Unpredictability of Hurricane Intensity Forecasts:

[ncforecaster89]

As we are all aware, intensity prediction is the most problematic aspect of hurricane forecasting. To highlight the difficulty, I decided to review the intensity forecasts provided by the NHC for all hurricane landfalls that have occurred on the U.S. mainland during the past 4 seasons; 13 separate hurricanes in total.

With that in mind, here’s a statistical listing of the intensity forecast errors by the NHC corresponding to the 72, 48, 24, and 12 hour lead times prior to landfall of all U.S. mainland hurricanes since 2017:

Hurricane...Landfall....T-72 .....T-48 hours...T-24.....T-12 hours

Zeta........115 mph....75 mph....80 mph....80 mph....100 mph
Delta.......100 mph...120 mph...100 mph...105 mph...110 mph
Sally........110 mph...100 mph....85 mph....85 mph.....80 mph
Laura.......150 mph...105 mph...110 mph...120 mph...150 mph
Isaias........90 mph.....70 mph....65 mph....70 mph.....75 mph
Hanna.......90 mph.....50 mph....60 mph....75 mph.....80 mph

Dorian......100 mph...100 mph....90 mph....100 mph....90 mph
Barry.........75 mph....90 mph.....80 mph....75 mph.....75 mph

Michael.....160 mph....90 mph....115 mph...125 mph...140 mph
Florence.....90 mph...130 mph....120 mph...105 mph....85 mph

Nate.........75 mph.....80 mph.....80 mph....90 mph....105 mph
Irma.........130 mph...140 mph...145 mph...140 mph....125 mph
Harvey......130 mph....75 mph.....75 mph...125 mph....120 mph

Based on this data, the average forecast errors were 28.5 mph (t-72), 24.2 mph (t-48), 16.2 mph (t-24), and 12.3 mph (t-12), at each interval, respectively. The most egregious were the way too conservative predictions for Michael and Harvey, and the overestimations for Florence at the 72 hour forecast points. More astonishing are the errors at less than 48 hours of >/= 25 mph for nine of the 13 (69.2%) hurricane landfalls during the past four seasons. Even at T-24 or less, the NHC errors were >/= 15 mph on eight of the 13 (61.5%) landfalling hurricanes!

These figures simply reflect the degree of difficulty that exists in such an inexact science. It shouldn’t be seen as a critique of the NHC forecasters’ relative expertise, as they are by far the very best in the field.

[ncforecaster89]

Amended the table and recalculated the forecast percentage errors following the official upgrade of hurricane Zeta.

With the Zeta reclassification, that makes six of the 13 (almost one-half) U.S. mainland hurricanes that the NHC missed by at least two full SSHWS categories...within 48 hours of their respective landfalls! That doesn’t even include the two borderline category one/category two hurricanes (Hanna and Isaias) that were forecast to be tropical storms at those lead times.

In all, 10 of the 13 (76.9%) hurricanes were mis-forecast by at least one SSHWS within 48 hours and 7 of 13 (53.8%) at t-12 hours! #intensityforecastsarehard

[tolakram]

I wonder how much improvement can be made? Intensity is so dependent on small scale factors we're going to nee a lot more data and get that data faster in order to improve much. In my amateur opinion anyway.

[ncforecaster89]

I wonder how much improvement can be made? Intensity is so dependent on small scale factors we're going to nee a lot more data and get that data faster in order to improve much. In my amateur opinion anyway.

I totally agree with you that intensity forecasting will very likely continue to be problematic with only relatively small improvements in the science. The main issue is that we’ll probably never fully understand, much less be able to accurately replicate in the models, all the microscale physical processes involved in a TCs internal dynamics. As a result, there’s only so much advancement that can be obtained...as you alluded to.

[ncforecaster89]

As time permits, I’m going to go back through the NHC advisory archives and the BT database to review the performance for all the other U.S. mainland hurricane landfalls dating back to 1995.

[ncforecaster89]

Since I’m currently unable to locate the advisory archive for the 1995-1997 seasons, I’ve chosen to begin with 1998. From here, I will review the NHC intensity forecast errors in the same format as the period of 2017-2020, but in five-year increments, instead.

1998-2002:

Hurricane...Landfall....T-72 .....T-48 hours...T-24.....T-12 hours

Lili............90 mph...120 mph...120 mph...125 mph...145 mph

Bret.........115 mph....80 mph....105 mph...140 mph...140 mph
Floyd........105 mph...150 mph...140 mph...140 mph...125 mph
Irene.........80 mph....N/A.........90 mph.....85 mph.....80 mph

Bonnie......110 mph...125 mph....125 mph...115 mph...115 mph
Earl...........80 mph....N/A..........75 mph....75 mph...105 mph
Georges.....105 mph...115 mph...100 mph....90 mph....105 mph
Georges.....105 mph...115mph....115 mph...115 mph...110 mph

Avg. error...N/A......24.2 mph...15.0 mph...16.9 mph...16.9 mph

This five-year period saw the U.S. mainland endure 8 separate hurricane landfalls by 7 individual hurricanes. Hurricane Georges (1998) struck the Florida Keys prior to its second landfall on the northern Gulf Coast...near Gulfport, MS.

