Model Behavior- NOAA is testing 2 new weather models - WMBFNews.com, Myrtle Beach/Florence SC, Weather

Model Behavior- NOAA is testing 2 new weather models

MPAS Hexagon grid design MPAS Hexagon grid design
fv3 cubed-sphere grid design fv3 cubed-sphere grid design

MYRTLE BEACH, SC (WMBF)- More and more people are becoming aware of different forecast computer models. Hurricane Sandy introduced America to the European forecast model (Euro) when it drastically outperformed the American model (GFS) in predicting Sandy’s track.

Since Hurricane Sandy, both global forecast models have undergone some upgrades, but the GFS is still lacking in a few areas. Congress gave NOAA a little money after Hurricane Sandy and NOAA put the funds to work. They have been testing two experimental global models and will make a decision over the next few months as to which one they feel best fits the needs of the meteorology community in the US.

The two models are very different. The MPAS (Model for Prediction Across Scales) was developed by scientist at the National Center for Atmospheric Research (NCAR). The MPAS uses a revolutionary grid of hexagons to map out the globe instead of traditional cubes. This new technique allows for very high resolution of weather features.

The other model being tested is the FV3 and was developed by NOAA. This model uses a “cubed-sphere” grid pattern that is similar to a traditional grid pattern. Since FV3 was developed by NOAA, it’s considered an “in-house” model and would be a significant upgrade to the GFS.

There are some strong opinions in the scientific community about which model would be best for us to adopt as our go-to global model to replace the GFS. Time will tell- most likely a lot of time. Changes of this scale take a lot of time to fully adopt, especially when life and property are on the line during major weather events. Bottom line, our forecast super computers will get a major upgrade in the next couple of years that will hopefully put us back in contention with the Euro.

Powered by Frankly