The best new hurricane model you may not know about

HAFS model leaps ahead in hurricane prediction

HAFS was the most accurate model to predict Cat 5 Hurricane Dorian as it hit the Bahamas Sept 2019. (NOAA)

For most people outside the weather research community, few are unaware of the changes being made to forecast models.

The GFS and Euro names are commonly pushed across social pages and ignite discussion as to which ones are the most accurate.

Pay attention to the Hurricane Analysis and Forecast System (HAFS), which shows great promise as the next-generation forecast tool for reliable predictions.

This model is still in development but took top place for track accuracy during the 2019 hurricane season, according to hurricane scientists at NOAA’s Atlantic Oceanographic and Meteorological Laboratory (AOML).

Models have trouble combining both world views of weather and smaller atmospheric features down to a couple of miles wide all at once. The HAFS architecture can see both the large scale weather patterns around a hurricane along with smaller inner core weather features by nesting the domains together.

AOML’s real-time experiment during the 2019 Atlantic hurricane season evaluated the model’s performance and found that the regional version called HAFS-globalnest showed was equal to or better than other models.

The superior forecasts of HAFS compared to other global models was primarily due to the nested parameters over the tropical cyclone.

So how good were the results?

HAFS regional shows better forecast skill for weak storms’ intensity forecasts and strong storms’ track forecasts. The results are encouraging given that the high-resolution TC models tend to over-predict the intensity of weak storms

HAFS regional showed track improvements compared to GFS, HWRF and HMON at almost all forecast lead times and as much as roughly 20% over HWRF.

The regional and global HAFS had lower track errors during the 2019 hurricane season compared to other NOAA models.

In cases like Hurricane Dorian, the regional HAFS track outperformed the GFS, HWRF and HMON. which incorrectly predicted the Florida landfall of Dorian in multiple cases. The forecast error is the lowest among all models during the 24–120 h period.

HAFS never predicted hurricane Dorian to hit Florida while the GFS (A), HWRF (B), and HMON (C) all did.

It also predicted the recurvature along the Florida coast more accurately, with none of its cases making landfall in Florida.

Researchers in the near future hope to make the models more accurate by improving how initial data are assimilated into the model.

These changes will help clear up how the inner circulation center or hurricane vortex is structured which is crucial to reducing intensity forecast errors.

About the Author: