8 May 2011
Violent Twisters in A Warmer World
Posted by Dan Satterfield

Multiple radar images of the tornadic supercell that produced the Tuscaloosa tornado on April 27th. This one storm took over 75 lives.
In eighty years, most of us will be long gone (except for a lucky few who will be soiling adult diapers in an assisted living facility). However, our great-grandchildren will be in their prime and looking forward to a new century where Microsoft Windows no longer produces the blue screen of death. After the Super Swarm of tornadoes on April 27th, it’s worth asking if our descendants will be dealing with more tornadoes in the warmer world they will almost certainly inhabit.
Make no mistake, there’s little doubt that it will be a much different world, climate-wise. What we call really hot summers will be the norm, and it will be either much wetter or much drier than it is now, depending on the location. Sea level will be about a meter higher (perhaps 1.5 meters) and the beaches we know and love will be underwater. This is all known with a rather high degree of confidence, despite what you may see on certain cable networks.
About 97.5% of scientists publishing in the field (of climate physics) agree on climate change. If nine doctors tell me I need an operation and one says no, I’m going under the knife (and you would too). You have likely heard all of this before, and many seem to have unconsciously decided that their political world view makes it much more comfortable to believe the 2%.
That’s fine, but just remember the laws of physics do not care about political views. If you add CO2 into a planet’s atmosphere, the planet warms, and this has been understood for over a century. The warming estimated by Svante Arrhenius over 100 years ago is remarkably similar to the average of the sophisticated numerical models that are run for the IPCC reports today.
WAS THE APRIL 27 EVENT CLIMATE RELATED?
With the super swarm of tornadoes last week here in Alabama, it was only a matter of time before someone asked if it were climate-change related. The best answer to this (in my opinion) is from climate scientist Kevin Trenberth at NCAR who said (paraphrasing) that climate change is now involved in any extreme weather event.
This makes sense because we are already about 0.8C warmer and this results in the atmosphere holding more moisture (around 5-7%). There is no doubt in this regard (it is basic physics): warmer air holds more water vapor. There is already very good evidence for increased heavy rain events worldwide, but the question of tornadoes in a warmer world is much more difficult.
WHAT THE BEST SCIENCE SAYS
The weekend before the outbreak in Alabama, I was at the Storm Prediction Center in Norman for a seminar on climate change. The seminar was organized by Bud Ward at the Yale Center for Climate and The Media, and was designed to answer TV weathercasters’ questions about climate change science. This is the 4th seminar I’ve been asked to participate in and I hope to do more.
Harold Brooks, of the National Severe Storms Lab (NSSL-NOAA), gave a fascinating discussion of climate induced changes in regard to severe weather in a warmer world. The answers actually surprised me, because my assumption has been that we would probably see fewer severe super-cell storms by 2100. I had good reasons for believing this might be the case and it’s a story I want to share with my readers.
SUPERCELLS
Severe weather events like tornadoes and large hail are nearly always the result of long-lasting thunderstorm complexes that we meteorologists call supercells. Supercells need two basic ingredients:
1.Wind shear
2. Instability.
A super outbreak, like the one last week, had very high amounts of both. Tornadoes can (and have) developed with high instability, and weak wind shear (and vice-versa), but the big EF 4 and EF 5 tornadoes almost always have copious amounts of both. As the climate warms, it is almost certain that the stronger winds aloft, and the jet stream itself will shift poleward. This will lead to a decrease in the amount of wind shear in the favored tornado alleys. That alone is a factor arguing for fewer severe storms in the future.

Imagine lifting a parcel of air from the ground to 10km in height. The warmer the parcel is compared to the surrounding air, the greater the CAPE. In this image, the cape is the green area. For calculus students the cape is the integral of the positive area of the sounding. You can learn more about his here.
Instability is based on heat and moisture in the atmosphere and is most often measured by an index called the CAPE. CAPE (Convective Available Potential Energy) is a measure of how much energy a parcel of air rising from the surface will have. A CAPE of 3000 or 4000 joules/KG of air is a very unstable air-mass and was typical on the day of the tornado swarm and in some cases it was much higher. As the world gets warmer and more moisture evaporates into the air, the CAPE will likely climb, and this would support more super-cell storms.
SO WHICH INGREDIENT IS MOST IMPORTANT?
Harold Brooks of the NSSL in Norman presented some model data that indicates that the CAPE is most important and that we may see more super-cell storms in a warmer world. He is a co-author of a fascinating paper published in the Proceedings of The National Academy of Sciences, and here is a quote from that paper:
“However,when jointly evaluated, the increase in CAPE more than compensates for the decrease in shear such that the environment would still be considered favorable for severe convection. The result is a net increase in NDSEV, the number of days on which meteorological conditions would support the formation of severe thunderstorms.”
