Showstopper: Documentation of the time averaging, interpolation and regridding methods

[susburrows]

Documentation of the time averaging, interpolation and regridding methods (which algorithms were used, how were weights calculated, including the name of tool, and methods used to calculate the weights (e.g. SCRIP low/high order), and once weight is calculated what tool was used to do the regridding) is needed both for the climatology script and the diagnostics scripts. Presumably standard algorithms were used, but we would like to know which ones and to have this documented somewhere where we can find it easily.

[painter1]

The meta-issue is where to put the documentation. Some of these things have been documented, but nobody knows where to look. I'd like to see some opinions on this, keeping in mind that it should be convenient to both ACME and other users.

[susburrows]

I know there is a lot of documentation of the UV-CDAT capabilities, but what we are asking for is some documentation of the ACME workflow tools, i.e. what choices were made in the ACME regridder, and what choices were made in the uvcmetrics package for each calculation. I'm not sure if that is documented anywhere at the moment. Maybe I just don't know where to find it. If that documentation already exists, could you point me to it?

We're not as concerned with the documentation being convenient to non-ACME users as with it existing in a place where ACME users can easily locate it.

[painter1]

I wrote messages and various times, Confluence and email, about some of these issues. Sometimes they were correct. It would be some effort for me to find them, though! The basic story is that averaging and regridding methods are almost always the UV-CDAT defaults.

The questions remains of where to put documentation so that it will be found.

[susburrows]

Jeff, we would like the documentation to be in an ACME Confluence page. You can also mirror somewhere that would be convenient to non-ACME users if you like, but since Confluence is the central location for all ACME documentation, we'd like it there. Comments on Confluence pages (or emails) are not useful as a permanent location for documentation -- the important information needs to be in an organized, and relatively permanent, documentation page. The SE team has done a really nice job of this by first getting the structure and basic documentation in place and then whenever questions or additional information come up, immediately updating the page to include the new information. We do have some UV-CDAT diagnostics documentation in the ACME wiki already, but it needs to be more organized and more complete.