There are a couple of new papers that essentially analyse the various analyses of the supposed “pause” in global warming. One, lead by James Risbey considers fluctuation[s] in surface temperature in historical context: reassessment and retrospective on the evidence. The other, lead by Stephan Lewandowsky, considers the “pause” in the context of model-observation comparisons.
The basic results are that there isn’t really any indication of a model-observation discrepancy, and that global warming didn’t “pause”, at least not in the sense of it having done anything unusual, given our understanding of how surface temperature are expected respond to increasing external forcings. I wrote quite a few posts about the “pause” and it always seemed pretty clear that it wasn’t anything particularly unusual, even if it did provide an opportunity to study variability in the temperature record.
Credit: Risbey et al. 2018
The papers also highlight a number of other issues. For example, it was never entirely clear what was meant by a “pause” in global warming. Was it an actual pause? Did it just mean that the trend was lower than the long-term trend? Was it based on the uncertainty being large enough that we couldn’t rule out that there had been no warming? Did it refer mainly to a model-observation mismatch? In fact, there wasn’t even a clear definition of what time period was being considered; papers considered quite a wide range of different start dates and durations. This is in addition to the “pause” referring only to the surface temperature datasets, and not to warming of the entire climate system.
Probably the most controversial aspect of these new papers is the suggestion that the declaration of a “pause” created additional confusion for the public and policy-system about the pace and urgency of climate change and that constant public and political pressure by climate contrarians may have caused scientists to take positions that they would not have taken in the absence of such opposition. I can see why some might be irritated by this, but I don’t see why it’s not an unreasonable thing to consider. I also think there is some truth to these suggestions.
We don’t do research only because it’s interesting, we also do it to have some kind of broader impact. There’s therefore nothing wrong with later considering the impact of some research. We also don’t do research in a vacuum; we often do what seems interesting at the time. There’s nothing wrong with this, but there’s also nothing wrong with considering what might have influenced some research direction. Even though researchers should be (mostly) free to research whatever they wish, there is also nothing wrong with thinking about how it should be presented. I never liked the term “pause” because it implied something that probably wasn’t true. It would seem important to frame research in a way that minimises the chance of it being interpreted in a way that isn’t consistent with what is actually being presented.
Of course, a lot of the research associated with the “pause” was very interesting; I wrote about quite a bit of it. It was also a good opportunity to better understand how variability influences surface temperatures. Even though the “pause” narrative did influence the public/policy discourse, I don’t think it really had all that much overall impact. The Paris agreement still aimed for quite a stringent target. Also, if it there hadn’t been a focus on the “pause”, there would probably have been something else. I don’t think that our lack of policy action is really due to scientists studying something that provided a convenient narrative for those who oppose doing anything. That doesn’t mean, however, that there isn’t any merit in reflecting on this.
Links:
John Kennedy has a post about the Risbey et al. and Lewandowsky et al. papers.
...and Then There's Physics 
“what was meant by a “pause” in global warming.”
A topic of controversy due to the fact that skeptics are supposed to use it to deny global warming from CO2.
A pause should be simple to define but no one wishes to accept common sense definitions because of the ramifications.
–
“Was it an actual pause?”
Well if someone can find a start and stop date on a set of observations and draw a flat trend line that would be a pause??
–
“Did it just mean that the trend was lower than the long-term trend?”
No.
Here at least we have the concept of a short term trend being accepted. A short term trend can be flat [AKA as a pause], down, yes down, or up at a rate lower or higher or equal to the long term trend. My answer would be a pause is a pause, a flat trend, not an upward trend lower than the long term trend.
Semantically I have seen this used by people on both sides [Lucia for example] but it is only and always referred to a slower upwards trend rate, not an actual pause which has no upwards trend at all..
Hope this helps.
” Was it based on the uncertainty being large enough that we couldn’t rule out that there had been no warming? ”
No.
The pause, if such there was, was based on up to 20 years of data which showed a flat trend at some stage. This argument conversely could equally be used to deny an upwards trend by those so inclined or blinded.
–
“Did it refer mainly to a model-observation mismatch?”
No
While models and observations must by necessity be out of kilter the pause refers to observations only.
–
“, there wasn’t even a clear definition of what time period was being considered;”
–
Hence the need for a simple pause definition.
The problem with interpretation is that a pause is dated backwards from when it occurs.
Hence if it is seen and persists the starting point of the pause [which is at the end] moves forwards. At the same time the pause can also extend back further in time as it lengthens if it has a downward continuing trend.
–
“global warming didn’t “pause”, at least not in the sense of it having done anything unusual,”
–
Short pauses occur all the time when an end point brings the trend over the preceding interval to zero [or flat]. They are a natural occurrence. Pauses of a week a month or 6 months can be found easily. They occur all through the temperature record documented from 1850 on. There are even periods of falls for years.
Was this pause unusual?
We just do not know. We can surmise, speculate,wish as much as we like but it will take 30 years minimum to actually know somewhat.
OK, so no hiatus or pause, in a post hoc sort of way.
I’m not sure how one goes about ‘splanin’ away all the scientific papers and the IPCC AR5 WG1 report that use the words ‘pause’ or ‘hiatus’ though.
I also don’t trust Risbey/Lewandowsky since their 2014 paper …
Well-estimated global surface warming in climate projections selected for ENSO phase
(they never tested for systemic selection biases, their “so called” method preferentially selects for the low hanging fruit in the CMIP5 database).
I guess we can all wait for AR6 WG1 where they say nevermind (in an Emily Litella weirding sort of way) what we said in AR5 WG1.
IMHO the more post hoc papers written about ‘hiatius/pause’ the more ammo given to deniers.
There never was any “pause”, it was a fantasy made up by someone in the media. The IPCC AR5 report discussed what they called a “hiatus” in the temperature series, and they were specific about what they were talking about.
It was Propaganda 101 all over again – some jackass with an agenda created a red-herring that successfully confused a lot of the public and resulted in a grand ol’ waste of time.
As the originator of the PAWS I feel one of my major accomplishments is being stolen from me.
Global warming stopped 16 years ago, reveals Met Office report quietly released… and here is the chart to prove it
Everert wrote “OK, so no hiatus or pause, in a post hoc sort of way.”
There never was statistically significant evidence for the existence of a pause (at least not that was robust to structural uncertainties such as the choice of dataset), that is the point. See my post here for a discussion of what this would (and would not) require.
“I also don’t trust Risbey/Lewandowsky since their 2014 paper ”
an ad-hom. how convincing.
” For example, it was never entirely clear what was meant by a “pause” in global warming.”
This is exactly the problem for me. The terms “pause” and “hiatus” both imply a change in the underlying rate of warming (i.e. the forced response of the climate system), where the thing we are actually talking about is related to unforced internal climate variability. However climate skeptics interpreted this as an indication that the forced response had changed somehow.
It is a pity that it got used as a synonym for model-observation discrepancies. Whether or nor there is a “pause” in warming in the observations is a property of the observations and is entirely independent of the models, so it only obfuscates things IMHO to conflate the two. “If the climate of a planet pauses and there is no model to predict it, is there still a pause?*” ;o)
* yes, obviously!
JCH –
I had forgotten what a work of rhetorical art that article was.
How sad it is that Judith endorsed such shit
My favorite part is when Rose got Jones to ‘admit” that models aren’t perfect.
This is in addition to the “pause” referring only to the surface temperature datasets, and not to warming of the entire climate system.
That’s always what has stuck out for me.
That “skeptics” spoke of a “pause in global warming” when they were referring to a short-term decline in a longer-term trend of increase in surface temps only, exclusive of ocean heat content which comprises the vast majority of the Earth’s thermal mass (is that the right term?), stuck me as a sign of “skepticism” and NOT skepticism.
I was very disappointed that Judith failed to be more precise in her language when she testified before Congress shout a “haitus in global warming” and declined to accept accountability for the impreciceness of her language (which, ironically, failed to be explicit about the uncertainties of climate change science).
The most accountability I’ve ever seen from “skeptics” in response to being asked about the rhetorical gamesmanship of a “pause in global warming” is the lame response of “well, they tod it first, ” pointing to an over-reliance by “mainstream” climate scientists on surface temps to describe trends in “global warming.”
Prof Jones (in 2012) said : “However, he said he was still convinced that the current decade would end up significantly warmer than the previous two.”
A testable prediction, which is looking good so far. It is a shame that a scientist can’t openly talk about the uncertainties in the science, or the limitations of what we can conclude on a finite period of observations, without being misrepresented in this way.
Prof Curry said: ” it was clear that the computer models used to predict future warming were ‘deeply flawed’. ”
An assertion, which isn’t even true (as the obs were within the spread of the models).
Prof Curry said “‘The new data confirms the existence of a pause in global warming,’”
Factually incorrect, the data at that time confirmed no such thing. I’ll have to remember that the next time Prof. Curry complains about the SkS Misinformation by Source page. That is misinformation.
Perhaps it is just that I have worked with Prof. Jones (and think he is a “good egg”) but this article to me casts him in a much better light than it does Prof. Curry, but I don’t think that is what I was supposed to conclude.
The thing about the “pause” was that it’s definition was never nailed down (perhaps somewhat deliberately).
So it’s worthwhile thinking about what would be a good definition of a “pause”.
Firstly in surface temperatures alone. We could consider, in order of increasing significance
(1) Any period where current temperature is below the historic maximum.
(2) Any period of negative slope of trend in surface temperatures
(3) Any period where the slope is below the long term projected mean slope
(4) Any period where the slope is below the 95% lower bound for the long term projected slope
(5) A long period (30 years?) for (1), (2), (3) or (4)
Of these, only (5) applied to (4) seems at all reasonable.
But even that represents a cherry pick because we are continuously monitoring the data for a “pause” – it’s not a one-off test.
Since 1950 we have ~70 years of data, so ~40 different 30 year periods to choose from. We would *expect* two of these to fall outside the 95% bounds! Additionally, we have multiple different ways of calculating the actual trend.
So perhaps a good definition of a pause might be something like
“A 30 year period where, without known cause, the trend is significantly below the long term trend expected, taking into account multiple testing across the whole dataset available”
Perhaps others could improve on this definition – paging Dikran.
Of course, the “pause” was nowhere near such a test.
A 30 year period where, without known cause, the trend is significantly below the long term trend expected, taking into account multiple testing across the whole dataset available”
Trend in what? Surface temps? If so, it wouldn’t be a pause in “global warming.”
For me the important thing is to explicitly make the distinction between the forced response of the climate system and internal climate variability. If we are talking about a change in the forced response of the climate, then I would base a definition on the existence of statistically significant evidence for a *change* in the underlying rate of warming – but AFAIK no climate skeptic every bothered to perform that test (it isn’t even that difficult, there are packages in R that work for annual data – e.g. the Chow test).
If you want to talk about internal climate variability. then that is a different matter, however then if you were using it to argue the models were flawed, it would be that the spread of the model runs is implausibly narrow (and I would agree) rather than that the ensemble mean is wrong.
Joshua, I don’t think we should be too obsessed with terminology. I think it is O.K. to refer to GMST when discussing global warming (as it is measuring how warm it is at the surface), as long as you don’t use it to make strong arguments about energy balance or the accumulation of heat in the climate system as a whole (i.e. ignore ocean heat content). Otherwise we spend too much time quibbling about exact words whilst ignoring the meaning of what people are arguing.
Joshua
I refer you to my post
I may, of course, never get around to the “secondly” part.
JCH, the Mail would just have found something else to lie about. Like this. Giving that this action was by the self-regulating body affectionately known this side of the pond as “Paul Dacre’s poodle”, they had to be really, really out of line.
dikran and vtg, I’ve never considered there to be a pause, because the only valid definitions involve a null hypothesis of no pause. That means that to demonstrate a true pause, the upper (call it 95th) percentile has to have a zero or negative slope, the central estimate a negative slope, and the 5th percentile a strong negative slope. None of the datasets come remotely close. A weaker claim, a pause in the rate of warming, or more properly a slowdown, requires the null hypothesis to be flipped to an extrapolation of the long-term rate. If the 5%-95% range encompasses the long-term trend, there is no statistically demonstrable slowdown. I’ve seen it demonstrated a number of times that if you cherry-pick a short interval to get the central trend down, the uncertainty bounds widen and still encompass the extrapolated long-term slope. And that’s before you even get to the multiple-testing problem.
I like Hanson’s better graph. The 11-year average takes out almost all the natural variability leaving straight-line warming since 1970.
http://www.columbia.edu/~mhs119/Temperature/
Dave – indeed, that would indeed be a stronger test, however the evidence for the pause doesn’t even pass the less stringent “has there been a reduction in the rate of warming” test, and I like to be charitable! ;o)
Don’t cha know, your bacon was saved – and at the very last second – by a gargantuan El Niño and a super duper spike in the PDO: a sort of stadium wave of heat. Short of those lucky happenstances, the La Niña events would have kept right on frying your butts.
dikran –
Maybe the problem is my lack of understanding, but…
I think it is O.K. to refer to GMST when discussing global warming (as it is measuring how warm it is at the surface), as long as you don’t use it to make strong arguments about energy balance or the accumulation of heat in the climate system as a whole (i.e. ignore ocean heat content).
Well.. seems to me that arguing that there has been a pause in global warming by referencing surface temps only isignoring OHC – particularly troublesome when people reference that “pause” to suggest we should be less concerned than otherwise about continuing CO2 emissions (as if the trend in OHC is not a very important consideration for assessing the risks posed by climate change).
Joshua – at the time, and I believe it was Gavin Schmidt who said it, the GMST is the metric by which global warming was defined for the public. Pielke Sr. argued for switching to using zillions of pearls, the jewels of the oceans, and Gavin sort of resisted that.
Regardless, the warmists’s position is hopeless. Abrupt climate change is coming soon. I am told it may have already started, and I can assure you, abrupt climate change hates progressive pissants and it’s going cover much of the earth in ice. Soon, if not sooner.
JCH –
I get that we live in the surface and not in the ocean, and so surface temps are a relevant measure for assessing the impact of ACO2 emissions. I wouldn’t suggest that we ignore thst metric. But I don’t understand why people in either side would use that metric as a measure of “global warming.”
Isn’t it possible that RPsr. has a point?
And I say that independent of my complete confidence that nothing progressive pissants might do will prevent the coming cooling in the next decade or three (btw, nice did you see the other day when Atomsk dug out Judith’s predictions of cooling from a few years back?)
Joshua.
I think RPsr does have a point. I think the OHC is a much stronger measure of overall climate warming than the surface temperatures. It’s less susceptible to short-term variability and provides and indication of how much we’re out of energy balance. RPs and I wrote a joint post about this a number of years ago.
However, it’s my understanding that when RPsr was first suggesting this, the OHC data was not nearly as reliable as it is today. So, it’s probably a good way to assess things now, but may not have been when RPs first suggested this. Also, there’s a difference between trying to assess how our emissions are leading to warming of the climate, and trying to asses the impact where we live (the surface).
I think R Sr’s real point, at the time, was his belief that OHC numbers were going to invalidate James Hansen and GISS Model E.
