Defense Against the Dark Arts

Last summer we examined some first-class misinformation about global warming from the “Heartland Institute,” namely their NIPCC report. We showed that their fake take on Arctic sea ice revealed that they’re not real skeptics, they’re fake skeptics.

There’s an update to that report (full version here), in which they’ve modified their discussion of the state of Arctic sea ice. Mostly they’ve eliminated it. Except for some vague generalities which don’t amount to a hill of beans, the closest they come to discussing northern hemisphere sea ice is to talk about ice conditions in Hudson Bay and on the Great Lakes — which together constitute a tiny fraction of ice-covered waters in the northern hemisphere. But Arctic sea ice is one of the most important, and most obvious, evidences of climate change, so why would you issue a 416-page report on the state of the climate but say almost nothing, and certainly nothing substantive, about Arctic sea ice?

It surprised me (and it takes a lot for the “Heartland Institute” to surprise me) how blatantly they misrepresented the changes in Great Lakes ice cover. They only reference one paper (Wang et al. 2010, Severe ice cover on Great Lakes during winter 2008-2009, EOS, Transactions, American Geophysical Union 91: 41–42), which discusses factors influencing the extensive ice cover during the 2008-2009 season. These factors include ENSO (el Nino southern oscillation) and AO (Arctic oscillation), and they demonstrate interesting facets of the interplay of these teleconnections with the year-to-year ice cover fluctuations. It’s worth noting that although the 2008-2009 season witnessed extensive ice cover, its maximum value was not the greatest on record (since the 1972-1973 season), only 5th-largest.

The paper only briefly mentions the trend in Great Lakes ice cover, saying

Previous studies show that Great Lakes ice cover had a significant downward trend,about -1% per year, for the period between the onset of winter in 1972 and the end of winter in 2001. Nevertheless, during the entire period of the winters of 1972-1973 to 2008-2009 (Figure Ib), the downward trend disappears or even reverses. This indicates that (I) natural variability dominates Great Lakes ice cover and (2) the trend is only useful for the period studied.

Their figure 1b shows annual maximum ice cover (filled circles), annual average ice cover (open circles), and winter-season average surface air temperature (red open squares):

The NIPCC report gives a modified version of their figure

It further draws this conclusion:

Wang et al. conclude from their study that “natural variability dominates Great Lakes ice cover,” and that any trend in the data — of which there are some of a few years and one that is lengthier — “is only useful for the period studied.” Given this finding, there is no reason to attribute any change in the annual average ice area of the North American Great Lakes to anthropogenic global warming.

How should one evaluate this argument? Unfortunately it’s work to do so — but fortunately, it’s simple. There’s a single method which exposes the misdirection in almost every case of fake claims by fake skeptics, namely: look at the actual evidence. In detail.

Be sure to read as much of the scientific research as you can — don’t limit yourself to the sources they give. Be sure to study the data yourself — don’t take their word for anything. Perhaps most important, don’t limit yourself to looking only at what they show you — be especially careful of fake skeptics turning the spotlight on a time span that’s too brief or an area that’s too small. If you use these safeguards, the truth of the matter will usually become clear. It’s work — but it works.

Step 1: study more of the scientific research than they refer to. We could, for instance, examine more recent work from the same authors: Wang et al. 2012, Temporal and spatial variability of Great Lakes ice cover, 1973-2010, Journal of Climate 25(4):1318-1329, doi:10.1175/2011JCLI4066.1. There’s a nice powerpoint presentation which summarizes this research. It shows trends in ice cover for all the Great Lakes (including Lake St. Claire) separately and combined:

Note that there are sizeable trends, and the change in total Great Lakes ice over the years has been one of significant decline.

Why, you might wonder, is there such a strong declining trend in Great Lakes ice cover? Wang et al. answer that very question, thus:

That kinda says it all.

Step 2: study the data yourself. There’s quite a bit available from Environment Canada. They provide two principal measures, the maximum ice cover for each winter season, and the “Total Accumulated ice Cover” or “TAC,” which combines not only how much area is covered but for how long as well. Analyzing the data shows that all the trends identified in Wang et al. (2012) are statistically significant. Here, for instance, is the trend in TAC for the Great Lakes:

Here’s the trend in seasonal maximum ice cover:

Here are both trends on one graph:

Not only are both trends statistically significant for the Great Lakes as a whole, they are so for all five Great Lakes individually. Study the data yourself, the result is clear: there is a strong, statistically significant decline in ice cover on the Great Lakes. All of ’em.

Step 3: look at more than they show you, and be especially wary of time spans that are too brief and areas that are too small. It’s obvious that the NIPCC report has done exactly that, because Hudson Bay and the Great Lakes combined are only a small fraction of the area covered by sea ice in the northern hemisphere.

Earlier, I asked “why would you issue a 416-page report on the state of the climate but say almost nothing, and certainly nothing substantive, about Arctic sea ice? Maybe it’s because of the graph they didn’t show:

When it comes to Arctic sea ice, we should note what the NIPCC report did not do. They didn’t survey the scientific literature to give a fair report of the state of current research. They didn’t present an array of data, or even present data for the Arctic overall, in order to give a realistic portrayal of the situation.

