干旱区生态环境知识资源中心

Located in a 200-400 mm precipitation zone, the Zulihe basin (Zulihe River basin) in China is highly sensitive to climate change and land management. The relatively concentrated rainfall during the wet season (June-September) often leads to severe soil and water erosion. With a dataset collected at the outlet of the basin during 1955-2013, we first examined the trends in precipitation, runoff generation, and sediment transport with the Mann-Kendall (M-K) method. Then, we analyzed the impacts of various soil and water conservation practices on soil and water erosion with regression analysis. M-K results suggest overall downward trends in soil erosion and water runoff discharge from the basin with some similarity to the trends in precipitation. However, precipitation can only explain 30% of variability in sediment load and runoff discharge out of the basin. Various soil and water conservation practices were found to play a significant role in forming the downward trends in soil and water erosion. Multi-variate regression analysis confirmed that terraces construction, afforestation, grass land establishment, natural wildland preservation, and establishment of runoff-holding ponds together could explain up to 50% of variations in sediment load and runoff discharge. With an increase from 0 to 59% of land coverage protected by various soil and water conservation practices within the basin, sediment load reduced by at least five times and runoff by up to three times. It should be noted that the effects of soil and water conservation practices varied at both spatial and temporal scales, possibly due to interactions among different practices in terms of their distribution along the landscape and topographic gradients, and effectiveness of each practice. The findings from this study will not only explain the long-term variability in soil and water erosion changes over time but also provide insight for guiding future soil and water conservation programs in this basin and other regions with similar conditions. (C) 2017 Elsevier B.V. All rights reserved.