Furrow length

D3 D2 D1 C3 C2 C1 F n°

0.32 0.5 0.75 0.35 0.52 0.78 qin (l/s)

0 0 0 0 0 0 qout (l/s)

139 92 30 195 204 102 tco (min)

2.7 2.76 1.38 4.1 6.36 4.77 Total V (m3)

2.7 2.76 1.38 4.1 6.36 4.77 Vin (m3)

80 80 80 80 80 80 Zreq (mm)

106 110 55 117 181 136 D (mm)

102 102 51 96 162.6 130 Zlq(mm)

106 110 55 117 181 136 Zavg(mm)

75.45 72.7 92.7 68.4 44 59 Ea (%)

96.5 93 92.7 82 90 95.5 DU (%)

Average1 Average1 Average1 Average1 Average1 Average1 Infilt. curve

- - Deficit irrig - Overirrig Overirrig. Note

42

0

10

20

30

40

50

60

70

80

90

%

B1.2 B1.2 B1.1 B2.1 B2.2 B3.2 B3.1

Ea DU

Figure 4.12.The performance of the first irrigation for furrows 100 m long.

As for the characteristics in Table.4.4 b, the treatments C1 and C2 were over-irrigated with water logging at the end of the furrow. D1 was in deficit irrigation, so Zavg < Zreq, due to the large discharge and short length even thought it has the highest Ea and DU but with storage efficiency < 70%. The table indicates that treatments D2 and D3 have about the same result (Table.4.4 b) with higher application time in D3, 139 min, compared with D2, 92 min. When comparing the treatments B3.2, D3, and C3 (Fig.4.13), which have different lengths, respectively 100, 50 and 36 m and about the same q of 0.32-0.35 l/s, where nearly the same water amount and application times per unit distance was applied, the poorest irrigation performance is for B3.2, and D3 has very high Ea and DU > 96%.

0

20

40

60

80

100

120

140

B3.2 C3 D3

mm

0

20

40

60

80

100

120

%

Zr mm D mm Zlq Ea DU

Figure 4.13.the characteristics and the performance indicators for B3.2, C3, D3.

43

The characteristics and performances of the second irrigation for 100m, and alternate furrow F for 89 m are shown in Table.4.5. The Table.4.5 shows that treatment E1, with the same length as E3 but high discharge and small tco, has the best performance, much better than F (alternate furrows). Treatment E3 with q=0.37 l/s has also good Ea and DU but Zavg > Zreq because it uses a higher water amount. Table.4.5.Characteristics and performance of the second irrigation and traditional irrigation (zigzag basin) for the first irrigation

Characteristics of E1 are shown in Fig.4.14 and refer to observations (a) and to simulations when using advance and recession or advance data only (b). results show that much better simulation results are obtained advance and recession data are used, which allows to take into consideration the irregularity of the slope. The case for the traditional zigzag furrow within a basin 50 m long and 5 m wide where it was applied a large inflow rate of 4.78 l/s provided a good advance time; the farmer was looking for 150 mm required depth. With this application the system produced very high deep percolation and Ea < 60% (Fig.4.15); the infiltrated depth is 300mm at the upstream end, which denotes a very much non-uniform condition and a cause for low application efficiency.

Zigzag basin L=265, FS=0.9m(1st irri)

89 m, FS=1.4m 100 m, FS=0.7m Furrow length

A0 F E3 E1 F n°

4.78 0.36 0.37 0.74 qin

0 0 0 0 qout

214 527 400 124 tco

63.56 11.4 8.83 5.51 Total V

63.56 11.4 8.83 5.51 Vin

150 80 80 80 Zreq mm

260 91 126 79 D mm

198 19 93.3 60 Zlq

260 91 126 79 Zavg

58 21 63.5 76 Ea

75.3 21 74.1 76 DU

Rz R2.3 R2.2 R2.1 Inf. Curve

44

-150

-100

-50

0

50

100

150

200

250

0 20 40 60 80 100

dept

h m

m

T

ime

(min

)

Advance time rec time min infilatrated depth required depth a)

-200

-100

0

100

200

300

400

0 20 40 60 80 100

depth

mm

tim

e m

in

adv time min rec time min simulated Adv time simulted rec time

simu infli depth requier depth mm simu Z

b) Figure 4.14 Furrow E1, second irrigation: a) Observed advance and recession time, and required and infiltrated depths; b) simulated advance and recession time, and infiltrated depth considering advance and recession or only advance data

-400

-300

-200

-100

0

100

200

300

400

0 30 60 90 120 150 180 210 240 270 300

Zigzag length (m)

dept

h m

m

T

ime

(min

utes

)

adv_time(min) rec_time(min) Zapplied Zreq

Figure 4.15. The advance and recession time, and required and infiltrated depths for the zigzag furrow

45

The irrigation in Bab Al Faraj was evaluated in several locations having different lengths and slopes. The discharge is controlled by the water head in the supply ditch and the diameter of the pipes. The actual performance is shown in Table 4.6. Cutoff is applied when the advance time is complete, so because furrows are very long it is also much longer than usual. Discharges are also high. Treatments with small discharge have higher Ea and DU (Figure 4.16). Table 4.6.Characteristics and performance in Bab Al Faraj

