Discussion of Intense Tropical Cyclones

[MarioProtVI]

 https://twitter.com/doomhaMwx/status/1629129334257762305



More evidence for Lorenzo not being a C5 at all. Supposedly “935 mb” or whatever according to NHC but this buoy estimate indicates an intensity of only 115-120 kt.

[Yellow Evan]

That buoy supports 938 mbar or so using the scho equation. Definitely stronger than 115-120 but go ahead.

[ljmac75]

One intense storm nobody seems to have mentioned here is Cyclone George. Per the Australian BOM it peaked with a minimum central pressure of 902 mbar. The cyclone report for it does not seem to explain what that estimate is based on though, and the JTWC has it being much less intense so I don't know if it counts.

[1900hurricane]

One intense storm nobody seems to have mentioned here is Cyclone George. Per the Australian BOM it peaked with a minimum central pressure of 902 mbar. The cyclone report for it does not seem to explain what that estimate is based on though, and the JTWC has it being much less intense so I don't know if it counts.

George is one of the last storms that had a pressure assigned to it using the old Atkinson & Holliday Wind/Pressure relationship, known with the shorthand of AH77. George impacted Bedout Island as it approached the coast, where it recorded at the time the highest 10 minute sustained wind in all of Australia. Using that reading and the strengthening trend, BoM estimated a AUS C5 landfall with winds of 125 kt. JTWC on the other hand was using satellite estimates. Regardless, George was a a strong TC at landfall, but it probably did not have a 902 mb pressure.

That wind record at Bedout Island I mentioned? That was actually broken just this past April at the same recording station by Cyclone Ilsa.

[1900hurricane]

Here's a few images of George near peak intensity/landfall.





Also, here's a few of Ilsa.

[ljmac75]

One intense storm nobody seems to have mentioned here is Cyclone George. Per the Australian BOM it peaked with a minimum central pressure of 902 mbar. The cyclone report for it does not seem to explain what that estimate is based on though, and the JTWC has it being much less intense so I don't know if it counts.

George is one of the last storms that had a pressure assigned to it using the old Atkinson & Holliday Wind/Pressure relationship, known with the shorthand of AH77. George impacted Bedout Island as it approached the coast, where it recorded at the time the highest 10 minute sustained wind in all of Australia. Using that reading and the strengthening trend, BoM estimated a AUS C5 landfall with winds of 125 kt. JTWC on the other hand was using satellite estimates. Regardless, George was a a strong TC at landfall, but it probably did not have a 902 mb pressure.

That wind record at Bedout Island I mentioned? That was actually broken just this past April at the same recording station by Cyclone Ilsa.

I saw how the record was broken by Ilsa, I wonder what the highest record ten-minute speed is worldwide. Highest I can think of at the moment is Maria on Dominica at 130 kt. Glad someone was able to explain that George pressure, without recon and the sparse population I guess there's not much else that can be done to estimate pressure but base it on wind.

[CrazyC83]

One intense storm nobody seems to have mentioned here is Cyclone George. Per the Australian BOM it peaked with a minimum central pressure of 902 mbar. The cyclone report for it does not seem to explain what that estimate is based on though, and the JTWC has it being much less intense so I don't know if it counts.

George is one of the last storms that had a pressure assigned to it using the old Atkinson & Holliday Wind/Pressure relationship, known with the shorthand of AH77. George impacted Bedout Island as it approached the coast, where it recorded at the time the highest 10 minute sustained wind in all of Australia. Using that reading and the strengthening trend, BoM estimated a AUS C5 landfall with winds of 125 kt. JTWC on the other hand was using satellite estimates. Regardless, George was a a strong TC at landfall, but it probably did not have a 902 mb pressure.

That wind record at Bedout Island I mentioned? That was actually broken just this past April at the same recording station by Cyclone Ilsa.

Based on the images, I would estimate (1-min) winds around 130-135 kt. It kind of reminds me of Eta and Iota, very deep convection but not a particularly warm eye. There's no way the pressure was 902 if that was the case - such would support around 924-932 mb.

[MarioProtVI]

I know it’s been mentioned before, but with the 10th anniversary tomorrow I feel some fresh input could be necessary on Haiyan:



It really does suck that U.S.-funded reconnaissance flights ended in the WPac in the 80s because this is probably THE reason why we still need it there because of storms like Haiyan. The true intensity will never be known, but I do give credit to JTWC for putting this at 170 kt. However, I think it was likely a bit stronger (and when you compare to Patricia 2 years later), so I’d go 175 kt with maybe a slight chance for 180 kt. Pressure though was likely high because of low latitude so it’s definitely not what Dvorak was saying it was (definitely not 858), but it probably was around 880 or so. In my analysis I have it as 175 kt / 878 mb.

