Hurricane Forecasting is inaccurate

From:
http://www.sciencedaily.com/upi/?fee...c-us-wilma.xml

Hurricane Wilma baffling prediction models
MIAMI, Oct. 20 (UPI) -- Hurricane Wilma, the strongest hurricane ever
recorded, had winds of 150 mph Thursday morning, but its behavior was
baffling projections about its path.

Forecasters at the National Hurricane Center in Miami said their most
reliable computer projection systems had enormously wide swings on
where the storm could go over several hours Wednesday. At one point,
data indicated it would eventually hit Maine, then it changed to Cuba,
then back to Rhode Island.

snipend


Other forecasting models are having similar problems. They've all
failed at predicting Wilma's behavior.

The one data point every 4 kilometer grid sucks balls and it's not
surprising. That's a very low resolution for a chaotic system. It's a
fiendishly difficult problem to model and the ability to do so won't be
possible with current computing methods.


thanks,

K. Gringioni.
awaiting the arrival of the qbit

Kurgan Gringioni wrote:


From:
http://www.sciencedaily.com/upi/?fee...-1-20051020-07
084700-bc-us-wilma.xml

Hurricane Wilma baffling prediction models
MIAMI, Oct. 20 (UPI) -- Hurricane Wilma, the strongest hurricane ever
recorded, had winds of 150 mph Thursday morning, but its behavior was
baffling projections about its path.

Forecasters at the National Hurricane Center in Miami said their most
reliable computer projection systems had enormously wide swings on
where the storm could go over several hours Wednesday. At one point,
data indicated it would eventually hit Maine, then it changed to Cuba,
then back to Rhode Island.

snipend


Other forecasting models are having similar problems. They've all
failed at predicting Wilma's behavior.

The one data point every 4 kilometer grid sucks balls and it's not
surprising. That's a very low resolution for a chaotic system. It's a
fiendishly difficult problem to model and the ability to do so won't
be possible with current computing methods.


thanks,

K. Gringioni.
awaiting the arrival of the qbit



You need to start reading the discussions at www.nhc.noaa.gov. The problem
is not the resolution of the models in general but that this particular
storm is being affected by a low pressure system over the Great Lakes.
There are a couple of steering systems that are predicted to interact after
the storm crosses Florida. In general the prediction models do a good job,
but sometimes the solution gets stiffer than others, and then the solutions
from the different models diverge. This is one of those times. (Note that
the article kind of confuses that the situation is not that same model is
producing different tracks, but that the track solutions from several
different models are very different. See the above mentioned website's
discussion archive for more details.)

The models aren't all bad, they aren't all good either but still, weren't
you the one who once claimed nothing is black or white.

It is also not the strongest, but most intense, as determined by the
central pressure.

"You suck!" said the wind field to the eye. "Blow me!" said the eye.

--
Bill Asher

On 20 Oct 2005 10:03:36 -0700, "Kurgan Gringioni"
wrote:

Other forecasting models are having similar problems. They've all
failed at predicting Wilma's behavior.

If Wilma was a dumbass, you could predict her behavior - please note
the predictability of rbr.

In fact, most newbies don't pick this up, but sometimes all the
regulars but one take an entire month off and that one person posts
for everyone else. Except H squared, 'cause we the rest of us don't
have the sensitivity thing down and we can't ask our wives or
girlfriends as that would require admitting that we are posting to
rbr.

So if everyone but H squared is posting, probably no one here but one
regular with time on his hands and a bunch of newbies. Was H squared
posting in July? Anybody here back then?

Curtis L. Russell
Odenton, MD (USA)
Just someone on two wheels...

William Asher wrote:
Kurgan Gringioni wrote:


From:
http://www.sciencedaily.com/upi/?fee...-1-20051020-07
084700-bc-us-wilma.xml

Hurricane Wilma baffling prediction models
MIAMI, Oct. 20 (UPI) -- Hurricane Wilma, the strongest hurricane ever
recorded, had winds of 150 mph Thursday morning, but its behavior was
baffling projections about its path.

Forecasters at the National Hurricane Center in Miami said their most
reliable computer projection systems had enormously wide swings on
where the storm could go over several hours Wednesday. At one point,
data indicated it would eventually hit Maine, then it changed to Cuba,
then back to Rhode Island.

snipend


Other forecasting models are having similar problems. They've all
failed at predicting Wilma's behavior.

