David Whitehouse, who is Science editor for the Global Warming Policy Foundation, has an article in the The Spectator, apparently presenting the truth about the global warming pause. It is, to be polite, very confused. For starters, it defines the pause as being a period over which the trend was no longer statistically significant.
Well, this is silly, because (as this post higlights) the intrinsic variability means that there will always be a period over which the trend will not be statistically significant (defined as the uncertainty in the trend being large enough that you can’t rule out that the trend is 0). If you give yourself the freedom to change the start point of your trend calculation, then you can always claim to be in period where warming has paused. This is essentially what Skeptical Science’s Escalator was developed to illustrate. Even though there are likely to be periods when surface warming is slow does not mean that there is not some long-term surface warming trend. Just to be clear, the long-term trend is just over 0.18oC per decade, and there are no indications that this is slowing.
What’s slightly worse, is that to supposedly illustrate his point, David Whitehouse showed two graphs of global surface temperatures; one from January 1997 to December 2014, and the other from January 2015 to May 2017. However, what he didn’t make clear is that the y-axis scale on the latter graph was not the same as that on the former graph. Hence, his presentation (as Roger Jones pointed out) hid the incline. His two graphs made it appear (as was clearly his intent) that there had been little warming between January 1997 and December 2014, and that there was then a rise, and a comparable drop, between January 2015 and May 2017 (i.e., little warming overall).
However, if you look at the figure on the right (which shows both periods on the same graph) it’s clear that there has been warming, and if you use the Skeptical Science Trend Calculator, it is statistically significant (i.e., it is almost certainly not 0). Furthermore, HadCRUT4 shows less warming than all the other surface temperature datasets (it suffers from coverage bias). Berkeley Earth, GISTEMP, and HadCRUT4 with kriging all show warming of around 0.18K/decade, while NOAA suggests around 0.17K/decade. So, it’s clear that even though surface warming may have been slower than was expected over the last decade or so, there is still a long-term warming trend and there is no real indication that it is slowing overall.
This whole the pause may not be over argument is extremely disingenuous. If we continue to pump CO2 into the atmosphere, it will continue to accumulate. If the surface doesn’t warm, or warms slowly, then we will build an increasing planetary energy imbalance. As the Ocean Heat Content indicates, the system continued to accrue energy even during the period when surface warming was slower than had been expected. The only way the system can return to energy equilibrium is through surface warming. The only way that it can be suggested that we may be returning to a pause/hiatus is by changing the start point of the trend estimation.
When it was clear that 2015/2016 were likely to be record warms years, many mocked organisations like the Global Warming Policy Foundation by suggesting that they will start to promote a narrative that there had been no warming since 2015/2016. Even I thought that this was so ridiculous that not even the Global Warming Policy Foundation would be silly enough to go ahead with such a strategy. Of course, nothing really surprises me anymore, so I should probably not be surprised that they are choosing this strategy, even though it is patently nonsensical. It’s almost as if they don’t even listen to their Academic Advisors. Hold on, maybe I have that the wrong way around?
Update:
As Magma points out in this comment, Whitehouse’s article claims that
for the past decade or so, although average global ocean temperatures have slightly increased, the oceans of the northern hemisphere and indeed most of the southern hemisphere have not warmed at all.
Credit: Wang et al. (2017)
The Figure on the right is from Wang et al. (2017) and shows Ocean Heat Content trends for the different oceans, for different time periods, and for different datasets. Although there are some differences between the different datasets (which is partly what the Wang et al. paper was addressing), they all show warming for the period 1998-2012.
...and Then There's Physics 
We face a grim situation with global warming. Implementation of sensible adjustments is hampered by scientists who have decided that it makes sense to rake in the bucks by propagating misinformation and maximizing confusion in the general population with regard to global warming.
I don’t know what can be done with folks who are acting in bad faith. Name-calling won’t work. It’s like mud-wresting and it makes no sense to mud-wrestle with these folks because you both get dirty and these folks love that.
I think it’s best to go with the truth: this scientist is being paid by special interests to wrap errors, half-truths and misinformation in scientific packaging. This is really special interest propaganda presented by a person who has scientific credentials, but is not acting in good faith. Follow the money.
“confused” is indeed “polite”. I have no doubt that Whitehouse willfully misled his audience, knowing the vast majority would just accept is as ‘fact’ anyway.