ARGO and the first papers on deep ocean warming dashed those hopes.
I think he also makes a similar argument on using energy content of the land surfaces rather surface temperatures.
Prof Curry said: ” it was clear that the computer models used to predict future warming were ‘deeply flawed’. ”
Well, it is becoming increasingly clear that this is true, even though perhaps it was not quite so clear in the past. Here’s a recent RSS analysis of their product. It’s entirely consistent with energy balance TCR estimates being much lower than those for GCM’s.
http://www.remss.com/research/climate/#Atmospheric-Temperature
This is similar to what Real Climate has been showing for a while. Especially in the tropics the discrepancy is hard to explain any other way than as model inadequacies. There is measurement error, but RSS agrees pretty well after it was adjusted with the warming rate in HADCRUT4 , which is very well vetted. UAH shows quite a bit less warming and GISS shows more. The period of comparison is 40 years, making it well outside Santer’s 17 year window. There may be forcing errors for the last few years even though one would expect them to be accurate before roughly 2010.
Of course we know that tropical convection and precipitation models omit many important phenomena such as aggregation that has nonlocal effects. Further the overall sensitivity of models is quite sensitive to differences in plausible parameters of these models. See points 22-25 of the following.
Click to access briefing-note-on-climate-sensitivity-etc_nic-lewis_mar2016.pdf
The real clincher here however is that papers are starting to appear that explain the discrepancy between models and observations. My favorite is the pattern of warming argument. Basically, GCM’s get that wrong over the instrumental period but for some reason we are supposed to believe that in the long term (unspecified in the papers), GCM’s will be right. Most people I think would say that the fact that a model that is wrong over a 50 year period makes its predictions over a 150 year period more suspect.
Everett is of course right that there are many climate science papers saying there was a pause and trying to explain it. This whole issue is probably over too short a time frame to tell us a lot and I’m not going to spend much time on the issue.
Joshua, I really doubt if anyone besides you cares about your “concerns” about Judith Curry.
dpy,
You’re ignoring that there are quite large uncertainties in the tropospheric temperature trends from satellite data.
David –
Trust me, I fully expected that you (at least) would be totally unconcerned that Judith, who champions herself as the keeper of the uncertainty monster, would ignore the relevance of uncertainties related to OHC – the major thermal mass of the planet – as she provided expert testimony (in a non-advocate form, of course) about the “pause in global warming” at the behest of Republicans in Congress.
Nor would I expect thst you’d be concerned that any other “skeptic” might do something similar – RPSr.’s concern about the inadequacy of surface temperatures as a measure of “global warming,” notwithstanding.
As for anyone else’s concerns, it’s obviously up to them. If I’m the only one concerned about impreciseness in expert testimony given to Congress, so be it.
Attp, that’s why I mentioned hadcrut, uah, and giss. Hadcrut has almost the same warming rate as RSS’s new product. Hadcrut is very well vetted.
dpy,
Huh? If you consider the surface temperature datasets, there is little discrepancy between the models and the observations.
I’ll add that you also misunderstand my ui>concerns.
I’m not concerned about Judith Curry. Judith will do what Judith does, and obviously I have no influence on that.
I do find it somewhat concerning that Judith hypocritically derides advocacy, something I consider a very important aspect of our society, and offers weak and rhetorically misleading testimony – the hallmark not of advocacy per se, bur the hallmark of problematic advocacy.
Judith’s impact, within the larger scheme of the societal wrestling with climate change, is obviously iignificantly small. But given the “post truth” presssues on how our society deals with global-scale problems, the symbolism of a respected expert failing accountability concerns me, as an example of a larger, troubling pattern. But sure, she’s far from the only one who acts like that – although it is a shame that “skeptics” won’t call her on it.
Maybe you should reconsider?
Don’t have the foggiest notion of where anybody would ever get the idea of a ” so called” pause or hiatus or slowdown … oh wait …
@dpy6629
Re: “Especially in the tropics the discrepancy is hard to explain any other way than as model inadequacies.”
Please read the RSS page you linked to. It rebuts what you said:
“In fact, there is hard scientific evidence that all four of these factors contribute to the discrepancy, and that most of it can be explained without resorting to model physics errors. For a detailed discussion of all these reasons, see the post on the Skeptical Science blog by Ben Santer and Carl Mears, and the recent paper in Nature Geoscience by Santer et al.”
http://www.remss.com/research/climate/#Atmospheric-Temperature
I’ve had contrarians repeatedly cite that page to me, and it’s like none of them actually read it. I also suggest you read the “Nature Geoscience” mentioned there, since it supports error in forcings as the primary explanation, while debunking the “model error” explanation folks like you and Christy offer. Model error is not the primary explanation for the difference. Furthermore, I suggest you not appeal to the tropics if you’re trying to argue for a lower climate sensitivity. After all, enhanced tropical tropospheric warming is a sign of the negative lapse rate feedback, which would limit climate sensitivity.
ATTP, What you link to from Real Climate has several plots in it. One needs to compare apples to apples and go beyond the first plot which is CMIP3.
The GMST CMIP5 plot has only 7 years of model forecasts as the rest is hindcast. The TLT plots near the end look to me to show something a little different that what RSS shows, but to basically confirm the general idea from RSS. Schmidt likes to show yearly averages and has some long baselining periods that are a little nonstandard. He didn’t show GMST comparisons for the tropics. In any case, bottom line RealClimate shows pretty much the same thing for TLT data with the rates of warming disagreeing a lot with the models. The discrepancy is particularly large in the tropics.
If you look carefully at global rates of warming from HADCRUT4 surface and RSS TLT the rate since 1979 is quite similar. I think that was part of the reason RSS did their recent adjustment so they were consistent with surface temperature products. UAH shows a lower rate (by a lot) and GISS shows a higher rate (which Schmidt used in his plots). I also don’t know what the source for Gavin’s “forcing adjustments” is. We’ll see what’s in AR6.
Another line of evidence is the energy balance estimates which use a much longer time period than any of the plots we are talking about here. They also show a much lower TCR than the models and that’s another line of conciliance with the shorter term trend information.
Joshua, Life is short. Selection bias causes many to waste a lot of their precious mental energy chasing things that are not significant. Pulling quotes from others which may or may not be fully representative of that person’s complete work history is such an idle pursuit.
Conclusion from your RSS link.
“We call to these four explanations “model physics errors”, “model input errors”, “observational errors”, and “different variability sequences”. They are not mutually exclusive. In fact, there is hard scientific evidence that all four of these factors contribute to the discrepancy, and that most of it can be explained without resorting to model physics errors. For a detailed discussion of all these reasons, see the post on the Skeptical Science blog by Ben Santer and Carl Mears, and the recent paper in Nature Geoscience by Santer et al.”
Me? I always expect the grist of an argument to be contained within the last paragraph .
Strangely I have often seen persons of a contrary persuasion ignore the conclusion and cherry pick details out of context .
Willard might be able to inform me if this is demonstrably true or just my own bias.
As to concern about our Judith.
She is well know to have a unidirectional monster haunting her .
The ones in my nightmares travel in both directions .
What concerns most of us is the conditions on the surface.
Not 10km up were the only life is space cadets heading into orbit.
http://www.realclimate.org/index.php/climate-model-projections-compared-to-observations/
DPY – “energy balance methods use longer time period …..” The recent paper by Adams and Dessler (2018) shows that models emulating EBM get similar TCR results. This shoudn’t be surprising because models do a reasonable job of matching the observed temperature record from the 19’th century to present. In other words, differences between EBM and climate models are mainly due to the method not the observations.
https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2018GL080714?af=R
Everett F Sargent says:
Don’t have the foggiest notion of where anybody would ever get the idea of a ” so called” pause or hiatus or slowdown … oh wait …”” The escalator”
Thanks Everett. I always forget about that chart of pauses.
Well pauses by one definition at least.
I would regard each of those, as did the creator of the escalator, as genuine pauses.
Mathematically speaking. A flat or zero trend for an extended period of time.
It is only possible with hindsight to see that each pause was of limited nature.
Very instructive for both sides of the argument.
“There is measurement error, but RSS agrees pretty well after it was adjusted with the warming rate in HADCRUT4 , which is very well vetted. UAH shows quite a bit less warming and GISS shows more.”
Not so sure about that, David: http://woodfortrees.org/plot/rss/from:1979/trend/offset:0.34/plot/gistemp/from:1979/trend/plot/hadcrut4gl/from:1979/trend/offset:0.14
(I’ve left out UAH, which indeed gives a much lower rate).
And then there’s your “Especially in the tropics the discrepancy is hard to explain any other way than as model inadequacies.” which KiwiGriff already addressed. Selection bias, much?
Joshua wrote “Well.. seems to me that arguing that there has been a pause in global warming by referencing surface temps only isignoring OHC – particularly troublesome when people reference that “pause” to suggest we should be less concerned than otherwise about continuing CO2 emissions (as if the trend in OHC is not a very important consideration for assessing the risks posed by climate change).”
My point is that you might want to talk about GMSTs and surface warming without making an argument about energy balance or planetary accumulation of heat. The point is whether you need to mention OHC depends on the argument someone is trying to make out of the apparent pause in surface warming.
Joshua wrote “Isn’t it possible that RPsr. has a point?”
Yes, he did, but he wasn’t even the first to make it.
I think we discussed the whole issue of who might have been to suggest using OHC in this comment thread.
dpy6629 “Joshua, I really doubt if anyone besides you cares about your “concerns” about Judith Curry.”
I’m not unduly surprised that you ignored the quote from Prof. Curry in my post that was factually incorrect and picked on the one that was merely unsupported by the observations at the time (and as ATTP pointed out, your analysis ignores the uncertainty in the observations). Why do you think we should care about your “concerns” when you do that?
dpy6629 “Joshua, Life is short. Selection bias causes many to waste a lot of their precious mental energy chasing things that are not significant. “
If selection bias wastes peoples time, the logical thing to do is not do it yourself, rather than repeatedly use it as an excuse not to address criticisms that are obviously valid.
Angech, if you have a noisy time series, you can always find “genuine pauses” according to your definition, but they will just be random artifacts of the noise and are entirely meaningless. For analysis of noisy data, you need statistics. Of course you know this already as that isn’t the first time I have pointed this out to you, and it is a pity that you can’t learn from your errors.
angech ” I would regard each of those, as did the creator of the escalator, as genuine pauses.”
This is completely incorrect, and demonstrates that you do not understand the message of the escalator diagram. Note also Dana wouldn’t have created the escalator diagram if “skeptics” hadn’t first been making claims of a pause.
why would anyone expect temps to go up in a straight line anyway – they certainly don’t between Jan and June, yet we know June will be warmer that Jan
to me the argument is no more sophisticated than the flat earther argument that the “earth looks flat” – indeed it does, so what
an_older_code,
You need to be slightly careful since the global surface temperatures are typically anomalies, which means that what is presented is relative to some baseline. Hence, things like June being warming than Jan is removed by this process (i.e., the June anomaly is relative to what would be expected for June, and the Jan anomaly is what we’d expect for Jan).
‘Note also Dana wouldn’t have created the escalator diagram if “skeptics” hadn’t first been making claims of a pause.”
You may be right, I have not been involved as long as you have and may have missed the starting points and discussion.
I did feel that while the pause from 1998 would predate Dana’s graph the issue of the pause did not really bite until after Dana’s graph, say 2008??.
Thanks for the explanation. Makes sense.
“It is only possible with hindsight to see that each pause was of limited nature.”is the message of the escalator diagram.
” if you have a noisy time series, you can always find “genuine pauses” according to your definition, but they will just be random artifacts of the noise and are entirely meaningless”
Unless and until a pause persists for long enough you are right, again.
dpy –
None of Gavin’s charts are apples-to-apples. His model values are surface temperature, while observations use SST. If you use SST in creating the model averages and restrict to locations where observations are made, then the model trends increase by roughly 20%. So Gavin is underselling model performance.
@Joshua
Re: “(btw, nice did you see the other day when Atomsk dug out Judith’s predictions of cooling from a few years back?)”
If people are curious, then I think this is what you’re referring to:
“The credibility of your sea level rise claims”
https://judithcurry.com/2018/11/27/special-report-on-sea-level-rise/#comment-885742 (http://archive.is/8104b)
It turns out Curry’s prediction was nonsense; warming continued, at about the rate projected by the IPCC, contrary to what Curry said. Of course, that doesn’t stop Curry from continuing in her false claims regarding a “pause”. But at this point, I think she’s beyond rational persuasion with evidence.
Atomsk –
Yah. That was it. Although I was wrong. It wasn’t a prediction of cooling, but another couple decades of no warming.
Maybe we should just follow David’s advice and ignore anything she says that doesn’t fit with his ideological preferences – no matter if it is imprecise, rhetorically jiggered, or just flat out wrong?
The level of toadyism she inspires is really quite impressive.
This was the specific quote that I found interesting:
A year earlier, Jan 2011, I made it pretty clear that I supported Tsonis’ argument regarding climate shifts and a flat temperature trend for the next few decades.
I think I can probably predict what David might have to say in 2031…
Ultimately, the stadium wave is a prediction for 21st-century cooling. Stay tuned. Right around the corner.
dikran –
My point is that you might want to talk about GMSTs and surface warming without making an argument about energy balance or planetary accumulation of heat.
No doubt. If what I wrote suggests otherwise, it wasn’t what I was intending to say.
The point is whether you need to mention OHC depends on the argument someone is trying to make out of the apparent pause in surface warming.
Again, no doubt. Again, my point is that IMO, it is highly problematic to talk about a “hiatus in global warming” without talking about the uncertainties introduced by OHC. Seems to me that OHC is necessarily a part of “global warming.”
And I think it’s problematic coming from when it comes from either side of the great climate change divide, actually. Not sure if I think there’s a distinction in the level of problem if someone’s (1) describing a “hiatus in global warming” when only referencing surface temps (not to mention, not acknowledging that the “pause” is actually, arguably, a short-term slowdown of a longer-term trend)., or (2) describing the risks of future climate change by pointing to trends in surface temps only. I’m inclined to think that 1 is more problematic, but I’m not sure if my “motivation” is kicking in when I think that way.
@Joshua
Re: “It wasn’t a prediction of cooling, but another couple decades of no warming.”
You weren’t wrong. It was both. She vacillates on this topic:
“JC note: Attention in the public debate seems to be moving away from the 15-17 yr ‘pause’ to the cooling since 2002 (note: I am receiving inquiries about this from journalists). This period since 2002 is scientifically interesting, since it coincides with the ‘climate shift’ circa 2001/2002 posited by Tsonis and others. This shift and the subsequent slight cooling trend provides a rationale for inferring a slight cooling trend over the next decade or so, rather than a flat trend from the 15 yr ‘pause’.”
http://archive.is/uczqT
Josh and Dikran, I don’t want to belabor the point. I’ll just wrap up my contribution by noting that there are 2 elements of selection bias in quote mining (and this applies to everyone who does it).
1. People tend to focus on a few statements from an often huge public opus of work. It’s much more objective and also much more charitable to try to add the context of that opus. Everyone is occasionally wrong or misspeaks or holds questionable views.
