CV storm climo 1851-2004: US hitters that formed 9/15+

[LarryWx]

Exact dates of FORMATION (TD+) for the 44 1851-2004 CV storms* that later hit the U.S:

7/5, 7/15, 7/31, 8/3, 8/3, 8/5, 8/7, 8/7, 8/15, 8/15, 8/16, 8/16, 8/17, 8/17, 8/18, 8/19, 8/20, 8/20, 8/21, 8/23, 8/25, 8/25, 8/27, 8/28, 8/28, 8/29, 8/29, 9/2, 9/3, 9/4, 9/6, 9/6, 9/7, 9/8, 9/10, 9/10, 9/10, 9/10, 9/11, 9/15, 9/16, 9/21, 9/21, 9/25

*My def. of CV storm: tropical storm/hurricane that first became at least a TD E of 50W and S of 20N

So, 154 years of history clearly show a sharp peak on 9/10. Being that it looks like no new TD's will form through 9/14, here are some stats for 9/15+ formations:

- From 1851-2004, there were 36 CV storms that formed within the interval of 9/15-9/25.

- Five later hit the U.S. and all five were majors at some point along their trek. Here are their formation dates: 9/25/1893 (storm #9 hit SC), 9/21/1966 (Inez hit FL), 9/16/1985 (Gloria hit NC to NE), 9/15/1998 (Georges hit FL keys and MS), and 9/21/2002 (Lili hit LA)

- So, five of 36 (13%) 1851-2004 CV storms that formed 9/15-9/25 later hit the U.S.

- This compares to ~19% for the entire hurricane season.

- So, the chances for a U.S. hit from any one CV storm that forms 9/15-25 are ~2/3 as high as they are for CV storms that form throughout the season as a whole. That makes sense since it is more difficult to make it across the Atlantic by this point in the season vs. earlier. But each still has about a 1 in 8 chance, a nontrivial chance of hitting the U.S.

- From 1851-2002, there were 36 T.S.+'s that were first declared a T.D.+ east of 55W and south of 20N ON OR AFTER 9/26. NONE of these later hit the U.S. So, once past 9/25, climo says the chances are extremely low for a storm that forms east of 55W to later hit the U.S.

[Anonymous]

SOME MEMORABLE AFTER SEPTEMBER 15TH HURRICANES:::
Cat 3 into Key Largo, FL:
Cat 3 into Port Charlotte/Sarasota area:
Cat 2-4 into Tampa region:

[LarryWx]

Thanks Floydbuster. Keep in mind that the three storms that Floydbuster showed are NOT CV storms. But showing them actually helps to make a nice point relevant to my post: a large majority of the worst late season U.S. landfalling storms landfalls have NOT been CV storms.

[Regit]

Here's a good CV storm... Formed Oct 5. Still not too late this year. It ain't over till it's over.

[HurricaneBill]

Also keep in mind, while the CV season winds down, that's the time to start paying attention to the Western Caribbean.

Many powerful late-season storms have formed there.

[LarryWx]

Here's a good CV storm... Formed Oct 5. Still not too late this year. It ain't over till it's over.

Regit,
I had said, "From 1851-2002, there were 36 T.S.+'s that were first declared a T.D.+ east of 55W and south of 20N ON OR AFTER 9/26. NONE of these later hit the U.S. So, once past 9/25, climo says the chances are extremely low for a storm that forms east of 55W to later hit the U.S."

Hazel of 1954 doesn't fit the above mentioned criteria of forming into a TD+ EAST of 55W. It formed at 59W. However, like Floydbuster's three storms, it is another good example of a late season major U.S. storm that did NOT form east of 55W. I'm not at all saying the season is anywhere close to being over as far as U.S. threats are concerned. I'm not even saying the CV storm threat to the U.S. is necessarily over. That wouldn't be until 9/26 although we are admittedly past the meat of the CV season. Climo still suggests about a 1 in 8 chance for any single CV storm that forms 9/15-25 to later hit the U.S. That's way too high a chance to ignore imho.

