Assessment of Actual Irrigation Management in Kalâat El Andalous District ( Tunisia ) : Impact on Soil Salinity and Water Table Level

The objective of this work is to assess water and soil salinity evolution in the irrigated area of Kalâat El Andalous. Soil salinity, crops yield, water table level and drainage water flow were monitored during the period May 2008-June 2010. The results showed that during irrigation season (May-September 2008), the supplied water amounts for drip irrigated crops (tomato, melon and squash) were higher than crop water requirements. In fact, the soil water content was always equal or higher than the field capacity. Average root zone (0-60 cm) electrical conductivity of the saturated past extract (ECe) was 2.3 dS m, 2.8 dS m and 3.0 dS m in May 2008, May 2009 and May 2010 respectively. But at the irrigation season end, higher electrical conductivity (8.4 dS m) was recorded in the upper layer (0-30 cm). Along rainfall season, a soil salinity decrease was recorded in fact the average electrical conductivity reached 2.0 dS m. In order to reduce soil salinization (due to accumulated salts during irrigation season), farmers use crop rotation including rain fed crops and bare soil. Allover irrigation season, the highest drainage discharge (192 l/h) was recorded on July 2008 when the maximum irrigation water amount was diverted. Water table level shows a sustained rise when irrigation is relatively frequent during summer.


Introduction
In arid and semi-arid areas, irrigation is used to maximize crop yields by minimizing water stress in the root zone.However, this is often done an ad-hoc manner.Excess of water supplies may cause rising of ground water table which may carry salts from subsurface to surface layers through capillary rise and evaporation (Turhan & Baser, 2001).Soil salinization induced by capillary rise of shallow groundwater into the rooting zone plays a major role, nullifying pre-season salt leaching efforts, entailing yield decrease and seriously threatening economic growth and development (Grieve et al., 1986;Smets et al., 1997;Singh, 2004;Murtaza, 2006).Such evapo-concentration phenomenon associated with saline irrigation water is the main cause of soil salinization in irrigated districts (Stuyt, 2000).The salt accumulation in the soil profile is a widespread problem that seriously affects crop productivity throughout the world.More than 50% of the salinized areas in the Mediterranean basin are located in Algeria, Morocco, Spain, Tunisia and Turkey (Aragüés et al., 2011).
The use of drip irrigation may bring about a potential threat of the secondary soil salinization because no salt can be discharged from soil profile and salt build-up on the soil surface may be on the rise after long-term application of drip irrigation (Zhou & Ma, 2005).Hence, it is essential that farmers have a clear understanding about irrigation practices' impact on the soil moisture content, on soil salinity and on crop yields.In fact, optimal irrigation management is supposed to maintain favorable soil water content, prevent salinity stress, and save water resources as much as possible.In Tunisia, Kalâat El Andalous irrigated district is one of the most affected area by salinization due to shallow groundwater level.This study aims to assess water and soil salinity evolution under the main frequently irrigated crops (tomato, melon and squash), rain fed crop (wheat) and bare soil in Kalaât El Andalous district. www.ccsen