2. Why do you choose a particular individual for attention? Are they really worse than average? Do you even know their opus of work that well? If you have spent the time to read their opus of work, you have wasted many years of your life.
In modern science, we are dealing with a situation where perhaps half of papers are wrong (I would limit that to those that attempt to quantify small effects). In my field, selection bias is a given. Incorrect science adversely affects the health and well being of millions. Many press releases are also exaggerated or wrong. It’s a better use of time to focus on this broader problem because it has effected public confidence in science and public institutions generally. BTW, I exclude qualitative “explanations” from this indictment. This is an institutional problem and is not due to scientists being unusually morally culpable even though some are certainly culpable. Nic Lewis I think has shown how individuals can make contributions to improving this situation. One of Prof. Curry’s contributions has been to provide Lewis with a public forum for his work.
David –
I pretty much agree with everything you wrote in that comment…
… well, at least until you get to the peddling part (some agreement is better than none, eh?) … where for the record I think your argument has some serious flaws… but don’t want to piss Willard off…so I’ll wait to see if some day Anders writes a post about the impact to society of errors in science.
Please forgive me, but really …
You can argue the science from here to eternity, and it only leads to more of the same. Scientists think if they get it just right it will blow the house down, and spend valuable time dotting i’s and crossing t’s. But what is going to blow the house down is the actual effects of climate change/global warming. Here’s an argument nobody can deny:
https://www.theguardian.com/environment/ng-interactive/2018/dec/21/deadly-weather-the-human-cost-of-2018s-climate-disasters-visual-guide
Graphics action is a mite annoying, but that’s what ordinary people are experiencing.
For continuing information on what is happening worldwide, this:
https://www.wunderground.com/cat6/Earth-Had-Its-Fifth-Warmest-November-2018-Lock-4th-Warmest-Year-Record
Wunderground provides a range of regularly updated lists. This one includes these headers:
Two billion-dollar weather disasters in November 2018 (37 year to date)
On the verge of an El Niño event
Arctic sea ice: ninth lowest November extent on record (heading lower as winter goes on)
Notable global heat and cold marks set for November 2018
Major weather stations that set (not tied) new all-time heat or cold records in November 2018
Six all-time national heat records tied or broken in 2018
Fifty-five monthly national/territorial heat records so far in 2018
One national monthly cold record set so far in 2018
Continental/hemispheric records in 2018
In addition, the comment section includes reports from all over the world; also, Cyclone Cilida is currently churning east of Africa with sustained winds of 150 mph.
Kiwi and Atomski, I mentioned all the sources of error mentioned by RSS and gave reasons why I though some of them were small. If you want to give your reasoning, I’d be interested in why you think I’m wrong. I’ll wait until AR6 for credible updated forcings taking into account all the published work. I particularly think measurement error is not a likely explanation given Marco’s point.
Chubbs, Nic Lewis already showed that Dessler’s EBM statements are just wrong. Dessler’s internal variability in his paper is actually a little less than what Lewis used in his EBM calculations. It’s a recent post at Climate Etc if you are interested.
Marco, if as you show RSS now has a higher rate of warming than GISS or HADCRUT it makes the divergence more concerning, not less. It seems to me also to pretty much rule out RSS being off on the low side as an explanation of the divergence.
Atomski, Another reason forcings are a red herring here is that CMIP5 used historical forcings up to 2005. The divergence was just as bad in 2005 as it is now. 2005 only gives a 25 year trend but that’s still pretty long. Corrected historical forcings for 2005 to today won’t affect the divergence in the earlier period. I also found another “commentary” in Nature by Schmidt that tried to correct the forcings to 2012. It looks to me that the differences are not going to make much of a change in the divergence, perhaps decrease it a little. It’s a complex area and there are lots of different papers and opinions. For example, if aerosol forcings are really less negative than thought as some experts are saying, then all model forcings have been too small.
RE: UAH, RSS and NOAA STAR (yes there is a 3rd one)
New generation of U.S. satellite microwave sounder achieves high radiometric stability performance for reliable climate change detection (open access published 17 Oct 2018)
Abstract
“Observations from the satellite microwave sounders play a vital role in measuring the long-term temperature trends for climate change monitoring. Changes in diurnal sampling over time and calibration drift have been the main sources of uncertainties in the satellite-measured temperature trends. We examine observations from the first of a series of U.S. new generation of microwave sounder, the Advanced Technology Microwave Sounder (ATMS), which has been flying onboard the National Oceanic and Atmospheric Administration (NOAA)/NASA Suomi National Polar-orbiting Partnership (SNPP) environmental satellite since late 2011. The SNPP satellite has a stable afternoon orbit that has close to the same local observation time as NASA’s Aqua satellite that has been carrying the heritage microwave sounder, the Advanced Microwave Sounding Unit-A (AMSU-A), from 2002 until the present. The similar overpass timing naturally removes most of their diurnal differences. In addition, direct comparison of temperature anomalies between the two instruments shows little or no relative calibration drift for most channels. Our results suggest that both SNPP/ATMS and Aqua/AMSU-A instruments have achieved absolute stability in the measured atmospheric temperatures within 0.04 K per decade. This uncertainty is small enough to allow reliable detection of the temperature climate trends and help to resolve debate on relevant issues. We also analyze AMSU-A observations onboard the European MetOp-A satellite that has a stable morning orbit 8 hours apart from the SNPP overpass time. Their comparison reveals large asymmetric trends between day and night in the lower- and mid-tropospheric temperatures over land. This information could help to improve climate data records for temperature trend detection with improved accuracy. The SNPP satellite will be followed by four NOAA operational Joint Polar Satellite System (JPSS) satellites, providing accurate and stable measurement for decades to come. The primary mission of JPSS is for weather forecasting. Now, with the added feature of stable orbits, JPSS observations can also be used to monitor changes in climate with much lower uncertainty than the previous generation of NOAA operational satellites.”
Bottom line? 0.04C/decade accuracy for era 2002-2018 (or 19 years with similar accuracy going forwards). Data before that time not useful for climate study purposes (1979-2001 are GIGO).
IMHO still not a good idea to combine channels in the hope of getting anything that one would call TLT …
Spencer and Christy, et. al. (and similar UAH deniers) can now RIP knowing they wasted their entire lifetimes.
Oops, last post needs link …
http://advances.sciencemag.org/content/4/10/eaau0049
Re: “Kiwi and Atomski, I mentioned all the sources of error mentioned by RSS and gave reasons why I though some of them were small. If you want to give your reasoning, I’d be interested in why you think I’m wrong.”
You gave your RSS link, after writing:
“Prof Curry said: ” it was clear that the computer models used to predict future warming were ‘deeply flawed’. ”
Well, it is becoming increasingly clear that this is true, even though perhaps it was not quite so clear in the past. Here’s a recent RSS analysis of their product. It’s entirely consistent with energy balance TCR estimates being much lower than those for GCM’s.”
Yet your RSS link explains that the differences are primarily not due to model error. So you shouldn’t cite the RSS page as if it supports the claim that the models are “deeply flawed”.
Re: “Atomski, Another reason forcings are a red herring here is that CMIP5 used historical forcings up to 2005.”
You likely didn’t read the Nature Geoscience paper RSS linked on their page. Here’s the paper:
“Causes of differences in model and satellite tropospheric warming rates”
As the paper explains, the discrepancy between the model-based projections and the satellite-based mid-to-upper tropospheric warming analyses, is due to errors in inputted forcings. Your point about “up to 2005” is irrelevant, since the discrepancy arises in the early 20th century, not before then (see page 483 of the paper).
So why are you citing an RSS page as if it supports your point, when both the page and the research it cites refute the point you’re making?
Re: “I also found another “commentary” in Nature by Schmidt that tried to correct the forcings to 2012. It looks to me that the differences are not going to make much of a change in the divergence, perhaps decrease it a little. It’s a complex area and there are lots of different papers and opinions.”
I’m not interested in appeals to disagreement. Anyway, Schmidt’s paper is looking at near-surface analyses, with updated forcings. There are plenty of papers that have done that. But, of course, near-surface analyses are not the same as the satellite-based tropospheric analyses on the RSS page you linked to. And no, the updated forcings do significantly reduce differences for near-surface analyses vs. model-based projections. This has been quantified, so there is not need to rely on your personal opinion on what the effect would be. If you want a list of papers covering updating forcings for near-surface analyses, then see (this includes the Schmidt piece you’re likely referencing):
“The ‘pause’ in global warming in historical context: (II). Comparing models to observations”
“Robust comparison of climate models with observations using blended land air and ocean sea surface temperatures”
“Reconciling warming trends”
“Natural variability, radiative forcing and climate response in the recent hiatus reconciled”
“Reconciling controversies about the ‘global warming hiatus’”
How much hotter would have been if there was no ‘pause’ ?
There are multiple observations that show a consistent warming trend in the climate over the last few decades.
Ignore the modulz… ignore the correlation with a credible physical cause.
Given the observed historical pattern of behaviour of the climate on what basis would anyone predict or expect that the warming trend will change.
The historical record and paleoclimate proxy indicators do not refute the possibility of large changes. Over longer timescales we know it has been colder and warmer than now, with periods of rapid warming.
If just half the warming we have observed is the result of internal variation/uncertainty rather than CO2 forcing, why would it stop ?
Unless you invoke some mystical faith in ‘cycles’, ‘waves’, or a strict climate homeostasis, none of which have any observational support.
The only reason to have any expectation that the warming trend could slow or stop is IF it is caused by our cumulative CO2 emissions and we take action to constrain further increases.
Re: “I particularly think measurement error is not a likely explanation given Marco’s point. […] Marco, if as you show RSS now has a higher rate of warming than GISS or HADCRUT it makes the divergence more concerning, not less. It seems to me also to pretty much rule out RSS being off on the low side as an explanation of the divergence.”
It depends on which RSS analysis you’re talking about. If you’re talking about their TTT analysis (a modified version of the TMT analysis, that focuses more on the mid-to-upper troposphere), then observational error is not the major issue. Error in inputted forcings is the major issue, as discussed in the paper I mentioned before, and which was cited on the RSS page you linked to:
“Causes of differences in model and satellite tropospheric warming rates”
If you’re instead talking about their TLT analysis (which focuses more on the lower troposphere), then observational error/uncertainty plays a larger role. In that case, RSSv4 TLT would be on the low side, contrary to what you claim. The RSS team admits this, and comparisons to other analyses show this:
“We find large systemic differences between surface and lower troposphere warming in MSU/AMSU records compared to radiosondes, reanalysis products, and climate models that suggest possible residual inhomogeneities in satellite records. We further show that no reasonable subset of surface temperature records exhibits as little warming over the last two decades as satellite observations, suggesting that inhomogeneities in the surface record are very likely not responsible for the divergence.”
http://adsabs.harvard.edu/abs/2017AGUFMGC54C..05H
Page 7715 of: “A satellite-derived lower-tropospheric atmospheric temperature dataset using an optimized adjustment for diurnal effects”
[Figure 2.7 of page S17 of “State of the Climate in 2017”]
Might not work but …
https://twitter.com/LabradorIce/status/1075445319087853570
Cut-Off Dates for literature to be considered for AR5
Updated 12 December 2012
Click to access ar5-cut-off-dates.pdf
Working Group I Report [31 July 2012 Papers submitted]
Working Group I Report [15 March 2013 Papers accepted]
First published: 12 November 2013
Oops, forgot link to “12 November 2013” date …
https://rmets.onlinelibrary.wiley.com/doi/full/10.1002/qj.2297
The ‘pause’ in global warming in historical context: (II). Comparing models to observations
http://iopscience.iop.org/article/10.1088/1748-9326/aaf372/meta
So where is part (I)?
Oh wait …
“The present analysis sets the stage for an exploration of the mechanisms by which this may have occurred, which will be the focus of future contributions.”
Part (I) comes after part (II)?
It is a prequal to Lewandowsky’s Seepage paper titled … wait for it …
“Sh-e-e-e-e-e-page” 😉
Well Atomski, you definitely like to cite a subset of papers that claim that forcings are in error for CMIP5. I’d prefer to wait for AR6 where they will consider all the evidence, not just the papers you select. If you believe Schmidt, who is usually pretty reliable, forcing errors will make only a small difference. You didn’t respond to any of my specific arguments which makes me suspicious that you don’t have good responses.
Further Atomski, your quoted assertion that “We further show that no reasonable subset of surface temperature records exhibits as little warming over the last two decades as satellite observations” is quite clearly wrong. Blindly citing papers can lead to mistakes. As Marco pointed out earlier, RSS warms quite a bit faster than both HADCRUT and GISS.
Here are some model/obs comparisons for 1975 – 2017 (C/decade)
Apples-to-apples there is a good trend match.
GISS 0.183
HADCRUT 0.180
Cowtan+Way 0.191
Best 0.190
RCP6 0.219
RCP6Blended 0.189
Model data from KNMI explorer. RCPblended uses SST instead air temperature over ocean to match observations which also use SST. Blending ratio is 71% SST and 29% Land
PS – DPY I was referring to a more recent Dessler paper not the one blogged by NL. Worth a read. It uses a climate model to emulate an EBM and shows that the differences between models and EBM can be reconciled.
@-dpy6629
Do you have any reason to think the climate will NOT continue to warn at the same rate observed for the last 5 decades ?
> there are 2 elements of selection bias in quote mining
The first element you provide needs to be established, DavidY. If you think some context is missing, it’s your job to provide it. Asserting that it lacks context is seldom enough, and in fact what you’re doing right now echoes a common Lobstersonian trick:
https://twitter.com/nevaudit/status/1074470339470602241
Your second element may not work either. How is the focus on a specific author cherry picking? That argument only floats if there’s some claim of representativeness at stake.
While you may wish to preserve your energy for your own pet topics, Da Paws is the topic of this very thread, and it’d be hard to argue that Judy hasn’t been long on Da Paws for most of her blogging life. In fact, she’s still committed to it:
“Da Paws” happened. I was there.
dikranmarsupial says on December 21, 2018
One man’s noise is another man’s signal. Many researchers are treating the ENSO part as separable and thus compensating the ENSO fraction out, with the result showing a much smoother trend. This is less a statistical approach than elementary signal processing.
dpy wrote “In my field, selection bias is a given.”
In my field we would try to acknowledge and eliminate such biases, rather than exploit them.
“The only reason to have any expectation that the warming trend could slow or stop is IF it is caused by our cumulative CO2 emissions and we take action to constrain further increases.”
A bit like Wiley Coyote really, off the road over the canyon and no way back only down.
“a mantra mystical faith in ‘cycles’, ‘waves’, or a strict climate homeostasis,”
Some form of the last is really, really appealing in these circumstances.
Particularly while it is not the fall that hurts.
JCH: “The Earth is flat I’m there. I can see it”.
Sorry, that’s not how it works in science. Or statistics. In scientific terms, your gut feel doesn’t even amount to a hypothesis. Because a hypothesis is an attempt to explain evidence. And you ain’t got no evidence.