The seven individual hurricane landfalls was well below the 13 hurricanes that crossed the U.S. coastline during the four year time span we examined previously. Of those 7, only one officially made a direct landfall at major hurricane (MH) intensity...despite five of those seven (71.4%) forecast to slam the shore as a MH within 24 hours (T-24) of actually doing so! More astoundingly, three of the five were expected to be strong category-four landfalls.

In comparison with the most recent four-year period, the forecast errors were as follows for each respective interval:

T-72 = - 4.3
T-48 = - 9.2
T-24 = + 0.7
T-12 = + 4.3

Overall, these figures show that the NHC did better two decades ago in the long range, but have improved in the short range, comparatively.

[Audrey2Katrina]

Amended the table and recalculated the forecast percentage errors following the official upgrade of hurricane Zeta.

With the Zeta reclassification, that makes six of the 13 (almost one-half) U.S. mainland hurricanes that the NHC missed by at least two full SSHWS categories...within 48 hours of their respective landfalls! That doesn’t even include the two borderline category one/category two hurricanes (Hanna and Isaias) that were forecast to be tropical storms at those lead times.

In all, 10 of the 13 (76.9%) hurricanes were mis-forecast by at least one SSHWS within 48 hours and 7 of 13 (53.8%) at t-12 hours! #intensityforecastsarehard

As one who went through Zeta completely (eye passed directly over--I remember that 10-15 minute lull - then winds from opposite direction) I gotta believe they'll revisit it one day too-- I may not even be around (after all they reclassified Audrey over 50 years later) but you'll NEVER convince me that storm (Zeta) was a Cat 3--and I can assure you it wasn't a Cat 2 passing over New Orleans. I have many friends who live down near Cocodrie who rode it out, the storm was just not as powerful as they said it was. I am certain that we never even had strong Cat 1 winds WITH the eye passing over us, and I have a picture from my TV set (never lost power--and I know most people in the area DID lose it, but for some reason, and I thank God, I never lost the power) I watched as they showed all the highest sustained and highest gusts on TV,... not a single GUST shown was over Cat 1 status and the NHC report itself states that the highest sustained winds after landfall were in Golden Meadow, I believe, and those were CAT ! winds. This was no Cat 3 storm--or if they're going to say it was, then they need to re-re-evaluate Audrey which was WAY worse than this thing.

A2K

[AnnularCane]

As one who went through Zeta completely (eye passed directly over--I remember that 10-15 minute lull - then winds from opposite direction) I gotta believe they'll revisit it one day too-- I may not even be around (after all they reclassified Audrey over 50 years later) but you'll NEVER convince me that storm (Zeta) was a Cat 3--and I can assure you it wasn't a Cat 2 passing over New Orleans. I have many friends who live down near Cocodrie who rode it out, the storm was just not as powerful as they said it was. I am certain that we never even had strong Cat 1 winds WITH the eye passing over us, and I have a picture from my TV set (never lost power--and I know most people in the area DID lose it, but for some reason, and I thank God, I never lost the power) I watched as they showed all the highest sustained and highest gusts on TV,... not a single GUST shown was over Cat 1 status and the NHC report itself states that the highest sustained winds after landfall were in Golden Meadow, I believe, and those were CAT ! winds. This was no Cat 3 storm--or if they're going to say it was, then they need to re-re-evaluate Audrey which was WAY worse than this thing.

A2K

That's very interesting. There were a couple of things about Zeta that have continued to puzzle me to this day. Granted, I don't live there anymore and all my information is secondhand at best, but from what I heard and read about, everything seemed a bit less dramatic than you might expect from a presumably major hurricane directly hitting the NOLA area. I think I might have expected considerably more flooding and damage. Also I don't recall there being any evacuation of the city. Granted they did strengthen the levees, and I guess Zeta wasn't expected to hit as a major, but you might still think they'd do it just to be on the safe side. My aunt in Ponchatoula said she never lost power either, although my dad in Abita Springs did (he said it got a little scary for about 30 minutes but other than that it wasn't too bad). I don't know, the whole thing just seemed somewhat undramatic in some ways...although there were some cool videos of the eye passing over. But again, maybe I just don't know enough about what went on since I wasn't actually there.

[CrazyC83]

Regarding Zeta, it was a much smaller storm and also very fast moving. The category 3 winds only took place in a very small area around Port Fourchon. Even in New Orleans, the winds were probably at category 1 level and those lasted maybe 15-30 minutes. Almost had a derecho-like feel. The stronger winds would have been to the east of the center - perhaps in St. Bernard and parts of Plaquemines.

Gulfport had a higher wind report (both sustained and gust) than any New Orleans-area station.

A small, fast-moving category 3 is also considerably different than a large, slow-moving category 3. In fact, Isaac had much higher storm surges than Zeta despite being barely a category 1, since it was much larger and crawling.

[AnnularCane]

Regarding Zeta, it was a much smaller storm and also very fast moving. The category 3 winds only took place in a very small area around Port Fourchon. Even in New Orleans, the winds were probably at category 1 level and those lasted maybe 15-30 minutes. Almost had a derecho-like feel. The stronger winds would have been to the east of the center - perhaps in St. Bernard and parts of Plaquemines.

Gulfport had a higher wind report (both sustained and gust) than any New Orleans-area station.

