A Thousand Bastrops


My birthday falls around Labor Day, and last year I asked Irene for an overnight in San Antonio as a present. We stayed at the historical Menger Hotel, saw the sights, ate delicious food, drank delicious drinks, and a fine time was had by all. But on the day we drove back, it seemed like the whole of Texas was on fire. Great clouds of smoke appeared from wildifres on the horizon in every direction as we drove up Interstate 35.

By far the most serious of them turned out to be the blaze in Bastrop Texas. On the worst day, its smoke was to shroud the city of Austin, 30 miles away, in reduced visibility and, I would say, a peculiarly evocative pine scent, the following day. Only two people died, but 1700 homes were destroyed, and a vast area remains desolate today.

Wikipedia says:

In the months before the Bastrop County Complex fire, Texas was affected by a series of wildfires amid several distinct record-breaking meteorological conditions conducive to combustion. During 2011, Texas endured its most severe single-year drought since the 1950s, received the lowest single-year rainfall since 1895, and experienced the hottest June–August period of any U.S. state at any point in time on record—exceeding that of even the Dust Bowl of the 1930s.[17][18][19]

Circumstances were further exacerbated by Tropical Storm Lee, which produced strong winds over the Labor Day weekend, creating ideal conditions for wildfires to spread.[20] Between September 4 and September 6, reports indicated that 63 new fires were started.[21] On September 6, the Texas Forest Service released a statement, describing the fire’s behavior as “unprecedented” and stating that “no one on the face of this Earth has ever fought fires in these extreme conditions”

The Bastrop fire burned for four weeks.


Continuing, if you’ll indulge me, with a personal perspective, I have been away for a month, in the chilly climes of Montreal for a week and then London. Both experienced uncommonly cool and rainy weather during my sojourns there. I’ve barely seen my shadow in a month.

Meanwhile, not to say coincidentally, the Arctic has reached an all-time observed minimum June ice cover. So it’s hard not to get back to Broecker’s original “angry beast” scenario: freshening of the Arctic surface diverting the Gulf Stream southward, swinging the northern Atlantic into a cool regime while the rest of the world warms.

When I left, Texas had come off a relatively hopeful spring, not the nicest Spring Texas has seen but far from the worst. The weather had been lovely, a bit on the warm side, but with a decent amount of rainfall, and Texas crawled slowly out drought. On my travels, I didn’t pay much attention to the weather in Texas until Tuesday, the day of my return.

On Tuesday it hit 109 at Camp Mabry.

It was noticed here as the hottest day ever recorded in the calendar month of June. Perhaps the nation was distracted by the fact that part of Kansas had hit 115 on the same day.

So I looked at the drought map last night, and was more than a little taken aback. The somewhat abated situation in Texas is now matched by drought in all but perhaps four states in the southern 2/3 of the country.



The first I heard about the current events in Colorado was some buzzing in email about evacuation of the NCAR facility at Table Mesa in Boulder. Anyone who’s been inor near the climate science fiels in the US for a while is very familiar with the beautiful site, perched up above Boulder on a steep slope in the front range. (It must have been built during a brief period when the society actually wanted to encourage people to become scientists.)

Just behind the ridge a wildfire burned yesterday, and the possibility that NCAR would be torched looked all too real. Fortunately, this risk (the irony of which Joe Romm was quick to pick up upon) has abated, b ut unfortunately, a far greater risk has taken its place on the front range above Colorado Springs, Colorado’s second city and home of the US Air Force Academy.

So it appears that much of Colorado is in danger. The main point I’d like to address here is that it’s the front range, abutting most of the population, that is most of all in danger, and that this is not a coincidence..


In his masterpiece of research, the book Normal Accidents, sociologist Charles Perrow makes the point that the most consequential industrial accidents usually result from a multiplicity of adverse factors and errors. As we enter the age of increasingly unforeseen climate variations, the lesson begins to apply to natural catastrophes as well.

The greatest catastrophe of recent years in North America remains the chaos in New Orleans after hurricane Katrina. A fair amount of secondary debate goes on as to whether or not the disaster can be attributed to “global warming” or more precisely, to anthropogenic climate change. It was, after all, an extraordinary storm in an an extraordinary year for Atlantic hurricanes. As people line up in mostly predictable clusters of opinion around this question, something important was lost in the shuffle.

