A Stochastic Approach to Scouring Events Promoting Removal of Riparian Vegetation

The evolution of riverine habitats is governed by the mutual interactions between the hydrological regime, sediment transport and riparian vegetation dynamics. In particular, riparian vegetation succession, establishment or mortality due to flow uprooting strongly depends on its resilience to flow magnitude, hydrograph duration and bed scouring depth. In this work, we apply a probabilistic approach to model plant uprooting and correlate its probability to the return period of flood events able to promote vegetation uprooting by riverbed erosion. We model the stochastic occurrence of flow discharges by considering the Compound Poisson Process. Then, we apply the Peak-Over-Threshold (POT) analysis to calculate the magnitude and duration of the reference scouring event and a very recent stochastic model to determine the uprooting probability of riparian vegetation due to such erosion event. By correlating the uprooting model to the POT analysis, we define the parameters of both the hydrological regime, the morphological scouring process and the biomechanical characteristics of plants that play a key role in determining the removal dynamics of riparian vegetation. We validate the combined approach using field data from measurements of mechanical uprooting of pioneer woody species. Our analysis shows that the existing correlations between hydro-morphological processes and riparian vegetation uprooting may be the key aspect for species recruitment and survival according to plant characteristics and flow magnitude and duration. Outcomes suggest that flow regime alterations due to climate change and human impact may lead to changes in the hydrological system and promote ecosystem shifts.