Spectral indices for monitoring soil heavy metal contamination: Direct soil and vegetation proxy approaches, current trends and future prospects

Soil heavy metal (HM) contamination is a critical environmental challenge with significant implications for ecosystem health, agricultural productivity and food security. Traditional laboratory-based monitoring methods are precise but costly and inefficient for large-scale, dynamic assessment. Consequently, remote sensing using spectral indices has emerged as a powerful, non-destructive alternative for detecting and mapping contamination. This review synthesizes the current science and applications of spectral indices for monitoring HM contamination in soils and vegetation. We explain the fundamental mechanisms underlying the spectral detection of metals, which primarily occur indirectly through metal-induced changes in soil physicochemical properties and plant physiological stress. The article categorizes and evaluates a range of vegetation-based, soil-based and emerging metal-specific indices, benchmarking their accuracy, transferability and sensitivity under diverse environmental conditions. A critical analysis identifies persistent challenges, including signal interference from soil moisture and organic matter, spectral overlap with other plant stressors and the lack of standardized, universally applicable indices. Finally, we highlight key future directions, emphasizing the integration of machine learning for intelligent index design,the fusion of optical data with thermal and radar sensors and the significant potential of next-generation spaceborne hyperspectral missions. It is concluded that the convergence of these advanced technologies could enable operational, global-scale monitoring networks, ultimately yielding valuable tools for evidence-based environmental management and remediation strategies.