This study aims to assess the impact of terrestrial protected areas (PAs) on deforestation in Madagascar, focusing on PAs established between 2002 and 2022. Utilizing a counterfactual framework, we will apply coarsened exact matching and difference-in-differences methods to evaluate forest cover loss, drawing on 24 years of high-resolution satellite data. To enhance comparability between treatment and control groups, we will account for confounders affecting both deforestation and PA creation. Additional controls will be incorporated to improve estimation precision. Furthermore, we will include buffer zones to assess spillover effects. The impact will be estimated by PA and by year, allowing for an analysis of the heterogeneity of PA effectiveness, its patterns, and plausible explanations. Our study seeks to address limitations of previous research, including inaccurate data on PA locations and creation dates, inappropriate definitions of forests, limited temporal scopes, and potentially biased estimation methods.