Suggestions

[arango]

Use this thread for any suggestions that will enhance the ROMS/TOMS user manual.

[tim_cera]

It appears that characteristics of the grid have been added to the "Ice" section. Maybe a "Grid" heading is missing?

[kate]

Would this be for the fortran grid stuff or for Seagrid? Both? How to excavate your fjords with NCL?

Then you need all the tools for preparing winds, tides, boundary conditions, etc., all in a 32-bit address space.

[tim_cera]

The Ice section in the Beta Documentation ( https://www.myroms.org/documentation/docs.php ) lists grid characteristics.

This:

Code: Select all

Ice

  Hibler viscous-plastic dynamics. 
  Mellor-Kantha thermodynamics. 
  Orthogonal-curvilinear coordinates. 
  Arakawa C grid. 
  Smolarkiewicz or third-order upwind advection of tracers. 
  Optional ridging scheme. 

Should be something like this:

Code: Select all

Ice

  Hibler viscous-plastic dynamics. 
  Mellor-Kantha thermodynamics. 

Grid
  Orthogonal-curvilinear coordinates. 
  Arakawa C grid. 

Tracers
  Smolarkiewicz or third-order upwind advection of tracers. 

Whatever This Is
  Optional ridging scheme. 

This would be alot easier if we could either make direct changes to the documentation (for example a WikiWiki), or make embedded comments in the documentation (see my earlier comment at viewtopic.php?t=107 )

[kate]

Ah, I get it. The point is that the original Hibler model was on an Arakawa B grid and was not curvilinear. We did couple an ocean model to it at one point. Adding the curvilinear terms to the ice model was not a trivial exercise. My guess is that our current ice model isn't curvilinear either.

Likewise, the advection scheme in the coupled model described in the draft manual was different for the ice and the ocean. The ice used MPDATA while the ocean did not.

Some ice models include ice ridging - that's what it is referring to.

[wmartin]

All the examples that come witH ROMS could be significantly more useful if the people who set them up would write a short note on why they chose the options that they did. For example, why choose one advection scheme over another for an estuary, or why use certain values of AkV_BAC. Why use SPLINES.

Get all the modelers with examples to do two pages on why the did what they did, plus whaterver else they think would be useful, like, "in this situation I would not use such and such because so and so". That seems pretty easy and would be hugely helpful. In a similar vein it would be great if people who write papers using roms, especially for process studies, for example for different closures etc. would post their actual setup files somewhere. I've tried to reproduce some of these results but I never get the details quite right.

Thanks Wayne

[kate]

ROMS at least was such a moving target that this is easier said than done. It would take someone committing to a setup and making sure it worked with each new release. One of my toy problems is currently broken, but not a high enough priority for me to fix it. The available options keep changing, as does the input file. Still, it's a worthy goal. We'll get Hernan's clone to start it off.

[cmoffat]

Hi,

I am a really new user, and I've been trying to figure out the basics of ROMS use, compilation, etc.

Apart from having a User Manual, I'd like to suggest creating a 'ROMS Wiki'. People could contribute to it with all kinds of tips, test cases and other things, without adding overhead to the developers. The Forum would still be a great place to ask specific questions, but when problems are solved, they could be summarized/added to the Wiki. This allows to 'build' documentation iteratively.

I would be happy to help with this, as I've done wiki-like projects before.

Cheers,
Carlos

[andres]

Quite a good idea. My students are building a notes on the class I teach using wiki and it is sometimes surprising the quality and dedication they put in the contributions.

The point is new/recent/intermediate users can learn from trying to explain how tu use ROMS

Andres

Hi,

I am a really new user, and I've been trying to figure out the basics of ROMS use, compilation, etc.

Apart from having a User Manual, I'd like to suggest creating a 'ROMS Wiki'. People could contribute to it with all kinds of tips, test cases and other things, without adding overhead to the developers. The Forum would still be a great place to ask specific questions, but when problems are solved, they could be summarized/added to the Wiki. This allows to 'build' documentation iteratively.

I would be happy to help with this, as I've done wiki-like projects before.

Cheers,
Carlos

[m.hadfield]

andres wrote:

Quite a good idea. My students are building a notes on the class I teach using wiki and it is sometimes surprising the quality and dedication they put in the contributions.

Good, it's agreed then: your students can write the ROMS wiki

[andres]

No problem! Well, a couple. It will be in spanish and where... Is there a Wiki installed in Rutgers server? I could ask for space in our deparment's server, if it's ok.

Andres

andres wrote:

Quite a good idea. My students are building a notes on the class I teach using wiki and it is sometimes surprising the quality and dedication they put in the contributions.

Good, it's agreed then: your students can write the ROMS wiki

[arango]

Yes, we had discussed the ROMS Wiki in the past. We think that it is a good idea. We want to have several levels of documentation since our Users have different degrees of expertise. Perhaps we can have Novice, Intermediate and Expert levels of documentation. I don't know if we will be able to have the ROMS Wiki in several languages because we need to be certain that information loaded is accurate. We have some ideas about its design but suggestions are encouraged.

I believe that David Robertson (ROMS webmaster) already started one. Let me check its status. We can host this ROMS Wiki at http://www.myroms.org.

We also developed a ROMS weblog for developers. Notice that we have posted a lot of messages on this blog, but they are associated to ROMS adjoint-based algorithms. We are also thinking about an User Blog, but we have been not able to differentiate it with ROMS Forum. Perhaps, the User Blog can be used for applications and technical information and the ROMS Forum for more interactive discussions.

Anyway, suggestions are welcome.

[cmoffat]

I think a Forum like this is a much better way to conduct discussions than a blog, although is nice to have the developer's blog to follow the coding progress.

As for the Wiki, I didn't know there were plans for it. However, I'd like to suggest that some of the interested users (starting with Andres, me, and others that hopefully will come along) get the Wiki up and running, as there's no need, I believe, to burden the ROMS webmaster with this task, as I've been looking at a few places to host this and I don't think it will be a problem.

As for the accuracy of the information, I think an open Wiki is the best way to go. Documentation will be written iteratively as people add more details and test cases and so on. That way we ensure that the production of documentation is not an excessive burden on the developers as can be done by users themselves.

Anyways, I'm looking into a place to host the Wiki which I think will make most people happy, and would save the webmaster a lot of work.

How do you guys feel about putting the documentation under a GNU FDL license?

[Kirang]

Hello everyone!

I have a small suggestion for the ROMS manual. Currently,the manual includes the directory structure and the role of each routine.However,for a neophyte like me,it is difficult to figure out how exactly the equations get evaluated using these routines in detail.

Regards,
Katherine