ufs-community / UFS_UTILS

Utilities for the NCEP models.
Other
21 stars 104 forks source link

chgres_cube - vertical velocity bug #626

Closed GeorgeGayno-NOAA closed 2 years ago

GeorgeGayno-NOAA commented 2 years ago

When using certain GRIB2 data as input, the vertical velocity must be converted from omega to dzdt. This conversion is controlled by the logical 'conv_omega'. However, that logical is not being set on all MPI tasks. As a result, vertical velocity is only being converted on part of the grid.

GeorgeGayno-NOAA commented 2 years ago

This is a problem in the following regression tests -

The GFS GRIB2 data does not need to be converted.

GeorgeGayno-NOAA commented 2 years ago

Ran the consistency tests on WCOSS-Dell. Three tests failed:

consistency.log02:<<< 3-km CONUS HRRR W/ GFS PHYSICS GRIB2 TEST FAILED. >>>
consistency.log03:<<< 3-km CONUS HRRR W/ GSD PHYSICS AND SFC FROM FILE GRIB2 TEST FAILED. >>>
consistency.log05:<<< 13-km CONUS RAP W/ GSD PHYSICS AND SFC FROM FILE GRIB2 TEST FAILED. >>>
GeorgeGayno-NOAA commented 2 years ago

Files from the failed tests are on Mars: /gpfs/dell1/stmp/George.Gayno/reg-tests/chgres-cube.

The difference between and test and baseline file is 'atm.diff.nc'.