Not sure what JCH was saying, but reminds me of:
[Carl Sagan, “The Demon-Haunted World”]
@-angtech
“Some form of the last is really, really appealing in these circumstances.”
As is the idea that TCR might be less than estimated, or Natural Variation will save us from more extremes.
But our lived experience contradicts that.
For every adult alive today, next year has always been one of the top ten hottest years they have experienced. With the record advancing every decade.
But the bottom ten COLDEST years we have lived through were all in our childhood or early teenage years.
That may be unique in human history. Or at least since the rapid warming ~8000 years ago at the end of the glacial cycle.
We have no realistic prospect of every experiencing a year as cold as those in our youth, and every prospect of encountering years that set new records for the warmest year.
There has been no pause, or prospect of one, that would negate that pattern.
Izen – Yes, this warming is a freight train with the direction of the tracks is well established over the past 40 years. Models not needed to see where the train is headed. EBM a distraction.
“Da Paws” is a completely different thing than the pause. “Da Paws” will never die because it was already dead.
Back in the heyday of the pause I used to tease acolytes of the witchdoctors at Climate Etc. with announcements that the pause had gone PAWS UP. And in no time people in blog comments around the super information highway started calling it the PAWS.
When animals die they bloat. Sometimes the gas pressure rolls the carcass over on its side or back, and the paws/hooves stick up in age air. It’s a common death pose: paws up.
@dpy6629
Re: “If you believe Schmidt, who is usually pretty reliable, forcing errors will make only a small difference.”
I actually read Schmidt’s paper, so I know what you’re saying is false. I cited his paper to you, along with other research that applied his forcing estimates. So I don’t need to be gullible enough to believe the false claims you’re making about what he said.
Re: “You didn’t respond to any of my specific arguments”
No, I actually responded to what you said. Anyone reading this discussion can see that.
Re: “Further Atomski, your quoted assertion that “We further show that no reasonable subset of surface temperature records exhibits as little warming over the last two decades as satellite observations” is quite clearly wrong. Blindly citing papers can lead to mistakes. As Marco pointed out earlier, RSS warms quite a bit faster than both HADCRUT and GISS.”
We scientists actually cite scientific research; that’s what I do when I write papers. If that bothers you, then that’s not my problem. And what I quoted from you came from Carl Mears of the RSS team. I think he’d know better than you what the RSS analysis shows.
Your citation of Marco is irrelevant here for at least three reasons. First, the quote you’re responding to is for over the last two decades. But Marco was looking at post-1979 trends, not trends over the last two decades:
Second, HadCRUT4 under-estimated warming due to it’s poorer coverage. This has been covered in the literature over and over. For instance:
“Coverage bias in the HadCRUT4 temperature series and its impact on recent temperature trends”
“A reassessment of temperature variations and trends from global reanalyses and monthly surface climatological datasets”
“Arctic warming in ERA‐Interim and other analyses”
“The ‘pause’ in global warming in historical context: (II). Comparing models to observations”
“Recently amplified arctic warming has contributed to a continual global warming trend”
“An investigation into the impact of using various techniques to estimate arctic surface air temperature anomalies”
Third, Marco is using WoodForTrees, which I don’t consider to be a very credible source, given what it says about itself:
“The algorithms used on this site have not been formally peer reviewed and hence should not be used unverified for academic publication (and certainly not for policy- making!). This site is only intended to help find interesting directions for further research to be carried out more formally.”
http://woodfortrees.org/credits
There are more reliable sources for temperature analyses, such as:
https://www.esrl.noaa.gov/psd/cgi-bin/data/testdap/timeseries.pl
“Web-Based Reanalysis Intercomparison Tools (WRIT) for analysis and comparison of reanalyses and other datasets”
https://climexp.knmi.nl/selectindex.cgi?id=someone@somewhere
http://images.remss.com/msu/msu_time_series.html
http://www.ysbl.york.ac.uk/~cowtan/applets/trend/trend.html
Here are the 1998-2017 trends using Cowtan’s trend analyzer (you can double-check that with the other sources listed above):
RSS TLT: 0.16
GISTEMP: 0.19
Berkeley Earth: 0.19
Cowtan+Way: 0.18
HadCRUT: 0.15
NOAA: 0.18
So RSS TLT shows less warming over the past two decades that each analysis, except for HadCRUT4, which under-estimates warming because of the coverage issues I mentioned over before. Thus the source I cited for you before was right, contrary to what you said. You can go check this with near-surface analyses such as ERA-I, JRA-55, and JMA (though note that JMA suffers from coverage issues as well). So it looks like RSS’ Carl Mears knows more about his own analysis than you do, as expected.
To paraphrase you:
Blindly repeating talking points, without checking the scientific literature or checking what you’re saying makes sense, can lead to mistakes.
Being somewhat pedantic Paul, noise is everything other than the signal you’re looking to detect or characterise. It doesn’t matter whether that “other” is random, chaotic or periodic, or even whether it’s another, interfering signal. If I have a single, perfectly periodic pendulum on a trolley, and I have no sensor on the trolley, just on the pendulum, that nice, regular, periodic oscillation is noise to me, if I’m trying to find out whether the trolley is stationary or moving very slowly.. Same if I’m an astronomer looking at a star close to the line-of-site of the Sun, or an audiophile trying to hear faint sections of music through mains hum. If you can characterise the coherent noise well enough to subtract it: well that’s another matter and well worth having.
“In my field we would try to acknowledge and eliminate such biases, rather than exploit them.” Well of course that’s the goal we want to work toward. The problem here is that so far the attempts to get people to acknowledge and eliminate these biases have not been tremendously successful. In one small area of medicine, preregistration of trials has helped. In fact it has shown a shocking drop in positive results indicating that there were and still are a lot of false positive results out there.
https://www.nature.com/news/registered-clinical-trials-make-positive-findings-vanish-1.18181
I’m not sure who you think is “exploiting” these biases. It is true that false positive results can generate big splash press for scientists especially if their press release is misleading. There are a lot of strategies to prevent these biases. One is workshop formats where everyone obtains and submits their results without knowledge of the test results. But really one needs something else where people document the sensitivities of their results to parameters of their codes and methods. That’s harder to do and requires a lot more investment of time pre publication. However, there has been progress in my field over the last 15 years or so.
DPY wrote “I’m not sure who you think is “exploiting” these biases. ”
you are by refusing to acknowledge the inconsistencies and then using the existence selection bias to justify refusing to discuss them, examples given on this and the previous thread.
dpy, there has been progress in the climate science field over the last 150 years or so. ISTM it’s about a century ahead of your field. Which makes your apples-to-oranges comparison worthless, I’m afraid.
willard:
If the mountain won’t come to Muhammad…
Still on Peterson? I like lobsters and a lot of the people he’s hanging out with. I thought this one was good:
Jordan Peterson and Ben Shapiro: Religion, Trans Activism, and Censorship
I’d like to see you go right for his heart. He must be weak right? The mob has been trying that for a few years. I do recall some of our last exchanges on the subject. It wasn’t that memorable. But not we have more to work with given the passage of time. You ought to look at his first videos as far as I know. He’s set up a camera pointing somewhat upwards. It’s the most low rent video one has ever seen. And he’s teaching is class and he isn’t famous. Chloe Valdary was on Rubin’s report. She described him as a story teller. I agree. With good intentions. The pushback is interesting. A sport. Most of it foolish and well below your level. Your above cartoon isn’t that convincing. Of course there’s some of that. It’s not new. Aren’t people dredging up his past to out him on something? We’re doing good. It might have been Joshua. My point to him here as far as I know was that Peterson is misunderstood. What is the most tired old plan when someone new comes along? Are we still doing that? If we could see where he’s coming from rather than just know he’s wrong and then try to prove that. Now this is a point a number of his supporters make. Where is he coming from? Heaven I guess. The far North. We mostly trust the Canadians. And I for one thank them, not for Peterson but for being our allies all these years. Not happy about that whole dairy pricing thing but let’s let bygones be bygones.
This would be great ground to plow willard. His biblical series. You could be Sam Harris. It’s a story. And the question is, what value has it had? And do we have a better story? If not the bible and not Peterson, what? The left’s current incoherence?
He does do some cool stuff with his biblical series. Lots of images: https://en.wikipedia.org/wiki/Flammarion_engraving?fbclid=IwAR2sf- 0ZukRyY6Hn2I1hQDQHYeM5rZOehM5NSQPU4bk3f_zp-d6iHJyjxL8#/media/File:Flammarion.jpg
@Ragnaar
Re: “Still on Peterson? I like lobsters and a lot of the people he’s hanging out with. I thought this one was good:
Jordan Peterson and Ben Shapiro: Religion, Trans Activism, and Censorship”
I’m pretty sure the topic of discussion is climate science. If you really wanted a discussion on Peterson, then he’s quite easy to debunk, since he makes silly claims. For example, he displays anti-atheist prejudice and a deep misunderstanding of meta-ethics, when he claims that intellectually-consistent atheism is on par with murder. And I say that as a Christian.
Bringing this back to the actual topic of discussion:
Jordan Peterson makes ridiculous claims on climate science, which is unfortunately common among many political conservatives like him. His problem largely stems from low scientific meta-literacy: he relies on unreliable non-peer-reviewed contrarian sources such as WattsUpWithThat and NoTricksZone.
You and I have been over this before; you had no adequate defense of Peterson’s position. There’s a reason why people like Gavin Cawley (I think that’s dikranmarsupial), I, and others have corrected Peterson’s claims:
“For example, he repeats nonsense […]”
http://archive.is/ab085
Ragnaar –
We’re doing good. It might have been Joshua.
Please translate?
Btw, Jim D linked to the article Judith linked to, decrying “academic activism.” What you might find interesting is that if you read the comments, the author points to Peterson as an example of problematic academic activism.
The author also had some interesting stuff to say about climate science and academic activism.
Check it out.
Joshua,
I actually thought that article was a bit confused, but somewhat better than the title indicated. I ended up in a discussion on Twitter about it, but it was only after the discussion had ended that I realised it was the author I was engaging with. They said
which, as far as I’m aware, is simply not true. They didn’t seem willing to clarify.
Joshua:
As I recall it, and I am old, I thought I was telling you that he is misunderstood. There’s a 25% chance that never happened. I could’ve been someone else.
Chubbs, I did find this clear explanation from Nic Lewis for the effect that Dessler seems to be referring to. I can’t access the full text of their paper though.
“An important issue has come to light recently, relating to the dependence in AOGCMs of cloud feedbacks, and hence climate sensitivity, on patterns of surface temperature change – primarily in sea surface temperature (SST).40 In most CMIP5 AOGCMs, two or three decades after a CO2- based forcing increase is imposed, their effective climate sensitivity increases, largely because short-wave cloud feedback becomes (more) positive. This leads, on average, to their diagnosed ECS being ~10% higher than their effective climate sensitivity when estimated from global changes corresponding to those that took place during the instrumental period. In typical CMIP5 models, intensified warming develops in the eastern tropical Pacific a few decades after an imposed increase in atmospheric CO2. This appears to be the main reason for the increase in short-wave cloud feedback, and hence the principal underlying cause for effective climate sensitivity increasing over time in most AOGCMs. Moreover, even in the first two decades after imposing a step increase in atmospheric CO2, CMIP5 models show strong warming in the deep tropical eastern and central Pacific. This is associated with a robust weakening of the Pacific Walker atmospheric circulation (easterly near the surface, westerly high in the troposphere) in these models, which inter alia reduces the upwelling of cool water in the eastern tropical Pacific. There is a theory as to why the Pacific Walker Circulation may be expected to weaken under greenhouse gas warming,41 but reasons to expect the opposite have also been put forward.42 In reality, CMIP5 models have been proved wrong to date: the Pacific Walker Circulation has fluctuated, but does not show a long term weakening trend and, far from warming strongly, the eastern and central Pacific have warmed little (apart from during El Nino events).”
Dessler’s statement that “We find that TCR estimated from the 20th century simulations may indeed be much lower than the model’s true TCR. This arises from biases in the methodology of estimating TCR from 20th century warming, as well as biases in the construction of the observational temperature data sets. We therefore find no evidence that models are overestimating TCR.” seems to me like a stretch.
Well Dikran, I think you and I disagree on the importance of searching for inconsistencies in the vast public OPUS of Prof. Curry. Perhaps we should leave it at that.
dpy,
Doesn’t it bother you that Nic seems to spend quite a lot of time defending his claims that climate sensitivity is probably lower, rather than spending time challenging this? Isn’t science about trying to test your hypotheses, rather than accepting them? To be clear, even mainstream results don’t rule out that it is one the low side. The main thing that Andrew Dessler points out is that if clouds are not a negative forcing, then the ECS is probably above 2K. Much other work (see Kate Marvel’s work, for example) indicates that clouds are probably not a negative forcing. Hence, we would expect it to be more likely that the ECS is above 2K, rather than below. I don’t recall seeing Nic providing a convincing physically-motivated argument as to why it’s more likely below 2K than above.
DPY you do realise, don’t you, that you have just demonstrated the validity of my point?
@dpy6629
Re: “You didn’t respond to any of my specific arguments which makes me suspicious that you don’t have good responses.”
I find this comment of your’s pretty ironic for a few reasons.
First, as I noted to you before, I responded to you the specious claims you made on this thread. You couldn’t muster a cogent response here, though I did rebut the response you made at Curry’s blog (my response seems to have ended up in the wrong place):
“Same mistakes I already corrected”
http://archive.is/Gzmk4
Second, when I rebutted distortions you made on satellite-based analyses on Curry’s blog, you didn’t respond. So you did the very thing you’re complaining about. By the way, do you now understand that both RSS and UAH changed their homogenization methods, not just RSS?:
“RSS and UAH “both” changed their homogenization”
http://archive.is/FXUcm
Third, I’ve addressed your evasion of evidence on water vapor feedback over at Curry’s blog:
“ambiguous on the water vapor feedback”
http://archive.is/sbSoh#selection-8379.246-8379.283
So no, I’m really not afraid to respond to the claims you make. In contrast, as dikranmarsupial noted above, you’re prettying good at finding reasons not to discuss issues with what you say.
On the subject of story telling, Peterson is kicking the consensus’s butt. We realize how lean is operation is right? He’s exploring new delivery options while cashing revenue checks. In a hostile environment. We could probably say that about Jimmy Swaggart too at one time. Now we can say he’s bamboozling people, like Rubin and the Weinsteins for instance. And a number of others from the left who we recognize and may have agreed with in the past. There is this thing that has grown from the middle. In reaction to many things in the system. And what I see is the same tired attacks. Lacking much of worth. Predictable. We don’t have to agree with him. But can we learn something about what is going on. Someone like Peterson draws out reactions. Very generally the right goes, Yep. And the left goes, he’s a bad person with some turning up the dial beyond that. I think that there are parallels with the climate debate. That being the left’s reaction to criticism. And the way a center has grown enough to be some concern to the left as some rhetorical shots are fired into it. We’ve done that enough in many areas. We’ll call the middle – extremists, and that will work. We look at the reaction to Peterson and compare that to past reactions to climate lukewarmers. Do we know anything and can we predict anything? The way I see it, taking an extreme position must fail eventually. I am going to anticipate that, there are not many extremists and that’s just a device of the right. Meaning they are middle left, as they lob rhetorical shots at the middle. I don’t know who is attacking the middle if it’s not the left? I think the left defines itself with who it attacks. Back to story telling. He’s good at that. People can take some new understanding with them after watching his videos. What do we take away from the climate debate?
