I've just completed a little reanalysis on Super Typhoon Tip from 1979. It's nothing crazy complex, but I figured It could give a better representation than the current best track indicates. My mock b-deck in its current state is pictured below.
Estimated maximum sustained winds rely heavily on the rounded off KZC values, but I did make a few slight up or down tweaks in a few places depending on the situation. I adjusted upwards a little bit on some of the rapid/explosive deepening phases and downwards in places where Tip had weakened after peak intensity and was dealing with eyewall replacement/structural issues. Below, I have plotted the original best track wind values, the raw KZC wind outputs, and two slightly different versions of winds expected from Atkinson-Holliday: one using the default environmental pressure and the other using my analyzed environmental pressures. As you can see, the original best track intensities closely follow the AH values for a significant stretch of Tip's life. Interestingly, it's also worth noting that the KZC and AH expected winds are basically the same for a good chunk of time following peak intensity. This is largely due to massive TS wind radii.
Speaking of TS wind radii, I actually made the average TS wind radii estimates myself based on regression between JMA's listed 30 kt wind radii and JTWC's 34 kt wind radii for the 2016 Typhoon Season. I wanted to include more seasons originally in my regression analysis, but JTWC did not quality control their 34 kt wind radii prior to 2016, and it's entirely absent prior to 2004. In stark contrast, JMA has good 30 kt wind radii data dating back to 1977, conveniently provided by Digital Typhoon
(which is also where I gathered my OCI data, as well as viewed satellite images). I used the windcode 'AAA' in my mock b-deck, meaning extending all the way around the circulation. Unlike a normal b-deck however, mine is an average rather than a maximum extent, since that is what I used with KZC. I might go back and try to do semicircle data at some point based on the JMA data. NotoSans is also putting together a hurdat type file for the WPac
, so I'll probably do a second estimate based on his wind radii when I get the chance.
Tip had good recon coverage for basically its entire life, which resulted in a good central pressure dataset. I followed it very closely, finding little reason to deviate from the pressures listed in the 1979 JTWC report
. Unfortunately, some of the biggest gaps in coverage came during Tip's two most prolific deepening phases. The second one in particular is most troubling since it involves Tip's maximum intensity. At 0353Z October 12th, 1979, recon flew into Tip for the first time in a little over 12 hours and recorded the record low 870 mb central pressure, a pressure 30 mb lower than the previous plane found. However, subsequent passes found that the pressure was rising fairly rapidly, and the pressure ended up rising back above 900 mb about as quickly as it fell. The real question is how much deeper, if any, was the pressure prior to the 870 mb pass? This is one I struggled with for a little while, but I ended up settling on an 868 mb pressure at 03Z October 12. This is after the coldest CDO cloud tops observed between 18-00Z (a time period where I nudged the winds up a little above the KZC values), but also coincides with the warmest eye temperature measured by geostationary satellite imagery, about 22ºC. Even with today's high resolution satellites, that is an extremely high value. Furthermore, the 30ºC 700 mb temperature measured by recon on the 870 mb pass indicated that Tip was likely still near peak intensity. Satellite presentation began to more quickly degrade after 03Z.
My 175 kt peak intensity estimate matches up well with the 173 kt estimated by Veldon's ADT study, which makes me optimistic that my values aren't garbage, especially since we arrived with similar numbers with very different methodologies. I'm looking forward to reanalyzing more systems in a similar fashion, but I'm not quite sure if I should try analyzing some of the other more impressive systems like Rita '78, Forrest '83, and Vanessa '84 or if I should try finishing off the other systems from 1979 like Alice, Hope, Judy, Owen, and Sarah before moving onto other systems. Regardless I'm looking forward to reanalyzing more systems!