A study on characteristics and causes of flood season water pollution in a plain river network area based on grid online monitoring of water quality

Water pollution during the flood season has become an important factor restricting the improvement of water environment quality in China. The spatiotemporal changes and flood season pollution characteristics of surface water quality in Pinghu, a typical plain river network city in the Taihu basin, were studied based on the data from 26 automatic water quality monitoring stations. The key pollution control area was identified through correlation analysis, and the impact of different rainfall characteristics on surface water quality in the key control area was explored. In addition, the causes of pollution during the flood season in the key control area were analyzed using the absolute principal component-multivariate linear regression (APCS-MLR) model combined with field investigation. The results showed that the water quality of the river network in Pinghu City deteriorated significantly during the flood season. The comprehensive pollution indices of East Lake and its inflow and outflow rivers were 35.8%-67.9% higher than those during the nonflood season, and dissolved oxygen and total phosphorus (TP) were the principal over-standard indices. The middle section of the Pinghutang River, a river flowing into East Lake, was the core area for pollution control during the flood season of the entire river network in Pinghu City. The concentrations of ammonia nitrogen and TP in the Pinghutang River were greatly affected by rainfall. The decline in the Pinghutang River's water quality directly impacted East Lake (r=0.464, P<0.01) and indirectly affected downstream rivers (r>0.445, P<0.01). In terms of rainfall characteristics, rainfall amount had the most significant effect on ammonia nitrogen and TP in Pinghutang River (r>0.695, P<0.01), followed by the number of days of rainfall duration and the type of rainfall (r> 0.514, P<0.05). In contrast, the number of dry days before rainfall had no significant effect (r<0.245). The APCS-MLR analysis revealed that the increase in ammonia nitrogen and TP concentrations in the Pinghutang River during rainfall was mainly attributed to domestic sewage overflow (25.7%) and surface runoff pollution (24.9%). In addition to known factors, unknown sources, including river sediment or pipe network sediment, accounted for 40.0%. Due to the small design scale of some pumping stations and the lack of scientific scheduling, domestic sewage overflow was the primary cause of river pollution. The results from this study have important implications for the precise policy of water quality improvement in the river network during the flood season.