Long term vision for this package?

[glwagner]

I'm curious what the long-term vision for this package is. There are some features here that I think would be very useful to the Oceananigans community (eg drag laws). Their impact is limited by being in this package because they aren't as visible (for example it would be nice to describe them in the Oceananigans documentation.

Also, I think they could be further developed since there is a substantial distance one can go with drag laws, including impactful cutting-edge research. Other features partially overlap with features of other packages like ClimaOcean. Combining effort can then make sense, rather than duplicating them, which seems more wasteful.

Here's a longer list of thoughts:

• Drag models for LES and also for large scale ocean models (which share a lot of the same mathematical form) are broadly useful and can be developed to a high degree of sophistication, including adding solvers for viscous sublayer roughnless length. More community power might be devoted to this if we move the drag models to Oceananigans.
• This work may be more impactful in Oceananigans, where its visibility can be increased through usage in examples and documentation.
• wind stress and surface heat fluxes are computed by bulk formula in ClimaOcean, which uses a similar model as this package (except with the ability to diagnose the MO stability length). Implementing the models in this package as a simplicification of the ClimaOcean bulk formula (eg a prescribed MO length) could make sense and have many benefits, including more powerful functionality (since ClimaOcean will also eventually include models for diagnostic surface temperature, high quality atmospheric thermodynamics, spatial averaging for LES, etc)
• radiative heating would also be a great feature for Oceananigans, since it is needed for BGC models and also for physics in realistic context (like those considered by ClimaOcean). I think the scientific community would benefit from a range of models, from simple prescribed attenuation length, to climatology-based models, to data-driven models, to models that depend on prognostic tracers such as chlorophyll.
• More generally, the more that we collaborate rather than disperse efforts, I think the more science we can do and stronger we make the community, making us all better off.

[jagoosw]

This package wasn't originally meant to be a package, I made it because I was copying over all of the things in here between a load of experiments and thought it would be easier to just put them in a package.

Happy to move the wall model over to Oceananigans when I have time

[glwagner]

That makes sense! I think what you're describing is good motivation for providing a source code implementation though, so putting code into a package makes sense to me.

There's no need to do any work either now or in the future. I'm wondering if you agree with what I wrote. For example, a different strategy could be to put effort into developing this package independently instead.