Changes to Model Data for 2007

[wxman57]

The most important changes:

2. THE SHALLOW BETA ADVECTION MODEL /BAMS/ WILL BE ADDED.

3. THE STATISTICAL TRACK MODELS /A98E/ AND /P91E/ WILL BE
DISCONTINUED.

http://www.nws.noaa.gov/om/notification ... ce_msg.txt

Here's an example of the new guidance package:

http://www.nhc.noaa.gov/whxx01_sample_2007.shtml

Good riddance to the climo models. Don't know why they kepe LBAR, though.

[caneflyer]

The models being discontinued in this message are not climatological models. A98E and P91E are statistical-dynamical models.

[wxman57]

The models being discontinued in this message are not climatological models. A98E and P91E are statistical-dynamical models.

But they're much more statistical (climatological) than dynamic.

[caneflyer]

You are confusing statistical with climatological. Models such as A98E are called statistical-dynamical models because they are statistical in structure, meaning they are derived from linear regression, but use as their key predictors inputs from dynamical models. Statistical refers to the framework or structure of the model, while dynamical refers to what kind of predictors are used within that framework.

CLIPER, on the other hand, is a statistical model that uses climatology as its predictors. (It also uses persistence predictors, hence the name CLI-PER.) The difference between the statistical-climatological model CLIPER and the statistical-dynamical model A98E is that CLIPER doesn't know anything about the current state of the atmosphere and A98E does.

[brunota2003]

They better keep the LBAR, what other model would provide us with such enjoyment and laughter when we need it most? I really dont pay attention to what models do what though. I have enough of an explanation of what they are doing in the forecast discussion for my liking.

[MWatkins]

If you ask me, that entire suite of data, other than the decay ships info, is almost worthless.

The Beta Advection guidance has error rates almost double the globals, and the official forecast.

There is valuable information there (for example, the much discussed by me indicator of directional shear evidenced by the difference between the BAMM and BAMD guidance) etc.

I suppose they are keeping the LBAR in there because it does reasonaly well in environments that are as close you can get to purely baratropic (IE the deep tropics under a strong ridge) in nature.

On the other hand, beyond looking at the SHIPS guidance, the intial strength and position estimates, and the paramaters used to initialize the models, I would throw out all of the BAM guidance if LBAR were to get chucked too.

Wouldn't it be nice to get NOGAPS, UKMET, GFDL, GFS, and the CONU guidance all on one easy to get to report?

Yeah...I know...idealist.

MW

[Derek Ortt]

there's no change in the model data being run. Only a change on what is released on those public releases

[wxman57]

If you ask me, that entire suite of data, other than the decay ships info, is almost worthless.

The Beta Advection guidance has error rates almost double the globals, and the official forecast.

There is valuable information there (for example, the much discussed by me indicator of directional shear evidenced by the difference between the BAMM and BAMD guidance) etc.

I suppose they are keeping the LBAR in there because it does reasonaly well in environments that are as close you can get to purely baratropic (IE the deep tropics under a strong ridge) in nature.

On the other hand, beyond looking at the SHIPS guidance, the intial strength and position estimates, and the paramaters used to initialize the models, I would throw out all of the BAM guidance if LBAR were to get chucked too.

Wouldn't it be nice to get NOGAPS, UKMET, GFDL, GFS, and the CONU guidance all on one easy to get to report?

Yeah...I know...idealist.

MW

I don't think that the beta and advection models should even be released to the public. They are the most misused and misunderstood models. They might provide a good first guess as to where a disturbance near the African coast might go - assuming there will be no change in the steering pattern in the next 5 days. I.E., a developing system trapped by a strong Bermuda high. But once a disturbance is far enough north or west that the steering flow could change over the 5 day forecast, you just can't use them for projecting where a storm will go. So they're pretty bad for most systems approaching the Caribbean Sea, the Gulf, or the east U.S. Coast.