0

50

100

150

200

250

G H L

dept

h m

m

0

10

20

30

40

50

60

70

80

%

Zr mm D mm Zlq Ea

Figure 4.16. Characteristics and performance for Bab Al Faraj irrigations This would be explain better by the simulation with SIRMOD model that was done for 100 m length, zero and 0.01 m/m slope and different inflow rates (0.2, 0.4, 0.6, 0.8 l/s) using the average infiltration parameters of the R2 curve (Fig.4.17). For the slope 0.01m/m it is difficult to apply > 0.6 l/s that has

100 m, FS=0.7m Slope=0.7%

90 m, FS=0.7m Slope=0.97%

L= 224 m, FS=0.7m Slope=1.9%

Furrow length

I H G F n° 0.36 0.52 1.3 qin 268 347 411 tco 5.9 11.45 33.2 Total V 5.9 11.45 33.2 Vin 100 100 100 Zreq mm 84 170 205 D mm

57.6 60.4 63 Zlq 84 170 205 Zavg

68.5 35.5 31.1 Ea

68.5 35.5 31.1 DU B1 B1 B1 Infilt. Curve

46

better Ea and DU than with 0.4 l/s with less amount of water. For zero slope, as the inflow rate increases until 1.2 l/s Ea also increases with gradually less water application.

0

10

20

30

40

50

60

70

80

90

0,4 0,6 0,8 1 1,2

0

2

4

6

8

10

12

14

16

Ea S0

Ea S 0,01

TW S 0

TW S 0,01

Figure 4.17. Simulated characteristics and application efficiency for 100 m in Bab Al Faraj for 0.01 m/m and zero slope as a function of the inflow rate. Water saving may be achieved when irrigation performances such as the distribution uniformity (DU) and the application efficiency (Ea) are improved. DU and Ea depend upon a large number of factors such as the unit inflow rate, the hydraulics roughness, the intake characteristics of the soil, the cross-sectional characteristics of the furrow, cut off time and the longitudinal slope of the furrows. In addition, Ea depends on the soil water deficit at time of irrigation (Pereira, 1999; Pereira and Trout, 1999). However, attention must be given to land levelling conditions since these play a major role for achieving uniform (Pereira et al., 2002b). Therefore, the factors by which a farmer may manage a system in order to improve the distribution uniformity and the application efficiency may be expressed by simplified functional relationships (Pereira and Trout, 1999) such as:

),,(

),(

SMDtqfEa

tqfDU

coin

coin

=

=

For the actual zigzag furrow irrigation system, the available discharge is divided into several strips that are irrigated simultaneously; the discharge flowing in each strip is generally small, approaching 5 l/s. Because the land is not leveled and the soils intake rate is low, the crop beds are made across the width, thus obliging the water to follow a zigzag that slows down the velocity and produces more time for infiltration, Irrigation cutoff is practiced when the water arrives to the downstream end or to some 5-10m of it. The system does not require intensive labour but only little manpower to deliver the water into the strip basins, and later to cut it off and start the irrigation of

47

the next basin/border set by Pereira (2006). The system with small and large discharge has low Ea and DU, wasting quite amount of the water as deep percolation; it may be improved by reducing the basin size or adopting a new approach to the basin or border with regular geometry and well levelled lands (Pereira, 2006). Based from the experiments and the simulation above, for Arnan it is recommended to have diked furrowed basins with 100 m length and not more than 0.75 l/s inflow rate for the first irrigation and between 0.6 and 0.4l/s for the second irrigation. For Bab Al Faraj, because long furrows have very low efficiency and some part of the field is water logged due to the irregularity of the slope, based on favourable Ea and DU results for less long fields, it is advisable to adopt 100 m length and relative small discharge up to 1.0 l/s. 4.3 Alternatives for improving surface irrigation. SADREG application 4.3.1 Projects: build up improved irrigation scenarios The projects considered improving surface irrigation in are described in Table 4.7 and were built using the information obtained from field observations and model analysis described in the preceding sections. Different irrigation methods are compared: for cotton, graded furrows (P1) and graded borders (P2) are considered, and for wheat, in addition to graded borders (P3) also zero-level furrowed and flat basins (P4 and P5) are studied. For all projects (P1 to P5) improved land preparation techniques are considered, including a frequent land smoothing operation, according to the irrigation method. Table 4.7 - Project characteristics for improved systems project method crop soil surface condition Zreq

(mm) Nbr. irrigations

P1 graded furrow cotton furrows, with 0.70 m spacing

80 12

P2 graded border cotton flat soil surface 80 12 P3 graded border wheat flat soil surface 60 5 P4 level furrowed

basin wheat furrows, with 1.0 m spacing 60 5

P5 level flat basin wheat flat soil surface 60 5 The projects were applied in several irrigated fields, defined according the typical characteristics of Ras El Ain area, namely the dimensions, the topography and the soil intake and water storage characteristics. The workspaces considered and the correspondent projects are described in Table 4.8. It is assumed that the actual field slopes are kept when an improved scenario is considered for that field; thus, no slope changes are required but only land smoothing to produce an uniform slope. This assumption allow the comparison of several alternative projects without

