Over at The Conversation Andrew Glikson asks Fact check: has global warming paused? citing an old Skeptical Science favorite graph, and that’s the problem; it’s old data. He writes:
As some 90% of the global heat rise is trapped in the oceans (since 1950, more than 20×1022 joules), the ocean heat level reflects global warming more accurately than land and atmosphere warming. The heat content of the ocean has risen since about 2000 by about 4×1022 joules.
…
To summarise, claims that warming has paused over the last 16 years (1997-2012) take no account of ocean heating.
Hmmm, if “…ocean heat level reflects global warming more accurately than land and atmosphere warming…” I wonder what he and the SkS team will have to say about this graph from NOAA Pacific Marine Environment Laboratory (PMEL) using more up to date data from the ARGO buoy system?
Sure looks like a pause to me, especially after steep rises in OHC from 1997-2003. Note the highlighted period in yellow:
From PMEL at http://oceans.pmel.noaa.gov/
The plot shows the 18-year trend in 0-700 m Ocean Heat Content Anomaly (OHCA) estimated from in situ data according to Lyman et al. 2010. The error bars include uncertainties from baseline climatology, mapping method, sampling, and XBT bias correction.
Historical data are from XBTs, CTDs, moorings, and other sources. Additional displays of the upper OHCA are available in the Plots section.
As Dr. Sheldon Cooper would say: “Bazinga!“
h/t to Dr. Roger Pielke Sr. for the PMEL graph.
UPDATE: See the above graph converted to temperature anomaly in this post.
Phobos says:
“I’m not incompetent…”
It says right in the link, “Climate4You.com graph”. Sorry you couldn’t find it.
Phobos says:
March 4, 2013 at 11:55 am
That’s all you got? That was your waste of time?
All you did was complain of my use of Energy instead of temperature, which I agreed with.
Maybe, just maybe you might look at the difference between daily temp increases and the following night time drop, or is that too much work for you.
Of course anyone with a brain who see that there’s no loss of night time cooling would have to come to the conclusion Co2 could not possibly be the cause of the rise in temp, that would make you have to wonder about what your whole religion says wouldn’t it.
It’s far easier to call it amateurish, and ignore it.
Phobos says:
March 4, 2013 at 9:30 am
————————-
It’s nice that scientists are doing their best to determine feedback. As I’ve noted before, I’m certain scientists are working very hard on this, and I think that’s commendable. Unfortunately, in the world I work in, doing one’s best and five bucks buys a cup of coffee at Starbucks. It doesn’t mean you’ve got a result worth looking at.
I assume (and I hope I’m wrong, frankly) that your 50% uncertainty figure refers to the IPCC TAR’s notion of 3C sensitivity, with a range from 1.5 to 4.5? As I expect you know, AR4 gave us 2.0 to 4.5 degrees with 3.0C most likely. However, the IPCC treatment of uncertainty based on an author’s expert opinion is categorically unacceptable. In my field (engineering), certainty is not quantified by opinion, no matter how expert, but by statistical analysis on test results. I would be laughed out of the building if I made an argument in my professional capacity for accepting such subjective assessment as rigorous and acceptable, and rightly so. I can only assume that the IPCC indulges in this disgraceful charade in order to plunder the confidence the public normally extends to scientific certainty by pretending quantatative numerical certainty. At any rate, if you have some source for a climate sensitivity figure that provides an objective mathematical treatment of uncertainty, I’d be pleased to review it. As far as I’m concerned, nobody has shown that the total sum of the feedbacks is even positive.
🙂 If you want to argue policy instead of science, I’d be glad to wipe the floor with you Phobos. There is uncertainty on the scientific question of climate change, but the stupidity of policies intended to mitigate via CO2 reduction rather than adapt is much easier to demonstrate.
@Mark Benson
“I do science with numbers, not by eyeballing graphs of unknown origin.”
When you do radiative heat transfer equations with CO2 what emissivity do you use for 500 R at 1 atm?
MiCro says:
March 4, 2013 at 10:55 am
Phobos commented
Yes; again, a couple of amateurish spreadsheets are simply no comparison to detailed, peer reviewed, published science.
