Differences with CROCO

[lcbernardo]

Dear all,

I had been a using Rutgers ROMS/COAWST for a number of years, but due to a recent change in my work place, I have now also been assigned to use the CROCO system. And though my previous experience has allowed me to understand the general usage, I'd like to ask for any advice or help regarding the following:

1) With regards to the grid configuration, I am more familiar with Vtransform, Vstretching specification, while in Croco, it involves activating the or deactivating the NEW_S_COORD key in cppdefs.h. If NEW_S_COORD is defined, I'd like to know what the closest configuration in Rutgers ROMS/COAWST would be. Looking at the documentation, it seems to be either Vtransform=2 and Vstretching=2, or Vtransform=2 and Vstretching=5, but perhaps someone familiar with both systems can point out the correct one?

2) I'm using Croco mainly due to the interest in using the PISCES model. If I were to continue to stick to ROMS/COAWST, would anyone be able to suggest some alternatives? I imagine I might also be able to adapt my configuration into Croco, but there might be some minute differences that would cause some issues that would be difficult to resolve.

I would greatly appreciate any help and advice. Thanks!

Lawrence

[kate]

I don't know what the capabilities of PISCES is, but my ROMS "master" branch has the COBALT model. The person who put it into ROMS has moved on, taken a different job, and I'm hardly an expert. The files for a Gulf of Alaska domain are in my Apps_master repository (time to rename some of these things...). Anyway, it shouldn't be that hard to merge into COAWST, but isn't for the faint of heart. Plus with 30+ tracers ROMS is a more costly platform for it than MOM6.

[Kosa]

CoSiNE might be another one to consider.

[jsouza]

Another one is the FENNEL model. From my limited experience with geochemical modelling, more complexity doesn't mean better model/results - sometimes the opposite.

[lcbernardo]

Thank you so much for your responses. They give me an initial idea at least of some of the options that I can look into.

I don't know what the capabilities of PISCES is, but my ROMS "master" branch has the COBALT model. The person who put it into ROMS has moved on, taken a different job, and I'm hardly an expert. The files for a Gulf of Alaska domain are in my Apps_master repository (time to rename some of these things...). Anyway, it shouldn't be that hard to merge into COAWST, but isn't for the faint of heart. Plus with 30+ tracers ROMS is a more costly platform for it than MOM6.

I'm not familiar with the COBALT model, but I think it can help me to look at your setup. I'll try to access the files, and also keep in mind that it might be challenging to adapt. Indeed, I very much need to mind the potential computational requirements.

CoSiNE might be another one to consider.

Another one is the FENNEL model. From my limited experience with geochemical modelling, more complexity doesn't mean better model/results - sometimes the opposite.

As for CoSiNE and FENNEL, I think I've come across them before. And indeed, more complexity doesn't automatically translate to better results. I'm trying to find a good balance, between my institution's current capabilities, the datasets which might be available to me, and my own skill at implementing the models.

[patrickm]

NEW_S_COORD in CROCO corresponds to Vstretching=4 in ROMS/COAWST (first introduced in ROMS-UCLA in 2010)

[lcbernardo]

NEW_S_COORD in CROCO corresponds to Vstretching=4 in ROMS/COAWST (first introduced in ROMS-UCLA in 2010)

Thanks Patrick! I really hoped to clarify this. I think I can with more confidence now look deeper into how I may be able to use both versions for my work.