Observed Variability and Projected Change in South American Flood Regimes

Assessing future changes in hydrological extremes requires accounting for both externally forced climate change and internal climate variability, which can substantially modulate flood magnitude and frequency. Here, we synthesize observational and modelling evidence to examine how these drivers jointly shape flood regimes across South America, with implications for both flood risk and ecosystem dynamics.Using multi-decadal streamflow observations, we show that internal climate variability associated with the El Niño–Southern Oscillation (ENSO) strongly alters the likelihood of extreme hydrological events. Flood probabilities increase by more than 120% during El Niño in the La Plata Basin and during La Niña in the northern Amazon. Streamflow extremes respond more strongly than precipitation, indicating cumulative hydrological amplification of climate variability.Complementing these findings, hydrodynamic–hydrological simulations forced by the CMIP6 ensemble reveal heterogeneous future flood responses under climate change. Flood magnitude and frequency are projected to intensify markedly in southern Brazil, where events may become up to five times more frequent, while major wetlands such as the Amazon and Pantanal are projected to experience reduced flood occurrence, with potential negative ecological consequences. These contrasting responses arise from competing influences of increasing extreme precipitation and enhanced evapotranspiration, as well as substantial spread across climate model realizations.Together, these results demonstrate that internal climate variability can amplify, mask, or temporally offset forced changes in flood regimes, leading to divergent but physically plausible outcomes.