-
ESMValTool enforces CMIP/CMOR format for datafiles. Most CMIP-formated data have at least lat and lon coordinates, but some have a single dimension for the horizontal, which is simply a site number in a predefined sites list; this is the case e.g. for calipso profiles emulators. Also, some in-situ observations datasets have a similar structure (as e.g. soil humidity profiles measured at a series of observation sites) Can ESMValTool data load engine handle such datafiles, and forward it to diagnostic scripts ? Is there any working example ? |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 1 reply
-
Some CMIP6 variables are in this category, namely those that use the S-na spatial dimensions; it might be worth checking out what happens when one of those is loaded. Apart from that, there has been prior discussion/work in #1119, #1655, and #496. The general thrust of it is to format also in-situ data in the spirit of CMIP following the CF Conventions, most likely particularly CF Convenvtions 1.9, Section 9, "Discrete Sampling Geometries". I have work planned in this direction for next year as well, so if you are getting into that now, let's talk early next year? |
Beta Was this translation helpful? Give feedback.
Some CMIP6 variables are in this category, namely those that use the S-na spatial dimensions; it might be worth checking out what happens when one of those is loaded.
Apart from that, there has been prior discussion/work in #1119, #1655, and #496. The general thrust of it is to format also in-situ data in the spirit of CMIP following the CF Conventions, most likely particularly CF Convenvtions 1.9, Section 9, "Discrete Sampling Geometries".
I have work planned in this direction for next year as well, so if you are getting into that now, let's talk early next year?