Extreme precipitation events play an important role in shaping the landscapes. The effect of these events on catchment erosion is expected to significantly depend upon the precipitation intensity and duration, as well as vegetation density and plant functional types (PFT). This study applies the Landlab-SPACE landscape evolution model (LEM) to investigate the vegetation-dependent erosional response at a daily time scale along a steep climate and ecological gradient in the Chilean Coastal Cordillera. The selected study areas range from (arid) barren land in the north (~26 °S) to (humid-temperate) dense forest in the south (~39 °S). Simulations were conducted with precipitation observations from 1982 to 2019 (repeated 50 times). The (area-normalized) runoff rates (predicted using LEM) at the catchment outlet were cross-checked with predictions from the Soil and Water Analysis Tool (SWAT) hydrological model. A strong positive correlation (Pearson r > 0.8, > 95% significance level) was observed between the runoff rates predicted by the LEM and the hydrological model. The results suggest that during extreme precipitation events, erosion rate changes in (arid) barren land are ~ 14%, ~ 86%, and ~ 93% more sensitive to precipitation changes relative to (semi-arid) grassland, (Mediterranean) shrubland, and (humid-temperate) dense forest, respectively. This sensitivity is attributed to (1) a higher prevalence of low-frequency, high magnitude precipitation events during El Niño periods; and (2) the absence of erosionally resilient plant function types (PFT) in northern catchments (i.e., arid and semi-arid settings).