Re: “There is measurement error, but RSS agrees pretty well after it was adjusted with the warming rate in HADCRUT4 , which is very well vetted. UAH shows quite a bit less warming and GISS shows more.”
Sorry I forgot to mention this (the moderators can feel free place this in my previous comment in the prefer, so I’m not cluttering the thread):
Suppose one accepts your logic. Then by your logic, UAH under-estimates warming, since it shows less warming that the “well vetted” HadCRUT4 analysis. Yet you’ve previously claimed that:
“There is no evidence UAH’s current processing method is in error.”
http://archive.is/FXUcm#selection-7479.0-7479.65
So it looks like your position is internally inconsistent, even if one accepts your logic.
I don’t accept the logic you’re using for the reasons I went over before: HadCRUT4 under-estimates warming, RSS TLT under-estimates warming over the past two decades (both by indirect tests with water vapor, and in comparison to other analyses such as radiosonde-based analyses and re-analyses), one would expect satellite-based lower tropospheric warming estimates to show a bit more warming than near-surface analyses, etc.
For Freedom Fighters stuff it’s next door, Ragnaar:
Izen:
“For every adult alive today, next year has always been one of the top ten hottest years they have experienced. With the record advancing every decade.
But the bottom ten COLDEST years we have lived through were all in our childhood or early teenage years.”
Very interesting. Do you have any links so I can read a bit more?
@-KenH
“Very interesting. Do you have any links so I can read a bit more?”
No, full disclosure, I estimated by eyeball from the graph of GMST.
It might not be true for every age or year, but I bet it is close, and the analysis to detect any exceptions will make one thing clear, the relentless trend.
Anders –
I think Wells’ arguments need specificity, clarification, and more supporting evidence. I indicated so in the comments.
I did do this a few years back;-
Well ATTP, Nic has done a more thorough job than I’ve seen anywhere else at an individual feedback analysis. Nic also has done a better job of communicating by very carefully answering all people who have legitimate questions on his blog posts. Here’s an example of that analysis:
“As well as ‘top down’ ECS estimates reflecting equation (1), one can use a “feedback analysis” approach, estimating individual climate feedbacks in models and/or from observations, and hence ECS from equation (3). However, it has recently been found that F2xCO2, the ERF from increased CO2, varies greatly between CMIP5 AOGCMs, which implies inaccurate radiation codes and increases uncertainty regarding estimates of feedbacks in AOGCMs. For a quadrupling of atmospheric CO2, for which the expected ERF is ~7.4 Wm−2, CMIP5 models have ERF varying between 5.5 and 9 Wm−2.26 Feedback analysis is in practice closely related (via effective climate sensitivity) to ECS as exhibited by AOGCMs, so I will not separately critique the raw ECS and TCR ranges exhibited by CMIP5 models, which as already shown are centred substantially above best estimates from sound instrumental-observation studies. The mean ECS and TCR values of climate models used for IPCC projections of future warming are respectively ~3.4°C and ~1.9°C. Their estimated ECS and TCR values lie in the ranges 2.1–4.7°C and 1.1–2.6°C respectively.
Turning to feedbacks, the Planck radiative feedback λp is fairly consistently estimated by AOGCMs at ~−3.2 Wm−2 °C−1, so in the absence of other feedbacks ECS should be ~3.7 / 3.2 = 1.15 °C. The minor albedo feedback is estimated reasonably consistently by AOGCM simulations at ~+0.3 Wm−2 °C−1, equating to another ~0.1°C on ECS. AOGCM estimates of both these feedbacks are generally considered to be reasonably realistic. . . . .”
There is a lot more on clouds and water vapor which you can read as I linked to it above.
More to your point, ATTP, I haven’t any definite idea what ECS or TCR are. Perhaps later in life I will have the time to do what Lewis has done on climate. I would say that AOGCM’s are not satisfactory evidence for it one way or another.
The basic point is that these models have very large truncation errors which are much larger than the small changes in energy fluxes they are attempting to model. Sub grid models for clouds, convection, and precipitation are unlikely to get even the global effects of these processes correct on the grids used. It’s exactly the same kind of thing as turbulence modeling (RANS calculations are done on vastly more well resolved grids) and eddy viscosity models are solving global PDE’s for eddy viscosity. Older algebraic turbulence models which are similar to what is in GCM’s for convection were quite clearly globally defective. One must conclude that the tuning of the models is adjusting cancelling errors to get some few measures on the fluxes right such as top of atmosphere net flux. And that’s borne out by some recent papers on model tuning.
Well Atomski, I don’t find your argument about measurement errors persuasive. Indeed RSS TLT does warm substantially faster than GISS over the 40 year record as Marco showed above. That agrees with my examination of the records themselves. You have not addressed that and have introduced a whole host of other issues concerning shorter time periods. It seems to me that the forcing issues are not going to make a difference of more than about 0.1 C for the period 2005 – present and much less over 1979 – 2005 for which the forcing data is much better established. That still leaving a significant divergence for both the 26 and 39 year periods.
I don’t know which of RSS/UAH is closer to the truth. It’s not important for this discussion. I just found it striking that even after a large upward revision with their new dataset ( at least 0.2C) RSS still diverged with CMIP5 as much as it did.
The other thing of course is just that AOGCM’s are pretty weak on things like predicting the pattern of warming, cloud fraction by attitude, etc. and there are first principles reasons to doubt their skill.
Well Dikran, I kind of see your point. It seems to me a real stretch. Fixing bias in science is a huge task with many requirements in the way science is done and published and models are run.
Searching for inconsistencies in vague and voluminous public statements of one particular scientist who you selected for scrutiny does nothing to further the cause of reforming science and just wastes time.
Jordan Peterson has jumped the shark – actually he has jumped a school of sharks. It was all very well when he made up outlandish claims about lobsters and Canadian Marxists, but he has now straw manned Climate Science and used PragerU as a reference so the gloves are off.
dpy – a couple of comments:
1) The Lewandowsky et. al. paper that is the subject of this post, compares model predictions and observations in great detail. The match is quite good. Per my comment above, I have downloaded model data and also find good agreement between models and observations. It is very important to make apple-to-apple comparisons when evaluating models.
2) You haven’t addressed the Dessler paper that I linked. This paper showed that models can replicate EBM results when the models predictions and observations are matched properly. The paper matches my experience using data downloaded from KNMI explorer. Per my checking, the temperature delta between the 1869-1882 and 2007-2016 periods used in the recent L&C paper is: Hadcrut 0.79, Best 0.96, CMIP5 mean SST blended 0.81. So the recent L&C paper cannot be used to criticize climate models. The paper results are exactly what CMIP5 predicts.
3) It is tedious to go through the rest of your comments. There are many acknowledged flaws in climate models, but it is very hard to tie specific flaws to bias. Qualitative arguments on a blog are not persuasive. Quantification is needed to prove bias. There are a number of emergent constraint papers that try to improve model ECS estimates by discarding or giving less weight to models with poor performance. So far these papers haven’t indicated that model errors were causing ECS to be overestimated. On the contrary, many of these papers show underestimation of ECS in the better performing models.
Re: “Well Atomski, I don’t find your argument about measurement errors persuasive. Indeed RSS TLT does warm substantially faster than GISS over the 40 year record as Marco showed above”
Same mistake from you that I already addressed elsewhere:
http://archive.is/wXIfn#selection-5517.0-5517.34
Once again:
RSS TLT is not a near-surface analysis. It is a satellite-based analysis focusing on the lower troposphere. It would be expected to show a bit more warming than the near-surface analysis, not exactly same rate. So pointing out that it shows more warming since 1979 than GISS, a near-surface analysis, is consistent with RSS still under-estimating warming. This includes under-estimating warming over the past two decades, which RSS clearly does for the reasons I’ve repeatedly explained to you in this thread.
Since you seem to be having trouble with this, I think an example might help. Suppose, for the sake of argument:
actual near-surface warming = 0.20 K/decade
GISS = 0.19
And for the lower troposphere:
actual lower tropospheric warming = 0.24 K/decade
RSS = 0.21
average of the re-analyses = 0.23
average of the radiosonde analyses = 0.23
In that scenario, RSS still under-estimates lower tropospheric warming, even though it shows more warming than GISS. Thus, pointing out that RSS shows more warming than GISS does nothing to show that RSS does not under-estimate warming. However, the reverse is not true: if RSS showed less warming than GISS, the re-analyses, and the radiosonde analyses, then that would be evidence of RSS under-estimating warming, as Carl Mears noted in the link I cited to you before.
Re: “You have not addressed that”
Who do you think you’re fooling?
Re: “It seems to me that the forcing issues”
It’s irrelevant what your personal opinion is on the effect of the forcing issues. What matters is what the published evidence shows. The published evidence actually quantified the effects of forcing for near-surface and mid-to-upper tropospheric comparisons. I’ve already cited the literature on that for you; it shows how correcting the forcing issues largely resolves the differences between model-based projections and observational analyses, especially since the relevant differences between projections and observational analyses predominately occur in the early 21st century period when the forcing differences occur.
Forcing issues that affect the near-surface estimates would also affect lower tropospheric estimates and mid-to-upper tropospheric estimates, since warming from the near-surface is transferred up through the troposphere, as per the lapse rate feedback. This transfer is relatively greater in the tropics, leading to a negative lapse rate feedback there. For further context on that, see:
“Relationship of tropospheric stability to climate sensitivity and Earth’s observed radiation budget”
Re: “I don’t know which of RSS/UAH is closer to the truth”
That’s not the inconsistent position you held before. Before you claimed there was no evidence that there was something wrong with UAH, and then claimed UAH showed less warming than the “well vetted” (your terms) HadCRUT4 analysis. You’re now changing your position, when the contradiction was pointing out. That’s progress, I guess.
Re: “The other thing of course is just that AOGCM’s are pretty weak on things like […]”
I’m not interested in evidence-free Gish gallops.
Re: “Well ATTP, Nic has done a more thorough job than I’ve seen anywhere else at an individual feedback analysis. Nic also has done a better job of communicating by very carefully answering all people who have legitimate questions on his blog posts. Here’s an example of that analysis”
Lewis’ own cited sources rebut his position. For example, the sources he cites on the Last Glacial Maximum in order to justify his low ECS estimate of ~1.76K, either argue against his low estimate or argue for a greater estimate:
http://archive.is/JNlx8#selection-2589.0-2589.28
So simply reading some of the material Lewis cites would have undermined his position. Did you do that?
Re: “More to your point, ATTP, I haven’t any definite idea what ECS or TCR are. Perhaps later in life I will have the time to do what Lewis has done on climate”
If you don’t know the basics of the topic (knowing what ECS and TCR are is pretty basic to this topic), then how are you are in position to evaluate if Lewis is right in the face of the large number of scientists who rebut what he says? Are you simply going with what Lewis says because it’s what you want to hear?
DPY no, inconsistencies in science are important, and ignoring them when they are pointed out to you shows a lack of the self skepticism that is a basic requirement of science.
Dave_Geologist says:
Calling it “coherent noise” is quite a contortion of terminology. Coherence is implying a phase relationship between two waves. The term you are searching for is likely a nuisance parameter or confounding factor, i.e. a signal that gets in the way of another signal that one is ultimately interested in. So the daily or annual temperature variation may be a nuisance or confounding factor to determining a long-term trend. But apparently this is something that you would rather call coherent noise.
Ragnar wrote “Peterson draws out reactions. Very generally the right goes, Yep. And the left goes, he’s a bad person with some turning up the dial beyond that.”
No, the video he promoted contains factual errors and obvious misrepresentation of the sources it quoted, this is not a matter of partisan opinion. This reflects more badly on Lindzen, who ought to know better, but it does show that Peterson is gullible and lacks the ability (or will) to perform basic fact checking (which makes his taunt effectively bullshit).
Sagan didnt live long enoug to read “the second brain”
you cant help but think with your gut
Based on Figure 7 in the RSS pdf that dpy linked, the RSS TPW measurement indicates that RSS TLT could be biased low by 20% or more. We have no way of knowing whether TPW or TLT is correct, but diurnal drift and a cooling stratosphere should be much less important sources of error for TPW.
So, assuming Nic’s right (which he probably isn’t, what with him not being a GCM expert and those who develop and run the models and invented the term ECS being GCM experts), that explains 10% of the discrepancy between him the mainstream. What explains the other 90%?
Just for the record, I will quote the entire abstract from the paper from RSS about their new version. Bottom line, they now have a higher warming rate than most satellite or radiosonde based products.
“Temperature sounding microwave radiometers flown on polar-orbiting weather satellites provide a long-term, global-scale record of upper-atmosphere temperatures, beginning in late 1978 and continuing to the present. The focus of this paper is a lower-tropospheric temperature product constructed using measurements made by the Microwave Sounding Unit channel 2 and the Advanced Microwave Sounding Unit channel 5. The temperature weighting functions for these channels peak in the middle to upper troposphere. By using a weighted average of measurements made at different Earth incidence angles, the effective weighting function can be lowered so that it peaks in the lower troposphere. Previous versions of this dataset used general circulation model output to remove the effects of drifting local measurement time on the measured temperatures. This paper presents a method to optimize these adjustments using information from the satellite measurements themselves. The new method finds a global-mean land diurnal cycle that peaks later in the afternoon, leading to improved agreement between measurements made by co-orbiting satellites. The changes result in global-scale warming [global trend (70°S–80°N, 1979–2016) = 0.174°C decade−1], ~30% larger than our previous version of the dataset [global trend (70°S–80°N, 1979–2016) = 0.134°C decade−1]. This change is primarily due to the changes in the adjustment for drifting local measurement time. The new dataset shows more warming than most similar datasets constructed from satellites or radiosonde data. However, comparisons with total column water vapor over the oceans suggest that the new dataset may not show enough warming in the tropics.”
Atomski, This is getting a little bit over the top. You “select” from Lewis’ lengthy article one statement (on a secondary point) where another scientist disagrees and then ignore the other thousands of statements in Lewis’ article. That’s called quote mining and does not discredit what Lewis says. It casts doubt however on your ability to fairly evaluate these issues.
dpy, someone like Peterson draws out reactions. Very generally the right goes, Yep. That’s what I want to hear. And the left goes, Nope. That’s not what I wanted to hear. And scientists go, that’s interesting, he’s claiming to base it on science, not just on the Truth Of Christianity. maybe I’ll have a look at the science. And then the scientist goes, Gosh, that’s worse than Wakefield. Why do people fall for it?
Dikran, Inconsistencies in the scientific literature or scientific theories are important. Inconsistencies in vague statements pulled from a vast body of work that may or may not reflect a scientists real position are irrelevant to “science.” It’s a critical distinction, don’t you think?
Mainly because that’s what it was called in seismic processing Paul, in my experience. But call it what you like.