—————————-
Uhm, are you really arguing this? Tell me, is the truth scrawled on the back of a napkin less valuable than a polished powerpoint deck that contains a bunch of errors? Does the validity of science depend on who’s performing it? This is a stupid argument Phobos. The quality of the presentation or the credentials of the researcher have got nothing to do with the validity of the content, obviously.
BTW – I do no original science. Maybe another way to put this is, I conduct no scientific research. I’m an engineer; I apply scientific results and principles in a methodical way to obtain reliable, consistent results over predictable boundaries and to within required degrees of confidence.
D.B. Stealey says: “It says right in the link, “Climate4You.com graph”. Sorry you couldn’t find it.”
I want access to the raw data, to understand its meaning and use it to calculate trends.
Clearly you don’t have it or know where it is, and just bought some graph off some Web site. Typical.
@Mark Bofill: I figured you were an engineer — you have that ‘I work with numbers and know them better than scientists do!’ attitude.
In climate science (which is not an experimental science), uncertainties of climate sensitivity come from the range seen in ensembles of model runs. You can read all about it lots of places, such as:
Why Is Climate Sensitivity So Unpredictable?
Gerard H. Roe and Marcia B. Baker
Science 318, 629 (2007).
Uncertainty in predictions of the climate response to rising levels of greenhouse gases
D. A. Stainforth et al
Nature v433 (27 January 2005) 403
Climate Sensitivity Estimated from Temperature Reconstructions of the Last Glacial Maximum
Andreas Schmittner, et al.
Science 334, 1385 (2011);
Bart says:
“dCO2/dt = k*(T – To)
Apparently, you do not know what a derivative is.”
That’s the thing with the epsilon, right?
This fails in several ways, besides what those I outlined above. During the surface temperature haitus 1945-1975, dCO2/dt was positive and statistically different from zero.
The big problem with thinking that temperature change is causing the increase in CO2 (among many) is that the ice ages show a CO2 increase of about 100 ppmv for a warming of about 10 K:
http://en.wikipedia.org/wiki/File:Vostok_Petit_data.svg
But modern warming of 0.8 K is supposed to cause an increase in atmospheric CO2 of about 120 ppmv.
These two rates are wildly inconsistent.
Phobos says:
March 3, 2013 at 8:48 pm
BTW, I’m also not calculating DTR.
Phobos says:
March 4, 2013 at 11:31 am
Anthony says: “I’m skeptical of your claim, where do you do science with numbers?”
[lots of links back to cut & paste]
——————————————————————————————————————
But that’s not doing science, Phobos. That’s a child going to Wikipedia for their homework and hoping the teacher doesn’t notice that they haven’t actually learnt anything. If you’d ever actually done any further education, you’d know that approach would fail any first degree at any even slightly reputable university because it does nothinhg at all to show understanding of the subject – anyone can cut & paste!
“Doing science” involves UNDERSTANDING what you’re talking about, not just repating things that you’ve been told are relevent. You’ve still shown no sign at all of being capable of doing that, and we’re all still waiting……
Phobos says:
March 4, 2013 at 1:51 pm
@Mark Bofill: I figured you were an engineer — you have that ‘I work with numbers and know them better than scientists do!’ attitude.
In climate science (which is not an experimental science), uncertainties of climate sensitivity come from the range seen in ensembles of model runs. You can read all about it lots of places, such as:
Why Is Climate Sensitivity So Unpredictable?
Gerard H. Roe and Marcia B. Baker
Science 318, 629 (2007).
Uncertainty in predictions of the climate response to rising levels of greenhouse gases
D. A. Stainforth et al
Nature v433 (27 January 2005) 403
Climate Sensitivity Estimated from Temperature Reconstructions of the Last Glacial Maximum
Andreas Schmittner, et al.
Science 334, 1385 (2011);
——————————————-
Thanks for the references, I’ll look at them if they aren’t pay walled.
As far as my attitude goes, well, what can I tell you. I don’t think I know numbers better than the scientists do. I do know that my job demands of me that I absolutely never pretend to be certain of anything I know hasn’t been properly verified and validated. I’d get fired for that, the systems I work on would misbehave with possible catastrophic results, etc., and as a result, I have a very hard time taking the IPCC’s methodology seriously. It’s what I do for a living. ~shrug~
Phobos says:
March 4, 2013 at 1:51 pm
@Mark Bofill: I figured you were an engineer — you have that ‘I work with numbers and know them better than scientists do!’ attitude.
