IN040-07
Portable Python-wrapped FV3GFS global atmospheric model
Tuesday, 15 December 2020: 19:18
Virtual
Jeremy McGibbon1, Noah Brenowitz2, Mark Cheeseman2, Spencer Clark1, Johann Dahm2, Eddie Davis2, Oliver Elbert2, Rhea George2, Brian M Henn2, Anna Kwa2, Andre Perkins1, Oliver Watt-Meyer1, Tobias Wicky2, Christopher S. Bretherton1,2 and Oliver Fuhrer2, (1)Vulcan, Inc., Climate Modeling, Seattle, WA, United States, (2)Vulcan, Climate Modeling, Seattle, WA, United States
Abstract:
Some of the most challenging and time-consuming tasks in the atmospheric sciences involve downloading, running, and modifying weather and climate models. fv3gfs-python is a python-wrapped version of the FV3GFS global atmospheric model, used by NOAA as the US operational weather forecasting model. The model main loop executes in Python, making it easier to understand model execution, insert diagnostic routines, or augment the model with machine learning. In this presentation, we will show how you can download the model in Docker and reproduce a prepared experiment on your laptop in minutes.
