Study On Estimation Method Of Pollutant And Suspended Mass Load In River

Accurate estimation of river material load can help us grasp the temporal and spatial distribution of river material transport,and provide decision-making basis for river water environment control projects.In order to accurately calculate the load of river material timely,it is theoretically necessary to continuously observe the flow and concentration of the controlled section for a long time.In the actual monitoring work,this can not be achieved.In China,the water quality observation period of large and medium-sized river water quality stations is generally 15 days or 1 month.Based on these observations,the accuracy of different annual load estimation methods has always been a hot issue of concern.Based on the observed data series,this paper attempts to analyze the errors of four kinds of flux estimation methods(Average Method,Ratio Method,Numerical Integration Method and Regression Method)by selecting data at different time intervals,so as to provide reference for the estimation of water quality flux in practical work.The basic data used are:daily water and sediment data of main hydrological stations in the Yangtze River Basin in 1980;flow and water quality monitoring data of Hankou Station in the Yangtze River from January 1 to September 11,2018(flow data monitoring interval is 1 h,water quality data is 3 h).From the aspects of unbiasedness and stability of the algorithm,the accuracy and scope of application of each method are analyzed,and the main conclusions are as follows:(1)In the 30-day sampling period,the average method,ratio method and regression method are applied to estimate the nitric acid nitrogen and DOC fluxes at Hankou Station.The estimated mean is close to the reference value,the estimated value fluctuates in a small range,and the algorithm has relatively good unbiasedness and stability.The numerical integration method has a good performance in estimating the ammonia nitrogen fluxes at Hankou Station.(2)Regression method is applied to estimating suspended load flux in Yangtze River basin in 30-day sampling period.When the sampling period is larger than a critical value and the area controlled by the station is larger than a critical value,the performance of regression analysis method varies sharply.The estimated value of flux may be several times different from the real value,and the algorithm is basically invalid.In this study,the critical values of sampling cycles at different stations are quite different,while the critical values of watershed area are between 10,000 km~2 and 100,000 km~2.(3)Except for regression methods(reg01,reg02),there is a significant correlation between the performance of the other nine algorithms and sampling period and the area controlled by stations.The longer sampling period and the smaller area of the basin controlled by the station will lead to the worse unbiasedness and stability of the algorithm.For regression method,this correlation only exists in a specific interval.(4)Based on the calculation results of suspended load flux,the optimal flux estimation method for each station under the conventional sampling period(30 days)is determined,and it has been successfully applied to the analysis of interannual and along-course variations of three pollutant fluxes in the Three Gorges reservoir area.