• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Hurricane computer models

G

glen

Lifer
storm_96.gif

Are we to pretend there is some sort fo predictive value in that?
 
Gibson486 said:
can you do any better?
Click to expand...
Well, from the looks, it is totaly random.
So, a monkey with darts could do better 50% of the time.


But, the point is the models imply they contain information and have predictive value.
But, from the looks of this spread, they dont.
It might be better for the weather tracks to simply say, "We have no idea," than to imply they have the ability to track.
 
I think you're complaining about the length of time it predicts for. As long as it gives warning earlier than looking out your window, it has value.
 
MagnusTheBrewer said:
I think you're complaining about the length of time it predicts for. As long as it gives warning earlier than looking out your window, it has value.
Click to expand...

I agree. The models may work well, say 24 hours in advance. But this model doesn't say anything. We don't know if peopel in Mexico need to worry, or people in SC>
 
Errr... you're complaining that they can't accurately predict the path of a hurricane which doesn't even exist yet? 😕

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

THE CENTER OF CIRCULATION OF THE BROAD AREA OF LOW PRESSURE OVER THE
GULF OF MEXICO APPEARS TO BE BECOMING BETTER DEFINED ABOUT 275
MILES SOUTH-SOUTHEAST OF THE MOUTH OF THE MISSISSIPPI RIVER. IN
ADDITION...SHOWERS AND THUNDERSTORMS HAVE BEEN INCREASING OVER THE
EASTERN GULF OF MEXICO...AND A NOAA BUOY LOCATED ABOUT 130 MILES
EAST-NORTHEAST OF THE CENTER HAS PRODUCED TROPICAL-STORM-FORCE
WINDS WITHIN THE PAST COUPLE OF HOURS. A TROPICAL DEPRESSION...OR
MORE LIKELY A TROPICAL STORM...COULD FORM LATER TODAY OR TONIGHT IF
THE DEVELOPMENT TREND CONTINUES. THIS SYSTEM HAS A HIGH CHANCE...
90 PERCENT...OF BECOMING A TROPICAL CYCLONE AS IT MOVES SLOWLY
NORTHWARD DURING THE NEXT 48 HOURS.
Click to expand...
 
glen said:
Not at all.
I am complaining that they are implying that they CAN track it.
Click to expand...

You obviously didn't read the multiple explanations and disclaimers on the website that you grabbed the track image from.

When they think they have an accurate model of the storm's path, you'll see it show up via NOAA.
 
I think that by releasing all the models on a single map, they're NOT implying that they CAN track it. They're showing that the models aren't in agreement with each other. I can see how the average person might use this to say "see, they don't know what they're talking about." Actually, they do know what they're talking about, but people think that because the mathematics behind these prediction models works perfectly for one tropical storm 3 days out, they should work perfectly for ALL storms over the same span of time. This is not true. I can explain mathematically for you, giving you specific math equations and data to plug into the equations - you'll see that over a pretty wide range of values for the initial input (let's call it a measurement that can be off by as much as 0.7), that you will be able to predict the output after 4 or 5 iterations. However, for some very narrow neighborhoods for the inputs, you'll see that by changing the input just very very slightly - by as little as one ten-thousandth, after 4 iterations, you'll get one value; change the value by another ten-thousandth, and you'll get a radically different result a few iterations down the road. The mathematics is boring to most people, but if you're interested in at least a little bit, perhaps a good starting place is understanding a bifurcation diagram, or see if you can find an intro to chaos theory.

But, perhaps a better way to explain it: imagine a pencil that's perfectly balanced on its very sharp point. You see a person walking across the floor toward the pencil and know that it's going to tip over from the vibrations. Which way is it going to go? Now, compare this to a pencil that's already leaning just a tiny amount. Pretty easy to predict which way it's going to fall, and you can predict nearly perfectly where it will land. Compare yet again to a pencil that's tipped over at a 45 degree angle, but now, toss in a slight breeze perpendicular to the plane over which the pencil will be falling - still pretty easy to predict, though there's a little more uncertainty due to the breeze. Now, factor in dozens of other things and certainty will change. But, for some factors (how vertical the pencil is initially), occasionally, you'll run into a case where all the other calculations follow from a very uncertain measurement at the beginning.
 
