Simulation of Standing Waves Using ROMS

General scientific issues regarding ROMS

Moderators: arango, robertson

Post Reply
Message
Author
yong.liang
Posts: 6
Joined: Wed Apr 08, 2015 3:16 pm
Location: University of California Berkeley

Simulation of Standing Waves Using ROMS

#1 Post by yong.liang » Thu Apr 09, 2015 12:06 am

Hi Everyone,

I am trying to simulate standing waves using ROMS in a linearly stratified fluid. The initial conditions I use is free surface elevation and a corresponding nonzero initial velocity field everywhere in the domain(to excite a specific mode of internal waves). The boundaries of the rectangular domain are all walls except the free surface. Can anyone tell me if I can realize this in ROMS? Thanks.

User avatar
kate
Posts: 3770
Joined: Wed Jul 02, 2003 5:29 pm
Location: IMS/UAF, USA

Re: Simulation of Standing Waves using ROMS

#2 Post by kate » Thu Apr 09, 2015 12:23 am

I don't see why not. Do you have a case with an analytic solution? You could use the solution to get the initial conditions which are consistent for free surface and velocity. If you turn off rotation, you might be able to find something with zero initial velocity, just set up the zeta and initial density field appropriately. I would start with a 2-D case, no stratification, unless you want to set up internal waves.

yong.liang
Posts: 6
Joined: Wed Apr 08, 2015 3:16 pm
Location: University of California Berkeley

Re: Simulation of Standing Waves using ROMS

#3 Post by yong.liang » Wed Apr 22, 2015 6:21 pm

I am able to simulate my case with ROMS now. But I have not get the correct numerical results that can match the linear analytic solutions. So I started from a simple case: homogeneous fluid(no stratification), rectangular domain(all wall-type lateral boundaries), the initial condition is a free surface elevation only(standing wave amplitude). I used the 3D solver since the 2D version only solves shallow water equation. Then I examined the output but only got good results at the first time step. From the second step, zeta becomes zero all the time. Could anyone help look at what the problems might be? Thanks.



kate wrote:I don't see why not. Do you have a case with an analytic solution? You could use the solution to get the initial conditions which are consistent for free surface and velocity. If you turn off rotation, you might be able to find something with zero initial velocity, just set up the zeta and initial density field appropriately. I would start with a 2-D case, no stratification, unless you want to set up internal waves.
Attachments
ocean_internal.in
(116.85 KiB) Downloaded 82 times
internal.h
(723 Bytes) Downloaded 81 times
ana_initial.h
(32.2 KiB) Downloaded 83 times
ana_grid.h
(34.73 KiB) Downloaded 83 times

yong.liang
Posts: 6
Joined: Wed Apr 08, 2015 3:16 pm
Location: University of California Berkeley

Re: Simulation of Standing Waves using ROMS

#4 Post by yong.liang » Thu Apr 23, 2015 1:27 am

I just noticed that Rutgers ROMS is a not a non-hydrostatic model. This does not work for my cases.

User avatar
kate
Posts: 3770
Joined: Wed Jul 02, 2003 5:29 pm
Location: IMS/UAF, USA

Re: Simulation of Standing Waves Using ROMS

#5 Post by kate » Thu Apr 23, 2015 4:59 am

The French have a team working hard to make ROMS_Agrif have a non-hydrostatic option. Coming soon?

User avatar
patrickm
Posts: 24
Joined: Fri Apr 30, 2004 6:41 pm
Location: IRD, FRANCE
Contact:

Re: Simulation of Standing Waves Using ROMS

#6 Post by patrickm » Sun May 17, 2015 10:15 am

yes Kate, we have done the TANK test case in the homogeneous and two-layered fluid (internal soliton) using a non-hydrostatic version of ROMS_AGRIF. We can accurately reproduce both the hydrostatic and non-hydrostatic analytical solutions in the homogeneous case and lab experiments in the two-layer case. There is more work needed before we want to release this code (optimization, realistic cases capability ...), but that will not be too long, hopefully before the end of the year ...

Post Reply