Statistics

Stream: xkcd’s Global Warming Time Series Mistakes

The first portion of the cartoon, presented for the sake of illustration.

The first portion of the cartoon, presented for the sake of illustration.

Today’s post is at the Stream: xkcd’s Global Warming Time Series Mistakes.

The popular web cartoon xkcd has provided a wonderful opportunity to plug my must-read (and too expensive) book Uncertainty: The Soul of Modeling, Probability & Statistics. Buy a copy and follow along.

In this award-elligible book, which has the potential to be read by millions and which has the power to change more lives than even the Atkins Diet, I detail (in the ultimate chapter) the common errors made in time series analysis. Time series are the kind of data you see in, for example, temperature or stock price plots through time.

The xkcd post—thanks to the many readers who emailed about it—“A Timeline Of Earth’s Average Temperature” makes a slew of fun errors, but, and I want to emphasize this, it isn’t xkcd’s fault. The picture he shows is the result of the way temperature and proxy data are handled by most of the climatological community. Mr Munroe, the xkcd cartoonist, is repeating what he has learned from experts in his attempt at being humorous (but nobody bats a thousand), and repeating things from experts when you yourself don’t know the subject is a rational thing to do.

The plot purportedly shows the average global temperature, presumably measured right above the surface, beginning in 20,000 BC and ending in the future at 2100 AD. Mr Munroe misspells “BC” as “BCE” throughout the cartoon, incidentally, and leaves out “AD”.

No, I’m kidding. “BC” means “Before Christ”, which some academics, sensitive creatures that they are, find offensive on behalf of people they haven’t met, and so they change it to “Before the Common Era”. And how do they demarcate the “Common Era”? By the birth of Christ, a.k.a. BC. The same people who gave us “BCE” gave us “safe spaces”. Skip it.

Now I’m going to show exactly why xkcd’s plot fails, but to do so is hard work, so first a sort of executive summary of its oddities.

(1) The flashy temperature rises (the dashed lines) at the end are conjectures based on models that have repeatedly been proven wrong—indeed, they’ve never been proven right—by predicting temperatures much warmer than today’s. There is ample reason to distrust these predictions.

(2) Look closely at the period between 9000 BC until roughly 1000 AD, an era of some 10,000 years which had, if xkcd’s graph is true, temperatures much warmer than we had the Internet. And this was long before the first barrel of oil was ever turned into gasoline and burned in life-saving internal combustion engines.

(3) There was no reason to start the graph at 20,000 BC. If xkcd had taken the timeline back further, he would have had to have drawn temperatures several degrees warmer than today’s, temperatures which outstrip the threatened warming promised by faulty climate models. And don’t forget that warmer temperatures are always associated with lush and bountiful periods in earth’s history. It’s ice and cold that kill.

(4) The picture xkcd presents is lacking any indication of uncertainty, which is the major flaw. We should not be looking at lines, which imply perfect certainty, but blurry swaths that indicate uncertainty. Too many people are too certain of too many things, meaning the debate is far from “settled.”

The temperature at 20,000 BC was, Munroe claims, surely after referring to expert sources, about 4.3 C colder than the ad hoc average of temperatures from 1961-1990.

Was it actually 4.3 C cooler? How do we know? Forget the departure from the ad hoc average, which is a distraction. How do we know what the temperature was all those years ago? After all, there were no thermometers.

The answer is—get a pen and write this down, it’s crucial …

Go there to read the answer and discover why you’ve been looking at time series the wrong way.

Categories: Statistics

30 replies »

  1. Are you having trouble understanding the graph? Whether it is accurate or not, it does not indicate that the “period between 9000 BC until roughly 1000 AD” had “temperatures much warmer than we have now”. Even if it did, of course, that would be quite beside the point, as are the warm temperatures earlier in the Earth’s past.

  2. Lee,

    Glad to hear you say it. Many who fear global warming have been predicting the most terrible doom for a 1 degree temperature increase (which that period saw over ours), temperatures which they call “much warmer.” But you knew they were wrong.

  3. I was wondering how long it would be until I saw a commentary about this from you. Good article.

    Still on my way to that last chapter!

  4. Dr. Briggs,

    I’m afraid I have no idea what the intended meaning of your reply could be.

    But one ineluctable conclusion is that you’re having inordinate difficulty in parsing a simple time-series plot, and can’t see your error even after it’s pointed out to you.

  5. I note Briggs is a bit embarrassed to have to ask for money to impart knowledge.

    Nobody should feel shy or embarrassed for asking or requesting payment for their work. Of course they do but it is important to remember
    It’s your work and your money which is hard earned and the price reflects it’s real value. In terms of hourly rate it’s very cheap. That’s how people should view this book.

    You will find it’s only people who know you who will quibble about the price. Strangers will pay without batting an eyelid.

    “People must pay for knowhow” my Dad.

  6. Dr. Briggs,

    In your silent haste to correct the error that I pointed out above in your stream article, you’ve substituted a phrase that makes no sense whatsoever, I’m afraid. The garbled version of it reproduced here is even worse.

