Modeling the distribution of heavy metals in lands irrigated by wastewater using satellite images of Sentinel-2

Abstract

The pollution of heavy metals is considered as one of the main problems of using wastewater for irrigation purposes. Frequent experimental measurements, time, and cost are essential for evaluating the pollution of heavy metals in a large area. Thus, using satellite images and establishing a relationship between the images and concentration of heavy metals can be regarded as a solution for estimating the polluted area. In this study, the image of the Sentinel-2 satellite was used to evaluate the heavy metals pollution of the area irrigated with wastewater in south of Tehran. For this aim, 30 soil-surface samples irrigated with raw wastewater were collected in the area. After preparing the samples, the concentration of Pb, Cu, and Ni was determined using atomic absorption spectroscopy. Then, the relationship between the heavy metal concentration, reflection on the bands or rating of the bands at the corresponding sampling points was determined by applying the stepwise regression method. In addition, the developed models were applied to the satellite image for zoning the heavy metal concentrations in the studied area. Finally, the accuracy of the developed models was examined by Root-Mean-Square Error (RMSE) and Pearson correlation coefficients. The results indicated that the amounts of RMSE for the equations of Pb, Cu, and Ni were 1.90, 2.54, and 1.59 respectively, while the amounts of R were 0.81, 0.75, and 0.73 for these metals representing a promising match between the predicted and measured results of the models.

Keywords

• Tehran
• Sentinel-2
• Heavy metals
• Statistical analysis
• Spatial distribution