[kevin]

I know it’s been mentioned before, but with the 10th anniversary tomorrow I feel some fresh input could be necessary on Haiyan:

https://upload.wikimedia.org/wikipedia/commons/6/69/Typhoon_Haiyan_2013_landfall_loop.gif

It really does suck that U.S.-funded reconnaissance flights ended in the WPac in the 80s because this is probably THE reason why we still need it there because of storms like Haiyan. The true intensity will never be known, but I do give credit to JTWC for putting this at 170 kt. However, I think it was likely a bit stronger (and when you compare to Patricia 2 years later), so I’d go 175 kt with maybe a slight chance for 180 kt. Pressure though was likely high because of low latitude so it’s definitely not what Dvorak was saying it was (definitely not 858), but it probably was around 880 or so. In my analysis I have it as 175 kt / 878 mb.

With KZC and the following inputs I also get an intensity of 878 mb.

V_max = 170 kt
C (translation speed) = 22.3 kt
R34 (gale-force winds radius) = 215 nm
Latitude = 10.2 deg
P_env = 1006 mb

Pressure estimate (KZC < 18N) = 878 mb

If the true intensity was 175 kt then the pressure estimate becomes 871 mb. Because of the uncertainty between 170 and 175 kt I'd put the pressure estimate at ~875 mb. One of the most intense TCs in modern history, but imo just a little bit less intense than Tip. But indeed very unfortunate that there are no recon planes anymore. I think there are many TCs in the 870s mb (Haiyan, Surigae, Meranti) since the 1980s but we just don't know due to the lack of recon. And KZC for Patricia results in 862 - 870 mb at peak intensity in between recon flights, but I understand that NHC doesn't want to make an official world record without in-flight measurements and thus went with the slightly more conservative 872 mb. I might do a KZC reanalysis at some point to make imo more accurate pressure estimates of WPAC TCs than the ones we currently have.

[Nimbus]

I know it’s been mentioned before, but with the 10th anniversary tomorrow I feel some fresh input could be necessary on Haiyan:

https://upload.wikimedia.org/wikipedia/commons/6/69/Typhoon_Haiyan_2013_landfall_loop.gif

It really does suck that U.S.-funded reconnaissance flights ended in the WPac in the 80s because this is probably THE reason why we still need it there because of storms like Haiyan. The true intensity will never be known, but I do give credit to JTWC for putting this at 170 kt. However, I think it was likely a bit stronger (and when you compare to Patricia 2 years later), so I’d go 175 kt with maybe a slight chance for 180 kt. Pressure though was likely high because of low latitude so it’s definitely not what Dvorak was saying it was (definitely not 858), but it probably was around 880 or so. In my analysis I have it as 175 kt / 878 mb.

With KZC and the following inputs I also get an intensity of 878 mb.

V_max = 170 kt
C (translation speed) = 22.3 kt
R34 (gale-force winds radius) = 215 nm
Latitude = 10.2 deg
P_env = 1006 mb

Pressure estimate (KZC < 18N) = 878 mb

If the true intensity was 175 kt then the pressure estimate becomes 871 mb. Because of the uncertainty between 170 and 175 kt I'd put the pressure estimate at ~875 mb. One of the most intense TCs in modern history, but imo just a little bit less intense than Tip. But indeed very unfortunate that there are no recon planes anymore. I think there are many TCs in the 870s mb (Haiyan, Surigae, Meranti) since the 1980s but we just don't know due to the lack of recon. And KZC for Patricia results in 862 - 870 mb at peak intensity in between recon flights, but I understand that NHC doesn't want to make an official world record without in-flight measurements and thus went with the slightly more conservative 872 mb. I might do a KZC reanalysis at some point to make imo more accurate pressure estimates of WPAC TCs than the ones we currently have.

There may be Chinese military data from that era that is still classified, any country that regulates their internet to the point that they have to wait for the official party to open their umbrellas, probably has no need to know.

[CrazyC83]

I know it’s been mentioned before, but with the 10th anniversary tomorrow I feel some fresh input could be necessary on Haiyan:

https://upload.wikimedia.org/wikipedia/commons/6/69/Typhoon_Haiyan_2013_landfall_loop.gif

It really does suck that U.S.-funded reconnaissance flights ended in the WPac in the 80s because this is probably THE reason why we still need it there because of storms like Haiyan. The true intensity will never be known, but I do give credit to JTWC for putting this at 170 kt. However, I think it was likely a bit stronger (and when you compare to Patricia 2 years later), so I’d go 175 kt with maybe a slight chance for 180 kt. Pressure though was likely high because of low latitude so it’s definitely not what Dvorak was saying it was (definitely not 858), but it probably was around 880 or so. In my analysis I have it as 175 kt / 878 mb.