“Hey you Whitehouse
ha ha, charade you are…”
-Roger Waters
The competence or honesty of Whitehouse, who has a doctoral degree in astrophysics, may be illustrated by the following statement from his latest GWPF piece:
“Yet a recently published analysis shows that for the past decade or so, although average global ocean temperatures have slightly increased, the oceans of the northern hemisphere and indeed most of the southern hemisphere have not warmed at all. Warming, the Argo buoys show, is coming from just one region of the South Pacific.”
From the likely (but unnamed) paper in question:
During the 1998–2012 period, trends of global OHC are much more consistent, varying from 0.81 to 1.0 × 10^23 J/decade (Fig. 2). It is still a question which ocean basin has sequestered more heat during the recent hiatus period than prior to that time? In the Atlantic and Indian Ocean, all datasets show a robust acceleration of OHC increase during the recent decade compared with 1983–1998. The Pacific Ocean shows a slowdown of OHC increase during the 1998–2012 period for the IAP and Ishii data (not for EN4-GR10). And the Southern Ocean experiences a slight slowdown of OHC increase in the IAP data but significant increase for EN4-GR10 and Ishii analysis. This finding indicates that, although heat accumulation is evident in the global ocean, the basinal OHC change is still uncertain among different datasets.
Wang et al. (2017), Consensuses and discrepancies of basin-scale ocean heat content changes in different ocean analyses, Climate Dynamics
https://link.springer.com/article/10.1007/s00382-017-3751-5
Magma,
Wow, so all the basins show warming over the 1998-2012, and Whitehouse has claimed that most of them did not warm at all? It’s getting harder and harder to think that Whitehouse is simply confused.
That this type of Whitehouse nonsense is in the mainstream media is worrisome.
@ATTP, two years ago you had a joint post with Roger Pielke Sr., and in the comments I dug up a post from 2007 in which he considered an annual increase of global OHC of 1×10^23 J/year as a “global warming litmus test” of IPCC projections. To his credit, Pielke replied in non-contrarian mode. (Although to be clear, basic civility while discussing scientific research is a low bar, and it would have been difficult for him to disavow his own comments, even if made eight years earlier.)
Note: I had a typo in the 2015 comment, writing 10^22 instead of 10^23 J/yr
Magma,
Thanks, here is Roger’s original post. Based on the Wang et al. paper, we’ve essentially satisfied that.
jsam,
Indeed. If you’re presenting the right narrative, they’ll publish any old rubbish.
I enjoyed this exchange between Michael Brown and Fraser Nelson.
I used to say the pause fooled a lot of really smart people. I was wrong.
They’re not confused. Not in the slightest.
Psychopathic is too severe, but it covers it.
no confusion, bad faith. Call it when you see it. Scientific gibberish for sale.
It’s difficult to reconcile Whitehouse’s claims with something like this:
And so easy for him to have checked.
BBD,
Indeed. It is possible that he has confused “the rate hasn’t increased” with “it hasn’t increased”, but if he did, he really should give back his PhD.
@Wayne – lol!
Looks to me like Whitehouse doesn’t know what statistical significance actually means. Now if his definition included sufficient statistical power and an attempt to compensate for the multiple hypothesis testin involved, then he might have a workable definition.
Dikran,
Indeed. I was a bit worried about using that term in the post, which is why I tried to define what was meant by its use, even if it isn’t how it should be used.
No problem with you article AFAICS, ATTP. Ironic that it is called climate skepticism as Whitehouse’s definition has no self-skepticism whatsoever, and that is the most important kind. He is saying we assume there is a pause, and will declare unless the data falsify that assumption, and of course if you make the period short enough, falsification becomes impossible (due to uncertainty). Just like calling a two headed coin “fair” because you only saw it come up heads four times, and equally meaningless!
Dikran,
This is a great example
Professor Curry’s gold standard ended up being gold-plated pot medal.
…The changes result in global-scale warming (global trend (70S-80N, 1979-2016) = 0.174 C/decade), ~30% larger than our previous version of the dataset (global trend, (70S-80N, 1979-2016) = 0.134C/decade). …
“For starters, it defines the pause as being a period over which the trend was no longer statistically significant.”
Seems reasonable as being a definition of a pause.
