Greg Krehbiel's Crowhill Weblog - Content

Thoughts on life — News, culture, politics, beer, art, science, education, religion and ethics

Other Crowhill sites:
Crowhill PublishingGreg's Book Publishing blog
Greg's Marketing blogGreg's Home Brewing blog

Don’t get too excited about arctic ice

by Greg Krehbiel on 9 September 2013

There’s a proverb that goes “the first to plead his case seems just, until another comes and examines him.”

You may have seen that arctic ice increased this year, despite some projections that the arctic would be ice free.

Part of the problem here is the “report the craziest story you can find” ethic at newspapers.

“Polar bears will be dying next year for lack of ice” sells more ads than a sober story on what may be happening in the arctic.

Last year was a low year for arctic sea ice. This year there’s a lot more ice. “Look! Global cooling!”

No, it doesn’t work that way. The part in this article about regression toward the mean helps clear it up a little.

There’s a principle in statistics known as “regression toward the mean,” which is the phenomenon that if an extreme value of a variable is observed, the next measurement will generally be less extreme. In other words, we should not often expect to observe records in consecutive years. 2012 shattered the previous record low sea ice extent; hence ‘regression towards the mean’ told us that 2013 would likely have a higher minimum extent.

That article doesn’t tell the whole story either, but at least it puts it in some perspective.

I am still very skeptical about AGW. There is reason to believe that CO2 is a lagging indicator of changes in climate, not a cause. There are also reasons to believe that CO2 is a minor factor in climate. IMO there are a lot of good reasons to doubt the alarmist story, most particularly because they don’t seem to want to deal honestly with longer time periods.

The absolute biggest reason I doubt the AGW case is the shrill, nutty, dishonest rhetoric the advocates employ.

Still … you have to be honest about these things. This year’s increase in arctic ice no more disproves AGW than last year’s dearth of ice proves it. In another ten years we might have a decent chance of knowing if the AGW theory is correct — but only if scientists allow an honest dialog. In the current environment, there is very little hope of getting an honest answer.

-- 2013-09-09  »  Greg Krehbiel

Talkback x 16

  1. Dave Krehbiel Dave Krehbiel
    9 September 2013 @ 5:52 pm

    as Greg says above, and as John has said in the past, you can’t make good decisions based on one or two years worth of data. I think you can get a more helpful perspective on climate trends by doing a Google image search on “interglacial”.

  2. Greg Krehbiel Greg Krehbiel
    10 September 2013 @ 8:11 am

    Dave, that is actually a very interesting search. It highlights one of my main complaints with the AGW crowd, which is that they refuse to look at longer time frames. In the context of longer time frames, nothing that has happened recently seems all that odd.

  3. John Krehbiel John Krehbiel
    14 September 2013 @ 9:15 am

    Actually, in terms of the longer-term changes, and their causes, the current situation is even more odd. We should have been in a cooling period, according to things like orbital changes and such, for about a century. Here’s a graph that shows hundreds of thousands of years worth of change, for instance.

    The thing about current ice cover is that it is largely young ice, and therefore very likely to melt again next summer. The amount of ice older than, say, 5 years is getting a lot smaller.

    Also, since climatologists usually operate on 20-30 year periods, the average that is being compared is from 1989-2010 (I think– the site is down today), rather than the older 1979-2000 average. If you compare graphs from a couple of years ago to the ones prepared now, it looks like its not as bad, when it’s actually just as bad if not worse.

    Also, CO2 change lags behind other effects in natural changes, but not this time. That was a major message in that book Ice, Mud, and Blood we read a couple of years ago. (although the book was a bit badly written, and it was hard to figure out exactly what his point was for most of the book.)

  4. Greg Krehbiel Greg Krehbiel
    14 September 2013 @ 10:12 am

    John, your first sentence highlights one of the reasons I remain skeptical about AGW. Everything is odd. The models have to be adjusted all the time. Reality refuses to do what the computers tell it to do.