@-dpy
“Inconsistencies in vague statements pulled from a vast body of work that may or may not reflect a scientists real position are irrelevant to “science.””
Very few scientists in any field have the privilege of reading evidence into the congressional record of the US legislature. JC has been invited several times. One might assume that those are not vague statements and do reflect the real position she holds.
How well do you think they would withstand due diligence ?
David –
It’s a critical distinction, don’t you think?
Sure. That’s a critical distinction. But that doesn’t mean that inconsistencies and vague statements should be above scrutiny, or immune from criticism. And it doesn’t mean that scientists, vested with, and relied upon because of their, “epistemic authority” should not be accountable for their inconsistencies and vague statements.
Judith regularly denigrates the actions of those she seems “activist,” attaching a bevy of negative outcomes to their “activism” (such as a public crisis of trust in scientists) while defending her own actions by saying she doesn’t consider what she does to be biased at all. And in that stance, she has near uniform (and IMO, uncritical) support from the “skeptic” community and others.
Thus the problem, IMO, when she testifies before Congress and reports the science in an imprecise manner – such as describing a “haitus in global warming” without discussing the uncertainty in OHC (or even, in fact, discussing OHC at all as if it would be meaningful to desire “global warming” without describing the bulk of the earth’s thermal mass) ), even as she talks about the importance of including uncertainties when describing the risks of continuing ACO2 emissions.
Not only is that activism, IMO (obviously so, as she is testifying at the behest of politicians who are trying to implement specific policies), but it is irresponsible activism if (1) the testimony (willfully?) distorts the science by ignoring important uncertainties and (2) is not followed by clarification of inconsistencies and vague statements.
Don’t you agree? Are you always similarly inclined to dismiss inconsistencies and vague statements made at the behest of politicians who are invested in implementing specific policies?
(That’s a yes or no question, btw).
If not, what are the criteria you use to distinguish when you would and wouldn’t be similarly dismissive?
Keep in mind, I am not criticising Judith for inconsistencies and vague statements. As you note, it happens a lot. Priba ly everyone is prone to such issues. I am being critical of the inconsistencies and the vagueness in what she has said, and saying it is important that people whose Congressional testimony – at the behest of politicians pursuing specific policies – contains inconsistencies and vague statements, should take pains to clarify their inconsistencies and vague statements. That is particularly true if they are concerned about potential negative outcomes from scientists being advocates.
DPY no, I don’t agree. Inconsistencies in a leading scientists’ public statements on scientific matters of societal importance should not be ignored.
Here is another one, where Prof. Curry is promoting a GWPF report in which the very first criticism is based on a misleadingly selective quote from the IPCC report, but is directly refuted by the following sentence that was omitted from the quote. Prof Curry continued to defend the report even after that was pointed out.
To how many of these inconsistencies should we turn a blind eye?
Dave said:
OK, so as I was saying, removing a 60 Hz or 50 Hz (for UK) line hum has very little to do with statistics and is mainly a signal processing filtering exercise.
Chubbs, The Dessler paper appears to be paywalled. Based on Dessler’s earlier errors about energy balance methods that Lewis rebutted very effectively, I’m not going to pay to access the paper.
Figure 7 in the RSS piece is about total column water vapor and in my view is not as reliable as actual temperature measurements from radiosondes.
I’d be cautious about the Lewindoswki results. For CMIP5 everything prior to 2005 is a hindcast. Whether or not modelers tuned specifically for global mean temperature, they knew very well what the data showed. The TLT results seem more interesting to me because I doubt the modelers knew about them when they tuned their models. The problem here for climate models is that there are an almost infinity of output functions to look at and only a limited number of model parameters to vary. A balanced picture must evaluate a broad range of outputs, such as pattern of warming, cloud fraction, etc. Models do a poor job on at least these as well as the TLT data.
> Based on Dessler’s earlier errors about energy balance methods
Citation needed.
Chubbs, I looked in detail at the RSS total water vapor section. It is so typical of climate science and the biases in the field. We have direct measurements of temperature and humidity from instruments, But those measurements are inconsistent with climate models which “suggests” that the measurements are wrong. Of course, climate models models of convection, clouds, and precipitation are clearly wrong but rather than creating better models that agree with the data, you claim the data must be wrong. It’s really sad in my opinion.
Willard, I even think Dessler essentially admitted he was wrong.
DavidY, I even think you haven’t provided a quote for that admission.
Here’s what a quote looks like:
Oh, and do note that I recall that thread, e.g.:
What is two decades prior to the data cutoff of a research paper versus the entire satellite record?
DPY wrote “Chubbs, The Dessler paper appears to be paywalled. Based on Dessler’s earlier errors about energy balance methods that Lewis rebutted very effectively, I’m not going to pay to access the paper.”
Funny how when it comes to Prof Curry we should look at her papers, but when it comes to Prof Dessler, supposed on-line errors/inconsistencies matter.
I don’t want to “rain on your parade”, but I think that you should read this article, before you claim that the recent slowdown wasn’t a real phenomenon.
https://agree-to-disagree.com/alarmist-thinking-on-the-slowdown
Sheldonjwalker if you want to use statistics to argue for the existence of a pause, then you *do* need to investigate the statistical significance of the evidence for it’s existence. AFAICS there is none.
First this…
Debate in a non-hostile environment
Then this…
Alarmist thinking on the recent slowdown is one dimensional
I love it.
dikranmarsupial, I didn’t say that I wanted to use statistics to argue for the existance of a pause. There is plenty of evidence that isn’t purely statistical. I can manage without most statistics (lies, damned lies, and statistics). But I like to use averages, standard deviations, and linear regressions, etc. By the way, I prefer the term “slowdown” rather than “pause”. I think that it is more accurate, and harder for warmists to dismiss/disprove.
In my article, I explain why statistics is not useful when looking at intervals of less than about 15 to 20 years. With short intervals, you can’t prove that there was a pause, and you can’t prove that there wasn’t a pause. But if you want to use statistics and hypothesis testing, then what will you use as the null hypothesis. And are you prepared to accept the null hypothesis, if you don’t get a statistically significant result?
Joshua, “one dimensional” is NOT hostile.
It is a statement of fact.
I can give you hostile, if you would like it.
I don’t always engage in non-hostile debate, but when I do, the first thing I always do is characterize my interlocuturs with pejoratives, and characterize their thinking as one-dimensional.
It is a statement of fact.
Interesting perspective on “fact, ” btw.
I can give you hostile, if you would like it.
If I’m not mistaken, I may have an internet tough guy on my hands.
Joshua, I am not a “tough” guy!
But after being called a “denier” for over 9 years, and being abused by Alarmists and Warmists on a regular basis, I have developed a “shell”.
I prefer being pleasant.
I like the saying, “speak softly, and carry a big stick”.
It’s mostly about removing multiples, reflected refractions and offline reflections and reflected refractions, Paul. Aperiodic coherent noise, or interference if you prefer.
“one dimensional” is NOT hostile.
It is a statement of fact.
I tend to think that a person who is</strong= interested in non-hostile debate would not charcacterize that statement as one-dimensional thinking.
On the other hand…
A person who is notinterested in non-hostile debate….
Well, you get my drift.
Joshua, I meant “one dimensional” literally.
Alarmists are NOT thinking in more than one dimension.
When they look at the recent slowdown, they only consider the warming rate. They don’t look at the length of the recent slowdown in comparison to other climate events. See my article for a more detailed explanation.
First this….
But after being called a “denier” for over 9 years, and being abused by Alarmists and Warmists on a regular basis,
Then this…
I prefer being pleasant.
Anecdote is not evidence. Nor is personal belief, nor the Mk I Eyeball. Have you forgotten that the easiest person to fool is yourself? That’s why we have statistics. So you don’t fool yourself.
So you’ve done the work, demonstrated that the interval concerned is too short to demonstrate the presence or absence of a pause, and yet here you are, claiming that there is evidence of a pause. As Judith would say: Interesting.
And as I pointed out earlier, it’s even harder than that. If you want to claim a pause, the null hypothesis has to be that there was no pause. On that basis, the pause goes from epic fail to the epicest fail ever.
Interesting dpy. So explaining 10% of the discrepancy (assuming it was a real explanation and not just hand-waving) is an effective rebuttal. I wouldn’t recommend trying that in accountancy.
Joshua, would you rather argue about whether “one dimensional” is hostile or not, or discuss global warming.
Can I say, in a non-hostile way, that you seem to be avoiding the topic of global warming.
Sheldon –
Joshua, I meant “one dimensional” literally.
When people call you a “denier,” they might/probably explain that they have used the term “literally” (as in, you “literally” deny that the risks from global warming are alarming).
I could call such a view of “literally” as one-dimensional – because it ignores many dimensions of the context of the exchange taking place.
If I were interested in hostile debate, I might characterize your view of “literally one-dimensional” as one-dimensional. Who determines which dimension of meaning is the literal dimension? I think that the pejorative dimension is a pretty operational (e.g., literal) one here, and your description of you historic legacy reinforces that impression.
Dave_Geologist, I have a Mk 2 Eyeball.
Do you realise that “pauses” can NOT be statistically significant. The T-statistic is calculated from the warming rate. The warming rate for a “pause” is zero. Therefore the T-statistic is zero = not statistically significant.
People who believe in global warming love statistics, because nobody can ever show a statistically significant “pause”. So they claim that there was no “pause”. Nice one.
Have you ever seen my “global warming contour maps”? Based on hundreds of thousands of linear regressions, with the warming rates colour coded. There are some on my website.
Here is one example:
https://agree-to-disagree.com/gistemp-and-uah
Joshua,
YOU deny that global warming is NOT a serious problem.
YOU are a denier.
Using the label “denier” for somebody, instantly makes them your enemy, and they will probably not do anything that you want them to.
How many millions of people are anti-global warming, because they have been insulted.
At least I am still talking, in a reasonably civil way.
Sheldon –
I’m exploring your logic. My understanding of statistics is extremely imited, so I look at other elements to help me evaluate the probablities of the arguments of people that do.
Btw, if you look above, you will see that I tend to refer to the “pause in global warming” as a short-term slow down in a longer-term trend of increase in surface temps only , which is of limited use because it ignores the earth’s largest thermal mass when characterizing “global warming.”
When I have done so, dikran had cautioned me to consider statistical significance, (because, I assume, to not do so might be one-dimensional thinking).
Joshua,
there is more of my logic about statistics, here:
https://agree-to-disagree.com/is-tamino-dishonest
WARNING – it might be too “hostile” for you.
Sheldon –
Using the label “denier” for somebody, instantly makes them your enemy,
Think about that as you engage in non-hostile debate.
How many millions of people are anti-global warming, because they have been insulted.
My guess is zero millions. I think there are preceding factors that better explain how the vast majority of people have formulated their views on global warming. The use of pejoratives, IMO, isn’t causal for the vast majority, but a symptom of the larger causalities in play.
Let me ask you, how many millions of people view global warming the way that they do because someone has called them an alarmist, or characterized their thinking as one-dimensional?
Sheldon –
At least I am still talking, in a reasonably civil way.
I’m going back to bed now (sometimes I wake up at 4 un the morning, take a break from sleeping, and then go back for a second shift of sleep).
But before I do, I’ll ask you to consider how reasonably civil it is to use pejoratives and characterize people’s thinking as one-dimensional. How do you quantify what is reasonably civil?
Sheldon –
Btw,
YOU deny that global warming is NOT a serious problem
It is a good rule of thumb that before characterizing someone’s thinking, you should first clarify what their thinking is.
Joshua,
I can only tell you what being called a “denier” means to me.
It means that the person calling me a “denier” is calling me stupid, and is telling me that what I think, doesn’t matter.
They are not interested in finding out what I think, or why I think it.
They are basically telling me to bugger off.
My reaction, is to treat them the same way that I feel they are treating me.
I am a friendly person, when not being attached. But my version of the golden rule, is
Do to others, what they have done to you.
The normal golden rule has a logical flaw. It favours bullies.
I always try to start out friendly.
Sheldon wrote ”There is plenty of evidence that isn’t purely statistical.”
Specifically, what is that evidence? Note this excludes the eyecrometer, which is essentially just the application of bad statistics.
”With short intervals, you can’t prove that there was a pause, and you can’t prove that there wasn’t a pause.”
That is entirely the point. However, if you are going to claim that there is some unexpected phenomenon that needs explaining, then the onus is on you to “prove” (neither statistics nor science can prove anything) that the phenomenon exists.
”But if you want to use statistics and hypothesis testing, then what will you use as the null hypothesis. And are you prepared to accept the null hypothesis, if you don’t get a statistically significant result?”
The null hypothesis should normally be the opposite of what you are arguing for. Thus if you are arguing that there has been a reduction in the rate of warming, then the obviou null hypothesis is that the rate of warming has remained constant.
https://www.skepticalscience.com/statisticalsignificance.html
I always try to start out friendly.
Err, no. If you start out by posting an article about alarmists. That isn’t starting out friendly, that is being hostile from the outset. You will notice that I haven’t responded in kind.
Sheldon wrote “Do you realise that “pauses” can NOT be statistically significant.”
No, that isn’t true at all. There are several ways in which you could establish statistically significant evidence for a pause, for instance you could show that a null hypothesis of a constant warming rate could be rejected, or you could show that the statistical power of a test for a lack of statistically significant warming was high.
dikranmarsupial,
did you see my reply to Dave_Geologist:
Do you realise that “pauses” can NOT be statistically significant. The T-statistic is calculated from the warming rate. The warming rate for a “pause” is zero. Therefore the T-statistic is zero = not statistically significant.
People who believe in global warming love statistics, because nobody can ever show a statistically significant “pause”. So they claim that there was no “pause”. Nice one.
==========
I worked this out for myself, I suspect that most people don’t understand it. I am prepared to be proved wrong, if somebody can explain my mistake.
This is why the null hypothesis must always be “that there was a slowdown”. Because you can not “prove” the opposite with hypothesis testing.
==========
Some evidence:
Have you ever seen my “global warming contour maps”? Based on hundreds of thousands of linear regressions, with the warming rates colour coded. There are some on my website.
Here is one example:
https://agree-to-disagree.com/gistemp-and-uah
==========
My website is full of heretical articles about global warming. There is something on my website to offend just about everybody (but done in the nicest possible way).
For a first look, try this (one of my early articles about the recent slowdown):
https://agree-to-disagree.com/how-to-look-for-slowdowns
sheldon,
This is just nonsense. Let’s imagine we weren’t actually warming (we are). The trend in global temperature data would then be close to zero with an uncertainty that would depend on the period we were considering. As we collected more data, the uncertainty would decrease. Eventually we could constrain the trend to be close to zero with high confidence.
Now, we are warming. The problem is that if we consider a short time period, then the uncertainty in the trend will be large enough that we can’t reject that it hasn’t warmed (we also wouldn’t be able to reject that it’s been warming rapidly). As we consider longer and longer time intervals, the uncertainty will decrease and the trend will emerge. This is essentially what has happened. People considered short time intervals during which the uncertainty in the trend was high, but if you consider a much longer time interval, there is little evidence to support that the trend during the early 2000s was significantly different from the long-term trend.