Certainly, an extraordinary storm in an extraordinary storm season is relevant to the disaster. Whether and how much to attrib ute this to “global warming” is a question which invokes both mathematical and philosophical complexities. It’s a tough call. What is not at all a marginal decision, though, is whether the disaster could be attributed to warnings unheeded. There had been discussion of the vulnerabilitgy of New Orleans to a Katrina-like event for decades. The risk was well-known. The political will to reinforce the levees was not to be found before the disaster. Even the political will to set up a realistic evacuation plan was absent, and in the event, the competencies to do adequate disaster relief were shockingly absent in a domestic disaster despite the competence of Americans and westerners in responding to comparable events in places like Haiti or Bangladesh.

Business-as-usual trumped foresight. The ordinary compromises and petty corruptions of the marketplace and of government overrode the widely understood necessity for preventive action. Thousands of people died and most of a great and unique city lay in ruins. Arguably climate change had a hand in it, but a systematic failure to heed rational warnings was the root cause of the disaster.


In the last few years a phenomenon of intense and persistent medium-scale heat anomalies has clearly emerged. In the past decade, notable events have occurred centered in France, eastern Australia, central Russia and Texas. WHile most of these occur in summer, this spring we had an astonishing warm anomaly centered in WIsconsin and Minnesota.

It’s fair to say that this phenomenon was not clearly predicted in climate models. Consequently, it is fair to say that it is not fully understood, and whether it will remain a persistent feature of the climate is speculative. Strictly speaking, formal attribution will be difficult. But it being a global response, a global cause is to be sought and identification of a prime suspect doesn;t require much imagination. Global warming appears to be manifesting in large part as extreme regional anomalies, rather than as the gentle trend that some people are so eager to presume. And this leads, among other things, to stress on forests.

But, every forest that is with us is a forest on a crowded world. Few are in anything resembling a natural state. In particular, natural forests are susceptible to fires, fires which people living in forested regions are eager to suppress. Consequently, over trhe years, the combustible fuel in the forests (and to a lesser extent in grasslands) accumulates to an extent that would not occur in nature. Consequently, fires become hotter, larger, and more devastating when they do occur. This risk was well understood vefore anthropogenic climate change became a palpable part of our daily lives. The connection between the one and the other was not really considered.

Consequently, the extreme damage we saw in Moscow, in Bastrop Texas just a few months ago, and potentially in Colorado this week, was a Perrow-like interaction of two separate and disconnected unheeded warnings, just as the mortality from Katrina was.

The twenty years we have thrown away on the climate change issue since Rio has amounted to a wandering cosmic heat lamp, heating now one forest, now another, to the point of ignition. How much damage that ignition does comes down to local forest management practice. The building of wealthy suburbs in forested places like Boulder and Bastrop, with a strategy not of forest clearing but of fire suppression, has provided a great supply of tinder.

These errors are fundamentally alike. They result from pressures of the marketplace with its systematic discounting of the future. Discounting the future amounts also to a systematic devaluing of low-probability events. The errors combine in a way that is far worse than the additive sum of the risks considered separately.


The pervasive nature of climate change exacerbates many other risks. Failure to account for those other risks occurs for similar reasons to the increasingly obvious failure to account for climate risks. Combined disasters combine worse than additively.

This is how it hits the fan.


Bastrop wildfire: Wikipedia

Drought Map: Drought Monitor (University of Nebraska at Lincoln)


Waldo Canyon fire: Wikipedia


  1. Today in England insurance companies are dealing with the novel circumstance of widespread hail damage to automobiles and houses, part of a day's weather that included fatal flash flooding and tornadoes. A convective style of day, a parcel of energy we're accustomed to here in the US but which is hair-raising in other places.

    Roger Pielke Jr. is a standard bearer for the notion that connecting extreme weather with climate change is folly. He's quite forthright:

    The detection of changes in climate requires looking at actual data — and the data on tornadoes, large-scale river floods (in unaltered river basins), and landfalling hurricanes shows no evidence of trends in the direction of more extreme events. This should not be surprising, because even if we assume a strong signal in extreme events from human-caused climate change, the statistics suggest that it would take many decades, and probably longer, before such signals would be detected.

