Monitoring Of Water Resources Changes In Typical Agricultural And Pastoral Regions And Driver Analyses Based On Remote Sensing Cloud Computing

Water is the premise of the sustainable development of agriculture and animal husbandry.At present,water security is facing global challenges.Affected by climate changes and human activities,the typical agricultural and pastoral regions in China are facing enormous pressure on water resources.How to quickly realize continuous monitoring of water resources changes is of great significance to sustainable water management.Remotely sensed big data and cloud computing provide advanced methods and technologies for monitoring water resources changes at large scale,but the applications in the macro monitoring of water resources dynamics in agricultural and pastoral areas in China still remain very limited.This paper constructs the continuous change process of surface water area(SWA)with 30 m spatial resolution in the North China Plain(NCP)and Mongolian Plateau(MP),by using all the available Landsat imagery,improved water body mapping algorithm,and Google Earth Engine(GEE)cloud computing platform.Then,we study the spatio-temporal changes of terrestrial water storage(TWS)in the study areas by using Gravity Recovery and Climate Experiment(GRACE)data.Finally,based on the analyses of changes of surface water bodies and TWS,we analyze the driving mechanisms of water resources changes from the aspects of climate change and human activities,quantify the relative contribution of each factor to water resources changes.The main findings are as follows:(1)GEE cloud computing platform provides significant technical support for quick ly analyzing the changes of SWA and TWS.We form the automatic technical process of remote sensing-based continuous monitoring of SWA by using GEE,all the available Landsat data and water index-and threshold-based water body mapping algorithm,and then realize the continuous monitoring of SWA changes in the NCP since 1987 and MP since 1991 at inter-annual scale.In addition,we assess the dynamics of TWS in the NCP and MP since 2002 by using GRACE data.(2)For the NCP,we find the decreasing TWS(8.9 mm/yr)in spite of expanded SWA(109.1km~2/yr),and identify a hotspot with the most rapid loss of TWS(>12 mm/yr)across the Shanxi-Hebei border.Though agricultural irrigation has been recognized as the main driver of TWS loss in the NCP in the former studies,however,we find the decreasing agricultural water use due to the measures of water-saving irrigation and so on,and furtherly we find that the implementation of ecological restoration programmes(e.g.,“Grain to Green”project)has accelerated the water crisis in the NCP to a certain extent.In the western NCP(mainly in Shanxi Province),the increasing forest coverage induced by large-scale implementation of ecological restoration projects has made evapotranspiration(ET)significantly increased at a rate of 9.2 mm/yr,accounting for more than60%of the loss of TWS per year.In addition,the construction of reservoirs,artificial lakes and coastal aquaculture ponds caused the expansion of SWA in the central and eastern NCP,making the water evaporation significantly increased at a rate of>3 mm/yr and bringing new threats to local water security.(3)For the MP,both SWA and TWS generally showed decreasing trends during 1991-2017.The region with the rapidest TWS decline occurs in the central and southern part of Inner Mongolia,which is closely related to intense agricultural irrigation.The areas and numbers of lakes continuously decreased before 2009,followed by a recovery process.Specially,Inner Mongolia of China experienced more drastic lake variations than Mongolia.Limited administrative regions dominated the variations of lakes on the MP.Specially,Hulunbuir City and Xilin Gol League dominated the lake changes in Inner Mongolia,contributing to 60.0%and 57.6%of the variations of areas and numbers of lakes,respectively;while the lake changes in Mongolia were mainly controlled by Ubsa Aymag,Dornod Aymag and Hovsgol Aymag.The varied drivers of lake variations in different regions showed the complexity of factors impacting lakes on the MP.Specially,both natural(decreased precipitation)and anthropogenic(grazing and irrigation)factors induced the lake shrinkage before 2009,precipitation played increasingly important role for the recovery of lakes on the MP after 2009.Generally,this study develops the automatic process of remote sensing-based continuous monitoring of SWA by using remote sensing cloud computing platform,all the available Landsat data and water body mapping algorithm.Then,we assess the dynamics of TWS in the study area by using GRACE data.This paper realizes the comprehensive monitoring of water resources changes in the NCP and MP,which provides scientific technical methods and important data for understanding the continuous process of spatial changes of water resources and sustainable water management in the study area.