Seamount modifications

[bperfect]

Hello,

My goal here is to modify the seamount test case to have a mean flow going through the domain (i.e., to resolve the seamount wake). I'm not quite sure how to accomplish this. My research into similar issues leads me to believe that this is actually a much more complicated procedure than I initially thought.

(This thread viewtopic.php?f=24&t=3562 has been a primary source of information, but is certainly incomplete for my level of knowledge). It seems like I have to

1. Modify the ocean*.in file to have different boundary conditions (Cla Clo Rad Clo)

2. Edit the ANA_M3OBS file to have a defined inlet condition, say, 1m/s in u and 0m/s in v, in the plane on intended inlet, corresponding to the Cla boundary

3. Perform something similar in ana_m2obs.h. As a side note, the 2D versions of everything seem to just be solving equations that have been averaged over depth. Is this accurate?

3. Edit the ana_initial.h file so that the domain starts with the proper velocity vector inside the domain (should I use 1m/s in U, and 0 in V and W?)

4. Define both ANA_M3OBS and ANA_M2OBS in seamount.h to have the files in (2) and (3) parsed

5. Make similar boundary condition changes for LBC(isUbar). Do isUbar and isUvel need to have the same boundary conditions specified in general?

Is there anything else? Is there some way of looking up which options I need to define in my "my_application".h file, and what changes need to be made for each one once I enable it?

[kate]

Hello,

My goal here is to modify the seamount test case to have a mean flow going through the domain (i.e., to resolve the seamount wake). I'm not quite sure how to accomplish this. My research into similar issues leads me to believe that this is actually a much more complicated procedure than I initially thought.

(This thread viewtopic.php?f=24&t=3562 has been a primary source of information, but is certainly incomplete for my level of knowledge). It seems like I have to

1. Modify the ocean*.in file to have different boundary conditions (Cla Clo Rad Clo)

Yes, or RadNud instead of Rad.

2. Edit the ANA_M3OBS file to have a defined inlet condition, say, 1m/s in u and 0m/s in v, in the plane on intended inlet, corresponding to the Cla boundary

Yes, perhaps at the outlet too if you use RadNud.

3. Perform something similar in ana_m2obs.h. As a side note, the 2D versions of everything seem to just be solving equations that have been averaged over depth. Is this accurate?

Yes to all.

3. Edit the ana_initial.h file so that the domain starts with the proper velocity vector inside the domain (should I use 1m/s in U, and 0 in V and W?)

Sure, but you don't need to (can't) set W.

4. Define both ANA_M3OBS and ANA_M2OBS in seamount.h to have the files in (2) and (3) parsed

Yes.

5. Make similar boundary condition changes for LBC(isUbar). Do isUbar and isUvel need to have the same boundary conditions specified in general?

On Ubar, one typically uses Fla or Shc instead of Rad(Nud). Clamped is the same.

Is there anything else? Is there some way of looking up which options I need to define in my "my_application".h file, and what changes need to be made for each one once I enable it?

The most complete document is the manual. Let me know if you have suggestions for it.

[bperfect]

Thanks! For whatever reason I only just saw this post now. I managed to independently find the manual, but it wasn't trivial. I decided that the best thing was to just sit down and read it. It's been an incredible resource. Is there a way of having that somewhere really obvious on the wiki, or even included with the ROMS download? I think encouraging people to read it before monkeying around with anything other than just running the default test cases would go a long way.

I'm currently in the process of adopting your recommended directory structure for my modified case. Getting some errors of the form "missing '(' after predicate" and "/bin/sh: 1: -c: not found". I'm hoping that I'll discover it's something silly rather than a more sinister installation problem

[kate]

I think the second of your problems is in the FAQ on the wiki.

The ROMS manual has a link from the tutorials page.