Introduction

Water quality and quantity have become of great importance in Jordan in the last four decades. The available water from the existing renewable resources are projected to fall to less than 91 m3/c.d by the year 2025, which is very low in comparison with the international water poverty line of 1000 m3/c.d (Al Smadi et al. 2010). Water shortage has forced Jordan to take in consideration all available water resources to bridge the gap between demand and the available supply (Al-Tabbal and Al-Zboon 2012).

This study focuses on the Azraq groundwater basin which is considered one of the most important basins in Jordan. The groundwater abstraction from Azraq basin exceeded 50 MCM/year and is used to supply Amman, Zarqa and Jordan-northern governorates with fresh water supply (Amro et al. 2001). Azraq Oasis is considered the heart of the basin where millions of migratory birds reset in every year during their travelling from Siberia to Africa and vice versa, therefore, any water shortage is expected to negatively impact the ecosystem in this basin (Al Eisawi 2005, Ta’any et al. 2014).

Climate change is likely to be the predominant scientific, economic, political and moral issue of the twenty first century (Hansen and Sato 2011). Climate change is defined as a long-term change in the statistical distribution of weather patterns over periods of time that range from decades to millions of years. This change may be in the average weather conditions or a change in the distribution of weather events with respect to an average values (EPA 2014).

The change in climate directly affect water resources quantity and quality; in some areas causes floods transport huge quantities of sediments leading to lower water quality, while other regions suffer from increasing groundwater withdrawals due to draughts.

Internationally, a study in the United States concluded that because of the alterations in precipitation patterns, types and intensities along with increased temperature; the effect on water resources is clearly evident (Nolan et al. 2013).

Annual runoff simulated model for Colorado River basin showed expected decreasing in runoff by 14–17% through the period from 2010 to 2098 (Christensen et al. 2004).

Climate change impacts on Chalk aquifer of the Marlborough, Berkshire Downs and South–West Chilterns in the UK were studied for the year 2080 and it was found that there will be 4.9% reduction in annual potential groundwater recharge (Jackson et al. 2011). Another study investigated the impact of climate change on water resources in north-western England for (2070–2100) indicated a reduction in overall yield by 18% (Fowler et al. 2007).

A study to assess the potential impacts of climate change on the hydrology and water resources of the Nile River basin found that in the period 2010–2039, the streamflow is expected to increase due to increase in precipitation, but in the next periods 2040–2069 and 2070–2099, precipitation is expected to decrease and evaporation to increase resulting in decreasing streamflow (Beyene et al. 2010). Future climate change scenarios in Iran using global climate change model were studied for the period of 2010–2100 and indicated that ground water recharge will decrease by 50–100% in the eastern region in the half of the century (Abbaspour et al. 2009). Effects of climate change on flow regimes of twenty-five catchments (with area ranging from 6 to 1293 km2) in central Sweden were investigated referring to incremental rainfall and temperature increasing and decreasing scenarios. The study found an increase in evaporation by 23% and a decrease in snow water equivalent by 79% for 15 climate change scenarios expected in the future (Xu 2000).

Abdulla and Eshtawi (2015) studied climate change impacts on runoff, sediments and ground water recharge for King Talal Dam watershed located in Jordan using SWAT model under different scenarios. The scenarios  included: ± 20% change in rainfall, and 1, 2 and 3.5 °C increases in average temperature. The study showed that climate warming can impact runoff, groundwater recharge and sediment yield in the basin and greatly influenced by significant changes in rainfall volume. Climate change impacts were studied in Mujib basin in Jordan using SWAT model. Ten incremental scenarios representing dry, normal, and wet events were tried. Findings indicated that dry scenarios lead to about 20–50% reduction in annual precipitation and surface runoff and the wet scenarios estimated that annual precipitation and surface runoff to increase up to fourfold the baseline values (Abu-Allaban et al. 2014). Climate change impacts on Azraq basin were studied by Ta’ani et al. (2014) using water evaluation and planning program (WEAP) based on three estimations; increasing 2 °C in temperature, decrease of 20% in rainfall and 20% decrease in ground water recharge. It was found that these changes are expected to directly affect the water resource balance in Azraq basin. At Jordan River, the global climate change model was used for precipitation simulation, the model showed decreasing trend in rainfall by 11.5–20% through the period (2031–2099) (Smiatek et al. 2011). A simulation study for period 2040–2099 and based on precipitation records (1961–1990) for eastern Mediterranean and Middle East region, found that Jordan’s precipitation will decrease 17% by mid of this century and 21% by the end of the century (Chenoweth, et al. 2011).

