THE WEATHER TRAP

January 12, 2014 9:45 am4 commentsViews: 744

Climate change is something that’s difficult to experience directly.  How many people can say, from personal experience, that it feels like the average temperature in your hometown, let alone the globe, has changed by a degree or two over the past century?

Instead, people tend to try to detect climate change on their own from their experience of individual, usually extreme events.  For example, “It never seems to get as cold as it used to.”  Communicators have learned this, and have taken to emphasizing individual events as evidence for or against climate change.

Weather Channel producer Shawn Reynolds @WCL_Shawn tweeted this incredible photo taken by pilot Hank Cain of Chicago on 1.7.2014 during the "Polar Vortex."

Weather Channel producer Shawn Reynolds @WCL_Shawn tweeted this incredible photo taken by pilot Hank Cain on 1.7.2014 during the “Polar Vortex.” It went viral on Twitter. Published by Michael Quirke under fair use.

I call this “the weather trap”.  Weather extremes will always take place, no matter how the climate changes.  Some will become more frequent, some less frequent.  Many weather extremes are so erratic that a statistically significant increase or decrease might not be detectable for many decades, even though the very real monetary consequences from such changes in extremes start accumulating from the get-go.  This means that just about any such attempt to link extreme events with climate change is doomed to logical failure.

For example, suppose that climate change has caused 60% of all temperature records to be record highs and only 40% to be record lows.  If Person A points out each record high temperature as evidence of climate change, Person B can point out each record low temperature as evidence of no climate change.

Who wins such an argument?  It seems to me that both persons come off badly.  A person hearing from Person A and Person B would be likely to notice that record lows still happen almost as frequently as record highs, and if the highs are evidence for climate change and the lows are evidence against, the evidence is almost 50-50. Not very compelling, so Person A loses.  Still, record lows certainly don’t mean the globe is getting cooler, so Person B looks stupid too.

Person A is on more solid footing saying that record highs are becoming more frequent (if the evidence supports such a statement) or are expected to become more frequent (if the evidence supports such a statement).  Still, if Person A wishes to relate such a statement to a particular weather event, he or she had better make sure that such events are increasing at that particular location.

Texas A&M AgriLife Research photo taken on 5/11/2011. Caption: "The Texas drought continues as the extreme dry weather is causing large cracks in the ground -obvious sign of lack of rainfall." AgriLife via Flickr.

Texas A&M AgriLife Research photo taken on 5/11/2011. Caption: “The Texas drought continues as the extreme dry weather is causing large cracks in the ground -obvious sign of lack of rainfall.” AgriLife via Flickr.

During the 2012 election there was severe drought in Texas.  Some filled Person A’s role by implying that Rick Perry’s state was being directly harmed by climate change.  Yet, an analysis we published in 2011 showed that rainfall in Texas had increased by 10%-15% over the past century.  At best, global warming contributed to drought intensity by increasing evaporation through warmer temperatures.

There’s strong evidence that drought is increasing globally because of climate change, but this does not mean that drought is increasing in every single location because of climate change.  According to the latest IPCC report, Texas is one of those locations, but until the observed trend turns around, it’s difficult to be confident about future changes, and even more difficult to conclude that those changes have already started.

Does Person B have any sort of solid position to stand on?  Perhaps the most defensible position in pointing out things like record lows is to argue or imply that either (a) record lows are bad things, so global warming wouldn’t be all bad, or (b) since record lows keep happening, global warming can’t be having a substantial impact.  Statement (a) is true, while statement (b) requires that all the potential impacts of global warming are measurable by the local change in temperature, and is thus more questionable.

The polar vortex invades eastern North America.  Graphic from University of Washington

The polar vortex invades eastern North America. Graphic from University of Washington

During the current “polar vortex” episode, we’ve seen another trap for Person A: arguing that an extreme event is caused by climate change because a scientific study finds a link between them.  As a general rule, a large fraction of apparent links detected by correlation are later found to be spurious, and statistical correlation is the main evidence linking the polar vortex excursion to global warming.  Meanwhile there’s other evidence pointing the other way, and most scientists in the field (such as myself) are waiting for physical proof of a viable mechanism connecting the two before they’ll accept that the correlation is physically based.

This is not the only time the press or advocates have trumpeted a connection between severe weather and climate change based on a number of studies that can be counted on one hand (tornadoes have a similarly tenuous link).  If the connection turns out to be nonexistent, we will again hear Person B saying scientists got another thing wrong about global warming, when in fact the problem is public officials and the media talking up what is at best a tentative connection.

If you’re a scientist talking about the possible connection between a particular extreme event and climate change, I recommend: (a) wait for a consensus to develop before even mentioning the idea; (b) be specific about whether the connection has been observed or is merely expected; and (c) be specific about whether it’s known that the specific location of the particular extreme event is or should be seeing the frequency or severity change.  Even following those guidelines, whatever you say will be picked up and translated by Person A and Person B.

If you’re a member of the general public, you can bet that the truth is a lot more nuanced than what either Person A or Person B is telling you.