A small, fast-moving category 3 is also considerably different than a large, slow-moving category 3. In fact, Isaac had much higher storm surges than Zeta despite being barely a category 1, since it was much larger and crawling.

Oh yes, I guess that could explain it. I remember the fast movement but I always thought she was larger.

[Hurricaneman]

There probably will not be a nailed forecast on intensity as any small thing can disrupt or cause RI and here are some examples of initial forecast compared to what happened

Harvey: initially expected to be 60mph at Texas landfall but was 140mph

Joaquin: was expected to only peak as a depression but was 155mph near the Bahamas

Michael: was initially expected to be a 65mph Ts in Panama City but was 160mph

Dorian: was initially expected to be a 45mph TS in the Bahamas but was 180mph

Hanna: was initially expected to be a 50mph tS but was 100mph

And many more from 2020

So as we can see intensity is hard to forecast and can be missed by a country mile

[chaser1]

Since I’m currently unable to locate the advisory archive for the 1995-1997 seasons, I’ve chosen to begin with 1998. From here, I will review the NHC intensity forecast errors in the same format as the period of 2017-2020, but in five-year increments, instead.

1998-2002:

Hurricane...Landfall....T-72 .....T-48 hours...T-24.....T-12 hours

Lili............90 mph...120 mph...120 mph...125 mph...145 mph

Bret.........115 mph....80 mph....105 mph...140 mph...140 mph
Floyd........105 mph...150 mph...140 mph...140 mph...125 mph
Irene.........80 mph....N/A.........90 mph.....85 mph.....80 mph

Bonnie......110 mph...125 mph....125 mph...115 mph...115 mph
Earl...........80 mph....N/A..........75 mph....75 mph...105 mph
Georges.....105 mph...115 mph...100 mph....90 mph....105 mph
Georges.....105 mph...115mph....115 mph...115 mph...110 mph

Avg. error...N/A......24.2 mph...15.0 mph...16.9 mph...16.9 mph

This five-year period saw the U.S. mainland endure 8 separate hurricane landfalls by 7 individual hurricanes. Hurricane Georges (1998) struck the Florida Keys prior to its second landfall on the northern Gulf Coast...near Gulfport, MS.

The seven individual hurricane landfalls was well below the 13 hurricanes that crossed the U.S. coastline during the four year time span we examined previously. Of those 7, only one officially made a direct landfall at major hurricane (MH) intensity...despite five of those seven (71.4%) forecast to slam the shore as a MH within 24 hours (T-24) of actually doing so! More astoundingly, three of the five were expected to be strong category-four landfalls.

In comparison with the most recent four-year period, the forecast errors were as follows for each respective interval:

T-72 = - 4.3
T-48 = - 9.2
T-24 = + 0.7
T-12 = + 4.3

Overall, these figures show that the NHC did better two decades ago in the long range, but have improved in the short range, comparatively.

This suggests to me that the result in error must be attributed to one of two things. A "fall-off" in intensity forecasting..... OR intensity forecasts which err on the side of heightened alerts and "over-warning" so to mitigate the potential for public apathy and threat of greater harm. At least with regard to Florida Watches, Warnings, and actual severe WX verification, I feel this follows a fairly consistent pattern by both NHC and NWS offices with regard to most severe weather threats (including tornadic events and potentially severe thunderstorms). It is interestingly rare that T-72 down to T-12 hr forecasts actually "under-forecast" hurricane intensity (as shown in the above research done).

[ncforecaster89]

Amended the table and recalculated the forecast percentage errors following the official upgrade of hurricane Zeta.

With the Zeta reclassification, that makes six of the 13 (almost one-half) U.S. mainland hurricanes that the NHC missed by at least two full SSHWS categories...within 48 hours of their respective landfalls! That doesn’t even include the two borderline category one/category two hurricanes (Hanna and Isaias) that were forecast to be tropical storms at those lead times.

In all, 10 of the 13 (76.9%) hurricanes were mis-forecast by at least one SSHWS within 48 hours and 7 of 13 (53.8%) at t-12 hours! #intensityforecastsarehard

As one who went through Zeta completely (eye passed directly over--I remember that 10-15 minute lull - then winds from opposite direction) I gotta believe they'll revisit it one day too-- I may not even be around (after all they reclassified Audrey over 50 years later) but you'll NEVER convince me that storm (Zeta) was a Cat 3--and I can assure you it wasn't a Cat 2 passing over New Orleans. I have many friends who live down near Cocodrie who rode it out, the storm was just not as powerful as they said it was. I am certain that we never even had strong Cat 1 winds WITH the eye passing over us, and I have a picture from my TV set (never lost power--and I know most people in the area DID lose it, but for some reason, and I thank God, I never lost the power) I watched as they showed all the highest sustained and highest gusts on TV,... not a single GUST shown was over Cat 1 status and the NHC report itself states that the highest sustained winds after landfall were in Golden Meadow, I believe, and those were CAT ! winds. This was no Cat 3 storm--or if they're going to say it was, then they need to re-re-evaluate Audrey which was WAY worse than this thing.

