Spatially Explicit Machine Learning for Mapping Soil Heavy Metals in Abandoned Rare Earth Mines Using Multisource Remote Sensing Data

This dataset contains two geospatial layers used in the study “Spatially Explicit Machine Learning for Mapping Soil Heavy Metals in Abandoned Rare Earth Mines Using Multisource Remote Sensing Data.”The database is provided in ESRI Shapefile format and includes: 1. Study area boundary (gannan_ree_polygon_of_study_area.zip)A polygon dataset defining the spatial extent of the study area.It was used for raster clipping, masking, and generating prediction maps. 2. Sampling points (gannan_ree_sample_points.zip)A point dataset representing soil sampling locations within the study region.Attribute fields include sample ID and associated soil heavy-metal measurements used for model training and validation. Both layers are spatially referenced and ready for direct use in GIS and geospatial machine learning workflows.