A Spatial Non-Stationary Based Site Selection of Artificial Groundwater Recharge: a Case Study for Semi-Arid Regions

In this study, a methodology was developed for AGR assessment, including site selection, non-stationary kriging, numerical modeling, and the long-term supply and demand sustainability. To achieve this, spatial variations maps of effective parameters were prepared using kriging techniques along with the Fuzzy logic method and GIS analytical functions for AGR site selection. To estimate how AGR is changing the water balance in the region, the Soil Water Assessment Tools (SWAT) hydrological model was coupled with an in house 2-D finite difference groundwater model. Based on the SWAT model results, by applying AGR, infiltration increased dramatically from 101 to 146 mm. In order to improve the model's accuracy, transmissivity was regionalized through non-stationary kriging by splitting the region into sub-regions where the characteristics of each variogram are different. By using non-stationary kriging, the statistic R-2 from single variogram to multiple variograms increased from 0.27 to 0.95, and the groundwater level was improved by 15%. Furthermore, for evaluating the sustainability of water resources, an index called Planning for Sustainable Index (PSUI) was utilized to measure AGR effects. PSUI, after applying AGR without changing in water withdrawal, was increased from 0.11 to 0.17 during 27-years time horizon. The case study is the aquifer in Qorveh Dehgolan sub-basin in Kordestan province, Iran. The results indicate that by applying AGR to the region, the rate of groundwater level reduction would decrease significantly for about 30%. However, for preventing the groundwater depletion, AGR should be accompanied by 50% reduction in withdrawals/pumping. The results show the significant value of utilizing a platform to assess the attributes and benefits of AGR. The proposed algorithm can be used for other geographical settings.