    Fair enough, but let's recognize that data points include burnt homes. How many are enough and how do we include cost versus benefit with this meticulous approach to forming conclusions?

    How about another way of recognizing "extreme weather?" Are longitudinally compressed and increasingly frequent record breaking weather events a kind of larger weather? In some parts of the US this year's June has resembled later months last year, when records fell across large swathes of the country. If we say that 15 or 30 years of weather data is necessary to recognize climate shifts, what's the corresponding upper limit of age on a "weather event?"

    For my part if I see a steadily plunging Arctic ice volume figure I'm going to conclude that I'm looking at a long weather event, one that may or may not represent a change in climate. If the ice continues to deteriorate after 15 or maybe 30 years I'll pull the chip from the "inclement weather" bucket and drop it into the "climate change" receptacle.

  2. If the ice continues to deteriorate after 15 or maybe 30 years I’ll pull the chip from the “inclement weather” bucket and drop it into the “climate change” receptacle.

    I could have done a better job with that. Extreme weather is excursive weather; 15-30 years of excursive and hence extreme weather seamlessly merges into climate change. I think Pielke Jr. could do a better job of integrating this into his thinking and to do so would be advantageous in terms of not wasting time.

  3. Further monopolizing the conversation...

    I've been struggling to figure out what's bothering me about the discretion of Pielke Jr.

    We know we're essentially adding energy to the atmosphere by our habit of burning hydrocarbons. We've known for a long time that energy can't be "hidden" in a dynamic system; add kinetic energy to a ball and you will see it travel faster though not necessarily in a straight line. If the ball is traveling across a natural, imperfect landscape with lumps the ball will bounce a bit higher and a bit wider as it travels over its course.

    That's not a perfect analogy. Let's try another. Let's imagine that we're in a big field and we've got a supply of chickens. We're experimenting with chicken displacement. There's a varying wind. We start by simply dropping chickens a number of times. Depending on the instantaneous wind, the chickens land more or less at our feet. Occasionally one goes quite a distant, blown by a fortuitous gust. We see nature at work. Now we toss a few chickens, carefully consistent. Again, the chicken landing zone varies, sometimes quite a bit; we're picky so we can't conclude if we're seeing the energy we add to the chicken ourselves or just the work of nature, "natural variability." Now we begin loading chickens in a cannon and firing them downwind and downrange. Still variation in where the featherless corpses are landing but at last we can definitely see that we're adding energy to the chickens; they're clearly going much farther than the wind could possibly carry them. It was a messy process but conclusive several klicks beyond a reasonable doubt.

    Point is, the moment we start tossing chickens we -know- we're changing their landing zone all by ourselves. We don't have to fire the chickens from a cannon to know this and to insist on doing that is rather dense even if experimentally more perfect.

    We're adding energy to the atmosphere. We know it now. How could we possibly not be changing the trajectory of the weather? Why would we want to fire our weather through a cannon to make quite sure?

  4. And lest it be a distraction, it ain't the burning, it's the blanket; we're adding energy to the atmosphere by trapping energy that otherwise would arrive and depart.

    Also, the variable ability of chickens to fly (less than more) is a pleasing thought to add to the "can we add energy to a chicken?" experiment.

  5. Pingback: This is how it hits the fan: Michael Tobis on superfires and climate change attribution « Watching the Deniers

  6. MT - brilliant article. Few edits in there to clean up...

    As an Australian I watched aghast back in 2009 as fires ripped through my state killing close to 200 people. Then we had cyclone Yasi in Queensland which flooded one of our major cities (but not as destructive as Katrina).

    And yet the mainstream press is still "debating" (denying, misleading, confusing) the reality of climate change. Record temperature, record precipitation, major cities in the Western/developed world being flooded and destroyed, epic droughts, super-fires...

    I think the old "perfect storm" analogy has some merit: lack of understanding, lack of mitigation and inadequate adaptation measures will cost "us" dearly.

  7. Pingback: A Thousand Bastrops | Planet3.0 – A Few Things Ill Considered

  8. You would think there would be a way to figure out how to 'split the difference', so to speak, using models based on the skeptical thoughts of Veblen. I doubt without a major catastrophe, the deficit will ever be made up completely. It brings up the question of how our current forest management actually helps in the long run. The 'long run' seemingly getting shorter, of course.

  9. Don't forget the bark beetles. Due to global warming, fewer are killed off by winter's cold. And fire suppression leads to dense forests, making it easier for the beetles to move from tree to tree.

    When considering fire risks, think about a forest with many dead trees in it...

  10. Grypo: I doubt without a major catastrophe, the deficit will ever be made up completely.

    Perhaps it's because of the dull weather here but this morning I wonder whether we'll be capable of recognizing a "major catastrophe" when it comes along.

    The Waldo Canyon fire is the most destructive in Colorado's history in terms of property, taking 346 homes. Just a few days before Waldo, Colorado's previous most destructive fire happened, the High Park with 259 homes. These come not so long after the Hayman Fire which was the previous record-setter.

    Early innings but it does not seem that this collection of lives reduced to ashes and a fresh start is enough to count as a wake-up call.

    Expand the scope. Last year's Bastrop fire; is it part of a major catastrophe?

    Question: When thinking about "major catastrophe" is it more productive to consider these fires together, or separately? Why is the window of "catastrophe" confined to a single fire boundary? To analogize, no single arrival at a hospital of a person ejected from a car is a "catastrophe" but the annual statistics of an entire state or country for the cumulative mayhem were indeed catastrophic and hence were addressed.

    Insurance premiums in Colorado are creeping up in response to bad statistics that include extreme weather events. This year's fires will cause a bulge in those rates a few years from now. It's the "creeping" part that makes it easy to miss a major catastrophe even as it is happening around us. We're not good at mentally integrating a burgeoning disaster if it's slow to reveal itself.

  11. Interesting article and comments.

    Michael: "It’s fair to say that this phenomenon was not clearly predicted in climate models. Consequently, it is fair to say that it is not fully understood, and whether it will remain a persistent feature of the climate is speculative. Strictly speaking, formal attribution will be difficult."

    Agreed on the first point, but the seemingly fundamental nature of the change (slowing and increasing amplitude of Rossby waves associated with polar amplification) really makes me want to see an explanation as to why that was, and for that matter some idea as to the current efforts of modelers to catch up with reality.

    Sure, it's not fully understood, but recent work (e.g. Francis & Vavrus 2012) seems to indicate good progress.

    As to persistence, well, since we can't predict ENSO cycles I suppose we can't know they won't just stop, but I think the smart money is on such fundamental things continuing.

    Re formal attribution, that would seem to be just another way of stating that the models have failed to show this phenomenon.

  12. grypo, I don't think there's any difference to split or, contrary to the implication of the last few paragraphs of the article, any dispute about the results of the paper or about the fact that additional research will be needed to draw more fine-grained conclusions. I hate it when journalists who ought to know better (Tom Yulsman in this case) do things like that.

  13. Nothing against my grad school classmate Steve Vavrus, who is about as nice a guy as you can imagine and pretty darn sharp (like the rest of the gang) too.

    But ex post facto explanations are not that hard to come by. See the financial pages.

    Coming up with a proposed dynamical theory that accounts for something unexpected after the fact isn't nearly as hard as predicting it. Coming up with a theory that holds water and makes useful extrapolations is a very different matter than coming up with a proposal.

    People following along at home would do well to distinguish between observational data on the one hand, and theory and modeling on the other. An observation is an observation. Colorado is burning and Greenland is melting. There's no second guessing that. But more abstract knowledge takes a long time to emerge. There's no progress without first attempts, of course, but first attempts should be taken with due skepticism.

  14. As a person following this from home and looking at this from a different (and admittedly ignorant) perspective, I find it strange that it should be necessary to propose and then prove that adding energy to a dynamic system will be detectable.

    Rather, it seems the burden of demonstration would be on those who propose that energy can be added to a dynamic system and not be visible in its effects and rather immediately so. I've not seen any hypothesis for how this may happen, how energy added to a dynamic system can be hidden in a way we cannot see.

    So, turning the Pielke Jr. argument on its head, why not propose a means for hiding energy in a dynamic system then wait and see if the hypothesis is borne out by observation?

    In the meantime-- while the "hidden energy" crowd are waiting for confirmation-- the rest of us can get on with the application classical physics to the problem confronting us.