The SWAT model was used to study the impact of climate change on water resources in north of Jordan by applying different 20 scenarios. The simulation results of these scenarios showed that precipitation is the major factor that affects the vulnerability of surface runoff, while the temperature plays a minor role in changes of surface runoff amounts (Hammouri et al. 2016). In 2008, Abdulla and Al Omari studied climate change in Amman-Zarqa basin. The study analyzed twelve scenarios and utilized the surface infiltration base flow (SFB) model. The results indicated that total runoff will be reduced by approximately 12% (Hadley model) and 40% (Max Planck Institute MPI model). A study in Azraq basin based on different scenarios showed that the increase in temperature of 2 °C with precipitation fluctuation will decrease the overall annual ground water recharge. The results indicated that Azraq basin may be under stress within 18 years because of climate change (Al-Zu’bi 2009). Trend analysis using different statistical analysis tests were investigated for temperature over Jordan using historical data series from 30 to 83 years before 2005. The study concluded that there is no statistical trend of the annual precipitations. Annual maximum air temperature records do not show clear trends, but annual minimum temperatures have significant increase trends (Hamdi et al. 2009). Incremental scenarios for precipitation and temperature for Zarqa River watershed for years 1980–1994 with applying hydrological model Basin-HSP,F concluded that if temperature increase by 3.5 °C and no change in rainfall, the impact on runoff is insignificant but with pronounced effect on ground water recharge (Abdulla et al. 2008).

Hydrological modeling for two time periods (1961–1990 and 2070–2099) indicated mean annual temperature increase by 4.5 °C, and 25% decreases in the mean annual precipitation, while total runoff at the outlet of the upper Jordan catchment was predicted to decrease by 23% and it is accompanied with a significant decrease of groundwater recharge (Kunstmann et al. 2007). Amman precipitation trend analysis for years from 1922 to 2003 using change point method showed a change in the annual number of rain days and annual total rainfall in mid of 1950s and concluded that the impact of a decline in the rainfalls may result in serious effects in the desertification, drought, water resources and agriculture (Smadi and Zghoul 2006).

However, very few studies have been conducted in Jordan concerning climate change, most of them assumed different scenarios for temperature and rainfall change, and they did not predict the magnitude of the changes.

In this paper, possible climate change in a very important basin is studied utilizing RClimDex for six rainfall stations and two metrological stations that almost cover the study area. Impact of climate change on water streamflow is simulated using SWAT model.

Methodology

Data

The data used in this study are grouped into two main categories; hydrological which includes rainfall and streamflow and metrological which is comprised of: temperature, relative humidity, evaporation and wind speed. All data are collected from the Ministry of Water and Irrigation (MWI). Four rainfall stations, two streamflow gauges and two meteorological stations are used to study the effect of climate change on temperature, rainfall and streamflow (Fig. 1). The daily data for the period 1980–2014 cover the observations of six variables: minimum air temperature, maximum air temperature, total rainfall, relative humidity, wind speed, and evaporation. Two wadis were chosen to study the streamflow: Al Butum and El Janab, with data covering the period from 1996 to 2014. The data for the period from 2007 to 2009 is not available for unknown reason related to the device monitoring; therefore, to avoid any mistakes these data was considered as missing and excluded from all the analyses (MWI 2015). The missing values were coded with the appropriate value according to the software used in the analysis. Outliers’ analysis was conducted using RClimDex software package, the outliers are daily values outside a region defined by the user and the procedure are discussed within the methodology section then the results are used for further analyses.

Fig. 1
figure 1

Hydrological and meteorological stations in Azraq Basin (Palestine Grid Belt)

Full size image

Data quality control

Measurements reliability of the hydrological and climatological data is the essential foundation for the quantitative climate analyses. Although there are universal standards and recommendations for instruments installation, calibration and observation; there are differences in the observations from station to station in the same country and in the same station due to many factors such as changes in instrumentation and observation practices, and the modifications of the environmental conditions of the site which include rapid urbanization or development of irrigation in the study area (Kang and Yusof 2012; Sahin and Cigizoglu 2010).

