Forecasting is tough stuff..

[Aquawind]

Just a reminder..lol

http://vh10066.v1.moc.gbahn.net/apps/pb ... 40395/1075

[dhweather]

That's why I use the IMH model, it seems to work pretty well.


That's In My Head !

[WeatherEmperor]

That is a very interesting article. Thanks!

<RICKY>

[hookemfins]

"The problem is most people don't understand the models, including meteorologists. They take the worst-case scenario for their location and jump on it. That's a terrible mistake. These models are only for guidance. You have to figure out which one is working best for the situation you're in now."


Naw, no one from these boards does that


Common sense is the best. Take Ivan for example, I thought all along that the ridge was stonger than what the modles indicated. So I had Ivan staying on a WNW track a little longer around the western edge of Cuba.

This year I don't think the modles knew hadle the Atlantic ridge's strength.

[vbhoutex]

I have to agree with you on the models and the ridge this year. It became pretty obvious IMO with each successive storm that they appeared to be underestimating the ridge. That is the reason I kept saying the same thing you did, especially with Ivan. I wonder if the modelers will try and tweak them to get rid of this apparent "problem". Overall the models did a decent job, but as stated, they are for guidance, not the be all end all in forecasting.

[hookemfins]

GFS used to be good but they did some tweaking to it, I think the made it a higher resolution, and since then it's become weaker. So to answer your question, yes but you can make it worse.

For Ivan the CMC was the only modle that consistenly kept Ivan as a nother gulf coast threat. UKMET was the next one to.

[MWatkins]

This comment, as much as any, is a GREAT argument for the TPC one voice policy. If all or most local TV mets think this, then take them out of the process.

" (from Dr Steve Lyons) The problem is most people don't understand the models, including meteorologists. They take the worst-case scenario for their location and jump on it. That's a terrible mistake. These models are only for guidance. You have to figure out which one is working best for the situation you're in now.

A local met responds:

WINK-TV meteorologist Jim Farrell takes exception to that comment, referring to Steve Lyons' quote:
"I know Steve; I'm a friend of his, but I don't know how one person can speak for the meteorological community at large," he said. "We at WINK look at every single tool available to us. We've certainly never taken a worst-case model. It's like the Olympics, you throw out the worst and throw out the best and look at what's in the middle.

Dr Lyons is 100% dead on with his note that you have to lean toward the model that is working best at the time, if you lean toward one at all. There were local met reports, time after time during Jeanne, that wanted to take Jeanne to SC because a model (the GFS) said so. But clearly the GFS was weakening the high presssure ridge way too quickly to be accurate. It is not like the Olympics at all, you can't just toss the outliers because the outlier may be right (see NOGAPS durng Jeanne and Frances). And how many times did folks in central FL hear about the GFS during Jeanne? Or OH no the GFDL has it coming right at us? Answer...ALL the time.

Models are an art and a science and each set of guidance is unique. I think the WINK guy is missing the point because he thinks Steve may have taken a shot at him. I wonder how many local mets even know to look at the upper air charts, especially 500MB charts, to make conclusions. Because based on everything I read from this season, and saw on TV, the answer could be in the 30% range at best, I really don't know.

I had people who know ZERO about models telling me what the BAMM was saying. But without knowing that the BAMM is a mid-level model that does not account for pressure heights above 400MB (and we know that strong hurricanes like Jeanne are better off being looked at up to a level of 200MB), or that the model only looks at horizontal wind trajectories, and not direct pressure relationships etc, lend the model relatively useless. Yet, average clowns reading the output of this model and saying the models don't work right don't really know ANYTHING AT ALL to qualify that statement.

Most people outside of this board or the met field in general DO NOT KNOW HOW TO READ THE MODELS. This is the problem. The models arent crazy or screwed up. The people saying the models are screwed up are the problem

MW

[WeatherEmperor]

I couldnt agree more with you Mike.

<RICKY>

This comment, as much as any, is a GREAT argument for the TPC one voice policy. If all or most local TV mets think this, then take them out of the process.

" (from Dr Steve Lyons) The problem is most people don't understand the models, including meteorologists. They take the worst-case scenario for their location and jump on it. That's a terrible mistake. These models are only for guidance. You have to figure out which one is working best for the situation you're in now.

A local met responds:

WINK-TV meteorologist Jim Farrell takes exception to that comment, referring to Steve Lyons' quote:
"I know Steve; I'm a friend of his, but I don't know how one person can speak for the meteorological community at large," he said. "We at WINK look at every single tool available to us. We've certainly never taken a worst-case model. It's like the Olympics, you throw out the worst and throw out the best and look at what's in the middle.

Dr Lyons is 100% dead on with his note that you have to lean toward the model that is working best at the time, if you lean toward one at all. There were local met reports, time after time during Jeanne, that wanted to take Jeanne to SC because a model (the GFS) said so. But clearly the GFS was weakening the high presssure ridge way too quickly to be accurate. It is not like the Olympics at all, you can't just toss the outliers because the outlier may be right (see NOGAPS durng Jeanne and Frances). And how many times did folks in central FL hear about the GFS during Jeanne? Or OH no the GFDL has it coming right at us? Answer...ALL the time.

Models are an art and a science and each set of guidance is unique. I think the WINK guy is missing the point because he thinks Steve may have taken a shot at him. I wonder how many local mets even know to look at the upper air charts, especially 500MB charts, to make conclusions. Because based on everything I read from this season, and saw on TV, the answer could be in the 30% range at best, I really don't know.

I had people who know ZERO about models telling me what the BAMM was saying. But without knowing that the BAMM is a mid-level model that does not account for pressure heights above 400MB (and we know that strong hurricanes like Jeanne are better off being looked at up to a level of 200MB), or that the model only looks at horizontal wind trajectories, and not direct pressure relationships etc, lend the model relatively useless. Yet, average clowns reading the output of this model and saying the models don't work right don't really know ANYTHING AT ALL to qualify that statement.

Most people outside of this board or the met field in general DO NOT KNOW HOW TO READ THE MODELS. This is the problem. The models arent crazy or screwed up. The people saying the models are screwed up are the problem

MW

[Aquawind]

Well Said Mike. I hope we get plenty of input related to how the NHC can improve it's communication to Meteorologists, the NWS and the Public during your Talkin Tropics show tomorrow night..

http://www.talkintropics.com/

[cape_escape]

See , just one more reason why I love this place! Very interesting MW, thanks!

[birdwomn]

Most people outside of this board or the met field in general DO NOT KNOW HOW TO READ THE MODELS. This is the problem. The models arent crazy or screwed up. The people saying the models are screwed up are the problem

MW

Thanks for this and the many other great commetns on this topic. I will be the first one to admit I don't know the details of each model and just enjoy looking at what they say in comparison to what I see actually happening. That said, I DO know enough to realize that the people who work with them regularly must understand the strengeths and weaknesses of each and when one will be more useful than another, that's why they are the experts, and I am just an interested observer.

[Janie34]

Models are an art and a science and each set of guidance is unique. I think the WINK guy is missing the point because he thinks Steve may have taken a shot at him. I wonder how many local mets even know to look at the upper air charts, especially 500MB charts, to make conclusions. Because based on everything I read from this season, and saw on TV, the answer could be in the 30% range at best, I really don't know.

Thats an excellent question, MW. I too have to wonder about that, or even if some would know how to read a difax chart. Its very easy to copy information from a NWS teletype or NHC discussion. Thats not a slam against all broadcast mets. Some of them do seem quite clueless, though.