    There are so many factors involved — all intersecting in weird ways. And it still seems to me that people are only looking at the processes and possible effects on a very short timetable.

    They are definitely making tremendous progress. The amount of research on climate issues has pushed us way ahead of where we were when Jim Hansen did his big show for Congress. But I still don’t believe they can accurately project what’s going to happen.

    Of course — as we’ve discussed before — that doesn’t mean we should keep dumping CO2 into the atmosphere.

  5. John Krehbiel John Krehbiel
    14 September 2013 @ 10:30 am

    Greg, that “criticism” of the models– that they are changed to better reflect what really happens, is part of the systemic misunderstanding of how science is conducted.

    You make a model, you see if it works. It probably doesn’t work very well, so you try to figure out why. You make a change that you think will help, and try again.

    The trouble is, this scientific way of thinking runs completely counter to the politicians way of thinking, where changing your mind is a sign of weakness, and people are praised (as Steven Colbert satirically said of Dubya) when “What this man believes on Monday is the same thing he believes on Wednesday, no matter what happens Tuesday.”

  6. Greg Krehbiel Greg Krehbiel
    14 September 2013 @ 10:34 am

    Of course that’s how you improve models. But don’t jigger them to reflect what happened yesterday and then tell me I’m supposed to believe what they say about tomorrow. It doesn’t make any sense at all — especially given how chaotic the factors are. It’s like “models” for the stock market.

    You like to say that “climate” only deals with short increments of time. I’m not sure you’re right about that — see, for example, this page — but let’s say, for the sake of argument, that it’s true.

    Okay, let’s let the models run for 20 years and see how well they do.

  7. John Krehbiel John Krehbiel
    14 September 2013 @ 11:07 am

    How would you test a model in terms of predicting the future, except to see how well it “predicts” things that have already happened?

    For instance, if I had a model that supposedly predicts lottery winners, I would test it by seeing what I would have predicted last month, to see if it came up with correct numbers. IOW, do the inputs from last month correctly predict what actually happened.

    On standing in a scientific question: one of the things that always happens (in my experience) is exemplified by this:
    A customer at a grocery store wants to buy a roast, cook it at home, and bring it back to the store so that the meatcutter can slice it for her. Without asking if this can be done, she selects her roast, pays for it, and then asks the kid gathering carts in the parking lot if they will do that for her. He tells her that they will.

    Except that they can’t. Health department rules won’t allow store equipment to be used on something cooked in someone’s home (for obvious sanitation reasons).

    Or someone asks (often even in my presence) how my name is pronounced. Before I can reply, someone chimes in with Cry-bull, or Kree-bull, or Kreh- beel, or somne other completely wrong answer.

    Or a student asks me a question, and before I can answer another student pipes up with a completely wrong answer.

    There are correct answers to questions, and only some people know them. If the questions are scientific, then only scientists in the specific field are qualified to answer them.

    A friend of mine used to say “opinions are like belly-buttons; everybody has one.”

    My reply was “No, opinions are like bank accounts; most people have them, but they are not all of equal value.”

  8. Greg Krehbiel Greg Krehbiel
    14 September 2013 @ 12:33 pm

    You’ve illustrated exactly why I don’t trust the so-called consensus on AGW.

    You’re right that you have to ask people who know the subject. If you have a legal question, ask a lawyer.

    But there are questions that by their very nature require a multi-disciplinary approach. There is no single discipline that can answer the AGW question.

    You have to know about ice ages and sun spots and physics and weather and fluid dynamics and all kinds of things.

    As soon as somebody says “we can only consult this group of specialists,” my BS meter screams at me. That is very likely to result in a narrow, insular perspective on the issue. (Which is, of course, exactly what we see with AGW.)

    I was interacting with an AGW advocate the other day who didn’t know that we are currently in an ice age, and that the normal state for the planet is to have no ice at the poles. He also didn’t know that there have been times in the past when CO2 levels were not just higher than today, but many multiples of what they are now.

