Syifa, M.; Park, S.-J.; Achmad, A.-R.; Lee, C.-W., and Eom, J., 2019. Flood mapping using remote sensing imagery and artificial intelligence techniques: A case study in Brumadinho, Brazil. In: Jung, H.-S.; Lee, S.; Ryu, J.H., and Cui, T. (eds.), Advances in Remote Sensing and Geoscience Information Systems of Coastal Environments. Journal of Coastal Research, Special Issue No. 90, pp. 197-204. Coconut Creek (Florida), ISSN 0749-0208.
Floods are considered to be among the most devastating disasters and can threaten human life and environmental ecosystems. On January 25, 2019, the Brumadinho dam wall collapsed and waste material from the Córrego do Feijão mine flooded the area beneath the dam. At least hundreds of people were killed, animal habitats were swamped, and the flood invaded the river and agricultural fields. Brazilian authorities are examining how this destructive flood might threaten the water quality of the contaminated river. It is important to determine the flood distribution to prevent additional contamination by dangerous material from the flood. In this study, remote sensing was effectively used to map and calculate the dimensions of the flood. Pre- and post-flood images from Landsat-8 and Sentinel-2 were employed to devise a pixel-based classification using two artificial intelligence techniques: artificial neural network (ANN) and support vector machine (SVM). The flood area was successfully determined using the two classifiers. The resulting post-flood damage map should be beneficial for mitigating damage from a future flood event.