Under the backdrop of global climate change and intensified human activities, groundwater pollution remains a serious problem worldwide. Assessing the quality of groundwater involves considering a non-linear system containing several factors and hierarchical structures, which still faces challenges. The indicators used to evaluate water quality are complicated and diverse. The large amount and irregular nature of the data on water quality introduce several challenges for evaluators, and create considerable uncertainty in the results of evaluation. Reasonably selecting the indicators of evaluation and assigning different weights to them is thus important in this regard.
In this study, we propose a coupled method to assess the quality of groundwater. We chose seven datasets on pore water in loose rocks from the Quaternary in the City as the basic data. We then used rough sets (RS) for parameter reduction, and calculated the weights of the indices by the RS in conjunction with the analytic hierarchy process (AHP). An experience factor was introduced to couple these methods to obtain the comprehensive weights of the indices, which were then used to dynamically evaluate the quality of groundwater.
Results showed that the determination of weights had a significant impact on the results of evaluation of groundwater quality. The RS was calculated based only on the weights of the dataset, but this was found to be inadequate for assessing the quality of water in the Jiangnan and Qijiazi areas. Moreover, the weights determined by the AHP relied on subjective experience, and excessively emphasized the role of NO2 - , which is inconsistent with the actual status of water quality in the Longtan Mountain and Jiangbei districts. The coupled RS-AHP model reduced the number of chemical parameters from nine to six (33.3%).