So, there’s a tropical storm out in the Atlantic.
Or, at least there was one recently; the storm named Arthur (not to be confused with other storms named Arthur in recent years). Although Arthur itself was not especially destructive and never achieved hurricane status, it’s remarkable in that hurricane season won’t actually start for another week and a half. It’s May, a time of year when the Atlantic Ocean is supposedly too cold still to feed this kind of storm.
So, this here is a slam-dunk bit of evidence of climate change, right?
Well, it is and it isn’t.
Trying to Reason with Hurricane Season
(Yes, I titled this section with the name of a Jimmy Buffet song. You can go hear the song here)
Six years ago, I wrote a post on hurricanes and climate change that did a good job of explaining certain basics. Otherwise unattributed quotes come from that post.
“Hurricane” technically refers to only one subset of a whole category of storms that share the same structure.
“Tropical cyclone” is the generic term that covers tropical storms, hurricanes, typhoons, and cyclones. All these storms have a distinct eye and draw their energy from the evaporation of water, rather than from temperature differences between adjacent air masses as extra-tropical cyclones do.
“Tropical storm” refers to a tropical cyclone with sustained winds of anywhere from 39 MPH to 74 MPH. Once a storm intensifies to 75 MPH or beyond, it is called a typhoon in the Northwest Pacific, a cyclone in the South Pacific or the Indian Ocean, and a hurricane everywhere else. I have not found any explanation for this diversity of names for the same kind of storm. Perhaps it is a relic from a time before we knew they were all the same.
So what we normally call the hurricane season should be called the tropical cyclone season–after all, Arthur wasn’t a hurricane, but its formation outside of the season still attracts attention.
Each storm basin has its own season. In the North Atlantic, the season officially runs from June 1st to November 30th, but tropical cyclones outside of those dates in other parts of the world aren’t necessarily remarkable.
On May 16th, 2020, a large, multi-day rainstorm in the Florida Strait was recognized as a tropical depression, meaning it had an eye and drew its energy from the evaporation of water. It was thus the first tropical cyclone of the 2020 North Atlantic season. A few hours later, its sustained winds topped 39 MPH, making it a tropical storm. It was given the name Arthur–every tropical storm found in the North Atlantic gets a name from an alphabetized list of alternating male and female names, so the first storm of this year would have been named Arthur no matter when it occurred. Each list gets re-used every six years (indeed, I remember being rained on by the last Arthur), although the names of particularly notable storms are retired. There will never be another Katrina, for example.
This year’s Arthur moved north as its winds intensified to around 50 mph. It did not make landfall but brushed the Outer Banks before heading further out to sea and then, oddly, turning south towards Bermuda. On May 19th, the storm’s designation was changed to “post-tropical cyclone,” as it was no longer gaining strength from evaporating water. However, a storm does not need to be tropical or dangerous, and Arthur’s story is not necessarily over yet, as of this writing.
Arthur is not the first North Atlantic tropical cyclone to occur in May. In fact, tropical cyclones can form in the Atlantic any month of the year–and have. Hurricane season is not a law of physics but rather a rule of them; meteorologists, government officials, tourist agents, and anyone else who needs to think about the likelihood of hurricanes know it’s best to keep an eye out from June through the end of November. The occasional unseasonal storm doesn’t change the pattern, especially since out-of-season storms are usually weak and rarely make landfall.
But this is the sixth year in a row that the first named storm has occurred before June 1st.
2016 was particularly odd, as it ha two pre-season named storms, the first an actual January hurricane. But over the past 17 years, nine have had at least one pre-season North Atlantic tropical cyclone.
We’re at the point where meteorologists are starting to talk seriously about extending the season, though the change hasn’t been formally proposed, yet. The arguments for and against are interesting in several different ways.
The argument for is fairly clear; if tropical cyclones often form in May, then shouldn’t the season start in May?
The arguments against are several:
- We don’t know yet that May storms are actually typical. We could have a few unusual years in a row by chance, in which case we could next have a decade or so of late first storms. In that case, an earlier start to the official season will be both silly and confusing.
- It’s possible that May storms are typical, and have always been typical, we just didn’t notice most of them until we started tracking storms using satellites. The early storms we see these days tend to be weak and of short duration, and they don’t often make landfall, meaning that there could have been lots of similar May and even April storms in the past that nobody knew about. The point of having a hurricane season has never been to include all months when tropical cyclones can happen–nobody is proposing extending the season to include January and December. The point is to include the months when these storms are likely to become problems. Maybe May storms aren’t usually problems.
- If we changed the hurricane season, someone might think climate change is real.
More on that last point shortly.
In an article about Tropical Storm Arthur and other early storms, the Florida Sun Sentinel recently quoted a meteorologist as saying he could understand not wanting to change the season “because you’d suddenly get all these existential political arguments about oh they’re just doing that because of climate change or something.”