Experi
The irrigat the end pa 1400 mm 1992 on a (about 100 l/s.All the with a len level, and  This study was carried out during May 2008-June 2010 in a farm plot of 2.38 ha (170 m x 140 m) equipped with drip irrigation system and drained by three subsurface pipes D 1 , D 2 and D 3 .Tables 3 and 4 showed the cropping pattern and the irrigation system characteristics respectively.Irrigations were practiced daily from 05/05 to 07/08 for tomato and from 15/05 to 25/07 for melon.Subsequently, irrigations were made once every two days.For squash, irrigation frequency was maintained on alternate days from 31/05 until 26/06, it was daily from 27/06 to 11/07 then once every two days during the rest of the irrigation period.Irrigation duration ranges between 1.5 h/day and 4 h/day for tomato, between 3 h/day and 3.5 h/day for melon and squash.The supplied volumes V (m 3 ) are determined as: Where N is the number of emitters per hectare, q is the average emitter discharge (l/h), and T refers to the irrigation duration.The fixed emitters' discharges were measured weekly.The irrigation time was estimated according to farmer declaration and our survey.Daily climatic data were collected on a meteorological station located near the experimental plots.Reference evapotranspiration (ETo) was calculated using Penman-Monteith method (Allen, 1998).Crop evapotranspiration (ETc) was calculated as: where K c is crop's coefficient (Allen, 1998).
As given in Figure 2, rainfalls are negligible during the irrigation season (May-September).
Figure 2. The repartition of precipitation and reference evapotranspiration (ET 0 ) during the study period The water table levels were measured monthly using two piezometers.The first is localized in the plot, and the second is 50 m near the plot.Groundwater was sampled monthly for Electrical Conductivity (EC) measurements.
In order to compute the removed (from the study area) salts amounts, monthly drained water discharge measurements were made on the subsurface pipes (D 1 , D 2 and D 3 ) ends.Furthermore, soil sampling was carried out in order to assess soil salinity (ECe) evolution in irrigated (tomato, melon and squash), rain fed (wheat) and in bare soil plots.Within each plot, three spots were randomly chosen where samples were done on the 0-30 cm, 30-60 cm, 60-90 cm, 90-120 cm, 120-150 cm and 150-180 cm layers.These measurements were fortnightly during the irrigation period and monthly for the rest of the season.
For stored water computing, soil water contents (SWC), before (on 06/14) and after (on 07/12) irrigation, were determined gravimetrically in tomato and melon plots.Within each plot, three spots were randomly chosen where sampling was made on the 0-10 cm, 10-30 cm, 30-50 cm, 50-70 cm and 70-90 cm.For roots distribution and crop yields' estimates, three representative plants from each plot (tomato, melon and squash) were randomly chosen where some agronomic parameters were determined namely: the root length, average fruit number per plant and average fruit weight.Soil, groundwater and drained water samples were analysed to determine electrical conductivity and pH.The calcium (C ca ), magnesium (C Mg ) and sodium (C Na ) concentrations were determined on saturated soil extracts (Black, 1965) and then sodium adsorption ratio (SAR) was calculated using the relationship: Gathered data were analysed using descriptive statistics (average value, minimum value, maximum value, coefficient of variation and standard deviation).

Water Irrigation Volume, Salt Amount and Soil Water Content
The daily supplied water volume varied during the irrigation season, from 4 to 16 mm for tomato, from 3 to 9 mm for melon and from 4.0 to 5.5 mm for squash crops.In tomato and squash plots, the maximum supplied water volumes were recorded in July: 4143 m 3 /ha and 1130 m 3 /ha respectively.The applied water amount decreased in August, mainly because the crop water requirement decreased.But these diverted amounts remained higher than the crop water requirements (Table 5).
Allover the irrigation season, the cumulated rainfalls were only 17 mm.Thus, the applied salt amounts (by irrigation water) reached 24, 12 and 7 tons/ha in tomato, melon and squash plots respectively (Table 5).These results agree with those previously obtained in this region.In fact, Slama et al. (2004) recorded a total water amount equal to 10000 m 3 /ha for drip irrigated tomato.Allover irrigation season, the water content was always higher or equal to field capacity (34%) (Figure 3).In fact before irrigations, the average water content profile was higher than 30% and 33% in tomato and in melon plots respectively.The most important changes were observed on 12/07 within 0 -10 cm where the soil water content increased from 34 % to 39% underneath the emitter, from 32% to 36% at 10 cm and from 32% to 35% at 20 cm away of the emitter.In the deep layers (60-90 cm), water content distribution was approximately constant (≈ 37%) throughout the soil profile.Rawlins and Rotas (1975) and Guohua et al. (2009) reported that compared with the border-irrigation, frequently drip and sprinkler-irrigated field help to maintain higher soil water content.
www.ccsen Figure 3.In irrigated plots, the soil electrical conductivity (EC) changed allover crop growth stages.In fact, the root zone (0-60 cm) EC averaged value increased (between the beginning and the end of irrigation season) from 2.0 dS m -1 up to 5.3 dS m -1 , to 3.3 dS m -1 and to 6.8 dS m -1 in tomato, melon and squash plots respectively.But, the highest EC values: 8.4 dS m -1 , 7.0 dS m -1 and 7.7 dS m -1 were recorded on 19/07 and on 16/09 in melon, tomato and squash plots respectively.So irrigation water supply caused a progressive salt accumulation over irrigation season.In the three (melon, tomato and squash) plots, such salinization was especially recorded in the top layer (0-30 cm).
In the three irrigated plots, EC within the rooted layer (0-60 cm), ranged between 2.7 and 6.0 dS m -1 .While in the rain fed wheat and bare soil plots, EC (within the 0-60 cm layer) ranged between 1.6 and 4.1 dS m -1 and between 2.9 and 4.0 dS m -1 respectively.Therefore, this salt accumulation was essentially due to an inadequate irrigation management.The inherent alkalinisation risks are obvious because the recorded maximum SAR values reached: 16.1, 15.1, and 10.7 in the irrigated squash, tomato, and melon plots respectively.Whereas in the rain fed wheat and in bare soil plots, SAR recorded values were: 4.5 and 6.2 respectively.

