Assessment of Actual Irrigation Management in the Irrigated Area Of

Home , Irrigation management, Tile drainage, Water table

Estimation of root-zone salinity using SaltMod in the irrigated area of Kalaât El Andalous (Tunisia) 1*Ferjani Noura, 2Dhagari Hedi and 3Morri Mabrouka 1Faculty of Sciences of Bizerte, 7021 Zarzouna Bizerte, Tunisia, University of Carthage 2Resources Management and Conservation Research Unit, National Agronomic Institute of Tunisia (INAT), 43 avenue Charles Nicolle, Cité Mahragène -Tunis 1002-Tunisia, University of Carthage 3Resources Management and Conservation Research Unit, National Agronomic Institute of Tunisia (INAT), 43 avenue Charles Nicolle, Cité Mahragène -Tunis 1002-Tunisia, University of Carthage *Corresponding author. E-mail adress: [email protected]

ABSTRACT

The objective of this study was to predict the root zone salinity in the irrigated area of Kalaât El Andalous using the SaltMod simulation model over 10 years. En general, the model predicts more or less similar values of electrical conductivity in the root zone. Highest electrical conductivity values equal to 7.7 dS m-1 would be reached during the irrigation season. Following the rains fall particularly during the winter season there would be a decrease of the soil salinity when the average minimum value of electrical conductivity would reach 3.1 dS m-1. This seasonal variation of root zone salinity was confirmed from the measured values. Investigations were made also to study the impact of varying drain spacing and drain depth on root zone salinity and water table depth and drain discharge. The simulation show that the decrease in the depth of the drain does not change significantly the root zone salinity which reaches 6.21 dS m-1 during the irrigation season and 3.67 dS m-1 during the non irrigated season. The depth of the drain can be reduced to 1.6 m without any damage to crops. A water table depth (Dw= 1.8 m) appears to be excessive and would increase the cost of system installation. We can see a slight reduction in drainage flow when the depth of the drain goes from 1.8 m to 1.2 m. A decrease of the drain depth decreases the water table level. Although there was no variation in root zone salinities due to change in drain spacing. The predicted drainage flows have registered an evolution related to the occurrence of irrigation and rainfall. This evolution was confirmed by observed values. In this study, calibration of SaltMod for water-salt balance parameters shows the reliable validity of the model for the local conditions.

Key words: Root zone salinity; subsurface drainage; SaltMod; water table; drain discharge.

INTRODUCTION mainly chloride, carbonates and sulphates of sodium, calcium and magnesium. There are various factors High soil salinity is a serious problem often that cause salinization which include natural or encountered in the arid and semi-arid regions of the inherent and human induced factors and these are world where there is insufficient rainfall to leach generally categorized into primary and secondary accumulated salts from the crop root zone, especially salinization respectively (Shrestha, 2006). Primary in irrigated areas. Other factors contributing to the salinization results from natural weathering of parent problem are high evaporative demand and shallow material (e.i rock and minerals) and is influenced by ground water which cause accumulation of salts at factors related to climatic, topographic, hydrologic, the soil surface. Since soil salinity imposes a stress and geologic and soil condition. Secondary on the crop, it results in decreased yields, and in salinization develops from mobilization of the stored some cases, complete crop failure (Bresler et al., salts in the soil profile and/or groundwater due to 1982). Soil salinization results from accumulation of human activities (Greiner, 1997) and practices which water soluble salts in the soil surface and sub-surface include cultivation of marginal lands, inappropriate Agric. Biol. J. N. Am., 2012, 3(3): 105-116

