A rapid soil Chromium pollution detection method based on hyperspectral remote sensing data

Chromium (Cr) pollution in soil can cause serious harm to both human health and the environment. Cr (Ⅵ) is of particular concern due to its strong carcinogenic, teratogenic, and mutagenic effects on humans. It can also lead to soil hardening and crop yield reduction. Recently, hyperspectral approaches have become a popular way of analyzing heavy metal pollution in soil, and are both convenient and cost-effective, compared with chemical analysis. However, the complex composition of soil makes it challenging to directly analyze the influence of heavy metals on soil spectra and achieve efficient and rapid identification of heavy metal contamination in soil. In this research, a simulation experiment was designed for soil Cr content. Threshold detection-Hilbert-Huang transform (TD-HHT) was implemented to remove the complex environmental noise of soil spectra, to extract the weak heavy metal information. The results showed that the TD-HHT amplitude fluctuates at different Cr concentrations, and heavy metal pollution can be directly detected from the amplitude curve at around 600 mg/kg. To verify this threshold value, the correlation between the soil optical constant and the soil heavy metal concentration was explored through the bidirectional reflectance distribution function, based on the threshold validation-Hapke (TV-Hapke) model. At around 600 mg/kg, Cr can be directly detected using the soil radiative transfer model, which can be utilized to rapidly detect soil heavy metal pollution.