Atlantic Tropical Action 2013

[HurricaneJosh]

For me, damage is a non discussion point when looking at meteorological return rates.  I certainly understand the viewpoint of the '38 champions, but the window of opportunity to create a Sandy is much less than that of a 1938.  One doesn't get full latitude troughs that strong and that far south very often in a 100 year period even in late October.  If I recall, it was a -5SD event.  Even at that, you'd be hard pressed to get one outside late October/November and during tropical season.  Then, you need a late season hurricane passing by and a blocking high to its NE in order for said hurricane to become captured by the trough to its west at such a low latitude and be further enhanced baroclinically while thrown into the coast at such a perpendicular angle. 

 

The difference to me is that a cat 5 can occur at any time from July-Oct, and has a much higher frequency of occurrence than the mid latitude trough we saw last year.  Granted, 1938 hit a meteorological sweet spot of rare events itself, but to me, it all starts with the initial conditions.  A cat 5 can occur for a three month period.  A -5SD trough that far south can only occur during a much narrower window during tropical season.

 

Having a Cat 5 E of the Bahamas with a mechanism for shooting it due N at speeds up to almost 60 knots is just as unusual as the Sandy setup.  If you think it isn't, please show me another example of it (or something like it) having occurred.

[HurricaneJosh]

Yeah that's what I said... they were both catastrophic and quite similar.  1635 probably was a couple ticks lower in pressure and certainly had the higher surge.  Landsea estimated it met or exceeded Hugo's max surge, which is the highest ever recorded on the EC.  

 

What you originally said was "1635 redux would be worse than 1938 for many areas.  That'd be the ultimate SNE-centric weenie redux."  I disagree with that-- a 1938 track would be worse for more people, except maybe Boston Metro.  A 1938 'cane exposes a lot more people the right side.

[HurricaneJosh]

Josh, your INBOX is full, fyi.

 

 

Don't be so negative about Hurricane Season 2013. I still think we'll see hurricanes, obviously. Probably 6-8 of them.

Remember in 2000? One day we had Hurricane Isaac (Cat 4) turning east of Bermuda, Hurricane Joyce (Cat 1) forecast to be a dangerous Caribbean cruiser, and Hurricane Keith (Cat 4) coiled up against Belize. Of course, Joyce fizzled, but all of that activity (I was 12 at the time...) was very exciting.

The dates of that activity were September 28th-October 2nd of 2000.  That would be nearly 40 days from today.

I will say though, I have little faith in "violent" hurricane landfalls (Charley, Andrew) in the U.S. after September 18th. Once they go to the Carolinas it's 80-95 kt cattails in the green foam. Compared to tiny violent 130 kt coconuts blowing/2x4 through palm tree microcanes, I know they aren't your niche.

Mike

 

Thanks for the words of encouragement.  It all seems so dark right now.  

 

King 1950-- which hit Miami with winds of 110 kt-- is a good example of an Oct microcane in the USA.  I agree they're  less likely after late Sep, but they do happen.  And of course the NW Caribbean is great breeding ground for late-season micros, like Janet 1955 and Iris 2001.

[Sunny and Warm]

Having a Cat 5 E of the Bahamas with a mechanism for shooting it due N at speeds up to almost 60 knots is just as unusual as the Sandy setup.  If you think it isn't, please show me another example of it (or something like it) having occurred.

you seem too married to your thoughts, so maybe we should just let it go.  For the record though, I do believe what you describe above is inherently more feasible than Sandy.

[HurricaneJosh]

you seem too married to your thoughts, so maybe we should just let it go.  For the record though, I do believe what you describe above is inherently more feasible than Sandy.

 

It's not a matter of being married to it.  It's a matter of the official database (going back to 1851) and what we've observed even before then.  Cat 5s in themselves are extremely unusual. (The basin hasn't seen one in years.)  Cat 5s E of the Bahamas are a tiny subset of that.  Cat 5s E of  the Bahamas that have a mechanism for shooting them due N-- not NE, not NNE, but due N-- at speeds over 50 kt are an infinitesimally tiny subset of that.

[wxmx]

Also, It's very possible a Cat 5 has an anomaly > 5 SDs of the sea level pressure for that region it's in.