Originally Posted by Gibson486
can you do any better?

mxrider said:
Dave can
Click to expand...

Yeppers, the Dave Model has always been extremely accurate.

By the way, the current storm in the Gulf is the fastest four storms has ever formed before July 1st since record keeping began in 1851.

A couple of days ago the big expensive models had the thing heading to Texas. I don't know what they were smoking but it's heading to the western edge of the Florida Panhandle, the more normal route for a storm formed in southeast quadrant of the Gulf.

It's a big rainmaker and I think the region needs it?

The trade winds are just starting to kick up in the Atlantic so expect the first named storm coming off Africa.

NostaDaveous
 
I think I once read that an zettaflop supercomputer would be able to fully model weather for the planet for a 2 week time span. Only a couple of more decades to go.
 
DrPizza said:
I think that by releasing all the models on a single map, they're NOT implying that they CAN track it. They're showing that the models aren't in agreement with each other. I can see how the average person might use this to say "see, they don't know what they're talking about." Actually, they do know what they're talking about, but people think that because the mathematics behind these prediction models works perfectly for one tropical storm 3 days out, they should work perfectly for ALL storms over the same span of time. This is not true. I can explain mathematically for you, giving you specific math equations and data to plug into the equations - you'll see that over a pretty wide range of values for the initial input (let's call it a measurement that can be off by as much as 0.7), that you will be able to predict the output after 4 or 5 iterations. However, for some very narrow neighborhoods for the inputs, you'll see that by changing the input just very very slightly - by as little as one ten-thousandth, after 4 iterations, you'll get one value; change the value by another ten-thousandth, and you'll get a radically different result a few iterations down the road. The mathematics is boring to most people, but if you're interested in at least a little bit, perhaps a good starting place is understanding a bifurcation diagram, or see if you can find an intro to chaos theory.

But, perhaps a better way to explain it: imagine a pencil that's perfectly balanced on its very sharp point. You see a person walking across the floor toward the pencil and know that it's going to tip over from the vibrations. Which way is it going to go? Now, compare this to a pencil that's already leaning just a tiny amount. Pretty easy to predict which way it's going to fall, and you can predict nearly perfectly where it will land. Compare yet again to a pencil that's tipped over at a 45 degree angle, but now, toss in a slight breeze perpendicular to the plane over which the pencil will be falling - still pretty easy to predict, though there's a little more uncertainty due to the breeze. Now, factor in dozens of other things and certainty will change. But, for some factors (how vertical the pencil is initially), occasionally, you'll run into a case where all the other calculations follow from a very uncertain measurement at the beginning.
Click to expand...
And ultimately not every model is equally useful, even though the NHC runs all of them anyhow. Certain models are better suited for predicting storms out on the open seas, other models are better for predicting land interaction, etc. A big part of storm forecasting is knowing which models to listen to and why.

On that note, a bit of discussion from Jeff Masters on the models for Debby.

The National Hurricane Center is forecasting Debby to remain a tropical cyclone over the next 5 days as it drifts north and then west toward Texas. The Hurricane Center is forecasting a very slow progression of the storm, with a potential landfall not occurring until Friday. However, most of the models that predict the turn to the west suggest landfall will happen before or around Wednesday. The models are still generally split on the forecast for Debby; by Monday, the majority of the reliable models, including the ECMWF, NOGAPS, HWRF, and UKMET, agree that a ridge of high pressure will build in over the Southern U.S., forcing Debby west across the Gulf of Mexico and into South Texas by Wednesday. However, the GFS model, which has been our 2nd most reliable track model over the past two years (behind the ECMWF), has consistently been predicting that a trough of low pressure pushing off of the U.S. East Coast will be capable of grabbing the disturbance and accelerating it to the northeast across Florida north of Tampa Bay on Monday. The GFDL model splits the difference between these extremes and takes Debby north to a landfall near the Alabama/Florida border on Tuesday. The predicted track west to Texas is still the most likely outcome, though it remains a low-confidence forecast. In terms of intensity, none of the models is predicting Debby will become a hurricane, nor is the Hurricane Center. Though sea surface temperature is warm (and around 1°F above average), the actual heat content of the Gulf is relatively low. Wind shear is predicted to remain moderately strong through Sunday, but will increase to 30+ knots by Tuesday.
Click to expand...
 