With KZC and the following inputs I also get an intensity of 878 mb.

V_max = 170 kt
C (translation speed) = 22.3 kt
R34 (gale-force winds radius) = 215 nm
Latitude = 10.2 deg
P_env = 1006 mb

Pressure estimate (KZC < 18N) = 878 mb

If the true intensity was 175 kt then the pressure estimate becomes 871 mb. Because of the uncertainty between 170 and 175 kt I'd put the pressure estimate at ~875 mb. One of the most intense TCs in modern history, but imo just a little bit less intense than Tip. But indeed very unfortunate that there are no recon planes anymore. I think there are many TCs in the 870s mb (Haiyan, Surigae, Meranti) since the 1980s but we just don't know due to the lack of recon. And KZC for Patricia results in 862 - 870 mb at peak intensity in between recon flights, but I understand that NHC doesn't want to make an official world record without in-flight measurements and thus went with the slightly more conservative 872 mb. I might do a KZC reanalysis at some point to make imo more accurate pressure estimates of WPAC TCs than the ones we currently have.

What would 185 kt for Haiyan result in with the KZC? That is my guess using the limited data from Patricia and Megi (2010), and Haiyan looked to be similar to Patricia and clearly stronger than Megi.

[kevin]

I know it’s been mentioned before, but with the 10th anniversary tomorrow I feel some fresh input could be necessary on Haiyan:

https://upload.wikimedia.org/wikipedia/commons/6/69/Typhoon_Haiyan_2013_landfall_loop.gif

It really does suck that U.S.-funded reconnaissance flights ended in the WPac in the 80s because this is probably THE reason why we still need it there because of storms like Haiyan. The true intensity will never be known, but I do give credit to JTWC for putting this at 170 kt. However, I think it was likely a bit stronger (and when you compare to Patricia 2 years later), so I’d go 175 kt with maybe a slight chance for 180 kt. Pressure though was likely high because of low latitude so it’s definitely not what Dvorak was saying it was (definitely not 858), but it probably was around 880 or so. In my analysis I have it as 175 kt / 878 mb.

With KZC and the following inputs I also get an intensity of 878 mb.

V_max = 170 kt
C (translation speed) = 22.3 kt
R34 (gale-force winds radius) = 215 nm
Latitude = 10.2 deg
P_env = 1006 mb

Pressure estimate (KZC < 18N) = 878 mb

If the true intensity was 175 kt then the pressure estimate becomes 871 mb. Because of the uncertainty between 170 and 175 kt I'd put the pressure estimate at ~875 mb. One of the most intense TCs in modern history, but imo just a little bit less intense than Tip. But indeed very unfortunate that there are no recon planes anymore. I think there are many TCs in the 870s mb (Haiyan, Surigae, Meranti) since the 1980s but we just don't know due to the lack of recon. And KZC for Patricia results in 862 - 870 mb at peak intensity in between recon flights, but I understand that NHC doesn't want to make an official world record without in-flight measurements and thus went with the slightly more conservative 872 mb. I might do a KZC reanalysis at some point to make imo more accurate pressure estimates of WPAC TCs than the ones we currently have.

What would 185 kt for Haiyan result in with the KZC? That is my guess using the limited data from Patricia and Megi (2010), and Haiyan looked to be similar to Patricia and clearly stronger than Megi.

Pressure depening on Vmax (the other variables stay the same).

170 kt = 878 mb
175 kt = 871 mb
180 kt = 864 mb
185 kt = 856 mb

[1900hurricane]

Just based on the chatter I've heard, it seems more people lean towards the higher Shimada pressure estimate than the lower ones for Haiyan.

[1900hurricane]

After a couple of years of neglect, I've started updating the "Table of Contents" in the parent post again. I'm aiming to have it up to date again in the next day or two.

And as always, the way we look at and estimate the most intense tropical cyclones continues to evolve. I've heard of some exciting new techniques for TC estimation currently being evaluated. With SFMR having as many questions as ever, the reconnaissance aircraft tail doppler radar (TDR) has shown some serious promise in assessing the vertical wind profile of a TC. I've also seen some cool new intensity techniques being developed behind closed doors. One is similar to ADT, but also factors in metrics like CDO size and radial symmetry, both of which have shown skill in intensity estimation. Another is also satellite based, but derives pressure, and makes use of data from multiple satellite bands, including the water vapor channels. I am certainly excited to see these techniques continue to develop and mature, and hopefully become publicly available at some point in the future down the road.

