This is just one of dozens of responses to common climate change denial arguments, which can all be found at How to Talk to a Climate Sceptic.
Objection:
Sea ice at the north pole recovered a whopping 9.4% from 2007 to 2008 despite the doom and gloom predictions of the alarmists. Yet another wheel falls off the global warming bandwagon.
Answer:
It is true that the minimum summer ice extent in the arctic ocean in 2008 was 9.4% higher than the minimun in 2007. But calling this a recovery is simply not justifiable, not even by a long shot. Firstly, at 4.52 million square kilometers, this measurement is 2.24 million square kilometers below the average minimum observed since 1979 when accurate satellite observations began, so we are nowhere near getting back to normal levels of ice cover. (Source: NSIDC) Secondly, year to year flucuations are very large and simply reflect the chaotic nature of weather, the change over a single year does not say anything meaningful about climate trends.
So what is the trend?
(image courtesy of NSIDC)
There are several things worth noting that we can see from this image. Not only is 2008 below the average as noted above, it fell well below the downward trend line, the fourth year in a row to do so. So hardly showing sings of recovery, this year is consistent with an accelerating ice loss. We can also see that an almost 10% jump from one year to the next is not unprecedented. The jump up from 1995 to 1996 was even larger, nearly twice so. The differences between 1989-1990, 1994-1995, 2001-2002, and 2006-2007 all were larger as well than the difference between 2007 and 2008.
We can also see that 2007 was really an exceptional record setter and aside from that year, 2008 is lower than any other. This is hardly the "warming is over" news the climate denialist organisations and websites have been proclaiming or at least implying.
So that's what has happened, but what were the "alarmist" expectations? It is true that a small number of media reports quoted people saying there might be another record this year, maybe even total ice loss (can anyone show me an actual prediction?), it makes a nice sensational headline after all, but if you look to climate science you see a predictive failure in precisely the opposite direction. No research papers from scientists in the field have been predicting ice loss at the rate it has been happening. Checking in the IPCC report from 2007, in the Summary for Policymakers[PDF] we find this on page 15:
"In some projections, arctic late-summer sea ice disappears almost entirely by the latter part of the 21st century."This is nowhere near a prediction of total summer ice loss by 2008 and it is looking more and more like it will be very wrong by being too conservative. This is not the hallmark of an "alarmist"!
So, I guess the reality is that 2008's summer arctic ice extent observation is not a wheel off the GW bandwagon, it is one more nail in the coffin of denialism.
This is just one of dozens of responses to common climate change denial arguments, which can all be found at How to Talk to a Climate Sceptic.
"Summer ice in the arctic has recovered" is also posted on the Grist website, where additional comments can be found, though the author, Coby Beck, does not monitor or respond there.
Environment
Politics
Comments
That graph is its own worst enemy. Since joining the dots from year to year is meaningless, the dots should be left alone, with the regression line running through them.
Is the raw data anywhere on the NSIDC site? I looked a few days ago, and gave up.
Posted by: derek | September 19, 2008 11:56 AM
Obviously you should put cubic splines through the points!
http://scienceblogs.com/deltoid/2008/03/remember_eg_becks_dodgy.php
Posted by: Chris Noble | September 19, 2008 2:40 PM
Glad you are still posting these. Deniers and delayers will never run out of talking points, so I'm glad someone is willing to do the heavy lifting on gathering responses to them in one place.
I came from the Gristmill post. Something you might note about Gristmill. If you want comments to only be over here, there is now a selection box that lets you disable comments. (If you apply it to an existing post, it hides all existing comments as well.)
Posted by: Gar Lipow | September 20, 2008 2:21 PM
Hi, Gar,
For a while I thought the denialism was fading (after Gore's AIT), but clearly they will not give up that easily!
Re comments: I figure I should let Gristmill readers speak their minds where they choose...I will follow those threads for the first burst of activity at least.
Thanks for the visit and feedback!
Posted by: coby | September 20, 2008 3:17 PM
The other thing is that the average thickness is greatly decreased. The remaining older ice is thick, up to 5 meters. Ice that forms this winter will be only 1 meter. The reduction of the thick ice cover is conditioning the Arctic for additional record melts.
Here is a very large NSIDC data page from last year.
http://nsidc.org/arcticseaicenews/2007.html
3/4 of the way down an animation runs. Figure 4, brown landmasses. The ice is color-coded by thickness. Months and years spin past in the lower LH corner. Compared to 1981, today much of the thicker ice is gone.
Posted by: Jay Alt | September 20, 2008 4:55 PM
One of the interesting things is to look at temperatures around the Arctic circle. Those on the Russian side (esp towards the east) are still well above freezing. Something tells me that the story is not over.
Posted by: Eli Rabett | September 20, 2008 8:21 PM
When your audience (gw deniers) wants to grab at any shred of hope, comments like 'the wheel fell off' will be all that's remembered and the actual science or lack thereof will fade very quickly. A general feeling of 'one less worry' is all that will remain.
A very small percentage of the (mostly) highly educated folks I run into at work or daily life know or care enough about the science to enable a sound judgment on any set of facts. Scary. Marx called religion the opiate of the masses. Orwell used a lottery. 21st century humans use politics scrapping over inanities rather than facing issues.
Posted by: Mike | September 22, 2008 9:50 AM
Posted by: llewelly | September 24, 2008 12:22 PM
"holds up to peer review". Good god, I must have missed that meeting. Essentially, you think people can guess how the most complex of complex systems will behave. Good luck with that.
I'm going to open an airline, with the marketing theme that none of the planes have been tested, but their construction has instead been thoroughly peer reviewed. Can I count on you guys using it? I GUARANTEE things would snap into focus double quick then.
Posted by: paul | September 24, 2008 3:16 PM
Paul, what 'testing' is to aircraft (which, by the way, is primarily done through computer models), peer review is to science. If you miss this point, you don't understand what either is.
Posted by: Brian D | September 24, 2008 3:37 PM
paul,
How do you propose we test how much methane the seabed will release and what that effect on climate will be? Please be aware that due to relativistic constraints and the size of university research facilities, your proposal must not involve duplicate planet earths and time machines.
Posted by: coby | September 24, 2008 5:32 PM
I don't think the methane content in the seabed (and many other climate variables) is testable in that sense, and so I'm being simplistic of course, some kind of modelling is required. My point is that the models are just that - models - and you can't be sure about the behaviour of a complex system (climate, software, AEROPLANE etc.) until you actually run it and see. As someone who works on one, and is constantly proved wrong by it, this point is so obvious that I find it hard to even discuss.
Sorry, I take that back, Brian D seems to be able to do this with aeroplanes. Ok, I should have been more precise - by testing, I meant actual flight time. I ask you Mr D, would you go on a plane that had not been flown yet, only thoroughly peer reviewed?
Posted by: paul | September 25, 2008 1:39 AM
Yes, modeling the climate is very difficult, but the level of confidence you can have in the model results really depends on how much detail you are after. The conclusion of warming of around 3oC at 2x CO2 is a very robust result now. CH4 is similarily complex in terms of its effect as a forcing agent, alos well understood.
To continue with your testing/aeroplane analogy: there is only one model aeroplane that has been tested and that is the one with relatively stable CO2 and CH4 at around preindustrial levels. People who are unconcerned about greenhouse gas emissions are the one's with the new and untested model, the one with rapidly rising GHG levels.
Oh, and as for choosing to fly on it or not, there is no choice, we are all aboard right now and there are no parachutes.
Posted by: coby | September 25, 2008 8:58 AM
This is just evidence that some arctic ice has melted.It does not show that CO2 is the cause.
Posted by: timwells | September 29, 2008 4:40 AM
Paul:
Coby's right. There are no emergency exits. If I were on a plane in flight and unable to contact the plane's designers or pilots (as is essentially the case with Earth), and all the aeronautics experts on the plane, who had been individually studying the plane's motion for quite a while, argued amongst themselves on the merits of their observations until a clear picture emerged... if they were saying it wasn't stable, I'd pay attention. If one greybearded retired expert who hasn't followed aeronautics for a decade and several wealthy businessmen with business to conduct at our destination, say that everything's fine, forgive me if I am less than assured.
Posted by: Brian D | September 30, 2008 9:48 AM
That's a good one, I'll have to remember that - a "clear picture". If I were on the plane in the situation you posit Mr D, I might feel the same way IF a clear picture emerged.
But when these aeronautics experts told me I had 5 mins to live one too many times, then kept insisting all their predictions had in fact come true in spite of the fact that they obviously hadn't. And insisted the plane was diving even as my drink stood motionless on my table ("this is only a temporary suspension of a diving trend")....
Hmm, I'd go back and speak to the "greybearded retired expert who hasn't followed aeronautics for a decade and several wealthy businessmen with business to conduct at our destination" and see if we could get that radio working.
And anyway, if we can return to the point, I assume the refusal of either of you to answer my question directly (not a good sign I might add) means that you accept that planes are flown extensively during testing before being released to the real world? And that this is because, no matter how much time and thought is given by engineers, and no matter how clever they are, the system is simply too complex for this. And so you agree that there are systems so complex that you can't ever be sure that your models of them are valid - and only OBSERVATIONS will do. This is obvious of course but is the level one has to sink too when faced with argument by assertion.
Maybe some good models are in place, maybe they're not. But anyone making statement like "the level of confidence you can have in the model results really depends on how much detail you are after" is not the person to ask. No, it's to do with how well they fit with, ie. can predict, observations - what you say is, charitably, a secondary factor.
Posted by: paul | September 30, 2008 1:20 PM
As is par for the course for me, I tend to get trapped in poor analogies. Through Michael Tobis, I found that Herman Daly had a better one:
To make the point more simply, if you jump out of an airplane you need a crude parachute more than an accurate altimeter.
Now, Paul, would you prefer to bicker about how you're not falling and the ground's too far off and landing isn't going to hurt anyway?
Posted by: Brian D | October 2, 2008 10:12 AM
No Brian, I would not prefer to "bicker" about hypothetical scenarios. My point about planes was a specific one about complex systems in general, not an analogy - one which you appear to accept. I certainly didn't request duelling thought experiments. No, I would much prefer to discuss the matter at hand ie. how certain we can be that predictions about climate are correct when they rely on models and peer review rather than observations.
That you link you posted hinges on one statement “Can we systematically continue to emit increasing amounts of CO2 and other greenhouse gasses into the atmosphere without eventually provoking unacceptable climate changes? Scientists will overwhelmingly agree that the answer is no. The basic science, first principles, and directions of causality are very clear. Arrhenius discovered the basics a century ago.” All the rest is irrelevant to the matter at hand – I’m interested in is whether that statement is true. He spends no time on this at all, he simply asserts that it is true. No different to Coby’s earlier statement in this thread, “The conclusion of warming of around 3oC at 2x CO2 is a very robust result now.”
Yawn. I’ve heard the assertions before – they just don’t make sense to me. Why? 1) the 30% increase in CO2 so far has resulted in a temperature increase so small as to be within the experimental error. What you say is “robust” relies on being sure about second order effects, when we’re not even clear of the first order one yet!!! 2) no matter what painful-to-read contortions you go through, the temperature has stayed constant for at least 7-8 years (maybe 10) which NO climate model predicted 3) drop between 1940 and 1970 – aerosols? The “models” say so. Ha ha ha 4) I could go on….
… but what’s the point? I came here for information – but when the arguments are just assertions, I’m learning nothing.
Posted by: paul | October 3, 2008 1:44 AM
The idea that "no models predict constant temperatures" is demonstrably false.
9 out of 55 models projected flat or cooling temperatures between 2000 and 2007 for the AR4 A1B scenario. http://www.realclimate.org/images/trends_dist.jpg
If you ran those same models again, with the same inputs, you'd get approximately the same number of models projecting flat/cooling, but they'd likely be different models. The models that had projected flat/cooling would likely show warming the second time around. That is because these outputs are single realizations, and just like reality, short periods are dominated by "natural" variablity -- primarily energy sloshing between the ocean and atmosphere. But when you average multiple outputs together, natural variability drops out, leaving only the anthropogenic signal.
As far as CO2 sensitivity is concerned, both "simple" models and full-blown AOGCM's point to about 3 degrees. The instrumental record points to about 3 degrees. The temperature change since the Last Glacial Maximum points to about 3 degrees. The temperature change immediately following explosive volcanic eruptions point to about 3 degrees. When you have multiple independent observations and experiments all pointing to the same range of values, you can have reasonable confidence that they are correct.
Posted by: cce | October 5, 2008 9:37 PM
Okay, I had not seen such data, models predicting the levelling off of the temperature are news to me. Why is this?
I was certainly interested to know more about them – but then you said "If you ran those same models again, with the same inputs, you'd get approximately the same number of models projecting flat/cooling, but they'd likely be different models.".
If this is true, then I'm confused. You're not saying that there are inherent assumptions in these 9 models that allow them to better predict temperature variability (which I would want to know more about) - you're saying that (no doubt very complex versions of) random seeding mean that you'd expect 9 out of 55 to do the same next time, just not the same 9.
