Problem restarting ROMS in COAWST (WRF-ROMS coupling)

[nicholls]

Hello everyone,
I am a COAWST model user running a WRF-ROMS coupling configuration. I am currently testing whether I can start, stop, and restart the COAWST model. Specifically I am running two six hours segments. The first starts at model initialization, and the second from a restart time six hours later (via an ocean_rst.nc file). The first segment runs perfectly, but in the latter, ROMS successfully reads the restart file and I changed NRREC in roms.namelist.in from 0 to -1, but ROMS fails because it tries to find a file ocean_avg_0000.nc. I hope this is a newbie question, but I am confused as to why ROMS would look for this file to restart itself. From the ROMS users' guide, my understanding was that all I needed to do was to change ININAME to ocean_rst.nc, change NRREC to -1, and increase NTIMES to the end time of the next period (i.e., 4120, I am running a time step of 10 seconds), so I am at a loss to find an explanation as to why ROMS would look for an ocean_avg file. Even when I renamed my ocean_avg_0001 file to ocean_avg_0000, the ROMS read the ocean_avg file but then failed there after because ROMS then said it was also missing a ocean_dia_0000.nc file. Considering that in my roms.namelist.in file I set NDEFAVG = 2160 (or six hours), it does not make sense to me why ROMS would look for ocean_avg_0000, when I do correctly generate a file ocean_avg_0001.nc. For reference I am using COAWST rev 727 and ROMS version 455 and I have attached my error and ROMS configuration files to this message.
Thanks for your time and consideration of this post.
--Stephen Nicholls



ROMS error output (subsection containing first error)
**********

Maximum grid stiffness ratios: rx0 = 8.938246E-01 (Beckmann and Haidvogel)
rx1 = 1.373953E+01 (Haney)


Initial basin volumes: TotVolume = 2.3137323492E+17 m3
MinVolume = 1.5508442673E+07 m3
MaxVolume = 1.6246901264E+11 m3
Max/Min = 1.0476165535E+04

ROMS recv data from WRF
ROMS sent data to WRF


NL ROMS/TOMS: started time-stepping: (Grid: 01 TimeSteps: 00000001 - 00017286)

GRID STEP Day HH:MM:SS KINETIC_ENRG POTEN_ENRG TOTAL_ENRG NET_VOLUME

1 0 1 00:00:00 5.227080E-03 2.075019E+04 2.075020E+04 2.314783E+17
DEF_AVG - inquiring average file: ocean_avg_0000.nc

NETCDF_OPEN - unable to open existing NetCDF file:
ocean_avg_0000.nc
call from: def_avg.F

ROMS/TOMS - Output error ............ exit_flag: 3


Elapsed CPU time (seconds):

Node # 0 CPU: 24.722
Node # 1 CPU: 28.402

[jcwarner]

i see your post on the coawst site, so i will answer it there. think this might just be a ROMS issue with restarting.
-john

[nicholls]

Following your suggestion from John Warner, I recompiled COAWST with the changes seen below to my .h file. With these
changes, ROMS no long demands the ocean_avg_0000.nc or ocean_dia_0000.nc. Without these demands, I managed to get COAWST to restart successfully. That said, this problem has been resolved. Thanks John for the help!
--Stephen

old:
#define AVERAGES
#define DIAGNOSTICS_UV
#define DIAGNOSTICS_TS

new:
#undef AVERAGES
#undef DIAGNOSTICS_UV
#undef DIAGNOSTICS_TS

[balaji]

Hi everyone....

I'm new to this forum..
I have written mail to JC Warner regarding the address to download the Model COAWST..
I have no idea that whether my mail have reached him...
Please help me on this issue..
How to get access to download COAWST model ...

Thanks in advance