Wave absorption boundary condition

[hpd14thu]

Hi all,

I used OpenFOAM recently and ran a case with wave absorption boundary condition.

When I turn to ROMS, I found that they were no wave absorption boundary condition. I don't have open boundary data, and I only want to absorb wave of open boudary and avoid reflection.

I tried cha for 2dzeta, fla for 2duv, and gra for 3d, but it worked bad. After several hours, lots of reflection happened near open boundary and model blowed up.

Look forward to your suggestion.

Thanks

Peida

[kate]

This is a classic problem with regional ocean models. You can try the radiation boundary conditions instead, if you like. However, what is working for us in practice is using values from a global ocean model on the boundaries to keep things behaving better (if not entirely well).

[hpd14thu]

This is a classic problem with regional ocean models. You can try the radiation boundary conditions instead, if you like. However, what is working for us in practice is using values from a global ocean model on the boundaries to keep things behaving better (if not entirely well).

Thanks Kate,

I wonder that if there are some general rules when using Rad? For example, when I use Rad, if I need to set all of 2d-uv, 2d-zeta and 3d-uv as Rad || or Fla for 2d-uv, Cha for 2d-zeta and Rad for 3d-uv.

Cause I barely remember that I seemed to use Rad long long ago, and I made a bad conbination of different condition and failed to run model. (When I use Rad, I don't need to input any varibales or define RADIATION_2D?)

Peida

[kate]

If your barotropic solution is working for you, don't change that.

Using Rad alone will not require external fields. You may or may not do better with RADIATION_2D. Just try it and see if it helps.

If you want more specific advice, you should probably tell us more about what you are trying to do.

[hpd14thu]

If your barotropic solution is working for you, don't change that.

Using Rad alone will not require external fields. You may or may not do better with RADIATION_2D. Just try it and see if it helps.

If you want more specific advice, you should probably tell us more about what you are trying to do.

Thanks and I'm sorry I didn't make myself clear，

What I plan to do : I try to calculate zeta and 3duvw in a semi-closed coastal area, which north boudanry open, south/east/west boudanry closed.

Forcing : forcing condition is large wind, may result in large zeta and uvw. So maybe Rad is suitable for me to deal with the open boudanry.

Boundary : My present boundary conditon (I wonder if it's Ok) and results of zeta near north boundary (reflection) (run time=60s)shows below (PS : I set ana_fsobc and ana_m2obc as zero):

Now the issue is reflection appeared near the north boundary.

Peida
[/quote]

[zhangtianhao]

Maybe you should change tke boundary to gra

Hoty

2020-01-17

[hpd14thu]

Thank you, Hoty:

After some trials, I found I could expand the area to decrease the influence of boundary reflection.

And I use: