Yes, the Globe is Still Warming
| Fri Jul. 24, 2009 12:21 PM PDT
Jim Manzi defends George Will: "There has not been a lot of measured warming for the last ten years." But that's just not true. A fifth of a degree in the trendline is still a fifth of a degree in the trendline, even if you post a chart that uses a shorter timespan and displays the results by month instead of by year.
Now, it is true that the single year 2008 was cooler than the single year 1998. It's also true that there hasn't been much warming over the past six years. Noisy data can usually provide any answer you want if you cherry pick it well enough. But neither of these things is the same as demonstrating that there's been no warming during the past decade. There has been. About a fifth of a degree.
Advertisement
Advertisement
Comments
1/5 of a degree? By all
Submitted by MacGruber on Fri Jul. 24, 2009 12:50 PM PDT.
1/5 of a degree? By all means, sound the alarm. Surely a sign of man-made, CO2 global warming. I mean, when talking about something as vast, complex and chaotic as the world's climate, 1/5 of a degree is well above the margin of error for such things.
In other news, India told the West to shove it, the science isn't settled.
Longer period is better
Submitted by Anonymous (not verified) on Fri Jul. 24, 2009 12:57 PM PDT.
If you look at Kevin's earlier chart, a few posts back, over 100 years, there isn't much change if you measure between 1930's and '70's, for example, but the trend over time is unmistakable. Just looking at a decade is ridiculous (especially if you cherry-pick end points), unless you've got some actual scientific explanation for the variation, and there isn't any theoretical basis for why the earth would be cooling. Just the usual anti-scientific crap and wishful thinking.
Certain people
Submitted by David in NY (not verified) on Fri Jul. 24, 2009 1:01 PM PDT.
Certain people, like those in charge in India and those invested in the status quo above all here, will of course always have incentives to ignore or misunderstand evidence. Some of these comment here and write for major magazines, which do not fact-check their work.
We interrupt this important
Submitted by Anonymous (not verified) on Fri Jul. 24, 2009 2:13 PM PDT.
We interrupt this important subject matter with a news alert. The TOTUS has just issued what is short of an apology to Sgt. Crowley for his "unfortunate choice" of words regarding Henry Lewis Gates arrest.
We now understand why he reads directly from the teleprompter. This important message is being brought to you in the name of fairness because we know Kevin will never blog about it.
You may now your climate change food fight.
Kevin is doing the same thing here.
Submitted by BigTom (not verified) on Fri Jul. 24, 2009 2:36 PM PDT.
Just grabbing a couple of points and drawing a line through is just not right. Only a statistical regression makes any real mathematical sense. Now maybe Kevin is just pointing out the cherry-picking foolishness (I can pick some points and get a different result), but clearly proper procedure is important in these matters. And .2 degrees in five years would be pretty alarming I think the longer term trend is somewhere in the .1 to .2 per decade range.
Still Warming?
Submitted by Trashhauler (not verified) on Fri Jul. 24, 2009 3:04 PM PDT.
Can someone tell me the correct mean temperature of the Earth? How far away from that correct temperature are we?
BigTom: My point all along
Submitted by Kevin Drum on Fri Jul. 24, 2009 3:43 PM PDT.
BigTom: My point all along has been that the long-term trendline is what's important, not any particular year or even any particular decade. However, I also wanted to point out that even if you *do* restrict yourself to just the past decade, as Manzi did, global temps have still risen. The trendline really hasn't changed much for over 30 years -- though it's likely to pick up speed in the future as CO2 concentrations rise and feedback loops really start to kick in.
Manzi also has a larger point in his post that I didn't address. I might do so later.
Manzi points are well addressed
Submitted by Jeff S. on Fri Jul. 24, 2009 8:44 PM PDT.
in the comments. Not sure which point you consider "larger", but his statistical significance point are addressed somewhat briefly by Eric L at Jul 24, 07:44 PM. Manzi's really breathtakingly ignorant point:
So, naturally we just go to the escrowed set of AGW models with their predictions made over the past 20 years or so, enter in all data for actual emissions, volcanic activity and other model inputs for the time from the prediction was made until today, and then run the mdoels and compare their outputs to actual temperature change in order to build a distribution of model accuracy, right? Ha ha. Needless to say, no such repository exists.
