Behind the hurricane hype
The fifth anniversary of Hurricane Katrina, coupled with the mild hurricane Danielle tracking toward Bermuda, turns thoughts toward cyclones.
In May, before the current Atlantic hurricane season began, forecasts were for Armageddon. This year’s hurricane season could be “very active” (Jane Lubchenco, head of the National Oceanic and Atmospheric Administration) or “very very active” (CNN) or “a hell of a year” with “quite high” numbers of intense storms (William Gray, head of the hurricane prediction center at Colorado State University).
What has actually happened so far? A below-average season of two hurricanes, neither one intense.
The totals should change, though. September can be the peak month for Atlantic hurricanes. But three of the last four years have shown below-average hurricane activity in the Atlantic, and 2010 is shaping up as another below-average season.
What’s going on here? Wasn’t global warming supposed to spawn ultra-monster hurricanes? That’s what Al Gore started claiming five years ago. Similar assertions have been heard from other quarters, too.
Of course, just because Al Gore says something doesn’t mean it’s untrue. So what is going on?
The media loves predictions of deadly hurricanes. Each spring NOAA, CSU and other organizations issue hurricane forecasts; in recent years nearly all such forecasts have been alarming; the alarming forecasts are played up. If hurricane season ends without calamity, there are never follow-up stories noting the predictions were wrong, because newspapers and newscasts want to feature the next spring’s alarming predictions.
The media loves hurricanes, period. Hurricanes make great television! Fierce winds, lapping waves, correspondents in heavy rain gear shouting to be heard. Plus because hurricanes take days to form, there’s time to put newscasting resources in their paths: good luck positioning cameras in the path of a tornado. I don’t think it is too cynical to say that cable news is rooting for destructive hurricanes, though rooting for a kind of Hollywood fantasy hurricane which causes widespread cinematic-quality destruction but doesn’t kill anyone.
Predictions are worthless. The below mini-chart shows the pre-hurricane-season predictions of NOAA and CSU, followed by actual results. (I use the upper bound for NOAA, which sometimes issues vague predictions such as “three to seven” hurricanes, which is like predicting, “the Dow Jones will either rise or fall.” I use the seasons-start predictions by CSU, which has a sneaky habit of altering its forecasts once the season is nearly over and most of the trend is already known.)
|8 hurricanes, 3 intense.||8 hurricanes, 3 intense.||9 hurricanes, 6 intense.|
|9 hurricanes, 5 intense,||8 hurricanes, 4 intense.||15 hurricanes, 7 intense. (This was the year of Katrina and Rita.)|
|10 hurricanes, 6 intense.||9 hurricanes, 5 intense.||5 hurricanes, 2 intense.|
|10 hurricanes, 5 intense.||9 hurricanes, 5 intense.||6 hurricanes, 2 intense.|
|9 hurricanes, 5 intense.||8 hurricanes, 4 intense.||8 hurricanes, 5 intense.|
|7 hurricanes, 3 intense.||5 hurricanes, 2 intense.||3 hurricanes, 2 intense.|
|14 hurricanes, 7 intense.||10 hurricanes, 5 intense||So far: 2 hurricanes, neither intense.|
Note that in the six years before the current season, only two of the 12 major predictions turned out to be correct. Since the likely numerical range of hurricanes is fairly small – the 50-year average is six hurricanes, three intense – you’d think some predictions would be right by sheer chance. Instead the leading experts in the field have been wrong on 10 of their last 12 projections, and are on track to be totally wrong this year.
Predictions get attention anyway. Last year Colorado State analyzed its predictions and for the last 25 years discerned only a “modest” improvement over simply predicting every season would be average. As best I could determine, no major media outlet covered the release of this report showing that hurricane predictions are a complete waste of everyone’s time.
The sillier the prediction, the better. Colorado State has gotten press for this website, which claims to generate a scientific likelihood of hurricane landfalling by U.S. county. The page offers absurdly hyper-specific predictions, such as a “18.7 percent probability” that a hurricane will strike Harrison County, Mississippi, this year, or a “2.4 percent probability” that a hurricane will cross Essex County, New Jersey. Numbers such as these are gibberish, not science — since what’s to the right of the decimal point cannot possibly have statistical significance, and what’s to the left is pure guesswork. But the pseudo-science feel makes for a nice source of stories for local newscasters. (“Researchers Say Hurricane Might Strike Maine.”)
Hurricanes have been awful long before artificial global warming. The extremely strong Great Hurricane of 1780 killed about 27,500 people, at a time when the Western Hemisphere was far less populous — and when artificial greenhouse gas emissions were not a factor. The Category Four Galveston hurricane of 1900 killed 8,000 people, and occurred when greenhouse gas levels in the atmosphere were much lower than today. The 1938 Long Island hurricane left 800 dead and $4.6 billion in property damage (stated in 2010 dollars), also occurring before coal and oil use could have altered nature.
Hurricanes don’t show any pattern clearly linked to greenhouse gases. A terrible hurricane year (2005) or an eerily quiet year (2009) doesn’t prove anything one way or the other. Nor does the larger trend. For the last 60 years, decade-by-decade averages have been roughly the same. A good summary is here.
But you should still worry. Just because climate change hasn’t yet caused more or stronger hurricanes does not mean they will not happen. The evidence for climate change is strong. If sea surface temperatures are the key to hurricanes, as some researchers think, then hurricanes should get worse, because sea surface temperatures are rising. Then again, climate change might make cold fronts less cold, while reducing the difference between high- and low-pressure areas, and those factors could reduce hurricane incidence or intensity.
This 2005 paper by a researcher at the National Center for Atmospheric Research, published just before Katrina, takes the view that rising sea-surface temperatures will cause stronger hurricanes and also more rain. This 2009 paper, from a researcher affiliated with the same organization, takes the view that hurricane activity will decline in a warmer world.
This recently released study, from the Geophysical Fluid Dynamics Laboratory at Princeton — gotta love that name — splits down the middle, forecasting that climate change will increase hurricane activity but slowly, requiring decades or longer. (The Princeton paper is also a rare example of presenting scientific information in a way intended to be comprehensible and accessible.)
Perhaps the Princeton paper would cause you to think, “If hurricanes won’t get worse until 2050, then I don’t need to care about this.” You do.
One of the big questions of global warming is whether there will be tipping points — natural thresholds that cause climate change to accelerate. If the Princeton paper’s view is correct, by the time the tipping point for hurricanes is reached, it will be too late to reverse the effect — because greenhouse gas levels in the air will be too high.
Bottom line: instant-doomsday hurricane panic isn’t supported by science. But greenhouse gas regulation is.
Photo caption: Waves crash upon rocks as weather conditions worsen due to a low pressure system passing through the area along the coastline in Port Fourchon, Louisiana July 5, 2010. There was a “high chance” it will become the second named storm of the 2010 Atlantic hurricane season before it makes landfall in the Terrebonne Parish area near Caillou Bay early Monday evening, the U.S. National Hurricane Center said . REUTERS/Sean Gardner