Discussion of Intense Tropical Cyclones

[supercane4867]

Not sure if it's been posted here before, but thought some might like to read this unofficial reanalysis of the most powerful tropical cyclones, using the most up-to-date (as of 2017) version of the Advanced Dvorak Technique (unlike previous unofficial reanalyses, which either compared apples-to-apples but used older versions that no longer allow for comparisons to Patricia or Haiyan, or simply compared apples-to-oranges), as well as compensating for several biases (such as latitude and eye size): https://journals.ametsoc.org/mwr/article/145/3/971/103374/Reprocessing-the-Most-Intense-Historical-Tropical Click on the PDF button to read the whole thing!

The reanalysis' final calibrated ADT rankings:

1 Patricia (2015) 8.4 182 876
2 Haiyan (2013) 8.2 176 878
3 Tip (1979) 8.1 173 873
3 Gay (1992) 8.1 173 883
5 Gilbert (1988) 8.0 170 887
5 Yuri (1991) 8.0 170 887
5 Nida (2009) 8.0 170 892
8 Linda (1997) 7.9 167 884
8 Allen (1980) 7.9 167 886
8 Vanessa (1984) 7.9 167 886
8 Wilma (2005) 7.9 167 888
8 Angela (1995) 7.9 167 889

Note that this reanalysis retains Tip as having the lowest minimum pressure (although it underestimates it by 3mb), although Patricia and Haiyan beat it for windspeed! I am faintly thrilled Gilbert was analysed so high; I always thought Gilbert peaked in between recon passes (888mb, 889mb six hours later) - the original 885mb the NHC scrapped in post-season analysis was probably closer to its true peak - and always thought its satellite presentation was the most beautiful and powerful-looking I'd ever seen, alongside Haiyan and Tip. Nice to know there's an objective satellite analysis to support that subjective observation!

As for Gilbert, years ago Dr. Jeff Masters wrote an article documenting his experience of the recon mission during peak intensity. The plane at the time did not carry any dropsonde so that pressure readings were only extrapolated from flight level. Operationally, the NHC reportedly said they couldn't believe their numbers which had been as low as 879mb. Later study determined the actual lowest pressure in Gilbert was 888mb, but that still leaves some wiggle room for debate since there was no ground truth to the measurement.

https://www.wunderground.com/blog/JeffM ... ars-l.html

To this day, Gilbert remains the most impressive Atlantic hurricane in the satellite era and the only one to receive a T8.0 on subjective Dvorak. Its measured flight level wind of 173kt is also the highest in the Atlantic and the second highest overall in the WHEM only behind Patricia. I have no doubt that we would see 165kt+ SFMR measurements in Gilbert had the instrument available back then.

[1900hurricane]

Not sure if it's been posted here before, but thought some might like to read this unofficial reanalysis of the most powerful tropical cyclones, using the most up-to-date (as of 2017) version of the Advanced Dvorak Technique (unlike previous unofficial reanalyses, which either compared apples-to-apples but used older versions that no longer allow for comparisons to Patricia or Haiyan, or simply compared apples-to-oranges), as well as compensating for several biases (such as latitude and eye size): https://journals.ametsoc.org/mwr/article/145/3/971/103374/Reprocessing-the-Most-Intense-Historical-Tropical Click on the PDF button to read the whole thing!

The reanalysis' final calibrated ADT rankings:

1 Patricia (2015) 8.4 182 876
2 Haiyan (2013) 8.2 176 878
3 Tip (1979) 8.1 173 873
3 Gay (1992) 8.1 173 883
5 Gilbert (1988) 8.0 170 887
5 Yuri (1991) 8.0 170 887
5 Nida (2009) 8.0 170 892
8 Linda (1997) 7.9 167 884
8 Allen (1980) 7.9 167 886
8 Vanessa (1984) 7.9 167 886
8 Wilma (2005) 7.9 167 888
8 Angela (1995) 7.9 167 889

Note that this reanalysis retains Tip as having the lowest minimum pressure (although it underestimates it by 3mb), although Patricia and Haiyan beat it for windspeed! I am faintly thrilled Gilbert was analysed so high; I always thought Gilbert peaked in between recon passes (888mb, 889mb six hours later) - the original 885mb the NHC scrapped in post-season analysis was probably closer to its true peak - and always thought its satellite presentation was the most beautiful and powerful-looking I'd ever seen, alongside Haiyan and Tip. Nice to know there's an objective satellite analysis to support that subjective observation!