[Sunny and Warm]

It's not a matter of being married to it.  It's a matter of the official database (going back to 1851) and what we've observed even before then.  Cat 5s in themselves are extremely unusual. (The basin hasn't seen one in years.)  Cat 5s E of the Bahamas are a tiny subset of that.  Cat 5s E of  the Bahamas that have a mechanism for shooting them due N-- not NE, not NNE, but due N-- at speeds over 50 kt are an infinitesimally tiny subset of that.

 

 

Honestly, it doesn't matter that it was a cat 5 east of the Bahamas.  It matters what it was further north. 

 

Here is a study that discusses cat 3 or higher hurricanes in the NE, including the Long Island Express:

 

An empirical study of 20 past hurricanes that have impacted the New York City and Long Island coast regions by Scheffner and Butler (1996) found that the return period of a category 3 or greater hurricane is approximately 80 years. A strong category 3 or minimal category 4 hurricane has a return frequency of approximately 200 years. (Click graph to the right for larger image.) Therefore, it is not unlikely that another "epic" hurricane will strike the Long Island coastal region in the coming decades.

 

http://www2.sunysuffolk.edu/mandias/38hurricane/hurricane_climatology.html

 

 

Here is what they say about Sandy:

 

Abstract 

Hurricane Sandy’s track crossed the New Jersey coastline at an angle closer to perpendicular than any previous hurricane in the historic record. This steep angle was one of many contributing factors to a surge-plus-tide peak-water level that surpassed 4m in parts of New Jersey and New York. The lack of precedent in the historic record makes it difficult to estimate the rate of Sandy-like events using solely historic landfalls. Here we use a stochastic model built on historical hurricane data from the entire North Atlantic to generate a large sample of synthetic hurricane tracks. From this synthetic set we calculate that under long-term average climate conditions a hurricane of Sandy’s intensity or greater (category 1+) is expected to make NJ landfall at least as close to perpendicular as Sandy at an average annual rate of only 0.0014 yr-1 (95% confidence range 0.0007 to 0.0023); i.e., a return period of 714 yr (95% confidence range 1429 to 435)

 

http://www.ldeo.columbia.edu/~sobel/Papers/hall_sobel_grl_submitted.pdf

 

 

I'll give the Long Island Express a minimal 200 year return period and add on additional years for the way it got there.  However to me, they probably all got propelled at great speed up the coast to attain the wind speed they were clocked at along the coast.  I'd give it a 300 year return period, which is less than the estimate for Sandy.  While smaller than the 1938 hurricane, Sandy was more unique. 

[Ed Lizard]

Working hard for my WOTY nomination even though I don't post in OT anymore.

 

While y'all discuss whether 1938 or Sandy is more rare and remarkable, may I show the remarkable and wonderful similarities between the 12Z FIM 9 and 12Z GFS in about 2 weeks.

 

 

That is worth two  

 

[Sunny and Warm]

Also, It's very possible a Cat 5 has an anomaly > 5 SDs of the sea level pressure for that region it's in.

The trough for Sandy was -9SD according to some data I saw at the time, but I am hesitant to claim that since a couple Mets questioned it.  I believe it may have been though.  A cat 5 is a rare bird, but not more rare than the trough that captured Sandy in late October.  For late October and that far south, that was rare. 

[Sunny and Warm]

Working hard for my WOTY nomination even though I don't post in OT anymore.

 

While y'all discuss whether 1938 or Sandy is more rare and remarkable, may I show the remarkable and wonderful similarities between the 12Z FIM 9 and 12Z GFS in about 2 weeks.

 

 

That is worth two  

Thanks for sticking to the real stuff Ed.  Although, you'd probably get the weenie tag for posting a 336 hr map.  I do believe September starts off strong and goes the whole way through.

[HurricaneJosh]

Honestly, it doesn't matter that it was a cat 5 east of the Bahamas.  It matters what it was further north. 