    Here’s hoping you eventually get it right. I’m rooting for you! And, as usual, no acknowledgement will be necessary.

  7. Joy: You’re very diplomatic. Unfortunately, Briggs appears to have wondered into an area he is unfamiliar with—sales—and is winging it. Perhaps consulting with an expert or two might not be a bad idea.

    Lee: The reply, I believe, means the world was hotter in the past and we’re all still here. You are dismissing the past as if it’s irrelevant to today, which is quite in contrast to global warming theory, that insists tomorrow is going to be too hot for humans to survive and it will be hotter than any time in the past when humans existed. The past is very important in global warming theory. Without it, there’s no “unprecedented” warming.
    One cannot properly tack instrumental temperatures on a proxy graph. It must be all proxies. Even Marcott understood that (though the media did not). Plus, the graph doesn’t go as far back as proxies are claimed to (seems one needs to go back approximately 200,000 years for that—or note they don’t have the data so no one knows if it was hotter way back then), but there is no explanation for why older data is not part of the graph—which will lead to the claim of “cherry-picking” by the person creating the graph. Why truncate the data at all?

  8. Mysteriously, the Common Era began 2016 years ago. I wonder what major historical or geophysical event occurred to distinguish 1 CE from 1 BCE? No one seems to know.

  9. And Briggs, the academic world was a better place, and certainly more interesting, when academics were required to sell their own lectures, at whatever price the market would bear in the absence of helicopter money.

  10. Ms/Mr Warnings:

    For good or ill, we are stuck with calendrical terminology based on various mythologies, even after civilization has moved on. Our days of the week are named after Roman and Norse gods — but most rational people know that the persistence of these names doesn’t mean that those gods were real.

  11. Lee Phillips,
    It depends what you call civilisation.
    It’s not civilised to be sarcastic all the live long day.

  12. We’ve “moved on” yet no attempt was made to obliterate the names of Roman and Norse gods from our weekdays as is being attempted with the name of our Lord from our system of years.

    Perhaps from Jesus Christ we haven’t moved on; we seemed determined to love Him or hate Him.

  13. Dr. Briggs,

    I appreciate your sisyphean efforts. One in ten thousand may be persuaded, while the rest pat each other on the backs, congratulating themselves on how smart they are. XKCD, generally, is on the better end of the spectrum, but still has problems. On one hand, it gets some things right: https://xkcd.com/882/ But on the other it commits big philosophical sins, like this global warming comic, and this one: https://xkcd.com/1478/

    Generally, I think it’s because the author often appears to be one of the many now who believe that everything is quantifiable.

  14. xkcd sometimes projects a certain smugness (as with this one) that’s unappealing. Calvin and Hobbes was a more clever critic of human foibles.

  15. Dr. Briggs,
    Your losing comment about the missing uncertainty indicators on the projections reminded me of Pat Franks work as seen in the video at Heartland(https://www.heartland.org/topics/climate-change/index.html).
    Could you possibly comment on his analysis of the models?
    If he is correct in his error propagation calculations these models are worse than you so adamantly assert they are (if that is possible).

  16. You only need to look at the sources Munroe cites on the top right of the diagram to know that his graph is a crock, especially Marcott 2013, HADCRUT4 and the IPCC:

    Munroe is a right-on, rabid warmist and has hilariously created his own hockey stick by disappearing the Roman WP (no mention), the MWP (‘too regional’ = little squiggle) and even the LIA (little squiggle) and then splicing in model forecasts.

    He has done some funny stuff in the past but his best days are over IMO.

  17. Prof Briggs,
    Thank you for your book and blog. While I am not sure if you are right in the things you say, you challenge me to think. And, for me, that is more important. I appreciate it and thank you.

    Question I have had on my mind reading your writings: You make a good argument for the climate models not being 100% precise, but what if they are “directionally” accurate? If they have the direction right, doesn’t that count?

  18. FYI – Climate Fraud Whistleblower Rewards Program | The Deplorable Climate Science Blog http://realclimatescience.com/2016/09/climate-fraud-whistleblower-rewards-program/

    Guest post by Kent Clizbe. Government employees willing to tell the truth about climate can be handsomely rewarded.

    The federal government has a vigorous program to identify and prosecute fraudulent grant applications and administration–the False Claims Act. It provides rewards for insiders who reveal grant fraud.

    As detailed daily on Real Science, it is clear that there is wide-spread fraud in “climate research.”

  19. “I wonder what major historical or geophysical event occurred to distinguish 1 CE from 1 BCE? No one seems to know.”

    If you’re trying to hint that a supernatural revolutionary was born in that period, you would be betraying a profound lack of knowledge of the collection of primitive drivel that is the canonical source of witterings about him and his putative Dad (a genocidal maniacal foreskin-obsessed Sky Monster). (“The Bible”, for those playing at home).

    Although Jeebus (a.k.a. Ye[ho]shuah ben Yusuf) is almost certainly a fictional composite (like Emmanuel Goldstein, Abraham, Noah, Moses, David, and Solomon), the ‘background’ is clear: he was putatively born when Herod was still alive, and Herod died in 4 BCE.

