Koskimäki, TeemuEklund, JohannaMoulatlet, Gabriel MTuomisto, Hanna2023-09-120376-8929http://hdl.handle.net/1885/299444Summary Protecting tropical forests from deforestation is important for mitigating both biodiversity loss and anthropogenic climate change. In Amazonia, a common approach to protected area (PA) impact studies has been to investigate differences among broad PA categories, such as strictly protected, sustainable use and indigenous areas, yet these may be insufficient for the management of PAs at local scales. We used a matching method to compare impacts and carbon emissions avoided during 2011-2016 of individual PAs in the state of Acre (Brazil). Although most PAs had a positive impact and effectively prevented forest loss, we observed substantial variation among them in terms of impacts, pressures and emissions during our study period. The impacts varied from 3.6% avoided to 15.6% induced forest loss compared to expected levels of deforestation estimated for each PA using the matching method. All but a few PAs helped avoid substantial amounts of emissions. Our results emphasize the need for more PA impact studies that compare multiple PAs at the individual level in Amazonia and beyond.Our work was supported by the Kone foundation (grant 201608614 to JE) and the University of Turku Graduate School (grant to GMM).application/pdfen-AU© The Author(s), 2021. Published by Cambridge University Press on behalf of Foundation for Environmental Conservation.arc of deforestationcounterfactualdeforestation fronteffectivenessREDDþtropics202110.1017/S03768929210002292022-07-31