Attribution of Extreme Events

It’s, literally, a great shame that this sort of stuff appears on left-leaning sites like “The Nation” almost to the exclusion of most other media. Steve Vavrus is being pretty darn careful, maybe even erring on the side of excessively restrained here, despite the headline on the video. But the mainstream press keeps evading and muddying a rather simple question.


  1. As does Vavrus, come to it. It is true that no particular event can be attributed to AGW, but it's also true that no particular event can be said any longer to have solely natural causation. Why is that so hard to say?

    IMHO it should also be noted that given present science we still wouldn't be able to make an AGW attribution even if extreme events become an everyday experience. Of course, if that happens the distinction will have ceased to matter to anyone but scientists.

  2. I think a big part of the problem is that people generally are asking the wrong questions and expecting a simple yes/no answer.

    When someone asks if this or that event caused by climate change we have a good opportunity to explain some of these complexities.

    If they are determined to get a yes/no answer then I think simply not answering might be the best option.

  3. Grypo, while it would be nice if somebody actually collected things like this, it is difficult to be fair. I had not heard of the floods in BC - but there are many thousands of places on earth. Someplace or other is always going to be having a hundred year flood. One could compile a list rather like this even in the absence of climate change.

    The question has two parts: HOW odd the events are, and HOW many of them there are. Just listing odd events is not proof of anything. It's a big world.

    Regarding the derecho, I have found this which indicates that derechos, while common enough in the midwest, are extremely rare on the east coast. This fits in nicely with what Dr. Vavrus is saying in the video.

  4. I realize what you are saying, that's why it's like baseball 🙂

    Each event needs to be analyzed individually. For instance, a link I did not provide is about the flooding in Vietnam. The rains may or may not be unprecedented, and the flooding may be due to infrastructure deficiencies. Or not.

    I think, the question, although less interesting than direct attribution, is, what are the odds of these extraordinary events happening all at once on a global scale?

  5. Ah, there's another complication. Large scale anomalies are correlated because the system is closed.

    To take a simple example, if the jet stream wanders too far south in one place it has to hustle back north in another place just to close the loop. In particular amazing hot/dry in America and amazing cold/damp in England actually do go together.

  6. That would show in the historical record then could help or hurt the thesis that this set of events is extraordinary. It makes the questions of how much of the observed jet stream anomalies and changes in hydro-cycles are tied to global energy increase

  7. Hmm, not sure I like "global energy increase" any more than I like "global warming".

    "Global radiative imbalance" anybody?

    Anyway, there's no serious doubt left that what we are seeing already is extraordinary in aggregate.

    I would propose that the practical way to get to that is not by endless statistical hair-splitting, which in a case like this can be made to turn out any way you want. It's by asking the best operational meteorologists. Ten years ago they were saying "ho-hum". They are not saying that anymore.

    The real issue is whether the costs are about to make a real impact on a global scale. The trouble with that is you have to rely on economists, and economics is defined in terms that are peculiarly oblivious to resources. So they don't know.

    But anyone who doesn't see that climate change is already causing many tragedies on individual and local scales and much damage to local ecosystems is just not paying attention.

    That all said, it would be nice to have a systematic global compendium of disasters, natural, artificial, and this new kind, with some systematic way of weighing them. It might help us judge whether we are approaching the point where it becomes overwhelming in aggregate.

    Honestly, I have no idea whether we are or aren't approaching that point, but I don't find that reassuring.

  8. KT gets a good chunk of time on the Lehrer show. Note that he manages to avoid the "no single event" meme. The interview was perhaps not his best overall, but he made up for it at the end after the interviewer tossed out a ridiculous bromide about climate change-induced extreme weather being something to "keep you scientists busy."

  9. Note that Jason Box is predicting that Greenland surface melt will reach the summit soon, very possibly this year. It's a striking milestone, albeit not an example of extreme weather as such.

  10. There's lots o talk of dot connecting lately and it got me to thinking of impacts like those discussed in the 1000 Bastrop thread. For instance, the Derecho, caused by stationary front (Arctic amplification connections), excess moisture from Gulf moving north (ocean heat), and excess heat (temperature increase) all came together to cause a fairly unique event in size and impact. But also, impacts following, such as downed power grid for over a week right in the heat wave zone. A problem I am having is understanding our incessant need to ask scientists if we can attribute this to AGW. Who can really answer answer that question with science precision? Who is writing a paper in the climate field to quantify statistically all these impacts and their causes? Is this not impossible and we may have to get people to understand the science field is not going to be able to give answers and we just have to figure this out ourselves, preferably yesterday?

    Another example for the near future, John Fleck has been following the fire season transition to monsoon season in New Mexico. So you have wildfire (excess heat, energetic storms and low humidity) burning out mountainsides leaving long shoots for rain (excess precipitation in warmer atmosphere) to cascade down causing damaging flood impacts (increase in hydrolic cycle). How do we quantify that in statistics and begin attributing it?

    How do we get across the dangers of multiple, exacerbating impacts whose conditions are significantly effected by the changing climate?

    PS If we get scientists to stop saying that we can't attribute any single event to AGW with absolute confidence, that would be nice. It's already well understood and reinforces meaningless questions.

  11. Here's an interesting article about increasing water supply problems in Uganda. This is perhaps not an extreme in terms of weather, noting that it's not the area affected worst by the East Africa drought, but the ultimate outcome may be some shade of catastrophic.

  12. grypo wrote, "For instance, the Derecho, caused by stationary front (Arctic amplification connections), excess moisture from Gulf moving north (ocean heat), and excess heat (temperature increase) all came together to cause a fairly unique event in size and impact."

    Regarding your temperature increase (which shifts the probability distribution to the right), I would also include increased temperature variability. From Hansen's "climate dice" paper:

    "In addition, the probability distribution broadens, the warming shift being greater at the high temperature tail of the distribution than at the low temperature tail."

    Oddly, though, all these connections appear to have been missed by someone who has published extensively on extreme weather. In a recent post by Jason Samenow at Capitol Weather Gang, he quotes Harold E. Brooks stating in response to the question of whether climate change may have contributed to the derecho, "Not to a particularly significant extent. The hot surface temperatures and high lapse rates aloft directly contributed. I'm not sure how much of either of those goes to long-term warming trends. An important aspect was the set up of the vertical wind profile in relationship to storm motion."

    The right answer is that while global warming may not impact all of these factors, it makes some factors more likely and thus would appear to make derechos more likely and make it more likely that those that do form will be more severe.

  13. It appears we won't have to wait to see if Fleck's warnings are realized, as it is already happening in Colorado

  14. But the model results she refers to are absent the blocking event increase, and her concluding remark that it won't ever be possible to attribute a single extreme event or even season to human influence misses the point in the usual way.

  15. It's aslo complete nonsense that it will "never" be possible to attribute any event to human influence. Urban heat islands are uncontroversially anthropogenic and happen every day, for instance.

  16. Thanks for this. This is exactly what I'm talking about. There is nothing "wrong" with Dr. Brook's answer. But let's break it down. When asked about "connecting global warming to the derecho" Brook's answer is he "doesn't think it’s 'significant'". Why? "we don’t know how much greenhouse gases contributed to the heat wave and to the amount of instability in the atmosphere."

    But is that an important reason(?), and when I say 'important' I mean meaningful in regards to future safeguards. So by deduction, if GHG contribute to the heat wave and atmospheric instability, then shouldn't the answer be something like, "Yes, the warmer and unstable atmospheric conditions could contribute to larger and more dangerous derechoes"?

    Logic check
    1. Warmer and unstable atmospheric conditions can lead to larger and more dangerous derechoes
    2. We are experiencing warmer and unstable conditions and expect more in the future due to GHG's
    Conclusion: Derechoes can pose a larger threat now and in the future as GHG's accumulate

    I believe the problem is more with the reporter's question, than the scientist's answer in this case.

    I'm basically retyping what Timothy said, but I think it bears repeating. 😉

  17. But Harold did say this:

    "The hot surface temperatures and high lapse rates aloft directly contributed. I’m not sure how much of either of those goes to long-term warming trends."

    Which I would submit is missing the point in the usual way.

  18. "... missing the point in the usual way." I more or less have to guess what you mean by this.

    I presume what you mean is that for any given weather event there are specific causes that you can trace that event to that are equally specific and weather-related, e.g., the so-called blocking events that result in extended droughts that, after evaporating what moisture is in the soil raise temperatures further due to the lack of moist air convection. And as viewed from this context, there is nothing left to explain by reference to climate change. Essentially, this is the same sort of fallacious reasoning that I see coming from experts at NOAA at times, e.g., with respect to the Russian heat wave. Expanding on this, they might argue that the more severe the heat wave the less global warming could have contributed to it since the average rise in temperature due to global warming of roughly 1F is dwarfed by the magnitude of the heat wave.

    However, as Trenberth would point out, global warming forms the backdrop for all weather events at this point. It provides the atmosphere with more heat, moisture and energy. It alters atmospheric circulation and weather patterns. It has altered the background against which natural variability is expressed.

    It makes more extreme events far more likely, first, by shifting the probability distribution of temperatures to higher temperatures, and second by widening that distribution, so that what was in the rapidly narrowing wing moves closer the center of the bell. And with polar amplification, it decreases the temperature gradient between the Arctic and mid-latitudes, increasing the amplitude of Rossby waves, the incursion of warm moist air to the north and cool dry air to the south, and it slows the jet-stream, making blocking patterns more likely, which then amplify droughts and flooding due to positive feedbacks, such as when moisture is baked out of soil, reducing moist air convection so that the surface warms even more.

    Of course, I haven't been terribly concise, but is this what you meant, and do you have a better way of stating it?

  19. Quick point...

    On the attribution of extreme events. In case anyone has missed it, NOAA has put out a State of the Climate here:

    And there is a separate paper on the same page under the heading of Real Time Attribution paper:

    Explaining Extreme Events of 2011 from a Climate Perspective

    Seems decidedly on topic.

  20. Pretty much, Timothy, and nicely stated. The only quibble is that I think Rossby waves are also slowing down (in association with the amplitude change), and I'm not sure about the jet stream itself. Come to think of it I need to refresh my memory as to whether the Rossby waves are also decreasing in length (wavelength, i.e.). Will check and report back.

  21. Steve Bloom wrote:

    I'm not sure about the jet stream itself.

    Please see:

    As high latitudes warm more than mid-latitudes, however, this north-south temperature difference weakens, which has two impacts on the jet stream.

    The first effect is to slow the west-to-east speed of the jet stream, a phenomenon that already appears to be occurring....

    Some regions exhibit even larger drops in wind speed, such as over North America and the North Atlantic, where winds have slowed by about 14 percent since 1980. Theory tells us that a decrease in the west-east flow tends to slow the eastward progression of waves in the jet stream.

    Linking Weird Weather to Rapid Warming of the Arctic
    by Jennifer Francis, 05 MAR 2012

    Jennifer A. Francis coauthored "Evidence linking Arctic amplification to extreme weather
    in mid-latitudes" (2012) with Stephen J. Vavrus

    If you don't have a copy of the paper let me know.

    timothychase at g mail dot com

  22. Interesting. Being ill informed, I read through the Wikipedia (!) entry on Rossby waves, and combining that with other input, it's quite clear they are slowing.

    I think it must have been Francis at Rutgers that got me going, this one, very thorough and clear for those used to scientific rigor, and totally dismissible by anti-reader readers.

  23. really? I thought the usual confusion was about "global warming" qua trend in GMST vs "global warming" qua informal name for anthropogenic climate change. Denial makes use of the more convenient meaning at any moment, which is why I try to stick to the narrower meaning myself.

    Of course "long term warming trends" do not cause derechos. So in that sense it is ridiculous to say any event at all is "caused by" "global warming". And anyone making such an argument is willfully taking advantage of a poor trahectory of the vernacular, not seriously arguing science or consequences.

    As for Steve's geophysical fluid dynamics etc., I am in the large either out of date or unconvinced, I am not sure. But I do have the expertise to promise you that a derecho is nothing like a Rossby wave at all nor is it directly connected to the jet stream in terms of an energy source.

    However Timothy's "Essentially, this is the same sort of fallacious reasoning that I see coming from experts at NOAA at times, e.g., with respect to the Russian heat wave. Expanding on this, they might argue that the more severe the heat wave the less global warming could have contributed to it since the average rise in temperature due to global warming of roughly 1F is dwarfed by the magnitude of the heat wave." is right on the money.

