River Ecomorphodynamic Models Exhibit Features of Nonlinear Dynamics and Chaos

• Published in: Geophysical research letters 2024-06, Vol.51 (11), p.n/a
• Main Authors: Cunico, I., Bertoldi, W., Caponi, F., Dijkstra, H. A., Siviglia, A.
• Format: Article
• Language: English
• Subjects: Complexity; Dynamical systems; Flood predictions; Mathematical models; Nonlinear dynamics; Nonlinear feedback; Nonlinear systems; Numerical models; Numerical simulations; Predictions; River beds; Riverbeds; Rivers; Scour; Sediment; System dynamics; Uncertainty; Uprooting; Vegetation; Weather forecasting
• ISSN: 0094-8276; 1944-8007
• DOI: 10.1029/2023GL107951

Modeling the nonlinear interactions between flow, sediment, and vegetation is essential for improving our understanding and prediction of river system dynamics. Using simple numerical models, we simulate the key flow‐sediment‐vegetation interaction where the disturbance is intrinsically generated by the presence of vegetation. In this case, biomass growth modifies the flow field, induces bed scour, and thus potentially causes vegetation uprooting when erosion exceeds root depth. Our results show that this nonlinear feedback produces deterministic chaos under a wide range of conditions, with complex aperiodic dynamics generated by a period‐doubling route to chaos. Moreover, our results suggest relatively small values of Lyapunov time, spanning 2–4 growth‐flood cycles, which significantly restrict the predictability of riverbed evolution. Although further spatial and temporal processes may add complexity to the system, these results call for the use of ensemble methods and associated uncertainty estimates in ecomorphodynamic models. Plain Language Summary Simple models that mimic the interplay between river dynamics and vegetation show that changes in the shape of riverbeds can be unpredictable. Our research suggests that we can make skillful predictions on time scales of the order of a few decades for many typical rivers, but we cannot accurately predict what riverbeds will look like in the long term. This has important implications for how we manage and restore rivers. It means that we need to run multiple numerical simulations to generate different scenarios, similar to the methods used in weather forecasting, to more accurately represent the uncertainty in our predictions. The chaotic behavior we found is not unique to rivers; it may be important in other systems where the interaction of flow, bed, and vegetation is key to how they change over time. Key Points Coevolution of riverbed and vegetation, mediated by intrinisic biogeomorphic feedbacks, exhibits nonlinear dynamics and chaos For chaos to be possible, the erosion during a flood must be comparable to the root depth The time scale on which the riverbed/vegetation evolution becomes unpredictable is short, ranging between 2 and 4 growth‐flood periods