Estimating heavy metal concentrations in Technosols with reflectance spectroscopy

Reflectance spectroscopy in the visible-infrared and shortwave infrared (450–2500 nm) wavelength region is a rapid, cost-effective non-destructive method that can be used to monitor heavy metal (PTE, potential toxic elements) contaminated areas. Due PTE pollution has accumulated course of wastewater treatment, existence Technosols presents an environmental problem, source for uptake by vegetation, or even release PTEs into groundwater. In this study, multivariate procedures using Partial Least Squares Regression (PLSR) Random Forest (RFR) are applied quantify relationships between soil concentration (Cr, Cu, Ni, Zn) reflectance data highly from former sewage farm near Berlin, Germany. Laboratory measurements 110 samples four different preparation steps were acquired with HySpex hyperspectral cameras. The impact steps, namely “oven-dried”, “sieved”, “ground”, “LOI”, was evaluated its enhance performance reduce time-consuming sample preparation. Furthermore, spectral pre-processing methods regarding improvements modelling their ability minimise noise multiple scattering effects. Considering optimal coefficient determination (R2), PLSR shows improving accuracy increasing such as ground LOI all metals interest (R2_Cr: 0.52–0.78; R2_Cu: 0.36–0.73; R2_Ni: 0.19–0.42 R2_Zn: 0.41–0.74). RFR weaker estimation metals, when higher levels 0.36–0.62; 0.17–0.72; 0.20–0.35 0.26–0.67). results show application prediction still challenge but provides more robust estimations than user-friendly method. Additionally, study heterogeneous improved increased laboratory further steps.