 

Here is a study that discusses cat 3 or higher hurricanes in the NE, including the Long Island Express:

 

An empirical study of 20 past hurricanes that have impacted the New York City and Long Island coast regions by Scheffner and Butler (1996) found that the return period of a category 3 or greater hurricane is approximately 80 years. A strong category 3 or minimal category 4 hurricane has a return frequency of approximately 200 years. (Click graph to the right for larger image.) Therefore, it is not unlikely that another "epic" hurricane will strike the Long Island coastal region in the coming decades.

 

http://www2.sunysuffolk.edu/mandias/38hurricane/hurricane_climatology.html

 

 

Here is what they say about Sandy:

 

Abstract 

Hurricane Sandy’s track crossed the New Jersey coastline at an angle closer to perpendicular than any previous hurricane in the historic record. This steep angle was one of many contributing factors to a surge-plus-tide peak-water level that surpassed 4m in parts of New Jersey and New York. The lack of precedent in the historic record makes it difficult to estimate the rate of Sandy-like events using solely historic landfalls. Here we use a stochastic model built on historical hurricane data from the entire North Atlantic to generate a large sample of synthetic hurricane tracks. From this synthetic set we calculate that under long-term average climate conditions a hurricane of Sandy’s intensity or greater (category 1+) is expected to make NJ landfall at least as close to perpendicular as Sandy at an average annual rate of only 0.0014 yr-1 (95% confidence range 0.0007 to 0.0023); i.e., a return period of 714 yr (95% confidence range 1429 to 435)

 

http://www.ldeo.columbia.edu/~sobel/Papers/hall_sobel_grl_submitted.pdf

 

 

I'll give the Long Island Express a minimal 200 year return period and add on additional years for the way it got there.  However to me, they probably all got propelled at great speed up the coast to attain the wind speed they were clocked at along the coast.  I'd give it a 300 year return period, which is less than the estimate for Sandy.  While smaller than the 1938 hurricane, Sandy was more unique. 

 

*  Actually, it does matter what it was E of the Bahamas, because the fact that it was a Cat 5 down there is why it was so strong at 42N. 

 

*  You're lumping 1938 with all other majors in the NE USA without realizing that 1938 1) was much stronger than all the others, 2) was moving much faster, 2) exposed a lot more population to major-hurricane conditions than all the others.  The other majors were strong; 1938 was the strongest.  They all moved fast; 1938 was by far the fastest.  You can't lump it in with the others and then say it's really not that unusual.

 

*  It seems that you're  only focusing on track-- I'm focusing on track and intensity.  I guess we can pick any track arbitrarily track and say it's rare.

 

We're going to go around in circles with this boring topic all night.  If you want to think that Sandy was this unbelievable, nuclear, one-in-1,000,000-year event, unparalleled in the Earth's history, please do.  Sandy shills are becoming new genre of weather-board zombie.

[HurricaneJosh]

P.S.  That study you cited (Scheffner and Butler (1996))-- Re: return periods of majors to NYC/Long Island-- is ancient.  That's from before the data had even been reanalyzed.

[thewxmann]

You know the tropics are boring when we're talking about return periods of storms and 336 hr FIM plots.

[Drz1111]

The thing with 38 is that it missed all the areas that mattered. Did a number on RI, eastern LI, and then interior western New England. I went to school in NH surrounded by trees of the exact same size, which no one realized was because all of them had been planted in 1939....

...but 38 missed any population center of real consequence. Split the proverbial NY/BO'S uprights. If it happened today, some poor putzes in Westerly would have a hell of a storm, but 38 redux wouldn't shut down NYC for a week. 38, 60 miles to the east or 30 miles to the west is a storm with vastly more impact, but it is worth repeating; 38 flooded the hell out of RI, was deeper, had stronger winds, did horrific damage to the rural New England forests.... And if it happened again today, wouldn't match Sandy's damage. Sandy was like the Christchurch earthquake - not strong, but exquisitely well located.

[Ed Lizard]

You know the tropics are boring when we're talking about return periods of storms and 336 hr FIM plots.

 

 

We have a 10% lemon with no model support, and the new GFS shows a single TC the next 16 days, a fish at that.

 

 

I'd like to hear how Labor Day RI'd into a Cat 5 in one or two days myself.  And would chasers go to Monroe County to tape/video if a major was heading for the islands?  Would local law enforcement let them?