[HurricaneQueen]

Although Hazel occurred before most of the members (and in some cases their parents) were born but I remember her well as a life changing experience when she roared through PA. She took out our steel bridge that was the link to the nearest big city making my home area a virtual ghost town. Houses along the Delaware River were under water with just the tip of their chimneys showing. One of my father's businesses declared bankruptcy the following year because there were no customers.

I was only eight but well remember the boil water advisories. We were forbidden to play in the standing water (thought our parents were REALLY mean). Personally, the worse part was all the shots we had to get to avoid the various diseases that can occur after severe flooding.

Perry, as always, thanks for your terrific historical information. Thanks, also, to Floydbuster for his contributions.

Lynn

[LarryWx]

Also keep in mind, while the CV season winds down, that's the time to start paying attention to the Western Caribbean.

Many powerful late-season storms have formed there.

Absolutely!

[LarryWx]

Perry, as always, thanks for your terrific historical information. Thanks, also, to Floydbuster for his contributions.
Lynn

You're welcome Lynn although I'm Larry not Perry. But being mistaken for Perry is a true complement in itself! His historical posts are some of the best/most interesting/best written.

[Regit]

Here's a good CV storm... Formed Oct 5. Still not too late this year. It ain't over till it's over.

Regit,
I had said, "From 1851-2002, there were 36 T.S.+'s that were first declared a T.D.+ east of 55W and south of 20N ON OR AFTER 9/26. NONE of these later hit the U.S. So, once past 9/25, climo says the chances are extremely low for a storm that forms east of 55W to later hit the U.S."

Hazel of 1954 doesn't fit the above mentioned criteria of forming into a TD+ EAST of 55W. It formed at 59W. However, like Floydbuster's three storms, it is another good example of a late season major U.S. storm that did NOT form east of 55W. I'm not at all saying the season is anywhere close to being over as far as U.S. threats are concerned. I'm not even saying the CV storm threat to the U.S. is necessarily over. That wouldn't be until 9/26 although we are admittedly past the meat of the CV season. Climo still suggests about a 1 in 8 chance for any single CV storm that forms to later hit the U.S. 9/15-25. That's too high a chance to ignore imho.

Well if we're getting technical... Hazel DID form East of 55W. Notice that in its first advisory it was a 65MPH tropical storm. Theres no way it went from tropical low to nearly a cane just like that. Hazel had been around for a while and no one was aware because of the technology around in 1954. We do know about the last 40 years however, and 40 years is not enough time to build a model for determining the likelihood of a meteorlogical event.

[LarryWx]

Well if we're getting technical... Hazel DID form East of 55W. Notice that in its first advisory it was a 65MPH tropical storm. Theres no way it went from tropical low to nearly a cane just like that. Hazel had been around for a while and no one was aware because of the technology around in 1954. We do know about the last 40 years however, and 40 years is not enough time to build a model for determining the likelihood of a meteorlogical event.

Regit,
1) I'm getting technical because I'm following my very clear-cut criteria. Why shouldn't I?

2) Neither you nor I know what it was east of 55W. Whereas what you say sounds reasonable, we still don't really know. I'm going by the data that is available.

3) It was likely developing quite rapidily as evidenced by it going all of the way to a CAT 3 by 67W. Also, it was moving quite slowly for that area..just under 4 degrees longitude/day. So, at that rate, it could have easily taken more than 24 hours to move from 55W to 59.2W. Going from the first designation as a TD to a 60 knot TS or stronger within 24 hours is far from impossible and has occurred four times in just the last 10 years:

a) Humberto of 1995 went to a 60 knot TS within 24 hours of first being designated a TD:

http://weather.unisys.com/hurricane/atl ... /track.dat

b) Bill of 1997 went to a 60 knot TS within 24 hours of first being designated a TD:

http://weather.unisys.com/hurricane/atl ... /track.dat

c) Nicole of 1998 went to a 60 knot TS within 18 hours of first being designated a TD:

http://weather.unisys.com/hurricane/atl ... /track.dat

d) Florence of of 2000 went to a 65 knot hurricane within 24 hours of first being designated a sub-TD:

http://weather.unisys.com/hurricane/atl ... /track.dat

4) Plenty of storms were still detected east of 55W in 1954 and earlier. It isn't as if the data older than 40 years is useless. Regardless, there have been 36 storms that are recorded as having first formed east of 55W since 1851 on 9/26 or later and ZERO have hit the U.S. That tells me a lot.