The Veldon ADT study is interesting. Someone came to this thread and shared it back when it was first published (you can check that out here if you want), but that was literally 20 pages back. It's very cool to see how close it gets to the best track intensities of some of these systems, as well as providing estimates for some TCs without recon. There are some things worth pointing out though.

First the analysis uses only one intensity estimation method: ADT. As such, it is prone to all of ADT's possible biases, such as too much priority towards deep non-axisymmetric in some cases (think of cases like Bill '09 and Karl '10) while having no check for radial symmetry, which may underrepresent intensities in other TCs (cases like Meranti '16 and Dorian '19). ADT also handles storms differently between the eastern and western hemispheres.

Another thing worth pointing out is the storm selection that was used was somewhat limited. Some lesser-known C5 TCs from other basins or even "missed" C5 TCs were excluded from the study. I feel like the SPac in particular was underrepresented. Using only C5 TCs, and not even the entire selection of the available ones confined the scope of the study somewhat.

All in all, definitely a neat study with plenty to offer, but it's worth keeping in mind the limitations. Especially in basins not frequented by recon, it's important to use data from multiple estimation methods as some are stronger in particular situations than others.

[supercane4867]

Another thing worth pointing out is the storm selection that was used was somewhat limited. Some lesser-known C5 TCs from other basins or even "missed" C5 TCs were excluded from the study. I feel like the SPac in particular was underrepresented. Using only C5 TCs, and not even the entire selection of the available ones confined the scope of the study somewhat.

I'm quite perplexed about how they left Hina 85 out of their study. There were also plenty of other noteworthy TCs in the SPAC that could have been included in the study.

[1900hurricane]

Here's something I think is interesting. In the Willoughby, Landsea, & Masters publication on Gilbert '88, they came to the conclusion that the 700 mb height extrapolation estimation of central pressure devised by Jordan in 1958 (0.115 * z700 + 645) ran 1.6 mb deeper than dropsonde splash pressures for TCs under 900 mb. Given the analyzed minimum 700 mb height in Gilbert of 2102 m, raw Jordan Equation output is 886.7 mb, which was corrected up to the ~888 mb finalized in best track.

However, after comparing with more recent storms with a pressure below 900 mb, the Jordan Equation appears to do rather well compared to dropsonde splash pressures. The sample size is a bit smaller, but the extrapolation ran on average 0.6 mb above splash pressures below 900 mb. Factor in wind reductions from modern dropwindsondes, and that gap further increases to 2.7 mb. As to why this is, maybe modern recon practices are better at a more accurate minimum splash pressure. Maybe instruments are better calibrated overall. Whatever it is, it's something that I personally cannot easily explain or conclude. Given this though, I do think it may be possible that 886 mb is perhaps a better representation of Gilbert's overall minimum pressure.

[1900hurricane]

Sentinal-1's SAR once again got a clean pass, this time on Hurricane Isaias a little before its CONUS landfall. There was once again recon at the time, and data lines up very nicely between the two data sources. I wonder if the recon data is actually being actively used to calibrate the SAR data, which is why the two line up so nicely.

The SAR pass lined up almost exactly with the first recon pass of that mission, but it would have been interesting to see it come an hour or so later to compare against the second pass. That's when recon measured the very unusual 117 kt 700 mb wind.

AL, 09, 202008032313, 70, SEN1, CIR, , 3268N, 7911W, 10 , , 72, 1, , , SEN1, 34, NEQ, 78, 107, 0, 0, , , , , , , , L, STAR, SAR, SAR, 202008032313, 202008032314, , , SENTINEL-1A, STAR Synthetic Aperture Radar 3KM Wind Speed Analysis
AL, 09, 202008032313, 70, SEN1, CIR, , 3268N, 7911W, 10 , , 72, 1, , , SEN1, 50, NEQ, 19, 41, 38, 18, , , , , , , , L, STAR, SAR, SAR, 202008032313, 202008032314, , , SENTINEL-1A, STAR Synthetic Aperture Radar 3KM Wind Speed Analysis
AL, 09, 202008032313, 70, SEN1, CIR, , 3268N, 7911W, 10 , , 72, 1, , , SEN1, 64, NEQ, 0, 0, 31, 0, , , , , , , , L, STAR, SAR, SAR, 202008032313, 202008032314, , , SENTINEL-1A, STAR Synthetic Aperture Radar 3KM Wind Speed Analysis

[WAcyclone]

Not sure this is the right thread given these are relatively weak storms, but we just got another two recon-verified SMAP passes.

1) Marco at 2020-08-23 1151Z. SMAP estimated 58 kt. Highest unflagged SFMR at around this time was 59 kt. NHC estimate 60 kt.

2) Laura at 2020-08-23 2243Z. SMAP estimated 57 kt. Highest unflagged SFMR at 20Z was 55kt. NHC estimate 50 kt. However, all of these estimates are quite uncertain due to significant land interaction occurring at the time.

Hopefully we get some more passes once Laura strengthens significantly in the Gulf.

[1900hurricane]

Not sure this is the right thread given these are relatively weak storms, but we just got another two recon-verified SMAP passes.

1) Marco at 2020-08-23 1151Z. SMAP estimated 58 kt. Highest unflagged SFMR at around this time was 59 kt. NHC estimate 60 kt.

2) Laura at 2020-08-23 2243Z. SMAP estimated 57 kt. Highest unflagged SFMR at 20Z was 55kt. NHC estimate 50 kt. However, all of these estimates are quite uncertain due to significant land interaction occurring at the time.

Hopefully we get some more passes once Laura strengthens significantly in the Gulf.

I'll allow it.

Looks like some of the newer satellite estimates discussed in this thread have been preforming very well compared to recon observations this year, which is awesome! However, we really haven't gotten an intense TC for comparison yet, so it'll be interesting to see if the good performance continues or not, as well as see if there are any possible biases as we climb the intensity ladder. I was hoping to get some good data from Genevieve, but that one kind of fell on its face. Laura might be our next opportunity, and I feel like it won't be the last one in the NAtl this year.

[WAcyclone]

Not sure this is the right thread given these are relatively weak storms, but we just got another two recon-verified SMAP passes.

1) Marco at 2020-08-23 1151Z. SMAP estimated 58 kt. Highest unflagged SFMR at around this time was 59 kt. NHC estimate 60 kt.

2) Laura at 2020-08-23 2243Z. SMAP estimated 57 kt. Highest unflagged SFMR at 20Z was 55kt. NHC estimate 50 kt. However, all of these estimates are quite uncertain due to significant land interaction occurring at the time.

Hopefully we get some more passes once Laura strengthens significantly in the Gulf.

I'll allow it.

Looks like some of the newer satellite estimates discussed in this thread have been preforming very well compared to recon observations this year, which is awesome! However, we really haven't gotten an intense TC for comparison yet, so it'll be interesting to see if the good performance continues or not, as well as see if there are any possible biases as we climb the intensity ladder. I was hoping to get some good data from Genevieve, but that one kind of fell on its face. Laura might be our next opportunity, and I feel like it won't be the last one in the NAtl this year.

Yes, SMAP's accuracy for high-intensity systems is what will ultimately determine its true usefulness for TC intensity estimation. We just don't have a lot of recon-verified passes for intense TCs so far so the sample size is still quite small. I also wonder if the more recent V1.0 version reduces the amount of significant errors the like of Patricia (underestimated by 40 kt) and Meranti (overestimated by...probably a lot ). It is possible that it tends to underestimate small systems and overestimates large ones which would make sense given its relatively coarse spatial resolution. Also looking forward to see more passes of SAR which has a much higher resolution but less coverage and possible lower accuracy.