A Closer Look at Cons
At first glance, that quote about not changing hurricane season dates really does sound climate-denial-ish, and in fact I don’t know that it isn’t meant that way. I can believe there are those who don’t want to change the season because they don’t want to appear to believe in climate change. But I don’t know that this meteorologist meant it that way–and that’s why I’m not including his name here. You can find his name in two seconds by clicking on the link to the article, but it’s possible the article takes his words out of context.
Climate change is real, but it’s difficult to demonstrate that fact using hurricane data alone.
Tropical cyclone records are being studied, but the problem is the data are “noisy.” That is, there are so many variations that are not related to the greenhouse effect that it’s hard to spot the variations that are….Some of the noise in tropical cyclone data is the natural variability in storminess from year to year. Normally scientists can tune out such noise by looking at a large enough dataset. The basic procedure is to let random variations cancel themselves out–years with a lot of hurricanes are balanced by years with very few, if you look at enough years. What variation doesn’t get cancelled out is actually the climate changing.
But with tropical cyclones that standard procedure doesn’t work very well because there are problems with the data:
- We don’t have good records of tropical cyclones before the Industrial Revolution. Scientists only started realizing that some large storms are spirals around 1820. Modern weather forecasting based on networks of weather stations didn’t begin until the 1860’s and most of the technology used to monitor hurricanes was only invented in the 20th century. It’s hard to do a before-and-after comparison if you have no “before” shot.
- The United States has been conducting aerial reconnaissance on hurricanes for decades, but since similar flights into typhoons have stopped, the data on storms in different parts of the world are not directly comparable. That makes it hard to really get a global picture.
- A lot of research on tropical cyclones is done by satellite, especially in the Pacific, but satellites are a relatively new technology so, again, we don’t have a good picture of how storms change over time.
- Which information we get about which storm is a little random. For example, getting a measurement of a storm’s highest winds at landfall depends on getting the right instrumentation into the right part of the storm at the right time. For obvious reasons, that doesn’t always happen.
- The conventions on how researchers analyze data and how they make estimates can change, subtly but definitely changing the numbers they record.
Scientists can and do work around these limitations, but they can’t make the limitations vanish.
And while it seems like a no-brainer that a warmer world will have more tropical cyclones, hot water is not the only requirement for storm formation; certain atmospheric conditions are also necessary, and some models show the frequency of these conditions–and thus the frequency of tropical cyclones–holding steady or even decreasing.
So while climate change is real, it’s far from clear that increased pre-season storm activity is related–or even happening at all. Whatever’s happening with early tropical storms might have nothing to do with climate change and much more to do with figuring out which rules-of-thumb are useful for disaster preparedness. And it’s easy to imagine even scientists who fully support climate action being irritated by having their work misinterpreted by climate activists.
But regardless of what that one un-named meteorologist meant when quoted by the Sun-Sentinel, some of the articles I’ve been finding on early tropical cyclones seem a bit disingenuous, being focused on the idea that the links between climate change and tropical cyclones is unclear and anyway these storms are usually quite weak and barely tropical in structure at all.
“Weak” and “barely tropical” don’t actually mean much, for one thing.
Weak, in a tropical cyclone, generally means it doesn’t have very high maximum sustained wind speeds. Arthur’s winds, for example, never exceeded 74 MPH, so it never counted as a hurricane. But wind speed is not as important as we might assume; most of the death and destruction in these storms is caused by flooding, not by wind. So the fact that pre-season storms rarely develop windspeeds over 74 MPH doesn’t tell us much. I want to know how big they are, how much rain they carry, and how slowly they move–all information not provided by most reports. Even tropical characteristics are not necessary for a storm to be dangerous. Nor’easters, which are non-tropical cyclones, can be as destructive as hurricanes because they can cause as much flooding, and their more moderate winds can cover a very large area. So I don’t know what “barely tropical” means, but it’s not comforting.
Finally, the connection between tropical cyclones and climate change is no longer as mysterious as it seemed when I wrote my posts on the subject back in 2014. Yes, the data of the past are still noisy, but new research methods are starting to give us a much clearer picture, and the picture isn’t pretty. No, we still don’t know whether early-season storms are, in general, a sign of climate change, Arthur particularly developed in unusually warm water. That is, the storm didn’t occur in typical-May conditions that we just didn’t know could produce tropical cyclones, nor was it the result of unusual atmospheric conditions that might have occurred irrespective of water temperatures. We had a tropical storm in May because ocean temperatures more closely resembled those of June.
It behooves us to think carefully, to not jump to conclusions, to not assume that a storm in May is a sign of the Apocalypse. But it also behooves us not to ignore the fact that climate change is making the ocean warmer–and it seems that whenever an unusual tropical cyclone occurs, unusual water is below it.