A2K

Hi A2K! I relocated from the NE eyewall near Jesuit Bend, La and drove NNW through the eyewall into New Orleans as the peak was hitting the city. Like you, I wasn’t that impressed, relatively speaking, as the winds were likely no greater than 90 mph at the most. That said, the winds were definitely exceeding 100 mph (in gusts) as I drove back NNE over Lake Ponchartrain towards the Mississippi coast. Those winds out over the water and on the bridge were possibly the most extreme velocity I have ever encountered in a vehicle.

As Crazy stated, the Cat 3 winds of Zeta were only felt in a very small area right at the coast to the E of where the eye crossed the shoreline. Although the winds were still of Cat 2 strength as Zeta moved directly through the city of New Orleans, the stronger winds were still well to the east as the eye pushed towards the MS coastline.

I will add that Zeta was one of the weaker, if not the weakest, borderline Cat 2/3 eyewall experiences of my chase career (dating back to 1989). In contrast, Sally provided the most extreme eyewall experiences of my career for a borderline Cat 2/3 encounter.

[ncforecaster89]

Since I’m currently unable to locate the advisory archive for the 1995-1997 seasons, I’ve chosen to begin with 1998. From here, I will review the NHC intensity forecast errors in the same format as the period of 2017-2020, but in five-year increments, instead.

1998-2002:

Hurricane...Landfall....T-72 .....T-48 hours...T-24.....T-12 hours

Lili............90 mph...120 mph...120 mph...125 mph...145 mph

Bret.........115 mph....80 mph....105 mph...140 mph...140 mph
Floyd........105 mph...150 mph...140 mph...140 mph...125 mph
Irene.........80 mph....N/A.........90 mph.....85 mph.....80 mph

Bonnie......110 mph...125 mph....125 mph...115 mph...115 mph
Earl...........80 mph....N/A..........75 mph....75 mph...105 mph
Georges.....105 mph...115 mph...100 mph....90 mph....105 mph
Georges.....105 mph...115mph....115 mph...115 mph...110 mph

Avg. error...N/A......24.2 mph...15.0 mph...16.9 mph...16.9 mph

This five-year period saw the U.S. mainland endure 8 separate hurricane landfalls by 7 individual hurricanes. Hurricane Georges (1998) struck the Florida Keys prior to its second landfall on the northern Gulf Coast...near Gulfport, MS.

The seven individual hurricane landfalls was well below the 13 hurricanes that crossed the U.S. coastline during the four year time span we examined previously. Of those 7, only one officially made a direct landfall at major hurricane (MH) intensity...despite five of those seven (71.4%) forecast to slam the shore as a MH within 24 hours (T-24) of actually doing so! More astoundingly, three of the five were expected to be strong category-four landfalls.

In comparison with the most recent four-year period, the forecast errors were as follows for each respective interval:

T-72 = - 4.3
T-48 = - 9.2
T-24 = + 0.7
T-12 = + 4.3

Overall, these figures show that the NHC did better two decades ago in the long range, but have improved in the short range, comparatively.

This suggests to me that the result in error must be attributed to one of two things. A "fall-off" in intensity forecasting..... OR intensity forecasts which err on the side of heightened alerts and "over-warning" so to mitigate the potential for public apathy and threat of greater harm. At least with regard to Florida Watches, Warnings, and actual severe WX verification, I feel this follows a fairly consistent pattern by both NHC and NWS offices with regard to most severe weather threats (including tornadic events and potentially severe thunderstorms). It is interestingly rare that T-72 down to T-12 hr forecasts actually "under-forecast" hurricane intensity (as shown in the above research done).

Hi Chaser! Honestly, I wouldn’t necessarily say that there has been a marked “fall-off” in intensity forecasting in so much as there just hasn’t been any significant improvement in the science or its predictability.

In comparing the period of 1998-2002 with the most recent four-year time frame, the NHC actually under-forecast a full 8 of the 13 hurricane landfalls within less than 24 hours prior to the eye crossing the coast. As you noted, they actually did the opposite, by over-forecasting intensity, during the former time interval.

[Iceresistance]

As we are all aware, intensity prediction is the most problematic aspect of hurricane forecasting. To highlight the difficulty, I decided to review the intensity forecasts provided by the NHC for all hurricane landfalls that have occurred on the U.S. mainland during the past 4 seasons; 13 separate hurricanes in total.

With that in mind, here’s a statistical listing of the intensity forecast errors by the NHC corresponding to the 72, 48, 24, and 12 hour lead times prior to landfall of all U.S. mainland hurricanes since 2017:

Hurricane...Landfall....T-72 .....T-48 hours...T-24.....T-12 hours

Zeta........115 mph....75 mph....80 mph....80 mph....100 mph
Delta.......100 mph...120 mph...100 mph...105 mph...110 mph
Sally........110 mph...100 mph....85 mph....85 mph.....80 mph
Laura.......150 mph...105 mph...110 mph...120 mph...150 mph
Isaias........90 mph.....70 mph....65 mph....70 mph.....75 mph
Hanna.......90 mph.....50 mph....60 mph....75 mph.....80 mph

Dorian......100 mph...100 mph....90 mph....100 mph....90 mph
Barry.........75 mph....90 mph.....80 mph....75 mph.....75 mph

Michael.....160 mph....90 mph....115 mph...125 mph...140 mph
Florence.....90 mph...130 mph....120 mph...105 mph....85 mph

Nate.........75 mph.....80 mph.....80 mph....90 mph....105 mph
Irma.........130 mph...140 mph...145 mph...140 mph....125 mph
Harvey......130 mph....75 mph.....75 mph...125 mph....120 mph

Based on this data, the average forecast errors were 28.5 mph (t-72), 24.2 mph (t-48), 16.2 mph (t-24), and 12.3 mph (t-12), at each interval, respectively. The most egregious were the way too conservative predictions for Michael and Harvey, and the overestimations for Florence at the 72 hour forecast points. More astonishing are the errors at less than 48 hours of >/= 25 mph for nine of the 13 (69.2%) hurricane landfalls during the past four seasons. Even at T-24 or less, the NHC errors were >/= 15 mph on eight of the 13 (61.5%) landfalling hurricanes!