The one data point every 4 kilometer grid sucks balls and it's not
surprising. That's a very low resolution for a chaotic system. It's a
fiendishly difficult problem to model and the ability to do so won't
be possible with current computing methods.


thanks,

K. Gringioni.
awaiting the arrival of the qbit




You need to start reading the discussions at www.nhc.noaa.gov. The problem
is not the resolution of the models in general but that this particular
storm is being affected by a low pressure system over the Great Lakes.
There are a couple of steering systems that are predicted to interact after
the storm crosses Florida. In general the prediction models do a good job,
but sometimes the solution gets stiffer than others, and then the solutions
from the different models diverge. This is one of those times. (Note that
the article kind of confuses that the situation is not that same model is
producing different tracks, but that the track solutions from several
different models are very different. See the above mentioned website's
discussion archive for more details.)

Bah,

you all don't get the real reason for these storms:

http://www.weatherwars.info/katrina.htm

The storm pictures there are impressing, though.

Kurgan Gringioni wrote:

Other forecasting models are having similar problems. They've all
failed at predicting Wilma's behavior.

Will it affect the price of fish in Horotewhenua ?

William Asher wrote:

You need to start reading the discussions at www.nhc.noaa.gov. The problem
is not the resolution of the models in general but that this particular
storm is being affected by a low pressure system over the Great Lakes.
There are a couple of steering systems that are predicted to interact after
the storm crosses Florida. In general the prediction models do a good job,
but sometimes the solution gets stiffer than others, and then the solutions
from the different models diverge. This is one of those times.



Dumbass -

From:
http://quote.bloomberg.com/apps/news...r=news_inde x
``Two days before Katrina ripped through the Gulf of Mexico forecasters
were saying it would make landfall at Tallahassee, Florida,'' said Adam
Sieminski, chief energy economist at Deutsche Bank AG in New York. ``It
ended up making landfall south of New Orleans and we know the results.
Hurricane forecasting is still an art not a science.''

snipend


thanks,

K. Gringioni.

SNIP

you all don't get the real reason for these storms:

http://www.weatherwars.info/katrina.htm

The storm pictures there are impressing, though.

I like the horseshoe vortex video, though I disagree with the analysis
that it is man-made.

Maybe if we all start riding our bikes in a big counter-closckwise
circuit, all at the same time, we will be able to "steer" Wilma.

Ernst Noch wrote:

snip

Bah,

you all don't get the real reason for these storms:

http://www.weatherwars.info/katrina.htm

The storm pictures there are impressing, though.



He's just be pretending to believe they aren't natural and controlled by
the Russians so that you keep believing he's a nut job and they are natural
and you never figure out that they aren't natural and are controlled but
it's not the Russians that are controlling them. Maybe.

Think about it. A radially laced bicycle wheel looks like a hurricane if
you straighten out the rain bands and Tom Simpson died on Mt. Ventoux. Or
was he killed?

--
Bill Asher

Kurgan Gringioni wrote:


William Asher wrote:

You need to start reading the discussions at www.nhc.noaa.gov. The
problem is not the resolution of the models in general but that this
particular storm is being affected by a low pressure system over the
Great Lakes. There are a couple of steering systems that are
predicted to interact after the storm crosses Florida. In general
the prediction models do a good job, but sometimes the solution gets
stiffer than others, and then the solutions from the different models
diverge. This is one of those times.



Dumbass -

From:
http://quote.bloomberg.com/apps/news...u8WGh0WQA&refe
r=news_index ``Two days before Katrina ripped through the Gulf of
Mexico forecasters were saying it would make landfall at Tallahassee,
Florida,'' said Adam Sieminski, chief energy economist at Deutsche
Bank AG in New York. ``It ended up making landfall south of New
Orleans and we know the results. Hurricane forecasting is still an art
not a science.''

snipend


thanks,

K. Gringioni.



Read the archived discussions and look at the forecast tracks from the NHC
website. They got Katrina dead-bang on all things considered.

http://www.nhc.noaa.gov/archive/2005...graphics.shtml

You can see that the Deutsche Bank AG guy is talking through his hat. The
consensus forecast at three days out had Katrina ripping right through New
Orleans.

Really, you need to go look at the primary data before you can have a
reasonable discussion about this.

http://www.nhc.noaa.gov/archive/2005/KATRINA.shtml?

Eyewall replacement cycle forecasting is voodoo. Intensity forecasting is
an art. Track forecasting is much much better.

--
Bill Asher

Kurgan Gringioni wrote:

Hurricane Wilma baffling prediction models

Other forecasting models are having similar problems. They've all
failed at predicting Wilma's behavior.

The one data point every 4 kilometer grid sucks balls and it's not
surprising. That's a very low resolution for a chaotic system. It's a
fiendishly difficult problem to model and the ability to do so won't be
possible with current computing methods.

Dumbass,

Forget the 4 km grid bull**** and the supercomputers.
Just look: Katrina, Rita, Wilma. Scientists have difficulty
predicting their behavior, because as everyone knows,
scientists[*] don't understand women. And to you,
this is news?

Ben
[*] Male scientists, that is. I wouldn't want to be
sexist, now.