It was an irresistible factoid, widely covered and still found all over the web: in the three days after the September 11 terrorist attacks, when all commercial air traffic was banned from American skies, the average daily temperature range over the continental US got a sudden, substantial stretch (see chart).
The disappearance of jet condensation trails, or contrails, had unmasked their remarkable effects, researchers suggested (Nature News, subscription). Lack of contrails seemed to widen the gap between the lowest overnight temperatures and the daytime highs.
But this may be a mere coincidence, not evidence of aviation impacts, according to a series of studies since then. These doubts have gotten little if any press, though climatologists seem to be listening – the latest paper made Geophysical Research Letters’ top five downloads last week. The study (subscription) gets a story today in Nature News.
The odd shift in daily temperature range (DTR) first came up in a 2002 Nature paper (subscription) by the University of Wisconsin-Whitewater’s David Travis and colleagues, who concluded, “the absence of contrails was responsible for the difference between a period of above-normal but unremarkable DTR and the anomalous conditions that were recorded.” (They also published a follow-up analysis in the Journal of Climate.) It was the size of the effect that was surprising: scientists had already hypothesized that when jet vapour trails spread into high cirrus-like clouds, they trap escaping heat throughout night and day, and, to a lesser extent, reflect warm daytime sunbeams – so that they warm the Earth overall but can squash daily temperature ranges.
Gang Hong of Texas A&M University and his colleagues offer the latest counter-argument in GRL. They attribute the temperature pattern after September 11, 2001 to shifts in low clouds – which are thicker, warmer, and thus far more influential than high clouds such as contrails, they say. Key among their findings is that 2001’s hairpin turns in DTR fall within the range of natural variability for the early-Septembers of the previous 30 years.
But that’s not the last word, according to Travis. He stands by his assertion – which, he says, is “more like a hypothesis” than a firm conclusion – that the spike in temperature range was so out of character that contrails probably boosted it.
Hong et al.’s research, Travis points out, found that during 1971-2001 there was only one other year where natural weather patterns made daily temperature ranges surge wildly around September 11. (The other variability identified by Hong et al. doesn’t follow this pet pattern, the one that “makes physical sense” when contrails are removed, he says.) Travis says that other meteorological data explain the change in 1975, but not in 2001.
But that brings the debate back around to another batch of Travis’s critics. Arizona State University’s Adam Kalkstein and Robert Balling found in their 2004 paper in Climate Research that unusually clear weather in September 2001 accounted for the pattern post-terrorist attacks.
In fairness, says Travis, his hypothesis hinges on a once-in-a-lifetime accidental experiment brought on by a national emergency. For those seeking to support or refute the September 11 contrail effect, he says, “It’s impossible to repeat this study”.
That shouldn’t make his widely echoed claim unfalsifiable. The question is how the Nature paper’s provocative idea weighs against data from alternative approaches by Hong et al., Kalkstein and Balling, and others. Any comments as to how they balance out?
Chart: Departure of average diurnal temperature ranges (DTRs) from the normal values derived from 1971–2000 climatology data for the indicated three-day periods in September 2001. Reproduced from Nature 418, 610 (2002).