irrigation practices, deforestation and mining may reach values up to 8-10 dS m-1 near to the activities. south-east sebkha. The hydraulic conductivity varied between 0,2 cm/h and 3,6 cm/h. Soil pH ranged from In order to control and monitor the process of 7.3 to 8.9. The average bulk density of the soil is salinization for the purpose of recovering damaged 3 about 1.5 g/cm . Shallow water tables of about 1.4 m land and preventing further expansion, information on depth are present in the lower parts of the district, its spatial distribution, its trends of expansion and with very high salinity values that make them severity levels is essential (Metternicht, 2001). A unsuitable for irrigation or other municipal and number of computer models have been developed industrial uses. which are useful in making predictions of soil salinity accumulation. The SALTMOD program was Properties of SaltMod: In order to estimate the root developed to predict the long-term effects of varying zone salinity, the SaltMod simulation model was water management options on desalinisation or salt used. The hydro-salinity model "SALTMOD" accumulation in the soil of irrigated agricultural lands. developed by Oosterbaan and Pedrose de Lima The water management options include irrigation, (1989), computes the salt and water balance for the drainage, and the reuse of surface drainage water or root zone, transition zone and aquifer zone. It is a subsurface drainage water from pipe drains, ditches computer-based model for simulating the salinity of or wells for the irrigation. In addition, predictions are soil and drainage waters, water table depth, and made on the depth to water table, the salt drain discharge in irrigated agricultural lands under concentration of the groundwater and of the drain or different hydrological and geo-hydrological well water. In Tunisia, Kalâat El Andalous irrigated conditions, varying water management options, and district is one of the most affected area by salinization several crop rotation schedules. due to shallow groundwater level. The objective of The model was developed to make long-term the present study was to predict the root zone salinity predictions of the impact of water management using the SaltMod simulation model. In addition, programs (including drainage) on the level of the investigations were made to study the impact of water table, and on the salt content of the soil, varying drain spacing and drain depth on root zone ground water, and drainage effluent. It can also salinity and water table depth and drain discharge. assess the impact of re-used drainage water. It MATERIALS AND METHODS includes area frequency distributions of salinity. SaltMod can also simulate farmers’ agricultural and Site description: The irrigated area of Kalâat El water-management responses to changes in water Andalous (latitude: 36° 37’ and 37°2’ N; longitude: table depth and soil salinity, which, in turn, can 10°5’ and 10° 10’ E) is located on the end part of the influence the salt and water balance (Oosterbaan, Medjerda watershed (Figure 1), with an average 1998). The program aims at using input data that are annual ETP of 1400 mm and an average annual generally available, or that can be estimated with rainfall of 490 mm. Irrigation area of Kalâat El reasonable accuracy, or that can be measured with Andalous was launched since 1992 on a flood area. It relative ease. covers an area of 2905 ha and the effectively irrigated area changes from season to season and In order to use SaltMod some factors should be the maximum was observed in the summer (about determined first. Determination of the data required 1000 ha). The irrigated area was divided into plots of could be done by entering different values for 5 ha supplied by a flow rate of 3 l/s. All the irrigated leaching efficiency (Flr), and natural drainage (Gn) area was equipped by a pressurized irrigation into SaltMod until values that match the actual network and a subsurface drainage system with a measured soil salinity and water table depths are length of 180 m and a depth of 1.8 m and spaced at obtained. Some of the data used as input parameters intervals of 40 m. The irrigation water is taken from in SaltMod were estimated or determined with in situ, the Mejreda River. Irrigation water salinity ranges laboratory, while the others were calculated by the between 3.1 g/l and 2.5 g/l in winter and between 2.3 model. g/l and 2.4 g/l in summer. The drainage outlet is Measurements: This study was carried out from Jun below sea level, and the drainage waters are 2008 to May 2010 in a farm plot of 2.38 ha (170 m x discharged to the Mediterranean Sea through a 140 m) equipped by three subsurface drainage pipes pumping station (SP4). The soils have a fine texture, D1, D2 and D3 and by drip irrigation system. There ranging from silty-clay to clayey-silt. Most soils have -1 are two cropping seasons per year i.e. winter electrical conductivity values above 2 dS m , and

106 Agric. Biol. J. N. Am., 2012, 3(3): 105-116

Water shed of Mejerda Irrigated areas Wadi of Mejerda

Alegria Governorate of Bizerte Tunisia

Douar Amar Ben Hamda Bizerte Kalâat El

Governorate of of Governorate Andalous

Legends Watershed Douar

boundary Khchim Sea Mediterranean Irrigated area Bridge of Bizerte

Douar Gbar Ejjahli Ennahli Régional road Highway National Highway Délégation Sidi Delegation of Raoued THABET Linear hydrography Perennial stream Temporary stream Irrigated area Public Echelle: 1/120 000 UsedPrivate water treated Urban area Garaat Ben Ammar Wetland

Fig. 1. Irrigated area of Kalâat El Andalous (December to May) and summer (June to (Cucumis mela) (1 ha) and squash (Cucurbuta September) with a fallow season in between (October maxima) (0, 38 ha) and rain fed wheat occupy the to November). In the field, summer irrigated crops are field during winter season. Table 1 indicated irrigation tomato (Lycopersicum esculentum) (1 ha), melon characteristics system.

Table 1. Fields and irrigation characteristics system during irrigation season (May 2008 - September 2008).

Average emitter Crops Field size (ha) Row spacing (m) Emitter spacing (m) discharge (l/h) Tomato 1 1.5 0.4 2,1 Melon 1 1.5 0.8 1,5 Squash 0.38 1.5 0.8 1,9

107 Agric. Biol. J. N. Am., 2012, 3(3): 105-116

Daily climatic data were collected by a Campbel content and salinity. The water supplied volumes V meteorological station located near the site and (m3) are determined as: reference evapotranspiration (ETo) was calculated V = N q T.103 (1) with the FAO Penman-Monteith method (Allen et al., 1998). Crop evapotranspiration (ETc) was calculated Where: N is the number of emitters per hectare, q the as ETc = ETo Kc, where Kc are crop’s coefficients average emitter discharge (l/h), and T irrigation taken from local information or the literature (Allen et duration. The discharge of emitters was measured al., 1998). The repartition of precipitation and ET0 weekly. The duration of irrigation was estimated during the study period is given in Figure 2. During according to farmer declaration and it varied between the irrigation season (May-September), rainfall is 3 and 3,5 hours/day for squash, between 1,5 and 4 negligible. hours/day for tomato and between 1 hour and 3 hours/day for melon. The applied water during the

irrigation season, ranged from 4 to 16 mm/day for tomato, from 3 to 9 mm/day for melon and from 4 to 5.5 mm/day for squash. The Amounts of irrigation water diverted to crops and the ETc values were indicated in Table 2. The water table levels were measured monthly using a piezometer localized in the plot. Also, samples were monthly taken to measure the Electrical Conductivity (EC) of the groundwater. To assess the amount of salts removed from the study area, drain discharge was measured monthly at the end of subsurface drainage pipes D1, D2 and D3. Furthermore, soil sampling were carried out to monitor soil salinity under irrigated crops (tomato, melon and squash), under rainfed crops

Fig. 2. The repartition of precipitation and reference (wheat) and bare soil. Sampling was done on three evapotranspiration (ET0 ) during the study period. spots at 0-30 cm, 30-60 cm and 60-90 cm every two weeks during the irrigation season and about once a month for the other periods. Soil, irrigation water, In the field, measurements included irrigation water drainage water and groundwater samples were volume and salinity, pipe drainage discharge and analysed to determine electrical conductivity. salinity, water table level and salinity, soil water

Table 2. Amounts of irrigation water diverted to crops and mass of salts induced by irrigation water.

Mass of salts induced by Crops ET (mm) Water amounts diverted (mm) c irrigation water (tons/ha)

Tomato 449 1030 24

Melon 332 503 12

Squash 285 299 7

The computation method SALTMOD is based on laboratory while the others were calculated by the seasonal water and salt-balance of agricultural lands, model. which can be expressed by the general water- Calibration of SaltMod: After collection of necessary balance equation as: Incoming Water = Outgoing input-data, it was then converted into the input-format Water ±  (where  is the change in water stored). as required by the model. The input-data for each Some of the data used as input parameters in year was given as average values over three SaltMod were estimated or determined with in situ, seasons i.e. irrigated season and non irrigated

108 Agric. Biol. J. N. Am., 2012, 3(3): 105-116

season and fallow season of each year, given transition zone (Flx), and the natural drainage (Gn) of separately to simulate the results for the next year. groundwater through the aquifer. All these factors The model was applied to calibrate for local should be determined prior to applying the SaltMod conditions of the irrigated area of Kalaât El Andalous, model. This can be done by trials with SaltMod using under existing irrigation and cropping practices, to different values of root zone and transition zone simulate seasonal watertable changes, root-zone leaching efficiency, and the natural drainage, salinity, quantity, and quality of tile drainage flow. The choosing those values that produce soil salinities and match of the data was obtained by optimizing and water table depths that correspond with the values varying the leaching-efficiencies and the natural actually measured (Oosterbaan, 1998, 2000). Field drainage to the aquifer, establishing the validity of the observations on water table and soil salinity of the model. The simulated results were analyzed and three seasons of the two years 2008-2009 and 2009- compared with field-data collected. 2010 were used for calibrating the model and subsequently the projections were made starting from Some factors could not be measured, notably the the year 2009 till 2019. The input parameters used in leaching efficiency of the root zone (Flr) and the model calibration are shown in Tables 3 and 4.

Table 3. Season-wise input parameter for use in SALTMOD

SI.N°. Parameters Season 1 Season 2 Season 3 1 Duration 1st Jun to 30th Sept 1st Oct to 30 Nov 1st Dec to 31 May Tomato, melon and 2 Crop grown Fallow Rainfed wheat squash 3 Water sources Irrigation water Rainfall Rainfall Fraction of area occupied by irrigated 4 1 - - crop other than rice Fraction of area occupied by irrigated 5 0 - - rice crop 6 Fallow - 1 -

7 Rainfall (m) 0.03 0.02 0.15 Water used for irrigation in crops other 8 1 - - than rice (m) Potential evaporation from irrigated land 9 0.264 - - (m) Potential evapo transpiration from un 10 - 0.03 0.093 irrigated area(m) Percolation losses from the irrigation 12 0.05 - - canal system (m) Storage efficiency of rain and irrigation 13 0.9 - - water in the root zone Storage efficiency of rain water in the 14 - 0.8 0.8 root zone of the un-irrigated land Incoming surface runoff into the un- 15 - 0 0 irrigated land (m) Outgoing ground water flow through the 16 0 0 0 aquifer (m)

109 Agric. Biol. J. N. Am., 2012, 3(3): 105-116

Table 4. Other input parameter for use in SALTMOD

S.No Parameter Value 1 Depth of root zone (m) 0.6 2 Depth of transition zone (m) 0.7 3 Depth of aquifer (m) 6 4 Total porosity of root zone 0.43 5 Total porosity of transition zone 0.46 6 Total porosity of aquifer 0.5 7 Effective porosity of root zone 0.02 8 Effective porosity of transition zone 0.02 9 Effective porosity of the aquifer 0.07 Initial salt concentration of the water in the transition zone when saturated 10 4 (dS/m) Initial salt concentration of the soil moisture in the root zone when saturated 11 2.8 (dS/m) 12 Initial salt concentration of the water in the aquifer when saturated (dS/m) 5.5 Initial salt concentration of the ground water in the upper part of the transition 13 3.9 zone (dS/m) Initial salt concentration of the ground water in the lower part of the transition 14 6.2 zone (dS/m) 15 Salt concentration of the irrigation water (dS/m) 3 16 Salt concentration of the rain water (dS/m) 0 17 Initial depth of the water table (m) 1.33 18 Critical depth of the water table for capillary rise (m) 1