Therefore, three absolute homogeneity tests are used to detect any variation in the time series to extract meaningful statistics and characteristics from these data, using AnClim software which has been specially created for climatologic purposes (Stepanek 2008). The tests that are used in this research are; Pettitt’s test, standard normal homogeneity test (SNHT), and von Neumann’s test (Taxak et al. 2014; Carvalho et al. 2014).

Indices calculations

The linear trend analysis model and the statistical parameters for 27 climate indices are found by RClimDex software package. RClimDex is recommended by World Metrological Organization (WMO) to determine climate change indices and was used successfully by many investigators in different countries (Baddour 2015; Gbode et al. 2015; Santos 2011; Tank et al. 2009).

SWAT model

SWAT (Soil and Water Assessment Tool) was developed from several earlier models by Arnold for the United States Department of Agriculture in the early 1990s (Krysanova et al. 2009). The model was developed to predict the impact of land management practices on water, sediment and agriculture chemical yields in large complex watersheds with varying soil, land use and management conditions over a long time period (Gassman et al. 2007). In this research, ArcSWAT-ArcGIS interface is used.

SWAT model requires several types of data consisting of: topography, land use, soil and climate, etc. SRTM data (90 m resolution) is used for watershed delineation (GIS-web, website). From GLC (2000) database Land use/land cover map was downloaded and re-projected to Palestine Grid Belt using ArcGIS tools with resolution of 30 m. Soil dataset has been downloaded using HWSD-Viewer which provides a simple tool to consult the data contained in the Harmonized World Soil database. The HWSD consists of 1.0 km raster image and an attribute database. It was also re-projected to the same projection and all necessary inputs for the model were extracted from the same database for each soil type including; name, soil texture, hydrological soil group (HSG), soil depth, rock fragments, and organic carbon content (FAO 2012). SWAT requires daily or monthly values of precipitation, maximum and minimum temperature, solar radiation, relative humidity and wind speed for modeling. The model needs at least 5 years of data for better estimation of the results through priming (Panhalkar 2014).

SWAT model was used successfully in many researches for different purposes such as for sediment and water yields in Taiwan (Huang and Lo 2015), evaluate climate change in agricultural watershed in the USA (Ficklin et al. 2009), predict surface water flow to the Danube Basin (Pagliero et al. 2014), assess the amount of sediment into the Kaw Lake in Oklahoma/United States (Srinivasan and Daggupati 2014), simulate the water flow in the Buyo Lake/west of Cote d’Ivoire (Koua et al. 2014), assess climate change impacts on surface runoff in Mujib basin (Abu-Allaban et al. 2014), long-term impacts of climate change on water resources in southern India (Nune et al. 2013), and to simulate hydrology, soil erosion and sedimentation in Wala dam catchment (Ijam and Tarawneh 2012).

Azraq basin is divided into main catchments and major streams that flow into Azraq depression but only three wadis are gauged and two of them are used in this research (wadi Al Butum and wadi El Janab). The total number of sub-watersheds that were created is 59 for both of the catchments Al Butum and for El Janab. The total catchment area is 439.96 km2 for Al Butum and 496.89 km2 for El Janab.

The period of the streamflow records covers the last 18 years since 1996 and with three missing years (2007–2009). The total used streamflow records were 15 years divided into two periods, one for calibration (1996–2006) and the next period for validation (2010–2013). The used model is soil and water assessment tool (SWAT). The selected model is very efficient and can analyze several hydrological processes in the same run. Soil and water assessment tool SWAT is used to calculate the streamflow for the gauged catchments in Azraq basin.

Model calibration

Model calibration is necessary for preliminary testing of a model results. Model calibration is also essential for the successful use of any hydrologic and water quality simulation. The observed versus simulated streamflow are drawn to give a set of data points, linear regression gives a formula for the line most closely matching those points, it also gives value to say how well the resulting line matches the original data points.

Based on the sensitivity analysis studies (Abbaspour et al. 2015; Arnold et al. 2012; Feyereisen et al. 2007), the following parameters were identified as sensitive in predicting the streamflow which include: curve number (CNII), plant uptake compensation factor (EPCO), initial soil–water storage expressed as a fraction of field capacity water content (FFCB), groundwater delay time (GW_Delay), base flow alpha factor (Alpha_Bf) and surface runoff lag coefficient (Surlag). Statistical parameters can be calculated to check and improve the correlation between the simulated and the observed streamflow data or any resulted data are needed to be calibrated. The sum of square differences SSDmo and relative error RE are calculated for the average monthly streamflow observed and simulated (Champathangkham and Pandey 2013). Other statistical coefficient used is Nash–Sutcliffe ENS coefficient; where it ranges from 0 to 1. When the value of ENS is greater than 0.65, this indicates very good model prediction, whereas lower ENS indicates poor model prediction (Getachew and Melesse 2012). Coefficient of correlation is also calculated to detect the stronger of the trend between the two datasets and improve the calibration process. Model validation is an important step to check if the model can predict streamflow for the study time period or the conditions to be studied. The coefficient of determination is used to validate the performance of the model over the validation period.

