MetOffice / opsinputs

JEDI library generating VarObs and Cx files
BSD 3-Clause "New" or "Revised" License
4 stars 1 forks source link

update to make varfield_clw name consistent with ufo naming convention #178

Closed brettcandy closed 1 year ago

brettcandy commented 1 year ago

This is a minor update to the expected meta data for Varfield 68 following the ufo naming convention. It is required in JOPA processing for a single obtype (SSMIS)

brettcandy commented 1 year ago

Hi Brett, changes look fine, and I have looked in the 068_VarField_clw.nc4 where I can see the liquidWaterContent in OneDVar. The only thing I could suggest, if it's not too much work, is to run sith with SSMIS and these changes to prove that VAR is happy with them.

I have built the change via bbb into an hpc build and then run ssmis sith. Output is here http://fcm1/cylc-review/taskjobs/frbt?&suite=sith-ssmis-varfieldsupdate&cycles=20210701T1200Z

The change in varstats compared to using the nightly build is shown below. The new run is on the right. Note a smaller O-B rms fit for channels 12-16 which are the imager channels. This is consistent with using the retrieved clw in each rttov calculation.

Screenshot 2023-08-17 145340

brettcandy commented 1 year ago

I have also added a plot of the clw in the varobs file (for level1) as an additional check

Figure_1

brettcandy commented 1 year ago

bump: can this be merged please? I want to trial impact in Batch 6

brettcandy commented 1 year ago

Grazias