The use of geomorphological descriptors and landsat-8 spectral indices data for flood areas evaluation: A case study of Lato river basin

Abstract

In the last few years, the scientific community has dedicated a strong effort for the rapid identification and mapping of flood risk. Last generation models have often taken advantage (even without of in-situ measurements) of the distributed information provided from remotely sensed data. In this work is proposed a multidisciplinary approach to reproduce maps of flooded areas. The method compared spectral descriptors to estimate the areas at risk of flooding in the Lato river basin (Puglia region - Southern Italy) using the ground effects caused by flood events. The inundated areas, obtained with a 2D hydraulic model, were used as reference for Landsat-8 spectral indices. The selection of the most appropriate spectral index was achieved using the binary classifiers test. Lastly, the adopted procedure provided also the calibration of different geomorphological descriptors for a rapid identification of areas at risk of flooding by using Digital Elevation Models.