Results and discussion

Data quality control

According to the homogeneity tests, all stations passed Von Neumann’s and SNHT tests for rainfall data (Table 1), while some stations failed in Pettitt’s Test which indicates change in the rainfall gauge locations or replacing the old devices with new ones. Regarding flood data, Al Butum station (F0023) passed all the tests, while El Janab station (F0022) passed all the tests except Pettitt’s test (Table 2).

Table 1 Homogeneity tests results for total annual rainfall data series
Full size table
Table 2 Homogeneity tests results for gauged flood wadis data series
Full size table

For maximum and minimum temperature, Al-Safawi station (F0002) passes the three tests while Azraq station (F0009) failed in the three tests, and accordingly the data of F0009 was analyzed and arranged using linear regression to get reliable data (Table 3). The regression is conducted relative to Al Safawi station where the statistical parameters are very close for both of them.

Table 3 Homogeneity test results for mean annual temperature data series
Full size table

Climate change trend analysis

The RClimDex-produced results are tables and figures that demonstrate the linear regression model. The software calculated the coefficient of determination as percentage, slope of the linear trend model with direction of the trend. In this paper, the trend is considered significant if p value is less than 0.05 (New et al. 2006; Zhang et al. 2005).

Temperature trend analysis

Table 4 presents the linear air temperature extreme indices for Azraq and Al Safawi stations with p value for each index. Accordingly, monthly max value of daily mean temp (TMAXmean), Tropical nights (TR20), monthly maximum value of daily maximum temp (TXx), monthly maximum value of daily minimum temp (TNx) and cool days (TX10p) were found to be statistically significant climate change indices.

Table 4 Annual linear trends of the extreme climate indices of air temperature over the study area (p value < 0.05)
Full size table

Warm nights (TR20) were the most noticeable indices (p < 0.02) with confidence interval of 98%. The stations showed positive trend with slope ranged between 0.847 and 0.424. The results indicated that in the last 46 years, the number of nights at which temperature exceeds 20 °C increased by 19 nights for Azraq and 39 nights for Al Safawi. It is also expected to increase to 23 nights and 47 nights at the end of the year of 2030 for Azraq and Al Safawi stations, respectively.

The monthly maximum value of daily maximum temperature (TXx) (p = 0.048) and the monthly maximum value of daily minimum temperature (TNx) (p = 0.03) showed noticeable statistically significant increasing in trend. In the last 46 years, there was an increase by 2 and 3 °C in average over the study area for TXx and TNx, respectively.

For indices of cold days (TX10p), both stations showed significant decreasing trend (p = 0.01) with decrease rate of 0.157. The results also indicated significant decreasing trend in cool nights (TN10p) (p = 0.10) with confidence interval of 90%, where the slope of trend was 0.117. Concerning temperatures of the coldest days and coldest nights in each year (TXn and TNn), the slope of the trend line ranged between (+ 0.046 to + 0.041) and (− 0.018 to − 0.016) for TXn and TNn, respectively, despite that the change is statistically insignificant. Cold spell duration indicator (CSDI) has negative trend of value 0.317 (p = 0.067). The stations indicated increasing (p < 0.15) trend in DTR indices, where the slope of trend ranged between 0.011 and 0.01 as shown in Table 4, but this change is statistically insignificant.

The percentage of days exceeding the 90th percentiles (TX90p) gave positive trend with slope of 0.068 but the trend is considered insignificant. Moreover, insignificant trends were found in hot days (SU25), the percentage of nights exceeding the 90th percentile temperature (TN90p) and warm spell duration (WSDI).

Mean maximum values and mean minimum values for monthly temperature were calculated and the results show significant trend in maximum monthly mean with p value was less than 0.05, while the mean minimum monthly did not give significant trend (p = 0.301). During the last 46 years, the mean maximum temperature increased with 1 °C in average while an increase of 0.5 °C in mean minimum temperature was detected.