I suspect a lot more modeling will be done in the future on this question and it may turn out that some areas of the world see fewer tornadoes and others see more of them. It’s really a symptom of the whole issue of climate change: the planet’s atmosphere is changing at a rate that is unprecedented in human history and we are heading into uncharted territory.
HAS THE WARMING UP TO NOW INCREASED THE NUMBER OF TORNADOES?
Brooks addresses this and makes a convincing case that you cannot answer that question by looking at the tornado statistics, and here’s why:
1.Yes, there are more tornadoes reported, but there are more people to see them and that is likely why.
2.There are more radars to detect them so that afterwards someone can go to an unpopulated area and confirm damage. Many past tornadoes were likely missed because no one detected them.
Brooks believes it’s just not possible to reliably use the tornado database to say that we’ve seen an actual climatalogical increase in tornadoes (He’s an expert on this database). That does not mean there has not been an increase, and there definitely has been an increase in severe floods and droughts. That has been well shown by researchers at the National Climate Data Center and it really should not be a surprise. It’s unrealistic to assume that the warming we have already experienced would have no effect on our weather patterns.
Did all this play some role in the Super Swarm on April 27? Almost certainly. The water temps in the Gulf of Mexico were well above normal on April 27, 2011, and that warm gulf air was the fuel for the outbreak.
Can you explain the alphabet soup in the second-to-last figure?
Yes, LCL is Lifted Condensation Level and is the altitude at which a parcel of air at the surface reaches condensation or 100% relative humidity. LFC is Level of Free convection. A parcel passing this point will rise to the top of the troposphere or nearly so. In other words the parcel of air will always be warmer than the air that surrounds it and thus is buoyant. The area in green is the area between the parcel temp. and the surrounding air temp at every level. The more green area you see, the greater the CAPE.
Hi Dan,
Thanks for posting this. In a follow-up paper to the PNAS one that you mentioned, the first two authors (Jeff Trapp at Purdue and Noah Diffenbaugh who is now back at Stanford) published a great follow-on paper in 2009 showing that the rate of increase in “NDSEV” has quite a bit of regional variability, too. The greatest increases over the next several decades exist in the Southeast, Northeast, and Midwest, with smaller increases in the Great Plains.
Even the classic Weisman and Klemp (1982 MWR) paper showed some evidence of this (Figure 11 was the low-level vorticity one), and Bill McCaul (NASA/USRA/UAH) and I have seen it in our research as well. It’s very possible for low-level vorticity to increase in the presence of decreasing shear and increasing CAPE: in the Weisman and Klemp figure, just look down and to the left.
It will also be interesting to see how storm morphology changes in those parts of the country that don’t currently see “significant” amounts of CAPE very often (higher latitudes, north of, say, 40 N). The 2009 Trapp et al. paper looked only at days with big CAPE (> 2000), but November in Alabama tells us that it’s very easy to get strong storms with much less CAPE.
Reference: Trapp, Diffenbaugh, and Gluhovsky, 2009 (Geophysical Research Letters, doi:10.1029/2008GL036203).
Thanks Cody! – I have the GRL paper on my computer and am looking for the time to read it!
NOAA says no:
“Neither the time series of thermodynamic nor dynamic variables suggests the presence of a discernable trend during April; any small trend that may exist would be statistically insignificant relative to the intensity of yearly fluctuations. A change in the mean climate properties that are believed to be particularly relevant to severe storms has thus not been detected for April, at least during the last 30 years. Barring a detection of change, a claim of attribution (to human impacts) is thus problematic”
http://www.esrl.noaa.gov/psd/csi/events/2011/tornadoes/climatechange.html
Actually NOAA says exactly what I wrote in the post.
I added in more info about how the tornado database cannot be used to ascertain whether or not the tornado threat is increasing. Also see the papers linked in the other comment.
Here is a quote from the ESRL link you sent.
“A recent analysis of climate change projections suggests that the number of days during which meteorological conditions are conducive for severe storms may increase during latter decades of the 21st Century as a consequence primarily of increased instability, though projected decreases in vertical wind shear may oppose thermodynamic destabilization”
Dan
“As the climate warms, it is almost certain that the stronger winds aloft, and the jet stream itself will shift poleward. This will lead to a decrease in the amount of wind shear in the favored tornado alleys. That alone is a factor arguing for fewer severe storms in the future.” Does this mean that we can expect more tornadoes further north in future?
[…] of atmospheric instability, and wind shear, which imparts rotation. (For more explanation, see here.) So one way of examining what will happen to tornadoes in a warming world is to examine how these […]