Study On Monitoring The Dynamic Change Of Plateau Lake Water Body Based On Multi-source Remote Sensing Data

As the main component of the terrestrial hydrosphere,the water level changes directly respond to the local topography and geological structure,ecology,climate and natural environment changes in the area where the watershed is located.The QinghaiTibet Plateau is the most densely distributed and largest area of lakes in China.In response to global climate change,the surface area of plateau lakes generally shows an expansion state.Due to the large extent of the Qinghai-Tibet Plateau,harsh climatic conditions,obvious regional differences,rugged terrain and inaccessibility,it is difficult to obtain geographic information,and there is little information about lakes within the plateau.The spaceborne single-photon lidar ICESat-2 can accurately measure inland lake levels and the depth of subglacial lakes.It can be used as the main data source for the lack of field observation areas,and can also provide accuracy verification for the water level data of rivers and lakes obtained by other remote sensing observation methods.This contributes to the understanding of global changes in water resource availability and provides reliable support and assurance for the monitoring of lake dynamics changes in long time series.In this paper,we take Qinghai Lake,Hala Lake and Ngoring Lake as the research objects,extract lake surface area data based on multispectral remote sensing images,ICESat-2/ATL13 standard data products and LEGOS HYDROWEB database,conduct long-term series dynamic monitoring of water level-surface area of the three lakes and estimate the change of lake water reserves,combined with the meteorological data and partial surface runoff data of the lake area,The main driving factors of lake water body change are analyzed.The main contents include:(1)Based on ICESat-2/ATL13 data product,the water level information of Qinghai Lake,Hala Lake and Ngoring Lake is extracted,and the water level measured by other satellite altimeters is fused and the accuracy is evaluated.The results show that the correlation R2 between ICE Sat-2 and the lake water level values extracted by HYDROWEB is above 0.97,which can meet the water level fusion standard and generate effective water level sequence results.(2)Based on the optical remote sensing satellite images such as Landsat series and Sentinel-2,four kinds of water body indexes are used to enhance the water body of the images,and the surface area and shoreline of three lakes are respectively extracted using the Canny algorithm,and the long-term water level sequence combination results of the lakes are generated.(3)Based on the relationship between water level and surface area,the variation of water reserves in the study area is estimated indirectly by constructing the relationship model between lake water level and water storage capacity.The dynamic change degree of lake water level and surface area,the change trend of lake shoreline and the stage change characteristics of lake water storage capacity in different periods are analyzed.The purpose of monitoring the long-term change of lake water on the plateau is achieved.(4)The trends of water body changes and the main influencing factors of Qinghai Lake,Hala Lake and Ngoring Lake are discussed in the context of natural and anthropogenic factors.The results show that,in general,all three plateau lakes show expansion trends influenced by the warming and humidification of climate on the Tibetan Plateau,but the main driving factors of each lake still have slight differences.The water body changes in Qinghai Lake and Hala Lake are mainly caused by natural factors such as precipitation and temperature,while in the Ngoring Lake basin,the magnitude of water body changes due to human activities such as reservoir construction has exceeded the degree of influence of natural factors on water bodies.