5 day Cone

[AJC3]

It was my understanding that the original request for NHC to issue routine 5-day forecasts came from the U.S. Navy.

[AussieMark]

I like the cone, it's the black line that needs to go.

Good point but who focuses on that line

the Media

remember Charley and the black line was showing somewhere near Tampa

[brunota2003]

I like the cone, it's the black line that needs to go.

Good point but who focuses on that line

the Media

remember Charley and the black line was showing somewhere near Tampa

I remember that...and the people Charley hit were under a hurricane warning, yet were not prepared because they watched the line...(no I'm not generalizing...)

[ziggy1122]

I had understood that the 5-day forcast was requested most by the DoD. I believe for their direction of naval craft at sea.

I wish it wouldn't be public information as I see it causing more problems than results for the general public who don't know how to handle its data.

[StrongWind]

I personally don't think that it is worth anything. But it does do two things. One, it does tend to keep the publics attention. Two, it sure generates some really interesting debates here on Storm2k, ie. It's Texas, NO it's Florida, NO it's Nova Scotia. LOL

It also gives the media at least 2 more days of 'hype time.' Wouldn't want to deprive them of their livelyhood would we.

[weatherwoman]

any one who lives along coastal areas knows when the hurricane season is, so why would anyone get caught off guard cone or no cone. i like the cone but hey lets face it they never know where one of these things is going. I love it each time we have one the weather forecasters will come on and say this is a hard one to predict. they make it sound like the rest were so easy but they say that about each storm. leave the cone its gives them something to play with when they really don't know where the storm is going makes them look good anyway

[DrCloud]

This is an interesting discussion for me, because I know well the limitations of the 5-day track forecast, its public misinterpretation, and its usefulness. I'm constantly trying to educate people about the first two and put the last to the best use I can.

One problem, of course, is that lots of folks assume that the storm starts at the apex of the cone as a little-bitty thing and then grows to the size of the cone as it moves along. Arrgh!

But the most useful thing I've found is the evolution of the forecasts over time. The one for Ernesto, for example (http://www.nhc.noaa.gov/archive/2006/ERNESTO_graphics.shtml as of the time of this post) shows how the forecast track rotated clockwise over time. I picked up on this early on and started mentioning it to people around here with the comment that we really need to pay close attention when the track stabilizes. That was helpful for folks who were starting to panic too soon.

I think that one way for the NHC to improve this product would be to combine (or "blend", as they like to say) the climatological statistics shown in an earlier post here with the envelope of the forecast tracks from the different models. At least that way, the size of the cone would change from storm to storm and provide more information about real-time uncertainties. HPH

[HomesteadHoney]

Funny you should mention this DrCloud!

"One problem, of course, is that lots of folks assume that the storm starts at the apex of the cone as a little-bitty thing and then grows to the size of the cone as it moves along. Arrgh!"

I was at my mother-in-laws community association meeting and this is exactly what was being explained. I did my best to challenge the park expert but was summarily dismissed as too young to understand weather dynamics (I'm 42!)

I suppose that way of thinking is better than assuming that it will only hit "on the line". At least they prepare.

[wxman57]

While NHC has changed the time period for their long-term means from 10 years to 5 years, they did not change how the cone is calculated. The cone is still (at least during 2006) drawn using the average errors from the last 10 years. And it only HAPPENS to contain the actual track 60% of the time - it wasn't drawn purposely that way.

That's now what I understood when I heard them speak at the NHC last spring. They said they were using a 5-year error cone this season. This is a slightly smaller cone than was used previously, to account for less error in the past 5 years vs. the last 10 years. These new 0-120hr 5-year error values are represented on the graphic I posted yesterday. Those values are used to determine cone size at various time steps.

As for the 60% bit, I tried getting that answer out of them but wasn't successful. The error values perpendicular to the cone do match up with the 60% 5-year probability, but the NHC is vague in describing what percentage error the cone represents. I've heard them say that the cone represents a 60-70% probability, but the error numbers say 60% not 65 or 70%.

[hurricanemike]

AJC3,

You stand correct. The Navy requested the 5 day forecast in 2004 to get them the ability to sortie to the ships out of harms way.

[Rainband]

Well according to our news local EOC insisted upon it too.

[DanKellFla]

Aren't the 5 day forecast as accurate as the 3 day forecasts from 5 or 10 years ago? I remember reading something like that.

[Patrick99]

We've been in an awful lot of 5-day cones.

Really, how much longer than 3 days do you need?

The only cone that means anything to me is the 1-day cone. hehe