Reality check for erosion experiments under simulated rainfall: a meta-analysis in China

The results reveal systematic discrepancies between simulated rainfall experiments and natural rainfall observations, indicating that simulated rainfall tends to overestimate runoff rates compared with natural rainfall. Although the erosive capacity per unit of runoff was similar between the two methods, the higher runoff under simulated rainfall led to systematically greater estimates of soil loss. These differences, along with varying abilities of vegetation covers to reduce runoff and soil loss, as well as the distinct roles and pathways of environmental factors in these processes, underscore that simulated rainfall experiments may not adequately represent the complexity of natural rainfall or the scale effects inherent in real-world erosion processes. This is largely due to their simplified rainfall patterns, shorter durations, and different plot scales used in these experiments compared to natural observations.