The mean annual temperature over Azraq basin during the studied 46 years has increased by 0.75 °C. These findings nearly comply with a study-discussed climate change implications in Jordan over the past 50 years and it was concluded that the local temperature in Jordan has increased by 1.5–2.0 °C (Matouq et al. 2013). Other study found that the temperature in the Mediterranean region for the period 1961–1990 increased by 3.5–7 °C (Lelieveld et al. 2012).

Monthly temperature extremes (January and July)

Figures 2, 3, 4 and 5 show the maximum of maximum, average of maximum and the minimum of maximum temperature for July for Al Safawi and Azraq stations, respectively. The result of T test indicated that the trend of the maximum temperatures is significant (T > 2.016), while insignificant trend for the minimum temperature is noticed (Stepanek 2008).

Fig. 2
figure 2

Monthly maximum temperature in July Al Safawi

Full size image
Fig. 3
figure 3

Monthly maximum temperature in July Azraq

Full size image
Fig. 4
figure 4

Monthly minimum temperature in January Al Safawi

Full size image
Fig. 5
figure 5

Monthly minimum temperature in January Azraq

Full size image

It is determined from the figures that for Al Safawi and Azraq, maximum of the maximum temperature will increase by 0.75 °C in July for the next 15 years till 2030 and the minimum of maximum temperature will increase by 0.9–0.97 °C for the same period.

Monthly minimum temperature shows variation in slope from increase to decrease for minimum of minimum and average of minimum, respectively. T test results for the slope showed insignificant trend of the minimum temperature.

Temperature frequency analysis

Frequency analysis for temperature is defined as the count for the temperature values exceeded or less than specific value over the study period. Trials were made to find this specific value according to the T test results. For maximum temperature, the count of values that is greater than 38 °C was selected and for minimum temperature, the count of values that is more than 20 °C was considered.

Frequency analysis indicated that the frequency of days in which the maximum temperature exceeded 38 °C increased and there was increase in the minimum temperature count of values that are exceeding 20 °C in the last 46 years (T values > 2.022) (Figs. 6, 7, 8 and 9).

Fig. 6
figure 6

Maximum temperature frequency (values exceeded 38 °C) in Al Safawi Station

Full size image
Fig. 7
figure 7

Maximum temperature frequency (values exceeded 38 °C) in Azraq Station

Full size image
Fig. 8
figure 8

Minimum temperature frequency (values exceeded 20 °C) in Al Safawi Station

Full size image
Fig. 9
figure 9

Minimum temperature frequency (values exceeded 20 °C) in Azraq Station

Full size image

Based on the slope of trend, it is expected that the count of days for temperature exceeding 38 will increase by 6 and 5 days during the next 15 years for Al Safawi and Azraq, respectively.

Rainfall trend analysis

Table 5 presents the annual linear trends of the extreme climate indices for precipitation on the study area. The current investigations indicate insignificant trend of all precipitation indices for almost all of the analyzed rainfall stations. Some stations show significant trend (decrease) in rainfall intensity (Deir El-Kahf, Al-Safawi). It can be concluded that there is no change in total annual precipitation over the study area which comply with previous study conducted by Matouq et al. (2013). Similarly, Hamdi et al. 2009 reported insignificant trend in the annual precipitation in Jordan.

Table 5 Annual linear trends of the extreme climate indices for precipitation over the study area
Full size table

Streamflow future forecasting model

Streamflow is affected by topographic and geological characteristics of the basin. The northern part of the basin is covered by Jebel El-Arab basalt flows which extends from Syria and continues through Jordan to Saudi Arabia (MWI 2015). The highest point of the basin is located at Jebel El-Rimah area with an elevation more than 1200 m asl (above sea level) and the lowest point of the basin is located at the Azraq depression with an elevation of about 500 m asl. Azraq basin on Google image is presented in Fig. 10.

Fig. 10
figure 10

Geological map for the study area (El-Naqa 2007)

Full size image

Only the mean maximum temperature has significant trend during the study period (Figs. 11, 12), so the climate change that will be evaluated in this study will rely on the significant increase in the maximum temperature that will be reflected in the simulation of streamflow for the next 15 years.

Fig. 11
figure 11

Indices monthly mean value of daily maximum temperature Al Safawi forecasting till 2030

Full size image
Fig. 12
figure 12

Indices monthly mean value of daily maximum temperature Azraq forecasting till 2030

Full size image

The model was calibrated according to the sensitivity analysis studies (Abbaspour et al. 2015; Arnold et al. 2012; Feyereisen et al. 2007) then trial and error was made using manual calibration tool combined with Excel until the best correlation and correspondence between simulated and observed streamflow was achieved (Table 6).

Table 6 Parameters used in SWAT model calibration
Full size table

The coefficient of determination for Al Butum is 0.96, coefficient of correlation is 0.98. The sum of square differences is 0.04 mm2 and relative error within good range of 14.8%. Nash–Sutcliffe (ENS) coefficient was more than 0.65. Coefficient of determination for El Janab was 0.81, coefficient of correlation was 0.90. Sum of square differences is 0.016 mm2 and relative error within good range 24.8%. Nash–Sutcliffe (ENS) coefficient is more than 0.65.

As well as for both catchments, the mean long-term monthly streamflow was very close to the simulated results in addition to the standard deviation. The coefficient of determination for the calibrated and validated periods are more than 0.5, this means that the model is calibrated and can be used to provide more data and forecast the streamflow.

The impact of increasing in temperature is used to predict the future changes in streamflow if present. The potential evapotranspiration increased in accordance to the increase in temperature and it is clear that the evapotranspiration will increase significantly in the predicted period; for Al Butum it will be 4.74% and it will be 5.32% for El Janab as shown in Figs. 13 and 14.

Fig. 13
figure 13

Al Butum Wadi average monthly potential evapotranspiration

Full size image
Fig. 14
figure 14

El Janab Wadi average monthly potential evapotranspiration

Full size image

In accordance with the expected temperature increasing within the next years, the streamflow was simulated. The streamflow will decrease by 1.51% for Al Butum and 1.02% for El Janab Wadis (Figs. 15, 16).

Fig. 15
figure 15

Al Butum Wadi total simulated yearly streamflow (1996–2030)

Full size image
Fig. 16
figure 16

El Janab Wadi total simulated yearly streamflow (1996–2030)

Full size image

The streamflow decrease is very small and can be neglected. The runoff analysis is mainly affected by rainfall and since no significant trends are found in rainfall, the decreasing is very small. Al-Zu’bi 2009, used the stochastic model and found that Azraq Basin as a whole, the change in temperature reduced the volume of surface water by 2.4%. Abdulla and Al Omari, 2008, found significant decrease in runoff of Amman-Zarqa basin up to 12% using Hadley model.

Conclusions and recommendations

In this paper, possible climate change in a very important Azraq groundwater basin in Jordan is studied utilizing RClimDex for six rainfall stations and two metrological stations.

The results indicate that the mean maximum annual temperature is projected to increase by 0.3 °C (2.0%) with confidence interval of 95% for the next 15 years (2015–2030). There is no significant trend for the mean minimum annual temperature during the study period. Accordingly, the mean annual temperature is expected to increase by 0.15 °C for the next 15 years (2015–2030) which is about 1.0% with confidence interval 95%. The frequency of days in which the maximum temperature exceeds 38 °C and the minimum temperature exceeds 20 °C is expected to increase significantly by 6 and 5 days for both of them, respectively, during the next 15 years. The trends for rainfall are insignificant over the study period (30–45) years.

SWAT model is capable to simulate the streamflow and provide highly accurate values after calibration and validation. The results show that evapotranspiration is expected to increase significantly in the predicted period with 4.74 and 5.32% for Al Butum and El Janab wadis, respectively. The runoff analysis was performed and it was found that the streamflow is projected to decrease slightly.

Climate change impact is insignificant for both rainfall and streamflow. Temperature as a driven force, it is expected to affect human life activities and agriculture both rain fed and irrigated areas, therefore, more attention should be taken in the design and analysis for both agriculture and urban planning projects. It is a need to have more observations for streamflow within the study area to cover all major wadis to ensure high simulation results in respect of calibration and validation.

More attention should be taken into account to the temperature and rainfall observations to provide resource of complete and reliable data. Climate change impacts on water resources should be taken into consideration in the future strategic water budget plans especially that this ground water basin is suffering due to high abstraction. The decrease in streamflow and the increase in temperature are predicted to affect the ground water abstraction and recharge. It is recommended to study impact of climate change on ground water in further researches.