Spatial prediction of soil–water transmissivity based on fuzzy logic in a Brazilian headwater watershed

The Mantiqueira Range region is one of the most important Brazilian headwater regions, from which a great number of springs flow to form one of the most important watercourses for São Paulo, Rio de Janeiro and Minas Gerais States. These natural features make the Mantiqueira Range an important hydrological region for research and development of technologies for management and environmental preservation. Areas in which the predominant land-use is native forest have proved to be strategic for groundwater recharge purposes, and thus, to mitigate drought effects that are being observed recently in Southeastern Brazil. In this context, to identify and map these areas lead to better and more successful decision making regarding soil–water conservation and management. In headwater watersheds, the knowledge about spatial distribution of soil–water transmissivity can be very important to sustainable use and conservation of environmental resources. The aim of this study was to evaluate the application of pattern recognition on landscape classification tools combined with digital soil mapping techniques for mapping soil–water transmissivity in a watershed located at one of the most important Brazilian headwaters, in Southeastern Brazil. The model used solum depth and saturated hydraulic conductivity as soil deterministic elements and topographic indexes. The results showed that the method proposed combining terrain attributes and Geomorphons, which is a mapping tool that allows for the identification of landforms within a landscape through digital elevation model analysis, was very efficient and, therefore, used to predict the soil–water transmissivity for the watershed. Areas with high to moderate soil–water transmissivity values were associated with the steepest slopes, shallow and moderately deep Inceptisols and native forest fragments. Since the relief and solum depth features of Inceptisols did not show optimal conditions for soil water percolation, the results demonstrated that land-use and management factors are crucial for groundwater recharge in the region of the Mantiqueira Range. The results also showed that areas with native forests play a fundamental role in water distribution processes in the soil profile in mountainous catchments in Southeastern Brazil. These results strengthen the importance of soil conservation practices and adequate land-use to sustain this kind of environment in Brazil.