search

noaa-ocs-modeling / EnsemblePerturbation

perturbation of coupled model input over a space of input variables
https://ensembleperturbation.readthedocs.io
Creative Commons Zero v1.0 Universal
7 stars 3 forks source link

Spinup run hurricane track #110

Closed SorooshMani-NOAA closed 1 year ago

SorooshMani-NOAA commented 1 year ago

According to test in https://github.com/oceanmodeling/ondemand-storm-workflow/issues/24 the spin up run always uses the best track, and downloads it from the internet instead of using the provided track.

SorooshMani-NOAA commented 1 year ago

The issues is how the spinup_configuration is created from the base configuration, while the perturbed configuration is using the dictionary that points to the correct track files for each of the hotstart runs. See:

https://github.com/noaa-ocs-modeling/CoupledModelDriver/blob/1cc7c129aba3b9ed562410a63af98da1db40caec/coupledmodeldriver/generate/schism/generate.py#L161

vs

https://github.com/noaa-ocs-modeling/CoupledModelDriver/blob/1cc7c129aba3b9ed562410a63af98da1db40caec/coupledmodeldriver/generate/schism/generate.py#L178

There might not be a good solution inside the ensemble perturbation for this, but this can be addressed in the client code by passing the right track to the BestTrack forcing created for generation of base (main) configuration. Then this file will automatically be overwritten by the code referenced above for all non-spinup tracks.

SorooshMani-NOAA commented 1 year ago

The issue is resolved in the client script.