    So… nice ‘zinger’, dude… but you really ought to go read that book again. If you read it cover to cover, with a rational post-Enlightenment hat on. you ought to be an atheist by the time you put it down.

  20. The post was about a cartoon which showed an unscientific view of historic climates
    Being a cartoon it had a single point . The belief in God has nothing to do with the case.
    Kratoclastes should get out his books on the Hellenistic Period. They are rather hard to read, though , especially that by Tarn. It will make your hair grey but be good for the soul. There were two Herods at least. A.D ( Anno Domini )is a convention. BCE (Before the common era.) is also a conventional date, both are used by scholars, especially those who don’t understand Latin.

  21. Let’s see how the scrutineers react…

    @M E ought to understand that the internet recognises the “I noes Latinz” angle is an attempt at gravitas: as Sterne, Shaftesbury and de La Rochefoucauld all pointed out, gravitas is “the essence of imposture”; “design, and consequently deceit”; and “a mysterious carriage of the person crafted to conceal shortcomings of the mind”. It is something of which I will never be accused, by design – I detest gravitas, eminences grises and solemnity because they are shallow control tropes.

    I also know full well how many Herods there were (8 in the direct lineage).

    If you are trying to pretend that the relevant Herod was Herod Archelaus – or any Herod other than Herod the Great – then you’re using the typical “True Believer” trope of trying to obfuscate the factual innaccuracies in that primitive congeries of ethnocentric genocide-rape-slavery supporting drivel that is the foundational literature for two of the three big cults that enable charlatans to live off the productive capacity of the gullible.

    In other words, you’re relying on the notion that by invoking “I noez Latinz”, people will assume you’re educated and nobody will fact-check the claim (that’s a trick the ecclesiasts tried to pull for almost twenty centuries).

    That’s a yuuuuge error in a world where information is almost free and you’re not allowed to burn heretics: it amounts to egging the pudding to an extent that will result in pudding-flavoured scrambled eggs, not eggy pudding… because it comes unstuck very rapidly.

    The document is clear: Herod the Great was the one who sought to killing of the male children; Herod Antipas was the one before whom Jeebus was brought for trial (and who had John the Baptist beheaded); Herod Agrippa I was the one who martyred the terrorist James; Herod Agrippa II ‘saved’ Paul.

    Now watch this… it’s where you lose both bishops and your queen in 3 moves (and I don’t even play chess… people who do are almost always Betas).

    (1) Matthew 2:19-20 states clearly that Jeebus was taken to Egypt when Herod died.
    (2) All of Herod the Great’s sons outlived Jeebus; ergo
    (3) the ‘Herod’ who ruled when Jeebus was putatively born, was a fortiori Herod the Great.

    QED, or perhaps ISOL (interdum stultus opportuna loquitur, for those who know their Fielding).

    Eggs… scrambled.

    As to your assertion that “the belief in [invisible sky maniacs] has nothing to do with the case” – why was the ‘BCE/BC’ issue brought up? Answer – for anyone with an IQ above ‘educable’: to push the Jeebus death-porn cult into the frame. By bringing the cult into the frame, the the inaccuracy of the whole ‘BC’ schtick becomes immediately relevant.

    This “We’re not talking about religion now [although we just did]” trope is seen in many religiotards – even though their foundational literature includes gems like “Don’t bear false witness”. The ‘righteous’ get all dishonest and politician-y any time the nonsense they pretend to believe is subjected to scientific dismantling. They also tend to start talking about stuff like ‘tolerance’ and ‘respecting people’s faith’ – as if they would not set atheists on fire if they still held the whip hand.

    Here endeth the lesson.

  22. Your 4 points.

    1. You are expressing distrust in models however your distrust doesn’t disprove them. There is a wealth of data out there within which these models are analyzed and tested. Yep, none are perfect, but the graph represented at XKCD hasn’t been disproven. In fact, its been confirmed. If you want to disprove it, do so. Your commentary doesn’t disprove anything….because that’s now how science works.

    If you want your arguments to hold ANY water, either disprove the c02 emissions of man or the physical properties of c02. Those are your choices. Failing those, you can’t disprove the overarching theme seen universally in all models and across every discipline.

    2. Point 2 is incoherent. Did you look at the graph? I assume you accidentally deleted some of the content.

    3. Point 3 is logical, but irrelevant. Do we measure our driveway in lightyears? No. The temperature changes of earth over millions of years has no bearing in this conversation. The resolution in temperature records isn’t detailed enough for that anyway. Was the earth warmer? Yep. Irrelevant. Should we discuss the temperature of the earth 6 billion years ago and compare? Yeah, no point.

    4. Point 4 is plain wrong. The graph shows a dotted line to represent exactly what you say they aren’t symbolizing. See point 1. Prove the models wrong, or find another way to contribute to human knowledge.

Leave a Reply

Your email address will not be published. Required fields are marked *