Discussion of Intense Tropical Cyclones

[Monisae]

The lowest SLP recorded in Japan during VERA is 929.2 mb, so the landfall pressure is very likely 928 mb. VERA was indeed very large when it made landfall, so the intensity estimated using modern re-analysis technique may not be particularly high, but its extensive wind field is what makes it dangerous.

Re-Analysis/Prediction Of Typhoon Vera (1959) Project
https://ams.confex.com/ams/pdfpapers/16 ... 168250.pdf

It has what Vera would of looked like on satellite. Hurricane Katrina had central pressure of 920 millibars when it made the first landfall on Southeast Louisiana. It had 125 mph or 200 km/h wind. Katrina was large when it made landfall on Louisiana and Mississippi.

It's almost impossible to get a CAT5 landfall on Japanese mainland. I belive the northernmost CAT5 landfall record was set by Camille just north of 30N. In fact it's much harder for a TC to maintain such intensity at that latitude anywhere near land in the WPAC due to its geographic environment.

Maybe Muroto Typhoon in 1934 was a cat-5 typhoon in Japan, 911.6hpa was recorded in the eyewall.
https://repository.kulib.kyoto-u.ac.jp/dspace/bitstream/2433/257138/1/mcskiu-a_19_2_79.pdf

[Zonacane]

Before the TCR comes out, what do y'all think the maximum intensities of Milton and Beryl were?

[Hurricane2022]

Before the TCR comes out, what do y'all think the maximum intensities of Milton and Beryl were?

Beryl: 150 kt/931 mb
Kirk: 140 kt
Milton: 160 kt/895 mb
Oscar: 85 kt

[cheezyWXguy]

Before the TCR comes out, what do y'all think the maximum intensities of Milton and Beryl were?

Beryl: 150 kt/931 mb
Kirk: 140 kt
Milton: 160 kt/895 mb
Oscar: 85 kt

I think you’re spot on with these intensities, but I’m not convinced the nhc will bump up beryl or kirk. Feel like Milton and Oscar have a better shot.

[vbhoutex]

Before the TCR comes out, what do y'all think the maximum intensities of Milton and Beryl were?

Beryl: 150 kt/931 mb
Kirk: 140 kt
Milton: 160 kt/895 mb
Oscar: 85 kt

I think you’re spot on with these intensities, but I’m not convinced the nhc will bump up beryl or kirk. Feel like Milton and Oscar have a better shot.

Perhaps my old mind is fuzzier than I think, but when did Beryl approach that intensity? Of course my memory of Beryl is focused on the GOM portion of Beryl and landfall since the eye went over my house. Milton may be a tad high, but not by much. Not sure NHC will have enough to bump it based on independent wind readings I am aware of. Sparse reliable certified weather stations near/at landfall are the problem.

[Monisae]

I feel like some official/agency stuff about Typhoon Judy (1979) is just.. weirdly false for some reason

On the 19th of August at 22:45, aircraft reconnaissance recorded a pressure of 887mb was recorded in the storm according to the JTWC's ATCR. Some time earlier, the warmest eye temperature (at 700mb height) ever recorded in a tropical cyclone (34C) was also recorded when the storm's barometric was at 889mb.

I went to look through satellite images of Judy from the JMA database out of curiosity
Why does it state the pressure at 910mb
that's not even rounded to the nearest 5mb or anything
https://media.discordapp.net/attachments/887577983508967455/919879276759744522/unknown.png?width=990&height=427

I also looked through the IBTrACS database for the other agency estimates
https://media.discordapp.net/attachments/887577983508967455/919880223577755668/unknown.png
It feels odd that agencies aren't paying attention to aircraft recon observations sometimes (or maybe they don't trust them idk)
you wouldn't be able to know about Judy's actual pressure if you didn't read the JTWC's ATCR

Anyways, because the JTWC used AH77 (Atkinson-Holiday wind pressure relationship 1977) for winds, which apparently tends to underestimate wind speed for stronger storms, I was wondering: what does everyone here think Judy's peak winds were? I'm not an expert but I'd say >=160kt from the pressure alone.

I guess they did't receive the aircraft observations data at 8.19

For example, China Meteorological Administration(CMA) receive the aircraft reconnaissance data and published them on annual Typhoon Report（台风年鉴） from 1949-1987, here's the record of typhoon Judy from CMA, BJT=UTC+8



Compared to record from JTWC,It's obvious that the data in 8.19 was missed by CMA, maybe JMA and HKO meet with the same problem.

