Projections of global warming over the coming century are driving a shift to renewable energy production, however the future rate of warming both globally and locally remains uncertain. Even for a specific greenhouse gas forcing scenario, such as the ‘RCP8.5’ scenario used in the current climate model intercomparison project CMIP5, changes from individual models differ considerably. This can be attributed to differences in both the global sensitivity to forcing, the local sensitivity to large-scale warming, and also unforced variability on decadal timescales. An approach to representing these three factors using probability density functions (PDFs) has been developed and is applied here to CMIP5 simulations from 25 models, using data processed for projections supported by the Australian Natural Resource Management Climate Change Fund. The approach provides both the best estimates, and high and low percentiles of change, as a function of time and also forcing scenario. Accompanying such warming are changes in precipitation, cloud cover and atmospheric circulations. Of specific interest to the energy industry are changes in wind, solar, and water resources. PDFs can be generated for these variables also, specifically for downward solar radiation, mean wind speed at 10 m, and simulated runoff. The results will be presented in the form of maps over Australia and the globe, and as time series for individual locations. While the details will depend on the models available and the method, the PDFs provide an indication of the potential for change in the energy resources. This could be useful for planning in both the near-term, particularly with regard to interdecadal variations, and in the long-term, in which forced changes may be more substantial.