Almost all humans resist management and audit, and climate modelers are no exception. Because they have been so poorly managed, we have no well-structured program to evaluate accuracy, and instead must rely only on back-testing (or what among climate modelers is termed “hindcasting”).
is, well, breathtakingly ignorant. Climate scientist, like Jim Hansen, and others have been putting predictions in "repositories" for some time now. These "repositories" include scientific papers, testimony before congress, IPCC reports, etc. Many of these "repositories" are available on this thing called "the Internets" which you can search with "The Google" . Perhaps Jim has heard about it.
Anyway, Eric L at Jul 24, 07:44 PM has a link to some of these "repositories".
I'd be careful about calling
Submitted by Avatar (not verified) on Fri Jul. 24, 2009 9:29 PM PDT.
I'd be careful about calling a response "breathtakingly ignorant" and then proceeding to miss the point.
We know that the predictions have been placed in the public record. What the original commenter was interested in are the actual models behind those predictions.
Illustrative example - say you had done a bunch of global warming analysis in 2000 and released the predictions based on your models. In 2020, someone takes your computer code, looks at what inputs you were modeling up until 2000, and then takes the same inputs and provides data for 2001, 2002, 2003, etc., all the way to 2010. This is one good way to tell how well your model is working - if it's an accurate model, then providing it with the next 10 years' worth of data should mean that its prediction for 2020 would become more refined - i.e. it's likely to move closer to actual conditions.
On the other hand, when you're modeling chaotic systems, it's far from a sure thing that your model is actually any good. In many of those cases, giving it additional data will change the prediction in ways that take it further from observed reality. Why is this? Because a lot of factors aren't actually well-understood... you know, little things like cloud formation and albedo effects on the climate... they're mathematically modeled instead. This means that values get plugged in as approximations where a true model would actually model the phenomenon. The values that get plugged in are the ones that let the model match observed reality - i.e. that make the model accurately predict the past. That's why it's pointless to run this kind of test the day the model is released - it was engineered to match past data. But if the way the model is built is flawed, then it may react differently to "new" observed reality than reality actually did.
It's one of the ways you can "disprove" a model... "It didn't predict properly!" And that's one of the reasons it's so bloody hard to find the necessary information.
It's not necessarily a sign of mendacity on the part of climate scientists. Nobody wants to be judged on work they did ten years ago, when the work they're doing today is better-informed and (hopefully) more precise and accurate. But at the same time, it's precisely the sort of thing that would give you a (probably-pessimistic) estimate of the value of climate models. ;p
ExxonMobil & The Ditto-Head Cult
Submitted by SecularAnimist (not verified) on Fri Jul. 24, 2009 3:56 PM PDT.
Of course the Earth is still warming. But what this discussion ignores is the simple fact that the CURRENT levels of warming are already dangerous, as evidenced by the effect that they are already having on the Earth's climate, hydrosphere, cryosphere and biosphere. Even if all CO2 emissions and all warming magically stopped right this minute, the existing anthropogenic excess of CO2 and the existing anthropogenic warming are already at dangerous levels.
ExxonMobil has done a very smart thing, from their point of view. Not only have they manufactured a fake "controversy" about the science, by funding frauds and cranks who churn out pseudoscientific drivel. They have cleverly tied the issue of global warming into the pseudo-ideological cult of hatred known as "conservatism" in America today.
Thus, thousands of weak-minded, ignorant, mean-spirited Ditto-Heads -- stooges like MacGruber -- who would normally never give a crap one way or the other about climate science or global warming, jumped on the bandwagon of ExxonMobil's denialism and deceit as soon as Rush Limbaugh told them it had something to do with Al Gore.
What would otherwise have been an obscure issue to them, thus becomes a focal point for their raging hatred.
Ten years isn't enough time
Submitted by Neil Bates (not verified) on Fri Jul. 24, 2009 4:55 PM PDT.
