I recently read Emily St. John Mandel’s Station Eleven, which didn’t quite overwhelm me the way it has overwhelmed many others — though I liked it. It’s good, but it could have been great. The post-apocalyptic world is beautifully and convincingly rendered: I kept thinking, Yes: this is indeed what we would value, should all be lost. But the force of the book is compromised, I think, by its chief structural conceit, which is that all the major characters in the novel’s present tense of civilizational ruin are linked in some way to an actor named Arthur Leander who died just before the Georgia Flu wiped out 99.9% of the human race. This conceit leads Mandel to flash back repeatedly to our own world and moment, and every time that happened I thought Dammit. I just didn’t care about Arthur Leander; I didn’t want to read fairly conventional realistic-novel stuff. I wanted to rush through all that to get back to the future Mandel imagines so powerfully.

All that said, I have one small thought, totally irrelevant to my feelings about the book as a whole, that keeps returning to my mind. In one of the book’s first scenes, a troupe of musicians and actors (the Traveling Symphony) is walking along an old road somewhere in Michigan, and it’s very very hot, over a hundred degrees. This is twenty years after civilization died, which makes me wonder: Would the world by then be any cooler? If all of our culture’s heat sources ceased functioning today — no more air conditioners emitting hot air, no more internal combustion engines, no more factories blowing out smoke — how long would it take before there was a measurable cooling of the world’s climate?


  1. The way that researchers have tried to answer your question is by seeing what happens to their climate models when they dial artificial carbon dioxide emissions and perhaps other man-made interventions down to zero while holding all other environmental factors equal. The models suggest that temperatures won't drop appreciably for 500 years ("eliminating CO2 emissions resulted in stable global temperatures for the following five centuries of model simulation") or 1,000 years ("the climate change that takes place due to increases in carbon dioxide concentration is largely irreversible for 1,000 years after emissions stop").

    There are obviously many assumptions and counterfactuals and what-have-yous, including all the usual caveats about regional differences, but the short answer seems to be that "twenty years after civilization died," global mean temperature would not be noticeably different.

  2. Thanks, Adam — some folks on Twitter earlier were pointing to similar documents. That definitely answers my BIG question.

    I guess I'm also wondering whether the other heat-producing elements of our civilization — other than the production of greenhouse gases, I mean — make much of a difference, locally if not globally. For instance, I'm in Chicago now. If humans went nearly extinct and our machines stopped working, then there would be no air-conditioners expelling hot air, nor the heat generated by internal-combustion engines. Those sources of heat production would happen immediately. Then, as plants started to reclaim more and more asphalt- and concrete-covered areas (especially roads, as they break apart without maintenance), the city heat-island effect would diminish. So I imagine that within 20 years urban areas, anyway, could be somewhat cooler … but that may be just my imagination, to coin a phrase.

  3. My layman's understanding is that, should all the operations of civilization that make it into a heat engine suddenly cease, immediate contributions to warming (combustion engines, AC, coal-burning electrical generators, etc.) would drop but their delayed effect would take roughly four decades to be fully felt. Loss of sea and glacial ice and thus a diminished albedo effect would also argue against any sort of quick recovery. So the Earth will likely be on a warming trend for centuries, if not millennia.

    Also, the result of planes being grounded for three days after 9/11 was a surprisingly sudden heat spike in North America. See


    Modeling environmental effects is far too complex for most of us to get our heads around, but that doesn't stop reckless arguments and counter-arguments about global warming and global cooling and its causes from, er, clouding the public sphere.

  4. While I agree with the main thrust of the comments of Adam and Brutus, I want to make a small but significant correction regarding the article Brutus cites at greenmedinfo. If you follow the link to the original short communication at Nature, published in 2002, you will see that it is not the case that the grounding of planes after 9/11 resulted in a "surprisingly sudden heat spike." There was no heat spike at all: the change studied in the article was a change in Diurnal Temperature Range (DTR), a measure of the difference between the daily max and minimum temperatures. There was a slight uptick in the DTR the three days planes were grounded, which could well have been due to the absence of contrails. But–and this is a big but–there are a couple of complicating factors here the authors do not discuss.

    First and most importantly, they fail to mention that DTR's have been steadily shrinking as a result of global warming for the past 100 years. (For a thoughtful recent treatment of DTRs and climate change, see http://www.sciencedirect.com/science/article/pii/S2212094714000346.) So to compare the three days post 9/11 against DTRs from just 1971-2000 is a bit misleading; to fail to mention the bigger picture of what DTRs indicate and that, in fact, a slight rise in DTR suggests that contrails might actually be damaging is very misleading.

    Second, the authors had to stagger the DTR data because of the timing of the event relative to that day's minimum temperature, and the lifting of the flight ban before the max temp had been reached on the 14th. It's hard to tell how this shift affects the story.

    Third, and most vexingly, the authors fail to indicate where they measured DTRs, referring only to "stations across the US," and also noting differences in OLR (an indicator of cloudiness) between western and eastern US. They simply do not show enough data for the reader to evaluate their approach (I'm surprised this got published in Nature, frankly; it must be because one year after 9/11 people were desperate for any seeming assurance that things weren't so bad). Fourth, it would have been nice if they had zeroed in on climate or DTR effects in the NYC region, because the ash dispersed from the event would have its own effects on the microclimate in the area, but coastal areas are subject to other factors that make such measurements difficult.

    Anyhow, as Brutus wrote, modeling environmental effects is indeed complex.

  5. I'm out of my depth to referee the veracity of claims pro and con, so I'll sit this one out.

Comments are closed.