This study aims to address these issues, by developing simple methodologies to assess the statistical significance of future soil erosion rates based on comparing relative changes to natural variability around present-day rates, and determine whether these changes may be problematic by comparing absolute soil erosion rates to tolerable thresholds. A modelling study across six hillslopes in Northern Ireland is conducted to demonstrate these methodologies, using the Water Erosion Prediction Project (WEPP) model. The direct impacts of climate change are modelled using statistical downscaling methods and a simple sensitivity analysis in the case of sub-daily precipitation data, whilst a scenario approach is taken in order to model the indirect impacts of changing land use and management.
Results indicate a mix of soil erosion increases and decreases, depending on which scenarios are considered. Downscaled climate change projections in isolation generally result in erosion decreases, whereas large increases are projected under many scenarios where changes in sub daily rainfall intensity and land use are accounted for. Only the most extreme scenarios reveal the potential for on-site problems of soil erosion, whilst the off-site impacts are likely to become a more considerable environmental issue with respect to water quality and muddy flooding under a wide range of future scenarios. A number of scenarios reveal statistically significant increases in soil erosion from the present day, but these are confined to only two out of the six sites. At the remaining four sites, high absolute rates are projected under many scenarios, despite not representing statistically significant increases, and there also exist a number of cases where statistically significant increases translate to low absolute erosion rates. This illustrates the importance of analysing both relative and absolute soil erosion rates, as the former allows us to isolate the impacts of climate change in contributing to erosion, whereas the latter allows us to determine whether those changes may be problematic by comparing them to tolerable thresholds. The methodologies outlined and demonstrated in this study provide a simple means of generating these additional pieces of information which could prove useful in decision-making contexts.