CRU : Research : STARDEX
STARDEX Deliverable D16
To recommend the more robust statistical and dynamical downscaling methods for the construction of scenarios of extremes
The choice of the best downscaling technique depends on the region and the application. There is no "one size fits all" solution to downscaling. Therefore this deliverable provides guidelines for those wishing to undertake their own downscaling, building on IPCC guidelines. The checklist of good practice provided here is also likely to be valuable for impacts scientists and other users who want to assess the robustness and reliability of downscaling results from other sources (e.g., to what extent is the recommended good practice followed in these other studies?). The recommendations and guidelines are based on the analyses performed during the project, but some proposals are also made for additional analyses, together with suggestions for how we might do things differently next time. These good practice guidelines form part of the STARDEX downscaling "toolbox" - the statistical downscaling models available in our toolbox are described in D15 and D12. Application of these tools to construct future scenarios is described in D18.
Last updated: November 2005, Malcolm Haylock