Peculiar measurements: Accumulated Cyclone Energy

Measurement Sep 10, 2024

Earlier this week I was reading something about how we've gone almost a week without any kind of tropical cyclone system in the North Atlantic, despite it being the height of the season for them. While such a lull is (only slightly) unusual, the article had noted that the 2024 Atlantic hurricane season had actually been quite active according to the "Accumulated Cyclone Energy" metric. Accumulated cycle energy? What the heck is that? So I did some light digging into whatever that metric is.

The formal definition of Accumulated Cyclone Energy (ACE) is as follows:
[ACE is] the sum of the squares of the estimated 6-hourly maximum sustained wind speed (knots) for all named systems while they are at least tropical storm strength.

There's a lot going on in that definition. First, you look at every tropical storm (storms that are not designated as tropical do not count no matter how powerful they may be). Tropical storm strength means it must have sustained winds of ≥ 34 kn; 63 km/h; 39 mph. Take that wind speed, square it (because research shows damage is more correlated with the square of wind speed). Do this for all tropical storms in the area of interest – typically an ocean, region, or the entire planet – and sum all the values. Make observations every 6 hours, and keep summing the values on a running total. Finally, divide by 10,000 to keep the numbers more manageable.

Yeah, it's a really weird metric! It's essentially a running tally of squared storm wind strength over time. It lets you more fairly compared a weak storm that hangs around for a long time with a powerful one that dissipates quickly. What's interesting is that the name sorta implies that it's measuring the amount of energy the storm is expending, but it doesn't take into account the size (aka mass) of the storm, so it's not really doing what it's name is saying I measuring "energy". Nor is it taking some integral of that energy value to "accumulate".

One of the uses of this metric seems to be in classifying how active a given hurricane season is in the North Atlantic. Taking ACE data from 1951-2020, meteorologists split the distribution of annual ACE values into 4 quartiles.

Graph of North Atlantic yearly ACE values (from Wikipedia)

It was always a bit of a mystery to me how meteorologists could say one year had a more or less active season than others, but this seems to be how they do it. It's definitely better than merely counting storms since it at least attempts to account for intensity.

Other uses for the metric is for comparing different individual storms against each other with a single number. I'm sure there's value in being able to compare storms by wind strength AND length of time with tropical storm winds.

It probably comes as no surprise that different regions of the world have different thresholds for ACE activity levels. All the oceans of the world have unique geographic aspects that mean storms form in different ways. So it stands to reason everything is considered locally.

But wait, there's more weirdness

So, having a quirky index to compare storm seasons is nice and all. But digging deeper, how is "maximum sustained winds" measured for a storm? Quirks in that measurement would have effects on the ACE metric.

So the definition of maximum sustained wind states that it's measured 10 meters above the ground (because being close to the ground can significantly slow the wind speed measured). Internationally, the wind speeds over a period of 10 minutes are averaged to find the sustained wind value. The US for whatever reason decided that they'd average wind speeds over just one minute, which can lead to about a 14% higher value than the 10-minute measurement.

Obviously with weather stations, you could measure that wind directly but most tropical storms get strong enough over vast empty ocean. How is that measured?

Direct measurement is always a possibility via planes dropping measuring devices, ships, buoys, radar, and other devices. But how do weather satellites measure wind speed? I always had assumed they used some special filters or magic satellite device to allow them to measure wind speed in a storm from orbit. But the answer is apparently... absolutely nothing of the sort.

Instead, satellite observations primarily use something called the Dvorak Technique (after Vernon Dvorak, who has nothing to do w/ the Dvorak keyboard layout). The Dvorak technique takes advantage of the fact that storms tend to evolve in appearance on satellite images in a way that is closely related to their actual strength.

Page 14 of NOAA Technical Memorandum NESS-45 A Technique For Analysis and Forecasting of Tropical Cyclones Intensities From Satellite Pictures, February 1973 (from Wikipedia)

Various observed characteristics of storms are give T-numbers. Through lots of research in the past, these T-numbers are then correlated with wind speeds, lowest air pressure, as well as the Saffir-Simpson hurricane wind scale that we're familiar with using in the US.

NOAA's conversion table for T-numbers to other properties of hurricanes.

Apparently over the history of this measurement method, they had to make tweaks to get things to agree well with reality. The current modern method seems to get within 5mph of wind speeds measured via direct plane measurement for about half the time. Considering it's largely derived from pictures of the storm structure, that's a surprising amount of accuracy. Given that many storms hang around over open ocean that's impractical to constantly send measurement aircraft missions to, it's a pretty practical compromise to get a pretty darn close approximation of what a storm is doing out in the middle of nowhere. When the storm gets closer to land and populated areas, it's much easier to directly measure the sustained wind speeds.

So yah, a curious little metric that few of us will ever need. I find it pretty fun to learn about how other fields solve very real problems about characterizing and predicting things using numbers. I certainly wasn't expecting that hurricane speeds can be estimated just by looking at the cloud patterns of the storm. It makes it so I'm slightly more willing to consider quirky index style metrics too.

Standing offer: If you created something and would like me to review or share it w/ the data community — just email me by replying to the newsletter emails.

Guest posts: If you’re interested in writing something a data-related post to either show off work, share an experience, or need help coming up with a topic, please contact me. You don’t need any special credentials or credibility to do so.

I’m Randy Au, Quantitative UX researcher, former data analyst, and general-purpose data and tech nerd. Counting Stuff is a weekly newsletter about the less-than-sexy aspects of data science, UX research and tech. With some excursions into other fun topics.

All photos/drawings used are taken/created by Randy unless otherwise credited.

randyau.com — Curated archive of evergreen posts. Under re-construction thanks to *waves at everything

All Tuesday posts to Counting Stuff are always free. The newsletter is self hosted, so support from subscribers is what makes everything possible. If you love the content, consider doing any of the following ways to support the newsletter:

Consider a paid subscription – the self-hosted server/email infra is 100% funded via subscriptions
Share posts you like with other people!
Join the Approaching Significance Discord — where data folk hang out and can talk a bit about data, and a bit about everything else. Randy moderates the discord. We keep a chill vibe.
Get merch! If shirts and stickers are more your style — There’s a survivorship bias shirt!