48

including land leveling costs, which would be very significant, but only land smoothing costs. Table 4.8 - Workspaces characteristics and applied projects Length Slope=0 Slope=0.5% Slope=0.8% Slope=1.0% Slope=1.5% L=50m w1.0

Arnan P4, P5

w1.1 Arnan P1,P2,P3

w1.2 Arnan P1,P2,P3

w1.3 Bab Al Faraj P1


L=100m w2.0 Arnan P4, P5

w2.1 Arnan P1, P2,P3

w2.2 Arnan P1, P2,P3



L=150m w3.3 Bab Al Faraj P1


L=200m w4.3 Bab Al Faraj P1


Note: w is for identification of workspaces and P for projects The workspaces (w) in Table 4.8 allow the analysis of the following factors:

(i) The field length (50 and 100m) for level basins for wheat, in Arnan area;

(ii) the longitudinal slope (0.5 and 0.8%) for graded furrows and graded borders, for cotton and wheat, in Arnan area;

(iii) the field length (from 50 to 200m) for graded furrows, in cotton irrigation, for Bab Al Faraj area,

(iv) the longitudinal slope (1 and 1.5%) for graded furrows in cotton irrigation, for Bab Al Faraj area.

The input data were obtained from field observations and published data relative to other study cases. The Infiltration curves used were obtained from those analyzed in the precedent Section 4.2 and were obtained with the double ring infiltrometer and with furrow and basin observations of advance and recession, and parameters were optimized through using SIRMOD with the reverse method, i.e. comparing the advance and recession curves simulated with those observed in the field. (Katopodes et al., 1990; Calejo et al., 1998). Thus, the average infiltration curve for Arnan, called raselain3, is

ττ ⋅+⋅= 00021.00229.0 1175.0Z and for Bab Al Faraj, called raselain2, is

ττ ⋅+⋅= 00010.00071.0 4304.0Z . where units for Z are m3/m and for τ are min. The Figure 4.18 shows the input form relative to field forms, including the size, slopes and infiltration curves. The Figure 4.19 refers to crop input data, for cotton and wheat. It shows that a soil-water yield function was applied as

49

described in last item of Section 3. The Figure 4.20 refers to the unitary cost sand financial data. The tasks duration data, for manpower calculation and for land levelling maintenance are presented in Figure 4.21.

a)

b) Figure 4.18 – SADREG window for field size, slopes (a) and infiltration curves (b)

50

Figure 4.19 – SADREG window for input of used crop data on cotton and wheat

Figure 4.20 – SADREG window showing the used unitary cost and financial data

51

a)

b) Figure 4.21 - Tasks duration data; a) irrigation; b) land leveling maintenance 4.3.2 Alternatives, selection and ranking Each project, for any applied workspace, produces a set of alternatives, which are organized by groups relative to the different farm water distribution equipments and tail end management options previously considered. Table 4.9 shows an alternative layout, with a detailed description of alternative characteristics and attributes. Figure 4.22 shows the "tree" structure to access alternatives’ data.

52

Table 4.9 - Example of a project alternative layout ALTERNATIVE IDENTIFICATIONnumber workspace project group Nr.sub-units7 w2.1 P3 3 9DELIVERY DATAnumber_outlets outlet (lps) number_field_splitsoutlet_head(m)1 10 1 1irrigation_event Nr.sub-units app_time(full length)App_eff. DeepPerc. Runoff1 9 82.4 67.2 4.7 28.12 9 82.4 67.2 4.7 28.13 9 82.4 67.2 4.7 28.14 9 82.4 67.2 4.7 28.15 9 82.4 67.2 4.7 28.1PERFORMANCE ATTRIBUTESTWU (m^3/ha) LP (kg/ha) LEP (EUR/ha) WP (kg/m^3) WEP (EUR/m^3) BW (ratio) Y/TIC (ratio)TIC/TWU (EUR/m^3)FIC/TWU (EUR/m^3)VIC/TWU (EUR/m^3)4448.48 5499.98 1188 1.23637 0.267057 0.672 3.43472 0.0777522 0.00862779 0.0691244IRRIGATION METHODSupply Side irrigat.Method Inflow Regime Tail end manag.X graded border continuous open,without reuseFIELD UNIT CHARACTERISTICSUnit Length (m) Unit Width (m) Longit.Slope (%) Cross Slope (%) furrow Spacing (m)100 50 0.5 0 1LAND LEVELLINGvolumes operated(m^3) avg.cut(cm) avg.fill (cm) max.cut (cm) max.fill (cm) initial operation (h) maintenance (h)Initial (EUR) Maintenace(EUR)