I have seen multitudes of definitions about the meaning of a pause, which seems to be an obvious thing to most people yet gets everyone upset when it contradicts or confirms their pet belief.
What definition is better than this?
“the intrinsic variability means that there will always be a period over which the trend will not be statistically significant (defined as the uncertainty in the trend being large enough that you can’t rule out that the trend is 0).”
Yes, so we can always find a pause, as defined above.
The fact that it might be meaningless does not detract from it being a pause.
If it goes long enough it can increasingly become significant.
Agreed so far?
angech,
Exactly. However, how can we always be in a period during which surface warming has paused, and yet we continue to warm?
No, not really. Given that we can always define a recent period as being a period during which surface warming has paused, it seems rather meaningless.
angech
The climate system as a whole is warming (it’s mostly ocean, remember). It’s unfortunate that Dr Whitehouse was mistaken on this point.
Decadal variability in the rate of tropospheric warming doesn’t impact the longer-term forced trend.
“Yes, so we can always find a pause, as defined above.”
In the past six months, by this definition, I have seen probably close to 200 “pauses” in the temperature rise, lasting several hours each. They seem to happen nearly every day – or sometimes more than once a day. Yet six months ago I was shoveling snow, and today I’m in shorts and I’m still too hot. How is that possible with some many regular “pauses” in temperature rise?
“If it goes long enough it can increasingly become significant.”
Yet time after time is does not go on “long enough”. If you keep drawing attention to the short, meaningless “pauses” without reference to the fact that “if it goes on long enough” isn’t happening, then you’re in denial.
https://skepticalscience.com/graphics.php?g=47
angech wrote “Seems reasonable as being a definition of a pause.”
that is because you don’t know what “statistical significant” means (see the link I gave earlier)
“What definition is better than this?”
that there is statistically significant evidence for a breakpoint (a change in the rate of warming – see the link I gave earlier).
What surprises me in reading this is that there are still those who believe that the scientific community is engaged in a civil debate based around the empirical evidence (or lack thereof) for anthropogenic global warming. The debate is over, or should be. Yet note how AGW deniers are continually moving the goal posts. First, when Hansen raised the alarm it was dismissed as a ‘doomsday myth’. This persisted for about a decade, and then suddenly the surface temperature rise was deemed ‘natural’, or within defined natural variability. Then the sun was implicated as the primary culprit, and finally by around 2013, dishonestly using 1998, an exceptionally powerful El Nino year, as a starting baseline, denier circles started increasingly arguing that there was a hiatus.
Now, as the appalling Whitehouse article shows, they are dividing the hiatus into two parts to continue mangling the truth. Once again, the goal posts are being moved around at will without any need to worry about that pesky little thing called science. That is because this is most certainly not a scientific debate – as I said, that was settled over 10 years ago – but we have one side interested in objectively assessing the scientific evidence and the other, which actually hates the science, mangling it to promote a pre-determined world view and nakedly political agenda. It’s time that the scientific community writ large made this point abundantly clear instead of trying to treat deniers as honest brokers. They aren’t. They are a despicable bunch of liars and deceivers. The sooner we make that clear, the better.
Pause? What, physics took a few days off?
I guess the ploy must be to throw out enough truthiness-sounding-quasi-science out there so that the sympathetic portion of the public buys “the Science is not settled”. As well, it could be, that there is a good deal of deliberation here, confusing the Twitterverse and Facebook-verse offering one Whitehouse, the David, to confound with the other Whitehouse, the Senator, namely climate change lecturer Sheldon Whitehouse.
But I don’t really know why David Whitehouse is being given any credence whatsoever.
First, he’s using a damn significance test, which is about as rejected as methodology as one can get. Even the mainstream textbook DeGroot and Schervish (3rd edition, 2002, Probability and Statistics, Section 8.9) describes these as having “foundational issues”. And I offer no defense of geophysicists or their statistical consultants who use these things, either. They should know better.
Second, the outcome of a significance test depends upon a choice of the appropriate level and the sample size. Where does David Whitehouse justify his choice? What loss function is he assuming? Moreover, technically speaking, because the interval is the floating term — length of series — they are completely incomparable.
Third, where is the sensitivity test David Whitehouse should do? Outcomes of significance tests are themselves random variables. This means if you bootstrap or cross-validation the inputs, you won’t get a point estimate out of the process, you’ll get a density, and the decision of where it is “significant” or not is a density, not a Yes/No.
