1. Introduction -The snowstorm of January 6-7, 2002
failed to quantify the amount of snowfall, areal coverage of snowfall,
and the timing of the snow event due to failed data assimilation and
model dynamics. The ETA model and SREF model failed by not representing
the following areas: 1. Shear and curvature vorticity. 2. Upper level diffluence and convection. 3) Model resolution, upper level dynamics, and data assimilation.
Questions:
Does an advanced forecaster have to review 4-5 COMET NWP modules to gain further understanding of this case study?
Can we FOCUS the NWP modules to gain insight on mid-upper level diabatic process within the NWP model?
Can this NWP case study provide me with a 3D visualization of the model underperforming or overperforming?
2. Discussion - The advanced weather forecaster assimilates and interrogates environmental data prior to displaying NWP models to provide a conceptual view of atmosphere. A thorough analysis is critical to asses the initial state of numerical models and to discover possible errors that may exist. A problem exists where the advanced forecaster does not have sufficient knowledge of NWP dynamics or structure. How can a forecaster obtain a better working knowledge of the inner working of a NWP model, "the black box".
3. Instructional Design - A suggestion would be to develop visualization software that can rerun the model and focus on the errors that this case study presents. (mid-upr level dynamics, convective processes, frozen precipitation processes). The advanced forecaster would have good and bad model runs from which to evaluate their impact on his forecast products. The case study area should finish this training scenario by providing educational content on how a forecaster adjusts model guidance using the digital forecast process.
4. Smart Tools - The digital forecast process is in place at NSW forecast offices. Weather forecasters account for errors in model output by running python scripts to adjust guidance. There are over 300 tools and most run basic routines that allow a forecaster to adjust or tweak guidance. These results are detached from the model. Most forecasters do not have complete understanding of the NWP models to understand the digital output in scenario driven situations.
5. Summary - The NWP case studies can be improved by providing 3D visualizations of good and bad model forecasts. A 3D visualization aids in graphically representing where a good model goes bad. Specific section sof COMET NWP models could be injected into the cast studies to focus the students learning and address the error in model guidance. Lastly, the NWP case studies needs to complete the end to end forecast training by providing content on the digital forecast process.
