Dynamics of soil fragment size distribution under successive rainfalls and its implication to size-selective sediment transport and deposition

Abstract

Changes in soil fragment size distribution (FSD) under the impacts of successive rainfall events can have strong effects on sediment export during erosion events. However, field observations on the temporal dynamics of FSD under changing rainfall conditions are rare. In this study, we applied five successive artificial rainfall events with two different rainfall intensities (25.8 and 49.1mmh−1) on two replicated agricultural soil plots over a period of five weeks. Surface soil samples were taken immediately before and after each rainfall event and also in between the rainfall events in order to monitor surface aggregate stability and soil FSD. The FSDs resulting from rainfall-induced aggregate breakdown were used as inputs into a size-selective sediment transport model in order to assess how the variations in FSD would affect the concentration and size of exported sediments. Aggregate stability at first decreased due to generally wet conditions, but then increased with soil drying, until reaching a plateau towards the end of the experimental period. It did not fully recover to initial levels though, probably due to loss of stabilizing agents over time. Independent of soil moisture, notable re-aggregation occurred after each event but again did not reach the full restoration of the initial state even after up to 10days of regeneration. In contrast to aggregate stability and irrespective of pre-event re-aggregation, the percentage of coarse fragments in post-event FSDs continued to decrease from event to event, suggesting that in addition to the previously re-formed aggregates also new coarse aggregates were broken down. The model simulations showed that the temporal variations in post-event FSD would have had as much influence on the amount and size distribution of sediments exported with surface runoff as a 3-fold variation in unit discharge, suggesting the importance of quantifying rainfall-induced FSD in soil erosion assessment.