Bill McKibben's Recent Op-Ed On Climate and Severe Weather Remixed Into Video

Check out this excellent video version of Bill McKibben’s recent Washington Post op-ed “A link between climate change and Joplin Tornadoes? Never!”

Narrated and illustrated by Stephen Thomson of, the video is a great visual representation of McKibben’s widely-circulated op-ed.

Watch here, and share this widely:


Don’t forget the massive drought in China.

Good video, although for any real skeptic (not talking about deniers here), this would not be compelling at all, and its important not to give deniers the idea that we consider this to be good evidence for climate change. That is, unless there were actual data available on the number and severity of weather events over say the last 100 years. Anybody know if any organization has attempted to compile this data?

There is nothing “excellent” about Bill McKibben’s op-ed. It is utter twaddle.

McKibben evidently has absolutely no understanding of climatology or meteorology, if he did he would realise that it was very difficult, if not impossible to attribute any individual weather event, severe or otherwise to global warming (I nearly wrote climate change - but climate is the result of weather, not the cause of it. Climate change does not cause changing weather, long term changes to the weather cause the climate to change).

To draw the conclusion that global warming is changing weather patterns around the world and so increasing (or decreasing) the likelihood of different weather types, severe or otherwise, effecting a region, you need to demonstrate that the frequency of occurence of a weather type of specified severity has changed. You also need to demonstrate a plausible physical cause for this change that can be demonstrated to be an effect of global warming on weather patterns.

This kind of statistical analysis has been done for some weather types (droughts, rainfall, hurricanes etc) for some regions of the globe and there is indeed some evidence that the frequency of occurrence of severe droughts, category X hurricanes, high rainfall events, is changing. But there is probably not enough evidence to draw confident conlcusions in most cases, so far. However, to draw knee-jerk conclusions as McKibben has done is junk science.

Lets consider the recent major outbreaks of tornadic storms over the Midwest and southern USA. Tornadoes occur there each year, some years there is more than others and which parts of this region most experiences them and when, changes too. This annual variation reflects temporal and geographic differences in the magnitudes of the thermo-dynamic and kinematic ingredients necessary to produce tornadic storms (supercells). There is as yet no strong evidence that global warming has resulted in long term persistent changes to weather patterns and thus the spatial-temporal variation in the magnitudes of the ingredients, although there are good reasons to think that continued global warming might do this and so change the tornado and convective storm climatology of the USA. However, the meteorology of tornado formation is complex. Tornadic storms can form under a wide range of environmental conditions and thus different absolute and relative values of the different requisite ingredients. This in itself, suggests global warming will not greatly effect tornado climatology to the degree that regions of the US that experience tornadoes frequently each year, will become tornado free. The exception to this is deep-layer and boundary layer windshear. High windshear through the troposphere and within the boundary layer (~lowest 2Km) is an essential ingredient for tornadic storm formation, providing other ingredients are present. If global warming changes the seasonal climatology of the strongest winds through the troposphere over the USA, then changes in tornado climatology there are likely. Numerical climate models suggest this may indeed occur in a future warmer world.


You seem to contradict yourself. You say McKibben’s piece is ‘utter twaddle’, but then go on to say that there is scientific evidence that extreme weather events (you did not mention heat waves) are on the increase around the globe. You did not mention that the increase in these phenomena is actually quite well documented in the scientific literature, and that there is also compelling evidence that they may be in part being exasperated by the warming and attendant increase in moisture in the atmosphere. Insurance data corroborate that– why would weather-related natural disasters be increasing so much more rapidly than other natural disasters…? So there is evidence claim that extreme weather events are on the increase, well most of them.

As for the tornadoes, there I do agree with you, at least in part. Had I been consulted on that I would have urged them to not refer to tornadoes, but rather to severe thunderstorms.

Yes, vertical wind shear, especially low-level wind shear is a necessary ingredient for supercells which may or may not spawn tornadoes. Now while model projections suggest that vertical wind shear in the mid-latitudes will decrease as the planet continues to warm, those same model projections show an increase in low-level moisture and a concomitant increase in convective available energy (CAPE). Several recent studies point out that the reduction in wind shear will be offset by the increase in CAPE, leading to an increase in severe storms (see below). This is from a post I made recently at SkepticalScience:

“In recent years a few papers have been published on how severe storm environments might change as the USA warms and as low-level moisture increases. Note that they do not speak to trends in tornado occurrence per se, and keep in mind that only a small percentage of supercells actually produce tornadoes, but a consistent pattern of increasing severe storm potential is evident.

Van Klooster and Roebber (2009, J. Climate):
“In this work, the authors present a “perfect prog” approach to estimating the potential for surface-based convective initiation and severity based upon the large-scale variables well resolved by climate model simulations. This approach allows for the development of a stable estimation scheme that can be applied to any climate model simulation, presently and into the future. The scheme is applied for the contiguous United States using the output from the Parallel Climate Model, with the Intergovernmental Panel on Climate Change third assessment A2 (business as usual) as input. For this run, relative to interannual variability, the potential frequency of deep moist convection does not change, but the potential for severe convection is found to increase east of the Rocky Mountains and most notably in the “tornado alley” region of the U.S. Midwest. This increase in severe potential is mostly tied to increases in thermodynamic instability as a result of ongoing warm season surface warming and moistening.”

Trapp et al. (2009, GRL)
“Our study shows that the frequency of severe thunderstorm forcing increases in time in response to the A1B scenario of GHG emissions. This is also true for severe-thunderstorm forcing that is constrained by the occurrence of convective precipitation. The rate of increase varies with geographical region and inherently depends on (i) low-level water vapor availability and transport, and (ii) the frequency of midlatitude synoptic-scale cyclones during the warm season. The current report provides further evidence of the effect of anthropogenic GHG emissions on long-term trends in thunderstorm forcing [Trapp et al., 2007a; Del Genio et al., 2007].”