[Hurricane2022]

An overview of surface observations and analyses:
*Weather data = https://meteologix.com/
*Sea/land conversion factors for 10-min/1-min/gusts = https://www.researchgate.net/figure/Draft-Recommended-10m-Conversion-Factors_tbl1_237826652
*Altitude correction = https://www.nhc.noaa.gov/aboutwindprofile.shtml

Conclusion (see observations and analyses further down for details)

Note that none of the surface measurements listened below entered the eye or even the inner eyewall and still measured hurricane-force winds and adjusted gusts well over 100 kt. Based on satellite presentation and a blend of the radar analyses combined with the high confidence of the MW sounder I'd go with the lower range of the latter estimate -> 140 kt. Using the KZC analyses for this value (remaining inputs taken from JMA) results in a pressure estimate of 893 mb (input: 140 kt, 23.7N, 10 kt translation speed, 323 nm TS winds, 998 mb environmental pressure). Based on these analyses I'd conclude that Gaemi's official peak intensity is significantly underestimated and is closer to 893 mb / 140 kt than the current 940 mb / 125 kt. Even if Gaemi's wind speed is 125 kt, the storm's latitude, translational speed and especially its anomalously large gale radius would still result in a very low pressure of 909 mb instead of 940 mb.

Surface observations

Sustained winds

Yonagunijima Island (30m)
07:00AM CEST
10-min average wind speed = 122 kmh / 66 kt
-> Converts to 132 kmh / 71 kt for 1-minute sea-level winds (1.16 factor for 1-minute & 1/1.07 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/wind-average-10min/20240724-0500z.html#obs-detail-TA1280

Lan Yu (325m)
11:00AM CEST
10-min average wind speed = 163 kmh / 88 kt
-> Converts to 144 kmh / 78 kt for 1-minute sea-level winds (1.16 factor for 1-minute & 1/1.31 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/wind-average-10min/20240724-0600z.html#obs-detail-595670

Pengjia Yu (102m)
11:00AM CEST
10-min average wind speed = 133 kmh / 72 kt
-> Converts to 130 kmh / 70 kt for 1-minute sea-level winds (1.16 factor for 1-minute & 1/1.19 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/wind-average-10min/20240724-0900z.html#obs-detail-589740

Gusts

Yonagunijima Island (30m)
07:00AM CEST
Maximum 10-minute gust = 181 kmh / 98 kt
-> Converts to 230 kmh / 124 kt for instantaneous sea-level gusts (1.36 factor for 1-minute & 1/1.07 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/gusts-10min/20240724-0500z.html#obs-detail-TA1280

Tokorono (15m)
06:30AM CEST
Maximum 10-minute gust = 165 kmh / 89 kt
-> Converts to 218 kmh / 118 kt for instantaneous sea-level gusts (1.36 factor for 1-minute & 1/1.03 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/gusts-10min/20240724-0500z.html#obs-detail-TA1280

Yonaguni Airport (19m)
06:23AM CEST
Maximum 10-minute gust = 165 kmh / 89 kt
-> Converts to 218 kmh / 118 kt for instantaneous sea-level gusts (1.36 factor for 1-minute & 1/1.03 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/gusts-10min/20240724-0500z.html#obs-detail-TA1280

Sea-level air pressure QFF

Yonagunijima Island (30m)
06:40AM CEST
Mininum sea-level pressure = 967.4 mb
https://meteologix.com/nl/observations/1220-e-213-n/pressure-qff/20240724-0700z.html#obs-detail-TA1280

(...)

MW Sounders (CIMSS AMSU)
896 mb / 152 kt at 02:58z

The fact that Gaemi likely reached a peak intensity of around 140 kt and minimum pressure quite possibly sub-900 mb makes me think about other large and powerful systems in this same basin that performed much better, such as Typhoon Dianmu. Like, assuming that Gaemi, a typhoon with the appearance of a low-end C4, reached something around 140/895, what can be said (again) about wide and deep typhoons like the aforementioned Dianmu '04, Hagibis '19, Nuri '14, Bolaven '23...?

[kevin]

An overview of surface observations and analyses:
*Weather data = https://meteologix.com/
*Sea/land conversion factors for 10-min/1-min/gusts = https://www.researchgate.net/figure/Draft-Recommended-10m-Conversion-Factors_tbl1_237826652
*Altitude correction = https://www.nhc.noaa.gov/aboutwindprofile.shtml

Conclusion (see observations and analyses further down for details)

Note that none of the surface measurements listened below entered the eye or even the inner eyewall and still measured hurricane-force winds and adjusted gusts well over 100 kt. Based on satellite presentation and a blend of the radar analyses combined with the high confidence of the MW sounder I'd go with the lower range of the latter estimate -> 140 kt. Using the KZC analyses for this value (remaining inputs taken from JMA) results in a pressure estimate of 893 mb (input: 140 kt, 23.7N, 10 kt translation speed, 323 nm TS winds, 998 mb environmental pressure). Based on these analyses I'd conclude that Gaemi's official peak intensity is significantly underestimated and is closer to 893 mb / 140 kt than the current 940 mb / 125 kt. Even if Gaemi's wind speed is 125 kt, the storm's latitude, translational speed and especially its anomalously large gale radius would still result in a very low pressure of 909 mb instead of 940 mb.

