Global analysis of cover management and support practice factors that control soil erosion and conservation

Cover management and support practices largely control the magnitude and variability of soil erosion. Although soil erosion models account for their importance (particularly by C- and P-factors in the Revised Universal Soil Loss Equation), obtaining spatially explicit quantitative field data on these factors remains challenging. Hence, also our insight into the effects of soil conservation measures at larger spatial scales remains limited. We analyzed the variation in C- and P-factors caused by human activities and climatic variables by reviewing 255 published articles reporting measured or calculated C- and P-factor values. We found a wide variation in both factor values across climatic zones, land use or cover types, and support practices. The average C-factor values decreased from arid (0.26) to humid (0.15) climates, whereas the average P-factor values increased (from 0.33 to 0.47, respectively). Thus, support practices reduce soil loss more effectively in drylands and drought-prone areas. The global average C-factor varies by one order of magnitude from cropland (0.34) to forest (0.03). Among the major crops, the average C-factor was highest for maize (0.42) followed by potato (0.40), among the major orchard crops, it was highest for olive (0.31), followed by vineyards (0.26). The P-factor ranged from 0.62 for contouring in cropland plots to 0.19 for trenches in uncultivated land. The C-factor results indicate that cultivated lands requiring intensive site preparation and weeding are most vulnerable to soil loss by sheet and rill erosion. The low P-factor for trenches, reduced tillage cultivation, and terraces suggests that significantly decreased soil loss is possible by implementing more efficient management practices. These results improve our understanding of the variation in C- and P-factors and support large-scale integrated catchment management interventions by applying soil erosion models where it is difficult to empirically determine the impact of particular land use or cover types and support practices: the datasets compiled in this study can support further modeling and land management attempts in different countries and geographic regions