{A, LI, L, OI, O, aero}clim variable groups are missing for PMIP

[jypeter]

When you go to http://clipc-services.ceda.ac.uk/dreq/u/PMIP.html, you see that the {A, LI, L, OI, O, aero}clim variable groups have 0 variables, e.g. PMIP-Aclim {PMIP}: 0 variables, 1 request links

This is probably due to the fact that we used the following way to request it in the PMIP request (https://pmip4.lsce.ipsl.fr/doku.php/database:pmip4request#the_pmip4_request), and it was probably missed by the automated tool:

We need ALL the variables selected in PMIP_standard_output.xls for Amon, and the variables in PMIP-Amon to be generated at the monClim frequency

This basically means cloning whatever is requested for {A, LI, L, OI, O, aero}mon to {A, LI, L, OI, O, aero}clim. Note that I'm also going to check the content of the mon tables

Note that we are adding variables to the DB, but this will have a very small footprint in the global DB, since we will have only 12 time steps for each variable. These clim variables have proved to be very useful for PMIP3/CMIP5

[jypeter]

I have noticed that some of the new variables we have requested for Amon have found their way to AERmon and Emon, so we will need the matching variables in AERclim and Eclim

[martinjuckes]

These were not added because the WIP decided that it would be an unjustified additional load on modeling centres -- the information is available in the requested monthly variables. A downstream service could calculate these products and do so with greater flexibility.