These figures simply reflect the degree of difficulty that exists in such an inexact science. It shouldn’t be seen as a critique of the NHC forecasters’ relative expertise, as they are by far the very best in the field.

One big issue is that the storms can defy forecasts! Dorian in 2019 was a very scary example (I called it "The Defying Dorian that will never Die!" for that reason) . . .

There has benn also other cases that the storms can rapidly fluctuate, either RI/VRI/ERI or RW/VRW/ERW (Rapid Weakening/Very Rapid Weakening/Extremely Rapid Weakening)

RI examples that I can think of: Imelda 2019, Lorenzo 2019 (First Peak Intensity) . . .
VRI examples: Eta 2020, Gamma 2020 . . .
ERI Examples: Wilma 2005, Matthew 2016, Florence 2018 (First time as a CAT 4), Delta 2020, Lorenzo 2019 (Secondary Peak Intensity) . . .

RW example: Florence 2018 at the North Carolina Landfall . . .

Also, it seems like that RI is more common than RW (Rapid Weakening) . . .

[Audrey2Katrina]

Regarding Zeta, it was a much smaller storm and also very fast moving. The category 3 winds only took place in a very small area around Port Fourchon. Even in New Orleans, the winds were probably at category 1 level and those lasted maybe 15-30 minutes. Almost had a derecho-like feel. The stronger winds would have been to the east of the center - perhaps in St. Bernard and parts of Plaquemines.

Gulfport had a higher wind report (both sustained and gust) than any New Orleans-area station.

A small, fast-moving category 3 is also considerably different than a large, slow-moving category 3. In fact, Isaac had much higher storm surges than Zeta despite being barely a category 1, since it was much larger and crawling.

Yes, and I took into account the "fast movement" which could account for a lesser surge and lesser wind damages as it passed over more quickly; but the fact remains that not a single instrument including those of the NHC, measured anything remotely as sustained cat 1 anywhere in New Orleans, gusts, yes, sustained, NO; and I have family in St. Bernard... they didn't have any winds (sustained) at Cat 1 either. Gulfport did have Cat 1 sustained by one report, but the official NHC report which was replete with "estimates" and other vagaries, (and given the dearth of reliable available data, I can certainly understand that) they still couldn't find so much as a Cat 2 sustained wind anywhere. (I'm NOT denying that it was a Cat 2... I'm only saying no station reported any such winds--because they didn't, and I can see areas they would have missed creeping into Cat 2 bursts---but never Cat 3.

In the NHC report they said "The strongest wind report received from a near-standard height was from the public in Golden Meadow, Louisiana at 2139 UTC - a sustained wind of 82 kt." That's high-end Category 1 and they only drop from there. 94 mph, you might argue is "close enough" to Cat 2 but it is NOT Cat 2... which is 96 mph. And this is also why I'm certain it was every bit of a 2--for a very short period close to the coast, as Golden Meadow is. Yet that was the highest recorded anywhere! The only reference made to Port Fourchon was also somewhat indefinite and understandably containing much speculation: "There were few strong wind reports at the immediate coast in southeastern Louisiana since the area that experienced Zeta’s peak winds (near and west of Port Fourchon in the East, which included the Timbalier Wildlife Refuge, is relatively unpopulated." Hence we rely on estimates--which I respect and equally respectfully feel over-rated the top winds.

This is all well and good for the claim that it was almost certainly a Cat 2 at landfall; but I reiterate, the people I know who were very close down there, said there were only about 70 to 80 mph winds tops. Cocodrie is VERY near the coast. As far as points east, as I said, no place in St. Bernard registered even sustained Cat 1, and yes Gulfport did register some higher winds... but the highest "sustained" wind was still only Cat 1, and not high-end Cat 1 at that. All the while the TV reports kept insisting it had winds of 110 mph... if anything it's forward speed might have shortened the duration of top winds; but not lessened them, neither would its size--these have bearing on impact, but not on top winds--if anything along the east wall where winds were coming from the south, the forward speed might actually have added to those speeds, making them greater. And yet the report says this about Mississippi: "In Mississippi, the strongest sustained wind report received was from Gulfport where a WeatherFlow station measured a sustained wind of 74 kt." That isn't remotely Cat 2. Hence all these reports of 110 mph winds well into Mississippi just did not happen from all the evidence I saw and see.

