Test cases

[drewa]

Finally, I have compiled both versions of ROMS (ROMS and ROMS_AGRIF) and am starting to play around with the test cases. I've started with the 'upwelling' test case for the ROMS version and have a simple question:
Where are the NetCDF input files (ocean_grd.nc, ocean_ini.nc, ocean_itl.nc...)?

The ROMS_AGRIF version is interesting to me,for one reason, because the ROMS_GUI interface seems pretty slick. I can't get it to run though. I have gone through make_gid, make_forcing, make_clim, and make_tides. When I go to run the model, this is my output that I'm having a bit of trouble wading though:

***************************************************
Roms_tools/Run$ ./roms roms.in

Southern Benguela
480 ntimes Total number of timesteps for 3D equations.
5400.00 dt Timestep [sec] for 3D equations
60 ndtfast Number of 2D timesteps within each 3D step.
1 ninfo Number of timesteps between runtime diagnostics.

6.000E+00 theta_s S-coordinate surface control parameter.
0.000E+00 theta_b S-coordinate bottom control parameter.
1.000E+01 Tcline S-coordinate surface/bottom layer width used in
vertical coordinate stretching, meters.
Grid File: ROMS_FILES/roms_grd.nc
Forcing Data File: ROMS_FILES/roms_frc.nc

WARNING: Unrecognized keyword: bulk_forcing --> DISREGARDED.

Climatology File: ROMS_FILES/roms_clm.nc

WARNING: Unrecognized keyword: boundary --> DISREGARDED.

Initial State File: ROMS_FILES/roms_ini.nc Record: 1
Restart File: ROMS_FILES/roms_rst.nc nrst = 480 rec/file: -1
History File: ROMS_FILES/roms_his.nc Create new: T nwrt = 480 rec/file = 0
1 ntsavg Starting timestep for the accumulation of output
time-averaged data.
48 navg Number of timesteps between writing of time-averaged
data into averages file.
Averages File: ROMS_FILES/roms_avg.nc rec/file = 0

Fields to be saved in history file: (T/F)
T write zeta free-surface.
F write UBAR 2D U-momentum component.
F write VBAR 2D V-momentum component.
F write U 3D U-momentum component.
F write V 3D V-momentum component.
F write T(1) Tracer of index 1.
F write T(2) Tracer of index 2.

READ_INP ERROR while reading block with keyword 'auxiliary_history_fields'.


READ_INP ERROR: A total of 1 configuration errors discovered.
***************************************************

Any suggestions on where the error is?

Thanks (again).

[kate]

I can answer this at least:

Finally, I have compiled both versions of ROMS (ROMS and ROMS_AGRIF) and am starting to play around with the test cases. I've started with the 'upwelling' test case for the ROMS version and have a simple question:
Where are the NetCDF input files (ocean_grd.nc, ocean_ini.nc, ocean_itl.nc...)?

The Rutgers ROMS has the ANA_INITIAL type switches. For any field on which the ANA_ gets defined, ROMS will use the corresponding ana_ routine to load up the fields. Take the grid, for instance. If you are doing a square box problem, you can make a square box grid file or you can create it internally to ROMS inside ana_grid. UPWELLING needs no external files since everything is done through the ana_ routines. These can be in ROMS/Functionals, User/Functionals, or anywhere else you care to put them, as long as you let the makefile know.

[drewa]

Thanks again Kate. That really clears things up for me.

[drews]

I'm still puzzling over this. When I run the windbasin example I get:

READ_PHYPAR - could not find input file: ocean_ini.nc

Kate, are you saying that the build.bash script for my windbasin has to specify Functionals/ana_initial.h?

Carl

[kate]

That's odd. The windbasin.h file should have ANA_INITIAL #defined so that it won't look for a separate file.

[drews]

Okay, here's what fixes the READ_PHYPAR problem for me. Comment out the definition of NPZD_POWELL on line 114 of build.bash, like this:

#export MY_CPP_FLAGS="-DNPZD_POWELL"

Now the initial conditions are all analytical, like Kate said they should be.

Curiously enough, NPZD_POWELL is not defined in the corresponding build.sh file. But it is defined in r139 of build.bash.