[Ncep.list.emc.fv3gfs_tickets] FV3GFS Ticket #19: Modify CHGRES to output boundary conditions for stand-alone nests

FV3GFS Trac Ticket ncep.list.emc.fv3gfs_tickets at noaa.gov
Tue Oct 24 19:28:00 UTC 2017

#19: Modify CHGRES to output boundary conditions for stand-alone nests
  Reporter:  george.gayno@…  |      Owner:  george.gayno@…
      Type:  task            |     Status:  accepted
  Priority:  major           |  Milestone:
 Component:  component1      |    Version:
Resolution:                  |   Keywords:

Comment (by george.gayno@…):


 Add Jim's updates to
 fv3gfs_driver_grid.sh] for the stand-alone regional case to output two
 sets of "grid" and "orog" files - one set with a halo of 3 and one with a
 halo of 4.  Jim's reasoning:
 "To remap the winds we need ps in an extra halo row. So we create the
 grid and orog files with halo of 4 and the the boundary files with halo=4.
 Finally I shave the grid and orog files back to a halo of 3".

 The regional case was run on Cray and produced both sets of files:

 $ ls -l
 total 26112
 -rw-r--r-- 1 George.Gayno global 9628707 Oct 24 17:04
 -rw-r--r-- 1 George.Gayno global 9815395 Oct 24 17:04
 -rw-r--r-- 1 George.Gayno global    8485 Oct 24 17:04 C96_mosaic.nc
 -rw-r--r-- 1 George.Gayno global 3438565 Oct 24 17:04
 -rw-r--r-- 1 George.Gayno global 3505141 Oct 24 17:04

 Jim's change did not affect results for the C96 'uniform', 'stretch' and
 'nest' cases.  On Cray, bit identical results were obtained against the
 trunk (at r98608).

 Also, correct shell initiation line in
 build_shave.sh], which causes the script to fail on Cray.

Ticket URL: <https://svnemc.ncep.noaa.gov/trac/fv3gfs/ticket/19#comment:13>
fv3gfs <https://svnemc.ncep.noaa.gov/trac/fv3gfs>
NGGPS FV3GFS Development

More information about the Ncep.list.emc.fv3gfs_tickets mailing list