Surface observations

Sustained winds

Yonagunijima Island (30m)
07:00AM CEST
10-min average wind speed = 122 kmh / 66 kt
-> Converts to 132 kmh / 71 kt for 1-minute sea-level winds (1.16 factor for 1-minute & 1/1.07 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/wind-average-10min/20240724-0500z.html#obs-detail-TA1280

Lan Yu (325m)
11:00AM CEST
10-min average wind speed = 163 kmh / 88 kt
-> Converts to 144 kmh / 78 kt for 1-minute sea-level winds (1.16 factor for 1-minute & 1/1.31 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/wind-average-10min/20240724-0600z.html#obs-detail-595670

Pengjia Yu (102m)
11:00AM CEST
10-min average wind speed = 133 kmh / 72 kt
-> Converts to 130 kmh / 70 kt for 1-minute sea-level winds (1.16 factor for 1-minute & 1/1.19 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/wind-average-10min/20240724-0900z.html#obs-detail-589740

Gusts

Yonagunijima Island (30m)
07:00AM CEST
Maximum 10-minute gust = 181 kmh / 98 kt
-> Converts to 230 kmh / 124 kt for instantaneous sea-level gusts (1.36 factor for 1-minute & 1/1.07 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/gusts-10min/20240724-0500z.html#obs-detail-TA1280

Tokorono (15m)
06:30AM CEST
Maximum 10-minute gust = 165 kmh / 89 kt
-> Converts to 218 kmh / 118 kt for instantaneous sea-level gusts (1.36 factor for 1-minute & 1/1.03 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/gusts-10min/20240724-0500z.html#obs-detail-TA1280

Yonaguni Airport (19m)
06:23AM CEST
Maximum 10-minute gust = 165 kmh / 89 kt
-> Converts to 218 kmh / 118 kt for instantaneous sea-level gusts (1.36 factor for 1-minute & 1/1.03 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/gusts-10min/20240724-0500z.html#obs-detail-TA1280

Sea-level air pressure QFF

Yonagunijima Island (30m)
06:40AM CEST
Mininum sea-level pressure = 967.4 mb
https://meteologix.com/nl/observations/1220-e-213-n/pressure-qff/20240724-0700z.html#obs-detail-TA1280

(...)

MW Sounders (CIMSS AMSU)
896 mb / 152 kt at 02:58z

The fact that Gaemi likely reached a peak intensity of around 140 kt and minimum pressure quite possibly sub-900 mb makes me think about other large and powerful systems in this same basin that performed much better, such as Typhoon Dianmu. Like, assuming that Gaemi, a typhoon with the appearance of a low-end C4, reached something around 140/895, what can be said (again) about wide and deep typhoons like the aforementioned Dianmu '04, Hagibis '19, Nuri '14, Bolaven '23...?

The reason for Gaemi's extreme estimated KZC pressure despite an intensity of 'only' 140 kt is its large gale radius which with 375/270 nm (avg of 323 nm) is above that of iconic storms like typhoon Gay and just below storms like typhoon Mangkhut. We can never be sure of past intensities without 24/7 recon, but here are the KZC estimates for the storms you just mentioned. Note that the largest uncertainty factor with KZC estimates in the WPAC is the wind speed input which, when it is also underestimated, can of course strongly affect the outcome. The other inputs are always quite well-known. Sometimes the one factor I still have my doubts about is the R34 gale radius by JMA which seems a bit high for some storms and could thus cause them to be overestimated. On a side note, one of the ways in which KZC could still be improved is by also including a RMW variable in addition to or as a replacement of the R34 gale radius.

For more information about KZC I'd recommend this paper: https://www.google.com/url?sa=t&source= ... d3m32X6Xpu

Typhoon Dianmu (2004)


Estimated KZC intensity: 890 mb / 155 kt
Edit with 0.65 factor for gale radius: 896 mb / 155 kt
Given intensity: 915 mb
Input: 155 kt, 8.6 kt C, 300 nm, 15.0 deg lat, 1006 mb
Notes: calculated intensity is more in-line with the SSHWS/JTWC estimate of 879 mb.

