Factors controlling groundwater chemistry in an agricultural area with complex topographic and land use patterns in mid‐western South Korea

Shallow and bedrock groundwater from granitic aquifers were investigated for the hydrogeochemistry of major and minor constituents in an agricultural area. Nitrate concentrations were observed up to 49 mg/l as NO₃‐N, with 22% of samples exceeding the drinking water standard, which could pose a significant threat because most residents rely on groundwater as their drinking water source. Principal component analysis revealed three principal components (PCs): (1) nitrate contamination, contributed by major cations, Cl^?, SO and NO , (2) reduction processes positively involving Fe, Mn and B, and negatively involving dissolved oxygen and NO and (3) natural mineralization, involving HCO and F^?. Cluster analysis, performed on the PC scores, resulted in seven sample groups, which were successfully identified by total depth, elevation and land use. The nitrate‐contaminated groups had mixed land uses, with locally concentrated residential areas. Uncontaminated groundwater groups were found in the natural environment, including high‐altitude spring water and bedrock groundwater with a higher degree of natural mineralization. Shallow groundwater groups in paddy fields in lowlands were affected by reducing environments, of which one group was characterized by high Fe, Mn and B, and negligible nitrate. Groundwater with intermediate nitrate and lower Cl^? and SO was found primarily in hilly terrains with orchards and vegetable gardens, indicating lower contaminant loadings than lowland areas. Higher concentrations of F^? and nitrate were observed in the nitrate‐contaminated water, which seemed unlikely to be explained by groundwater mixing. The strong acidity generated from nitrification may infiltrate deeper into the aquifer, induce accelerated weathering of bedrock and result in the coexistence of F^? and nitrate, which may be an evidence of intense nitrate loading, leading to soil acidification. Multivariate statistical analysis successfully delineated hydrochemical characteristics of groundwater attained by natural and anthropogenic processes in an agriculturally stressed area with complex topographic land use patterns. Copyright © 2009 John Wiley & Sons, Ltd.