Trapp et al. (2007, PNAS)
“We use global climate models and a high-resolution regional climate model to examine the larger-scale (or “environmental”) meteorological conditions that foster severe thunderstorm formation. Across this model suite, we find a net increase during the late 21st century in the number of days in which these severe thunderstorm environmental conditions (NDSEV) occur. Attributed primarily to increases in atmospheric water vapor within the planetary boundary layer, the largest increases in NDSEV are shown during the summer season, in proximity to the Gulf of Mexico and Atlantic coastal regions.”

Del Genio et al. (2007, GRL)
“For the western United States, drying in the warmer climate reduces the frequency of lightning-producing storms that initiate forest fires, but the strongest storms occur 26% more often. 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.””

Remember too that storms can still ingest sufficient low-level vorticity (in the presence of sufficient CAPE and bulk shear) to form mesocyclones if they interact with mesoscale boundaries (such as outflow boundaries, drylines) and those are almost all present.

What will be interesting to see is what happens with the great plains low-level jet. It is a major reason for the extremely strong low-level wind shear that is oftentimes present over tornado alley, and its strength is, IIRC, not necessarily related to meridional temperature gradient but is classified as a barrier-type jet.

Harold is that you?

“Insurance data corroborate that– why would weather-related natural disasters be increasing so much more rapidly than other natural disasters…?”

You say weather related disasters incidents are on the increase globally. This is moot. Are they? or is it that there better communication and more observations being made, greater awareness and records kept? I would not advise use of insurance claims as a measure of the increase in severe or extreme weather events. There are too many other factors influencing these for them to be a very reliable proxy. Such factors include, population growth, increase in numbers insured especially in the developing world, locally and globally to name but two.

Re Heat waves. I did not mention these, but did refer to weather types generally. But i agree. There was a recent study that suggested there is some extreme summer heatwaves in Europe are now becoming more frequent.

The point of my message was not to dispute that extreme weather events or severe weather is on the increase or that weather patterns are changing as a result of global warming. I expect they are. It would be surprising if they were not. But just because there has been a lot of weather related disasters recently does not mean there is a clear trend, or indeed the weather disaster rate is unusual. However, i also pointed out and i think this is important, some types of extreme weather event may become less frequent than now in a warmer world. The is a quid pro quo in this respect.

My point was that for McKibben to make the very easy and simplistic claim that these are the result of global warming (or climate change [sic]) is flawed science and statistical illiteracy. This is because you cannot make such claims about individual weather events or indeed temporal clusters of weather events (look up Poisson clustering) as this year (apparently). Extreme weather events by definition are infrequent, but they do happen and would still happen infrequently without global warming occuring. To make such claims requires careful statistical comparative analysis of weather records for at least two equally long periods in the recent and more distant past. At the moment although I agree there is some statistical evidence that some regions are experiencing a greater frequency of extreme weather events and/or more severe weather events, and perhaps of greater severity more frequently, the increases being seen are subtle and not so far greatly different from the historical climatological ranges of severities and frequencies of occurence.

To give you an example why one has to be careful about Poisson clustering, there has been an abnormal number of magnitude 8 or 9 earthquakes in recent years. Is this indicative of a change in plate motion patterns and rates? Should we be concerned. Or is it simply the random statistical outcome of a set of causally unrelated events, i.e., a Poisson cluster.

BTW the term “severe thunderstorm” is not a good one to use. Thunderstorms that produce severe weather are identical in structure and physics to thunderstorms that produce weather that does not exceed the formal severe weather definitions. They are just bigger, more vigorous and energetic, to put it simply. There is no such thing as a severe thunderstorm. There are only convective storms, some of which produce weather types that are more severe than the weather types produced by others. If the storm does not produce severe weather it is not a severe storm and storms are not distinguishable.


No, but I do know Harold. But flattery won’t get you everywhere…

I am busy right now, but re insurance claims…think about it, why would people be disproportionately affected by extreme weather events than other natural disasters? Populations are also rapidly increasing the vicinity of tectonic plates and volcanoes…and those factors you suggested as possible explanations apply to other natural disasters as well, why the discrepancy? Later you say “there has been an abnormal number of magnitude 8 or 9 earthquakes in recent years”, indeed, and yet weather-related costs are still outpacing other natural disasters. Not surprisingly Munich Re would disagree with you.

“However, i also pointed out and i think this is important, some types of extreme weather event may become less frequent than now in a warmer world. The is a quid pro quo in this respect.”

An unsubstantiated and highly confident claim (you say “this is”), and I beg to differ regarding quid pro quo– some numbers and peer-reviewed citations please. And Munich Re and other re-insurers would not be so concerned if that were the case.

“There is no such thing as a severe thunderstorm.”

Now that gave me a laugh. A supercell is very different to an airmass thunderstorm in terms of its structure (not to mention the environment in which it develops is very different)– it has a rotating updraft, amongst other things. Even severe storms without rotation often have WERs or BWERs, not so in an airmass thunderstorm. I’d love to be at the table when you try and convince Harold and Chuck and others in the know that “Thunderstorms that produce severe weather are identical in structure and physics to thunderstorms that produce weather”. Finally, one doesn’t necessarily need to rely on surface reports to warn a storm– NWS uses radar derived parameters all the time to warn severe storms.

Let us cut to the chase– do you agree that we need to dramatically reduce our GHG emissions?

Munich Re keeps track of disasters: