#23 Postby TheAustinMan » Sun Sep 20, 2020 4:34 pm
ACE is just a quick way to gauge (by proxy) the total energy output of a hurricane season. Take the max winds and duration of the storms combine them in a pretty simple way, and voila, you have some number that takes into account those (and just those) inputs. Like any other index, it has its uses, and may not be useful in other cases. There's a whole soup of them, from very similar Power Dissipation Index to the number of Named Storm/Hurricane/Major Hurricane days to the more mathematically and data involved Track Integrated Kinetic Energy (TIKE).
In general, the indices try to take some aspect of seasonal activity and simplify them to a single value. That always comes at the cost of losing some information. Same goes with the PNA, PDO, AMO, NAO, SOI, IOD, ONI, or any other environmental index that commonly looks at some combination of sea surface temperature and pressure anomalies. ACE doesn't say anything about impacts, but it was never designed to. It is concerned strictly with the seasonal energetics. Any self-respecting researcher will know to go beyond the labels and check the hurricane seasons more rigorously rather than taking those indices at face value.
If we concern ourselves solely with what ACE deals with, it does tell us a few things. ACE is about 140% of the average so far this season. That's no surprise, we've seen the sheer amount of cyclogenesis this year. ACE per storm is rather low this year, and that should also come as no surprise. We've had a lot of short-lived tropical storms, so they bring down the count quite a bit. If we were to use long-term averages, a 23/8/2 season would be expected to produce about 122, while we're at about 97. That ratio is a little better than 2007 and similar to 2009. So that tells us that whatever storms we had were shorter lived than average, and that lines up with what we've seen this year, especially with a lot of the East Coast subtropical action.
But that's about all ACE tells us, and that's really all it was designed to do. It doesn't tell us about the insane rate of cyclogenesis this year—c'mon, we're at Beta entering late September and we've just about trashed every cyclogenesis record from 2005 through Beta—nor does it have any description of where the storms formed, when they did, or where they hit. But if ACE is where you decide to draw the lines for an active, inactive, average, or hyperactive year, that's perfectly fine. The raw numbers that supply the indices with their values are always going to be there for future reassessment. Personally I think TIKE does a better role of filling the same niche that ACE does, but whatever. If you make some mega weighted composite of ACE, storm counts, storm days, damage and fatalities and numbers of retired storms, maybe you'll get something useful, too. But while calculation of these indices is objective, deciding what goes into those indices is not. And no matter what you do, you'll miss out on aspect X of the season by virtue of simplifying everything down to a single number.
"Hyperactive" or not, those are just simple labels at the end, and it'd be more valuable to dive deeper into what the storms themselves did when comparing seasons. Those labels are most useful when drawing up possible analogs and studying longer term trends, but again, those require a deeper dive than just the labels. A season of 7 Caetgory 5 Hurricane Michaels would be right at the 1981-2010 median ACE. 9 of them would hit above average and you'd need 13 of them to hit the 175% median threshold for hyperactivity, at least in one definition. By design, indices like ACE don't handle abnormal cases particularly well.
10 likes
Treat my opinions with a grain of salt. For official information see your local weather service.