IRON ORE MINE WASTE FLOODWATER EXTENT MAPPING UTILIZING REMOTE SENSING DERIVED INDICES

摘要

The collapse of two iron ore mine dams in south-eastern Brazil on 5 November 2015 is one of the recent environmental disasters. The clean-up and environment recovery would take several years and cost billions of dollars. Mapping the extent of the affected areas and monitoring the water quality deterioration is a challenge. In this study the new optical satellite SENTINEL 2 imagery along LANDSAT 8 were utilized to test the applicability of the Land Surface Water Index (LSWI) and Modified Normalized Difference Water Index (MNDWI) in the mapping process. The systematic coverage of the study area from the aforementioned satellites before and after the incident were studied and compared. The dynamics of LSWI, MNDWI were utilized in the delineation of the affected areas. Both quantitative and qualitative measures to assess the mine waste floodwater extent were developed. The high reflectance in both the 650-nm and 750-nm wavebands as an indication of Iron-Oxide precipitates occurrence was also tested. The study revealed that the mine tailings extended about 500 Km downstream. The affected areas and extent revealed from the study results were validated against the official figures from the Brazilian government. There was a good agreement between the study results and the published figures. The temporal variation in the Iron-Oxide precipitates occurrence was successfully mapped. In addition, visual interpretation go well with the study results. The findings of this study indicates that the proposed algorithm can be used in the timely mapping of the iron ore mine waste floodwater disaster.