[cheezyWXguy]

Beryl: 150 kt/931 mb
Kirk: 140 kt
Milton: 160 kt/895 mb
Oscar: 85 kt

I think you’re spot on with these intensities, but I’m not convinced the nhc will bump up beryl or kirk. Feel like Milton and Oscar have a better shot.

Perhaps my old mind is fuzzier than I think, but when did Beryl approach that intensity? Of course my memory of Beryl is focused on the GOM portion of Beryl and landfall since the eye went over my house. Milton may be a tad high, but not by much. Not sure NHC will have enough to bump it based on independent wind readings I am aware of. Sparse reliable certified weather stations near/at landfall are the problem.

Beryl reached peak intensity in the eastern Caribbean, and its operational peak was 145kt. This was based on recon observations that were taken just slightly after satellite appearance started to degrade from the apparent peak, so it is conceivable that its peak was closer to 150kt, but I don’t think there is a sufficiently compelling case to officially bump up the intensity.

As for Milton, it’s another close case with recon, but I think it missed the peak when they measured its central pressure at 897mb. I made a post about this later that night, in which I believed the peak was slightly higher than the obs taken because the storm’s inflow channel began to be impeded by the Yucatán peninsula about 30min to an hour before recon got in there. When they did, they eye had cooled slightly and had begun contracting at an accelerated rate, similar to post-peak Lee ‘23. Additionally, the pressure rose several mb between the first and second pass, which makes me think the peak pressure was a few mb lower, since satellite appearance had largely stabilized for a while until shortly before recon got in. However, I don’t think this is a super compelling case for a post season adjustment, but is slightly more so than the case for adjusting Beryl. If I were to spitball a percentage chance of adjusting each, I’d give a 30% chance for Milton and a 10% chance for Beryl, but again that’s just me spitballing.

[Monisae]

Patricia and Megi lose their 180+ knot claims is the biggest change if this analysis holds up.

Patricia’s TCR mentioned that the reconstructed SFMR readings from the two flights were considered reliable at the time. Therefore, the NHC will have to re-analyze these readings again, and try to see how well winds were translating down to the surface. FL conversion still supports winds of 200+ mph. What wind speed is supported by KCZ with a pressure of 872mb?

I don’t know much about Megi’s recon flights nor the analysis of its data, but I do know that Megi is not officially ~180 kt according to the JTWC — at least, that’s what Wikipedia would suggest.

CKZ supports 185 for 872 pressure.

Reprocessed with the 2015-2021 SFMR, Megi had SFMR’s of 187 knots and FL winds for Megi was 192(?) knots. Even the old SFMR algorithm had 171 knots. JTWC getting intensity estimates wrong is nothing new.

That's amazing, considering Megi was still intensifying at the last aircraft reconnaissance, maybe Megi has the wind speed so high as Hurricane Patricia at it's peak, though the pressure of Megi is about 15mb higher than Patricia. What's the new SFMR of hurricane Felix? I think it would be very impressive.

[WaveBreaking]

While looking back at some satellite imagery from Maria '17, I noticed an interesting feature around the time of peak intensity. Just before 00Z September 20th, there was a large and cold convective burst near the edge of the CDO (still image below, loop here). I'm not sure this phenomenon has a proper name, so for now, I'm going to call it lobing.

https://www.nrlmry.navy.mil/tcdat/tc2017/AL/AL152017/png/Infrared/201709192330_AL152017_abi_Infrared_80.0_covg100p0_res1km.png

Based on recon data, it appears like it might have coincided with a halt in intensification since the previously falling pressure evened out at 909 mb at the 2219Z pass and only varied by a millibar up and down from that number for the next six hours before rising. This lobing phenomenon could possibly be associated with structural changes. Lobing isn't exclusive to Maria '17 either. It appears it has occurred with a number of intense tropical cyclones across multiple basins. Some of the more clear examples are Nepartak '16, Patricia '15, Usagi '13, and Gilbert '88. Patricia '15 and Gilbert '88 each had recon near the time of lobing (near 18Z October 23rd for Patricia and 00Z September 14th for Gilbert), and both cases were similar to the Maria '17 data which showed no appreciable deepening from that time on. Patricia '15 actually began to weaken rapidly, although land interaction may have played a part in that case. Regardless, it may be a good sign to indicate whether an intense tropical cyclone has peaked.

Pretty sure this can also be seen with Jova ‘23 since its violent CDO quickly warmed and shrank after an outer band flared up during its peak. This band also started an EWRC and a long-term weakening trend.

[Ptarmigan]

Before the TCR comes out, what do y'all think the maximum intensities of Milton and Beryl were?

Beryl: 150 kt/931 mb
Kirk: 140 kt
Milton: 160 kt/895 mb
Oscar: 85 kt

I would not be surprised if Beryl, Kirk, Milton, and Oscar are stronger. I expect Milton to be upgraded have lower pressure and higher winds, ala Rita in 2005. Rita initially had 150 knots and 897 millibars. It was upgraded to 155 knots and 895 millibars.

[Yellow Evan]

Patricia’s TCR mentioned that the reconstructed SFMR readings from the two flights were considered reliable at the time. Therefore, the NHC will have to re-analyze these readings again, and try to see how well winds were translating down to the surface. FL conversion still supports winds of 200+ mph. What wind speed is supported by KCZ with a pressure of 872mb?

I don’t know much about Megi’s recon flights nor the analysis of its data, but I do know that Megi is not officially ~180 kt according to the JTWC — at least, that’s what Wikipedia would suggest.

CKZ supports 185 for 872 pressure.

Reprocessed with the 2015-2021 SFMR, Megi had SFMR’s of 187 knots and FL winds for Megi was 192(?) knots. Even the old SFMR algorithm had 171 knots. JTWC getting intensity estimates wrong is nothing new.

That's amazing, considering Megi was still intensifying at the last aircraft reconnaissance, maybe Megi has the wind speed so high as Hurricane Patricia at it's peak, though the pressure of Megi is about 15mb higher than Patricia. What's the new SFMR of hurricane Felix? I think it would be very impressive.

Even the algo that was in place at the time yielded 173 knots. Problem is that was in a meso.

[kevin]

I did some reanalysis of the peak intensity of hurricane Jova in 2023. Of course it's difficult for an official agency to be bullish regarding a TC when there are no land threats and direct measurements, so I understand why the NHC kept Jova at the intensity it had. Saying 'it was almost certainly a cat 5, but besides that we don't know exactly' is probably the safest bet. But since I don't have the same responsibilities as the NHC I'm free to be a bit more free in my estimates .

Anyways, I first calculated the expected wind speed of hurricane Jova based on the different satellite estimates and then used this new wind speed value with the KZC wind-pressure relationship to find the associated pressure.

ADT
2023SEP07 014020 5.9 948.8 112.4 5.9 6.0 7.8 1.3T/6hr OFF OFF OFF OFF 16.85 -77.29 EYE -99 IR 4.6 15.40 112.76 ARCHER GOES18 33.2
2023SEP07 124020 7.2 917.2 146.0 6.5 6.4 6.4 NO LIMIT ON FLG OFF OFF -38.62 -72.64 EYE -99 IR 24.7 16.80 115.26 ARCHER GOES18 31.7

Raw T# peaked at 7.8 with a peak in the CI# of 7.2. Since the Raw T# started dropping 12 hours after an eye became visible, I put more stock in the more consistent CI# and use its associated value of 146 kt, 917 mb for the ADT intensity estimate.

AiDT
20230907 074020 143 135 140
20230907 081020 146 136 140
20230907 084020 146 137 140
20230907 091020 146 137 140
20230907 094020 146 137 140
20230907 101020 146 136 140

During the same time period AiDT peaked with a slightly lower peak strength of 137 kt.

AMSU
, , , , , , , , , , , , , , , , , , ,
EP, 11, 202309070443, 30, AMSU, IP, , 1595N, 11342W, , 2, 157, 1, 909, 1, MEAS, , , , , , , , , , , , 4, , E, CIMS, , , , , , , , 909, , NOAA93, , , , , , , , , , , , , , , ,

AMSU on the other hand peaked at a whopping 157 kt, 909 mb.

D-MINT
20230907 0208 UTC SSMISF17 932 hPa 145 kts 137 kts 153 kts current 11E intensity image

D-MINT had a peak intensity of 145 kt, 932 mb.

D-PRINT
20230907 0500 UTC 935 hPa 144 kts 137 kts 151 kts

At the same time as D-MINT, D-PRINT peaked at 144 kt, 935 mb.

SATCON
2023 EP 11 250.278 2023SEP07 064020 16.09 113.98 3 923 148

In terms of wind speed SATCON peaked at 148 kt, 923 mb (a few hours later there was a slightly lower pressure estimate of 919 mb).

Wind blend
All of the above estimates show a peak intensity between 02:00z and 08:00z, which is close enough together that I think a direct blend of the above values is possible. This would thus result in a peak intensity of 146 kt, 923 mb.

KZC relationship
Now I use the KZC relationship to find a potentially more accurate pressure value associated with the wind speed of 145 kt (rounded down from 146 kt since we usually work in steps of 5 kt). For this I used the following input values during peak intensity of 03:00z.
Vmax = 145 kt
C = 14 kt
R34 = 121 nm
Lat = 16.2 deg
Background pressure = 1008 mb

This then results in a pressure estimate of 916 mb.

Conclusion
The official peak intensity for hurricane Jova is 140 kt, 926 mb. Based on a blend of all available satellite estimates and the KZC relationship, I estimate that Jova's peak intensity was slightly stronger at 145 kt, 916 mb in the early hours of September 7.

[kevin]

And since I did Jova I also have to do an analysis on hurricane Lee. Lee had direct recon coverage as a cat 5, but also possibly intensified a bit more after recon left.

ADT
2023SEP07 234020 6.0 948.9 115.0 6.0 6.4 7.0 1.3T/6hr OFF OFF OFF OFF 21.76 -72.12 EYE 18 IR 86.1 16.99 51.69 ARCHER GOES16 33.5
2023SEP08 064020 6.8 930.1 134.8 6.6 6.7 6.7 NO LIMIT ON OFF OFF OFF 15.84 -70.87 EYE 15 IR 86.1 17.51 53.13 ARCHER GOES16 32.5

Raw T# peaked at 7.0 with a CI# peak of 6.8. I'm going with CI and thus 135 kt, 930 mb for the ADT intensity estimate.

AiDT
20230908 061020 134 135 145
20230908 064020 134 135 145
20230908 071020 134 135 145
20230908 074020 134 135 145
20230908 081020 134 135 145

During the same time period AiDT peaked at 135 kt.

AMSU
There was no good AMSU coverage during peak intensity (analyses are from 14:00z on Sep 7 and 13:00 on Sep 8) so I'm skipping this.

D-MINT
20230908 0737 UTC SSMISF18 930 hPa 136 kts 129 kts 142 kts
D-MINT had a peak intensity of 136 kt, 930 mb.

D-PRINT
20230908 0400 UTC 930 hPa 140 kts 133 kts 146 kts

At a similar time as D-MINT, D-PRINT peaked at 140 kt, 930 mb.

SATCON
2023 AL 13 251.278 2023SEP08 064020 17.51 53.13 3 950 125

In terms of wind speed SATCON peaked at 125 kt, 950 mb.

Wind blend
All of the above estimates show a peak intensity around 06:00z with a blend of 134 kt, 935 mb.

KZC relationship
Now I use the KZC relationship to find a potentially more accurate pressure value associated with the wind speed of 135 kt (rounded up from 134 kt since we usually work in steps of 5 kt). For this I used the following input values during peak intensity of 06:00z.
Vmax = 135 kt
C = 12 kt
R34 = 90 nm
Lat = 17.6 deg
Background pressure = 1010 mb

This then results in a pressure estimate of 930 mb.

Conclusion
The official peak intensity for hurricane Lee is 145 kt, 926 mb. Based on a blend of all available satellite estimates and the KZC relationship, I estimate that Lee's peak intensity was slightly weaker at 135 kt, 930 mb in the early hours of September 8. In reality peak FL winds of 153 kt were found (reducing to 140 kt at the surface) with adjusted SFMR of 145 - 150 kt. Furthermore, a dropsonde found a central pressure of 928 mb with 20 kt winds -> 926 mb. So we know from direct observations that Lee was slightly stronger than the satellite estimate I just made. Note the following passage by NHC:

"Lee’s estimated minimum pressure of 926 mb was based on dropsonde data, with 928 mb
and 20 kt of wind recorded near 0455 UTC 8 September. Notably, there was a rise from 928 mb
at 1025 UTC 8 September to 940 mb just 76 minutes later -- quite a remarkable filling over open
water in such a short period of time"

I think that since Lee was at it peak for such a short period of time and then quickly weakened, satellite estimates were not able to align and/or catch up with the true intensity. While 135kt/930mb is still a reasonable estimate for Lee, I think this shows that such satellite/KZC techniques are more suited to TCs with a consistent long-duration peak instead of these rapid ramp-up/collapse systems.

