Research on site category classification method based on multi-source remote sensing and geographical information: A case study of Weining County, Guizhou Province

Site classification is of great significance for the site selection of major engineering projects and prevention of seismic geological hazards. This study considered Weining County as the research area. By collecting geological maps, digital elevation model data, and Landsat8-9ETM image data, we analyzed the main factors influencing the classification of site categories and determined three evaluation factors: rock category, slope, and normalized vegetation index. We used deterministic and fuzzy logic methods based on weighted superposition to build a model. We adjusted the model using measured borehole data, classified site categories, used research area survey data for verification, and compiled a spatial distribution map of the site categories in the study area. Results show that site categories are mainly related to lithology, slope, and NDVI. The softness and hardness of rocks play major roles in the classification of site categories. The classification method can be adjusted using the measured survey point data, and the deterministic method combining the slope and NDVI can more accurately classify the site categories. This research result is significant for the classification of large-scale site categories.