Typhoon Hagibis (2019)


Estimated KZC intensity: 885 mb / 160 kt
Edit with 0.65 factor for gale radius: 892 mb / 160 kt
Given intensity: 915 mb
Input: 160 kt, 11.7 kt C, 295 nm, 16.9 deg lat, 1008 mb
Notes: blend of >18N KZC and <18N KZC due to proximity to 18 deg latitude. This would make Hagibis one of the most intense cyclones in the historical record, which imo doesn't seem that weird considering its appearance. Hagibis' secondary peak around October 10 might even have a lower intensity in the upper 870s using KZC (inputs 160 kt, 12.1 kt C, 375 nm, 23.2 lat, 1008 mb), but I personally am not so confident of the wind estimates during that time (imo no longer 160 kt at that point), so that's why this calculation is based on the initial October 7 peak.

Typhoon Nuri (2014)


Estimated KZC intensity: 895 mb / 155 kt
Edit with 0.65 factor for gale radius: 900 mb / 155 kt
Given intensity: 910 mb
Input: 155 kt, 6.1 kt C, 270 nm, 18.4 deg lat, 1011 mb
Notes: blend of >18N KZC and <18N KZC due to proximity to 18 deg latitude. Calculated intensity is also significantly more intense than SSHWS/JTWC's estimate of 907 mb.

Typhoon Bolaven (2023)


Estimated KZC intensity: 892 mb / 155 kt
Edit with 0.65 factor for gale radius: 900 mb / 155 kt
Given intensity: 905 mb
Input: 155 kt, 9.1 kt C, 270 nm, 20.6 deg lat, 1010 mb
Notes: -

Edit: posted a new intensity estimate for all typhoons using a 0.65 conversion factor from JMA R30 to NOAA R34 for the gale radius.

[kevin]

I also performed the KZC analysis for the 2 strongest hurricanes in the Atlantic basin, Wilma and Gilbert. Background pressure is taken from the ECMWF re-analysis, while other data is taken from the official NHC reports. I might perform a more extensive analysis with some Python code and a whole database of TCs, but if I do that (if I ever have the time for it) I'll make a separate thread. One thing I'm noticing is that one of the main reasons why Atlantic hurricanes struggle to go below 900 mb compared to WPAC typhoons is their lower gale radius. The western part of the Atlantic simply has more land obstacles which prevent hurricanes from reaching the monstrous sizes of storms like Mangkhut and Gaemi. Even monsters like Wilma and Gilbert below don't have the R34 of most standard super typhoons. And a smaller size means less of a pressure gradient away from the background pressure and as such higher winds or more favorable environmental conditions are required for most NATL TCs to reach sub-900mb pressures than is the case for WPAC TCs. Just look at the image below of the WCar and Gaemi side by side with the same scale. For Gaemi to fit in the warmest portion of the WCar part of the storm would have to be over Florida and South America at the same time.



Hurricane Wilma (2005)

Estimated KZC intensity: 890 mb
Official intensity: 882 mb
Input: 160 kt, 6.1 kt C, 140 nm, 17.3 deg lat, 1005 mb
Notes: interestingly, KZC slightly underestimates Wilma's intensity. I think this is because of two reasons. Firstly, Wilma's unusually small eye resulted in an extremely steep pressure gradient, which I expect is not really taken in to account in the KZC equations. Furthermore, the 140 nm R34 from the NHC advisory feels a little on the small side imo when compared to that of Gilbert and other hurricanes. Afaik the 140 nm value was based on a recon pass, so maybe the pass wasn't perfect and there was 150 or 160 nm R34 along another axis. Either way, that could make a 1 - 2 mb difference.

Hurricane Gilbert (1988)

Estimated KZC intensity: 888 mb
Official intensity: 888 mb
Input: 160 kt, 13.3 kt C, 217 nm, 19.7 deg lat, 1006 mb
Notes: looks like KZC is spot-on here. It makes sense as well since Gilbert was a stable and typically-shaped high-end hurricane (unlike Wilma with its tiny eye).

[Hurricane2022]

I also performed the KZC analysis for the 2 strongest hurricanes in the Atlantic basin, Wilma and Gilbert. Background pressure is taken from the ECMWF re-analysis, while other data is taken from the official NHC reports. I might perform a more extensive analysis with some Python code and a whole database of TCs, but if I do that (if I ever have the time for it) I'll make a separate thread. One thing I'm noticing is that one of the main reasons why Atlantic hurricanes struggle to go below 900 mb compared to WPAC typhoons is their lower gale radius. The western part of the Atlantic simply has more land obstacles which prevent hurricanes from reaching the monstrous sizes of storms like Mangkhut and Gaemi. Even monsters like Wilma and Gilbert below don't have the R34 of most standard super typhoons. And a smaller size means less of a pressure gradient away from the background pressure and as such higher winds or more favorable environmental conditions are required for most NATL TCs to reach sub-900mb pressures than is the case for WPAC TCs. Just look at the image below of the WCar and Gaemi side by side with the same scale. For Gaemi to fit in the warmest portion of the WCar part of the storm would have to be over Florida and South America at the same time.