Determination of Leaching Efficiency: Leaching Fig. 3. Calibration of root zone salinity and leaching efficiency of the root (Flr) or transition zone (Flx) is efficiency in the test area defined as the ratio of the salt concentration of the Determination of the Natural Subsurface water percolating from the root or transition zone to Drainage: In SaltMod, natural subsurface drainage the average concentration of the soil water at (Gn = G0 - Gi) is defined as the quantity of saturation. A range of arbitrary values were given for horizontally outgoing ground water (G0, m3/season the leaching efficiency of the root zone (Flr) and per m2 total area) minus the quantity of horizontally transition zone (Flx); the corresponding root zone incoming ground water (Gi) in season. This value salinity results of the program were compared with was determined by setting the natural drainage the values actually measured. The arbitrary Flr values (Gi) at zero, arbitrarily changing the outgoing values were taken as 0.2, 0.4, 0.6, 0.8, and 1.0, ground water values, and finding the corresponding salinity levels of the root zone were obtained, and the values for water table depth (Dw) and drain discharge data are shown in Figure 3. Flr = 0.8 was the best (Gd). Since the third season (6 months) was longer matching value to the observed values and was used than the second season (2 months) and the first in all calculations. season (4 months), the arbitrary G01, G02 and G03 st nd 20 values, i.e. the Go values for the 1 , the 2 and the 18 third season respectively, are in pairs: (0.00, 0.00, 16 0.00), and (0.02, 0.01, 0.04), (0.05, 0.02, 0.07) and 14 FLr = 0,2 12 FLr = 0,4 (0.07, 0.04, 0.1), and (0.09, 0.06, 0.13). As inflow Gi 10 FLr = 0,6 values were taken to be equal to zero, G0 values for 8 FLr = 0,8 three seasons together gave Gn values (Table 3). 6 FLr = 1

4 Measured Results from the simulation showed that when the Rootzone salinity (dS/m-1) 2 annual natural drainage was set at 0.0 m for the three 0 seasons, drain discharge (Gd) was 0. 15 and 0.0511 4 16 28 40 52 64 76 88 100 112 124 Time (months) and 0.785 m for the first and the second and the third

110 Agric. Biol. J. N. Am., 2012, 3(3): 105-116

season, respectively. These discharge values are cropping-pattern, etc., from the observed data given much higher than those observed in the study area. as average input to the model. Drain discharge (Gd) was simulated by the model as Prediction of root zone salinity: Prediction of 0.0514 and 0.0166 and 0.711 m when the annual salinities for 10 years period is shown in Figure 4. En natural drainage was assumed to be 0.21 m. The general, the model predicts more or less similar value simulated values were close to the discharge values of the root zone salinity (Cr4) provided that the measured during the summer and winter seasons. current land use practices are maintained. In fact, RESULTS AND DISCUSSION highest electrical conductivity values equal to 7.7 dS m-1 would be reached during the irrigation season on Once the model has got verification-reliability for an the first year. During the first season due to irrigation area, it could then be applied with confidence for the salts accumulation is more than in the second further use and application, to make predictions and season. Following the rains fall particularly during for various alternative water-management scenarios. the winter season there would be a decrease of the The main objective is to determine how salinization soil salinity when the average minimum value of would change over long term basis in the study area electrical conductivity would reach 3.1 dS m-1 on the given the present land use practices continue. In second year. In fact, rainwater leaches down the order to achieve the objective, SaltMod was used to salts from the top layer and contribute to the model temporal changes of salinization over one desalination of the soil profile. By the end of 10 years, decade (10 years) periods. The root zone (≤60 cm the predicted root zone salinity is 6.21 dS m-1 and depth) salinity and the level of groundwater table and 5.84 dS m-1 and 3.67 dS m-1 during the first, the drain discharge were the main variables of interest second and the third season, respectively. If one predicted by SaltMod. However there are other compares the actual values (tows years) with the parameters that are predicted by the model but are simulated values, more or less similar trend is not of concern in the present study. For the prediction observed during the three season. Observed values period, it was assumed that there will be no of soil salinity during the study period were indicated significant yearly deviations of the input parameters, in Figure 5. such as rainfall, irrigation, evapotranspiration, 9 8 7 6 5 4 3

