Recent work has shown that the biodiversity of organisms can influence geophysical processes such as the transport of streambed sediments and the erosion of soils. Yet most of this work has been conducted in small-scale fluvial system mimics, demonstrating a clear need to investigate the relationship between biodiversity and erosion in natural systems. We conducted an observational field study across 3 rivers in forested watersheds in northern Michigan, U.S.A., quantifying streambank retreat rates using aerial photos and measuring riparian plant community biodiversity and abundance. We hypothesized that more diverse riparian plant communities would produce greater woody plant stem density and basal area, which in turn would reduce erosion rates of streambanks due to increased root production. We used structural equation modelling to compare causal networks using plant biodiversity metrics to predict streambank migration rate indirectly through effects on plant abundance, as well as models that used migration rate to predict plant abundance indirectly through effects on plant biodiversity. Although structural equation models explained both causal pathways successfully, models using biodiversity to predict migration rate were a better fit to data than models that used migration rate to predict plant biodiversity and abundance. The best performing models suggested plant biodiversity was indirectly negatively correlated with erosion rate (average standardized path coefficient = −0.22), after accounting for environmental differences between sites. This work adds to a growing body of evidence indicating that biodiversity can modify geophysical processes, demonstrating the need to explicitly account for biological variation when considering ecogeomorphic feedbacks.
All Science Journal Classification (ASJC) codes
- Ecology, Evolution, Behavior and Systematics
- Aquatic Science
- Earth-Surface Processes