Résumé
Deforestation is a major source of anthropogenic greenhouse gas emissions(1), and the greatest single driver of species extinctions(2). The reduction of emissions from deforestation and forest degradation (REDO) has been formally recognized as a climate change mitigation option. REDO might have important co-benefits for biodiversity conservation(3-10), yet the extent of these benefits will depend on as-yet untested associations between fine-scale spatial patterns of deforestation, species distributions and carbon stocks. Here we combine a global land-use model(11) and spatial data on species distributions(12-14) to explore scenarios of future deforestation within REDO-eligible countries, to quantify and map the potential impacts on species extinctions as increased by forest loss and decreased by carbon conservation. We found that the continuation of historical deforestation rates is likely to result in large numbers of species extinctions, but that an adequately funded REDO programme could substantially reduce these losses. Under our deforestation scenarios, the projected benefits of REDO were remarkably consistent across the four methods used to estimate extinctions, but spatially variable, and highly dependent on the level of carbon payments. Our results indicate that, if well designed, adequately funded and broadly implemented, carbon-based forest conservation could play a major role in biodiversity conservation as well as climate change mitigation.