Coupling Landscape Evolution and Dynamic Vegetation Models to Simulate Sediment Fluxes under Historical and Future Climate in the Laval Catchment (Draix-Bleone CZO, SE France)

Coupled interactions between climate, vegetation, and geomorphic processes control sediment export from rapidly eroding badlands; however, their relative roles under future climate scenarios remain poorly constrained. We present a coupled landscape evolution model (LEM) and a dynamic vegetation model, CLIMBAD, applied to the Laval catchment (Draix-Bléone CZO, SE France) in a badland setting, to quantify how fluvial and hillslope erosion, together with frost weathering and vegetation dynamics, drive historical and projected sediment fluxes. The LEM is forced by temperatures (acting on frost weathering) and precipitation events, including depth, duration, and peak intensity. The dynamic vegetation model is calibrated to 1982-2021 vegetation maps and driven by topographic and climatic variables. Future climate (2022-2099) is generated using a stochastic weather generator calibrated on observations from historical data and future projections obtained from a regional climate model.

Model evaluation for 1985-2021 shows that coupling dynamic vegetation to the LEM improves agreement with observed annual sediment fluxes at the catchment outlet (i.e., R² increased from ~0.60 to ~0.66), demonstrating the importance of vegetation-erosion feedbacks. To isolate climatic controls, we ran four scenarios for future climate: (i) changing temperature (T) and precipitation (P), (ii) constant T with changing P, (iii) changing T with constant P, and (iv) constant T and P. These results help disentangle the relative contribution of a change in the precipitation regime and a change in temperature on sediment fluxes in a coupled system, with direct implications for sediment hazard assessment in climate‑sensitive badland landscapes.