Statistical downscaling techniques link model outputs to observations providing a projection operator. This operator is usually trained with reanalysis data (e.g. NCEP or ERA) and historical records, being later applied to the outputs of sesonal-to-decadal (s2d) or climate change (cc) simulations. The following scheme illustrates the statistical downscaling process:

The portal works with daily data and its current features are:

Before using the portal, you must read the following documents:
Tutorial of the Data Access Portal <> Some demo slides
[D2B.13] ERA-40 based predictor data set for statistical downscaling (Task 2B.2.5)
[D2B.4] A first prototype of web service for downscaling at seasonal-to-decadal timescales (Task 2B.2.2)
[D2B.19] Extension of the ENSEMBLES web-based service for downscaling (Task 2B.2.9)
You can also consult the following additional documentation:
[pdf] Presentation at the third General ENSEMBLES assembly

The following scheme illustrates the portal's components and their relationship with ENSEMBLES's tasks: GCM forecasts (s2d and cc from RT2A and RT3) are downscaled to local statations or uniform observation grids (RT6) using any of the available downscaling algorithms (RT2B). This process is performed from a web browser following three steps: Predictor selection, predictand selection and downscaling method: