An attempt has been made in this study to analyze the hotspots and cold spots of groundwater potential zones. First, a total of nine significant factors were identified (Slope, distance to faults, distance to rivers, river density, altitude, landcover, geology, soil texture and rainfall) and integrated to obtain the groundwater potential zones (hotspots and cold spots). Second, these factors were used as explanatory variables to understand the spatial configuration of groundwater potential zones. The ordinary least squared (OLS) model diagnostics showed that the R-squared and adjusted R-squared values of the explanatory variables are 0.71 and 0.87 respectively. Variance Inflation Factor (VIF) values of the OLS ranges between 1.04 (for slope) and 1.49 (distance to rivers) indicating the absence of redundancy among explanatory variables. Koenker (BP) statistic is statistically insignificant (p > 0.005) indicating that relationships modeled are consistent. The OLS regression model was subject to different tests to confirm its reliability. Among them, the F-statistic, Wald, Koenker (BP) and Jarque–Bera statistics tests stand out. Spatial autocorrelation tool (Global Moran's I) result showed that the residuals exhibit a random spatial pattern with the z-score of 0.01. This study proved that all the explanatory variables played a significant role towards demarcation of groundwater potential zones.

本研究尝试分析地下水潜在区的热点和冷点。首先，总共确定了九个重要因素（坡度，到断层的距离，到河流的距离，河流密度，海拔高度，土地覆被，地质，土壤质地和降雨），并进行综合以获得地下水潜在区域（热点和冷点）。其次，这些因素被用作解释变量，以了解地下水潜在带的空间配置。普通最小二乘（OLS）模型诊断表明，解释变量的R平方和调整后的R平方值分别为0.71和0.87。OLS的方差膨胀因子（VIF）值介于1.04（对于坡度）和1.49（与河流的距离）之间，表明解释变量之间没有冗余。Koenker（BP）统计数据在统计上不重要（p> 0.005），表明建模的关系是一致的。对OLS回归模型进行了不同的测试，以确认其可靠性。其中，F统计量，Wald，Koenker（BP）和Jarque-Bera统计量检验突出。空间自相关工具（Global Moran's I）的结果表明，残差表现出z值为0.01的随机空间格局。这项研究证明，所有解释变量对划分地下水潜力区都起着重要作用。空间自相关工具（Global Moran's I）的结果表明，残差表现出z值为0.01的随机空间格局。这项研究证明，所有解释变量对划分地下水潜力区都起着重要作用。空间自相关工具（Global Moran's I）的结果表明，残差表现出z值为0.01的随机空间格局。这项研究证明，所有解释变量对划分地下水潜力区都起着重要作用。