Problem with ssh response to uniform winds.

[maxblu33]

Hi ROMS Community,

I am trying to set up an experiment to study upwelling/downwelling response to constant wind on an ideal continental shelf. The experiment is force with uniform surface stress of 0.1 N/m^2 in the alongshore direction, for a homogeneous fluid with constant vertical mixing. The bathymetry is constant in the alongshore direction, in the cross shelf direction a constant slope of 0.001 is used with minimum depth of 2m and maximum of 100 m. The experiment is run for 5 days.

The problem I found is that the response of the ssh to upwelling/downwelling favorable winds is not uniform along the coast, but instead is stronger closer to the northern boundary, so I suspect I have some issues with the boundary condition but I haven't been able to figure out what's wrong.

I try two different configurations for the open boundary. In every case the coastal boundary is closed, the west, south and north boundaries are open. For the first one, I try Chapman for the free surface and Radiation for all other variables. The second one is Chapman/Flatther combination for zeta/ubar-vbar with zero velocity and zero SSH but the problem was not fixed. I attached the header and roms.in files and the link to the model zeta variable for the last step of the run.

ssh:
https://drive.google.com/file/d/1GKf4BP ... sp=sharing

Any ideas on what could I do?
thanks in advance!

Luis Sorinas-Morales

[jcwarner]

is there a reason you dont use NS periodic? also there is an upwelling test case distributed with the code that you might want to try first, just to get an idea of what that case can do.
-j

[maxblu33]

Hi Professor Warner,

Thanks for the reply!

I did try periodic OBC for the NS and closed EW and that actually fix the problem. However, the experiment we had in mind was to show the formation of an inner shelf, and surface and bottom ekman layers, and I was expecting at some point far enough offshore that the surface and bottom ekman transport would balance and therefore ssh would not change offshore of that area, and that was not possible using a channel in the way I set it up (I attached a ssh picture).

Channel:
https://drive.google.com/file/d/11ekZ8S ... sp=sharing

So I guess my question is can I set the model to reproduce surface and bottom Ekman Layers with periodic conditions on the N and S boundary and OB in the west coast? Does that make sense physically?


I play a little bit with the upwelling test case. I kept everything as in the test case but modify the EW periodic conditions and use instead Chapman/Radiation and pretty much got to the same behavior that with my ideal coastal shelf experiment.

upwelling test case
https://drive.google.com/file/d/15wYQUv ... sp=sharing

[kate]

If you want a solution that is invariant along-channel, making it periodic that way is certainly the way to go. You can save some cycles by making the channel quite short, say only 3-4 points in that direction. Did you try a wall at the shallow side and open offshore?

[maxblu33]

Thanks Kate!

So, if it is not periodic there will always gonna be some along coast change, even for uniform winds? If that's the case then I do have to go with the periodic option.

I haven't try a wall in the shalow side and open in the deep side, I'll short the number of points in the NS direction and give it a try.

Luis