<p>Karstified aquifers respond rapidly to hydrological events, such as heavy rain storms or draughts. Our ability to predict the response of the aquifer after such events strongly depends on i) temporal and spatial resolution of the available monitoring data and ii) suitable modelling approaches to assess recharge at the respective level of detail. The study catchment, the Western Aquifer Basin (WAB), is Israel´s most important source for freshwater supply. The recharge area of the WAB has an area of 1,812 km<sup>2</sup>. Recharge is characterized by high spatial variability in topography and a high variability in precipitation and temperature, land use, and vegetation. Precipitation also shows a seasonal variability: while annual precipitation mainly occurs during the winter months accompanied by floods in the otherwise dry wadis (October to March, ca. 90 %), summer periods (April to September) are hot and dry, and precipitation decreases to nearly zero.</p><p>We employ SWAT to simulate the large-scale hydrological water balance (evapotranspiration, recharge, run-off) in the recharge area of the WAB on a daily and monthly temporal resolution. The SWAT model uses a SRTM DEM from NASA, soil maps from FAO, soil properties of the Harmonized World Soil Database, and land use maps from the ESA CCI project covering the time period from 1992 to 2015. These datasets are merged in SWAT into 361 Hydrologic Response Units with unique characteristics in soil, land use, and slope, respectively. The calibration of soil water balance model with SWAT-CUP employs monthly actual evapotranspiration and daily surface runoff data. Run-off was measured in hydrometric stations between 2004 – 2015. Evapotranspiration with a spatial resolution of 500 m x 500 m is obtained from the MODIS satellite mission and covers a period between 2001 and 2013 with individual time steps of 8 days. Calculated long-term groundwater recharge is compared with spring discharge measured during the period 1990 – 2013. Climate projections have been obtained with the RCM COSMO-CLM at resolution of 8km, under the IPCC RCP4.5 scenario, nested into the MENA-CORDEX domain.</p><p>The calibrated water balance model allows for scenario analysis for predicted shifts in climate until 2050 to address the impact of climate change on groundwater recharge. In addition to an increase in temperature, fewer but more extreme rainfall events are to be expected. Furthermore, the effect of future land use changes, such as expansion of farm land or urban areas, on recharge depth are analyzed. Finally, simulated high-resolution recharge provides an updated estimate for the currently developed groundwater flow model of the aquifer system. SWAT provides daily recharge for the equivalent porous medium model of the WAB, simulated by MODFLOW. One of our challenges is the calculation of recharge in the hilly region i) characterized by steep slopes and ii) vadose zones of several 100 meters of thickness. Our investigations are expected to provide information on the impact of shifts in climate and global changes on recharge processes and to illustrate the effect of short-term hydrologic events on water resources in large carbonate aquifers under Mediterranean climate.</p>

<p>Groundwater recharge is an important variable for sustainable groundwater resources management in regions affected by water scarcity. The specifics of the Mediterranean require adapted techniques to also account for climate change implying a higher frequency of extreme events. Appropriate techniques are highly relevant for recharge with low rates. We compare three methods for the Western Mountain Aquifer, a karst in Israel: soil moisture budget calculations at basin scale, empirical functions, and machine learning algorithms. Resulting recharge are compared with measured spring discharge.</p> <p><strong>Neural networks</strong> have the advantage of not requiring much knowledge about physical processes or hydrogeological and hydrological conditions, nor about model parameters. This data-driven machine learning algorithms learn the non-linear relationship between precipitation events and spring water discharge given a sufficient amount of training data is available. After training, the neural network could be used as a nonlinear function to model recharge of any predicted precipitation time series. However, this approach does not allow for any quantitative analysis of external forcing, such as land use, or internal parameter, such as soil characteristics, nor does it account for any expected future change in precipitation pattern.</p> <p><strong>Hydro-pedotransfer functions (HPTF)</strong> are based on empirical relationships between precipitation and recharge. HPTFs account for potential evapotranspiration, annual precipitation, land cover, and a critical water supply (a threshold when actual evapotranspiration depends only on atmospheric conditions). Resulting percolation rates consider i) vegetation types, ii) precipitation during the vegetation growth period, iii) runoff, iv) plant available soil water, and v) capillary rise. The application of HPTF to a karst aquifer has the advantage that only limited input data are required. However, our results indicate that HPTFs are not able to capture the rapid recharge component observed in karst systems and thus underestimate recharge.