Using tidal harmonics to impose elevation

[jmjaneiro]

Dear all,

I'm trying to run a real ROMS application that for now is just forced by the tide.
In the beginning I was using a elevation time series in the boundary and everything worked fine. I'm now trying to use tidal harmonics to do the same, but without luck. I understand that the tidal forcing in ROMS is super-imposed on the climatological sea surface elevation. I've created the harmonics netcdf file as well as a test climatological file using a constant mean sea level for my entire domain and took the elevation time series out of the boundary. The model doesn't run saying that it needs the zeta_south (in this case) in the boundary file. This is odd to me. Can't we force the boundary with the only with tidal harmonics?
Can you give me some tips on what I need to do to use tide harmonics to force my model.
Here are the options I'm using:

/* Options associated with momentum equations */

#define UV_ADV
#define UV_COR
#define UV_QDRAG

/* Options associated with tracers equations */

#undef SALINITY

/* Model configuration options */

#undef SOLVE3D
#define MASKING
#define STATIONS

/* Options for analytical fields configuration */

#define ANA_SMFLUX

/* Options for reading and processing of climatological fields */

#define ZCLIMATOLOGY

/* Southern edge open boundary condtions options */

#define SOUTH_FSCHAPMAN
#define SOUTH_M2FLATHER
#undef SOUTH_TRADIATION

/* Options for tidal forcing at open boundaries*/

#define SSH_TIDES

Thanks in advance

Joao

[kate]

This is a case of ROMS being too tricky for its own good. You will have to go into globaldefs.h and tell it not to read the boundaries. Here is what I have in mine:

Code: Select all

#if defined UV_TIDES && defined SSH_TIDES && (defined BERING ||\
    defined COOK_INLET) && !defined SOLVE3D
# undef OBC_DATA
# undef OBC
#endif

You may decide to use a simpler test:

Code: Select all

#if defined JOAO_TIDES
# undef OBC_DATA
# undef OBC
#endif

[jmjaneiro]

Hi Kate,

Thanks for your reply.
The model now runs with the tidal harmonics, but the results aren't great.The next figure show the elevation at the boundary in the first 11 hours of the simulation:

Picture 2.png (17.82 KiB) Viewed 1971 times

This one shows the velocity, in the same station, but for a 4 day period:

Picture 1.png (17.03 KiB) Viewed 1969 times

The model blow up after the 4th day, but I'm not sure why it does that.
I've check my harmonics netcdf file and it appears to be ok. I created it based on the CDL files available. I also noticed that the model converts the phase direction from degrees to rads and the period from seconds to hours in. I've changed that in varinfo.dat to suit my units.
Is there anything else I should have in mind that I've might had missed?
Do I need more than the mean sea level in my climatological forcing file to make it work?
Could this error be because I'm only imposing the tidal elevation and not adding the tidal velocity?
I would be very grateful if you could give me any advice on this please.

Thanks in advance.

Joao