Abstract: The accurate monitoring and analysis of land use changes in mining areas is a crucial part of environmental management in mining regions.Traditional methods have limitations in data acquisition and analysis accuracy,making it diffi cult to comprehensively reflect the temporal and spatial characteristics of land use changes in mining areas.Taking a mining ar ea in Shanxi Province as an example,a method for analyzing land use changes in mining areas based on multi-source remote sensing data fusion and Support Vector Machine (SVM) classification algorithm has been proposed.Firstly,satellite images from GF-2 and GF-3 satellites from 2020 to 2024,Landsat 8 optical remote sensing data,and Sentinel-1 radar data in the min ing area were obtained.The data were preprocessed through steps such as radiation correction,geometric correction,and noise filtering,and a multi-source data fusion technique using weighted averaging was adopted to generate comprehensive data with rich information.Then,the Support Vector Machine (SVM) algorithm was used to classify the fused remote sensing data of dif ferent time periods,accurately identifying and quantifying the changes in land use types.The research results show that multi source data fusion significantly improves the accuracy of land use classification,with a classification accuracy of 94.3%.This method reveals the temporal and spatial change characteristics of different land use types in the mining area,providing a scien tific method for land use planning and environmental monitoring in mining areas,and helping to promote the sustainable devel opment of the mining area′s environment.

Key words: multi-source remote sensing data,support vector machine,land use variation,environmental monitoring in mining area,data fusion