Sheldon,
I agree there was a slowdown in GMST in the 2000s on the decadal timescale. Do you agree there has been a recent acceleration on the same timescale?
Dave said:
Keeping track of multiple reflections is all that diffraction entails but we don’t call that coherent noise. That’s the signal of interest, and deconvolving diffraction is the goal to understand the real-world structure when there are few other alternatives. To rephrase my initial comment, one man’s issue is another man’s solution approach, lol.
dikranmarsupial,
I just read your comment, “You will notice that I haven’t responded in kind.”
Yes, I would like to thank you for that.
I am not perfect. I have had more than 9 years of being called a “denier” (and worse). After many attempts to talk to people without name calling, but still being called names, I have probably become exhausted by the whole situation.
For some reason, I am interested in global warming. I like science, maths, and computing. And I enjoy trying to work out what is happening. I always try to listen to what other people have to say, even if I disagree with them
But I find that many people who believe in global warming, are not interested in hearing a different view. In fact, they often become hostile, including name calling. My website, “agree-to-disagree.com”, was an attempt to have non-hostile discussion. It was a failure. Nobody wanted non-hostile discussion.
For easy discussion, we need to have names for groups of people. I prefer to let people choose their own name. I call myself a skeptic. But people who believe in global warming refuse to use that. They label me a denier. This causes more problems.
What group name would you like to be called?
Hyperactive Hydrologist,
you said, “I agree there was a slowdown in GMST in the 2000s on the decadal timescale. Do you agree there has been a recent acceleration on the same timescale?”
Yes, I agree with you about a recent acceleration. If you read my article, then you will see it on the bar chart at the end, following the recent slowdown.
https://agree-to-disagree.com/alarmist-thinking-on-the-slowdown
…and Then There’s Physics,
do you disagree with my statements about the T-test, and not being able to get a statistically signifant “pause” (warming rate = 0).
If only we had a long enough period of time. But we don’t.
Do you trust linear regressions?
The calculated variables are BLUE (best linear unbiased estimates), even in the presence of autocorrelation. So even if a result is not statistically significant, it is still the best estimate that you can get. So the low warming rates calculated for the recent slowdown may not be statistically significant, but they are the best estimates that we can get.
I have never heard many global warming believers admit that we can’t prove that there wasn’t a pause. Or am I not listening to the right people.
Sheldon,
You can’t really prove there is a pause. The point being made is that if one considers a suitably long time interval, there is little evidence to support claims that the warming trend during the early 2000s was significantly different to the long-term trend.
Sheldon,
Can I ask what your point is then?
This is expected on decadal timescales due to internal variability. I have not met any climate scientists that would argue otherwise.
Sheldon wrote: ”did you see my reply to Dave_Geologist:”
Yes, and I explained why it was wrong, there are at least two ways in which you could demonstrate statistically significant evidence for a pause. See my article at sceptical science.
Sheldon wrote ”What group name would you like to be called?”
If you want to talk about science, what need is there for a partisan label?
Sheldon wrote ”Have you ever seen my “global warming contour maps”? Based on hundreds of thousands of linear regressions, with the warming rates colour coded.”
If it is based on linear regressions, that is statistics. You claimed to have non-statistical evidence, let’s see it.
The take-away message from any ‘pause’, ‘hiatus’, or slowdown in warming during the 2000s is that it had no effect on the established rate of warming.
Internal variations are completely dominated by the warming trend on anything longer than a decadel timescale.
@-sheldonwalker
“So the low warming rates calculated for the recent slowdown may not be statistically significant, but they are the best estimates that we can get.”
This is wrong. The whole point of calculating statistical significance and deriving sigma ranges from the variability for short runs of data is to determine the RANGE of the best estimate. There is nothing ‘better’ about the mathematical mean, linear regression or central value from short data, the best estimate we can derive is limited by the statistical significance.
Any claim something WITHIN that range is better than another part of the PDF is a fundamental mistake about the conclusions we can draw from data.
DPY-
“Chubbs, I looked in detail at the RSS total water vapor section. It is so typical of climate science and the biases in the field. …………………………”
I have to strongly disagree with you. I find the RSS document balanced and well reasoned, not biased. Here is what RSS says about the total precipitable water (TPW) measurement:
“Because of the strong water vapor absorption line near 22 GHz, within the microwave range, we can use microwave radiometers to measure columnar (atmospheric total) water vapor. This is a very accurate measurement due to the high signal-to-noise ratio for this measurement. With little diurnal variation, the measurements from different satellites at the same location often agree to within a few tenths of a millimeter.”
Water vapor content is closely related to temperature. If water content is increasing by 1.14% per decade than temperature in the lower troposphere is increasing by roughly 0.23 C/Decade (using 6.2%/K the climate model estimate). However it is important to understand differences between the diurnal variation and distribution of temperature and water vapor in the atmosphere to appreciate what the data is telling us.
Satellites have many sources of error and are not as reliable as surface measurements. One major source of error is diurnal drift. Satellites only make one pass per day over any particular location. The timing of these passes changes very slowly at rates that vary for each satellite,. This is a problem because temperature has a strong diurnal cycle following the solar heating cycle. Since water vapor doesn’t have a strong diurnal cycle it is impacted less by diurnal drift.
Another issue with satellite TLT and TMT is cooling in the lower stratosphere. Models have underestimated this cooling. So an unknown portion of the discrepancy between climate models and satellite TLT or TMT is due to the stratosphere, not the troposphere. TPW on-the-other hand is weighted towards the portion of the atmosphere that contains water vapor: the tropics and near the surface. So, because they provide a different spatial weighting, it is possible that both TPW and TLT are correct, even though their estimated warming rates are different.
Note however, that water vapor could be a better metric for the “hot spot”; because, the tropical “hot spot” is where water vapor will be concentrated vs the mid-troposphere outside of the tropics. In contrast, TLT is contaminated by stratospheric cooling which is unrelated to the hotspot.
So to wrap up, I have exactly the opposite reaction to the RSS document that you do. I found the tone scientific and unbiased and the results generally in-line with climate models and climate science.
Sheldon, you give yourself amd your blog far too much credit. The ‘pause’ was nothing more than another denier meme to replace the other memes that had bitten the dust previously. It is worth repeating for the millionth time that the scientific community by-and-large is now focused on understanding the consequences of anthropogenic climate change on natural and managed ecosystems across the biosphere and on solutions. Process and causation are by now well established. There is no longer any debate in this area, except among those who believe that business as usual is the preferred option. Scientists have moved on.
It’s Christmas Eve, everyone. Let’s try to chill.
Enjoy this interview.
Nope, two different beasts.
Historical Comparison of TLT Trends
http://www.remss.com/blog/historical-comparison-tlt-trends/
Historical Comparison of TMT Trends
My TMT effort from public domain data sources, NOAA STAR, RSS, UAH and RATPAC,
ftp://ftp.ncdc.noaa.gov/pub/data/ratpac/
I saw your reply Sheldon. All it proved was that you don’t understand the t-Test.
Well Chubbs, I may have not stated clearly that I was referring only to the one sentence in the caption for figure 7. The RSS document as a whole is well written and I have no quarrel with it. It does show something important I think about temperature trends in the troposphere and climate models lack of skill.
It is indeed quite possible that both the temperature and water vapor satellite measurements are accurate. That would be my going in position especially as the TMT data seems to agree reasonably well with 3 radiosonde datasets.
My point is simply that GCM’s disagree giving yet another indication that their convection/precipitation/cloud models are quite suspect. This leads me to reject the idea (stated in the abstract for RSS’s paper I quoted above) that because flawed GCM’s say one of temperature and vapor measurements must be wrong that we conclude magically that indirect evidence indicates that TMT may be underestimate warming. That’s not persuasive.
Who can forget Cruz presenting his own temperature chart (Monkers RSS chart mind you) and Adm. Titley’s extremely effective rebuttal of Cruz’s totally bogus graph …
“Senator CRUZ. … I would note this chart on the right, which shows for the last 18 years that there has been no significant warming whatsoever. Now that is directly contrary to what the dataset showed. Now, Dr. Titley, I noticed in your written testimony that you took a moment to address what you described as the so-called ‘‘pause’’ in global temperatures. By the pause, are you referring to
the last 18 years of no significant recorded warming?
Admiral TITLEY. Well, unlike your previous witness, I am not referring to the 1940s. It is—I thought you would like that.
Senator CRUZ. Indeed.
Admiral TITLEY. Yes. The pause is very interesting. As you know, sir, 1998, big El Nino. So it is kind of interesting we start at 18 years. We don’t look at a 15-year dataset or a 10-year dataset or a 20-year dataset. We look at an 18-year dataset. But even if you do that, that is fine. Let us look at that. We have—this is low budget here. I have got to do my own charts.
[Laughter.]
Admiral TITLEY. Thanks, Amanda.
Senator CRUZ. Well, let me take a moment on——
Admiral TITLEY. So here, sir, just to answer your question, Senator. Here was 1998, and here is today. So, on me, I mean, I am just a simple sailor. But it is hard for me to see the pause on that chart. So I think the pause has kind of come and gone.
Senator CRUZ. Do you dispute the satellite measurement?
Admiral TITLEY. Let us not talk about the satellite. Here is——
Senator CRUZ. But, sir—sir, I am asking, do you—I understand that the global warming alarmists don’t want to talk about the satellite data, but I am asking——
Admiral TITLEY. OK, sure. I will talk about the satellite. Let us talk about the satellite measurements. Let us talk about orbital decay. Let us talk about overlapping satellite records. Let us talk about stratospheric temperature contamination. I think Dr. Christy and Dr. Spencer, when they put this out, they had been wrong I think at least four consecutive times. Each time the data record has had to be adjusted upwards. There have been several sine (sic sign) errors. So when—with all due respect, sir, I don’t know which data exactly your staff has, whether it is the first or second or third or fourth correction to Dr. Christy’s data. We used to have a negative trend, then we had no trend, and now we begrudgingly have an upward trend. So looking at those data, you know, it is OK. But here is where we live——
Senator CRUZ. Let me see if—let me see if I can understand. The first argument you gave in response to this, and it is an argument that a number of the global warming alarmists use is they say, well, 18 years ago was El Nino, and it is arbitrary to begin there. And I will confess I don’t understand that argument because we have 18 years of no significant warming. So if you don’t like an 18-year window, we can start in 1999. There is no significant warming for 17 years. If you don’t like a 17-year window, we can start in 2000. Then we don’t have a significant warming. It is true for any date across those 18 years. So I fail to see the significance——
Admiral TITLEY. Actually, Senator, it is not. If you take off that top really big spike and you take that out, you start getting the upward bias, and this is what people do when you start looking at these relatively arbitrary times is you start with a really high number at the left-hand side, and that kind of influences basically your linear trend. So when you start looking at things like every decade, you have an upward trend in the data, and that is from the World Meteorological Organization.
Senator CRUZ. And I would note you asked about the source of the data on the right chart. It is actually not Dr. Christy’s data. It is the Remote Sensing Systems, the RSS data that is up there. At this point, my time has expired. But we are going to have another round.
Admiral TITLEY. Thank you, sir.”
Click to access CHRG-114shrg21644.pdf
(pp. 90-2)
I’ll post a very small followup if necessary. (Like, why are you posting that?)
Rats, missed it by a few seconds RSS 3.3 bogus graph starts a ~1:02:28, before that, it is Christy’s bogus graph.
@dpy6629:
Re: “Just for the record, I will quote the entire abstract from the paper from RSS about their new version. Bottom line, they now have a higher warming rate than most satellite or radiosonde based products”
Oh, give me a break.
This is the 2017 paper you’re quoting:
“A satellite-derived lower-tropospheric atmospheric temperature dataset using an optimized adjustment for diurnal effects”
I already showed you a 2018 report (co-authored by RSS’ Carl Mears) showing that RSSv4 TLT shows less warming than the radiosonde analyses and the re-analyses. I even showed you an image from the report illustrating that:
[Figure 2.7 of page S17 of “State of the Climate in 2017”]
Instead of addressing that, you cherry-pick the older, outdated comparison from the 2017 paper, while ignoring the part of the paper (page 7715) where the RSS team says RSSv4 TLT under-estimates warming.
It’s like you just willfully ignore any information that shows you’re wrong. Amazing. I think you’re beyond rational persuasion by evidence on topics like this.
@dpy6629:
Re: “Atomski, This is getting a little bit over the top. You “select” from Lewis’ lengthy article one statement (on a secondary point) where another scientist disagrees and then ignore the other thousands of statements in Lewis’ article. That’s called quote mining and does not discredit what Lewis says. It casts doubt however on your ability to fairly evaluate these issues.”
Your evasions are becoming tedious.
First, you show you don’t what “quote-mining” is. Quote-mining involves selectively quoting someone in a way that misrepresents what they said. I did not misrepresent what Lewis said, so no quote-mine occurred. You falsely called it a quote-mine, because you can’t actually address what was said, so you look for some excuse to dodge it.
Second, this isn’t a secondary point; this is Lewis’ defense against climate sensitivity estimates based on paleoclimate data. The vast majority of those estimates show a greater ECS estimate than his work does. When I brought those estimates up to him, this is the specific section of the paper he brought up (the section on the LGM):
“By far the best understood and most studied paleoclimate analysis is of the LGM-preindustrial Holocene transition. Estimating ECS from that transition using an energy budget model is a well established approach. Lewis & Curry (2018) shows that, using modern estimates of the forcing and global temperature change, doing so gives and ECS estimate of 1.76 C, in line with the paper’s main ECS estimate.”
http://archive.is/JNlx8#selection-2485.0-2485.400
So it’s ridiculous for you to claim that I “select(ed)” this portion, when it’s the portion Lewis pointed me to. Are you even trying to engage in serious discussion, or just continually evading points you can’t address?
And a happy new year !
Dikran, Dessler claimed his paper showed EBM’s were flawed. Nic supplied a very well reasoned argument why that was obviously wrong. This latest paper Chubbs references is about much the same thing. Therefore I give it little credence. My experience with Dessler is that his work is not worth a lot of my time. That is a lot different than combing his public utterances for inconsistencies which is an even bigger waste of time.
You can waste your time on inconsistencies in Prof. Curry’s vast public opus if you want. I’ll use the time on science instead and learn from sources who have proven their credibility to me in the past.
Dikran, now you are also trying to point out nonexistence inconsistencies in my comments here. Can you just give the inquisitorial thing a rest?
[Let’s agree to disagree, Sheldon. Merry Christmas to you too. -W]
Sanakan, The point here is that you are quote mining for largely irrelevant points. LGM estimates of ECS are of little relevance to Lewis and Curry 2018 or Lewis’ other work. Also just because another scientist disagrees with this one statement means very little. It certainly doesn’t prove your point at all.
The LGM is interesting perhaps because there was a large climate change with essentially zero change in total forcing. The general thought is that this change occurred because of a change in the distribution of forcing and a lot of feedbacks.
If you have something scientifically interesting to say, I’ll read it. Otherwise, I’ll just stop reading your comments.
