Fix wave restarts and GEFS FHOUT/FHMAX #3009
Conversation
|
ATTN: @weihuang-jedi |
ush/forecast_postdet.sh
Outdated
| @@ -340,7 +340,7 @@ WW3_postdet() { | |||
| echo "Copying WW3 restarts for 'RUN=${RUN}' at '${restart_date}' from '${restart_dir}'" | |||
| local ww3_restart_file | |||
| for ww3_grid in ${waveGRD} ; do | |||
| ww3_restart_file="${restart_dir}/${restart_date:0:8}.${restart_date:8:2}0000.restart.${ww3_grid}" | |||
| ww3_restart_file="${restart_dir}/${restart_date:0:8}.${restart_date:8:2}0000.restart.ww3" | |||
There was a problem hiding this comment.
This line does not have a variables ww3_grid, so there is no need to loop
ush/forecast_postdet.sh
Outdated
| for (( vdate = "${restart_date:-${PDY}${cyc}}"; vdate <= forecast_end_cycle; | ||
| vdate = $(date --utc -d "${vdate:0:8} ${vdate:8:2} + ${restart_interval} hours" +%Y%m%d%H) )); do | ||
| restart_file="${vdate:0:8}.${vdate:8:2}0000.restart.ww3" | ||
| ${NLN} "${DATArestart}/WW3_RESTART/${restart_file}" "${restart_file}" | ||
| done |
There was a problem hiding this comment.
I think we should split this into 2 parts.
part 1 should be linking the restart for initialization. This is either the restarts at restart_date (if found above) or PDYcyc.
If it is PDYcyc, the restart_dir is COMIN_WAVE_RESTART. If it is a RERUN=YES, it is DATArestart/WW3_RESTART. This restart must exist.
part 2 is creating the links from restart_date+restart_interval to forecast_end_cycle
These are the restarts that will be created as the model integrates.
There was a problem hiding this comment.
There's already a separate part to copy the initial restart in, if necessary. Just need to adjust the start date for the links when that happens.
Fixes some issues that were preventing wave restarts from operating
correctly.
First, the wave restart files were not being correctly linked from
`$DATA` to `$DATArestart`. The files are placed in the root of
`$DATA` instead of in `${DATA}/WAVE_RESTART`, so now links for the
individual files are created.
Second, the incorrect filenames were being searches for and copied
as part of a rerun. Filenames were geared towards multigrid waves,
which use the grid names, but single grid just uses a `ww3`. Since
multigrid waves are deprecated in workflow and will soon be removed
(NOAA-EMC#2637), these were updated only supporting the single-grid option.
These fixes allow forecast segments (and emergency restarts) to
work correctly when waves are on.
Resolves NOAA-EMC#3001
Removes an unnecessary loop when staging wave restarts and stops creating output links for restart files that were already copied in as ICs. Also updates FHMIN_WAV when running with segments to stop creating unneeded links from earlier in the forecast period.
WalterKolczynski-NOAA
force-pushed
the
hotfix/segs
branch
from
a11e004
to
6bbaa2b
Compare
|
Will need to add in additional fixes for GEFS perturbed members, as they are ignoring the segments. |
The gefs `config.efcs` still contained code from the enkf version of the file that was overwriting the FHMAX set for segments.
WalterKolczynski-NOAA
changed the title
|
I have test the PR code on Hercules. I worked as expected. Fixed the segments issue. |
Description
Fixes some issues that were preventing wave restarts from operating correctly.
First, the wave restart files were not being correctly linked from
$DATAto$DATArestart. The files are placed in the root of$DATAinstead of in${DATA}/WAVE_RESTART, so now links for the individual files are created.Second, the incorrect filenames were being searches for and copied as part of a rerun. Filenames were geared towards multigrid waves, which use the grid names, but single grid just uses a
ww3. Since multigrid waves are deprecated in workflow and will soon be removed (#2637), these were updated only supporting the single-grid option.These fixes allow forecast segments (and emergency restarts) to work correctly when waves are on.
Additionally, the
FHMAX(andFHOUT) for perturbed GEFS members was being overwritten byconfig.efcsdue to remnant code from when it was copied from the EnKF version. This interfered with the segment setting for those GEFS members.Resolves #3001
Type of change
Change characteristics
How has this been tested?
Checklist