We are in the process of proposing new variables in the context of the CORDEX FPS-URB-RCC and were wondering if there is a place where we can clear out the variable names. As CF does not impose restrictions on variable names (other than the character set to be used), we'd like to try to follow the CMIP variables building rules. These variables will likely come to be used in CMIP global models as resolution increases. However, I cannot find an up-to-date explicit set of rules for variable naming. I found this treasure from AMIP https://pcmdi.llnl.gov/mips/amip/DATASTDS/VARNAMES/main.html (a website running since 1998!), which is quite enlightening. I wonder if a more recent set of rules and key letter acronyms exists. As far as I understand, this is the current database of variables for CMIP: https://clipc-services.ceda.ac.uk/dreq/mipVars.html but seems specific for CMIP6 (is a new one ongoing for CMIP7?). Very useful information can be retrieved from there, including the provenance information.
Some new variables seem to follow these rules (sftflf, sftgrf, sftgif, ...), but others which should be similarly defined have a new building rule (vegFrac, wetlandFrac, ...). Or maybe they are just named like this because they were borrowed from a pre-existing initiative. This seems to explain e.g. the sfcWind variable.
In CORDEX, sfturf was recently introduced for the urban fraction, following the "old" rules and assigning ur to urban. We are now considering to save an impervious area fraction, which could be called sftimf or imperviousFrac. We are also saving variables for particular components of the urban model, with proposed names such as tascan, taspav, tasgree for the air temperature in the urban canyon or over paved or green spaces. We plan to propose new CF area types to code these variables, but we'd like to have some advise as well on the variable naming, even if they are not strictly CMIP for the moment.
We are in the process of proposing new variables in the context of the CORDEX FPS-URB-RCC and were wondering if there is a place where we can clear out the variable names. As CF does not impose restrictions on variable names (other than the character set to be used), we'd like to try to follow the CMIP variables building rules. These variables will likely come to be used in CMIP global models as resolution increases. However, I cannot find an up-to-date explicit set of rules for variable naming. I found this treasure from AMIP https://pcmdi.llnl.gov/mips/amip/DATASTDS/VARNAMES/main.html (a website running since 1998!), which is quite enlightening. I wonder if a more recent set of rules and key letter acronyms exists. As far as I understand, this is the current database of variables for CMIP: https://clipc-services.ceda.ac.uk/dreq/mipVars.html but seems specific for CMIP6 (is a new one ongoing for CMIP7?). Very useful information can be retrieved from there, including the provenance information.
Some new variables seem to follow these rules (
sftflf,sftgrf,sftgif, ...), but others which should be similarly defined have a new building rule (vegFrac, wetlandFrac, ...). Or maybe they are just named like this because they were borrowed from a pre-existing initiative. This seems to explain e.g. thesfcWindvariable.In CORDEX,
sfturfwas recently introduced for the urban fraction, following the "old" rules and assigningurto urban. We are now considering to save an impervious area fraction, which could be calledsftimforimperviousFrac. We are also saving variables for particular components of the urban model, with proposed names such astascan,taspav,tasgreefor the air temperature in the urbancanyon or overpaved orgreen spaces. We plan to propose new CF area types to code these variables, but we'd like to have some advise as well on the variable naming, even if they are not strictly CMIP for the moment.Any hint is welcome!