Soilsalinity (dS/m) 2 1 0 0 1 2 3 4 5 6 7 8 9 10

Year

Cr4 (season 1) Cr4 (season 2) Cr4 (season 3) Cxa

Fig.4. Predicted root zone salinity and soil salinity above drain level (Cxa)

111 Agric. Biol. J. N. Am., 2012, 3(3): 105-116

Fig.5 - Average soil profile electrical conductivity observed during the study period (May 2008-June 2010)

crop if keeping the same condition for 10 year The soil salinity above drain level (Cxa) increased period. from 3.9 dS m-1 to around 5.5 dS m-1 during the first year, as salts from the top layers leached down thus SaltMod is a very useful tool for predicting root zone increasing the salinity above drain level. Thereafter soil salinity and the amount of drainage water from the Cxa value is practically constant and is land equipped with subsurface drainage systems. maintained at the same level (5 dS m-1). The level of The model successfully predicts water table depth root zone salinity predicted by SaltMod was and drain discharge rates across agricultural areas confirmed from field tests and can reduce crops yield. using various water management practices According to Reina-Sanchez et al. (2005), tomato (Oosterbaan, 2000; Srinivasulu et al., 2004; Bahceci fruit is the most sensitive organ to the salinity, and it et al., 2006; Bahceci and Nacar, 2007). Rao et al. shows significant yield reduction under irrigation (1990) revealed that the time-averaged depth of the water with salinity equal to 2.5-7.3 dS m-1. water table during the critical season need not be more than 0.8 m below the soil surface to allow for Prediction of water table depth: The model also adequate reclamation of saline soils. predicts the seasonal changes of the water table depth. The simulated water table depths in three Prediction of drainage water quality and rate flow seasons over the 10 years are given in Figure 6. The : The predicted values of drainage water quality in simulation shows a variation in water table Figure-8 show no significant variation of electrical level between 1.21 and 1.77 m. The water table conductivity of drainage water which vary between 4, depth exhibits seasonal variation, being influenced by 7 mS/cm and 5,7 mS/cm. These values are small recharge from rainfall and irrigation. The model tends than the measured values which varied from 7 to maintain almost the same depths for each season mS/cm to 8,4 mS/cm (Figure 9). The observed throughout the simulation period. By comparing the drainage flows have registered an evolution related to actual values (two years) with the simulated values, the occurrence of irrigation and rainfall. In fact the we found that simulated values of water table depths maximum values are observed during winter season. remains higher than measured values (Figure. 7). The predicted value of drainage flow changed from The depth of the water table is always less than the season. The minimum values were observed during depth of the root zone (0.6 m), and then the level of season 2 and the maximum were observed during the water table should have no adverse effect on season 3 (0.711 m/season.)

112 Agric. Biol. J. N. Am., 2012, 3(3): 105-116

Fig. 6. Simulated water table level fluctuation

Fig. 7. Comparison of actual and predicted of water table depth

Fig. 8. Simulated values of electrical conductivity of drainage water

113 Agric. Biol. J. N. Am., 2012, 3(3): 105-116

Fig. 9. Observed values of drainage flow rate and electrical conductivity of drainage water

6). We can see a slight reduction Effect of varying drains depth on root zone in drainage flow (Gd) when the depth of the salinity, on water table level and drain discharge: drain goes from 1.8 m to 1.2 m. Table 6 shows also In this scenario, keeping all the other parameters that a decrease of the drain depth decreases the constant, the effect of varying drains depth on root water table level which varies between 0.55 and 1.15 zone salinity, on water table level and drain discharge m for the first season and between 1.14 and 1.71 m was studied. The simulation shows that the decrease for the second season and between 1.1 and 1.71 m in the depth of the drain does not change for the third season. The optimal drain depth is the significantly the root zone salinity which ranges depth where the water table depth and salinity have between 5.84 dS m-1 and 6.31 dS m-1 during the first -1 -1 no adverse effect on crop. The critical values season and between 5.55 dS m and 5.87 dS m of Cr4 vary from 3 to 6.5 dS m-1; the water table during the second season and between 3. 28 dS m-1 -1 depth must be greater than 0.6 m (the depth of the and 3.82 dS m during the third season (Table root zone). Table 5. Simulated values of annual natural drainage (Gn m/year), seasonal average of water table depth (Dw, m), and quantity of drainage water (Gd, m/season).

Season 1 Season 2 Season 3 Gn annual Dw Gd Dw Gd Dw Gd 0 1.15 0.15 1.71 0.0511 1.71 0.785 0.07 1.17 0.11 1.73 0.0419 1.74 0.764 0.14 1.19 0.0807 1.74 0.0327 1.75 0.733 0.21 1.21 0.0514 1.77 0.0166 1.77 0.711 0.28 1.22 0.0171 1.85 0 1.79 0.691 Observed values 1.2 0.02304 1.5 0.00216 1.4 0.0864

114 Agric. Biol. J. N. Am., 2012, 3(3): 105-116

Table 6. Effect of variation of drain depth on root zone salinity, water table depth and drain discharge

Drain depth (m) Season Cr4 (dS/m) Dw (m) Gd (m/season)

1 6.31 1.15 0.0514 1.8 2 5.87 1.71 0.0148 3 3.69 1.71 0.711 1 6.31 0.945 0.0497 1.6 2 5.87 1.53 0.0057 3 3.69 1.51 0.722 1 6.03 0.74 0.0463 1.4 2 5.68 1.34 0 3 3.82 1.31 0.732 1 6.21 0.555 0.0429 1.2 2 5.84 1.14 0 3 3.67 1.11 0.723 1 5.84 0.356 0.0429 1 2 5.55 0.945 0 3 3.28 0.983 0.71 1 6.21 0.155 0.0429 0.8 2 5.84 0.745 0 3 3.67 0.713 0.71

According to the simulations the drain depth that is Effect of varying drain spacing on root zone most responded to these conditions is between 1.4 salinity: The drain spacing is represented by a and 1.8 m then the depth of the drain can be reduced parameter, QH1 in the input file. The higher the QH1 to 1.6 m (which seems the most suitable with a water value, the spacing between the drains will be the table level that varied between 0,94. and 1.53 m) narrower. The QH1 value under the present drain without any damage to crops. A water table depth spacing in the pilot area is 0.0095. In order to (Dw= 1.8 m) appears to be excessive and would simulate the effect of drain spacing on root zone increase the cost of system installation. salinity, the QH1 values were varied in the range of 60% (QH1= 0. 0057), 80% (QHl=0.0076), 120% The advantage of reducing the drain (QH1=0.014) of the present value. It was observed depth allows easy installation of the drain with that the results of all the simulations do not change less cost. Safwat Abdel-Dayem and Ritzema (1990) the root zone salinity. have shown that the seasonal average drain depth in the Nile delta should not be less than 0.7 m to avoid CONCLUSIONS decline in crop yield. Therefore, a minimum drain The present study applied long term prediction of root depth of Dd = 1 m is required to safeguard the crop zone salinity changes by means of deterministic production. A drain depth of Dd = 1.4 m appears to modeling using SaltMod. En general, the model be excessive. Safwat and Ritzema (1990) determined predicts more or less similar value of the root zone that a seasonal average drain depth of 0.8 m is also salinity (Cr4). In fact, highest electrical conductivity sufficient for good crop production. By employing the values equal to 7.7 dS m-1 would be reached during 0.8 m depth as a drainage criterion, one can avoid the irrigation season on the first year. Following the the design of an excessively extensive drainage rains fall particularly during the winter season there system. Rao et al. (1990) revealed that the time- would be a decrease of the soil salinity when the averaged depth of the water table during the critical average minimum value of electrical conductivity season need not be more than 0.8 m below the soil would reach 3.1 dS m-1 on the second year. The surface to allow for adequate reclamation of saline simulations show that the decrease in the depth of soils. the drain does not change significantly the root zone salinity. The depth of the drain can be reduced to 1.6

