obs4MIP catalog about remote_climate_data

[aaronspring]

Hi @jbusecke,

building the demo for the made me realize that this might not even need cmip6_preprocessing but maybe only because I choose a working example.

I took leaf area index lai from some models and got the cmorized observation form obs4mip. The observations is just treated as one of the models: https://gist.github.com/aaronspring/35391fc111be8a123f221c9fa91bf13b

Maybe cmip6_preprocessing is needed when doing gridpoint comparison, which is always messy. Didnt try xesmf or interp_like. went for the easy-just-plotting-the-mean example.
So my idea for this catalog is to collect cmorized observations in a catalog, so users can quickly fetch the data from esgf via http without any effort.

[jbusecke]

Hi @aaronspring , thanks for looping me in here. I am still a bit unclear about the connection to cmip6_preprocessing here. Do you want to be able to 'clean-up` the data so that intercomparison is easy? Or would you want to integrate the catalog building into the package?

[aaronspring]

that was the idea we come up with when I visited Boulder @andersy005
getting observations for CMIP analysis easily.

[aaronspring]

would that be an interesting addition for cmip6_preprocessing? @jbusecke From a user perspective wanting to use observations, maybe cmip6_preprocessing would also be needed to correct the data a bit.
A nice collection of observations, already cmorized, to be downloaded from an intake catalog. only once some manual work needs to be done to fill a catalog, but then everyone could use it easily. similiar to the thoughts about creating cmip6_preprocessing

[aaronspring]

US node:
https://dpesgf03.nccs.nasa.gov/thredds/catalog/esgcet/legacy/ana4MIPs.NOAA-NCEP.CFSR.atmos.mon.v20160517.html