comments_image Comments

The Link Between Deadly Weather and Global Warming Is Real -- and Conservatives Can't Just Wish It Away

Does extreme weather foretell the kind of climate that global warming is bringing us? Scientists say yes.

Continued from previous page


Extreme events are, almost by definition, relatively rare, and thus not the best place to look for strong scientific evidence. So the fact that they do produce such evidence ought to be more alarming. There are, for example, a growing number of studies showing increased hurricane intensity linked to global warming. These have been appearing for years now, with remarkably little notice in the corporate media, even when the timing and connections are stunning.  

For example, on July 31, 2005―a month before Hurricane Katrina devastated New Orleans―in the online edition of Nature, MIT meteorologist Kerry Emanuel reported dramatic increases in the amount of energy released in hurricanes in both the North Atlantic and the North Pacific oceans since the mid-1970s. Emanuel found that both the duration and highest wind speeds had increased by about 50 percent over the past 50 years. "My results suggest... a substantial increase in hurricane-related losses in the 21st century," he said.  

Then, shortly after Katrina, in the September 16 issue of  Science, four researchers published a study showing "A large increase... in the number and proportion of hurricanes reaching categories 4 and 5," over the previous 35 years. While American attention was focused on the North Atlantic, the study found that "The largest increase occurred in the North Pacific, Indian, and Southwest Pacific Oceans, and the smallest percentage increase occurred in the North Atlantic Ocean."  

Of course, hurricanes are extremely large, relatively infrequent storms.  There may be two or more at one time near maturity in any given oceanic region during their respective hurricane seasons, but that's about it. Ordinary rainstorms are vastly more common, but evidence concerning increased rainfall in the most intense one percent of such storms is equally compelling, according to Ekwurzel. For U.S. as a whole, there's “an average of 20 percent more extreme precipitation in your heaviest one percent” of rainstorms. But the increases are most significant in two regions: “the Northeast would be 57 percent  more, and in the Midwest about 31 percent on average heavier for the heaviest rainfall,”Ekwurzel said. “So essentially, the flood risk has gone up.” 

With all this in mind, tornadoes are different from other extreme storm events in at least two main respects. First, the data for them is a good deal more sketchy than for other storms.  Rainfall data is amongst the most basic and long-recorded of any form of weather observation. Tornado data is far sketchier, though much more comprehensive than it used to be―which makes comparing recent data to that of 50 years ago a bit dodgy at best. Second, unlike other extreme storms, global warming doesn't just increase their likelihood by increasing temperature and humidity. Global warming also has an opposite effect: by heating polar regions faster than the tropics, it reduces windshear on average.   

While this dampens the relationship with global warming, it remains to be seen by how much. A 2007 modeling study lead by Anthony D. Del Genio of NASA's GISS, found that “For the central-eastern United States, stronger updrafts combined with weaker wind shear suggest little change in severe storm occurrence with warming, but the most severe storms occur more often.” 

The Challenges of Tornadoes and Modeling 

There's another challenge in studying the relationship between global warming and tornadoes: current climate models don't have the small scale resolution needed to predict tornadoes, or other severe storms on the same scale. The best that can be done is to look at preconditions, and compare those with observational data and smaller-scale high-resolution models. Still, even with these uncertainties, the future predictions are troubling.  

See more stories tagged with: