Assessment of the influence of rainfall and landform on landslide initiation using physiographic compartmentalisation

Abstract

Physiographic compartmentalisation emerges as an important instrument in urban planning and risk assessment of mountainous areas, identifying regions where natural erosive processes are more likely to occur based on landform features. The Serra do Mar escarpments are naturally prone to landslide occurrences, due to its landform characteristics and climate, and studies that correlate triggering (rainfall) with controlling (landform) factors are fundamental in the development of urban planning and risk assessment programmes. In this context, this study aims to assess the landslide susceptibility of the Perequê and Mogi River watersheds, in Cubatão (São Paulo), by compartmentalising the study area considering its physiographic features and discussing the role of rainfall and landform on landslide initiation, according to the 1985 and 1994’s landslide events. Physiographic units were separated based on aerial photographs, following geomorphometric criteria such as water bodies and landform elements density, amplitude and slope. Rainfall distribution was based on pluviometric data from five rain gauges that cover the area. Six units were identified, as a result, and those at the northern slope of the Mogi River exhibit higher susceptibility to triggering landslides. This higher susceptibility can be attributed to steep slopes and thin soils, anthropic activities and, especially, rainfall concentration. Physiographic compartmentalisation, therefore, is an important auxiliary tool providing groundwork for more detailed studies in finer scales.