Weather is shitty here in North Tampa. Rain rain rain. I've been watching this storm closely, as it seems like it is the type of storm that if it does follow the GFS model, is going to catch this area completely off guard. Its almost as if a lot of people don't realize we have a trop. storm sitting right off our coast.
 
Cuda1447 said:
Weather is shitty here in North Tampa. Rain rain rain. I've been watching this storm closely, as it seems like it is the type of storm that if it does follow the GFS model, is going to catch this area completely off guard. Its almost as if a lot of people don't realize we have a trop. storm sitting right off our coast.
Click to expand...

Looks to me the GFS Model is closest to what I said a couple of days ago.

It is even more extreme showing it going over your head heading to the Atlantic.

http://mag.ncep.noaa.gov/NCOMAGWEB/...NTIC&currKey=model&returnToModel=&imageSize=M
 
DrPizza said:
I think that by releasing all the models on a single map, they're NOT implying that they CAN track it. They're showing that the models aren't in agreement with each other. I can see how the average person might use this to say "see, they don't know what they're talking about." Actually, they do know what they're talking about, but people think that because the mathematics behind these prediction models works perfectly for one tropical storm 3 days out, they should work perfectly for ALL storms over the same span of time. This is not true. I can explain mathematically for you, giving you specific math equations and data to plug into the equations - you'll see that over a pretty wide range of values for the initial input (let's call it a measurement that can be off by as much as 0.7), that you will be able to predict the output after 4 or 5 iterations. However, for some very narrow neighborhoods for the inputs, you'll see that by changing the input just very very slightly - by as little as one ten-thousandth, after 4 iterations, you'll get one value; change the value by another ten-thousandth, and you'll get a radically different result a few iterations down the road. The mathematics is boring to most people, but if you're interested in at least a little bit, perhaps a good starting place is understanding a bifurcation diagram, or see if you can find an intro to chaos theory.

But, perhaps a better way to explain it: imagine a pencil that's perfectly balanced on its very sharp point. You see a person walking across the floor toward the pencil and know that it's going to tip over from the vibrations. Which way is it going to go? Now, compare this to a pencil that's already leaning just a tiny amount. Pretty easy to predict which way it's going to fall, and you can predict nearly perfectly where it will land. Compare yet again to a pencil that's tipped over at a 45 degree angle, but now, toss in a slight breeze perpendicular to the plane over which the pencil will be falling - still pretty easy to predict, though there's a little more uncertainty due to the breeze. Now, factor in dozens of other things and certainty will change. But, for some factors (how vertical the pencil is initially), occasionally, you'll run into a case where all the other calculations follow from a very uncertain measurement at the beginning.
Click to expand...

i had to solve a 5-dimensional equation using the newton-raphson method...and if you didn't pick a good initial point, you would never find the answer 😀

anyway, to the OP - since these are predictions, you can issue a general warning to the area. as the hurricane moves, you can update your model based on how it moves to get more accurate predictions, and go from there.
 
OverVolt said:
Oh right. Hot winter probably means a ridiculous hurricane season.
Click to expand...

Not true on it's own but...one of the things that enhances rapid development is sea surface temps and we had a very mild winter (FL) with a very hot spring, if your running your AC in April that's unusual..
 
Well, looks like I will be getting some rain in the future because of this thing. The funny thing is seeing all the snowbirds stocking up on plywood, gas, generators, etc for something that will be a weak cat1 at worst. Locals don't stock up on those things for a 1. You stock up on rum instead.
 
My sister has a one year old and it's her first year in the Florida panhandle, Pensacola specifically. She's a bit anxious about it. Looks like a damn near direct hit (although a bit east) of her...but still a solid shot of the hardest part of it.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top