[jconsor]

1938 did not miss all the areas that mattered.  Providence was a large population center at the time and the downtown was totally submerged.  Also, the Boston area saw substantial damage.   Blue Hill just SW of Boston saw sustained winds of 121 mph with gusts to 186 mph.  See the following graphic, which estimates level of hurricane damage using damage to forests and structures. 

 

http://harvardforest.fas.harvard.edu/sites/harvardforest.fas.harvard.edu/files/data/p01/hf011/n1938dam.gif

 

See also the Sep 1815 hurricane for an example that shows severe damage can occur >75 miles east of the track of a large New England hurricane:

 

http://harvardforest.fas.harvard.edu/sites/harvardforest.fas.harvard.edu/files/data/p01/hf011/n1815dam.gif

 

These graphics are based on methodology in the following paper:

http://isites.harvard.edu/fs/docs/icb.topic281447.files/Hurricanes_NE.pdf

 

As far as potential damage from a 1938-like hurricane if it occurred today - the following study from leading risk management/catastrophe modeling company RMS estimates insured damage would be 40-55 billion (meaning total damage of 80-110 billion)

https://support.rms.com/publications/1938_Great_New_England_Hurricane.pdf

 

Also worth reading this presentation showing the strong correlation between forward speed and minimum pressure in extratropical transitioning hurricanes that impact the northeast US:

 

https://support.rms.com/publications/26meteor_muir-wood.pdf

 

The thing with 38 is that it missed all the areas that mattered. Did a number on RI, eastern LI, and then interior western New England. I went to school in NH surrounded by trees of the exact same size, which no one realized was because all of them had been planted in 1939....

...but 38 missed any population center of real consequence. Split the proverbial NY/BO'S uprights. If it happened today, some poor putzes in Westerly would have a hell of a storm, but 38 redux wouldn't shut down NYC for a week. 38, 60 miles to the east or 30 miles to the west is a storm with vastly more impact, but it is worth repeating; 38 flooded the hell out of RI, was deeper, had stronger winds, did horrific damage to the rural New England forests.... And if it happened again today, wouldn't match Sandy's damage. Sandy was like the Christchurch earthquake - not strong, but exquisitely well located.

[GaWx]

 Meanwhile, the Euro ensemble mean continues to predict the strongest amplitude phase 1/2 MJO by far for near the peak of the season since at least 1979, when David and Frederic formed. That is one strong MJO prediction!!

[HurricaneJosh]

The thing with 38 is that it missed all the areas that mattered. Did a number on RI, eastern LI, and then interior western New England. I went to school in NH surrounded by trees of the exact same size, which no one realized was because all of them had been planted in 1939....

...but 38 missed any population center of real consequence. Split the proverbial NY/BO'S uprights. If it happened today, some poor putzes in Westerly would have a hell of a storm, but 38 redux wouldn't shut down NYC for a week. 38, 60 miles to the east or 30 miles to the west is a storm with vastly more impact, but it is worth repeating; 38 flooded the hell out of RI, was deeper, had stronger winds, did horrific damage to the rural New England forests.... And if it happened again today, wouldn't match Sandy's damage. Sandy was like the Christchurch earthquake - not strong, but exquisitely well located.

 

I don't totally disagree with your take (which I know surprises you because we always seem to be butting heads ), but I feel-- like jconsor--  that you overstate the lack of population density in 1938's stricken region.  New London-- a sizable city in the cyclone's inner core-- was devastated.

 

A repeat of 1938 today would expose what are now densely-populated, vulnerable regions (with very expensive real estate) to Cat-3 (or near Cat-3) conditions, including NY's Hamptons and a dense string of medium-to-large-sized New England cities including New Haven, Hartford, New London, and Providence.  (Providence was probably just outside 1938's RMW-- which was ~40 nmi-- but it still got thrashed hard.)