So, in other words, 46 out of 55 will always predict a rise in temperature. Applying the "When you have multiple independent observations and experiments all pointing to the same range of values, you can have reasonable confidence that they are correct." principle, with 46/55 models predicting a rise, we should have reasonable confidence that a rise was likely. This is why these models were not widely publicised, and why I’ve never seen them before, because most people applied this principle and so thought there was a rise on the way.
But we know that the rise did not occur. So this reasonable confidence in the rise was unfounded - demonstrating that we did not in fact have "multiple independent observations and experiments all pointing to the same range of values" ie. I suggest that the models are fundamentally flawed. This should be stated clearly and what these fundamental flaws are should be the focus of the research and discussion. That they are not, and that instead AGW proponents simply resort to ad hoc arguments to maintain their position that we’re all going to be barbequed regardless of the actual observations, is the fundamental problem here.
So, ok, I retract my “No” models predicted it, and replace it by “almost no”, and my point stands. The models are fundamentally flawed - and your link proves this for me.
Posted by: paul | October 6, 2008 4:21 AM
As for your second point, about the 3 degree increase, let's say all the things you say are true.
So, can you explain how, after a 30% increase, there has been a rise of less than 0.6 degrees? (we know that much of this rise was before 1940 after only 20% of this CO2 had been released, so only some of it could possibly be due to human CO2).
Why not more like 1 degree? And, it is established physics (is it not) that the higher the co2 level, the more of an increase you need to get a unit rise in temperature. So really, by now we might have expected a rise of more like 2 degrees? Why has this not happened? You can go on about models all day - but observations are far more important.
Is this simplistic? Sure. But you must explain precisely WHY it is simplistic - a more complex explanation has to be earned and justified, not just asserted.
Posted by: paul | October 6, 2008 4:43 AM
Regardless, 7 or even 10 years is too short to determine climate changes. Temperatures had been increasing for about 20 years before climatologists said much about global warming.
Your point about flying in aircraft that have not undergone flight testing has me baffled. Are you arguing that computer models of the Earth's climate provide absolutely no information? Or are you suggesting that we try not reducing CO2 output, see what happens and if it causes major problems, we start at 1990 all over again? Just what are you getting at and what do you suggest should be done?
Posted by: Richard Simons | October 7, 2008 4:18 PM
Posted by: Richard Simons | October 7, 2008 4:25 PM
My brain isn’t being fooled by anything. The temperatures in the years from around 2000 to 2007 are, given the error inherent in the measuring process, statistically indistinguishable from each other – I’ve seen data sets and analysis that show a tiny fall in the gradient of the line of best fit, some that show a tiny increase (like you did), but these are too small to be of any significance. There has simply been no rise in this period. You say 10 years is too short to say anything about “climate change” – yet 20 years seems to be just fine. Maybe you could explain your certainty in the qualitative difference between these two time periods.
“Temperatures had been increasing for about 20 years before climatologists said much about global warming.”
And you’re wrong about this anyway. Hansen’s first studies were 1986-88 – the temperature had been rising for less than 10 years then.
My point about planes is I think fairly clear. No matter what design and computer modelling goes on in the design of a new aircraft (or any other complex system), these are only validated by actual flight time. And many problems are discovered and ironed out because of this. Of course, climate experiments are not possible. This doesn’t mean though that we treat the computer models as gospel – it means we should have a healthy respect for the complexity of the climate system and only treat these models as validated when they can make predictions. The levelling off of the temperatures over ten years is, I submit, not “weather”, but is something that these climate models should have predicted. They didn’t. I don’t trust them.
“Or are you suggesting that we try not reducing CO2 output, see what happens and if it causes major problems, we start at 1990 all over again? Just what are you getting at and what do you suggest should be done?”
I don’t know what “we start at 1990 all over again” means, but basically am guilty of the other things you accuse me of here. But changing the way we live, the way our society is organised, has a cost!!! We can’t just click our fingers and have all these things happen – money and time spent on a wild goose chase is money and time that is not spent elsewhere. I think there are better ways that we can spend our money and time – on REAL problems that exist NOW – in alleviating human suffering in clear and tangible ways. All these people flying around the world to “discuss” Kyoto, and the truly absurd amounts of money that lobbyists want to spend on its implemenation despite the fact it is clearly a waste of time - makes me sick, it is simply disgusting. These vast amounts of money and effort should be spent on real issues, ones where people are dying NOW – you do know this don’t you, not everyone cares about whether other people driving SUVs, some people care about what they’re going to eat that day? These people don’t have the luxury of sitting around and telling everyone else how to live their lives based on poorly understood science.