This is no criticism of the hard and fine work of the folks at the NHC, they've had to go back and re-evaluate a lot of storms. I'm just convinced this will be yet another. I do NOT believe it could have been a Cat 3 and there is no substantive evidence that it was a Cat 2--anywhere, yet I will concede readily that it was close enough to where some coastal areas almost certainly DID get some Cat 2 winds; but those died down rapidly as it approached the New Orleans area where NO category two winds were anywhere to be found--not even close. I reiterate that its forward speed would not have had any bearing on windspeed measurements other than to make them stronger on the east side as it moved quickly north-north east, and lesser on the west side. I caught it practically down the middle. I filmed the winds just before during and just after the eye passed. It wasn't Cat 1 sustained though there were certainly gusts that reached it. And I'm not ABOUT to second guess the NHC either; I'm offering my opinion from what I observed and noted first hand and directly from other first hand observers and I can honestly state that I feel the day will come when they realize this was no Cat 3 storm, and none of the televised claims of 110 mph sustained winds ANYWHERE once it was inland-- were accurate.

A2K

[Shell Mound]

Amended the table and recalculated the forecast percentage errors following the official upgrade of hurricane Zeta.

With the Zeta reclassification, that makes six of the 13 (almost one-half) U.S. mainland hurricanes that the NHC missed by at least two full SSHWS categories...within 48 hours of their respective landfalls! That doesn’t even include the two borderline category one/category two hurricanes (Hanna and Isaias) that were forecast to be tropical storms at those lead times.

In all, 10 of the 13 (76.9%) hurricanes were mis-forecast by at least one SSHWS within 48 hours and 7 of 13 (53.8%) at t-12 hours! #intensityforecastsarehard

As one who went through Zeta completely (eye passed directly over--I remember that 10-15 minute lull - then winds from opposite direction) I gotta believe they'll revisit it one day too-- I may not even be around (after all they reclassified Audrey over 50 years later) but you'll NEVER convince me that storm (Zeta) was a Cat 3--and I can assure you it wasn't a Cat 2 passing over New Orleans. I have many friends who live down near Cocodrie who rode it out, the storm was just not as powerful as they said it was. I am certain that we never even had strong Cat 1 winds WITH the eye passing over us, and I have a picture from my TV set (never lost power--and I know most people in the area DID lose it, but for some reason, and I thank God, I never lost the power) I watched as they showed all the highest sustained and highest gusts on TV,... not a single GUST shown was over Cat 1 status and the NHC report itself states that the highest sustained winds after landfall were in Golden Meadow, I believe, and those were CAT ! winds. This was no Cat 3 storm--or if they're going to say it was, then they need to re-re-evaluate Audrey which was WAY worse than this thing.

A2K

Hi A2K! I relocated from the NE eyewall near Jesuit Bend, La and drove NNW through the eyewall into New Orleans as the peak was hitting the city. Like you, I wasn’t that impressed, relatively speaking, as the winds were likely no greater than 90 mph at the most. That said, the winds were definitely exceeding 100 mph (in gusts) as I drove back NNE over Lake Ponchartrain towards the Mississippi coast. Those winds out over the water and on the bridge were possibly the most extreme velocity I have ever encountered in a vehicle.

As Crazy stated, the Cat 3 winds of Zeta were only felt in a very small area right at the coast to the E of where the eye crossed the shoreline. Although the winds were still of Cat 2 strength as Zeta moved directly through the city of New Orleans, the stronger winds were still well to the east as the eye pushed towards the MS coastline.

I will add that Zeta was one of the weaker, if not the weakest, borderline Cat 2/3 eyewall experiences of my chase career (dating back to 1989). In contrast, Sally provided the most extreme eyewall experiences of my career for a borderline Cat 2/3 encounter.

Officially, Zeta made landfall at 29.2°N 90.6°W at 21:00 UTC on 28 October 2021. Based on CIMSS’s SATCON, the peak MSW derived from a consensus of satellite estimates was 96 kt at 21:20 UTC on 28 October, right around the time of landfall. For comparison: Sally made landfall at 30.3°N 87.7°W at 09:45 UTC on 16 September 2021. Peak SATCON-derived estimates of the MSW were 104 kt at 06:50 UTC on 16 September. Estimates around the time of landfall averaged 99 kt. So it is interesting to note that satellite estimates provided more support for Sally being a MH than Zeta. SATCON alone supported a MSW of 95 kt at LF for Zeta and 100 kt at LF for Sally, with Sally actually having peaked at 105 kt several hours before LF, while offshore. However, note that the observation of 98 kt from Ingram Bayou, AL, during Sally was inside the RMW, did experience the strongest (easterly) quadrant (the eye made landfall just to the west, near Gulf Shores), and had proper marine exposure along a bayou while sited at 18 m (standard elevation is 10 m). So in that case the observation may well have been representative of the peak winds in Sally and supported 95 kt at LF. Regardless, SATCON for Zeta was clearly supporting a higher MSW at landfall than local surface observations suggested, given the negligible difference between 95 and 100 kt.