Tags:

THE SCIENTISTS' COMMENT THREAD

  • I think the last sentence is a good summarizer. Some of us are definitely concerned with the unwarranted certainty with which many claims regarding climate and weather events are made. In many cases, the problem originates with the press, or public statements, as you state.

    You bring up another issue that I think urgently needs discussion, and that is the relationship between temperature and precipitation, with respect to both magnitude and timing. The relationship between the two is the relevant metric for natural vegetation and crops. Temperature increases in the absence of P changes are a different matter from the same T changes in the presence of P changes, in terms of their ag and ecosystem effects. And we know why, mechanistically, from plant physiology.

  • Conveniently enough, there’s a paper in Nature Climate Change published yesterday that talks about the concept I outline in my first two paragraphs…called “attribute substitution”. See Zaval et al. 2014.

  • I find the detailed assessments as part of the BAMS Sate of the Climate reports in 2011 and 2012 (Peterson et al, 2013) particularly useful.

    In the latter study 18 different research groups examine 12 extreme from 2012 and four longer term notable events.

    In approximately half the studies anthropogenic climate change was noted as a contributing factor to the extreme event examined. This is in addition to the impact of natural weather and climate fluctuations. Each case review indicates the attribution difficulty that John Nielsen-Gammon has identified.

    They found almost no anthropogenic climate change influence in the case of a wet UK, Japan or eastern Australia in 2012. They found an anthropogenic in the case of sea level rise enhanced flooding for Superstorm Sandy and for extreme precipitation in the northern European summer.

  • I agree that it’s important to avoid the “weather trap.” It’s clear that both journalists and an emerging group of “climate science communicators” prefer to talk about unusual weather, because it’s visceral and gets people’s attention. I can understand the desire to bring climate change home, to emphasize the here-and-now phenomena. But the media drumbeat associating daily weather with climate change feels to me like “ambulance chasing.” It’s technically true that a shift in the distributions of temperature will produce many more record highs than record lows, but this shift is happening very slowly by normal human standards.

    The more important point is that CO2 has “only” increase about 40% since preindustrial times, but “Plan A” is to add another 100% or more. So the real concern about dramatic changes in climate are in the future. The reason we know this has nothing to do with tracking day-to-day changes in weather and extrapolating to a warmer future. Rather, we know that adding 4 or 6 or 8 Watts per square meter will warm the world far more than anything we’ve seen up to now, because we understand basic physics.

Leave a Reply

You must be logged in to post a comment.

PUBLIC COMMENT THREAD

  • http://ecologicallyoriented.wordpress.com/ Jim Bouldin

    I think the last sentence is a good summarizer. Some of us are definitely concerned with the unwarranted certainty with which many claims regarding climate and weather events are made. In many cases, the problem originates with the press, or public statements, as you state.

    You bring up another issue that I think urgently needs discussion, and that is the relationship between temperature and precipitation, with respect to both magnitude and timing. The relationship between the two is the relevant metric for natural vegetation and crops. Temperature increases in the absence of P changes are a different matter from the same T changes in the presence of P changes, in terms of their ag and ecosystem effects. And we know why, mechanistically, from plant physiology.

  • http://atmo.tamu.edu/profile/JNielsen-Gammon John Nielsen-Gammon

    Conveniently enough, there’s a paper in Nature Climate Change published yesterday that talks about the concept I outline in my first two paragraphs…called “attribute substitution”. See Zaval et al. 2014.

  • http://glacierchange.wordpress.com/ Mauri Pelto

    I find the detailed assessments as part of the BAMS Sate of the Climate reports in 2011 and 2012 (Peterson et al, 2013) particularly useful.

    In the latter study 18 different research groups examine 12 extreme from 2012 and four longer term notable events.

    In approximately half the studies anthropogenic climate change was noted as a contributing factor to the extreme event examined. This is in addition to the impact of natural weather and climate fluctuations. Each case review indicates the attribution difficulty that John Nielsen-Gammon has identified.

    They found almost no anthropogenic climate change influence in the case of a wet UK, Japan or eastern Australia in 2012. They found an anthropogenic in the case of sea level rise enhanced flooding for Superstorm Sandy and for extreme precipitation in the northern European summer.

  • Pingback: WeatherBrains 416: This is Only the Second One : The Alabama Weather Blog

  • http://biocycle.atmos.colostate.edu/ Scott Denning

    I agree that it’s important to avoid the “weather trap.” It’s clear that both journalists and an emerging group of “climate science communicators” prefer to talk about unusual weather, because it’s visceral and gets people’s attention. I can understand the desire to bring climate change home, to emphasize the here-and-now phenomena. But the media drumbeat associating daily weather with climate change feels to me like “ambulance chasing.” It’s technically true that a shift in the distributions of temperature will produce many more record highs than record lows, but this shift is happening very slowly by normal human standards.

    The more important point is that CO2 has “only” increase about 40% since preindustrial times, but “Plan A” is to add another 100% or more. So the real concern about dramatic changes in climate are in the future. The reason we know this has nothing to do with tracking day-to-day changes in weather and extrapolating to a warmer future. Rather, we know that adding 4 or 6 or 8 Watts per square meter will warm the world far more than anything we’ve seen up to now, because we understand basic physics.