It’s just hogwash, and, by talking about it, people are implicitly affirming these methods have merit.
angech:
Of course, science is all about being wary of mere seeming.
If the null hypothesis is ‘there is no long-term trend in GMST’, then due to the normal range of annual variability, a sample size of at least 17 years is required to detect a statistically significant trend. The WMO uses a 30-year baseline to define climatic normals.
A statistically significant warming trend of about 0.14 degrees C per decade from 1968 to 1998 has been verified. WRT to any alleged pause, therefore, the null hypothesis is not that there was no trend during that interval, but that there was no change in the long-term trend. Appropriate statistical tests fail to reject that null hypothesis for the interval 1998-2012. That is, GMST was observed to depart from the long term trend for as long as 14 years, but no statistically significant decline emerged from the noise. Indeed, the 30-year trend from 1982 to 2012 was about 0.20 C/decade; that is, the long-term accelerated despite the ‘pause’.
Even if it’s not statistically significant, however, an apparent departure from the previous long-term trend may have physical causes worth investigating.
Watts stand-in Andy May just landed on his head trying to follow Willie Soon and the brothers Connolly Up the Down Escalator:
https://vvattsupwiththat.blogspot.com/2017/06/tell-thereviewers-it-will-all-make.html
What’s really cool is that May isn’t in the fossil fuel business.- why bother burning carbon when you can make an honest dollar drilling CO2 wells ?
It might be worth reviewing the proper analysis of a coin coming up heads 4 times in a row. As many of us know, this result is unusual, but not in itself unusual enough to be “statistically significant.” Even if nothing is weird about the coin, 4 heads can be expected to occur on average 1 time in every 16 trials, or 6% of the time. Usually, the bar you have to pass is that the trial result is unusual enough that you can expect to see it less than 5% of the time in the original situation. So you can never say with a straight face to a statistician that ANY single trial of 4 throws proves that the coin is probably normal, or that it isn’t. If you try, he won’t be impressed with your savvy. It takes more data to prove it, not more confidence. If that fact upsets you, perhaps you should not be interpreting experimental data alone.
Jeff,
Indeed, his article even ends with a suggestion that it could challenge the fundamental assumptions of climate science. You could turn this around and argue that the fundamentals are so solid that the chances of an extended period of no surface warming is extremely unlikely.
robertomatthews,
Indeed. Similarly with surface temperature data; if you consider a very short period of time (decade or less) the trend could be statistically significant, or not, but noone would who understood this would regard this as suitable time interval over which assess anthropogenically-driven global warming.
@-” but we have one side interested in objectively assessing the scientific evidence and the other, which actually hates the science, mangling it to promote a pre-determined world view and nakedly political agenda. ”
And both sides convinced that it is the other side that is acting from a political agenda, not objective assessment.
If you are JC then you are generous enough to ascribe this to ‘group-think’ and ‘self-deception’ rather than malicious dishonesty and fraud.
Timely article:
David Spiegelhalter, president of Royal Statistical Society, says sloppy attitude to statistics leads to misleading claims and draws parallels to rise of fake news
Izen, I think that you misread my post.
Science has moved on. There are not 2 sides to the issue of AGW. We have the scientific community by-and-large agreeing on causes and arguing that failure to act could have serious consequences for humanity against a few deniers who claim to be interested in science but who use that as a camouflage to hide their right wing neoliberal political beliefs. Across the board virtually all scientists are focused on Working Group 4 of the IPCC e.g. solutions to AGW. A tiny number are trying to drag the debate back to Working Groups 1 (is it warming?) or 2 (are humans responsible?). Whitehouse is one of them, along with a few others on the corporate payroll or on the academic fringe. These deniers are not interested in science. They actually hate it. But they must engage in it to hold onto a scintilla of credibility. So they end up writing embarrassing screeds like the latest GWPF on a hiatus that never happened. I am writing a paper right now on how deniers focus on proxies for AGW – a few sexy topics – while ignoring volumes of other evidence for it. The aim is dismissal by association. Creationists use the same strategy to dismiss the volumes of empitical support for evolution. They feel that if they can successfully debunk a few lines of evidence for evolutionary theory, then they win the debate by default, even though their own theory lacks any evidence whatsoever.
