Spatial and temporal variability in infiltration and soil characterization using ERT

Infiltration of the aquifer is one of the primary processes of the hydrological cycle. It plays a crucial role in the estimation of artificial groundwater recharge, surface runoff, and soil erosion. The electrical resistivity tomography (ERT) surveys were carried out to characterize the subsurface soil, and double-ring experiments were carried out at different locations in the study area. The results estimated from the double-ring infiltrometer data were compared with the infiltration rates obtained from various models such as Horton, Green-Ampt, Kostiakov, etc. The parameters are to be calibrated under the same hydrological site conditions, and these models’ water content and constants were estimated using the Graphical method. The best-fit models were estimated for each site, and the performance of the infiltration models was assessed based on the correlation coefficient, coefficient of determination, and root mean square error (RMSE). Horton model showed minimum RMSE, maximum correlation coefficient, and coefficient of determination, i.e. 5.24, 0.99, and 0.98, respectively, at most sites. The ERT data showed that sand is the predominant material available in the subsurface, and the saturated infiltration rate lies in the range of the sand. Nonlinear regression modeling was used to develop novel infiltration rate models from infiltration data.

含水层入渗是水文循环的关键过程之一，其在人工地下水补给量、地表径流量与土壤侵蚀量的估算工作中发挥着至关重要的作用。本研究通过电阻率层析成像（electrical resistivity tomography, ERT）勘探对地下土壤特性进行表征，并在研究区的不同点位开展双环入渗试验。基于双环入渗仪实测数据得到的入渗估算结果，与Horton、Green-Ampt、Kostiakov等经典入渗模型计算得到的入渗速率进行了对比分析。所有入渗模型的参数均在同一水文场地条件下完成率定，各模型的含水率与常数项通过图解法进行估算。本研究为每个研究点位选取最优拟合模型，并基于相关系数、决定系数以及均方根误差（root mean square error, RMSE）对各入渗模型的性能开展评估。在多数研究点位，Horton模型均表现出最优性能：其均方根误差最低（5.24），相关系数与决定系数最高，分别为0.99与0.98。电阻率层析成像勘探结果显示，研究区地下沉积物以砂质为主，饱和入渗速率处于砂质介质的合理区间范围内。本研究通过非线性回归建模方法，基于入渗实测数据构建了新型入渗速率模型。