Rita LBAR, A98E, BAMM, GFDL?

A reader had some questions today about the weather forecast models from which we show forecast hurricane tracks on our Professional site. Below is an example of the storm tracks from this afternoon:

HURRICANE RITA MODEL TRACKS 9/21 Afternoon

All of these tracks are from government-run weather prediction models which are geared specifically to tracking tropical systems. With the exception of the GFDL, all models are two-dimensional, predicting a storm's track only.

Here's what the abbreviations mean:

A98E - National Hurricane Center Atlantic 1998 Early Model
BAMM - Beta and Advection Model (Medium Layer)
BAMD - Beta and Advection Model (Deep Layer)
LBAR - Limited Area Barotropic Model
GFDL - Geophysical Fluid Dynamics Laboratory Model

Which model does the best? It depends on who you ask :) Like other meteorological models, each seems to forecast better in certain situations. Notice that today's tracks are converged on the central Texas coastline which makes meteorologists feel confident about the forecast that we are issuing today. Should these tracks begin to diverge with different landfalls, the forecast will be more risky.

I don't want to get too geeky here, but some (admittedly dated) verification statistics and advanced descriptions of the models are available here, if you are interested. Although the ones above are the most popular models, there are actually over 30 models that forecast tropical tracks, and we are working to get all of them on our website.

Here's a brief description of the ones on our Pro site:

A98E - A statistical track model which looks at past storms which formed in the same place, and mixes this forecast with upper-level forecast model data from the GFS. Storms are classified based on location and movement, then specific meteorological equations are applied.

BAMM/BAMD - A track model which uses an altered version of forecast model data from the GFS with special meteorological equations that predict the evolution of tropical storms, including taking into account the "beta effect" of the earth's rotation. The BAMD uses an average of winds over the entire atmosphere from top to bottom (good for big storms), while the BAMM uses a medium vertical average (good for medium storms). There is also a BAMS (shallow) for small storms.

LBAR - A an altered version of forecast model data from the GFS with special meteorological equations and techniques (different from the BAM), including an average of winds over the entire atmosphere from top to bottom, that predict the evolution of tropical storms.

GFDL - An advanced 3-dimensional weather forecast model developed specifically for predicting hurricane movement. It includes 18 levels of the atmosphere and 3 nested grids (resolutions). It includes special meteorological equations and techniques for forming storms but obtains its initial conditions from the GFS.