Questions about the outcome of the test case riverplume2

[fengfengsister]

Hi, all
The distributions of salinity in 3 layer for the riverplume2 test case (https://www.myroms.org/wiki/index.php/Test_Cases) are shown in the attachments. Are there something wrong with the test case?
Who can help me? Thank you!

[ckharris]

Hi
I recently used riverplume2 for a test case in a class. And, yes, the salinity distribution evolves to look like that. I think to improve it you need to change the downstream open boundary condition for salinity.

Best wishes,
Courtney Harris

[fengfengsister]

Hi
I recently used riverplume2 for a test case in a class. And, yes, the salinity distribution evolves to look like that. I think to improve it you need to change the downstream open boundary condition for salinity.

Best wishes,
Courtney Harris

Thank you very much!Could you please tell me how to change the downstream open boundary condition for salinity?

[ckharris]

Hi
I used riverplume2 to teach a class and we added sediment transport to the case. We did not change the downstream boundary condition.

But this is how the case works: the grid has two open boundaries that are aligned across the shelf, which are upstream and downstream on a continental shelf. The “land” boundary is closed, and the “offshore” boundary is also closed. The upstream and downstream (across-shelf) boundaries are set up to actually be point sources of discharge – so they are treated as if they are rivers. The input files have to specify the discharge at these points, in the same way that they specify the discharge of the rivers – this may be done in the analytical file (ana_psource.h?). But the point source code may also require that the salinity be specified – and specifies it to be the “shelf water”, but since there is a river that puts freshwater into the grid, that results in all of the freshwater remaining trapped on the shelf.

To change this, I think you might change the downstream open boundaries to be more conventional open boundaries, instead of having them be these “river” sources. You could keep the upstream open boundaries as the point sources.

We did not try to do that in my class – it was enough work to add the sediment transport.

Best wishes,
Courtney Harris