[Category5Kaiju]

So, some interesting recent developments have come up. Cyclone Dikeledi, in the Southwest Indian Ocean, attained major hurricane strength (Category 3 equivalent) well south of Madagascar.

Is it just me, or is this unusually far south for a major strength SWIO cyclone? Unless I'm missing a historically similar-natured storm in the Australian/South Pacific basins, it's kind of crazy to think that Dikeledi might even be the closest tropical cyclone (of Cat 3 or higher strength) to Antarctica in recorded times.

[Hurricane2022]

Is it just me, or is this unusually far south for a major strength SWIO cyclone? Unless I'm missing a historically similar-natured storm in the Australian/South Pacific basins...

https://i.imgur.com/q5AorRC.png

It is really unusual to have a Major this far south.
 https://x.com/EKMeteo/status/1879886792134209581

[Hurricane2022]

And since I did Jova I also have to do an analysis on hurricane Lee. Lee had direct recon coverage as a cat 5, but also possibly intensified a bit more after recon left.

ADT
2023SEP07 234020 6.0 948.9 115.0 6.0 6.4 7.0 1.3T/6hr OFF OFF OFF OFF 21.76 -72.12 EYE 18 IR 86.1 16.99 51.69 ARCHER GOES16 33.5
2023SEP08 064020 6.8 930.1 134.8 6.6 6.7 6.7 NO LIMIT ON OFF OFF OFF 15.84 -70.87 EYE 15 IR 86.1 17.51 53.13 ARCHER GOES16 32.5

Raw T# peaked at 7.0 with a CI# peak of 6.8. I'm going with CI and thus 135 kt, 930 mb for the ADT intensity estimate.

AiDT
20230908 061020 134 135 145
20230908 064020 134 135 145
20230908 071020 134 135 145
20230908 074020 134 135 145
20230908 081020 134 135 145

During the same time period AiDT peaked at 135 kt.

AMSU
There was no good AMSU coverage during peak intensity (analyses are from 14:00z on Sep 7 and 13:00 on Sep 8) so I'm skipping this.

D-MINT
20230908 0737 UTC SSMISF18 930 hPa 136 kts 129 kts 142 kts
D-MINT had a peak intensity of 136 kt, 930 mb.

D-PRINT
20230908 0400 UTC 930 hPa 140 kts 133 kts 146 kts

At a similar time as D-MINT, D-PRINT peaked at 140 kt, 930 mb.

SATCON
2023 AL 13 251.278 2023SEP08 064020 17.51 53.13 3 950 125

In terms of wind speed SATCON peaked at 125 kt, 950 mb.

Wind blend
All of the above estimates show a peak intensity around 06:00z with a blend of 134 kt, 935 mb.

KZC relationship
Now I use the KZC relationship to find a potentially more accurate pressure value associated with the wind speed of 135 kt (rounded up from 134 kt since we usually work in steps of 5 kt). For this I used the following input values during peak intensity of 06:00z.
Vmax = 135 kt
C = 12 kt
R34 = 90 nm
Lat = 17.6 deg
Background pressure = 1010 mb

This then results in a pressure estimate of 930 mb.

Conclusion
The official peak intensity for hurricane Lee is 145 kt, 926 mb. Based on a blend of all available satellite estimates and the KZC relationship, I estimate that Lee's peak intensity was slightly weaker at 135 kt, 930 mb in the early hours of September 8. In reality peak FL winds of 153 kt were found (reducing to 140 kt at the surface) with adjusted SFMR of 145 - 150 kt. Furthermore, a dropsonde found a central pressure of 928 mb with 20 kt winds -> 926 mb. So we know from direct observations that Lee was slightly stronger than the satellite estimate I just made. Note the following passage by NHC:

"Lee’s estimated minimum pressure of 926 mb was based on dropsonde data, with 928 mb
and 20 kt of wind recorded near 0455 UTC 8 September. Notably, there was a rise from 928 mb
at 1025 UTC 8 September to 940 mb just 76 minutes later -- quite a remarkable filling over open
water in such a short period of time"

I think that since Lee was at it peak for such a short period of time and then quickly weakened, satellite estimates were not able to align and/or catch up with the true intensity. While 135kt/930mb is still a reasonable estimate for Lee, I think this shows that such satellite/KZC techniques are more suited to TCs with a consistent long-duration peak instead of these rapid ramp-up/collapse systems.

And what is your opinion about Hurricane Kirk, sir?

[kevin]

And what is your opinion about Hurricane Kirk, sir?

Let's see, I'll do the same analysis with Kirk as I did for Jova and Lee.

ADT
2024OCT03 221020 6.5 931.7 127.0 6.5 7.1 7.1 NO LIMIT OFF OFF OFF OFF 18.89 -73.09 EYE 23 IR 68.6 21.03 46.78 ARCHER GOES16 40.4
2024OCT04 061020 7.0 919.0 140.0 6.3 6.2 6.2 NO LIMIT ON FLG OFF OFF 18.35 -66.84 EYE 21 IR 68.6 21.80 47.86 ARCHER GOES16 40.0

Raw T# peaked at 7.1 with a CI# peak of 7.0. I'm going with CI and thus 140 kt, 919 mb for the ADT intensity estimate.

AiDT
20241004 014020 140 134 125
20241004 021020 140 134 125
20241004 024020 140 134 125
20241004 031020 140 134 125
20241004 034020 140 134 125

During the same time period AiDT peaked at 134 kt.

AMSU
AMSU peaked at 134 kt on October 5, but has no good coverage during the true peak intensity so I'm skipping this.

D-MINT
20241004 0957 UTC SSMISF16 931 hPa 137 kts 131 kts 143 kts
D-MINT had a peak intensity of 137 kt, 931 mb.

D-PRINT
20241004 0610 UTC 937 hPa 133 kts 127 kts 138 kts

At a similar time as D-MINT, D-PRINT peaked at 133 kt, 937 mb.

SATCON
2024 AL 12 278.215 2024OCT04 051020 21.68 47.71 3 936 127

In terms of wind speed SATCON peaked at 127 kt, 936 mb.

Wind blend
All of the above estimates show a peak intensity in the morning of October 4 with a blend of 134 kt, 931 mb.

KZC relationship
Now I use the KZC relationship to find a potentially more accurate pressure value associated with the wind speed of 135 kt (rounded up from 134 kt since we usually work in steps of 5 kt). For this I used the following input values during peak intensity around 06:00z.
Vmax = 135 kt
C = 10 kt
R34 = 145 nm
Lat = 21.9 deg
Background pressure = 1008 mb

This then results in a pressure estimate of 921 mb.

Conclusion
The official peak intensity for hurricane Kirk is 125 kt, 934 mb. Based on a blend of all available satellite estimates and the KZC relationship, I estimate that Kirk's peak intensity was significantly stronger at 135 kt, 921 mb in the early hours of September 4. I expect NHC to upgrade to 130 kt or 135 kt during their analysis based on similar data I just used. Some data such as the T7.0 ADT supports a minimal cat 5 and I do think it's a possibility that it briefly peaked at 140 kt, but not all estimates support this. Even if you exclude the lower SATCON estimate, a new blend would still 'only' result in an intensity of 136 kt. The blend of satellite estimates is below cat 5 intensity and based on this I personally wouldn't pull the cat 5 trigger so I think it's highly unlikely that the NHC would.

[Hurricane2000]

What do you think of Gaemi 2024?

[kevin]

What do you think of Gaemi 2024?

That's a very interesting one. At some point it might be better for me to make my own thread for these analyses (especially of non–recon typhoons). I've thought about that before and would also make it a bit more rigorous than what I'm doing now, but who knows. Anyways, let's make this the last of its kind for now so I don't spam this thread.

ADT
2024JUL23 220000 5.0 948.5 90.0 5.0 6.1 6.6 1.3T/6hr OFF OFF OFF OFF -26.86 -78.00 EYE/P -99 IR 41.8 23.13 -123.32 ARCHER HIM-9 33.4
2024JUL24 053000 6.0 927.2 115.0 5.9 6.0 6.0 NO LIMIT ON OFF OFF OFF -23.86 -73.28 EYE -99 IR 61.0 24.10 -122.54 ARCHER HIM-9 34.7

Raw T# peaked at 6.6 with a CI# peak of 6.0. I'm going with CI and thus 115 kt, 919 mb for the ADT intensity estimate.

