Recent land use changes in Central America have involved the abandonment of marginal farmland activities, the regeneration of secondary forest and the spread of high return crops such as oil palm plantations. The potential impacts of land use change on overland flow are evaluated using data from Tinoco Experimental Catchment (South Pacific Costa Rica). Our main hypothesis is that secondary forest overland flow is lower than the one generated under the other land cover types. For this purpose, runoff responses at plot scale are analyzed for different land uses: secondary forests, forest plantations, oil palm plantations and grasslands. Runoff plots were situated over former grasslands, abandoned 8-15 years prior to plot settlement Measurements were conducted at two complementary spatial scales i) the plot (150 m(2)) under natural precipitation and ii) rainfall simulation on microplots (0.0625 m(2)). The combination of natural and simulated rain runoff response measurements provides a more accurate picture of the overland flow generation in the study site. Secondary forest shows a significantly lower runoff response than grassland and oil palm plantations, although there are no significant differences among the plots in variables such as saturated hydraulic conductivity (Ks). The oil palm plantation plot presented the highest runoff coefficient (mean RC = 32.6%), twice that measured under grasslands (mean RC = 153%) and 20-fold greater than in secondary forest (mean RC = 1.7%). The runoff plots part of the Tinoco Experimental Catchment provide valuable data and coefficients for evaluating the influence on overland flow of secondary forest recovery and oil palm plantation expansion over hillsides, contributing to a better understanding of the effects of land cover dynamics on water resources in the humid tropics. (C) 2015 Elsevier B.V. All rights reserved.