[1900hurricane]

Looks like SAR got a little too excited for Laura. Don't know if it is because it needs more calibration points at high intensities, if it had to do with Laura's beefy RMW, or it was a partial pass, but definitely a big swing and a miss for the instrument.

AL, 13, 202008270010, 70, SEN1, CIR, , 2850N, 9290W, 10 , , 172, 1, , , SEN1, 34, NEQ, 93, 96, 0, 0, , , , , , , , L, STAR, SAR, SAR, 202008270009, 202008270010, , , SENTINEL-1A, STAR Synthetic Aperture Radar 3KM Wind Speed Analysis
AL, 13, 202008270010, 70, SEN1, CIR, , 2850N, 9290W, 10 , , 172, 1, , , SEN1, 50, NEQ, 64, 41, 0, 0, , , , , , , , L, STAR, SAR, SAR, 202008270009, 202008270010, , , SENTINEL-1A, STAR Synthetic Aperture Radar 3KM Wind Speed Analysis
AL, 13, 202008270010, 70, SEN1, CIR, , 2850N, 9290W, 10 , , 172, 1, , , SEN1, 64, NEQ, 40, 34, 0, 0, , , , , , , , L, STAR, SAR, SAR, 202008270009, 202008270010, , , SENTINEL-1A, STAR Synthetic Aperture Radar 3KM Wind Speed Analysis

[NotoSans]

My observation would be that partial passes tend to be inaccurate.

[NotoSans]

Not related to a particular intense TC, but this recent paper from Klotzbach argues TC should be classified using MSLP instead of MSW because (a) MSLP is a better predictor of TC damage and (b) MSLP is more accurately measured.
https://doi.org/10.1175/BAMS-D-19-0062.1

[MarioProtVI]

JTWC BT for 2019 is out and boy were there changes:


Hagibis was upped to 160 kt from 140 kt (MUCH better, although I think 1900’s analysis still fits my directory)
Wutip now has two C5 peaks and is now 895 per their BT making Wutip the strongest February typhoon ever recorded.
Bualoi now joins the ranks of C5 super typhoons with a big jump to 140 kt
Halong was upped to 165 kt which seems a bit weird consider the eye wasnt as clear for a T8 as you’d think
Lekima was upped to 135 kt

[aspen]

JTWC BT for 2019 is out and boy were there changes:


Hagibis was upped to 160 kt from 140 kt (MUCH better, although I think 1900’s analysis still fits my directory)
Wutip now has two C5 peaks and is now 895 per their BT making Wutip the strongest February typhoon ever recorded.
Bualoi now joins the ranks of C5 super typhoons with a big jump to 140 kt
Halong was upped to 165 kt which seems a bit weird consider the eye wasnt as clear for a T8 as you’d think
Krosa was upped to 135 kt

Halong’s new official peak of 165 kt/888 mbar is extremely close to the T#7.8/7.9 results from ADT during its peak appearance (165 kt/892 mbar). It maintained that ADT intensity estimate for six or seven hours, and had a nearly complete CDG ring with an >18C eye that entire time.

[1900hurricane]

Just put together an introductory post for some of the eye diameter vs deepening rate stuff I've been working on on and off for much of the past year. There's a lot more I'll have to do with it to make it something truly viable, but hoping that putting this out there helps keep it off the heap of neglected unfinished projects I seem to have.

Re-examining Rapidly Deepening Tropical Cyclones Using Eye Diameter as a Determinant

 https://twitter.com/1900hurricane/status/1312922424598507526

[Shell Mound]

Delta will almost certainly be a topic of discussion in this thread shortly. Given the environment, Gilbert and Wilma are definitely in play.

[aspen]

Delta will almost certainly be a topic of discussion in this thread shortly. Given the environment, Gilbert and Wilma are definitely in play.

This aged as poorly as everything in Delta’s peak intensity thread.

[euro6208]

So back in 2019, we highlight 2 of the most intense typhoons and probrably one for the record books but because of the lack of recon, we will never know.