Ten years isn't enough time to get a good fix on air temps, that's part of off-trend fluctuation. Any real thinker knows that. But the better indicator is ocean temps since they average lots of effects and have more thermal inertia. Aren't sea temps trending up, even over the last decade? That's what needs promoting.
trends
Submitted by Eric L (not verified) on Fri Jul. 24, 2009 5:16 PM PDT.
BigTom, (and anyone else interested)
If you want a more thorough runthrough of the numbers, I would go here. Or here.
MacGruber,
1/5 degree C per decade would be 2 degrees C per century (3.6F), which is right in line with the more modest of the IPCC projections. So this hardly invalidates the established science. And yes, over the period of 10 years, it is not statistically significant given the higher year-to-year variance of the climate, however it is in agreement with the longer term trend, which is significant.
You should look at
Submitted by Anonymous (not verified) on Fri Jul. 24, 2009 5:26 PM PDT.
You should look at tamino.wordpress.com for exquisite debunking of the cherrypicking denialist claim of short time span trend analysis. He's an actual statistician/regression analysist by trade so his posts are a great intro into the field of climate trend analysis. Just as an example of the kind of posts that he does basically addressing what you've touched on http://tamino.wordpress.com/2009/01/15/what-if/#more-1422
Kevin is right of course and
Submitted by JS (not verified) on Fri Jul. 24, 2009 10:38 PM PDT.
Kevin is right of course and Manzi is completely wrong on this.
But a reading of Will's original column shows that, at least in this case, Will is not as much to blame here as Manzi.
First - Will did not come up with the no - warming - in - the - adult - life of - a 29 - year - old statement. He just quoted from a a column by Mark Steyn. And the point of his use of the quote was not to argue the science or the statistics of AGW, but its politics: Will's column was on the actual difficulties the Obama administration is having in convincing developing countries to become alarmed about AGW, and also points out that, regardless of the science, young people are also not responding as some had hoped because they have not actually observed global warming in their adult lifetimes. (That's how the 29 year-olds came in).
Will may himself be a GW sceptic, but at least in this column he is discussing the politics, not the science. What he says does not seem unreasonable (the US is indeed having problems advancing GW policies). Will was quoted out of context by Manzi, who makes no sense at all, and Kevin seems to have read Manzi, not Will.
And no, I don't generally like Will, but there has been a journalistic distortion here.
Timescales and cherrypicking
Submitted by Tom Fuller (not verified) on Sat Jul. 25, 2009 7:14 AM PDT.
Kevin, I don't know if you're being ironic or looking for weekend traffic, but when you talk about cherrypicking data and reference a 30-year record, you will be accused of the same crime.
As mentioned yesterday, why not reference temperatures to the last 11,000 years since we escaped the most recent Ice Age? It's easy enough to do, and shows that current temperatures are:
a) fluctuating within the same band of temperatures of the past 10,000 years
b) not the highest in the chart, as Holocene, Medieval and Roman warming periods exceeded current temperatures
c) quite possibly rising at the 2.1 degrees Celsius rate per century due to increased CO2 concentrations predicted by Svante Arrhenius in 1905. He looks spot on with his prediction, made without the benefit of computer models...
What the skeptics among your readers may point out is that 2.1 degrees Celsius is non-apocalyptic. They may also note that 'warmists' were perfectly happy to cherry pick date ranges when it supported their position.
timescales
Submitted by Eric L (not verified) on Sat Jul. 25, 2009 11:21 AM PDT.
It isn't cherry picking. 20 years, 40 years, 100 years, all get the same result. If you want to look at time scales longer than 100 years, I have to ask, what question are you trying to answer? Because you certainly aren't trying to examine the impact of anthropogenic CO2 on the climate. Perhaps you are trying to examine orbital effects, but those are well understood and aren't at play in current warming. Paleoclimate studies are useful and ongoing, but the way you analyze it isn't by computing the mean and standard deviation of 100,000 years of temp. We don't see ice ages one year followed by palm trees in Maine the next, but that is the statistical model your are fitting Earth's climate to here. It is possible to do much better than that; we may not know absolutely everything about climate but we know much much more than that.