[Audrey2Katrina]

I concur with Shell Mound, and to some degree ncforecaster89 (excepting this part: "Although the winds were still of Cat 2 strength as Zeta moved directly through the city of New Orleans,", as I found NO evidence of Cat 2 winds anywhere in or through New Orleans, not from the weather bureau, not from any televised reporting station. That said I suppose there might have been isolated gusts barely reaching that level, however we bandy about terms like Cat 2 or Cat 3 too much without distinguishing "gusts" from "sustained" and I can assure you New Orleans had NO sustained Cat 2 winds. Given that ncforecaster89 brings a valid point about travelling over the water. I hadn't considered winds over water where there would be considerably less friction from the land. That said, I also agree that from every indication I saw, Sally looked more the viable candidate as a MH than Zeta. I forgot to mention that the report stated landfall for Zeta was near Cocodrie, which is where at least two of my friends were--and they still say no sustained winds of 110 mph or greater hit them. They DID say the surf was (obviously) rough... and the winds "in gusts" probably reached over 110, but having been through genuine Cat 3's before, this one was tame and that's what worries me. People will think that if that was a Cat 3, and they did just fine going through it-- they might think a bigger more powerful Cat 2 is nothing to worry about--when it is! And a bigger more powerful Cat 3 can be devastating along coastal areas. Yes, the winds were obviously around Cat 2 "In Gusts", but storms are classified by sustained winds and not gusts. I was glad to see that at least on my television they made that distinction. I can't post the picture, obviously, but I have one of all the reported highest "gusts" and few of them are mid to high Cat 1, and only Gulfport showed minimal Cat 2.

Zeta was such a non-event for the vast majority down here that I truly worry the young, in particular, will underestimate the next Cat 3 that is a REAL Cat 3 and much bigger, and that could be a fatal mistake. On the other hand, a friend of mine in Gulf Shores who rode out Sally, said that was worse than he remembered from Betsy which we both rode out in our homes back in '65. But I think his memory is fading as I remember seeing Betsy blow a roof off of garage like it was a toy. Ahhh, I digress; but my main point is that Zeta was borderline Cat 2/3 but I don't believe it was a bonafide Cat 3... that's my opinion. We may well talk of not having much better science than what we now have; however I'm sure the same was said of the science we had in 1970. People tend to think they live in a technology that has reached its zenith, --that is until the next level shows up and is a quantum leap forward. I hope so, for the sake of people caught in their paths, as I had a family member killed by Katrina. (He was a first responder.) These monsters (yes even Zeta types) need all the respect we can give them. Tropical Storms can be killers and this was, to be sure, much worse than a tropical storm. They may be intriguing and even fun to view, plot, follow, and discuss; but they are, in the final analysis, very dangerous. I hold hope for improvement as better still in years to come, in forecasting and analyzing these killers for the sake of humanity.

[Shell Mound]

I concur with Shell Mound, and to some degree ncforecaster89 (excepting this part: "Although the winds were still of Cat 2 strength as Zeta moved directly through the city of New Orleans,", as I found NO evidence of Cat 2 winds anywhere in or through New Orleans, not from the weather bureau, not from any televised reporting station. That said I suppose there might have been isolated gusts barely reaching that level, however we bandy about terms like Cat 2 or Cat 3 too much without distinguishing "gusts" from "sustained" and I can assure you New Orleans had NO sustained Cat 2 winds. Given that ncforecaster89 brings a valid point about travelling over the water. I hadn't considered winds over water where there would be considerably less friction from the land. That said, I also agree that from every indication I saw, Sally looked more the viable candidate as a MH than Zeta. I forgot to mention that the report stated landfall for Zeta was near Cocodrie, which is where at least two of my friends were--and they still say no sustained winds of 110 mph or greater hit them. They DID say the surf was (obviously) rough... and the winds "in gusts" probably reached over 110, but having been through genuine Cat 3's before, this one was tame and that's what worries me. People will think that if that was a Cat 3, and they did just fine going through it-- they might think a bigger more powerful Cat 2 is nothing to worry about--when it is! And a bigger more powerful Cat 3 can be devastating along coastal areas. Yes, the winds were obviously around Cat 2 "In Gusts", but storms are classified by sustained winds and not gusts. I was glad to see that at least on my television they made that distinction. I can't post the picture, obviously, but I have one of all the reported highest "gusts" and few of them are mid to high Cat 1, and only Gulfport showed minimal Cat 2.

Zeta was such a non-event for the vast majority down here that I truly worry the young, in particular, will underestimate the next Cat 3 that is a REAL Cat 3 and much bigger, and that could be a fatal mistake. On the other hand, a friend of mine in Gulf Shores who rode out Sally, said that was worse than he remembered from Betsy which we both rode out in our homes back in '65. But I think his memory is fading as I remember seeing Betsy blow a roof off of garage like it was a toy. Ahhh, I digress; but my main point is that Zeta was borderline Cat 2/3 but I don't believe it was a bonafide Cat 3... that's my opinion. We may well talk of not having much better science than what we now have; however I'm sure the same was said of the science we had in 1970. People tend to think they live in a technology that has reached its zenith, --that is until the next level shows up and is a quantum leap forward. I hope so, for the sake of people caught in their paths, as I had a family member killed by Katrina. (He was a first responder.) These monsters (yes even Zeta types) need all the respect we can give them. Tropical Storms can be killers and this was, to be sure, much worse than a tropical storm. They may be intriguing and even fun to view, plot, follow, and discuss; but they are, in the final analysis, very dangerous. I hold hope for improvement as better still in years to come, in forecasting and analyzing these killers for the sake of humanity.

