Climate change impacts are important in shaping large ecotones, such as the transition zone between the Cerrado and Amazon rainforest (CAT) biogeographical domains. The accelerating rate of conversion of native vegetation, the most important factor for biodiversity loss in the Anthropocene, compounded by projected climate change impacts, requires a review of the effectiveness of existing designated protected areas (PA) and indigenous land (IL), where low-intensity and low-impact land use prevails. We identified priority tree species for conservation and quantified changes in their projected spatial distribution in future climate scenarios to estimate the conservation effectiveness of the current network of PA and IL in the CAT. Applying niche-based models to compare the geographical range of species in current and future climates, we estimated the displacement of species from their current distribution owing to projected climate change. We used four different IPCC emission scenarios for 2050 and quantified the losses or gains in species richness in PA and IL. All species were projected to suffer a reduction of climatically suitable area and a consequent range reduction. Inside both PA and IL there was a projected decrease in richness of the target species under climate change. The current PA network and designated IL in the CAT do not appear to safeguard future conservation of the species they currently contain. The future ‘climate refugia’ that our work identified could form the basis of plans to expand the protected area network in a region that remains under ever increasing pressure of deforestation in Brazil.