https://i.imgur.com/H8dUxSm.png

Hurricane Wilma (2005)

Estimated KZC intensity: 890 mb
Official intensity: 882 mb
Input: 160 kt, 6.1 kt C, 140 nm, 17.3 deg lat, 1005 mb
Notes: interestingly, KZC slightly underestimates Wilma's intensity. I think this is because of two reasons. Firstly, Wilma's unusually small eye resulted in an extremely steep pressure gradient, which I expect is not really taken in to account in the KZC equations. Furthermore, the 140 nm R34 from the NHC advisory feels a little on the small side imo when compared to that of Gilbert and other hurricanes. Afaik the 140 nm value was based on a recon pass, so maybe the pass wasn't perfect and there was 150 or 160 nm R34 along another axis. Either way, that could make a 1 - 2 mb difference.

Gaemi's eye was also very unstable at peak, with eye temps flutuating between -65 and -35°C, never reaching anything close to 0°C and it wobbled many times as well. So maybe it's possible that Gaemi reached ~883mb at peak like this guy below estimates?
 https://x.com/ak_1515/status/1816050341646901627



÷÷÷÷÷÷÷÷
 https://x.com/CyanideCN_/status/1816371094099841383

[1900hurricane]

An overview of surface observations and analyses:
*Weather data = https://meteologix.com/
*Sea/land conversion factors for 10-min/1-min/gusts = https://www.researchgate.net/figure/Draft-Recommended-10m-Conversion-Factors_tbl1_237826652
*Altitude correction = https://www.nhc.noaa.gov/aboutwindprofile.shtml

Conclusion (see observations and analyses further down for details)

Note that none of the surface measurements listened below entered the eye or even the inner eyewall and still measured hurricane-force winds and adjusted gusts well over 100 kt. Based on satellite presentation and a blend of the radar analyses combined with the high confidence of the MW sounder I'd go with the lower range of the latter estimate -> 140 kt. Using the KZC analyses for this value (remaining inputs taken from JMA) results in a pressure estimate of 893 mb (input: 140 kt, 23.7N, 10 kt translation speed, 323 nm TS winds, 998 mb environmental pressure). Based on these analyses I'd conclude that Gaemi's official peak intensity is significantly underestimated and is closer to 893 mb / 140 kt than the current 940 mb / 125 kt. Even if Gaemi's wind speed is 125 kt, the storm's latitude, translational speed and especially its anomalously large gale radius would still result in a very low pressure of 909 mb instead of 940 mb.

Surface observations

Sustained winds

Yonagunijima Island (30m)
07:00AM CEST
10-min average wind speed = 122 kmh / 66 kt
-> Converts to 132 kmh / 71 kt for 1-minute sea-level winds (1.16 factor for 1-minute & 1/1.07 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/wind-average-10min/20240724-0500z.html#obs-detail-TA1280

Lan Yu (325m)
11:00AM CEST
10-min average wind speed = 163 kmh / 88 kt
-> Converts to 144 kmh / 78 kt for 1-minute sea-level winds (1.16 factor for 1-minute & 1/1.31 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/wind-average-10min/20240724-0600z.html#obs-detail-595670

Pengjia Yu (102m)
11:00AM CEST
10-min average wind speed = 133 kmh / 72 kt
-> Converts to 130 kmh / 70 kt for 1-minute sea-level winds (1.16 factor for 1-minute & 1/1.19 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/wind-average-10min/20240724-0900z.html#obs-detail-589740

Gusts

Yonagunijima Island (30m)
07:00AM CEST
Maximum 10-minute gust = 181 kmh / 98 kt
-> Converts to 230 kmh / 124 kt for instantaneous sea-level gusts (1.36 factor for 1-minute & 1/1.07 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/gusts-10min/20240724-0500z.html#obs-detail-TA1280

Tokorono (15m)
06:30AM CEST
Maximum 10-minute gust = 165 kmh / 89 kt
-> Converts to 218 kmh / 118 kt for instantaneous sea-level gusts (1.36 factor for 1-minute & 1/1.03 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/gusts-10min/20240724-0500z.html#obs-detail-TA1280

Yonaguni Airport (19m)
06:23AM CEST
Maximum 10-minute gust = 165 kmh / 89 kt
-> Converts to 218 kmh / 118 kt for instantaneous sea-level gusts (1.36 factor for 1-minute & 1/1.03 factor for altitude correction)
https://meteologix.com/nl/observations/1220-e-213-n/gusts-10min/20240724-0500z.html#obs-detail-TA1280

Sea-level air pressure QFF

Yonagunijima Island (30m)
06:40AM CEST
Mininum sea-level pressure = 967.4 mb
https://meteologix.com/nl/observations/1220-e-213-n/pressure-qff/20240724-0700z.html#obs-detail-TA1280

(...)

