ocean simulation
 

Hi,

I am trying to simulate a square-shaped ocean of constant depth in cartesian coordinate. I would like to include the coriolis force in it. However, it seems that the coriolis force is a built-in source only for polar-cylindrical coordinate. Anyway to work around it?

Thanks for your reply in advance.

Wenqing

Re: ocean simulation
 

You will need to introduce the coriolis forces by means of source terms in the momentum equations. This can be done by GROUND or PLANT coding. Go to CHAM's website for general information on how to introduce sources into the finite-volume equations:

http://www.cham.co.uk/phoenics/d_pol...eral/bcond.htm

PLANT is a method for writing GROUND coding. It is described here, and there may be releavnt examples:

http://www.cham.co.uk/phoenics/d_pol...lant/plan0.htm

Professor Urban Svensson did a lot of work on PHOENICS for Geophysical flows, for example:

Gidhagen L, Nyberg L & Svensson U A model system for marine circulation studies. part a: basic principals and comparisons with analytical solutions & part b: applications to the Baltic system Computer Aided Fluid Engineering Sweden. Presented at the 2nd International PHOENICS User Conference, November 1987. Published in Proceedings, CHAM

Look in the publications list on CHAM's website there might be something relevant:

http://www.cham.co.uk/website/new/support/publish.htm

Any PHOENICS Journal papers usually contain the Q1 and GROUND coding.

Re: ocean simulation
 

Dear Dr. Malin, Thank you very much! It works!

I have another problem, and I will be really thankful if you could give me some advice on this.

If I assume the ocean is horizontally homogeneous (but the two horizontal velocities are different due to diffent forces), and want to simulate how the two velocities change with the depth of the sea (thus a one-dimensional problem), can I use the single-slab solver to solve this problem? it seems PHOENICS does not allow me to solve (u1, w1) with only one-dimensional grid.

Best wishes.

Wenqing

Re: ocean simulation
 

I don't really understand the physical situation. Is the pressure gradient zero in the two horizontal directions? The single slab solver works by deactivating the appropriate pressure term (via the TERMS command)and the appropriate convection and diffusion terms (via the TERMS command or GP12 PATCH) in the momentum equations. If W1 varies with depth then we must deactivate the built-in pressure term. and then supply an integrated pressure force via a source patch unless it is a couette flow where the pressure gradient is zero anyway. I think it best that you read the POLIS entry on fully-developed flows and look at some of the single-slab examples in the flow library of q1 examples.