0 0 0 0 0 0 1.5 0 55.055IRRIGATION SUB-UNIT DATAirrigat.event Nr.sub-units inflow (lps/m) Field Appl. volume (m^3)sub-unit width (m) manpower duration(h)1 9 1.80 444.8 5.6 12.82 9 1.80 444.8 5.6 12.43 9 1.80 444.8 5.6 12.44 9 1.80 444.8 5.6 12.45 9 1.80 444.8 5.6 12.4PERFORMANCE INDICATORS (%)irrigat.event App.Effic. DeepPerc. Runoff Dist.Unif.1 67.2 4.7 28.1 93.42 67.2 4.7 28.1 93.43 67.2 4.7 28.1 93.44 67.2 4.7 28.1 93.45 67.2 4.7 28.1 93.4IRRIGATION TIMES (min)irrigat.event adv[1/4] adv[2/4] adv[3/4] adv[4/4] cutoff rec[1/4] rec[2/4] rec[3/4] rec[4/4]1 12.4 28.6 47.2 68.2 82.3792 102.4 117.1 123.4 136.52 12.4 28.6 47.2 68.2 82.3792 102.4 117.1 123.4 136.53 12.4 28.6 47.2 68.2 82.3792 102.4 117.1 123.4 136.54 12.4 28.6 47.2 68.2 82.3792 102.4 117.1 123.4 136.55 12.4 28.6 47.2 68.2 82.3792 102.4 117.1 123.4 136.5INFILTRATED DEPTHS (mm)irrigat. events target depth (mm)Appl.Depth (mm) ARZS (mm) Z[0] Z[1/4] Z[2/4] Z[3/4] Z[4/4]1 60 88.9695 59.7875 65.3 66.3 65.8 61.3 58.32 60 88.9695 59.7875 65.3 66.3 65.8 61.3 58.33 60 88.9695 59.7875 65.3 66.3 65.8 61.3 58.34 60 88.9695 59.7875 65.3 66.3 65.8 61.3 58.35 60 88.9695 59.7875 65.3 66.3 65.8 61.3 58.3FIELD DISTRIBUTION SYSTEMSupply systemTypePVC gated pipe 160mmFIELD REUSE SYSTEMpumping reuse (m^3) pumping Cost (EUR)

0 0INVESTMENT COST (present worth, EUR/field)Initial land levelingDistribution system (including 10% for equipment maintenance and repair)

0 141.5ANNUAL COST (EUR/field/year)Maintenance land leveling Water Labor Investment Operation Total

55.055 48.9332 49.7608 19.1903 153.749 172.939

53

Figure 4.22 – SADREG Window showing the tree to explore alternatives The criteria applied for selection of alternatives are presented in Table 4.10. The weights assigned to the various attributes in this Table represent a balance on priorities for water saving, minimizing costs and maximizing economical benefits. Table 4.10 - Criteria and weights for alternative selection Criteria Attributes Weights

Land Economic Produtivity

15

Benefits Water Produtivity 15

Water Economic Produtivity

15

Beneficial Water Use ratio

15

Cost Fixed Cost per Water Use

10

Variable Cost per Water Use

10

Environmental impacts Total Water Use 20

4.3.3 Projects comparison and discussion The alternatives produced for each project were ranked from multicriteria analysis, and the best alternative for each project was adopted to represent that project for comparison purposes. The utility concept is an integrative

54

score that expresses its global worth and is used to rank the alternatives and the groups. The utility values allow the project comparison, when they apply to the same crop. It must be clear that the cropping costs relative to non-irrigation operations are not considered in this analysis because they are invariant, the same, for all projects under comparison. A deeper analysis to compare the different projects is based on the performance indicators shown in Table 4.11. Table 4.11 - Indicators applied for projects’ comparison Indicator symbol unit Total water use TWU m3/ha Water productivity WP kg/m3 Beneficial water use BWU ratio Yield/total irrigation cost Y/TIC ratio Total irrigation cost / total water use TIC/TWU EUR/m3 Irrigation operation cost IOC EUR/ha The cotton irrigation utility analysis relative to Arnan (Figure 4.23) let conclude that furrow irrigation systems are quite sensitive to slope (0.5 and 0.8%) and length (50 and 100m) and that these two factors are inter-dependent: when the slope increases the length must also increase. Contrarily, for Bab Al Faraj (Figure 4.24), for furrow slopes 1.0% to 1.5%, we conclude that furrow lengths do not influence significantly the system utility. The graded border irrigation for cotton at Arnan has an utility close to furrow systems; the best results are obtained with a length of 50m and a slope of 0.5%, or a length of 100m and a slope of 0.8% (Figure 4.23). This result is very important because border irrigation enlarges the framework for improved surface irrigation solutions and allows a system for cotton compatible with wheat irrigation (Figure 4.25) relative to land preparation. Note that in this analysis changing the actual field slopes was not considered but this cannot be ignored when the system design takes place in a particular farmer field.

55

furrows S=0.5%furrows S=0.8%

border S=0.5%border S=0.8%

L=50m

L=100m

0.68

0.75

0.680.710.75

0.680.72

0.67

0.5

0.55

0.6

0.65

0.7

0.75

0.8

0.85

0.9

0.95

1

Figure 4.23 Utilities of cotton irrigation projects, in Arnan

L=50mL=100m

L=150mL=200m

S=1.0%

S=1.5%

0.670.67 0.68

0.68

0.670.67 0.68

0.68

0.5

0.55

0.6

0.65

0.7

0.75

0.8

0.85

0.9

0.95

1

Figure 4.24 Utilities of cotton irrigation projects, in Bab Al Faraj

56

Level furrowedbasin Level flat basin

graded borderS=0.5% graded border

S=0.8%

L=50m

L=100m

0.88

0.68

0.74

0.75

0.88

0.77 0.80 0.81

0.5

0.55

0.6

0.65

0.7

0.75

0.8

0.85

0.9

0.95

1

Figure 4.25 Utilities of wheat irrigation projects, in Arnan The indicators for graded furrows and borders for cotton irrigation in Arnan (Figure 4.26) let to conclude that the projects applied allow achieving acceptable scores on beneficial water use and irrigation cost indicators, thus becoming potential solutions to improve Ras El Ain surface irrigation systems. Higher performances are obtained when furrows or borders have a combination slope/length of 0.5% - 50m or 0.8% - 100m. The operative cost of border system is higher than furrows due to a major requirement of land smoothing operation. An additional cost would be considered for border systems if it would be necessary to adjust the cross-land slope to zero through land leveling operation.

57



L=50m

L=100m0

2000

4000

6000

8000

10000

12000

14000

16000

TOTAL WATER USE (m3/ha)



L=50m

L=100m0

0.1

0.2

0.3

0.4

0.5

0.6

WATER PRODUCTIVITY (kg/m3)



L=50m

L=100m0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Beneficial water ratio



L=50m

L=100m0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

Yield / Total irrigation cost (ratio)



L=50m

L=100m0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

Total irrigation cost / Total water use (EUR/m3)



L=50m

L=100m0

100

200

300

400

500

600

700

800

Irrigation Operation cost (EUR/ha)

Figure 4.26 – Performance indicators for graded furrows and borders for cotton irrigation in Arnan For Bab Al Faraj, results in Figure 4.27 allow to conclude that for long furrows (150-200m) the system performance increases when reducing the total water use and global costs. On other hand, results are not sensitive to the slope in

58

the range 1.0-1.5%. However long furrows would require a high investment on land leveling but are highly compatible with crop mechanization.

L=50mL=100m

L=150mL=200m

S=1.0%

S=1.5%13400

13500

13600

13700

13800

13900

14000

14100

14200

14300

14400

Total water use (m3/ha)

L=50mL=100m

L=150mL=200m

S=1.0%

S=1.5%0.41

0.415

0.42

0.425

0.43

0.435

0.44

Water productivity (kg/m3)

L=50mL=100m

L=150mL=200m

S=1.0%

S=1.5%0.645

0.65

0.655

0.66

0.665

0.67

0.675

0.68

0.685

0.69

0.695

Beneficial water use ratio

L=50mL=100m

L=150mL=200m

S=1.0%

S=1.5%3.5

3.6

3.7

3.8

3.9

4

4.1

4.2

4.3

Yield / total irrigation cost (ratio)

L=50mL=100m

L=150mL=200m

S=1.0%

S=1.5%0.042

0.043

0.044

0.045

0.046

0.047

0.048

0.049

Total irrigation cost / total water use (EUR/m3)

L=50mL=100m

L=150mL=200m

S=1.0%

S=1.5%580

585

590

595

600

605

610

615

620

Irrigation operation cost (EUR/ha)

Figure 4.27 - Performance indicators for graded furrows for cotton irrigation in Bab Al Faraj

59

The indicators in Figure 4.28 let know that the furrowed level basins produce the best beneficial water use ratio and the best water productivity but the graded borders have a lower irrigation cost. The flat level basins have the worst performances, however it is also an acceptable solution. It requires more land smoothing cost but tillage practices are easier as it does not require furrows.



S=0.8%

L=50m

L=100m0

1000

2000

3000

4000

5000

6000

Total water use (m3/ha)



S=0.8%

L=50m

L=100m0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

Water productivity (kg/m3)



S=0.8%

L=50m

L=100m0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Water Beneficial ratio



S=0.8%

L=50m

L=100m0

0.5

1

1.5

2

2.5

3

3.5

Yield / total irrigation cost (ratio)



S=0.8%

L=50m

L=100m0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

Total irrigation cost / total water use (EUR/m3)



S=0.8%

L=50m

L=100m0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

Irrigation operation cost (EUR/ha)

Figure 4.28 – Performances of basin and borders for wheat irrigation, Arnan

60

The wheat irrigation utility analysis for Arnan (Figure 4.25) let conclude that level furrowed basins show the best performance, while the level flat basin has the worst result. Border irrigation, with 0.5 and 0.8% slope, shows an interesting option, in particular for sloped fields, requiring a minimum land leveling impact. 4.3.4 Conclusive remarks The SADREG application demonstrates the usefulness of this DSS tool to support the planning and design process of surface irrigation improvement. It creates a friendly framework to build up improved irrigation scenarios, considering the main factors that determine the system performance, and decision-making selection of designed alternatives by application of multicriteria methodology. The DSS integrates different sources of knowledge applied to prepare the input data, to build up the projects and to evaluate and alternative selection. The parameterization of infiltration was carried out based on field observations and optimization with SIRMOD model, which allows the confidence on DSS results. However, a deeper survey of Ras El Ain soil characteristics, including more observation sites and a sensitivity analysis for infiltration data, is determinant to have results that are more consistent at a large scale (regional level). Regarding cotton irrigation in Arnan, results show that furrow and border irrigation have high potential for improvement because they achieve good beneficial water use ratio and acceptable irrigation costs. Furthermore, in Bab Al Faraj the long graded furrow systems reveal a high potential to improve irrigation performance in sloped areas. Wheat irrigation in Arnan has several feasible options, having the level furrowed basin the best performance for flat fields and the graded borders for sloping fields, being also the flat level basin an acceptable solution. A further DSS application would contribute to a deeper knowledge about the surface irrigation in Ras El Ain relative to land leveling requirements to prepare the fields for improved surface irrigation, the conditions to field water supply, relatively to discharge, head and unitary cost, and the evaluation of the economical feasibility of water runoff reuse techniques.