</p> <p>The <strong>Soil Water Assessment Tool (SWAT) </strong>employs a hydrological and soil moisture budget calculations. Objective functions are actual evapotranspiration and surface runoff. Evapotranspiration is obtained from MODIS remote sensing data. Calibration of actual evapotranspiration is especially challenging for summer periods due to the impact of vegetation and irrigation. However, the most relevant parameter determining daily recharge rates are water loss by surface-runoff and surface water storage in wadi beds generating episodic recharge.</p> <p>Impact of shifts in climate is considered by climate projections obtained with the RCM COSMO-CLM at resolution of 3 km, under the IPCC RCP4.5 scenario, nested into the MENA-CORDEX domain. However, we believe that changes in land use from natural vegetation (trees, grass-, and shrublands) to rain-fed agricultural area could possibly shift the water budget from deficit to surplus conditions (recharge dominated). During the period 1992 to 2015 natural vegetation decreased by 8% and urban areas increased by up to 6%, while (rain-fed) agricultural areas remained almost constant. We investigate if land use changes might have (a much) larger impact on percolation rates than the predicted climate change effect. Thus, in future recharge may be controlled and enhanced in regions with water scarcity by better management of land use employing an optimized combination between precipitation, irrigation, and crop type.</p>

<p>Karst aquifers provided 9.2 % of the world’s population with fresh water in 2016 (Stevanović, 2019), but due to their dual flow behavior they are highly vulnerable to anthropogenic impacts and shifts in climate. In the near future, 52 out of 356 Mediterranean aquifers will be exposed to more extreme climatic conditions, which will enhance their water stress if the water usage is not adapted to available water resources (Nußbaum, 2020). Therefore, accurate and high resolution numerical - and empirical models are essential to calculate the groundwater recharge and water availability in complex karst aquifers that cover ~ 14 % of the earth’s ice free land (Stevanovic, 2019).</p><p>During the last decades, several empirical equations have been developed to calculate the recharge for Israel´s most important source of freshwater, the Western Mountain Aquifer (WMA). These equations calculate annual groundwater recharge of the entire 1.812 km<sup>2</sup> recharge area based on annual or monthly precipitation data. We analyzed the applicability of several new methods, such as Soil & Water Assessment Tool (SWAT), HydroGeoSphere (HGS) and Hydro- / Pedo- Transfer Functions (HPTF) to estimate groundwater recharge with  a higher resolution as this is essential to calculate proper water fluxes though the vadose zone of karstic aquifers when precipitation is affected by a high variability in space and time.</p><p>The hydrologic balance models,  e.g. SWAT (Neitsch et al., 2009),  calculate the water balance on a daily basis for specified Hydrologic Response Units (HRUs), while generalized HPTFs (Wessolek et al., 2009) use soil-, land cover-  and climate data to calculate  annual percolation rates on a coarse grid, in our case 500 m grid size. The dual continuum model using the code HGS (Brunner et al., 2011) is able to simulated based on Richards’s flow equation down- and upward water fluxes in the unsaturated zone accounting for both, a rapid flow component though the high permeable conduit and a slow flow component through the rock matrix.</p><p>The comparison of these empirical and new methods for groundwater recharge estimation show significant differences for hydrological extreme years, while results are similar during years with precipitation rates near the average value. For example, the empirical equation of Guttman & Zukerman (1995) gives  highest recharge values of all approaches during wet years, while the equation of Abusaada (2011) and the SWAT-model calculates  highest recharge values of all approaches during  dry years. Overall, the mean recharge ranges from 120 to 177 mm/a which equals 25 – 37 % of the average precipitation between 1990 – 2018.</p><p>These recharge rates are calculated based on IMS climate data. However, for recharge values used in water resources management regional climate projections are needed. For Israel a high resolution CORDEX-MENA climate projection (Hochman et al., 2018) is available for RCP4.5, showing an increase in temperature and decrease of precipitation during the winter of 2.5 °C and 40 %, respectively. Based on these climate projections the  SWAT-model estimates, that the average groundwater recharge for 2050 – 2070 will be 16 % lower than the reference period between 1980 – 2000.</p>