Vegetation patterns in South America associated with rising CO2: Uncertainties related to sea surface temperatures

Spatially precise forecasts of the impacts of climate change on the distribution of major vegetation types are essential for the implementation of effective conservation and land use policy. However, existing studies frequently omit major sources of climate variability that can significantly increase the uncertainty of projections. In this study we demonstrate how different predictions for sea surface temperature (SST) for the first half of the twenty-first century increase the uncertainty associated with forecasts of the future distribution of major ecosystems in South America. This is demonstrated through a numerical experiment using a coupled climate-vegetation model (CCM3-IBIS) for IPCC emission scenario A2 that incorporates the SST data from ten different models. The study reveals an increasing uncertainty in the ability to forecast future vegetation patterns, such that by 2050 the simulation is unable to robustly forecast the vegetation cover in an area equivalent to 28 % in South America (5 × 106 km2). The future of the central and northeastern regions of Brazil is especially uncertain, with outcomes, ranging from savanna, and open shrubland to grassland. Recognizing and managing such uncertainty should be a priority for decision makers. © 2012 Springer-Verlag