干旱区生态环境知识资源中心

The semi-arid loess hilly-gully region in China has an extremely vulnerable ecological environment. Inappropriate land use - crop farming, overgrazing, and plantation forestry - has worsened soil erosion, intensified water shortage, and hence impeded the ecological conservation and agricultural development of the entire region in the past. Optimizing land use and vegetation cover and spatial pattern is conducive to achieving both ecological and economic goals in terms of controlling soil erosion, using water resources rationally and raising agricultural productivity. Changchuan Watershed, a typical small catchment in the semi-arid loess hilly region, was selected as the case study area to analyze the impacts of land use and land cover structure and associated spatial pattern on soil erosion and water consumption in the Watershed, through field investigation and model simulation. Land use structure was optimized by multi-objective programming, using remote sensing (RS) and geographical information system (GIS) techniques, analytic hierarchical programming (AHP) and expert consultancy. The digitized optimum spatial pattern embodying rationally-proportioned land use structure was obtained through GIS-aided redistribution of land use types. The optimized land use structure reapportioned woodlands, shrublands, grasslands, and croplands at 3.7%, 38.6%, 49.4%, and 6.3% of the land area respectively, compared to the current land use structure of 2.4%, 38.6%, 24.0%. and 12.6%, respectively, in the Changchuan Watershed. In the optimized land use spatial pattern, croplands are mainly located in the riverside plain and check-dammed valleys and grasslands are widely distributed in the lower reaches of the basin, while shrublands are appropriately established in the middle and upper reaches of the river. A comparative analysis shows that the optimized land use structure, with well-designed spatial pattern is able to reduce soil erosion, enhance the utilization of water resources and raise agricultural productivity. (C) 2010 Elsevier B.V. All rights reserved.