Forest impact on floods due to extreme rainfall and snowmelt in four Latin American environments 1: Field data analysis

► Flood peak data are analyzed for four focus catchments in Latin America. ► Forest cover is unlikely to reduce peak discharges generated by extreme rainfall. ► Forest cover may reduce flood peaks for moderate (more frequent) rainfall events. ► The effect occurs at scales up to 1500 km 2 if the forest cover change exceeds 30%. Flood peak data for focus catchments in Costa Rica, Ecuador, Chile and Argentina are analyzed to test the hypothesis that, as the size of the hydrological event increases, the effect of forest cover on the peak discharge becomes less important. Previous research suggests that this hypothesis may hold for small catchments (less than 1 km 2) but the pattern is less clear for large catchments. The principal study results are for small paired catchments (0.6–10 km 2) with different forest covers (forest/pasture) in highland Ecuador and a small (0.35 km 2) plantation catchment in southern Chile subjected to logging. The former were analyzed by comparing the corresponding peak discharges for given rainfall events, the latter by comparing the relationships between peak discharge and rainfall event size for the pre- and post-logging periods. In all cases there is relative or absolute convergence of the responses as discharge increases, with convergence likely for flood return periods of around 10 years. More limited data for larger catchments which have undergone either deforestation or afforestation (131 km 2 in Costa Rica and 94–1545 km 2 in Chile) suggest that the percentage change in forest cover must exceed 20–30% to provoke a measurable response in peak discharge; convergence of peak discharge response at high flows (return periods of around 5 years) for the different forest covers may then be observed. For a 12.9-km 2 snowmelt-affected catchment in Tierra del Fuego, Argentina, extreme floods require rain-on-snow events but the data are not sufficient to quantify the complex relationship between forest cover, event return period and peak discharge. In general, forest cover is unlikely to reduce, significantly, peak discharges generated by extreme rainfall but may still offer substantial mitigation benefits for moderate (i.e. more frequent) rainfall events.