Confirm model levels and pressures levels is computed correctly for all models

[dbekaert]

Model levels were incorrectly computed at half levels not at full levels. E.g.:

• ECMWF HRES pressure
• ECMWF HRES model Not providing consistent results.

This has been fixed in the code with ECMWF Support interaction, but needs verification over a region to show the large bias has been removed. We also want to ensure we are now more consistent with the GACOS 2021 paper in terms of mean bias.

[dbekaert]

@kam3545 can you a test over a small region to verify the mean biass issue has been resolved?

[kam3545]

Hi @dbekaert I used below commands with new RAiDER codes after update repo and environment

raiderDownloadGNSS.py --out products_ex2 --date 20210517 --returntime '18:00:00' --bounding_box '32 42 -125 -114' --cpus all

raiderDelay.py --model NCMR --date 20210517 --time 18:00 --station_file ./products_ex2/UNRcombinedGPS_ztd.csv --zref 30000 -v

I found attached error

Thanks log.TXT

[kam3545]

Hi @jlmaurer and @dbekaert New codes are working at global calculation, but I found larger errors as compared to the last version for NCMRRF/NCUM model. Thanks

[dbekaert]

@kam3545 Can we confirm that NCUM product structure identical to the ECMWF model? I understood its model is a modification of the ECMWF model hence; similar variable definitions etc.

[kam3545]

@dbekaert Please provide me a sample file of ECMWF HRES to identify changes in NCUM vs ECMWF model's variables. Both centers are using unified model (UM), definitely with large differences in implementation level.