Spatiotemporal distributions and mixing dynamics of characteristic contaminants at a large asymmetric confluence in northern China

Abstract

Previous studies of confluences have mainly concentrated on their hydrodynamic characteristics and morphological dynamics. However, the increasing environmental pollution in rivers has resulted in abrupt variations in water quality at confluences, thus affecting the ecosystems in river networks. Therefore, more attention should be paid to the pollutant distribution and mixing processes at confluences. In this study, several field experiments on pollutant concentrations were conducted at a large asymmetric confluence with a discordant bed and wide-shallow channel in northern China, where a heavily polluted tributary known as the Fen River joins the Yellow River. The results displayed that the heavily polluted tributary seriously affected the water quality at the confluence, especially in the downstream channel near the tributary. The contaminations at the confluence present a relatively strong temporal variability. The organic matter and dissolved oxygen show a similar seasonal variability with each other whereas ammonia nitrogen and total phosphorus are also similar to each other, but they have a different variability from the first group. These pollutants mostly demonstrated the same spatial distribution characteristics, with large horizontal concentration differences at the confluence. The standard deviation of the cross-sectional concentration gradually decreased with the downstream mixing of contaminants. Nonetheless, under conventional hydrological conditions, these contaminants were not completely mixed through a length of 27 river widths. Compared with a concordant bed, this discordant confluence exhibited a relatively fast mixing rate. The key factor affecting the mixing process at the confluence was the discharge ratio. Exceptionally, during the irrigation period (March–April), owing to the extremely low discharge of the tributary, the main stream flowed backward into the tributary, which enhanced the mixing motion to constitute an approximately full mixing regime at the mouth of the tributary. These results are of great importance in understanding the degree of contamination, spatiotemporal distribution and mixing dynamics of characteristic pollutants at complex river confluences.