61

5. Conclusions From the irrigation scheduling study for wheat and cotton applying the ISAREG model and observed field data, it could be concluded that the farmers apply less appropriate schedules: (i) for cotton, they use excessive gross irrigation depths (160-200mm) much

above the soil water deficits at time of irrigation, thus leading to high percolation; and

(ii) for wheat, they apply excessive water in the first irrigation and delay the second and third applications causing crop water stress.

Therefore, the following schedules may be considered obtained: (i) to maximize yields, 5 irrigation events of 60 mm (net irrigation depths,

NID) for wheat and 12 irrigation events for cotton with NID = 80 mm; (ii) for deficit irrigation, allowing 30% depletion below the soil water threshold

θp for no stress, i.e. assuming θMAD = 70% θp, 4 irrigation events of NID = 60 mm for wheat and 11 irrigation events for cotton with NID = 80 mm; and

(iii) for wheat, in case of high deficit irrigation, θMAD = 45% θp , thus 3 irrigation events of NID = 60 mm.

To better base improvements in irrigation scheduling advice it is required that further characterization of soil water properties be performed in the next future which could improved the adherence of model use to farming practices. The traditional zigzag furrowed basins were studied and it was concluded that it is a method that allows the farmer to better manage the water overcoming problems resulting from the irregularity of the slope and the lower permeability of the soil; however the distribution uniformity and application efficiency are low, labour consume is high, and it poses limits to mechanization. Several field evaluations provided for assessing the performance of furrow irrigation with different slopes and lengths for the first and second irrigation, including open and blocked furrows and over- and deficit irrigation. A variety of results were obtained showing the need for adopting improved design tools. Field evaluations produced data for parameterizing the models. Infiltration tests with the double ring infiltrometer and with furrow and basin observations of advance and recession led to define average infiltration curves for Arnan and Bab Al Faraj. The respective parameters have been selected using SIRMOD model in the reverse mode to optimize the matching of observed and simulate advance and recession curves. This allowed adequate parameterization for further using the DSS SADREG for design and ranking of alternative solutions for surface irrigation improvement. The SADREG application demonstrates the usefulness of this DSS tool to support the planning and design process of surface irrigation improvement. It

62

creates a friendly framework to build up improved irrigation scenarios, considering the main factors that determine the system performance and for decision-making selection of design alternatives by application of multicriteria methodology. Relative to cotton irrigation in Arnan, results show that furrow and border irrigation have high potential for improvement because they achieve good beneficial water use ratio and acceptable irrigation costs. The best results were obtained for the combinations slope - length of 0.5% - 50m and 0.8% - 100m. For cotton at Bab Al Faraj, the furrow irrigation performance is very sensitive to furrow length, with the best result with 150-200m and a not significant influence from slope 1.0 - 1.5% assuming adequate land smoothing conditions. For wheat irrigation in Arnan it was concluded that several options are feasible, with the level furrowed basin as the best performing system for flat fields, and the graded borders, with 0.5 and 0.8% slope, for sloping fields. The furrowed level basin gets the best beneficial water use ratio and the best water productivity and the graded border has the lower irrigation cost. The flat level basin is also an acceptable solution. A further DSS application would contribute to a deeper knowledge about the surface irrigation in Ras El Ain relative to the following aspects: (i) land leveling economic and environmental impacts, when changing field slopes is required to apply an improved irrigation method, studying its impact, in relation with actual topographic characteristics of fields; (ii) the water supply conditions to the field, in particular the upstream discharge and head, to determine the best condition to design the distribution system and to plan the labor requirements for irrigation; and (iii) the evaluation of the economical feasibility of water runoff reuse techniques, by field upstream pumping or by gravity to other fields.

63

References Allen R.G., Pereira L.S., Raes D., and Smith M. (1998) Crop Evapotranspiration. Guidelines

for Computing Crop Water Requirements. FAO Irrig. and Drain. Paper Nº56, FAO, Rome, Bielorai, H. and Shimshi, D. (1963). The effect of various moisture regimes on water

requirements and yields of cotton (in Hebrew). Ktavim 13:223-231. Bruce, R.R. and Shipp, C.D. (1962). Cotton fruiting as affected by soil moisture regime.

