Teban54 wrote:Beef Stew wrote:While we're going to have to see if there's any changes made in post-season analysis, as of now I personally believe that- owing to the most absurd dropsonde data I've ever seen- Melissa was quite possibly stronger than 160 kts/185 mph.
I wonder if Melissa will contribute to a revisit of FL-surface reductions (and maybe even SFMR's "high bias"), especially in intense storms.
With SFMR being almost completely thrown out due to the alleged high bias in high-end hurricanes, and eyewall dropsonades generally not used at face value due to measuring instantaneous winds, it seems that the community - if not the NHC - generally see FL winds as the most (or the only) reliable tool left. But Melissa had clearly gotten to the point where winds aloft were mixing very effectively down to the surface.
Are there situations where the "standard" FL reduction ratios should be modified or even discounted, and if yes, when? Should there be a more streamlined, standardized and publicized procedure for how the dropsonade data can be factored in, either for determining in-situ reduction ratios or for suggesting wind estimates directly?
(There may be such methods before - so apologies if I was simply unaware - but they're clearly not as well-known as the "standard 0.9 reductions" etc.)Perhaps more controversially... Are there situations where the "insanely high" SFMR readings actually have merit, and are not just errors in the algorithm?
My take is that if SFMR estimates are close to 90-95% FL reductions, they should be considered reliable. There are definitely cases where even unflagged measurements could be suspect, such as if they’re greater than FL winds or if they’re very inconsistent (ex: one 140+ kt SFMR reading shouldn’t be enough to upgrade something to a Cat 5). But they shouldn’t be as disregarded as they are nowadays. If dropsondes are showing effective mixing of FL winds to the surface, then unflagged SFMR readings should be taken as reliable.
The more I look at the recon data from Melissa, the more I believe it likely peaked at 170 kt sometime in the last 3 hours before landfall. Both of the final pre-landfall flights provide multiple lines of evidence supporting a peak intensity greater than 160 kt.
1.) The highest FL winds from either flight were 180 kt in the southern eyewall measured by NOAA2 (162 kt after a standard 90% reduction). Other passes in the S/SE yielded 165-170 kt FL.
2.) That infamous SE eyewall drop yielded 172 kt at the surface and 188 kt in the lowest 150m. Even if the 172 kt surface wind is treated as just a gust, a 90% reduction of the 188 kt average yields 169 kt. That drop also recorded very consistent winds all the way down, never once dropping below 170 kt, which indicates a >90% FL-to-surface conversion rate. The Dorian-esqe eye drops of sinking air reaching all the way down to sea level also support the idea of effective FL-to-surface translation.
3.) SFMR was running in the AF plane and yielded peak unflagged measurements of 170-175 kt. If we go under the assumption that the SE eyewall drop reveals a >90% FL-to-surface conversion rate, then these SFMR measurements are in line with both said drop and the peak FL winds mentioned earlier.
4.) All of these measurements were taken in either the NW or S/SE eyewalls — the presumably stronger E/NE eyewall was never sampled, likely due to safety risks. Therefore, it’s possible the strongest winds were never measured by either flight.
While I don’t think the FL-to-surface conversion rate was around 100% like Derek was saying yesterday, these upper-echelon storms are likely far more efficient at mixing their FL winds down than most other majors, perhaps more than even lower-end Cat 5s like Matthew (pointing that one out in particular because its solely SFMR-based upgrade is still a little suspect). 95% may be a better conversion factor for storms like Melissa, Dorian, and Patricia. There could be a case for 175 kt using the data and reasoning I listed, but I personally wouldn’t go that far.