  24. Not sold. For one thing the "high latitudes warm more than mid-latitudes" is mostly concentrated in the lowest layers of the atmosphere, which would not actually slow the jet stream.

    As I've said before, I like Steve Vavrus a lot but I think it's highly premature to call this settled science. Francis' blog article just amounts to a self-citation for the 14 per cent. So far there seems to be only one citable source for this result. It may well pan out but it's premature to be hanging your hat on it quite yet.

  25. I included the quote from the informal article at Yale not because I thought it was authoritative, I don't, but because it was one of the first articles I found, and she and Vavrus (the fellow in the video) coauthored a paper which I believe is cutting edge. The latter implies almost of necessity it can't be settled science. But this also makes it more likely to be considered interesting. I thought it a pity that Vavrus didn't get actually get into some of the specifics of their study in the video.

    It would seem they have a mechanism:

    The differential warming of the Arctic relative to midlatitudes is the key linking AA with patterns favoring persistent weather conditions in mid-latitudes. Two separate effects on upper-level characteristics are anticipated: weaker poleward thickness gradients cause slower zonal winds, and enhanced high-latitude warming causes 500 hPa heights to rise more than in mid-latitudes, which elongates the peaks of ridges northward and increases wave amplitude. Both of these effects should slow eastward wave progression.

    ... and data reanalysis that supports their theory:

    Wave features in 500 hPa fields are analyzed from 1979 through 2010.

    ... and the cite a fair amount of literature in support of their views. Furthermore 500 hPa is at the belly of the jetstream. The tropopause has been rising partly as the result of global warming.

    Please see:

    Observations indicate that the height of the tropopause—the boundary between the stratosphere and troposphere—has increased by several hundred meters since 1979. Comparable increases are evident in climate model experiments. The latter show that human-induced changes in ozone and well-mixed greenhouse gases account for80% of the simulated rise in tropopause height over 1979 –1999. Their primary contributions are through cooling of the stratosphere (caused by ozone) and warming of the troposphere (caused by well-mixed greenhouse gases).

    Santer et al. (25 July 2003) Contributions of Anthropogenic and Natural Forcing to Recent Tropopause Height Changes, Science, Vol 301, pp. 479-483

    I would expect it to rise more in the Arctic given arctic amplification. The jetstream resides in the troposphere, so it would be affected.

  26. I still need to look at the paper (public copy on her pubs page) to try to figure out the exact relationships, but I think upper-level winds are not the same thing as jet stream winds, rather she's saying the former drives the latter. The article also discusses what I take to mean *average* west to east jet motion, which would be distinct from the speed of the jet along its track. Basically I was speculating that if the track has lengthened overall (not sure that's even the case) the actual jet speed might not have decreased, or perhaps even could have increased, although even if I have the relationships right increased wavelength would be track-shortening. Someday soon I'll report back on this.

    Re the jet and the derecho, if anything a slowing of upper-level flow would seem likely to be a drag on the latter (although in the article Francis seems to be saying the change is concentrated in the autumn; maybe the paper has more specifics on seasonal changes). In any case it seems to be a matter of the upper-level flow driving both. What the jet (or the Rossby Waves, more precisely) do seem to drive is blocking events.

    David Roberts has a useful post on the aforementioned missing of the point and the proper role of journalists and climate "hawks" (how I dislike that term), but while more or less accurately tarring the scientific community for feeding the bad meme, fails to note the considerable grouping, including as p3 readers will know some very big names, who are fighting to change things.

    I was struck as I was writing the previous paragraph by the question of how many people in the U.S. are sufficiently aware of the issue to be able even to state it coherently. Tens of thousands, or even less? People talk about economic policy being an elite debate, but this would seem to be much more of one.

    Finally, back to the jet stream(s), there's a web site tracking them. It's interesting to see what the northern polar one really looks like, which is a pretty much a mess even though the basic pattern is clear enough. The animated loop at the bottom of the upper left box is worth watching. The influence of the jet on current large-scale weather patterns is quite apparent from eyeballing.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.