Land capability of multiple-landform watersheds with environmental land use conflicts

Abstract

Abstract This study introduces an improved method to assess land capability in the rural environment. Land capability, also defined as natural use, is a powerfull asset in land use planning and development of land use policies, because actual land uses that deviate from natural counterparts trigger extensive and severe environmental impacts, namely amplified soil erosion and fertility decline. The proposed method builds up on a previous model based on the ruggedness number (RN), which allocates natural uses as function of increasing terrain slope and drainage density. According to the RN model, agriculture should occupy the areas of low slope and sparse drainage density, while the forest is reserved for the areas of steep slope and dense drainage. Other activities such as pasture or pasture mixed with forest can be settled on the areas of intermediate slope and drainage density. The modeling improvements introduced with this study are in a first stage related to the combination of RN-based capability classes with dissection classes derived from slope curvature estimates and landform classification (convex, concave and straight srufaces). In a second stage, the modeling improvements are related to validation of RN and landform classification schmes in the field through comparison of soil magnectic susceptibility data across the RN-dissection classes. The improved model was tested in a large multiple-landform watershed: the Tiete River basin located in Sao Paulo State (Brazil). The results proved the need to refine the RN land capability classes using the dissection classes, because dissection patterns are different across lithologic types and weather trends. In the Tiete, landscapes are generally more dissected when shaped in basaltic bedrock than in sandsone cover, and increase with increasing annual rainfall. The implications for land use planning are evident because high dissected lanscapes require more exigent conservation measures than low dissected landscapes, while suitable crop types may also vary between the two environments. The field validation of capability-dissection classes using the magnetic susceptibility proved efficient in a large portion of cases, but failed in several other sub-basins. A main reason for the failure was inadequate land use, interpreted as environmental land use conflict in this study. Overall, the current study can be viewed as major advance in the development of a land capability classification, because the integration of relief (RN), landform (dissection) and soil properties (magnetic susceptibility) has not been tempted before. The gains for land use policy are apparent because land capability maps are fundamental instruments in the management of rural areas, namely to help the mitigation of a major global concern nowadays, which is land degradation.