Model simulation of the future climate is the only method that allows us to study how the climate may respond to increasing levels of greenhouse gases. However, using projections requires some familiarity with how the simulations were produced so that uncertainty is adequately considered (see also Understanding climate change projections ). A related topic is that of confidence in projections and confidence ratings, see Confidence ).
There are diverse uses of climate change information reflecting the many different application areas, such as studying impacts of future climate change on biophysical or built systems or identifying adaptation strategies for businesses and communities. Depending on the application, projection information is used somewhat differently; see further discussion on the page, Approaches to climate change impact assessment . For most impact assessments , climate projections serve a key role in the analysis, though many real world applications also need to consider non-climate related factors.
Information and data sets used should be as representative as possible of current knowledge of regional climate change, see further discussion on the page Data vs projections . All information and data must be placed in context of the emissions scenarios. In addition, current understanding of what regional changes are plausible must be considered (see regional projections described in Technical Report Chapters 7 & 8 and in the Cluster Reports ).
There are many different projection products available (see Deciding what data you need ) and for many applications summary statements on the range of regional change are adequate . The Decision Tree is designed to support users in finding the level of detail appropriate for their needs. If detailed data are required, a further Regional projections data pathway can be used.
If you do need to use future climate datasets, it is always best to use the simplest method that gives acceptable results. This means the uncertainties and the effect of different choices is as transparent as possible. Different applications require different inputs, and the choices are therefore also different. Here are some example case studies of selecting and producing Application-ready data for different applications:Hydrology example case study Biodiversity example case study Human health (heat) example case study
Page updated: 10th August 2016