On another note, as far as Audrey (1957) is concerned, the Best Track Committee officially determined that Audrey exhibited double wind maxima during landfall: an outer RMW of ~30 n mi and an inner of ~15 n mi. Radar fixes from aircraft suggested that landfall occurred at 13:30 UTC on 27 June 1957. Based on this map, the outer RMW was responsible for a peak storm tide of 13.9 ft at the mouth of the Mermentau River, along Lower Mud Lake near Grand Chenier. Based on complete records of surface weather data from Lake Charles and Port Arthur, TX, the location of the landfall can be calculated rather accurately, given that aircraft and radar were also available, along with records of wind, MSLP, and storm tide. So despite being embedded in low environmental pressures, Audrey clearly retained a rather small inner core at landfall, along with an outer, or concentric, eyewall. It is noted that a report from the National Hurricane Research Project (NHRP) estimated a possible MSLP in Audrey ranging from 919 to 960 mb at landfall, with the best estimate calculated to be 946 mb, but there seems to have been a rather large degree of uncertainty involved. A ship reported an unconfirmed reading of 924.5 mb and four oil-tender barges, each equipped with an anemometer (height: 65 ft / ~20 m), recorded peak gusts ranging from 104 to 130 kt, with two of the barges separately registering peak gusts of 126 kt. The positions of these barges and the ship relative to the centre are not known with precision, but the NHRP report notes that the barges were adrift to the southeast of Cameron, so they would have experienced the outer RMW, if not the inner, on the eastern side of the storm. Note that the inner RMW would have likely been stronger than the outer, so the barges might have just missed the peak winds in the strongest quadrant. On the western side, the U.S. Coast Guard Station at Sabine, TX, recorded a sustained wind of 90 kt from the southwest at 15:00 UTC, the station having full marine exposure at the standard (10-m) elevation. This is considerably higher than the isoline charts in the NHRP report show for this location. Taken together, the evidence suggests that Audrey may need to be revisited in the future and potentially upgraded back to Cat-4 status, at the very least

[ncforecaster89]

I concur with Shell Mound, and to some degree ncforecaster89 (excepting this part: "Although the winds were still of Cat 2 strength as Zeta moved directly through the city of New Orleans,", as I found NO evidence of Cat 2 winds anywhere in or through New Orleans, not from the weather bureau, not from any televised reporting station. That said I suppose there might have been isolated gusts barely reaching that level, however we bandy about terms like Cat 2 or Cat 3 too much without distinguishing "gusts" from "sustained" and I can assure you New Orleans had NO sustained Cat 2 winds. Given that ncforecaster89 brings a valid point about travelling over the water. I hadn't considered winds over water where there would be considerably less friction from the land. That said, I also agree that from every indication I saw, Sally looked more the viable candidate as a MH than Zeta. I forgot to mention that the report stated landfall for Zeta was near Cocodrie, which is where at least two of my friends were--and they still say no sustained winds of 110 mph or greater hit them. They DID say the surf was (obviously) rough... and the winds "in gusts" probably reached over 110, but having been through genuine Cat 3's before, this one was tame and that's what worries me. People will think that if that was a Cat 3, and they did just fine going through it-- they might think a bigger more powerful Cat 2 is nothing to worry about--when it is! And a bigger more powerful Cat 3 can be devastating along coastal areas. Yes, the winds were obviously around Cat 2 "In Gusts", but storms are classified by sustained winds and not gusts. I was glad to see that at least on my television they made that distinction. I can't post the picture, obviously, but I have one of all the reported highest "gusts" and few of them are mid to high Cat 1, and only Gulfport showed minimal Cat 2.

Zeta was such a non-event for the vast majority down here that I truly worry the young, in particular, will underestimate the next Cat 3 that is a REAL Cat 3 and much bigger, and that could be a fatal mistake. On the other hand, a friend of mine in Gulf Shores who rode out Sally, said that was worse than he remembered from Betsy which we both rode out in our homes back in '65. But I think his memory is fading as I remember seeing Betsy blow a roof off of garage like it was a toy. Ahhh, I digress; but my main point is that Zeta was borderline Cat 2/3 but I don't believe it was a bonafide Cat 3... that's my opinion. We may well talk of not having much better science than what we now have; however I'm sure the same was said of the science we had in 1970. People tend to think they live in a technology that has reached its zenith, --that is until the next level shows up and is a quantum leap forward. I hope so, for the sake of people caught in their paths, as I had a family member killed by Katrina. (He was a first responder.) These monsters (yes even Zeta types) need all the respect we can give them. Tropical Storms can be killers and this was, to be sure, much worse than a tropical storm. They may be intriguing and even fun to view, plot, follow, and discuss; but they are, in the final analysis, very dangerous. I hold hope for improvement as better still in years to come, in forecasting and analyzing these killers for the sake of humanity.

I agree with your post, but need to clarify that I didn’t mean to imply that Zeta brought Cat 2 winds to New Orleans, as it moved directly over the city. Those Cat two winds were certainly just to the E of the city. I’d estimate that no part of the city experienced wind gusts even reaching 100 mph. To your point, N.O. likely had a Cat 1 hurricane impact, although (officially) it will go down as a direct Cat 2 hit on the city.