115 Agric. Biol. J. N. Am., 2012, 3(3): 105-116

m without any damage to crops. A water table depth eastern Australia. Land Degradation & Development (Dw= 1.8 m) appears to be excessive and would (1997). 8(4): p. 281-303. increase the cost of system installation. We can Matternicht, G., Assessing temporal and spatial changes of see a slight reduction in drainage flow when the salinity using fussy logic, remote sensing and GIS. depth of the drain goes from 1.8 m to 1.2 m. Foundation of an expert system, in Ecological Furthermore, a decrease of the drain depth Modelling (2001). p163-179. decreases the water table level. Although there was Rao, K.V.G.K., P.S. Kumbhare, S.K. Kamra and R.J. no variation in root zone salinities due to change in Oosterbaan (1990). Reclamation of waterlogged saline drain spacing. In this study, calibration of SaltMod for alluvial soils in India by subsurface drainage. In: water- salt balance parameters shows the reliable Symposium on land drainage for salinity control in arid validity of the model for the local conditions and so it and semi-arid regions. Vol. 2, pp. 17-25. Drainage could be applied with confidence for other areas. Research Institute, Cairo. Reina-Sanchez, A., Romero-Aranda, R., Cuartero, J. ACKNOWLEDGMENTS (2005). Plant water uptake and water use efficiency of greenhouse tomato cultivar irrigated with saline water. This study was partially supported by the European Agricultural Water Management, 78: 54-66. Commission research project INCO-CT-2005- 015031. The authors gratefully acknowledge the Safwat Abdel-Dayem and H.P. Ritzema (1990). verification assistance of the technicians and students that of drainage design criteria in the Nile Delta, Egypt. actively participated in this work. Irrigation and Drainage Systems Journal, 4, 2, p. 117- 131. REFERENCES Safwat, A.D. and H.P. Ritzema (1990). Verification of Allen, R.G., Pereira, L.S., Raes, D., Smith, M. (1998). Crop drainage design criteria in the Nile Delta, Egypt. Irrig. evapotranspiration-guidelines for computing crop water Drain. System. 4: 117-131. requirements. FAO Irrig. Drain. Paper 56, Rome, Italy. Shrestha, D.P., Relating soil electrical conductivity to Bahceci, I., N. Dinc, A.F. Tarı, A.I. Agar and B. Sonmez remote sensing and other soil property for assessing (2006). Water and salt balance studies, using SaltMod, soil salinity in northeast Thailand. Land degradation & to improve subsurface drainage design in the Konya- development (2006).17 (6) : p. 677-689. Cumra Plain. Turkey, Agricultural Water Management. Srinivasulu, A., C.S. Rao., G.V. Lakshmi, T.V. 85: 261-271. Satyanarayana and J. Boonstra (2004). Model Studies Bahceci, I. and A.S. Nacar (2007). Estimation of root zone on Salt and Water Balances at Konanki Pilot Area. salinity, using SaltMod, in the arid region of Turkey. Andhra Pradesh, India Irrig. Drain. System. 18: 1-17. Irrig. Drain. 56: 601-604. Oosterbaan, R.J. (1998). SALTMOD, Description of Bresler, E., B.L. McNeal, and D.L. Carter (1982). "Saline Principles and Applications. ILRI, Wageningen, The and Sodic Soils: Principles- Dynamics-Modeling." Netherlands. Advanced Series in Agricultural Sciences 10. Oosterbaan, R.J. (2000). SALTMOD. Description of Greiner, R. Optimal farm management responses to Principles and Applications, ILRI, Wageningen, The emerging soil salinization in a dry land catchment in Netherlands.

116