——————————————————————————————————–
You reall do like demonstrating your complete lack of understanding across a whole range of matters, don’t you?
That is not at all an engineer’s view that yoou’ve posted, it’s just another cut & paste cliche.
In engineering, an engineer has to know the numbers are right, or else the bridge falls down or the car explodes.
In science, the scientist has to know the numbers are right or else the hypothesis fails to conform to nature and, potentially, his reputation sinks with it.
In climate science (which is not “a science” of any sort but a mish-mash of scientific disciplines thrown together) the Team Climate Astrologer really doesn’t care if the numbers show nothing like the natural world because they just keep extending the timescale. A bit like those end-of-the-world cults who keep saying “well, we were wrong about the exact date this time but it’s definitely soon, just you wait and see…..”
In the army there was a well-known phrase when out on a run: “Just over the next hill”. Everyone KNEW it was bull and that the end was no-where in sight, but it gave you something to keep going for. That’s how the AGW industry works – it’s always “just over the next hill” and it always will be.
Phobos says:
March 4, 2013 at 1:51 pm
“I figured you were an engineer — you have that ‘I work with numbers and know them better than scientists do!’ attitude.”
Engineers are scientists seeking practical application of their knowledge. Scientists are guys who never left the nest, and few of them actually end up doing anything significant. Those who cannot do, teach.
One of my favorite Dilbert strips has Dilbert arguing with “Dan the Illogical Scientist.”
Dan: That system will never work. I should know, because I’m a scientist, and scientists have done many wondrous things.”
Dilbert: But, those were other scientists. Not you.
Dan: Apparently, you do not understand science.
The number one variable in predicting how a pilot will fare in combat is hours logged flying. Not the plane with the highest climb rate or maneuverability, but the guy manning the stick. So it goes with scientific disciplines. And, the guys who get all the stick time are the engineers, not the “scientists”.
I certainly know more about “numbers” than 99.9% of self-styled “scientists”. I took graduate mathematics courses in pursuit of my PhD, including group theory, functional analysis, partial differential equations, and advanced topology and differential geometry. I made a perfect score on the GRE. My papers in my field, involving rather intricate mathematics, have been published in major peer-reviewed journals with global reach.
So, avoid making an ass of yourself, and do not assume you know everything about your interlocutors.
“In climate science (which is not an experimental science), uncertainties of climate sensitivity come from the range seen in ensembles of model runs.”
GIGO. If the models do not represent reality, then the model runs under different scenarios produce spurious statistics. The models have performed so poorly that there is little doubt that they have fundamental problems with reality.
Phobos says:
March 4, 2013 at 2:20 pm
“This fails in several ways, besides what those I outlined above.”
You have not outlined any way in which it fails above.
“During the surface temperature haitus 1945-1975, dCO2/dt was positive and statistically different from zero.”
And, so was the temperature. We only have reliable CO2 measurements going back to 1958, but clearly the plot shows dCO2/dt tracking the temperature in the timeframe 1958-1975.
“These two rates are wildly inconsistent.”
1) The derived measurements from the ice ages are uncertain and unable to be verified. If the case were otherwise, then there would be no use for the stations monitoring CO2 in the modern era.
2) The relationship is not guaranteed to be static in time – the affine parameters are subject to change. We only know the relationship for certain in the modern era, and it holds very well for that time. When doing science, you should rely on your best and most up-to-date measurements, which require a minimum of justification and have been carefully validated in a closed loop fashion.
3) This is an integral relationship, so you cannot do a linear comparison as you are. You would understand this if you had studied calculus.
I see that Phobos still hasn’t answered mkelly’s comment of March 4, 2013 at 12:46 pm. No doubt Phobos is scrambling to find someone to provide him with an answer he can post, since more than arithmetic is involved.
Next, after hand-holding Phobos and showing him where the chart came from, he says:
“I want access to the raw data, to understand its meaning and use it to calculate trends.”
Dr Humlum runs Climate4You.com. He is extremely knowledgeable, he answers emails, and Phobos can ask him any questions he wants.
