Imagem de radar para mapeamento de inundação na área urbana de Alenquer, Pará

Abstract

Flooding is the phenomenon that most affects people around the world and has been increasing considerably in Brazil, especially in urban areas, causing immediate damage to the environment and directly and indirectly affecting populations, especially the poorest who live in irregular environments such as the floodplains. The municipality of Alenquer, located in the mesoregion of Baixo Amazonas Paraense, presents historical records associated with the phenomenon of flooding, mainly due to the widespread urban development in the floodplain areas, places where seasonal floods occur due to natural river elevation, intensified by extreme events of precipitation. In 2017, due to the extreme precipitation event associated with the flood period of the Amazon River, the municipality of Alenquer declared an emergency due to the floods that occurred, causing damage to the population. In this sense, the present study aims to map floodable extensions in the urban area of the municipality of Alenquer using images from active sensors, satellite Sentinel-1, considering the low effectiveness of optical sensors due to the high concentration of clouds in the Amazon regions. Geoprocessing and remote sensing techniques were used to obtain the extent of flooding, area of extension, and quantification of urban lots and affected roads from the collaborative mapping carried out by OpenStreetMaps (OSM). The analysis was carried out using the Google Earth Engine cloud computing platform, evaluating the results obtained between the two types of polarization (VV and VH) available for the region associated with geoprocessing techniques through the free software QGIS version 3.28 to produce data. floodplain maps. The results using a threshold of 1.40 and 1.25 for the VV and VH polarizations, respectively, produced floodplain areas of 2 km² each. The extension coming from the VV bias detected 21 habitational infrastructure and 7 roads affected while the VH bias detected 12 habitational infrastructure and 7 roads. The results obtained in this work aim to contribute to studies about floods. The methodology was efficiently; however, it has limitations regarding the crossing of data at different scales. Thus, caution is needed for methodological reproduction for other Amazonian and Brazilian cities, observing limitations and provide support from field data.