Catchment scale land use and hydrogeomorphology relationship in assessing river water quality using SWAT and RDA modelling in the Damodar River basin, West Bengal, India

Identifying the sources of river water pollution is the initial step to ensure access to clean water for the population. Water drops from a distinct place in the basin that reaches the river, resulting from the interaction of rainfall, slope, soil, groundwater, etc., with the land use of the basin. Non-point sources of pollution are affected by local geomorphology controlled by local climate. The SWAT model considers all the land use classes and physical parameters to estimate the discharge, runoff, groundwater flow, etc. All these affect non-point sources of pollution. Multiple Linear Regression (MLR) and Redundancy Analysis (RDA) determine the most significant variables among the eight (industries and urban areas, open land, vegetation cover, discharge, groundwater flow, water yield, runoff, and distance from industries and urban areas) water quality factors. MLR and RDA (in step mode) filter only significant variables that control the 24 water quality parameters in pre-monsoon, monsoon, and post-monsoon seasons. In the pre-monsoon (dry) season, the adjusted R² value is 0.42 higher than the other seasons due to the dominance of point sources of pollution. Pollution load decreases with increased distance from the point sources of pollution. MLR and RDA identify industries and urban areas as the dominant pollution sources in the Damodar River basin. The contribution of groundwater reduces the pollution load in pre-monsoon season but increases the turbidity in river water. The agricultural area significantly affects pollution during the monsoon season due to the washout effect from non-point sources of pollution. Vegetation cover is capable of substantially reducing the pollution load. The research findings can help understand the importance of non-point sources of pollution that can be mitigated through land use management practices.