Agron. J. 54:15-18. Calejo, M.J., Sousa, P.L., Pereira, L.S., 1998. Evaluation of performance of surface irrigation

systems: optimisation of infiltration and roughness parameters. In: Pereira, L.S., Gowing, J.W. (Eds.), Water and the Environment: Innovation Issues in Irrigation and Drainage (1st Inter-Regional Conference on Environment–Water, Lisbon). E&FN Spon, London, pp. 140–148.

Calejo, M.J., Sousa, P.L., Pereira, L.S., 1998. Evaluation of performance of surface irrigation systems: optimisation of infiltration and roughness parameters. In: Pereira, L.S., Gowing, J.W. (Eds.), Water and the Environment: Innovation Issues in Irrigation and Drainage (1st Inter-Regional Conference on Environment–Water, Lisbon). E&FN Spon, London, pp. 140–148.

Сholpankulov E. D., O. P. Inchenkova, P. Paredes, L. S. Pereira, 2005. Testing the irrigation scheduling simulation model ISAREG for cotton and winter wheat in Central Asia. In: L.S. Pereira, V.A. Dukhovny, M.G. Horst (eds.). Irrigation Management for Combating Desertification in the Aral Sea basin. Assessment and Tools. Vita Color Publ., Tashkent, pp. 97-124.

Engineering Economic Organisation 1984, Sofia – Bulgaria Irrigation Project of Khabor River Catchment Basin. Stage two,.Irrigation with ground water from the Eocene Aquifer. Irrigation Ministry Department of The Land Reclamation.

Gonçalves, J.M., Pereira, L.S., 1999. Design of sustainable on-farm surface irrigation systems with a decision support system. In: Musy, A., Pereira, L.S., Fritsch, M. (Eds.) Emerging Technologies for Sustainable Land Use and Water Management, Presses Polytechniques et Universitaires Romandes, Lausanne: CD-ROM Paper 8.10.

Gonçalves JM, Pereira LS, Fang SX, Fabião MS, Campos AA, Mao Z, Paredes P, 2001. A computer package for simulation and decision-making on surface irrigation improvements for water savings. In: Zazueta FS and Xin JN (eds.) World Congress on Computers in Agriculture and Natural Resources (Foz do Iguaçu, Brasil), ASAE, St. Joseph, MI, pp. 835-842.

Gonçalves J.M., Pereira L.S., Campos A.A., Fabião M.S., Li Y.N., Liu Y., Cai L.G., 2004. Demand and delivery simulation and multicriteria analysis for water saving in the lower Yellow River basin. In: GH Huang, LS Pereira (eds.) Land and Water Management: Decision Tools and Practices (Proc. 7th Inter-Regional Conf. Environment and Water, Beijing, Oct. 2004). China Agriculture Press, Beijing, Vol. 1, pp. 275-289.

Gonçalves J.M., Horst M.G., Pereira L.S., Rolim J., Muga A., 2005a. SADREG, a DSS for Improving Surface Irrigation Systems. In: J. Boaventura Cunha and R. Morais (Eds.) Proceedings of the EFITA/WCCA 2005 Joint Conference on Information Technologies in Agriculture, Food and Environment and Computers in Agriculture and Natural Resources (5th EFITA Conference and 3rd WCCA, Vila Real), Universidade de Trás-os-Montes e Alto Douro, Vila Real, CD-ROM, pp. 160-165.

Gonçalves J. M., Horst M., Muga A., Rolim J., L.S. Pereira, 2005b. Impacts of soil infiltration on alternative furrow irrigation scenarios and performance using the DSS SADREG. In: A. Hamdy (ed.) Water, Land and Food Security in the Arid and Semi-Arid Regions (Proc. Int. Conf. CIHEAM/MAIBari, September 6 – 11, 2005), IAM, Bari, CD-ROM.

Gonçalves J. M., Muga A., Pereira L. S., Rolim J., Horst M. G., 2006a. SADREG, a client-server application DSS for design and management of surface irrigation. In: F. Zazueta, J. Xin, S. Ninomiya, G. Schiefer (eds.) Computers in Agriculture and Natural Resources (4th World Congress, Orlando, FL), ASABE, St. Joseph, MI, pp. 427-432.

Gonçalves J. M., A. Muga, L. S. Pereira, J. Rolim, M. G. Horst. 2006b. SADREG, a DSS for design and management of farm surface irrigation systems. In: Proceedings of "2006

64

World Congress of CIGR: Agricultural Engineering for a Better World", 4-7 September, Bonn.

Grimes, D.W., Miller, R.J. and Dickens, L. (1970). Water stress during flowering of cotton. Calif. Agric. 24:4-6.

Horst M.G., Shamutalov Sh.S., Pereira L.S., Gonçalves J.M., 2005. Field assessment of the water saving potential with furrow irrigation in Fergana, Aral Sea Basin. Agric. Water Manage. 77: 210-231.

Humpherys.A.S., Trout J.J and English M J.(1987) “Automation of border and basin” In Irrigation System for 21st centery. Proceeding of a conference by the Irrigation & Drainage Division of ASCE potland Oregon 293-301.

ISED, 1989. Surface Irrigation Simulation Software. User’s Guide. Irrigation Software Engineering Div., Dep Agricultural and Irrigation Eng, Utah State Univ., Logan, USA.

