Study On Remote Sensing Retrieval Of Honghu Lake In Recent Ten Years Based On GEE

Lake is an important human water resources,with the socio-economic development and changes in natural conditions,facing the problem of water quality and nutrition,a direct impact on the lake ecological environment and human production and life.Honghu,as the largest natural lake in Hubei Province,has great influence on the economic development of the Jianghan Plain and the life of the residents.However,Honghu has a large area and has a large variation range.It is not only time-consuming to use the artificial detection and analysis method,Is not economical,and it is difficult to fully and timely response to the entire lake water quality distribution.Remote sensing technology,as an important technology of modern environmental monitoring,has great potential in water quality inversion monitoring.It is a comprehensive,real-time,fast and dynamic water quality information from remote sensing water quality.It is an important supplement to traditional water quality monitoring,and can provide scientific basis for Honghu water environment monitoring and management.In this paper,Honghu is the research area,and the Google earth engine cloud remote sensing processing platform is used as the analysis and processing platform.First,the image segmentation threshold is automatically obtained by using the Otsu method.The watershed area of Honghu Lake is extracted by threshold segmentation method.At the same time,through the monitoring data of the 8-year water quality monitoring station in Honghu from 2007 to 2016,total phosphorus(TP),total nitrogen(TN)and chlorophyll a(Chla)were tested for the water quality of Honghu Lake,and the band information with high correlation was found,Because the relationship between water quality parameters and remote sensing bands is complex,the fitting accuracy obtained by using the normal regression fitting function cannot get the inversion requirement,Therefore,a support vector regression machine with strong reasoning ability on small sample and nonlinear problem is used to return the model.The remote sensing water quality inversion model of Honghu nutrient is established respectively,the temporal and spatial distribution characteristics and changes of Honghu water quality were systematically studied.At the same time,the degree of water quality of Honghu Lake was evaluated,from the hydrological changes,human activities and other factors to explore the reasons for changes in water quality,the main conclusions are as follows:(1)The watershed area of Honghu Lake is relatively small,mainly due to the change of water period and dry season.Due to the construction of irrigation facilities and the construction of Honghu dike in the 1950 s and 1970 s,the area of Honghu waters is stable and the interannual variation is not.(2)Based on the SVM inversion,the inversion of SVM can be achieved with high accuracy.According to the situation of SVM inversion,the inversion accuracy is higher than that of Honghu calendar year,which can reflect the water quality of Honghu.Therefore,the use of remote sensing inversion of Honghu water quality,is the traditional way of monitoring the water quality is an important supplement in the rapid and comprehensive understanding of Honghu water quality distribution situation has a greater advantage.(3)Using the established model,the TP,TN and Chla concentrations of Honghu Lake were deduced,and the distribution map of the degree of nutrient distribution of Honghu Lake was made according to the comprehensive nutrition index.It can be seen from the analysis that the water quality of Honghu Lake is between the slight nutrition The trend of water quality in the time showed a trend of gradual improvement,but there were still repeated in different years.In space,water quality nutrition showed a decreasing trend from northwest to southeast.Water quality management will continue to be a long and arduous task.The overall study shows that the area of Honghu waters has stabilized and the interannual variation is small.The main changes are in the water area with abundant water during the wet and dry periods.The SVM regression is used to invert the water quality of Honghu Lake Which can be used as an important supplement to the water quality of the whole lake outside the traditional water quality monitoring.At present,the water quality of Honghu is in a mild eutrophication state,the nutrition degree of the northwest is higher and the water quality protection and management work is still arduous.