How to relax T and S back to ECCO: Did something replace ECCORestoring?

[lgloege]

I am trying to run a simple global simulation off of this example. I was trying to use ECCORestoring to set up restoring forcings for temperature and salinity. However, it seems that ClimaOcean.ECCO: ECCO4Monthly is no longer available, and possibly the OceanSeaIceModel now handles restoring internally.

Could you please provide guidance on the recommended way to relax temperature and salinity back to ECCO in the current version? In the past I have used ECCORestoring, did something replace this function? Thanks

[simone-silvestri]

Hello @Igloege. Indeed, there has been some refactoring going on recently to allow restoring with datasets other than ECCO. The new restoring type is DatasetRestoring. You can see how it works here
or by entering julia and running

julia> using ClimaOcean

julia> ? DatasetRestoring

[lgloege]

Hi @simone-silvestri, thanks, happy to hear the functions are becoming generalized! Can I assume the same will be happening with JRA55PrescribedAtmosphere() at some point?

One project on my backlog is to run simulations with various wind products: CCMP, ERA5, JRA3Q, .... My plan is to modify JRA55PrescribedAtmosphere() to just use a different wind field, but keep JRA55 for the rest of the atmosphere

[simone-silvestri]

That's great! There is, for sure, interest in adding different atmospheres.
We already have a PrescribedAtmosphere, the JRA55PrescribedAtmosphere is just a constructor for a specific PrescribedAtmosphere loaded with FieldTimeSeries that use metadata = JRA55Metadata().

We are planning to generalize FieldTimeSeries to include atmospheric data also. You can take insipiration on how to build another PrescribedAtmosphere using other datasets from this PR that aims to build a prescribed atmosphere from ECCO data