@Jeff Harvey,
DOE’s Perry is doing this, too, when he asks for “a conversation”. To the public, that sounds like a perfectly reasonable thing. It’s not, because it immediately gives non-Science a stature they do not deserve. Similarly, I was asked to participate in a debate about climate disruption once, at a church I attended. I refused outright. Then I was asked, instead, to participate in a “discussion” about it. I again refused, but explained to the minister that the trouble was, the Science denier in question could throw out objections to Science which to the untrained audience sounded plausible, but would take 15 minutes on my part to explain, and there was no way of adjudicating that.
In the end I gave a long presentation on my own, and the denier eventually got his “equal time”. Didn’t matter. Got the church to divest from fossil fuels anyway.
follow the money. Who is supporting the folks who are peddling misleading information? Folks with impressive scientific credentials and accomplishments can be purchased if the price is right. In the marketplace of ideas, money talks.
The sad thing is knowing that these scientific guns-for-hire will be proved wrong over time, but we all pay the price for the costs of their errors.
Really? I would think that a good definition of a pause would be a period with low trend with relatively high statistical significance.
In other words, showing that there actually has been little change in temperature. That’s a pause.
On the other hand, “it looks like it’s been warming rigorously, but we can’t rule out zero trend” is most definitely not a pause; it suggests warming, not a flat trend.
@-Jeff Harvey
“Izen, I think that you misread my post.”
I think you misread mine.
Probably because it was cryptically ironic !
As a veteran of the Evolution/Creationist blog battles I am quite aware of the real asymmetry between the ‘sides’. As you observed, Creationists and deniers try to debunk a few lines of evidence and claim it dismisses the whole theory by association. But it is evident which side is backed by >97% of the science!
So the fall-back position is an attack on the epistemological basis of scientific knowledge and the accusation the whole thing is a ideological construct driven by Secular Materialism, or Global Socialism.
One common characteristic of Creationists and Climate Change contrarians is that although they dispute the science they very rarely attack it directly. The ‘Intelligent Design’ papers or the Monckton paper shows when they do the effort is notable for its inadequacy.
The David Whitehouse article is not constructed to have any traction with experts in the field. It is primarily directed at those that share the GWPF point of view. It may also hope to sway the uninformed. But as is evident from the response from the science-informed AGW mainstream it is not another ‘nail in the coffin’ of the theory.
Evolution and climate change deniers portray science as a house of cards where (to mix metaphors) if they pull on one thread the whole thing will unravel. But scientific knowledge is more like a jigsaw. The overall picture does not change because of doubts about one piece.
(Although there is a lot of the ‘sky’ still to do where most of the pieces seem to be black!)
Imputing motive to behaviour is always difficult and unwise.
Perhaps the GWPF see themselves as knights wielding the sword of truth and decency in defence of the benefits of continuity in social and economic structures. The importance of the status quo in religious and political beliefs.
Perhaps we should be as generous as Judith and avoid assuming those that attack science as a legitimate body of knowledge are acting from avaricious malice or sociopathic adherence to dogma.
@Dikran,
Yes, it is, but the concerns which Professor Spiegelhalter expresses are partly due to having to communicate with a public who ostensibly pays the scientific bills, but is scientifically and statistically illiterate. For instance, while Greenhouse Effect is really all that difficult to understand, for the purposes of public consumption, it has been tortured into simplicity. And, then, deniers seize upon mistakes introduced in the simplification to show why it can’t possibly be so. They are correct. It can’t, but the message was simplified.
Still, with deep respect for the great Dr Risk, champion of Bayesian methods, I think it’s a little unfair to zing university public relations offices when public policy on medicine in things like JAMA and in journals like those of the AGU and AMETSOC still tolerate p-hacking, and research funds are distributed in proportion to number of publications+citations.
Never mind the temperatures feel my holiday houses.
The discussion of how the global temperature is developing I think should be in the context of how the carbon dioxide concentration in the atmosphere is changing. The yearly change of carbon dioxide in the atmosphere now seems to be developing in an unpleasant way:
A couple of years ago, looking at the change during the period 1999-2012, it seemed to me that this yearly concentration change was approaching a plateau of around 2 ppm/year. This is no longer consistent with what has happened in this graph after 2012. Now it seems that there is a steady increase of the yearly concentration change.
