The beauty of a graph

Last week I wrote about how one image can say more than a thousand words, and this in the context of history. But it’s also applicable in other sciences, and it’s a great way to communicate science to a greater public.

Take for instance the famous and unjustly criticised “Hockey Stick” graph portraying the recent global warming of the Earth. In its current form first developed by Mann, Bradley & Hughes, it recently got a cool upgrade. Marcott and colleagues wrote an article in Science (8 March 2013), showing research into the how temperatures in the last 11300 years varied against the average.

(Note: if you are of the belief that global warming is a hoax and we are all duped by whatever brooding kind of evil scientists, please stop reading now. Don’t bother to continue, and don’t comment either. Read first through John Cook’s amazing website Skeptical Science, including rebuttals of any possible and imagined counterargument, and then come back to this article. I can wait.)

Although the article was rather full of calculations, blogger ImaGeo (Tom Yulsman) from Discoverblogs luckily found the following gem hidden in the supplementary material, and got the lead author to comment on it, too.


This is not only serious science, it is cool. ImaGeo called it “Art of the Antropocene: The Scythe”, and he’s right. It’s no longer a nice little hockey stick, but cuts right to the bone of the problem: the recent global warming is happening a lot faster than any other warming in the past 11300 years.

The researchers gathered 73 different datasets based on fossils of tiny organisms in the sea floor. But with each dataset comes uncertainty. In order to obtain a meaningful result that could be studied, the researchers let each of the 73 datasets vary within its margin of error, and that 1000 times. Each colour in the graph is one of those simulations (corresponding with a global temperature), in a technique called Monte Carlo simulation.

The result, plotted against the average global temperature shows that recent global warming, in absolute numbers not unprecedented, is happening at breakneck speeds. While contemplating the beauty of this graph, do keep in mind that uncomfortable message…

About Bruno Van de Casteele

Philosopher by education, IT'er by trade. Allround Armchair Skeptic, History Enthusiast, Father of Three. Twitter @brunovdc Personal website:
This entry was posted in Cool Stuff, Education, Science. Bookmark the permalink.

7 Responses to The beauty of a graph

  1. Eric Hall says:

    Nice find! This is excellent!

  2. Phil de Gruik says:

    Sadly this most comprehensive survey of paleontological proxies for temperatures globally of the last 2000 years tells us next to nothing about the last century.

    Yes, that’s right,
    “. . . the 20th century portion of our paleotemperature stack is not statistically robust, cannot be considered representative of global temperature change ”
    from the clarifying FAQ published 31 March on RealClimate.

    The dramatic ‘up-tick’ is little more than an artefact of the simulation method with limited data towards the end of the range.

    Such is the beauty of graphing it.

  3. This is about the tenth time the alarmists have come forward with a paper riddled with circular references which later must be withdrawn. Bull shit is bullshit regardless of what pretty pictures you make of it.

    • to Karabar: I’m sorry, I have must have missed something. “Science” still lists the paper as published and a quick search via Google does not give any more information. No withdrawal has been done.

      Could you elaborate on the circular references?

      Oh and one final thing. Using the word “alarmists” might come accross as a personal attack rather than an attack on the merits of the study. Unless you meant it as a compliment …

  4. Bruno Van de Casteele says:

    Yes, and as explained by the authors. Personally, I think the uptick should be analyzed also by comparing it with results from other studies. Science is never done in solitary or one paper at a time.

  5. Max says:

    The problem with the hockey stick graph was “Mike’s Nature trick of adding in the real temps to each series for the last 20 years.” If that’d done to all the graphs, then they’ll all agree in the end because they’re blended with the same “real temps” from the instrument readings. So I hope that wasn’t done here.

    • From what I understand, this graph isnt, and the uptick can be explained by having less data. Adding the latest data is not as such a problem (I have seen it used elsewhere correctly) but you need to specify and explain where you let the graph start. If that is done (and the explanation accepted by academic peers), then it becomes once more a very valid tool for public understanding. Because it is a real phenomenon.

Leave a Reply

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