🚀[FEA]: Use GEFS members as initial condition data source

[djzurawski]

Is this a new feature, an improvement, or a change to existing functionality?

New Feature

How would you describe the priority of this feature request

Low (would be nice)

Please provide a clear description of problem you would like to solve.

I would like to be able to select a GEFS member to use for initial conditions for a prognostic model. A few weeks ago I attempted to repurpose the GEFS_FX forecast source but failed.

[NickGeneva]

Hi @djzurawski

How did it fail can you provide some details?
Did selecting the member with the product string not work as intended?

[djzurawski]

This solution should have been obvious to me the first time I dont know why it wasnt. The DataArray returned from data.GEFS_FX calls contain lead_time as a coordinate which the initialization datasources like data.GFS do not have which was the source of my errors. I got it working by creating a child class of GEFS_FX which selects the 0 lead_time (or any lead_time you want for time-lagging).

class GEFS_INIT(GEFS_FX):

    def __call__(
        self,
        time: datetime | list[datetime] | TimeArray,
        variable: str | list[str] | VariableArray,
    ) -> xr.DataArray:

        da = super().__call__(time=time, lead_time=timedelta(0), variable=variable)
        return da.isel(lead_time=0)

Then you can follow the Readme quickstart example

model = DLWP.load_model(DLWP.load_default_package())
ds = GEFS_INIT(product="gep10")
io = NetCDF4Backend("output.nc")

run(["2024-01-01"], 10, model, ds, io)

[NickGeneva]

Thanks for the follow up with your solution!
Closing for now, can reopen if others would find a GEFS data source particularly useful.