5) Even so, I disagree with your statement that 40 years is not enough time to use for determining the chance for a met. event. It obviously isn't as good as 154 years, but it isn't bad imho.

[HurricaneQueen]

Sorry about the mistake, Larry. It was easily made, however. Your data is terrific- among the very best!

Lynn

[Regit]

Well if we're getting technical... Hazel DID form East of 55W. Notice that in its first advisory it was a 65MPH tropical storm. Theres no way it went from tropical low to nearly a cane just like that. Hazel had been around for a while and no one was aware because of the technology around in 1954. We do know about the last 40 years however, and 40 years is not enough time to build a model for determining the likelihood of a meteorlogical event.

Regit,
1) I'm getting technical because I'm following my very clear-cut criteria. Why shouldn't I?

2) Neither you nor I know what it was east of 55W. Whereas what you say sounds reasonable, we still don't really know. I'm going by the data that is available.

3) It was likely developing quite rapidily as evidenced by it going all of the way to a CAT 3 by 67W. Also, it was moving quite slowly for that area..just under 4 degrees longitude/day. So, at that rate, it could have easily taken more than 24 hours to move from 55W to 59.2W. Going from the first designation as a TD to a 60 knot TS or stronger within 24 hours is far from impossible and has occurred four times in just the last 10 years:

a) Humberto of 1995 went to a 60 knot TS within 24 hours of first being designated a TD:

http://weather.unisys.com/hurricane/atl ... /track.dat

b) Bill of 1997 went to a 60 knot TS within 24 hours of first being designated a TD:

http://weather.unisys.com/hurricane/atl ... /track.dat

c) Nicole of 1998 went to a 60 knot TS within 18 hours of first being designated a TD:

http://weather.unisys.com/hurricane/atl ... /track.dat

d) Florence of of 2000 went to a 65 knot hurricane within 24 hours of first being designated a sub-TD:

http://weather.unisys.com/hurricane/atl ... /track.dat

4) Plenty of storms were still detected east of 55W in 1954 and earlier. It isn't as if the data older than 40 years is useless. Regardless, there have been 36 storms that are recorded as having first formed east of 55W since 1851 on 9/26 or later and ZERO have hit the U.S. That tells me a lot.

5) Even so, I disagree with your statement that 40 years is not enough time to use for determining the chance for a met. event. It obviously isn't as good as 154 years, but it isn't bad imho.

I know all about all the technicalities of meteorology and also know the track of every single known hurricane since 1851 as I prepared research on them while in college.

Also, doing that research in South Carolina, Hazel was one of my favorite case studies since its category at landfall has long been argued. Any professor on the subject would argue that Hazel was alive before 55W. Of course you can't prove it, but not much in tropical meteorology can be proven. We can't prove what Ophelia's top winds are now, the NHC just uses its best estimate. A scientist has to realize that setting strict standards often poses problems with data that is only classified as the best estimate, rather than testable data.

Also, about her strengthening: going from Cat 1 to Cat 3 in two days isn't a major accomplishment for a hurricane.

Furthermore, I must stand steadfastly for my statement that 40 years isn't enough time to predict the chance of a meteorological event. It's not even CLOSE to enough time. Weather patterns are cyclical, often by cycles of 40-70 years. You need to know the behavior of several of those cycles before true and accurate guesses can be made about the storms based on history. There are also larger cycles of weather activity worldwide. I'd say to have weather patterns down to a science, we need to observe at least 11,000 years of data. We've got a pretty good guess now, but can anyone give you more than a guess of what kind of season 2010 will be? The same goes for track forecasting. We need to see thousands of tracks before seeing the ones that are most likely.

And I must say I am impressed by your research on the subject. A lot of good work in there. I just enjoy a scholarly debate with someone who has studied them as I have. I have actually seen meteorologists and geographers argue after a few drinks. It's even more fun then.