We should also make sure the research is carried out to watch of for hot OR COLD periods (cold is much worse) – and not assume we know the answer and move on (we don’t). The sunspot data at the moment really is anomalous.
We should only tackle problems that will exist in the future (GW is not a problem now – these temperatures are fine, and they’re not rising) when there is clear evidence. And if that clear evidence comes to late, then we’re out of luck. But I, personally, think the chances of catastrophic temperature changes leading to widespread death and famine are miniscule.
And, for the record, I think we should implement policies to cut carbon emissions into the atmosphere and should gradually lower our dependence on fossil fuels and seek alternative sources of energy, because these are good things to do. But this has nothing whatsoever to do with the question here, which is about whether human generated CO2 causing global warming.
Posted by: paul | October 8, 2008 1:55 AM
"Inertia. It takes a lot of energy to melt ice or to raise the temperature of oceans. No doubt someone more proficient than I at basic physics could give you the calculations."
Yes, yes, but as I said before on this post
http://scienceblogs.com/illconsidered/2006/06/observations-show-climate-sensitivity.php#comment-1101763
this would carry more weight if it was actually asked as a scientist would, who would also want to consider past effects which are only now, due to intertia, manifesting themselves. Those who consider only one side are not scientists, but sales people.
And, incidentally, in that post, Coby claimed I had misrepresented the IPCC graph. I clearly showed that I had not. He also suggested i looked elsewhere for the data - despite any number of answers to questions on this site being "go to the IPCC reports".
So, ok, what past effects are now effecting the temperature, and how are you incorporating this information?
Posted by: paul | October 8, 2008 2:07 AM
Paul your comments are very concise and accurate.I do,however,have to suggest that a warmer climate and higher CO2 levels will be wholly beneficial to the earth and it's population.If anything we should be increasing our output of CO2 to enhance crop yields and to green-up arid areas.Higher temperatures will also lead to far fewer deaths due to cold.
Posted by: timwells | October 9, 2008 3:40 AM
Tim - actually I agree with what you say, I could have phrased what I said better.
Posted by: paul | October 9, 2008 5:25 AM
Paul,
The model runs that I linked are those in the IPCC's fourth assessment report (AR4) describing one of the emissions scenarios (with hindcast of historical conditions). They are the most "publicized" models that exist. What skeptics don't understand is that the IPCC projections are not individual model runs, but the "ensemble mean," or an average of all the runs from multiple models. This eliminates (simulated) "natural variability." What is left is the anthropogenic signal, which is currently about 0.2 degrees per decade. The 9 model runs that showed flat or cooling temperatures are balanced by runs showing significantly more than 0.2 degrees per decade over similar short time periods. According to these models, 7 years of flat or cooling temperatures (9 out of 55 or 16%) is about as likely as flipping a coin 2 or 3 times and getting heads. It's going to happen.
The fact of the matter is that if computer models didn't "predict" short periods of cooling then we'd know that they were wrong. Natural variability dominates over short time periods and everyone knows this.
http://www.realclimate.org/index.php/archives/2008/05/what-the-ipcc-models-really-say/
Re: 3 degrees
The paleo climate data is independent of the models. The temperature increase of the historical record is independent of the models. The temperature drop after volcanic eruptions is independent of the models. You get a similar temperature rise with "simple" mathematical models, or full blown AOGCMs. All of these things point to a 3 degree rise in temperature for doubled CO2.
The reason that we haven't observed all of the surface warming that we are committed to is due to two things. 1) The thermal inertia of the ocean and 2) anthropogenic aerosols ("global dimming.")
And "all sides" HAVE been considered, and they have been rejected over the past 30+ years. The alternatives do not explain observations. AGW does.
Posted by: cce | October 15, 2008 11:49 PM
CCE - Ok. There is a fundamental problem with what you said.
I understand the concept and reasons for multiple model runs and ensemble means. What you say though ie.
"This eliminates (simulated) "natural variability." What is left is the anthropogenic signal"
is a massive leap. It does not just follow as a matter of course. Far from it. It is only roughly true if the models are, firstly, at least roughly correct, and secondly if a whole other host of assumptions (about the distribution of the errors) roughly hold.
