FAQ Data and Access
How do I access data from CORDEX?
You can find information on how to access the data following the links on the website under How to access the data. If you can´t find what you are looking for you can contact the Point of Contact listed for your domain at https://cordex.org under Domains.
How do I download data from ESGF?
Detailed instructions on how to download CORDEX data from ESGF are available on the CORDEX website under ESGF (https://www.cordex.org/data-access/esgf/). More user tutorials for ESGF can be found at the Earth System CoG page (https://www.earthsystemcog.org/projects/cog/tutorials_web).
How do I access data for a specific country/region?
Look for your region, in CORDEX called a domain (https://www.cordex.org/domains/). For example, the CORDEX-Africa domain has acronym “AFR-44” and covers the entire African continent. See Table 1 in the CORDEX Archive Design (http://is-enes-data.github.io/cordex_archive_specifications.pdf) for a list with acronyms for all CORDEX domains. Suffix “i” in the CORDEX domain acronyms (e.g. AFR-44i) means a regular grid (see Table 2 in the CORDEX Archive Design).
According to the CORDEX Archive Design only monthly and seasonal means are supposed to be remapped to the CORDEX regular grids. Daily and sub-daily output is available on the RCM native grids. At the moment there is no service at ESGF allowing cropping a sub-region from the CORDEX domains
Some data portals can provide such functionality, see for instance Climate4Impact (https://climate4impact.eu/). There are a number of RCM/GCM combinations for all CORDEX domains, more information on these combinations can be found at the first page on the website https://cordex.org. Note that this matrix is sparse, not all RCMs have downscaled all GCMs and some simulations might be missing.
Is CORDEX output data open source? Can I use this data for my academic research without any copyright issue?
Is CORDEX output data freely accessible?
No charge applies on any CORDEX data. However, some CORDEX datasets can have a restriction to be used for non-commercial research and educational purposes only. To download CORDEX data through ESGF all users have to first sign up to the CORDEX Research or CORDEX Commercial group (https://www.nsc.liu.se/storage/esgf-datanode/).
Has CORDEX data been updated with CMIP6 and are this data available for download?
No, all CMIP6 activities have been delayed and no CORDEX simulations driven by the CMIP6 GCMs are yet available through ESGF. CORDEX documentation for downscaling CMIP6 will be published on the web shortly. At the moment you can use the CORDEX simulations driven by the CMIP5 GCMs available on ESGF. Among those are new simulations uploaded recently – the XXX-22.
Is there a formula or algorithm I can apply to convert the real world co-ordinates to rotated ones for extraction of data?
RCMs use a number of different native projection/coordinate systems that include for example the rotated coordinates, Mercator and Lambert Conic Conformal (LCC) projections. The rotated coordinate system is the most common one and is used in many CORDEX RCMs.
In some CORDEX domains (e.g. Africa and MENA) there is no rotation but RCMs still use the rotated coordinates. In this case, the rotated coordinates coincide with geographical ones which is not the case for other CORDEX regions with rotation (e.g. Europe and South Asia).
One of the CORDEX archiving requirements is to provide both native RCM coordinates (rotated, Lambert Conformal etc.) and the geographical ones. It’s supposed that all CORDEX RCMs should provide the 1D native coordinates (rlon/rlat for the rotated coordinates and x/y for the Mercator and LCC ) and additionally the usual 2D geographical coordinates (lon/lat). The 2D lon/lat coordinates are curvilinear for almost all CORDEX domains.
Note, that not all software packages support such double coordinates in the CORDEX netcdf files and some additional configuration can be necessary. The CDO tool (https://code.mpimet.mpg.de/projects/cdo/) can be used for converting the rotated coordinates or LCC projection to the geographical ones. However, this tool is for Linux-based system (command line).