Remote Sensing Mapping And Evolution Analysis Of Coastal Beaches Based On The Time Series Image Analysis Of The Google Earth Engine Platform

Coastal zone is the transit area between land and sea,with numerous natural resources and pleasant environment,where millions of people reside and economic development booms.However,global climate change and frequent coastal exploitation have led to the decay and deterioration of coastal ecosystems,threatening the delivery of important ecosystem services.Defined as important components of the coastal zone,tidal flats possess enormous mineral,biological and marine resources with high economic value and serve as potential land resources that provide broad living space for human beings.Therefore,the study of the exploitation status and spatial distribution of tidal flats as well as the analysis of spatiotemporal change information over time can provide scientific basis for the resource assessment,rational development and conservation.In this paper,remote sensing monitoring and driving force analysis of tidal flats in the past30 years have been carried out focusing on the annual spatial distribution and spatiotemporal change of tidal flats situated in the north of Hangzhou Bay in China.Taking 1987,1997,2007and 2017 as observation years,this study used Landsat TM/OLI surface reflectance images acquired from the target year,plus or minus one year,which cover the entire study area.Google Earth Engine(GEE)cloud computing platform was deployed for image processing.Moreover,we integrated pixel-oriented machine learning algorithm and object-oriented method to work out the spatiotemporal change information of tidal flats in our study area during the last 30years.The result manifests that:(1)GEE platform has distinct advantages in processing the sheer volume of remote sensed data,based on which the utilization of time series images can eliminate the effect of tidal level on intertidal flats.Additionally,integrating pixel-oriented machine learning algorithm and object-oriented method enables remote sensing mapping for tidal flats to obtain the overall accuracy at over 90%.(2)in the past 30 years,the area of tidal flats in the study area has sharply lessened,from10086.78km~2 in 1987 to 5665.58km~2 in 2017.Among them,the most severely reduced area happened in Shandong Province,where the cumulative reduction was 1,742.22 km~2 during the period from 1987 to 2017.On the contrary,the tidal flats in Shanghai have increased by210.09km~2 in 30 years,which is the sole administrative region possessing a larger extent of tidal flats in 2017 than that in 1987.(3)the driving forces for the spatiotemporal change of tidal flats in the study area comprise anthropogenic and natural factors.Anthropogenic factors demonstrate population and economy booming as well as governmental policy for coastal development,while natural factors manifest the continuous development of soil and water conservation and afforestation in China in addition to the reduction of sediment discharge.From what we found,anthropogenic and natural factors interact to drive the spatiotemporal change of tidal flats.