People are asking me: Is the recent spate of tornadoes caused by global warming? The usual answer to that question is that you can’t answer the question because a tornado is not caused by climate … it is cause by weather … and global warming (which is real, and which is cause by humans) is climate change.
However, that is not really the best answer to the question. Ultimately, I want to propose an analogy for how to think about this question, but first, a stab at a good answer, which if modified could probably be improved:
Question: Is Anthropogenic Global Warming the cause of the current spate of deadly, powerful, and numerous tornadoes?
Answer: Eventually we’ll have enough data to answer that question, but the answer will not refer to a given tornado or even a given tornado season. Rather, it will refer to the relationship between two major climatic factors which are potentially influenced by global warming.
The first is the total energy available to do things like producing thunder storms, including the tornadoes they spawn, as well as other weather events. Much of that energy comes broadly from the tropics where excess energy from the sun builds up and is then redistributed through oceanic and atmospheric currents, and more directly from sea surface temperatures in the Gulf of Mexico. The warmer the surface of the gulf, the more moister and energy going into north and northeast trending air masses that is a key component in making these storms. In the end, it is impossible to imagine a scenario in which there is not more storminess of some kind, somewhere, because storm fronts are this energy dissipation manifest as weather. Whether that tuns out to be more rain, more thunderstorms, more severe thunder storms, or more tornadoes is as yet unclear.
The second is a bit more obscure: This is a change in the way air masses are divided up on our planet. Climate systems are ultimately driven by the sun’s energy, and in particular by the fact that there is excess energy at the equator which then moves towards the poles (while at the same time dissipating off the planet at the upper edge of the atmosphere). This is a simplification: These air masses are organized as giant rotating donuts surrounding the planet. The first set of donuts are on the equator, and rotate (as in twisting, not orbiting) in opposite directions, the northern one twisting upwards at the equator and downward north of the equator, the southern one doing the opposite. The next set of donuts out, north and south, twist in the opposite direction. And so on. There are several different such donuts.
The boundaries between these giant donuts of air are highly active regions. The reason for this is way beyond the scope of this post, and requires calculus so you’ll have to take my word for it. Let’s just say that the phenomena we call “jet stream” is the result of interactions between the giant twisting donuts of air.
The donuts are far from perfect. Those closest to the equator are very consistent and always present and fairly obvious, but as we work out from the equator they become less clearly defined. Even the distinction between them is less and less clear with increasing latitude, and as the distinction becomes less clear, the “jet streams” become less likely to occur. Now, and again this is an oversimplification, imagine that each giant twisting donut as you go from the equator to the poles is colder and dryer. This means that at the margin of two donuts, where warm moist and cool dry air contact, there will be storms, especially where the atmospheric machinations generate low pressure systems. This is where storms are born. Now, imagine taking the warm and wetness of one of these donuts and pushing it up against the coldness and dryness of a rotating donut at a much higher latitude, skipping an entire donut. The contrast between the two air masses would be much greater than normal, and thus the storms would be much more frequent and stronger than normal.
That’s one way to get a stormy year: Take two donuts that are normally distinct and combine them into own: Thus, the contrast between the southern margin of that air mass and the next air mass to the south will be greater, and the storms stronger and more numerous. That is what is happening in the US now. Those living in norther states are experiencing a cold spring and early summer, and the southern edge off this air mass is contacting a very warm and moist air mass, and this contact zone is where the severe thunderstorms and tornadoes are forming.
There are some models that suggest that the breakdown in distinction between these air masses is caused by anthropogentic global warming (or, really, any global warming but it happens that our current warming is anthropogenic). At the same time, we know that the sea surface temperatures in the Gulf are high, also because of AGW. These two things together would be the cause of an increase in frequency and severity of storm fronts, the thunderstorms that form along the front, and the tornadoes.
There could be other explanations for an increase in tornadoes. For example, the frequency of thunder storms may be higher because of AGW, but there is some other reason more storms are producing twisters. But again, as per the answer given above, this is the sort of thing that will be figured out in due time with more data and analysis.
Now, the analogy. We can’t say one tornado or another tornado is caused by your selfish and inconsiderate CO2 belching behavior. Which is important because you would prefer to not be blamed for the massive destruction and whispered death we are experiencing across this country right now. But let’s assume that the two factors discussed above are true: The breakdown of the donuts and the increase in sea surface temperatures in the gulf. How do we link, and is it even possible to link, global warming to a tornado via these mechanisms?
The analogy is robbery. Robbery is where people take other people’s stuff. Maybe they rob a store or a bank, your house, or they mug you on the street. It is known that this kind of economic crime increases in frequency when the economy is bad and unemployment is high. So, does this mean that if you get mugged that the cause of the mugging is the economy? No. A certain amount of robbery happens anyway, so you can’t say that a given robbery is caused by the economy (indirectly… obviously the economy is not robbing you, some dude is). Not everyone becomes a robber when the economy goes in the toilet, nor does robbery become randomly assigned among the populous as a novel behavior. There is number of reasons why a particular robbery happens, including the role of the victim, who must, after all, be in the wrong place at the wrong time, or have some stuff to get robbed, etc.
A tornado is less likely to happen if the terrain is not flat or the season is not right. Tornadoes will destroy more property and klll more people if more people are settled on flat landscapes down-wind, as it were, from the Gulf. And so on.
So you can’t say a given robbery or tornado was caused by higher unemployment or AGW. But if you look at a graph and see a rise in frequency of either kind of event, you could attribute that bulge in the data to these causal factors.
In the end, in my opinion, the answer will be yes. Tornadoes or other storm phenomena including hurricanes will have increased in either frequency and energy level or both, and possibly the extent to which they reach northern latitudes, because of anthropogenic global warming. The current debate (to the extend that it is a debate rather than senseless mud slinging) involves scientists on one side saying “Yeah, probably, it makes sense, and the data are starting to show this” and denialists on the other side saying “you can’t say that” because of sophistic pseudo-philosophical reasons.
Tornado and hurricane frequency (and other features) vary from year to year across quasi-cycles that are less than a decade long, so it is not unexpected to have a few bad years followed by a few years that aren’t so bad followed by a few years that are bad again. This is especially true of the most severe events. For example, the strongest hurricanes and the strongest tornadoes are so rare that counting them up annually and looking for trends over the years will probably not be very useful. We may be having a bad period with more tornadoes, and more powerful (and deadly) ones, right now. The question at hand is this: Are the bad years getting worse, and are the not-so-bad years getting bad, or are the bad periods getting more common? To address these questions empirically it will be necessary to have at least a few decades worth of good data under conditions of global warming. We probably have those data now, or almost so. This question will be settled by reasonable people reasonably soon.
Note: This blog does have a policy regarding commenting by AGW denialists. If you are an AGW denialist, read the About page. Then, if you’re good, you can comment.