This change now seems to have reached 3 ppm/year while it was around 1 ppm/year when the measurements started in 1959. If the yearly change of the carbon dioxide concentration will increase with the same constant rate as it looks like from the graph, the carbon dioxide concentration will reach 500 ppm before the year 2040.
Sorry Izen, my error. You hit the nail on the head completely. The point I have been making for some time now is that AGW deniers focus on only a tiny sub-set of the empirical evidence for warming; this is deliberate because they know that the data is simply overwhelming. They focus on what they consider to be the most prominent or popular topics, and like dominoes place them strategically in front of hundreds or even thousands of other dominoes, each of which represents a line of evidence for AGW. These ‘keystone dominoes’ as I refer to them are topics like Mann etc als hockey stick proxy reconstructions of historical temperatures, coral reef bleaching, Arctic ice and polar bear demographics or the alleged hiatus in surface temperatures. The aim is to debunk these prominent areas in discussions of climate change. If they can do this in the eyes of the general public, then in effect by knocking over the keystone domino all others get knocked over with no effort. It is a form of dismissal by association. Creationists do exactly the same thing. I am writing about this denier strategy right now. It underpins their attack on science.
@pehr_bjornbom,
We effectively already have reached 500 ppm. As former USDOE Secretary Stephen Chu points out, once all the atmospheric species that decompose into CO2 are accounted for, we are already at 490 ppm.
In a different forum, climate scientist Glen Peters, member of the Global Carbon Project Carbon atlas team, pointed out, one explanation for this upward tick in concentration may be that despite a leveling off of emissions, the natural land and ocean sinks which have accepted so much CO2 in the past are becoming saturated, not in capacity, which remains substantial, but in the rate with which they accept additional CO2.
@hypergeometric,
We are talking about different things. The graph in my comment shows the yearly change of the carbon dioxide concentration in the atmosphere. The carbon dioxide concentration in the atmosphere is currently around 410 ppm:
You are talking about the concentration of all greenhouse gases in the atmosphere represented as a concentration of carbon dioxide equivalents. This carbon dioxide equivalent is calculated by considering the global warming potential of each greenhouse gas compared to the global warming potential of carbon dioxide:
‘
https://www.theguardian.com/environment/2011/apr/27/co2e-global-warming-potential
The concentration of carbon dioxide equivalents in the atmosphere is, as you pointed out, currently around 490 ppm.
As far as CO2eq goes, my understanding is that much of the forcing due to the non-CO2 greenhouse gases is currently masked by aerosol forcing, so that the net anthropogenic forcing is similar to the forcing due to CO2 only. Also, these non-CO2 GHGs will eventually decay to CO2, and the aerosols will eventually precipitate, so that on moderate timescales the key forcing will be CO2.
Exactly right, our emissions need to fall to zero now and we need to figure out to start removing CO2 from atmosphere or we face a future where our species will see a population crash. There is the question about the extent to which historic carbon sinks can continue to absorb CO2 and there is also the question of tipping points such as permafrost warming, forest fires, etc where historic carbon sinks transition to carbon emitters. I think we are already seeing that starting to happen.
We are now facing the greatest technological challenge our species has ever seen and we taking baby steps. Baby steps are not going to get us where we need to be.
Warm regards
Mike
Last year there was a paper that found the grow rate for ACO2 had slowed due to, by their findings, a bloom in terrestrial uptake.
@hypergeometric, @ATTP,
There is information by NOAA on carbon dioxide and carbon dioxide equivalents on this web page:
https://www.esrl.noaa.gov/gmd/aggi/
Under the headline “In a nutshell:” is written: “In terms of CO2 equivalents, the atmosphere in 2016 contained 489 ppm, of which 403 is CO2 alone. The rest comes from other gases.”
That is illustrated in this graph:
Pehr,
Yes, I agree with that. However, at the moment I think that most of the nonCO2 forcing is balance by the aerosol forcing. However, reading the Guardian article, it may be that the nonCO2 forcing will decay much more slowly than I had realised.