And whether these two propositions are true is what I have doubts about. The observations of the last 7-10 years suggest to me that the models are not doing a very good job at predicting. And your explanation of why this is so is torturous beyond belief - you are admitting, if we assume the models are correct, that 7-10 years of level temperatures was unlikely - a probability of say around 15%, for the sake of argument.
So you are hanging your hat on a theory which predicted that the levelling off would have arrived by chance with only 15% probability. And yet at NO POINT do you suggest that there is any need for you to reevaluate any part of your theory - your argument is essentially nothing more than "these temperatures are not likely, but unless you show they are impossible or vanishingly unlikely, I will assume the models are correct."
That is not science, that is marketing, and I defy you to say otherwise.
Posted by: paul | October 17, 2008 1:46 AM
And as I said above, regarding the inertia in the oceans, how are you factoring in heat released from events many years ago into your current models.
[it is not really a question of "heat released" from previous events, it is where the heat goes. The models incorporate this by the physical properties of water and radiation. The heat grows in the system as soon as an imbalance is imposed (a forcing) but it is by far mostly going into oceans which absorb a lot before showing much temperature change - coby]
If the temperature increase of 0.2-0.3 or whatever over 20 years is enough to store heat to rise the temperature at some point in the future like you say, what about the heat stored from the similar but even stronger increase in the first half of this century? When was this released and where are you subtracting the consequent increase in temperatures from the measured temperatures when isolating the anthropogenic signal?
And finally, maybe I will be accused of being simplistic, as I hae before. I read the link you posted on RealClimate, which essentially says the same thing you say, and I point out that if "Gavin" is the best you've got to offer, you're in trouble.
[Paul, "Gavin" is an expert in the field working for NASA, the people who put robots on Mars. There is no one "better to offer"]
He bandies around plenty of technical terms and clearly knows a lot of stuff. But I did the identical undergraduate degree as him and did a phd in data modelling (in biomedical engineering) and i've seen his type before. He can't speak clearly in simple laymans terms NOT because what he is describing is so complex, but because he, the emperor, is not wearing any clothes.
[Your here response is not "simplistic" it is undeserved ad hominem]
Posted by: paul | October 17, 2008 2:00 AM
And so I can't be accused of dodging your points, to these "anthropogenic aerosols and the "global dimming" that you speak of. Hmmm. Seems like begging the question to me (ie. deflecting the question from one poorly understood area to another).
But I admit I have much more research to do before I would speak any more on this particular subject. But, nevertheless, convenient, isn't it, that these only put their head above the parapet when it's convenient for the AGW argument?
Do you have, for example, a link to any actual agreed upon measurements of such particles in the air for the last 50 years or so?
Posted by: paul | October 17, 2008 3:04 AM
paul, it is a 15% chance that it would have occured over these particular 7-10 years. The chance it would happen over some other 7-10 year period at least once approaches 100% as time goes on. We should expect it to happen an average of once every 7ish 7-10 year periods you might care to look at.
In other words the models do predict it, they just can not say when, nor are they designed to say when.
Posted by: coby | October 20, 2008 3:56 PM
Coby - no, that is not correct.
Forget climate science. Say I want to decide if a coin is biased before use it in an experiment. I can toss the coin 5 times. I get 5 heads.
The chance of this happening - given the null hypothesis that the coin is unbiased - is low (0.5^5). I therefore would conclude, on this data, that the coin is biased.
The fact that, if you toss the coin for a long time, you would eventually get 5 heads is true but totally irrelevant. On the data i have, the only proper conclusion is that the coin is biased.
Hmmm, in theory. But if I took a coin from my pocket right now and tossed it and got five heads, I wouldn't assume it was biased, I would assume that I was just unlucky. Why? Because I'm just assuming that all coins in my pocket are unbiased. I have no reason to believe otherwise – I’m basically, saying that, despite this run of 5 heads, I’m pretty sure that this won’t carry on.
Back to climate science. You are doing exactly that. You are seeing the result and just saying you were unlucky - because you can't even conceive that you are wrong or even that there is anything to prove. You are just taking it as read that CO2 causes global warming and any evidence to the contrary can be explained away with a wave of the hands.
I, in contrast, am interested in what the data says – whilst I am pretty sure that the model that suggests coins land heads/tails with 50/50 chance is correct, I am, erm, somewhat less sure about the IPCC climate models and so will go where the data takes me.
