Hyperspectral remote sensing for soil heavy metal inversion: insights and applications

Abstract

Researchers worldwide have increasingly recognised hyperspectral remote sensing (HRS) for its extensive coverage and high efficiency in monitoring the spatiotemporal distribution of soil heavy metal (SHM). However, accurately estimating SHM concentrations using HRS remains a significant challenge because of data complexity, spatial heterogeneity, and scale variability. In this review, we critically examine recent advancements in HRS for SHM monitoring, beginning with an overview of the mechanisms underlying the hyperspectral inversion (HI) of SHM content. We then discuss the application of HI technologies in SHM research, summarise key findings, and identify persistent challenges, including those related to inversion accuracy and large-scale mapping. Finally, implementation strategies are outlined to provide valuable insights and guidance for advancing soil pollution monitoring and regulation applications. This review aims to support further developments in the field and foster more effective monitoring and management of soil pollution.

Keywords

• Hyperspectral remote sensing
• Soil heavy metal
• Machine learning
• Air-space-ground integration