Driving along a US highway, this is not something you want to see out your side window. The United States has the highest rate of tornadoes per year compared to any other country, so unfortunately this is something many US citizens worry about. But how does climate change affect how often we’ll be seeing funnel clouds? A new study in Climate Dynamics indicates that, although we see less total tornadoes per year, they are appearing in greater density and severity, leading to more overall destruction in recent years.In the past, researchers have found no link between global warming and tornadoes. The past couple decades have actually seen the number of days with reported tornadoes decline and, in general, the annual rate is quite sporadic. Climate change deniers have used this fact to indicate that global warming is not linked to severe weather events. But lead author James Elsner decided to delve into the data, taken from the US Storm Prediction Center (SPC), a little further. Instead of looking only at number of tornado days, he analyzed the number of tornadoes that occur on a given tornado day. Tornado intensity is generally measured using the Fukita damage scale, ranging from F0 to F5, where the higher numbers indicate higher winds and more damage. Elsner only included tornadoes rated F1 and up, indicating moderate to severe damage. Here’s his data regarding the number of days with 4,8,16, and 32 tornadoes over the pasty 50 years: Doesn’t look as good now! Although there is still quite a bit of variability across the years, there is a trend in which days with small numbers of tornadoes are decreasing and days with more total tornadoes – especially 16-32 (bottom left) and 32-and-up (bottom right) – are increasing dramatically since 2000. This actually lines up with what climate scientists have been saying all along – global warming will lead to extremes in weather, not just more heat. If you want some numbers:
1) Pre-1980, the number of days per year with 16+ tornadoes was 3.5. After 2000, it has doubled to 7.
2) Pre-1990, almost no days existed with more than 32 tornadoes. Since 2001, there has always been at least one day.
Now, there are a couple reasons why this could be happening. One hypothesis is that, in recent years, there has been greater total area with favorable tornado conditions, thus leading to more tornadoes spread across a wide range of land. To test this hypothesis, Elsner et al calculated the cluster size of the tornadoes on days with more than four. Two different clusters are formed, for example, if the distance between the clusters is greater than the average distance between tornadoes within a given cluster. That can be a bit confusing, so here’s the basic idea visually:This is an example from May 2008, and you can clearly see two tornado clusters separated by wide distances (tornadoes are black dots, cluster outlined in black). So Elsner then calculated the average cluster size across the years – if cluster size increased, this would support this widening range hypothesis described above. But he didn’t – there was no statistically significant increase in clusters! So this must mean that these tornadoes are occurring in very high densities within the same geographical area (as shown above, for example).
Why would this be happening? Answering this question definitively is really the next step, but Elsner thinks that these dense clusters indicate local thermodynamics are crucial for understanding the mechanism. Understanding these trends in tornado behavior may be a great avenue for understanding the impact of climate change on a more general level in terms of atmospheric dynamics as well. Until then, look out for those tornado days!
JB Elsner, SC Elsner, TH Jagger (2014). The increasing efficiency of tornado days in the United States Climate Dynamics