Remote Sensing Monitoring Of Salinization In The Yellow River Basin Based On GEE

Soil salinization is one of the main types of land degradation in the world,and it is also a global resource and ecological problem.In 2015,the United Nations issued the 2030 Agenda for Sustainable Development.The goal of SDG15.3 is “By 2030,combat desertification,restore degraded land and soil,including land affected by desertification,drought and floods,and strive to achieve a land degradation-neutral world”.As the granary of China,soil salinization in the Yellow River Basin has attracted the attention of the country,so it is necessary to monitor the soil salinization by remote sensing.In this paper,based on feature space method,remote sensing monitoring of soil salinization in the Yellow River Basin was realized by using Google Earth Engine(GEE)Google cloud platform,and the inversion results and causes of salinization were analyzed.(1)By exploring different feature space types,it is found that the precision of threedimension feature space is more suitable for large-scale inversion.According to the principle of feature space,the feature space model is divided into two-dimensional linear feature space,two-dimensional nonlinear feature space and three-dimensional feature space.Based on three different feature space types,the feature space models were constructed to invert the salinization of the Yellow River Delta.The results showed that the three-dimensional feature space models were more stable and more accurate,which is suitable for large-scale inversion of the Yellow River Basin.The optimal three-dimensional feature space was used to complete the inversion of salinization distribution in the Yellow River Delta in 2015 and2020,and the spatiotemporal characteristics were obtained.(2)To explore the automatic inversion method of salinization based on GEE remote sensing cloud platform,and realize the rapid inversion of large-scale salinization in the Yellow River Basin.The inversion process of feature space salinization is divided into five steps: image acquisition and preprocessing,image cloud removal and clipping,surface feature parameter acquisition,feature space construction and inversion,and data export.The inversion process of feature space was realized on GEE platform,and the automatic inversion procedure of salinization was completed.The Yellow River Basin is divided into six sub-areas considering the factors of "climate-meteorology-topography".According to the characteristics of each sub-area,the corresponding optimal feature space model was constructed,and the inversion was completed on the GEE platform.The obtained inversion map of salinization in the Yellow River Basin was downloaded to the local area.According to the features of each subarea,the corresponding optimal feature space model is constructed,and the inversion is completed on the GEE platform.The inversion maps of salinization in the Yellow River Basin were downloaded and classified locally,and the distribution maps of salinization in the Yellow River Basin in 2015 and 2020 were obtained,with the accuracy of73% and 72.14%,respectively.(3)The spatial-temporal distribution characteristics of salinization in the Yellow River Basin in 2015 and 2020 were analyzed,and the attribution and policy recommendations were made based on regional characteristics.The results showed that:(1)The proportion of salinization in 2015 and 2020 was 9.14% and 8.13% respectively.The overall salinization area of the Yellow River basin showed an obvious decreasing trend,but the salinization degree increased slightly in some areas such as Guoluo Tibetan Autonomous Prefecture and Hetao area of Inner Mongolia.(2)Severe salinization was mainly distributed in the Yellow River delta and Hetao region of Inner Mongolia;Moderate salinization is mainly distributed in Guoluo Tibetan Autonomous Region,Hetao region of Inner Mongolia,part of Ningxia,Yellow River Delta and other areas in the upper reaches of the Yellow River.Mild salinization is mainly distributed in parts of Henan Province and Gansu Province,and Luliang City and Taiyuan City of Shanxi Province.(3)Combined with the distribution of salinization,local climate,terrain and other factors,according to the attribution of man-made and natural factors caused by salinization,put forward the Yellow River basin salinization policy suggestions.