  15. The point about the burden of proof is an interesting one. To a considerable extent I agree.

    But it's a very different proposition to say "something will happen" than to say "this is happening because of this and that aspect of the dynamics of the situation". And one paper making such a claim cannot be regarded as definitive - to do so is to fail to understand how science progresses.

    Admittedly, most people, including some successful scientists, don't really think about that and don't really get it. Getting published is a judgment that your ideas are worth considering, not a judgment that they are correct.

    The judgment of correctness is actually not formalized anywhere as a general matter. IPCC WG I can be regarded as an effort to do exactly that, given the stakes of the matter at hand. But that's atypical and even IPCC admirers like myself must admit at best the process is not extremely refined.

  16. Doug,

    When I say "catastrophic", I should say cataclysmic. Like, when the default state policy becomes "tuck tail and run" instead of "fight and defend property", where the resources used to fight the heat, wind, and dead trees far outpace the usefulness to stop people's homes and businesses from burning. It is at that point that I would think the fire deficit would begin to close substantially.

    This brings me to what Steve said.

    I don't think Tom Y did a bad job here. But I would have used another angle to tell this story. Perhaps Tom is looking at this from a more positive perspective than I, but I think the article he wrote leaves the reader with the question I wrote above, "how our current forest management actually helps in the long run"? That's where I would start from. If the deficit of burned biomass is so large, at what point do we cut our losses? It's the same question asked by sea level rise. In Carolina, the state legislature, which appears to be an extension of the development industry, says we never cut our losses, or at least never pay attention to information that tells we should. They are saying, we need your capital, don't worry about losses or insurance. This goes along with Michael's comment about discounting the future. It's so bad that the market tells us, through its puppets, that investment is good even when price will certainly turn to zero.

  17. Thanks for your reply, Michael.

    Leaving aside my "hidden energy" provocation, there's an underlying tension here, a rift between urgency and the sometimes mandatory glacial progress of science. On the one hand in this particular case properly done science can't be hurried when it comes to testing predictions against observations while on the other hand we're also facing a situation a little bit akin to beginning a slide down an icy mountain slope with a brief window in which to decide how to arrest our fall.

    The dichotomy between the need for quick decisions and scientific circumspection is about as bad as can be imagined.

    Extreme weather of the short wavelength variety we're accustomed to talking about and are intuitively equipped to deal with (today's absurd convection as opposed to 2-3 years' worth) won't show up if additional energy in the atmosphere is fully homogenized, so I suppose a key question is how likely is it that additional energy will be perfectly dispersed? Well, we already know it's not; polar amplification is predicted and apparently observed, energy dispersal is lumpy.

    Further to Steve Bloom, perhaps we laypersons need a primer on the state-of-the-art when it comes to indications for/against homogeneous dispersal of added energy at vastly smaller scales than Rossby waves. Again, for me as a layperson my default short of having expert advice will be to look out the window and notice convection, assume that with extra energy that convection will be intensified. To me that seems a duly cautious approach.

  18. Thanks for the clarification, Grypo. Unfortunately it seems likely that when we reach such a point we'll be ipso facto cooked (if you don't mind the pun).

    I suppose the doings over the notorious NC-20 legislation are the embryonic form of "cut and run," which I'll offer represents an elliptical way forward. A spat over integration of models, observations and a conflict with commercial interests of the kind Michael refers to has resulted in a general elevation of discussion. The issue is on the table-- good.

  19. This is a perfect example of the reasons for cognitive dissonance I think a lot of us are experiencing when it comes to keeping our powder dry on conclusions while simultaneously acknowledging facts on the ground:

    "The Environment Agency said today that the April to June period this year was the wettest on record across England and Wales.
    England and Wales have already received more than double the long-term average rainfall for the month of June.

    The figures, which do not include yesterday's torrential rain across parts of the country, show the UK had received an average of 5.1 inches up until June 27.

    The figure is just a fifth of an inch off the total for 2007, which was the wettest June on record dating back more than a century."

    Wettest April to June since records began and more unsettled weather to come

    Subsequent to that article's publication this June was confirmed as beating 2007. "Dating back more than a century" unpacks to "since records began."

