# What are the guidelines for constructing variable names that follow CF conventions?

I am attempting to develop variable names following Unidata's climate forecasting (CF) variable naming guidelines, but for biological variables used to parameterize and evaluate land surface and ecosystem models (See a first draft on Google drive). However, I can not find the documentation for these guidelines.

Until early 2014, the guidelines were here: http://cf-pcmdi.llnl.gov/, but now there is a link to http://cfconventions.org/, and then a list of standard names, but the guidelines page that this points to does not (currently) exist.

I think it would be helpful here to have a link to the appropriate documentation, but also a summary of the guidelines (i.e., how to construct variable names that are consistent with the CF approach). In addition, any feedback on the standard_name field in my draft would be appreciated.

Possible this is not what you need, but here's an archive of the Guidelines for Construction of CF Standard Names, which is "Version 1, 3 December 2008" and was cached on July 28, 2013; guessing that the page has not been updated since December 2008.

Also, the source code repository for CF Conventions website you linked to (cf-convention.github.io) is viewable from here; if you had questions about the status of the site, you might try contacting the developer listed on Github for the project; their email is listed on their Github profile.

UPDATE: Here's the contact info for the CEDA Data Scientist; their "main role in CEDA is the development of metadata standards, particularly the CF (Climate and Forecast) metadata conventions."

As a result of @blunders answer, I located the document, and will summarize here:

The general framework for constructing names is:

[surface] [component] standard_name [at surface] [in medium] [due to process] [assuming condition]

where

• surface is a horizontal plane, e.g. tropopause, or sea_level.
• Single word surface names go at the beginning; surfaces with more than one name goes after the standard_name
• tropopause_stuff
• stuff_at_sea_level
• component is a vector, e.g. upward, northward, x, etc.
• medium is, e.g. in_water, in_soil, etc.
• process starts with "due_to", e.g. due_to_longwave_heating
• assuming condition is a conditional assumption, e.g. assuming_clear_sky

Other conventions: there are a number of further conventions provided to standardize meaning, e.g. in the construction of standard_names when transformations, special phrases, chemical species, or generic names are used. For example,

• transformations imply a particular form of units, e.g ratio_of_X_to_Y implies a specific form for the units (in this case [x]/[Y].
• generic names imply specific units, e.g. amount implies kg/m2, area implies m2, etc.
• there are standard spellings for chemical species, e.g. nitrate, carbon_dioxide
• phrases: water refers to H$_2$O in all phases; wrt = with respect to, toa = top of atmosphere, etc.