MW Sounders (CIMSS AMSU)
896 mb / 152 kt at 02:58z

The fact that Gaemi likely reached a peak intensity of around 140 kt and minimum pressure quite possibly sub-900 mb makes me think about other large and powerful systems in this same basin that performed much better, such as Typhoon Dianmu. Like, assuming that Gaemi, a typhoon with the appearance of a low-end C4, reached something around 140/895, what can be said (again) about wide and deep typhoons like the aforementioned Dianmu '04, Hagibis '19, Nuri '14, Bolaven '23...?

The reason for Gaemi's extreme estimated KZC pressure despite an intensity of 'only' 140 kt is its large gale radius which with 375/270 nm (avg of 323 nm) is above that of iconic storms like typhoon Gay and just below storms like typhoon Mangkhut. We can never be sure of past intensities without 24/7 recon, but here are the KZC estimates for the storms you just mentioned. Note that the largest uncertainty factor with KZC estimates in the WPAC is the wind speed input which, when it is also underestimated, can of course strongly affect the outcome. The other inputs are always quite well-known. Sometimes the one factor I still have my doubts about is the R34 gale radius by JMA which seems a bit high for some storms and could thus cause them to be overestimated. On a side note, one of the ways in which KZC could still be improved is by also including a RMW variable in addition to or as a replacement of the R34 gale radius.

For more information about KZC I'd recommend this paper: https://www.google.com/url?sa=t&source= ... d3m32X6Xpu

Typhoon Dianmu (2004)
https://i.imgur.com/fOZmjfl.jpeg

Estimated KZC intensity: 890 mb / 155 kt
Given intensity: 915 mb
Input: 155 kt, 8.6 kt C, 300 nm, 15.0 deg lat, 1006 mb
Notes: calculated intensity is more in-line with the SSHWS/JTWC estimate of 879 mb.

Typhoon Hagibis (2019)
https://i.imgur.com/YKCUYSd.jpeg

Estimated KZC intensity: 885 mb / 160 kt
Given intensity: 915 mb
Input: 160 kt, 11.7 kt C, 295 nm, 16.9 deg lat, 1008 mb
Notes: blend of >18N KZC and <18N KZC due to proximity to 18 deg latitude. This would make Hagibis one of the most intense cyclones in the historical record, which imo doesn't seem that weird considering its appearance. Hagibis' secondary peak around October 10 might even have a lower intensity in the upper 870s using KZC (inputs 160 kt, 12.1 kt C, 375 nm, 23.2 lat, 1008 mb), but I personally am not so confident of the wind estimates during that time (imo no longer 160 kt at that point), so that's why this calculation is based on the initial October 7 peak.

Typhoon Nuri (2014)
https://i.imgur.com/IkibXla.jpeg

Estimated KZC intensity: 895 mb / 155 kt
Given intensity: 910 mb
Input: 155 kt, 6.1 kt C, 270 nm, 18.4 deg lat, 1011 mb
Notes: blend of >18N KZC and <18N KZC due to proximity to 18 deg latitude. Calculated intensity is also significantly more intense than SSHWS/JTWC's estimate of 907 mb.

Typhoon Bolaven (2023)
https://i.imgur.com/ySROBw9.jpeg

Estimated KZC intensity: 892 mb / 155 kt
Given intensity: 905 mb
Input: 155 kt, 9.1 kt C, 270 nm, 20.6 deg lat, 1010 mb
Notes: -

JMA lists 30 kt radii instead of 34 kt, so the JMA radii listed are going to be larger. I ended up creating a conversion factor by correlating the JMA and JTWC radii at the same time, but it breaks down for older systems before at least Quickscat in 1999, which almost always have listed JMA radii which must be way too large. It would be nice to have an apples to apples comparison, but that really isn't possible using that data.

[MarioProtVI]

So JTWC just released their BT for 2023, and three major changes were made in regards to three of the biggest storms we had:

- Saola is now a C5 rightfully, up to 140 kt from 135 kt originally.
- Mawar and Bolaven are both now 165 kt, up from 160 kt and 155 kt respectively. Seems like JTWC really liked the T8.0 fixes that we got with these two. This now I think makes 2023 the only year to have two 165 kt storms on record I believe. Amazing.