tables are not reflecting the global means in the lat-lon plots

[cecilehannay]

ADF run type

Model vs. Model

What happened?

I am a bit concerned about the numbers in the ADF tables. They don't match the numbers in the ADF lat-lon plots. We should try to see why this is the case.

ADF Hash you are using

Ask Justin

What machine were you running the ADF on?

CISL machine

What python environment were you using?

NPL (CISL machines only)

Extra info

No response

[brianpm]

My initial assessment of this suggests two small changes that should be made for the table:

• when taking the temporal average, the table script should weight by the length of month
• it does not look like RESTOM is doing an area-weighted average

The lat-lon plots correctly use the month length IF weight_season is True. I wonder whether there are ever cases in which we would not want to do this. I can't think of any; the only caveat would be that I'm assuming that xarray can figure out the number of days per month for any arbitrary calendar (thinking of 360-day calendar as the most exotic).

@cecilehannay -- Is the inconsistency across all variables? Is it big for RESTOM but small for other fields? I'm not immediately seeing anything else, but I will keep investigating.