    This is what I'm referring to when I mention longitudinal compression of records. The heat wave currently plowing across the US is doing the same thing with temperatures this year, knocking over shortlived previous records. That heat wave follows this spring's eye-popping phenomenon. This a stampede we're witnessing.

    Circling back to Pielke Jr., he insists that clean data will be found in places unaffected by development-- a high bar to cross. Ok, but while there's been development in England recently it's hardly analogous with the "Mountain Shadows" subdivision in Colorado Springs nor has the countryside just recently been stripped of timber or experienced other novel confounding factors yet we're seeing out-of-bounds statistics being recorded.

    What I see is lumpy energy visible now, today, exactly what I'd expect given what I've been told we're doing to the atmosphere.

    On an amusing note, the poetically named "Spanish plume:"

    Giant hailstones, terrible floods, homes evacuated. Was all of this caused by the 'Spanish plume'?

  20. Hmm, it wasn't at all clear to me that this heat wave was approaching record intensity until I saw this in Friday's post by WU's Chris Burt:

    "The only previous June heat waves in U.S. history that compare to the current one were those of 1934, 1936, and 1954. The summer of 1934 went on to be the warmest on record for the U.S. (74.6° June-August average) and July 1936 the single hottest month on record (77.4° average).

    "Ominously, some of the June records that have so far been set this month have eclipsed those of June 1934 and 1936 (1954 turned out be a summer of only slightly above long-term average normal temperature)."

    The combination with the March event does seem striking.

  21. Here in Victoria, Australia we have been living with serious wildfire risks since white settlement. Devastating fires with major loss of life in 1939 triggered improved fire management, addressing many of the issues discussed in the wildfire deficit article Michael cited. "http://wwwp.dailyclimate.org/tdc-newsroom/2012/06/wildfire-deficit"

    Victoria has scheduled fuel reduction burns in national parks to manage fuel load, well equipped, organised and trained fire fighters with wildfire expertise, major public education programs for wildfire preparedness and plan to leave on high risk days campaigns. We also had major fires with 173 deaths in Feb 2009. The deaths happened on a day of unprecedented fire risk (temperature, wind, fuel moisture levels) and affected areas where people had moved to live surrounded by bush, although one small town was entirely burnt out.

    Whether there is a wildfire deficit depends on the local ecology. Bushfire ecologies, like Victoria's Sclerophyll forests, build up an accumulation of fuel over the years if fire is excluded. Rainforests do not - forest litter is relatively rapidly recycled by insects and fungi. Exclude fire long enough and the ecology can change from fire oriented to rainforest.

    There are three effects combining to make fires more dangerous:
    - Climate change effects (we were in a 10 year drought and the conditions were extreeme)
    - People ignoring the risks to live in natural surroundings, some not even aware of them
    - People assuming their safety is someone else's responsibility, which cannot be true in wildfire prone areas. People won't even attend bushfire safety presentations.
    Two of these are direct Perrow-like behaviours. The global warming effects make them worse.

  22. Mr. Tobis, a most excellent article, Sir!

    I previously was unaware of this site, but will be bookmarking it as soon as I post this comment. I learned of it over the Australia-based WatchingtheDeniers blog, which I really like. I would have come here without the really strong endorsement on WtD, but that shoved me this direction all the more quickly. They've not steered me wrong before.

    BTW, though I imagine you're aware of this, I recently read that at whatever point we reach an irreversible tipping point (if that's in the cards, which I believe looms large, and within decades), it could take as long as 500 years, give or take, for the full impacts of climate change to hit. Sorry, but I can't remember *where* I read it, but the writer did add the caveat that in theory at least, the whole premise might turn out wrong. Maybe we'll discover a magic bullet. Maybe an asteroid 150-200 miles across will come out of nowhere and alter the equation drastically (though I personally think not on a "magic asteroid" anywhere NEAR that size just popping up out of nowhere).

    Meanwhile, I believe we should get cracking, in a major way, right now. Problem is, politicians globally are essentially worthless in this regard. (Well, in most regards, at least in my native U.S., come to think of it.)

    Keep up the top-drawer work. Sir.

  23. Pingback: Summer in the garden [Stoat] - Dennis Flint High CountryDennis Flint

Leave a Reply

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