Modelling of Acid Mine Drainage in Open Pit Lakes Using Sentinel-2 Time-Series: A Case Study from Lusatia, Germany

Strong acid mine drainage (AMD) processes in the flooded, formerly open pits Lusatia area present an enormous environmental challenge for rehabilitation of post-mining landscape. Extensive and costly monitoring is required optimal AMD management remediation planning control. Because large size dimension problem, regular sampling can only provide limited point data, which needs to be extrapolated entire area. Consequently, search effective approaches extrapolating data all water bodies essential success understanding dependencies between factors such as land use, weather conditions, geology, hydrogeology. The main aim this study was investigate suitability Sentinel-2 multispectral imagery artificial neural networks (ANNs) quantitative mapping constituents, dissolved iron, pH value, sulfate bodies, approximately 7220 km2 (the pit lakes about 185 km2). Correlations different chemical parameters were also investigated. An extensive dataset used train identification remote sensing quality ground measurements. Respective relationships have been identified, especially iron pH. These trained ANNs produce maps with high spatial (10 × 10 m) temporal (any cloud-free period) resolution, show wide variability parts mining region. Concrete sources identified using single lakes, sanitation measures liming visualized. approach opens many doors optimization both program technology.