Add a prescribed ocean to the src #412
Conversation
| ##### A prescribed ocean... | ||
| ##### | ||
|
|
||
| struct PrescribedOceanModel{G, C, U, T, F, Arch} <: AbstractModel{Nothing, Arch} |
There was a problem hiding this comment.
We call it PrescribedAtmosphere. Is this PrescribedOcean or PrescribedOceanModel?
I'm not sure really sure what a "prescribed model" is.
There was a problem hiding this comment.
Hmm, this is because we add a simulation on top, while the atmosphere is a "simulation" rather than a model. I can switch it to a simulation rather than a model
There was a problem hiding this comment.
this is because the infrastructure we have now expects an ocean with a pattern of ocean.model.stuff..., while there is no assumption on the atmosphere. I did not want to make a huge refactor here, I can call it PrescribedOcean though and just have one field called model inside of it
| τx = Field{Face, Center, Nothing}(grid) | ||
| τy = Field{Center, Face, Nothing}(grid) | ||
| Jᵀ = Field{Center, Center, Nothing}(grid) | ||
| Jˢ = Field{Center, Center, Nothing}(grid) |
There was a problem hiding this comment.
why this? I think this is wrong
|
|
||
| Arguments | ||
| ========= | ||
| - `timeseries`: A `NamedTuple` containing time series data for various fields. The named tuple can be empty |
There was a problem hiding this comment.
the name is confusing because when we say "timeseries" it means just one series. But this is many timeseries?
This PR adds the
PrescribedOceanModelto the src (shuffling around a couple of methods). This prescribed ocean is useful to perform single-column simulations of a sea ice model sandwiched between a prescribed atmosphere and a prescribed ocean, to verify the performance of the fluxes and the sea ice climatology