    That’s the kind of insular, narrow focus you get when only people from a certain discipline look at the question. “Oh, you mean sun spots might have an effect? We didn’t know about that.”

    AGW is an incredibly complicated issue and requires input from specialists in many different fields. Saying “you can only ask climatologists” is, IMO, to fundamentally misunderstand the question.

  9. Dave Krehbiel Dave Krehbiel
    14 September 2013 @ 6:12 pm

    the rigor of a science is its ability to predict

  10. John Krehbiel John Krehbiel
    15 September 2013 @ 9:15 am

    Right. And a study of past predictions shows that the models are remarkably accurate.

    The study itself is in Nature Geoscience, and is behind a pay-wall. The thing is, it’s not that current temperatures are lower than predicted. They are in fact very very close to what was predicted.

    1997 was a very hot year. The analysis shows that the only anomaly is that 1997 was so hot, not that warming has slowed.

  11. Greg Krehbiel Greg Krehbiel
    15 September 2013 @ 9:20 am

    Okay, that’s one thing to consider. Then we also have to consider all the stories about top AGW scientists being bewildered and embarrassed at the lack of significant warming in the past 15 years.

  12. John Krehbiel John Krehbiel
    15 September 2013 @ 9:47 am

    They’re not “bewildered.” 1997 was a very hot year. When you start with an outlier, you get weird results. And the explanation for 1997 appears to be the long term buildup of heat in the oceans, periodically released by huge El Ninos like the record breaker that year.

  13. Greg Krehbiel Greg Krehbiel
    15 September 2013 @ 11:50 am

    John, the story just few couple months ago was “oh no, global warming is worse than we thought” — from some quarters, anyway. Then other people were saying “there’s been no significant warming for 15 years.” Then other people are saying “gee, the models are perfect.”

    The most likely explanation is what Judith Curry said in a leaked email — “the science is clearly not settled, and is in a state of flux.”

    It is definitely true that some of the leading scientists have been bewildered and embarrassed by the lack of predicted warming. I don’t know how you’ve missed such stories, unless you only read the ones that confirm your bias.

    I can’t read German so I can’t read the original interview, but Hans von Storch said …

    We face a puzzling problem. Only 2% of the models foresaw this [the stagnation in temperature], and if nothing changes in the next 5 to 6 years, research will be faced with a serious problem as not a single model projected that.

    So “the models are right” depends entirely on who you ask.

    Saying that the case for AGW is falling apart would be going too far, but it is definitely in a bad way right now.

  14. admin GregK
    15 September 2013 @ 2:04 pm

    This is worth reading.

  15. John Krehbiel John Krehbiel
    16 September 2013 @ 4:28 pm

    1. Scientific knowledge is always in a state of flux. That has absolutely nothing to do with confidence in the predictions.

    2. Again, if you start with an outlier, you wind up with goofy statistics. The 1997 El Nino shot temperatures that year way above the overall curve, and then temperature changes returned to the predicted curve.

    3. the models shouldn’t be used to predict temperatures in particular years, or even for periods of a couple of years.

    As I said when this first came up, there is a sub-pattern that may bear investigation. Every once in a while, temperatures shoot up, then remain more or less stable, then shoot up again. As I said elsewhere, this may have to do with oceans holding heat, then releasing it in a big El Nino type event.

    But science is all about solving puzzles. To say that something is a “puzzling problem” may really mean “this is a really cool question I’d love to work on.”

  16. Greg Krehbiel Greg Krehbiel
    17 September 2013 @ 8:57 am

    John, it seems to me that you are always moving the goal posts.

    “The science is settled,” but it’s always in a state of flux.

    The very context of the “flux” remark was the fact that the predictions were wrong, but you say that has nothing to do with confidence in predictions.

    “Global warming is worse than we thought,” but the models are amazingly accurate. Oh, and we’re always changing them anyway.

    I really don’t understand your position on this.