Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Titre : Spatiotemporal Dynamic Monitoring of Soil Salinization in Alar Reclamation Area Based on Satellite Remote Sensing

Auteur : 代云豪

Grade : Master 2022

Université : Tarim University

Résumé partiel
Southern Xinjiang（hereinafter referred to as southern Xinjiang）is an arid region,which is one of the main distribution areas of saline soil in China.The problem of soil salinization is serious.Using remote sensing technology to timely and accurately grasp the information of soil salinization provides an important reference for the effective prevention and control of regional saline soil,and also has practical scientific significance for regional development.Taking Alar Reclamation Area in southern Xinjiang as the research area,based on Landsat 7 ETM+data in 2011,Landsat 8 oli data in 2017 and 2021 and measured soil conductivity data,this paper uses normalized difference vegetation index（NDVI）,modified soil adjusted vegetation index（MSAVI）,surface albedo The salinity index（SI）establishes the two-dimensional feature space of si-ndvi,si-msavi,Si albedo and albedo MSAVI respectively,and constructs the corresponding remote sensing salinity detection index（SDI）models SDI1,SDI2,SDI3 and SDI4,extracts the soil salinization information of Alar Reclamation Area,verifies the accuracy of the model,and selects the optimal model to dynamically monitor the soil salinization of Alar Reclamation Area.The main research results are as follows:1)Through the natural discontinuity method,the soil salinization grade of remote sensing salt monitoring index model is divided.The correlation between SDI1 model and measured soil conductivity is0.7579,and the overall classification accuracy is 83.45% ;The correlation between SDI2 model and measured soil conductivity is 0.8227,and the overall classification accuracy is 69.78% ;The correlation between SDI3 model and measured soil conductivity is 0.5034,and the overall classification accuracy is53.23% ;The correlation between SDI4 model and measured soil conductivity is 0.7400,and the overall classification accuracy is 71.94%.SDI1 model is more suitable for retrieving the degree of soil salinization in Alar Reclamation Area.SDI2 model and sdi4 model have certain reference significance for soil salinization monitoring in Alar Reclamation Area.2)The distribution of soil salinization in Alar Reclamation area is inversed by SDI1.The Alar Reclamation area is mainly composed of non saline soil and mild saline soil,with a wide distribution range,and its surface feature type is mainly farmland ;Moderate saline soil and severe saline soil are mainly distributed in the center of alar,the south of the reclamation area and the northeast of the reclamation area,and their surface features are mainly Tarim River Bed and grassland ;Saline soil is mainly distributed in the northeast and southeast of the reclamation area,and its surface features are mainly desert and desertification land.3)From 2011 to 2021,the change trend of soil salinization area in Alar Reclamation area showed that non saline soil,mild saline soil and saline soil increased,while moderate saline soil and severe saline soil decreased.

Mots clés : Alar Reclamation Area ;Soil salinization ;Spectral index ;Remote sensing salt monitoring mode ;

Présentation (CNKI)