Evaluation of geostatistical interpolation methods on spatial representation of groundwater depth and nitrate concentration of Elalla-Aynalem wellfield, Northern Ethiopia

Abstract

Abstract Water quality degradation is one of today's key environmental challenges. Once contaminated, its treatment becomes difficult and prohibitively costly. Therefore, efficient and sustainable groundwater management is essential to conserve this vital resource. This is particularly true for the Elalla-Aynalem catchment, which serves as the primary water source for Mekelle town. This study aimed to produce optimized maps of groundwater table depths and nitrate concentrations using geospatial models, based on 123 data points for groundwater levels and 71 data points for nitrate concentrations. Empirical Bayesian Kriging (EBK) proved to be the most effective interpolation technique for groundwater table depth, while Simple Kriging (SK) was the best choice for predicting nitrate concentration, as determined by the lowest root mean square error (RMSE). The ranking of interpolation models for groundwater table depths was EBK > SK > OK > UK > IDW > RBF > LPI, whereas for nitrate concentrations, the models ranked as SK > EBK > IDW > OK > UK > RBF > LPI. These results demonstrate that geostatistical methods outperformed deterministic interpolation techniques for both parameters. The northwest and central regions of the study area were found to have shallow groundwater table depths, areas that also exhibited high nitrate concentrations, likely due to their proximity to pollutant sources. This research provides valuable insights for promoting sustainable groundwater exploration and implementing effective measures to protect groundwater from contamination.

Keywords

• Elalla-Aynalem
• Geostatistical model
• Groundwater
• Nitrate contamination