Katopodes, N. D., J. H. Tang and A. J. Clemmens. 1990. Estimation of surface irrigation parameters. J. Irrig. Drain. Engng., 116(5): 676-696.

Katopodes, N.D., Tang, J.H., Clemmens, A.J., 1990. Estimation of surface irrigation parameters. J. Irrig. Drain.Eng. 116 (5), 676–696.

Liu Y., Pereira L. S., Fernando R. M., 2006. Fluxes through the bottom boundary of the root zone in silty soils: Parametric approaches to estimate groundwater contribution and percolation Agric. Water Manage. 84: 27-40.

Merriam, J.L., Keller, J., 1978. Farm Irrigation System Evaluation: AGuide for Management. Utah St. University,Logan.

Namken, L.N. (1964). The influence of crop environment on internal water balance of cotton. Soil Sci. Soc. Am. Proc. 28:12-15.

Oweis, T., P.N. Rodrigues, L.S. Pereira, 2003. Simulation of supplemental irrigation strategies for wheat in Near East to cope with water scarcity. In: G. Rossi, A. Cancelliere, L. S. Pereira, T. Oweis, M. Shatanawi, A. Zairi (Eds.) Tools for Drought Mitigation in Mediterranean Regions. Kluwer, Dordrecht, pp. 259-272.

Paredes P., E.D. Сholpankulov, O.P. Inchenkova, L.S. Pereira, 2006 Irrigation Scheduling Strategies for Cotton to Cope with Water Scarcity. In: Proceedings of "2006 World Congress of CIGR: Agricultural Engineering for a Better World", 4-7 September, Bonn.

Pereira L. S., 2006. Assessment of Farm Irrigation Systems and Perspectives for Improvement in Ras Al Ain Area, Syria. CIHEAM/IAM Bari and “Cooperazione Italiana” project on “Rationalisation of Ras El Ain Irrigation System”, Syria.

Pereira L.S., Gonçalves J.M., 2005. Assessment of Farm Irrigation Systems and Perspectives for Improvement in Ras Al Ain Area, Syria. CIHEAM/IAM Bari and “Cooperazione Italiana” project on “Rationalisation of Ras El Ain Irrigation System”, Syria.

Pereira, L. S. and T. J. Trout. 1999. Irrigation Methods. In: van Lier, H. N., L. S. Pereira and F. R. Steiner (Eds.) CIGR Handbook of Agricultural Engineering, Vol. I: Land and Water Engineering. ASAE, St. Joseph, MI, pp. 297-379.

Pereira, L.S., Cordery, I., Iacovides, I., 2002a. Coping with Water Scarcity. UNESCO IHP VI, Technical Documents in Hydrology No. 58, UNESCO, Paris (http://unesdoc.unesco.org/images/0012/001278/127846e.pdf).

Pereira, L.S., Oweis, T., Zairi, A., 2002b. Irrigation management under water scarcity. Agric. Water Manage. 57,175–206.

Pereira, L. S., P. R. Teodoro, P. N. Rodrigues, J. L. Teixeira, 2003. Irrigation scheduling simulation: the model ISAREG. In: G. Rossi, A. Cancelliere, L. S. Pereira, T. Oweis, M. Shatanawi, A. Zairi (Eds.) Tools for Drought Mitigation in Mediterranean Regions. Kluwer, Dordrecht, pp. 161-180.

Rodrigues P.N., Pereira L. S. and Machado T. G. (2000) KCISA, a program to compute time averaged crop coefficients. Application to field grown vegetable crops. In: M.I. Ferreira and H.G. Jones (Eds.) Proceedings of the 3rd International Symposium on Irrigation of Horticultural Crops. ISHS, Acta Horticulturae Nr. 537, Leuven, vol. I, pp. 535-542.

Smith, M., Pereira, L.S., Berengena, J., Itier, B., Goussard, J., Ragab, R., Tollefson, L., Van Hoffwegen, P., (Eds.) 1996. Irrigation Scheduling: From Theory to Practice. FAO Water Report 8, FAO, Rome, 384 pp.

Stewart, J.L., Hagen, R.M. and Pruit, W.O. (1976) Water production functions and predicted irrigation programs for principal crops as required for water resources planning and

65

increased water use efficiency. University of California, Dept. Land, Air and Water Resources, Davis, and USDI/Bureau of Reclamation, Denver, CO.

Strelkoff, T. 1993. SRFR, a Computer Program for Simulating Flow in Surface Irrigation Furrows-Basins-Borders. USDA-ARS Water Conservation Lab., Phoenix.

Teixeira, J. L. and L. S. Pereira, 1992. ISAREG, an irrigation scheduling model. ICID Bulletin, 41(2): 29-48.

Walker, W. R. and G. V. Skogerboe. 1987. Surface Irrigation. Theory and Practice. Prentice-Hall, Englewood Cliffs, NJ, 386 p.

Zhang, H. and Oweis, T. (1999). Water-yield relations and optimal irrigation scheduling of wheat in the Mediterranean region. Agric. Water Manag. 38, pp. 195-211.