I have a question regarding atmospheric forcing generation with the pytools preprocessing scripts.
From what I understood during Webinar 1, the current preprocessing tools support ERA5, GLOFAS, HYCOM and Mercator. However, I did not see support for GFS forcing.
In my workflow, I used to rely on the Matlab tools to download GFS data via OPeNDAP and generate the forcing NetCDF files. Unfortunately, since February 24, OPeNDAP access to GFS is no longer be available, and those tools have started to fail.
I was wondering whether there are existing scripts (or plans) to integrate GFS support into the pytools preprocessing framework? If not, I would also be interested in knowing whether anyone in the community is currently working on this or has alternative suggestions for operational use.
Thank you very much for your work on the tools and for any guidance you can provide.
Thank you for your message and for raising this point.
At the moment, there is indeed no ready-to-use support for GFS atmospheric forcing in the CROCO pytools preprocessing framework. Transitioning from the legacy Matlab tools toward a fully Python-based workflow is definitely something we would like to do, and GFS support is clearly part of that longer-term objective, but nothing is available yet.
For now, one possible workaround is to access GFS data through the NOAA NOMADS GRIB Filter: https://nomads.ncep.noaa.gov/
However, we do not currently provide any CROCO-specific, ready-to-use scripts to process GFS data from this source into forcing NetCDF files.
If members of the community are already experimenting with GFS workflows in Python, we would of course be very interested in feedback or contributions. Otherwise, this remains an open area for future development.
Thank you again for your interest in the tools and for sharing your experience—this kind of feedback is very valuable to us.
That makes perfect sense. I fully understand that the transition from matlab_tools to a fully Python-based workflow takes time, but its great to know that GFS support is part of the longer-term objective.
Since I rely on GFS for operational atmospheric forcing, I will try to implement a Python workflow similar to what the Matlab scripts were doing for GFS forcing generation. If it works well, I would be glad to share it with the community following the appropriate contribution process.