 

I agree that the Sandy track is worse for damage because it puts NYC, NY Harbor, and all of Long Island to the right of the center-- but I think a lot of people are just not realizing how strong 1938 was.  It was a heavy-duty hurricane-- not just a big surge but a real wind core raking a dense, industrialized region that never sees winds of that speed.  While 1938 and Sandy had roughly similar landfall pressures, the comparison completely ends there in terms of structure and severity. Granted, 1938 was not 100% tropical, but it was close enough-- it had a core and eyewall and a fairly finite RMW, with sustained winds estimated at 105 kt.   (I know you know this, I'm just pointing it out for others.)

 

So, I maintain that a 1938 redux would be pretty shocking, and I think the level of devastation would be worse than Sandy.  Keep in mind, Sandy's winds only gusted to about 80 kt.  So, away from the immediate shore, the impact of the event was limited to downed trees and power outages.  1938 caused very heavy wind damage over a wide swath of inland New England-- not just trees, but structures.

[tmagan]

Based on everything I saw, the tree damage from Hurricane Sandy was more extensive in northern Nassau, NW Suffolk than either Irene or Gloria. I believe the reason was the length of storm force winds. Gloria was devastating for the trees in Heckscher State Park, one of the great parks around. 

[HurricaneJosh]

Based on everything I saw, the tree damage from Hurricane Sandy was more extensive in northern Nassau, NW Suffolk than either Irene or Gloria. I believe the reason was the length of storm force winds. Gloria was devastating for the trees in Heckscher State Park, one of the great parks around. 

 

One thing that stood out about Gloria-- for me, in Huntington-- was how fast it blew through.  Damaging winds lasted maybe an hour or two at the most, with a short burst of really strong winds-- that took down a large tree on our property-- just minutes before the calm.  That cyclone was really, seriously haulin' azz went it reached the Island.

[Sunny and Warm]

I don't totally disagree with your take (which I know surprises you because we always seem to be butting heads ), but I feel-- like jconsor--  that you overstate the lack of population density in 1938's stricken region.  New London-- a sizable city in the cyclone's inner core-- was devastated.

 

A repeat of 1938 today would expose what are now densely-populated, vulnerable regions (with very expensive real estate) to Cat-3 (or near Cat-3) conditions, including NY's Hamptons and a dense string of medium-to-large-sized New England cities including New Haven, Hartford, New London, and Providence.  (Providence was probably just outside 1938's RMW-- which was ~40 nmi-- but it still got thrashed hard.)

 

I agree that the Sandy track is worse for damage because it puts NYC, NY Harbor, and all of Long Island to the right of the center-- but I think a lot of people are just not realizing how strong 1938 was.  It was a heavy-duty hurricane-- not just a big surge but a real wind core raking a dense, industrialized region that never sees winds of that speed.  While 1938 and Sandy had roughly similar landfall pressures, the comparison completely ends there in terms of structure and severity. Granted, 1938 was not 100% tropical, but it was close enough-- it had a core and eyewall and a fairly finite RMW, with sustained winds estimated at 105 kt.   (I know you know this, I'm just pointing it out for others.)

 

So, I maintain that a 1938 redux would be pretty shocking, and I think the level of devastation would be worse than Sandy.  Keep in mind, Sandy's winds only gusted to about 80 kt.  So, away from the immediate shore, the impact of the event was limited to downed trees and power outages.  1938 caused very heavy wind damage over a wide swath of inland New England-- not just trees, but structures.

Josh,

 

I still think as I read this post that you're inserting severity of damage into the discussion when I'm simply talking about anomolous events.  We agree that 1938 is the more devastating event, but it would be my contention that the return frequency is greater for 1938 as well.  There is no question in my mind that 1938 repeat today would cause more damage than Sandy.

[Sunny and Warm]

We have a 10% lemon with no model support, and the new GFS shows a single TC the next 16 days, a fish at that.

 

 

I'd like to hear how Labor Day RI'd into a Cat 5 in one or two days myself.  And would chasers go to Monroe County to tape/video if a major was heading for the islands?  Would local law enforcement let them?

I've always wondered that myself.  I'd love to here the science of it.

[HurricaneJosh]

Josh,

 

I still think as I read this post that you're inserting severity of damage into the discussion when I'm simply talking about anomolous events.  We agree that 1938 is the more devastating event, but it would be my contention that the return frequency is greater for 1938 as well.  There is no question in my mind that 1938 repeat today would cause more damage than Sandy.