One thing is for sure. Both are very intense.



Halong

A peak intensity of 165 knots.




Hagibis

Post season upgrade to 160 knots.

175 knots peak?

Personally i think Hagibis was stronger. Think Wilma and Patricia.

A lack of recon will prevent anything higher than 165 knots because that is considered the *limit* sadly although Haiyan and Meranti might have to say something about that. Patricia got lucky.

[aspen]

So back in 2019, we highlight 2 of the most intense typhoons and probrably one for the record books but because of the lack of recon, we will never know.

One thing is for sure. Both are very intense.

https://i.imgur.com/XBHXE9g.jpg

Halong

A peak intensity of 165 knots.


https://i.imgur.com/waEc5Fk.jpg

Hagibis

Post season upgrade to 160 knots.

175 knots peak?

Personally i think Hagibis was stronger. Think Wilma and Patricia.

A lack of recon will prevent anything higher than 165 knots because that is considered the *limit* sadly although Haiyan and Meranti might have to say something about that. Patricia got lucky.

165 kt is a very good estimate for Halong. It was extremely intense, but it was just a little below a perfect T#8.0/170 kt TC. Hagibis, on the other hand, was likely in the 160-175 kt range.

Speaking of WPac systems, it’s very likely we’ll be talking about Goni and/or Atsani here soon. Goni is going through a phase of ERI and is developing a rapidly warming pinhole eye. Models are very aggressive with Atsani, far more than Goni. It seems like Goni could be similar to Hagibis and Atsani could be similar to Halong, in terms of structure.

[euro6208]

So back in 2019, we highlight 2 of the most intense typhoons and probrably one for the record books but because of the lack of recon, we will never know.

One thing is for sure. Both are very intense.



Halong

A peak intensity of 165 knots.




Hagibis

Post season upgrade to 160 knots.

175 knots peak?

Personally i think Hagibis was stronger. Think Wilma and Patricia.

A lack of recon will prevent anything higher than 165 knots because that is considered the *limit* sadly although Haiyan and Meranti might have to say something about that. Patricia got lucky.

165 kt is a very good estimate for Halong. It was extremely intense, but it was just a little below a perfect T#8.0/170 kt TC. Hagibis, on the other hand, was likely in the 160-175 kt range.

Speaking of WPac systems, it’s very likely we’ll be talking about Goni and/or Atsani here soon. Goni is going through a phase of ERI and is developing a rapidly warming pinhole eye. Models are very aggressive with Atsani, far more than Goni. It seems like Goni could be similar to Hagibis and Atsani could be similar to Halong, in terms of structure.

Can you say the same for Dorian. Estimates didn't even reach 140 knots yet recon found 160 knots. Oh my...

[ElectricStorm]

So back in 2019, we highlight 2 of the most intense typhoons and probrably one for the record books but because of the lack of recon, we will never know.

One thing is for sure. Both are very intense.



Halong

A peak intensity of 165 knots.




Hagibis

Post season upgrade to 160 knots.

175 knots peak?

Personally i think Hagibis was stronger. Think Wilma and Patricia.

A lack of recon will prevent anything higher than 165 knots because that is considered the *limit* sadly although Haiyan and Meranti might have to say something about that. Patricia got lucky.

165 kt is a very good estimate for Halong. It was extremely intense, but it was just a little below a perfect T#8.0/170 kt TC. Hagibis, on the other hand, was likely in the 160-175 kt range.

Speaking of WPac systems, it’s very likely we’ll be talking about Goni and/or Atsani here soon. Goni is going through a phase of ERI and is developing a rapidly warming pinhole eye. Models are very aggressive with Atsani, far more than Goni. It seems like Goni could be similar to Hagibis and Atsani could be similar to Halong, in terms of structure.

Can you say the same for Dorian. Estimates peaked at below 140 knots. Yet recon....

Not all storms are stronger than estimates. Look at Zeta, it had Cat 4 estimates there for a bit and it clearly wasn't a Cat 4. Recon goes both ways