But Phobos doesn’t want information. He wants to endlessly move the goal posts. That chart refuted the assertion Phobos made, so now he’s trying to find a way to wiggle out of it, that’s all.
Despite his thread-bombing, Phobos is convincing no one. He is getting soundly thrashed in these comments. Almost every assertion he makes is shown to be wrong, off topic, or attempting to re-frame the discussion. That typically happens when climate alarmists try to hold up their end of the argument. They don’t have science on their side, so they work the strawmen.
Phobos says:
March 4, 2013 at 1:51 pm
…
In climate science (which is not an experimental science), uncertainties of climate sensitivity come from the range seen in ensembles of model runs. You can read all about it lots of places, such as: (emphasis added)
…
————
I’m sorry, I got distracted by the engineering discussion and missed this important point in what you said. The models. The models are “the fly in the ointment, the monkey in the wrench, and a pain in the ass” to quote Bruce Willis. Let’s talk about this for a minute.
First off, let me say that I have nothing but the deepest respect and sympathy for those brave souls trying to improve climate models. I’m no expert in computer modeling, but I know enough to know it can be dang hard to get a model right even when you completely and deterministically understand everything about the system. Even then, I’ll bet getting everything right is tough. Verification and validation are a tall order, even with simple systems; with climate modeling verification and validation looks essentially impossible. I don’t like it when people disrespect modelers, because it isn’t deserved; those guys aren’t idiots, quite the opposite. It’s just that they’re trying to do something thats darn near impossible.
But again, that’s the rub. Brilliant a job as they may do, hard as they may be working, they’re trying to do a job that’s darn near impossible. Because when modeling the climate, we don’t understand what’s going on deterministically. Even if we did, the system is far too immense to slog through stepwise in a deterministic way. Worst of all, as you correctly point out, climate science isn’t an experimental science; you can’t go out and run experiments in the real world easily to help you understand how to model, since this is the reason for turning to modeling in the first place. What I’m summarizing here is pretty obvious and I don’t think there’s much controversy or news in anything I’m saying. Yet we’re left with inadequate models which demonstrate little or no skill at … much of anything.
In a sense I think this is the Gordian Knot of climate science. Until a way is found to slice through this problem, I don’t see the science going anywhere. Core questions are unanswerable and suspected answers remain a matter of faith. Faster computers, better parameterization of clouds? Bah – I don’t buy it. Incremental advances might shove the problems back one step but they aren’t going to be solved that way. The field needs a breakthrough advance to get past this, something for climate science or modeling similar to what calculus was to mechanical physics. Unfortunately, breakthroughs don’t come for the asking.
@Mark Bofill
” I’m no expert in computer modeling, but I know enough to know it can be dang hard to get a model right even when you completely and deterministically understand everything about the system. Even then, I’ll bet getting everything right is tough.”
Ironically, I am, over 15 years supporting and developing models for electronics simulators. It was this that in part led me into studying climatology and gcm’s. CS is just modelers bias.
MiCro says:
March 4, 2013 at 6:19 pm
@Mark Bofill
” I’m no expert in computer modeling, but I know enough to know it can be dang hard to get a model right even when you completely and deterministically understand everything about the system. Even then, I’ll bet getting everything right is tough.”
Ironically, I am, over 15 years supporting and developing models for electronics simulators. It was this that in part led me into studying climatology and gcm’s. CS is just modelers bias.
————–
Wow! 🙂 That’s tres cool. Any glaring errors in what I said? I started looking at the source for one climate model once (forget which one, but it was a publicly available download) but didn’t have the free time to really get my teeth into it before my attention wandered a couple of weeks later. I’ve worked with some guys who’ve done software modeling of various systems, although no climate modelers.
@Mark Bofill,
” Wow! 🙂 That’s tres cool. Any glaring errors in what I said?”
No, thats a fair summary.
@Mark Benson,
It is a worth while exercise, but they aren’t ready to be used for policy. I wouldn’t care about any of this except for that reason.
Btw, I am published, got a library of congress number and everything, I’ll link a copy tomorrow, and explain how you have something it influenced.
Bart says:”I took graduate mathematics courses in pursuit of my PhD, including group theory, functional analysis, partial differential equations, and advanced topology and differential geometry.”
