Geological environment carrying capacity assessment using remote sensing and integrated modeling: A case study of Yi’an district, Tongling, China

Abstract

The central tenet of Regional Sustainable Development (RSD) is fundamentally reliant on the Geological Environment Carrying Capacity (GECC). An in-depth comprehension of a region’s current geological environmental status is crucial for formulating pragmatic and efficacious management strategies. This is essential for laying a robust groundwork for scientifically informed regional development planning. In this context, our research zeroes in on the Yi’an District of Tongling, utilizing a suite of nine evaluative metrics that span geological, ecological, and socio-economic dimensions. We adopt a methodological approach involving the Information Value (I), Random Forest (RF), and Extreme Gradient Boosting (XGBoost or XGB) algorithms to construct a comprehensive GECC assessment framework to improve the precision of GECC predictions. This framework is further augmented through the application of the Receiver Operating Characteristic (ROC) curve for nuanced model comparison and analytical purposes. Our findings indicate that factors such as slope, topographic relief, and the Normalized Difference Vegetation Index (NDVI) have a marked impact on GECC. The I-RF model demonstrates superior predictive precision over the I-XGB model, thus providing a more accurate evaluation of GECC. The overall GECC in the Yi’an District is assessed as favorable. However, areas with suboptimal GECC, predominantly located in the southern, southeastern, and western hilly terrains, are largely attributed to mining activities and urban developmental pressures. These insights are instrumental for local administrative bodies to enact proactive and efficacious mitigation strategies. This study significantly contributes to the scientific underpinning of geological environmental safety and bolsters the sustainable economic advancement of the region.