On b, this is true for some isolated regional climate data sets, but not most of them. The preponderance of data suggest this is false. Anyway, I'm not sure what your point is, it is not now and has never been a tenent of climate science that nothing other than human generated CO2 can change temperature.
On c, this is consistent with the low end of IPCC projections. Is this apocalyptic? Depends on all the regional changes that go along with that -- where we see desertification vs more rain, where mountain glaciers that are important water supplies dissapear, whether Greenland and the West Antarctic Ice Sheet see net melting and how quickly, and so on. On the last one, for example, it has become clear in recent years that Greenland is beginning to contribute to sea level rise, which IPCC projections have assumed would not begin to happen until the end of the century. As for whether c is correct, that depends on the various feedback effects in the climate. A look at the paleoclimate data does not suggest a climate more stable than scientists have been assuming.
global vs. US warming
Submitted by EB (not verified) on Sat Jul. 25, 2009 8:19 AM PDT.
The graph that Kevin shows is for global temperatures. When we are talking about US politics, and personal experience of American population with warming temps, would it make more sense to talk about average temperatures in the US over last 10, 20, 100 years? My understanding is that US hasn't experienced much cumulative warming in the past century, with 1930s being at least as warm as the 1990s, 1970s quite a bit cooler, and 2000s holding steady or getting cooler.
EB, You are totally correct.
Submitted by Trippp on Sat Jul. 25, 2009 5:14 PM PDT.
EB,
You are totally correct. Sitting here in MN I'm thinking - screw the idiots stupid enough to live on the coast, and screw the brown people too stupid to consume while the consuming is good. Oh yeah, and screw my kids, stupid enough to be born after me and to trust me. Suckers. I'm sitting pretty, so why should I give a damn about anyone else?
Oh, wait, I forgot to take off my Republican/Libertarian hat. Before I do, I'll just say "Problem? What problem? I've got mine and to Hell with you."
Tripp
US has coasts, you know
Submitted by EB (not verified) on Sun Jul. 26, 2009 9:50 AM PDT.
Tripp,
Surely you realize US has coasts too, and data from there makes it into US temperature graph?
EB, if you plot them on the
Submitted by Vince (not verified) on Sat Jul. 25, 2009 9:44 PM PDT.
EB, if you plot them on the same graph, you can see that the US data are actually fairly similar to the global temperatures (it's clearest if you fit a curve to each of them). There's a lot more year-to-year variability in the US data, and a higher bump in the 1930s-40s, but the trend is roughly the same. Most notably, both show a sharp rise of about .6 degrees C since the early 1970s, taking us to temperatures that are about .8 degrees warmer than they were when the data series started in the late 19th century.
warming in US in 2000s?
Submitted by EB (not verified) on Sun Jul. 26, 2009 9:48 AM PDT.
US data also shows a drop-off in the temperatures in 2000s. That's essentially the point I wanted to make - there is a dissonance in what US electorate is actually experiencing, and what it is hearing from people who want to pass cap-and-trade.
I'm not seeing a drop-off.
Submitted by Vince (not verified) on Sun Jul. 26, 2009 5:04 PM PDT.
I'm not seeing a drop-off. The 2000s are easily the hottest decade on record in the US. It's true that we haven't matched the 1998 high, but four years this decade have had temperatures close to the peaks from 1960-1997, and another four have been higher than the typical year in any prior decade. Is it the one cool year in 2008 that's causing the problem?
Not only have we not matched
Submitted by EB (not verified) on Sun Jul. 26, 2009 8:59 PM PDT.
Not only have we not matched 1998, we also haven't matched the highs from the 30s. Regarding the dropoff, 2008 is continuing the trend of 2007 in being cooler than the year before. In any case, if you are an American coming of age in 2000s, your experience is certainly not one of steadily rising temperatures. If you are a baby boomer, then your experience is of temperatures getting steadily hotter. If you are an old-timer who lived through the Great Depression, you would know that temperatures can come down as well as go up.
Post new comment
MoJo Comments: Send Us Your Feedback
We changed our spam software to better filter comments. Should you encounter any issues, please let us know.