 

I guess I'm using the severity of the damage as one of the indicators of how anomalous 1938 was.  As the end of the day, it's hard for anyone to say definitively that one is more anomalous than the other.  I respect where you're coming from and don't want to fight-- so I'm cool to leave it at that.  I am sorry I suggested you're a "Sandy shill"-- I didn't mean it, of course.  

[HurricaneJosh]

We have a 10% lemon with no model support, and the new GFS shows a single TC the next 16 days, a fish at that.

 

 

I'd like to hear how Labor Day RI'd into a Cat 5 in one or two days myself.  And would chasers go to Monroe County to tape/video if a major was heading for the islands?  Would local law enforcement let them?

 

Wilma 2005 went from Cat 1 (75 kt) to Cat 5 (150 kt) in less than 12 hr.

[Sunny and Warm]

I guess I'm using the severity of the damage as one of the indicators of how anomalous 1938 was.  As the end of the day, it's hard for anyone to say definitively that one is more anomalous than the other.  I respect where you're coming from and don't want to fight-- so I'm cool to leave it at that.  I am sorry I suggested you're a "Sandy shill"-- I didn't mean it, of course.  

Not a problem Josh.  I know how deeply you feel about 1938, so expected the teeth marks.  I just enjoy interesting weather these days, and find Sandy to be interesting.  Besides, with all this discussion, we just worked through two days of the calendar toward a September to remember. 

[HurricaneJosh]

Not a problem Josh.  I know how deeply you feel about 1938, so expected the teeth marks.  I just enjoy interesting weather these days, and find Sandy to be interesting.  Besides, with all this discussion, we just worked through two days of the calendar toward a September to remember. 

 

 

[Ed Lizard]

I'm not encouraged by the models on anything through the Labor Day Weekend.

 

 

Euro is showing a non-tropical origin low trying to develop at the end of the Labor Day Weekend, but it was about this time last month I got excited about a Euro non-tropical origin low in a very similar location.  That wasn't a complete failure, we did get a decent mid level feature come back this way with enhanced rain.  But nothing resembling a TC.

 

But so far, the beaches of Amity should have a good holiday weekend.

[Ed Lizard]

 

 

I think I can see something resembling a circulation in reflectivity. about 100 miles due South BVE on MOB and LIX radars.  Not doing trig with cyclinders and all on what level the 88D is scanning, but staring at ADD Gulf visible loop I don't see a suggestion of a circulation in the low clouds.

 

Last CIMSS 700 mb vorticity might be showing a little enhanced vorticity on the North end of the 700 mb trough.

 

Side note, CIMSS showoing two "I" in the Atlantic with satellite imagery, I'm guessing they are targets for the NASA drone.

 

 

One is a low cloud swirl, the other a decent wave coming off Africa.

 

 

[wxmeddler]

Side note, CIMSS showoing two "I" in the Atlantic with satellite imagery, I'm guessing they are targets for the NASA drone.

I would assume so. They have a mission planned for the environment drone on Saturday and the over storm drone on Sunday.

[BrianLaverty]

The thing with 38 is that it missed all the areas that mattered. Did a number on RI, eastern LI, and then interior western New England. I went to school in NH surrounded by trees of the exact same size, which no one realized was because all of them had been planted in 1939....

 

 

You do realize that Rhode Island is the 2nd most densely populated state in the country, outside of NJ, right?  Massachusetts is 3rd & Connecticut is 4th

 

Your misconnception that its pretty much all trees and open land is pretty far off.  

 

The houses on the beaches of Westerly, Newport and Narragansett are some of the richest in the country.  The dollar damage alone that it would do to the Newport mansions would be astronomical.  

 

And much like New Jersey... stupid as it is, the Rhode Island is not built to sustain a category 3 hurricane.

 

For example.. look what the surge from Sandy did to Masquamicut in Westerly:     http://vimeo.com/52513169

 

As much as New Jersey  has been talked about, the RI coastline was altered big time from the hurricane.   Just imagine what it would do to Rhode Island if it hit dead on.  It would be unthinkable, and probably the worst we've ever seen.