Yadda yadda — so what? Every one I know took all those classes along with me. So you’re smart — everyone I know is smart. What matters is what you do with it, and you apparently spend all day flitting around from one blog to another trying to show how smart you are. That doesn’t cut it with smart people, because they are only impressed by your knowledge and ideas and can sense B.S. immediately, and your ideas are ridiculous. That you think you’re so much smarter than everyone else, and insist on showing it, just makes them even more ridiculous.
Like I said, it’s too bad all your genius is wasted on inane blog comments. That keeps you from having to actually prove anything. You know that, even though you won’t admit it here.
Mark Bofill says: “In a sense I think this is the Gordian Knot of climate science. Until a way is found to slice through this problem, I don’t see the science going anywhere.”
And your better way to do this is what?
You don’t have one. Computer models are the only known way to estimate future climate. They’re just numerical solutions to the underlying PDEs that describe the physics, so deterministic in that sense. They have their successes and their uncertainties, and scientists have spent an enormous amount of time on developing them and verifying them. This is a good introduction to it all:
IPCC 4AR WG1: Chapter 8: Climate Models and their Evaluation
http://www.ipcc.ch/publications_and_data/ar4/wg1/en/ch8.html
Everyone — everyone — knows models could be a lot better. And lots of people are working on that. But they’re the only game in town, and the answer is sufficiently important that any guidance is worthwhile. NOT KNOWING the future isn’t any better — just waiting to see what happens could spell real trouble, and by then it could be too late.
So I don’t see the point of your complaints.
@DB Stealey: Your chart doesn’t prove *anything*. There’s no indication of what “OLR” in it even means, what wavelengths, how the numbers were obtained, by what instruments, what are the error bars, how was the data processed — nothing.
Thinking such a Web site chart proves anything just paints you as novice. Getting annoyed when someone asks you for the data to a chart you’re promoting paints you as unserious.
MiCro says:
March 4, 2013 at 6:19 pm
@Mark Bofill
” I’m no expert in computer modeling, but I know enough to know it can be dang hard to get a model right even when you completely and deterministically understand everything about the system. Even then, I’ll bet getting everything right is tough.”
You’d win that bet. It’s especially hard when the code cannot be validated in a closed loop, where a real world action is taken, and the real world reaction of the system is observed and compared to the output of the computer program. These guys learned that lesson the hard way.
Phobos says:
“Computer models are the only known way to estimate future climate.”
And every one of them has been wrong. No GCM predicted the current 17 year warming hiatus. Not one of them. They all predicted ever higher temperatures.
When you’re wrong 100% of the time; when both your premise and your conclusions are wrong, your models need to be chucked. Climate models are wrong. Deal with it, instead of constantly trying to tweak them to match reality. That has not worked.
As far as the chart I posted, Dr Humlum has forgotten more than a noob like you ever learned. At least I posted verifiable corroboration; you only posted your baseless opinion.
I get it. You don’t like that the chart showed your OLR assertion was wrong. But when you’re wrong, you’re wrong. And you were wrong. So you don’t like the chart, deal with it.
Phobos says:
March 4, 2013 at 6:42 pm
Yadda, yadda, indeed. So, basically, you don’t understand the argument, you have nothing to counter it, and so now you’re reduced to the backup plan of heckling. Meh.
Bart says:
“During the surface temperature haitus 1945-1975, dCO2/dt was positive and statistically different from zero.”
“And, so was the temperature. We only have reliable CO2 measurements going back to 1958, but clearly the plot shows dCO2/dt tracking the temperature in the timeframe 1958-1975.”
Sorry, no.
From 3/1959 – 2/1974, the OLS trend for GISS was 0.03 +/- 0.04 C/decade — statistically flat.
Over the same period, the trend in the 12-month change of atmospheric CO2 was 0.05 +/- 0.01 ppm/yr, a statistically significant increase.
Phobos says:
March 4, 2013 at 6:52 pm
“But they’re the only game in town, and the answer is sufficiently important that any guidance is worthwhile.”
In the first place, it isn’t the only game in town. And, who says you have to attend a game on that basis anyway?
“NOT KNOWING the future isn’t any better — just waiting to see what happens could spell real trouble, and by then it could be too late.”
But, you admit you don’t know the future, so this isn’t one of the options anyway. And, really, we could do without the melodrama.