Dynamic Monitoring And Spatio-Temporal Analysis Of Key Elements Of Resources And Environment In Mekong Basin Based On Google Earth Engine Platform

The Mekong River flows through six countries:China,Myanmar,Laos,Cambodia,Thailand,and Vietnam.It is an important trans-boundary water system in Asia.With abundant natural resource that not only provides important basic condition for the socio-economic development but also becomes an important link for people's social and economic activities and cultural exchanges in the Mekong Basin.In recent decades,due to the rapid economic development of the countries along the Mekong Basin and the differences in the ecological environmental protection policies of various countries,the ecological environment of the watershed has suffered various degrees of damage.Therefore,the dynamic monitoring of resources and environmental elements in the Mekong Basin based on remote sensing technology has become one of the most urgent tasks currently facing governments,Mekong basin organizations,non-governmental organizations,and local research institutions.In recent years,the emergence of the Google Earth Engine(GEE)remote sensing big data cloud platform has provided technical support for global,watershed and national scale resource and environment dynamic monitoring and analysis of spatio-temporal changes,and has solved the computational complexity and High computational costs issues of traditional remote sensing image processing.In this paper,we will realize the automatic extraction of key elements such as surface water,forests,urbanization,and population data in the Mekong Basin based on the GEE platform and multi-source remote sensing data,as well as spatial-temporal dynamic analysis.At same time,combined with meteorological data,thematic data of land cover/land use and socio-economic statistics to analyze the driving factors of key elements of resources and environment.This work will provide resources and environmental foundations and decision-making support for governments and research institutions in the basin.The main conclusions of the thesis include the following aspects:(1)Remote-sensing monitoring for spatio-temporal dynamic analysis of Surface Water Area in Mekong Basin.Using the hierarchical classification method of pixels and objects,complete the automatic extraction of surface water in the Mekong Basin from 1990 to 2019 based on Landsat image data and GEE platform.The research results show that the surface water area of the Mekong Basin has changed greatly in the past three decades,with severe interannual floods and droughts.The surface water area in the basin was the largest in 2011,totaling 16,596.58 km~2.In 1992,the surface water area was the smallest,totaling 8363.96 km~2.In addition,the influence of meteorological factors on the surface water area of the Mekong Basin has obvious regional differences.Temperature is the main driving factor in the upper Mekong Basin,precipitation is the main driving factor in the Middle Mekong basin,and evapotranspiration is the main driving factor in the Lower Mekong basin.(2)Remote sensing monitoring and spatial-temporal dynamic analysis of forest area in Mekong Basin.The hierarchical classification method was used to complete the automatic extraction and mapping of the Mekong forest cover data from 1990 to 2019Based on the GEE platform and Landsat remote sensing data.An analysis of the forest cover data of the Mekong River Basin in the past 30 years shows that the forest area of the Mekong Basin has declined first,then increased,and then decreased,it has been consumed at an average annual rate of 1832.37km~2/yr,and the decline in forest area in Cambodia has been most significant.In addition,driving factors such as the socioeconomic development of the basin,population growth,the exploitation of mineral resources,and the process of urbanization have become important factors influencing the change and reduction of forest cover in the region.(3)Quantitative inversion and spatio-temporal dynamic analysis of population Data of Mekong coastal countries based on DMSP/OLS night-time light imagery.A linear regression model was used to quantitatively invert the spatial distribution of the population data along the Mekong River and its spatio-temporal changes based on DMSP/OLS night-time light imagery and GEE platform.The research results show that the spatial distribution of the urban population in the study area has obvious circle characteristics,that is,the population density in the city center is the largest,and the further away from the city center,the smaller the population density.(4)Extraction of impervious surfaces area and urban spatial Expansion of Towns along the Mekong River based on multi-source remote sensing data.A hierarchical classification method that combines objects and pixels is used to complete the extraction of impervious surfaces area of countries along the Mekong River from 1990to 2019 based on GEE platform and DMSP/OLS,NPP/VIIRS night-time light imagery,Landsat and 30m SRTM DEM reomote sensing.By analyzing the spatio-temporal changes characteristic in the impervious surfaces area of cities and towns along the Mekong River in the past three decades and the expansion of urban space,we can see that in the past three decades,the impervious surfaces area of cities and towns along the Mekong River have moved outward from the city center in a"domain"expansion,and the area covered by the impermeable grounds of towns and cities in the region is increasing exponentially.Among them,the area of impervious surfaces in 1990accounted for 0.1291%of the total area,and in 2019 accounted for 0.6738%of the total area.In other words,the area of impervious surfaces has expanded by 4.22 times in the past three decades.