Remote sensing and data-driven assessment of environmental risk patterns in mining and metallurgical areas

Abstract

Mining and metallurgical areas are exposed to complex environmental risks associated with land-cover transformation, waste disposal, dust dispersion, tailings storage, and surface temperature anomalies. Since these impacts are spatially heterogeneous, their assessment requires methods that integrate large-scale observational data with quantitative spatial analysis. This paper proposes a remote sensing and data-driven framework for assessing environmental risk patterns in mining and metallurgical territories. The methodology is based on multispectral satellite data, geographic information system tools, and a set of normalized indicators describing vegetation degradation, anthropogenic land transformation, bare soil exposure, and land surface temperature variation. These indicators are combined into a composite environmental risk index to distinguish low-, moderate-, high-, and very high-risk zones. The proposed approach allows the identification of ecological stress hotspots and provides a reproducible basis for monitoring industrially affected landscapes. The study demonstrates that integrating satellite-derived indices with data-driven assessment methods can improve the spatial accuracy of environmental risk evaluation and support sustainable management decisions in mining and metallurgical regions.