Integration between SHALSTAB model and Multi Criteria Evaluation (MCE) for mapping landslide susceptibility areas in São Bartolomeu Stream basin, municipality of Viçosa (MG)
1Miranda Nunes, D.; 2Dias Coelho, C.; 3Bezerra de Souza, A.; 4Lúcia Calijuri, M.; 5de Paula dos Santos, A.; 6das Graças Medeiros, N.
The accelerated process of urbanization coupled with lack of public policies directed toward efficient planning, has led to the occupation of geotechnically unstable areas, contributing to the occurrence of natural disasters, such as landslides (mass movement), which directly or indirectly affect the population located around the risk areas. The landslides are part of the processes of earth’s surface and can be considered as one of the most widespread natural hazards types in mountainous areas, causing loss of lives and damage to properties. The identification and mapping of areas of susceptibility to mass movements are important procedures, for example, to watershed management, preparation of risk maps, planning structural measures for infrastructure protection, etc. With this perspective, the present study aimed to use the potential of Geographic Information Systems (GIS) for mapping susceptible areas to shallow translational landslide through integration between the deterministic model Shallow Slope Stability Model (SHALSTAB) and the Multi Criteria Evaluation (MCE) in São Bartolomeu River Basin, located in Viçosa Municipality, Minas Gerais State, Brazil. The methodology adopted in this study combines the results obtained by the SHALSTAB model - which returned susceptible areas to landslides considering slope, hydrological model, soil type in the basin (Latosols, Argisols, Cambisols) and their respective geotechnical parameters (effective cohesion, density soil saturation, depth, hydraulic conductivity and internal friction angle) - as one of conditioning factors to landslide, weighted with the factors of soil cover, distance from roads and downhill profile in the MCE. Using the technique of comparing by pairs, it was generated a set of relative weights for each factor, subsequently aggregated in the procedures of Weighted Linear Combination (WLC) and Ordered Weighted Average (OWA). Three OWA scenarios were performed with low, medium and high risk, all with high compensation. For each scenario, the continuous values of suitability landslides (0 to 255) were reclassified into five susceptibility classes to mass movement (very low, low, moderate, high and very high) through the natural breaks method. Taking the landslides occurrence points reported to the Civil Defense Department of Viçosa (recorded by GPS receptor in field survey), it was found that the best scenario was that of low risk and high compensation (risk: 0.63 and compensation: 80%), because most of this points were identified in the areas of high and very high instability (obtained by integration between SHALSTAB model and MCE), unlike the application of the SHALSTAB model individually. Through the proposed methodology, was obtained that 1.55% of São Bartolomeu River Basin area lies in areas of high and very high instability, and therefore the locations that should receive more attention from management agencies. The results also showed that there was a predominance of medium, high and very high susceptibility classes in areas of exposed soil, pasture and built up areas, as well as Latosols Red-Yellow soil type and concave downhill profile. The regions of medium susceptibility to mass movement corresponding to 15.45% of the basin area, are concentrated in urbanized locations, showing that soil sealing, embankments road poorly designed, occupation of inappropriate sites, among others, are directly related to potential landslides. With results this study, was possible showed that the integration between SHALSTAB model and MCE enables mapping landslides susceptibility areas in a more coherent way with the reality instead using only SHALSTAB model, because the proposed methodology allowed consider natural and anthropogenic factors, reducing subjectivity of the analyst and producing more consistent results.