This paper describes the way in which the socio-economic projections in the SRES scenarios were applied in a global-scale assessment of the impacts of climate change on food security, water stresses, coastal flood risk and wetland loss, exposure to malaria risk and terrestrial ecosystems. There are two key issues: (i) downscaling from the world-region level of the original scenarios to the scale of analysis (0.5°×0.5°), and (ii) elaborating the SRES narrative storylines to quantify other indicators affecting the impact of climate change. National estimates of population and GDP were derived by assuming that each country changed at the regional rate, and population was downscaled to the 0.5°×0.5° scale assuming that everywhere in a country changed at the same rate. SRES scenarios for future cropland extent were applied to current baseline data, assuming everywhere within a region changed at the same rate. The narrative storylines were used to construct scenarios of future adaptation to the coastal flood risk and malaria risk. The paper compares the SRES scenarios with other global-scale scenarios, and identifies sources of uncertainty. It concludes by recommending three refinements to the use of the SRES scenarios in global and regional-scale impact assessment: (i) improved disaggregation to finer spatial resolutions, using both “downscaled narrative storylines” and new technical procedures, (ii) explicit consideration of uncertainty in the population, GDP and land cover characterisations of each storyline, and (iii) use of a wider range of future socio-economic scenarios than provided by SRES if the aim of an impact assessment is to estimate the range of possible future impacts.