The models which are being discussed are from around 1996. So they have had a guess at their first ten year period – and the predictions are not good. If we assume they are correct, then what actually happened from 1996 to about now had only a (for the sake of argument) 15% chance. So, they are probably (ie. 85%) wrong.
Saying that they are not wrong because, eventually, such a 10 year flatlining would happen, assumes – like I would with the coins – that the data in the future will not do this. In other words, you are (not explicitly, but in substance) “proving” my hypothesis about the current data wrong by citing WHAT WILL HAPPEN.
This is absurd. Anyone doing such a thing either fundamentally misunderstands what statistical tests are or is simply wriggling out of a corner - eiher way, they have no business telling me anything about this problem.
If that is the best you can offer on all the words I wrote above, something is badly wrong. And if you think I’m being a little rude, which maybe I am, statements like this on your latest post make my blood boil.
“I'm not trying to claim that usually satisfies my skeptical visitors, but I don't often go to more trouble than that. I try not to bang my head too hard against any brick walls that come my way!”
What brick wall would that be, one of logic and statistics?
Posted by: paul | October 22, 2008 3:27 AM
As far as heat released by oceans goes, I don't know enough about it to argue. Maybe you're right. But I can tell a scientific argument from a sales pitch - and when people pull such arguments from the air, I don't think it is at all undair of me to request that they demonstrate that they are considering its effect on both sides of the argument.
And, yes, my comments on Gavin are ad hominem and, maybe, undeserved - I don't know him - but I have seen the tone writes in, which is sometimes a little out of place, and I object to be being given links to his posts like what says is gospel. Like I said, I just don't buy it - sorry.
Posted by: paul | October 22, 2008 3:43 AM
And to be triply clear, I'm not saying that this data is definitely enough to reject the null hypothesis (that the models are correct).
I'm just incredulous that there is not even the slightest doubt from the AGW proponents, not any admission that there may be some rethinking to be done, that maybe someone somewhere says they are now less sure than they were. I haven't seen this anywhere.
Posted by: paul | October 22, 2008 3:51 AM
"Do you have, for example, a link to any actual agreed upon measurements of such particles in the air for the last 50 years or so?"
Does anyone have this? Does anyone know where the agreed upon place is to view the raw particles-in-the-air measurements on which the Cooling Due To Dimming and Aerosols Theory is based?
Posted by: paul | October 23, 2008 9:09 AM
Hi paul, sorry for the delay.
I recommend looking at RC, here: http://www.realclimate.org/index.php/archives/2004/12/index/#Aerosols There are many substantive posts on this topic there and they generally contain links to the primary research. As a starting point, this post (http://www.realclimate.org/index.php/archives/2005/07/climate-sensitivity-and-aerosol-forcings/) seems to have some good links to source papers and data. (eg http://data.giss.nasa.gov/modelforce/F.indiv.data.txt and descriptions/methods for where that data came from that should be nearby)
As for the "agreed upon" qualifier, that may not exist to your satisfaction as it is still a poorly understood and controversial topic. I would guess the most conservative take on that issue (ie the most agreed upon material) will be found in the IPCC section on that topic - http://ipcc-wg1.ucar.edu/wg1/Report/AR4WG1_Print_Ch02.pdf [PDF] (here is the older, but HTML, TAR chapter: http://www.grida.no/climate/ipcc_tar/wg1/160.htm)
HTH.
Posted by: coby | November 5, 2008 11:25 AM
"it is still a poorly understood and controversial topic"
So, the temperature anomaly in question has been attributed to the effects of a poorly understood effect by the unvalidated models. Not sure I have much else to say on that really.
My point still stands. Gavin Schmidt said on realclimate that the models would be invalidated if temperatures changed by more than -0.2 degrees or 0.5 degrees over 10 years. A look at the recent climate record shows this is fairly/very unlikely to happen - would everyone agree?
So the models are almost unfalsifiable over a decade, and can only be falsified if a "trend" turns out not to exist over multi-decadal timescales. What a wonderful world some people live in, where their professional opinion is only falsifiable in 2020 - mine is falsifiable in the next set of smoke testing - somewhere around Tuesday.
How about a compromise? Is anyone prepared to admit that they still think AGW is a problem but they are less sure than they would be had the temperature risen by 0.2 degrees over the last decade?
Posted by: paul | November 12, 2008 8:19 AM