AiDT
20240724 050000 115 112 120
20240724 053000 115 113 120
20240724 060000 115 113 125
20240724 063000 115 113 125
20240724 070000 115 114 125

During the same time period AiDT peaked at 114 kt.

AMSU
WP, 05, 202407240258, 30, AMSU, IP, , 2384N, 12320E, , 2, 152, 1, 896, 1, MEAS, , , , , , , , , , , , 9, , W, CIMS, , , , , , , , 896, , NOAA18, , , , , , , , , , , , , , , , , , , ,

AMSU peaked at 152 kt and 896 mb on the 24th of July.

D-MINT
20240724 0512 UTC AMSR2 930 hPa 128 kts 121 kts 135 kts
D-MINT had a peak intensity of 128 kt, 930 mb.

D-PRINT
20240724 0900 UTC 923 hPa 125 kts 118 kts 132 kts
At a similar time as D-MINT, D-PRINT peaked at 125 kt, 923 mb.

SATCON
2024 WP 05 206.167 2024JUL24 040000 23.97 -122.84 3 914 137

SATCON peaked at 137 kt, 914 mb.

Wind blend
All of the above estimates show a peak intensity in the morning of July 24 with a blend of 129 kt, 916 mb.

KZC relationship
Now before continuing I have to add that this is a very difficult storm to estimate. ADT/AiDT are significantly lower than the other estimates and these 2 methods are known to have difficulty handling pinhole eye systems, especially when the eye fixes are a little bit off. For example hurricane Wilma would only be 135 kt, 932 mb according to the automated ADT due to this issue. I still have to think of a good way to take this into account for these analyses since it is a common issue with short-lived pinhole systems. But for now, in this case, I'm leaning towards discarding both ADT and AiDT for the intensity blend. This results in a blend of of 136 kt, 916 mb.

Now I use the KZC relationship to find a potentially more accurate pressure value associated with the wind speed of 135 kt (rounded down from 136 kt since we usually work in steps of 5 kt). For this I used the following input values during peak intensity around 06:00z.
Vmax = 135 kt
C = 8 kt
R34 = 182 nm
Lat = 24.2 deg
Background pressure = 998 mb

This then results in a pressure estimate of 907 mb.

Conclusion
The official peak intensity for typhoon Gaemi is 125 kt, 919 mb. Based on a blend of all available satellite estimates and the KZC relationship, I estimate that Gaemi's peak intensity was significantly stronger at 135 kt, 907 mb in the early hours of July 24. However, as mentioned before, I have to note that Gaemi is an exceptionally difficult storm to estimate. The AMSU sounder for example shows high confidence (an uncertainty of only 12 kt) for a peak strength of a whopping 152 kt, while even the high-end uncertainty window for D-PRINT already ends at 132 kt. Such large discrepancies are unusual and the result of Gaemi's, at times, pinhole eye combined with the fact that the eye never cleared for a prolonged period of time. Since the eye is so important for the satellite fixes, the different algorithms behave in a very different manner.

Just after Gaemi passed I also made an intensity estimate (https://www.storm2k.org/phpbb2/viewtopic.php?f=95&t=124162&start=60#p3076141) where I estimated 140 kt, 901 mb after some fixes to the gale radius. Since then, more accurate values for the translational speed and R34 have become available which have nudged the estimate down a bit. And in hindsight I think I was a bit overreliant on the 152 kt AMSU data to push my estimate to 140 kt. 140 kt might be possible, but looking at the data now months later I wouldn't pull the trigger on it myself. Other estimates such as ATMS (133 kt) and SSMIS (133 kt) also hint at an intensity around 130 - 135 kt, while estimates such as CIRA ATMS (101 kt) show the difficulty to analyze Gaemi. A gradient wind balance estimate I found even showed a peak intensity in the 880s mb, so in conclusion I'd put less confidence in my estimate (or any other estimate) than usual.

[Hurricane2000]

Thanks, I think Gaemi is a very interesting storm as well, I hope one day recon will return in the WPAC