Professor Chu: https://www.youtube.com/watch?v=9QUoN8unzR0
from R. T. Pierrehumbert, “Short-Lived Climate Pollution”, Annual Review of Earth and Planetary Sciences, Vol. 42:341-379 (Volume publication date May 2014)
Forgot to paste in the link for the above:
http://www.annualreviews.org/doi/full/10.1146/annurev-earth-060313-054843
ATTP,
I agree that the total forcing of non-CO2 greenhouse gases apparently is balanced by the anthropogenic aerosol forcing (although considering the uncertainty limits should justify exercising caution). This is shown in the following graph, that is Fig. 8-18 from IPCC AR5-WG1 2014:
However, note that during the last decade shown in this graph apparently there is no change in anthropogenic aerosol forcing, while the forcing from the non-CO2 gases seems to change somewhat (again caution should be exercised due to the widely spread uncertainty limits).
Pehr,
Yes, I agree that there are uncertainties. As you say, this approximate balance may not continue.
Back from Russia.
Able to communicate again with wi fi, great feeling of relief.
ATTP. “If it goes long enough it can increasingly become significant. Agreed so far? ”
You said
No, not really. Given that we can always define a recent period as being a period during which surface warming has paused, it seems rather meaningless.
Following through on this I beg you to reconsider.
A pause becomes significant with time extent, you feel that the current time is too short (recent) but I was asking if it goes long enough I.e. Longer than what you regard as recent.
In this context for instance the pause could return and include the recent pause with a prolonged drop in temperature.
I realise you do not consider this probable but, if it happened in the next 3 years and it went back to 1998 ie 22 years would this be a pause?
Or would it need to go to 2030 or 2040 for it to be significant in your opinion?
4 heads in a row has a statistical significance for all those statisticians out there.
Statistical significance varies according to many factors.
Number of possible outcomes
Number of times the test can be done.
Rigged dies or coins.
For a two outcome coin with no bias with an infinity of tosses it can be considered insignificant.
For an infinite sided dice the same side 4 times in a row in only 4 tosses would tend to prove Einstein wrong.
Hence significance and the pause.
A pause exists
” we can always define a recent period as being a period during which surface warming has paused”
The statistical significance is low
“but noone would who understood this would regard this as suitable time interval over which assess anthropogenically-driven global warming.”
But not non existent.
As it stands warmists are on a winner while the pause has gone.
I remain hopeful it will recur in the next 3 years dating back to 1998.
ATTP’s comment unfortunately for both points of view will still be totally valid re the significance.
angech,
I’ll try to explain the point. If you consider the last 30 years, surface warming has been about 0.18C/decade, which is consistent with what we would expect (physics, paleoclimate, etc). If, however, we only consider the last 10 years, then one can make a statistical argument that we can’t rule out that there’s been no warming. However, this will almost always be true, so it’s a rather meaningless test. Your argument seems to be: “what if it continues to show no warming?”. Well, what are you basing this on? A statistical test that is meaningless?
So, we have a good understanding of our climate that suggests that if we continue to pump CO2 into the atmosphere, it will continue to warm, and a rather meaningless statistical test based on a time period that is probably too short to draw any meaningful conclusions, and you want to focus on the latter, rather than the former?
If we stick with the initial time that was first presented as the beginning of a “pause” (97/98) then it is already clear that it’s still warming. The only way to claim that we might get another pause is to change the start point of the trend calculation, which is essentially continually moving the goalposts.
Angech wrote “4 heads in a row has a statistical significance for all those statisticians out there.”
Guess who didn’t read the article then, quelle surprise!
@ATTP,
Statistical intervals are not arbitrary. There needs to be an assessment of statistical power as well as effect size, and, after the fact, an assessment of predictive skill hopefully using something like a Brier score. It’s odd that people don’t insist on all this as a matter of fact as these are fundamental to Meteorology. So, it should be possible to calculate a minimum interval below which statistical power is meaningless for long term characterization. Indeed, I believe that has been done, if I recall (although I do not remember specific references), and it’s something like 30+ years.
Effect size is absolutely necessary to specify if a non-Bayesian approach is pursued. The size of an effect is important in Bayesian decision-making, as well as losses of misclassification, but because a posterior density is produced as a result of analysis, the combination with these can be deferred until the last moment. Unfortunately, effect size is also something with respect to climate disruption that there is maximal common disagreement, and, so, succumbs to a Wood Shed Problem. (Hat tip to statistician John D Cook who introduced me to the term.) This is when everyone agrees a wood shed needs to be built (specifying the loss function), but no one can agree on even its approximate shape.
hyper,
Indeed.
Pingback: 2017: A year in review | …and Then There's Physics