Crop Yield
Measured crops' yields are shown in Table 8.In 2008, Harvests began on August the first and the 15 th for melon and tomato, and on September the first for squash.Tomato recorded yield was only 50 tons/ha: this result is significantly lower than Tunisian national average yield (80 tons/ha).According to Reina-Sanchez et al., (2005), tomato fruit is the most sensitive organ to the salinity, indeed significant yield reduction was recorded with irrigation water electrical conductivity higher than 2.5 dS m -1 .Ayers (1977) reported that using irrigation water with EC equal to 2.3, 3.4 or 5.0 dS m -1 reduces tomato yield by 10, 25 and 50 % respectively.Cuartero and Fernandez-Munoz (1999) recorded a decrease of tomato fruit weight and number when irrigated with water which EC is equal to 2.5 dS m -1 .Campos et al., (2006) compared the effects of five water salinity levels (1, 2, 3, 4, and 5 dS m -1 ) on industrial tomato yield.They concluded that total yield was reduced by 11% upon each unit increase of water salinity while fruit quality improved with increasing water salinity.9. Measured drainage flow rate varied (between 1.5 and 3.2 l/mn) according irrigations amounts and frequencies.Indeed, maximum drainage discharge was observed on July: when the maximum water volume was supplied.After the irrigation season, the drainage flow rate decreased sharply and annulled on the beginning of October.Drainage water electrical conductivity ranged between 4 and 8 dS m -1 .Consequently, the salt masses leached by D 1 , D 2 and D 3 were 2.6, 1.9 and 1.7 tons/ha respectively.Tedeschi et al. (2001) found that average flow rate of the drainage water reached 26.7 l s -1 and 31.3 l s -1 respectively during non-irrigated and irrigation season.They concluded that exported salts mass was linearly correlated (p < 0.001) with drained water volume that depended (P < 0.001) on irrigation supplies.
During rainy season, the subsurface pipes outlets were flooded.10), the same value observed on May 2008.Hence, groundwater shallow depths likely contributed to salt build-up in the soil through evapoconcentration process.Feng et al., (2005) reported that after irrigation season, the groundwater level remarkably rose from 2.9 to 1.3 m below soil surface. .This decrease is due to the important irrigation water amounts that reached the groundwater.Since November the first, water table EC reached 5.8 dS m -1 on 06/02.It should be noted that the rainfall recorded along the year was 462 mm while the diverted water amount (was 1030 mm for tomato, 503 mm for melon and 299 mm for squash crops.As previously discussed, the broad irrigation water supplies were responsible for the significant water table rise which contributed to the soil salinity leaching.Therefore, Kalaat Landalous water table exhibits seasonal level and salinity variations, especially due to inadequate irrigation management.

Conclusions
This study was carried out during May 2008-June 2010 in a farm plot of 2.38 ha divided into: irrigated crops (tomato, melon and squash), rain fed crop (wheat) and bare soil.The supplied water amounts were higher than the total crop water requirements.Hence, during the irrigation season, the soil water content was always more than or near the field capacity.Therefore, water table level showed a sustained rise when irrigations were relatively frequent, and drainage flow rates increased accordingly.Whil along two years, the soil salinity varied slightly, but it was not the case over irrigation season.In fact, results showed an increase of the soil salinity essentially due to capillary rise of salt water from the water table.Soil salinity follow-up, on the three irrigated plots, showed that the most important concern was the top layer (0-30 cm) salinity increase.Fortunately, these accumulated salts (during irrigation season) were leached during the following rainfalls season.
In order to reduce an eventual soil salinization, farmers applied crops' rotation including rain fed crops and bare soil.Even though, using brackish water, wise irrigation management and regularly soil salinity monitoring are a sine qua non condition for soil and water resources sustainability. Figu

Table 4 .
Field cropping and irrigation system's characteristics during the period(May 2008-September 2008)Field measurements included supplied water volume and salinity, drained water discharge and salinity, water table level and salinity, soil water content and salinity.

Table 5 .
Diverted water volumes and accumulated salt mass

Table 6 .
Some statistics of the soil electrical conductivity (dS m -1 ) during the period: May 2008-June 2010

Table 7 .
Soil electrical conductivity variations during irrigation season

Table 8 .
Recorded and average national yields of some crops

Table 9 .
Irrigation water amounts and average drainage flow rates during irrigation season

Table 10 .
Water table level and salinity variationOn the beginning of irrigation season (05/05), the water table EC was equal to 5.5 dS m -1 .On 12/07 (current irrigation season), measured water table EC was 2.8 dS m -1 (equal to irrigation water EC).Allover irrigation season, water table EC remained lower than 3.6 dS m -1