We are working on the EMC FV3 member that is in the High Res window slot.
The grid is a cut out so there is missing/undefined values on the edges of the domain.
the problem is that the missing value when read in via pygrib is 9999. The 250hpa geopotential has data that range 9700-10800 or so, which overlaps with the missing value, resulting in missing grid points not on the domain edge.
we realized this and ran wgrib2 to change the missing value:
wgrib2 hiresw.t00z.fv3_5km.f01.conus.grib2 -rpn "-999:swap:merge" -grib_out OUT.grib2
and this got us a clean geopotential height field.
Is this a pygrib issue or grib2 issue or both?
We are working on the EMC FV3 member that is in the High Res window slot. The grid is a cut out so there is missing/undefined values on the edges of the domain. the problem is that the missing value when read in via pygrib is 9999. The 250hpa geopotential has data that range 9700-10800 or so, which overlaps with the missing value, resulting in missing grid points not on the domain edge. we realized this and ran wgrib2 to change the missing value: wgrib2 hiresw.t00z.fv3_5km.f01.conus.grib2 -rpn "-999:swap:merge" -grib_out OUT.grib2
and this got us a clean geopotential height field. Is this a pygrib issue or grib2 issue or both?
Thanks jimmyc