Suscetibilidade a processos erosivos lineares apoiada em análise estatística multivariada

• Ano de defesa: 2022
• Autor(a) principal: Neves, Monique de Paula
• Orientador(a): Lollo, José Augusto de
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus São Carlos
• Programa de Pós-Graduação: Programa de Pós-Graduação em Engenharia Urbana - PPGEU
• Palavras-chave em Português: Erosão linear; Regressão logística; Carta de suscetibilidade; São Pedro
• Palavras-chave em Inglês: Gully erosion; Logistic regression; Erosion susceptibility map
• Área do conhecimento CNPq: ENGENHARIAS::ENGENHARIA CIVIL::GEOTECNICA; ENGENHARIAS
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/16519

The present study aims to evaluated erosion susceptibility using multivariate analysis. The study area is Ribeirão do Meio watershed situated in São Pedro (SP). Gully erosion is an active and complex process in this watershed. In this study, gully erosion were investigated over 50 years, using aerial photographs for the years: 1960, 1972, 1988 and 2010. The procedures of the current study were developed in four main stepes: preparation of the erosion inventory; spatial analysis through GIS to obtain topographical and hydrological parameters; determination of topographic threshold (SxA) to obtain the gully initiations conditions; logistic regression (LR) analysis to evaluate the erosion susceptibility. The results showed that the logistic regression had overall fit and accuracy acceptable. Thus, these models were used to prepare erosion susceptibility maps. According to the results, the conditioning factors mainly were soil and sediments, slope, flow accumulation, rainfall erosivity and land use. The results also revealed that the map obtained is acceptable and indicated satisfactorily the erosion-prone areas. The susceptibility maps can represent an important tool for regional managers or planners. Finally, the procedures that were used, in terms of developing the logistic regression model was efficient and can be used other assessment of gully erosion.