Spatiotemporal Differentiation Monitoring Of Suspended Particulate Matter Concentration In Shengjin Lake-Yangtze River Connecting Composite System Based On Multi-source Remote Sensing

Suspended particulate matter（SPM）is an important part of inland water.As an important parameter of aquatic ecology and water environment assessment,it plays a key geochemical role in the aquatic ecosystem.Its transportation and accumulation have direct and indirect effects on the water ecosystem and human activities,which will directly lead to changes in the morphology of the river and lake systems,and further affect the changes in the primary production,nutrients,and oxygen saturation of the water.Therefore,effective monitoring of the temporal and spatial changes of SPM in inland water is of great significance for mastering the law of SPM migration,formulating water environmental governance measures,and effectively protecting aquatic ecological resources.We take the Shengjin Lake-Yangtze River connecting composite system in Anhui Province as the study area.Simulate the band reflectivity of Landsat-8 OLI and Sentinel-2 MSI sensors based on the measured hyperspectral data of the water,and establish a variety of SPM retrieval models combined with the measured data of water SPM in the same period.Then compare the consistency of different retrieval models on Landsat OLI and Sentinel-2 MSI images,and select the SPM optimal retrieval model.Finally,the optimal retrieval model was applied to a data set of 82 images of Landsat-8 OLI and Sentinel-2 MSI from 2013 to 2019.According to the remote sensing retrieval image of SPM on the water surface to analyze the temporal and spatial differentiation of s SPM concentration in Shengjin Lake-Yangtze River Lake System,and the main influencing factors are studied.The main findings were as follows:（1）A water SPM retrieval model suitable for OLI and MSI sensors is established.The corresponding models all use R_rs（865）/R_rs（560）as independent variables and have the same form.The prediction accuracy of the retrieval model on the ground scale is high（Landsat 8 image:R²=0.797,RMSE=32.962 mg/L;Sentinel-2 image:R²=0.802,RMSE=32.673 mg/L）,but due to the difference in imaging dates,the unstable atmospheric conditions lead to large errors in the retrieval value on the satellite scale（Sentinel-2 images with a difference of 3 days:R²=0.756,RMSE=70.920 mg/L）.（2）Among the existing water atmospheric correction algorithms,the reflectivity of the OLI and MSI image bands obtained by using the MUMM algorithm under Seadas software has a high consistency,which is a prerequisite for ensuring the consistency of remote sensing products for the concentration of SPM in water.For the Landsat 8 and Sentinel-2 images,the consistency of the SPM concentration values retrieved using the same form of models（R²=0.935,RMSE=21.812 mg/L,MAPE=6.841%）and using the same model（R²=0.935,RMSE=22.107 mg/L,MAPE=6.985%）were obviously higher than that achieved by applying the optimal models with different forms（R²=0.885,RMSE=37.568 mg/L,MAPE=22.533%）.The consistency of the SPM concentration retrieved using the retrieval models with the same form was slightly higher than that using the same retrieval model.（3）The concentration of SPM on the surface of Shengjin Lake varies greatly in time and space,with values ranging from 10 to 500 mg/L.The high value areas are mainly concentrated in the southern part of the upper lake area and the entire lower lake area（greater than 160 mg/L）.In season,the concentration of SPM in Shengjin Lake is higher than that of the Yangtze River.The difference was small in summer,but large in winter and spring.During the year,the concentration of SPM in Shengjin Lake was slightly lower than that of the Yangtze River except in March and August.In other months,the concentration of SPM in Shengjin Lake was significantly higher than that of the Yangtze River.From the perspective of inter-annual changes,the concentration of SPM in Shengjin Lake was lower than that of the Yangtze River from 2013 to 2014,Between 2014 and 2016,there was little difference between them,and from 2015 to2019,the concentration of SPM in Shengjin Lake was higher than that of the Yangtze River.（4）The main influencing factors that cause the temporal and spatial differentiation of surface SPM in Shengjin Lake-Yangtze River were:the previous day’s rainfall had a synergistic effect on the concentration of SPM in the Shengjin Lake-Yangtze River,and the rate of change of SPM concentration in Shengjin Lake was greater than that of the Yangtze River,Which was also the driving factor for the temporal and spatial differentiation of surface SPM in Shengjin Lake-Yangtze River.The operation of the sluice to control the water level of the lake has led to a decrease in the area of Shengjin Lake in the recent years,which has led to an increase in the concentration of SPM in Shengjin Lake,while the concentration of SPM in the Yangtze River will not be affected.The concentration of SPM in Shengjin Lake-Yangtze River showed opposite trends with the increase of vegetation coverage in the reserve,but the rate of change of SPM in the Yangtze River was greater than that of Shengjin Lake.