Main content area

Determination of erosive events in the Chinese Loess Plateau using the runoff threshold

Zheng, Mingguo, Cheng, Weiming
Journal of soils and sediments 2017 v.17 no.4 pp. 1182-1190
base flow, bedrock, drainage, highlands, rain, runoff, sediment yield, storms, subwatersheds, t-test, China
PURPOSE: Exclusion of non-erosive rainfalls is highly desirable in calculating the rainfall erosivity. It is common practice to use a rainfall threshold to separate erosive from non-erosive events. This study aims to determine erosive events in the Chinese Loess Plateau using the runoff threshold. MATERIALS AND METHODS: The data used mainly include h (runoff depth; total runoff volume normalized to drainage area) and specific sediment yield (SSY) of 1475 storm events observed at three scales of hillslope, subwatershed, and watershed. The linear h-SSY relationships (SSY = bh + a) were established. The x-intercept term (−a/b) was interpreted to be the h threshold between erosive and non-erosive events. The t test was used to test whether the constant term a is statistically zero or not. RESULTS AND DISCUSSION: The t tests demonstrated that the h-SSY relationship was proportional on upland slopes, but linear with a positive x-intercept in subwatersheds and watersheds. Accordingly, the h threshold of erosive events varied from zero on upland slopes, implying that any runoff event is erosive, to a positive value in subwatersheds and watersheds. The positive h threshold can be related to the preferential generation of runoff on the bedrock channel at the subwatershed scale and to the base flow at the watershed scale. Using the h threshold can remove approximately 85% of rainfall events from the calculation of rainfall erosivity. CONCLUSIONS: The h threshold not only is equally effective as the rainfall threshold in removing non-erosive events but also allows a more accurate estimation of rainfall erosivity.