Selecting scenarios and models

 Selection of appropriate scenarios and models for your needs is a critical step in conducting an impact assessment.

The CMIP5 climate model simulations used in Climate Change in Australia have been undertaken using four greenhouse gas and aerosol emission scenarios , called RCP8.5, RCP6.0, RCP4.5 and RCP2.6 (see Technical Report Section 3.2.1). As well as incorporating a range of emission scenarios into any assessment, users of climate projections are strongly advised to represent a range of climate model results.

Given the similarity of emission scenarios up to the year 2030, it may only be necessary to select one scenario for impact assessment. However, beyond 2030 the emission scenarios diverge, so at least two scenarios should be selected, e.g. RCP4.5 and RCP8.5.

CSIRO’s Climate Futures Framework has been developed to simplify understanding, communication and use of climate projections and to help capture the range of projection results relevant to a specified region. This enables a large amount of climate model output (for a particular time period and emissions scenario) to be condensed into a small number of categories, each of which is defined by a range of change in two climate variables such as temperature and rainfall.

Consideration of model performance is also important and is built in to the Climate Futures Framework approach to developing projections datasets for impact assessments. Some models share a similar heritage and may not be fully independent (This is discussed further in Chapter 5 of the Technical Report ). Therefore, if working with a small number of representative models, it is best to avoid closely related models if possible. The latter is not straightforward, however it was taken into account in the selection of the eight-model subset used for development of the 'application-ready' data available on Climate Change in Australia. 

Model selection is also influenced by the availability of relevant data, since some climate variables were not archived for some models or emission scenarios. The variables for which data can be obtained from Climate Change in Australia are listed here .

Where downscaling has been shown to add value, selecting data from these models would be appropriate. The Climate Futures: Detailed Projections Tool provides information on the performance of all models.

Further reading:

Knutti, R., et. al. 2013. Climate model genealogy: Generation CMIP5 and how we got there. Geophysical Research Letters, 40, 1194-1199.

Moss, R. H., et. al. 2010. The next generation of scenarios for climate change research and assessment. Nature, 463, 747-756.

VAN VUUREN, D. P., et. al. 2011. The representative concentration pathways: an overview. Climatic Change, 109, 5-31.