We should also note what they did do. They surveyed the scientific literature for the express purpose of finding a few quotes which they could use to discredit global warming science. Yet in spite of how easy it always is to execute this tactic, it’s surprising how little they managed to find.


18 responses to “Defense Against the Dark Arts

  1. Northern hemisphere ice-out dates are very sensitive to warmer and shorter winters. On large lakes you need a fairly long stable period of cold weather to set up the ice so it can freeze solid and get thick from shore to shore. It’s like a phase change. On small ponds it’s more linear; they can freeze over quick in just a couple cold nights. Big lakes can’t do that.

  2. Definitely not Cherry Season for the NIPCC

  3. What makes me mad, and I don’t have much of a temper, is that science deniers can get a tax write-off by donating money to the nonprofit, nonpartisan hoax called the Heartland Institute. Thanks for highlighting the Heartland’s efforts to confuse the public. They have a lot of guts to claim that inland lakes, including the Great Lakes show no evidence of warming. The trends for every lake that I have seen are dramatic. Smaller lakes, that freeze up much faster than the Great Lakes show much stronger trends. The ice fishermen have been complaining a lot.

  4. Loking at the volume rather than area of Arctic sea ice was an eye opener for me. Is there a reason to believe the volume numbers are uncertain enough to justify using area instead?

  5. Thanks, Tamino. It’s amazing how blatant these guys are–it’s not as if it is hard to discern what is really happening with Great Lakes ice cover.

    I guess the only nice thing that I can say about them is that they must read the literature, in order to do these types of cherry picks. Of course, the ‘diligence points’ thus gained are more than offset by the ‘dishonesty points’ they lose, since the can hardly help but know better–it’s not ignorance, clearly, that produces a result such as theirs.

    Well, except ignorance of the meaning of the word ‘integrity,’ of course.

  6. You always have to cloud the issue with facts. They’re very unpleasant little things that always get in the way.

  7. Thanks for posting this, Tamino. You will be very interested in two related, recent papers on Great Lakes ice cover:

    Temporal and Spatial Variability of Great Lakes Ice Cover, 1973–2010
    Wang et al 2012
    J. Climate, 25, 1318–1329

    Interannual variability of Great Lakes ice cover and its relationship to NAO and ENSO
    Bai et al 2012

    Additionally, as of March 5, Great Lakes ice cover was down to 12% of the long-term average:

  8. Gavin's Pussycat

    Tamino, now you made me curious. As you have apparently more than just leafed through the NIPCC report, which is what I did: does it contain any substantive discussion of the global surface temperature record?
    I seem to remember a brief reference in the section on the satellite data, but this new version of the report contains no such section… or any substantive discussion of the satellite data either, it seems. Searching on UAH returns empty.

    • Marco van de Weert

      I did some searching in the document also, and the answer to your question is “no”. It cites three papers to cast doubt on UHI being properly taken into account, and it cites a few papers to argue present warming in the arctic is nothing special. That’s it. I guess Idso and Carter have convinced Singer to drop his “no warming in the satellite record” meme. And then there’s nothing left.

      • Gavin's Pussycat

        One could dream, and believe that Singer was shamed into this by John Mashey’s putdown of Heartland’s rag E&CN, where Singer month after month after month continued to refer to the pre-correction UAH data…

  9. “study more of the scientific research than they refer to.”

    To me, this is the most important step. Whatever the scientific contreversy (ex: health effect of weak radiation, magnetic field, microwave, etc) reading more scientific litterature helps a lot. After you have read a dozen of papers, you get a good idea about what is the status of the knowlede at that time.

    • Unfortunately, whenever I do this, I run afoul of the Citations Theorem: “The more papers you read, the more papers you have left to read.”

      • You are right. But, after a few papers you generally get the feeling about was is the state of the art. I can tell that I have kick ass of some proponent of the idea that food+microwave=danger that used the fact that 350 papers were covering this issue.

      • Gavin's Pussycat

        Corrollary: …and the ones you didn’t yet read are the ones that matter most

  10. I’d encourage everyone to click on Tamino’s Heartland Institute’s update link and check out the bulleted points…. It looks like they seem to have forgotten the last bulleted point: All the pigs are fueled and ready to fly!

  11. > the volume numbers are uncertain …?
    > enough to justify using area instead?

    Area: more data, better validated, covering more years
    Extent: oooh, bigger!
    Volume: ooooooh! Shiny!

    Those three measures (and others) are useful.
    How you use them?

  12. Heartland Institute, funded as it is by major oil producers such as Koch Industries, may have a secondary reason to leave off denial that arctic sea ice is in a long term decline. Beyond the fact that they are wrong. Currently there is an all out race to see who will get to drill for oil in the arctic, now that there is so much less ice.

  13. I’ll do a little cherry picking of my own. The 2011-2012 ice season on Green Bay (Lake Michigan) was practically non-existent. Not only did it cut drastically into my ice fishing season but also had significant effects on the local economy.