Investigating heavy metal distributions in vineyard soils using satellite data and natural radionuclides

Heavy metal contamination in agricultural soils represents a significant environmental and agronomic challenge, particularly in vineyard systems where copper-based fungicides and phosphate fertilizers are intensively applied. These contaminants pose risks to soil health and food safety and may lead to export restrictions for agricultural products. Conventional soil monitoring approaches, which rely mainly on point-based and laboratory-centered analyses, are often time-consuming and costly.

This study proposes a multi-source and non-destructive framework to assess heavy metal distributions in vineyard soils by integrating natural radionuclide measurements, satellite-derived spectral indices, and soil physico-chemical properties. Soil samples were collected from conventionally managed vineyard parcels located in Western Anatolia. Activity concentrations of natural radionuclides (²³⁸U, ²³²Th, and ⁴⁰K) were measured by gamma spectrometry. Physico-chemical properties including pH, electrical conductivity, and moisture content were also determined in the soil samples. This data was then combined with corresponding Sentinel-2 spectral bands and indices such as NDWI. This data set was then used to train and evaluate an integrated predictive framework to predict ground truth heavy metal concentrations measured by ICP-OES.

This holistic approach offers a promising pathway toward non-destructive and rapid monitoring of heavy metal contamination in agricultural soils. By combining natural radionuclide measurements, remote sensing data, and soil physicochemical properties, this approach has the potential to provide cost-effective soil contamination screening.