Soil organic carbon as a function of land use and topography on the Loess Plateau of China

Land use and topography are two primary factors that affect the spatial distribution of soil organic carbon (SOC) at the watershed scale, particularly in areas of multiple land uses and complex topography on the Loess Plateau. The General Linear Model (GLM) combined with Duncan's Multiple Range Test (DMRT) were used to conduct variance tests and evaluate the effects of topography (summit, sloping areas and gullies), land use (farmland, orchard, grassland, shrubland and woodland), and their interactions on the vertical and horizontal distributions of SOC at soil depths of 0-10, 10-20, 20-40, 40-60, 60-80 and 80-100 cm. The effects of land use intensified from the summits to the slope areas to the gullies. The SOC concentrations throughout the 100 cm-thick soil profiles in farmland in slope areas and gullies were, respectively, 45.4% (P < 0.0001) and 55.2% (P < 0.0001) greater than those on the summits (3.1 g kg(-1)). In slope areas, the SOC concentrations in shrubland and grassland exceeded those in farmland (4.5 g kg(-1)) by 52.1% (P < 0.0001) and 15.8%, respectively. In addition, in the gullies, the mean SOC concentrations in shrubland and grassland exceeded those in farmland (4.8 g kg(-1)) by 80.6% (P < 0.0001) and 50.7% (P = 0.0039), respectively. The SOC concentrations in areas of semi-shaded slope aspects were greater than those in areas of semi-sunny aspects. The enhanced effect of topographic position on SOC may be attributed to the vegetation and the soil conditions associated with the land uses and to the interactions between these variables at the watershed scale. These results could enhance understanding of the effects of environmental factors and their interaction on the spatial variation of SOC in regions with complex topography. These conclusions could have important implications for ecosystem recovery through rational land use changes in the ecologically fragile areas of the Loess Plateau and for improving the accuracy of prediction of SOC in regions of complex topography and land use on the Loess Plateau. (C) 2015 Elsevier B.V. All rights reserved.