What you call a vast public opus I call a vast steaming public dump. Talk about a nothingburger of an argument though. Heck, it isn’t even an argument, people say stuff, some people say a lot of stuff. Some people say a lot of wrong stuff. JC is just one example of a person that says a lot of wrong stuff.
In case you missed it, perhaps you could refrain from using the word opus with regards to JC as you have done several dozen times now (and in several threads). It got really old the 1st time you used it. Because it isn’t even a valid argument. File that one under selective perception (cognition).
Oh, and Merry Christmas to one and all.
I check my email and see someone chasing their tail on “the paws”, so much goalpost moving, so much walls of text. So the faux-paws continues to be a useful diversion/red-herring for some, even though it never existed and was actually made up by a journalist somewhere (a point I made previously).
The reason people who know a bit of climate science dismiss “the paws” is because it is irrelevant. There is no evidence global warming stopped, slowed down or what ever during the poorly-defined period of “the paws”.
Re: “The point here is that you are quote mining for largely irrelevant points.
Already explained to you why it isn’t a quote-mine, and why it’s relevant. As usual, you just repeat false, debunked claims you’ve made, while showing little-to-no concern for evidence.
LGM estimates of ECS are of little relevance to Lewis and Curry 2018 or Lewis’ other work.”
You’ve clearly never read Lewis+Curry 2018. And it shows. Go back and read it. The following should get you started:
“The impact of recent forcing and ocean heat uptake data on estimates of climate sensitivity
[…]
Reasonably thorough proxy-based estimates of changes in surface temperature (Annan et al. 2013: 4.0 K; Friedrich et al. 2016: 5.0 K) and forcings (Kohler et al. 2010: total 9.5 Wm−2) are available for the LGM transition. These values imply, using (4), an ECS estimate of 1.76 K (averaging the two surface temperature increase estimates and taking F 2x CO2 per AR5, since the WMGG forcings were derived using AR5 formulae), in line with the median obtained by scaling this study’s ECShist estimate.”
http://archive.is/JNlx8#selection-2595.0-2599.499
“Also just because another scientist disagrees with this one statement means very little.”
I’ll try to simplify this for you again:
1) The vast majority of ECS estimates from paleoclimate data argue are greater than the ECS estimate from Lewis-Curry.
2) To get around this, Lewis cites paleoclimate sources on the LGM, in order to claim that the LGM supports his low ECS estimate.
3) Lewis’ own sources on the LGM argue against his low ECS estimate.
Let me repeat that again for you, to see if you actually get it this time: Lewis’ own sources undermine his low ECS estimate.
Seriously, do you pay attention when people explain things to you?
“If you have something scientifically interesting to say, I’ll read it. Otherwise, I’ll just stop reading your comments.”
Whether you read them or nor is irrelevant, since you’ve already shown you’re unable to make cogent responses. All you can do is invent reasons for dodging evidence. What dikran noted about your evasions is right.
Sanakan, I guess I was giving you too much credit. I just checked and here is the complete paragraph from Nic on the LGM. You conveniently omitted the last sentence and the footnote. That’s pretty blatant and quite unfair.
“Simple calculations in which the global temperature anomaly at the LGM is divided by the total estimated forcing relative to the preindustrial state have long been used to generate estimates of the equilibrium climate sensitivity.20 Although ECS estimates from earlier LGM studies that did so were typically around 3°C, current best estimates of GMST21 and of relevant forcings22 at the LGM imply an ECS estimate of ~1.75 °C. A more sophisticated approach using AOGCM simulations to relate temperature change post the LGM to ECS in the current climate state gave an ECS best estimate of 2.0 °C.”
In a footnote he says the 2.0 was adjusted for missing dust feedbacks. As I recall the Annan ECS number was something like 2.2 or 2.3 if my memory serves. I really doubt that Nic misstates anything in the references he gives.
@dpy6629
Re: “There is a lot more on clouds and water vapor which you can read as I linked to it above.”
Your knowledge of clouds seems to boil down to citing non-peer-reviewed blog articles from Nic Lewis, while claiming the literature supports your claims (even though you don’t bother to actually cite the literature). I’m beginning to suspect you don’t actually know the literature on this subject. Let me know when you can cite the literature.
http://archive.is/e5zyt#selection-9007.0-9007.160
@dpy6629
Re: “Sanakan, I guess I was giving you too much credit. I just checked and here is the complete paragraph from Nic on the LGM. You conveniently omitted the last sentence and the footnote. That’s pretty blatant and quite unfair.”
Be honest.
You haven’t read Lewis+Curry 2018, have you? You have not read this paper:
“The impact of recent forcing and ocean heat uptake data on estimates of climate sensitivity”
Your comments show you haven’t read it.
What I quoted for you was the last sentence of section 7 of that paper. The section that follows that is section 8 of the paper, which is the “Conclusion”. Your quote, as far as I can tell, appears nowhere in Lewis+Curry 2018. If you think it does, then cite the section in which it appears. I highly doubt you’ll be able to since, as I said before, you likely haven’t read the paper. Instead it seems you are once again quoting a non-peer-reviewed blog article from Lewis, because you simply refuse to read the peer-reviewed scientific literature, even when that is a peer-reviewed paper Lewis co-authored.
So basically, you confused a non-peer-reviewed blog article with a peer-reviewed paper, and then whined because my quote from the paper didn’t include your blog article quote. Amazing. Just amazing. I’m struggling to take you seriously anymore.
Re: “I really doubt that Nic misstates anything in the references he gives.”
And you likely would have no clue if he did. After all, you likely haven’t even read the paper in which he references them, let alone read the references. If this is the level of scholarship and due diligence you can muster on this subject, dpy, then do better. Actually read the scientific literature.
@-sheldonwalker
“Do I sound like a stupid denier, to you? [Snip. -W]”
You sound like a craps player who after a couple of hours notices that the dice come up double-six TWICE as often as snake-eyes, and the cumulative mean is well above seven. You correctly conclude the dice are weighted, imbalanced to give a biased result.
However in about the third hour of play you throw a score of less than 7 ten times in 15 throws.
This is statistically insignificant and could happen by chance, even with the bias you have detected on the dice.
You therefore might conclude that the dice are still biased despite the ‘real’ results of those throws. This would be backed up by special weight measurements of the dice between throws that shows the imbalance or weighting of the dice is constant despite the chance pattern of results. Just as OHC, sea level rise or land ice-mass balance shows near constant energy imbalance independently of the land surface temperature. The variation in the pattern of throws (temperatures) is just chance overlaying the (warming) bias.
Or you could claim that the short run of results justifies an assertion that the bias (rate of warming) has changed despite the clear statistical evidence that such a conclusion is unjustified by the data.
There seem to be a number of people, who believe that the recent slowdown was not a slowdown, if it was caused by “random” factors.
A slowdown is defined by a decrease in the warming rate, not by what caused it.
Please note that I refer to the slowdown as a decrease in the (global) warming rate. I never call it a decrease in the rate of global warming.
I am NOT claiming that the rate of global warming has decreased, or stopped.
The (global) warming rate is the sum of global warming, plus warming or cooling from other factors. The short term variation in the (global) warming rate, is not likely to be caused by global warming and CO2.
I believe that the recent slowdown was a short term event.
Click to access briefing-note-on-climate-sensitivity-etc_nic-lewis_mar2016.pdf
An honest person looks at someone’s relevant work before attacking them.
Peer reviewed literature is perhaps less reliable than you realize. Recall Resplandy? where Nic found some obvious and quite significant errors and the authors agreed. Nic did that in a blog article. The post here highlights a paper that claims that perhaps 200 recent peer reviewed climate science papers have fallen prey to confirmation bias.
Quoting short sentences from the literature as you do is not balanced or reliable. I’m interested in more balanced views.
DPY it is a shame you cannot acknowledge your inconsistency in looking for the errors/inconsistencies in what Prof. Dessler has said, whilst ignoring the errors/inconsistencies in what Prof Curry has said. The rhetoric about inquisitions is just that, if you don’t want your inconsistencies pointed out, don’t draw attention to them.
Sheldon “There seem to be a number of people, who believe that the recent slowdown was not a slowdown, if it was caused by “random” factors.”
If it is just a artefact of “random factors”, why is it worth discussing?
BTW the factors are not really random. Foster and Rahmstorf showed that the apparent slowdown can be explained by physical processes we already know about – ENSO and volcanic forcing.
dikranmarsupial
you said, “If it is just a artefact of “random factors”, why is it worth discussing?”
So that we assess a future possible slowdown, better than we have done with the recent slowdown. If we know what the random factors are, and their probable size, then we can hopefully make an intelligent decision about the non-random factors.
I personally believe that the recent slowdown was caused by ocean cycles, like the PDO, and the AMO. There are climate scientists who say the same thing.
Sheldon what is the importance of future apparent slowdowns? Easterling and Wehner (2009) showed that these kinds of apparent slowdown happen from time to time in both the observations and in the output of climate models. They tell us nothing about centennial scale climate change, which is the key long term concern.
Sheldon,
It is called natural or internal variability and is nothing new. Climate scientists have been researching these issues for decades.
Sheldon wrote “But I have more faith in linear regressions, than I do in many of the other more complicated statistical methods (including hypothesis testing). ”
Assessing the statistical significance of the regression coefficients is a key component of regresion methodology, and is discussed in most undergraduate textbooks on linear regression.
dikranmarsupial,
things that are random are still important. If the warming rate starts decreasing, then we need to decide whether this is due to random factors, or if there is a non-random factor responsible (or it could be a combination of both). It is hard to know what is non-random, unless you understand what could be random (and vice versa).
The calculated variables from a linear regression are BLUE (best linear unbiased estimates), even in the presence of autocorrelation. So even if a result is not statistically significant, it is still the best estimate that you can get. So the low warming rates calculated for the recent slowdown may not be statistically significant, but they are the best estimates that we can get.
Sheldon as I said, ENSO and volcanic forcing are able to explain the observed apparent “slowdown” (a regression analysis can be used to demonstrate that). But we already knew about both of those things, so why is it interesting?
If a non-significant result is the best statistics can give you, then that means there is not sufficient evidence to assert that there is something there that needs explaining.
dpy,
I think we’re all well aware of the weaknesses of peer-review. I don’t think anyone here thinks peer-review doesn’t have any problems. As far as this post is concerned, it’s certainly not about 200 papers that have fallen prey to confirmation bias; there were simply ~200 papers that discussed the “pause”. Something to bear in mind is that for a long time now many have argued that the “pause” was not actually real; at best it was a slowdown in surface warming and there is – in fact – little evidence that anything particularly unusual happened (we know that not all periods are the same and there is little to suggest any significant change to the long-term warming trend). In the case of Nic Lewis, he seems to argue strongly in favour of the results of his own studies despite there being many reasons to indicate that they might be biased (they might not be, but his analysis cannot demonstrate this to be the case). As far as I’ve seen, he has yet to provide a physically-plausible argument as to why the ECS is less then 2K; when we consider all the process associated with surface warming it seems more likely that it is above 2K, than below 2K.
Anyway, it’s Christmas Day, so I’m mostly going to be offline most of the day. Hope everyone has a good break and let’s try to keep it constructive.
dikranmarsupia,
I don’t believe that ENSO and volcanic forcing can explain the recent slowdown by themselves. They may have been partly responsible. Didn’t Fyfe et al say that it was due to ocean cycles?
Part of the reason for statistical insignificance, is noise (temperature data is noisy). We need to try and reduce the noise (by improving our knowledge), so that we can see the signal more clearly.
Imagine that you are walking outside on a foggy night. You think that you might have heard a bear coming towards you, but you can’t see the bear because of the fog. The noise might not have been a bear. You don’t have statistically significant evidence that it is a bear. Do you ignore the possible threat, or do something?
I believe that all knowledge, scientific or otherwise, is potentially valuable. You don’t necessarily know in advance, which bit of knowledge will be valuable, and which bit won’t be valuable.
Sheldon, Foster and Rahmstorf demonstrated that.
Well of course the statistical insignificance is due to noise, that is the point. Foster and Rahmstorf reduce the “noise” using regression to eliminate the plausible effects of ENSO and volcanic forcing and the apparent slowdown dissapears. Thus the slowdown can be explained by ENSO and volcanic forcing.
Wow, compare Figure 5 in the Nik Lewis document dpy linked and Figures 3 and 4 in Lewandowsky et. al. NL’s figure shows a large discrepancy between climate models and observations, while Lewandowsky et. al. figures show very good agreement. Adding a couple of years of observations explains some of the differences, but NL’s choices also play a big role.
[Refers to a deleted comment. -W]
[Refers to a deleted comment. -W]
Sheldon. No, you have it the wrong way round. If someone wants to assert there has been a slowdown, the onus is on them to show that something has changed. You should listen to Tamino, he is a very good statistician, so it is worth considering the possibility that when it comes to statistics, perhaps he is right and you are wrong.
@-sheldonwalker
“I believe that the recent slowdown was a short term event.”
Do you have an equal belief in the recent short term speedups in warming ?
In the 8 years between 2008 and 2016 the temperature rose ~0.5 deg C, or five times faster than the long-term trend.
There seems to be a disproportionate interest in statistically insignificant slowdowns in the rate of warming over a decade or so, compared with the attention paid to the obvious examples of warming that are far faster than the established trend.
Inevitably there will be a preference for aspects of the climate behaviour that could indicate less warming via slowdowns, than more warming from speed-ups. But just because we prefer there to be a slowdown in warming should not mean we ignore the clear evidence it can also speed up.
@dpy6629:
Re: “An honest person looks at someone’s relevant work before attacking them.”
You’re not fooling any sensible person. You clearly dodged the question I asked you, and we all know why: it’s because you never read the paper.
Let me know when you can finally muster a response to the question (though I won’t hold my breath).
You haven’t read Lewis+Curry 2018, have you? You have not read this paper:
“The impact of recent forcing and ocean heat uptake data on estimates of climate sensitivity”
Happy Christmas all…
Atomsk, I’m greatly enjoying your commentary 🙂
Thanks as ever to ATTP for all the fish & W for the patient shepherding.
Love light & peace
BBD
Sheldon Walker’s mind seems to be read-only. He’s here peddling the same ol’ hoke, after being definitively shown the error of his ways on RC.
Whatever — happy Dies Natalis Solis Invicti to all. It’s the real reason for the season 8^D!
Axial tilt for the win.
@BBD
Re: “Atomsk, I’m greatly enjoying your commentary”
You’re welcome. I sometimes like challenging folks like dpy6629, to see if they have any honest interest in the science. Their responses betray the kind of reasoners they are.
“Atomsk, I’m greatly enjoying your commentary”
he’s a grinder. if you get something wrong he will grind you to bits.
It takes all styles, I like watching him grind.
Atomsk, you have motivated me to read Lewis and Curry, just so I can properly follow the conversation. Merry Christmas!
Chubbs, if you read carefully, I think Nic’s Figure 5 is for GHG forcing only model runs. This section of his writeup is in my view not the strongest argument he makes. He also says: “It is well known that, in the mean, the evolution of GMST in CMIP5 models from the start of their historical simulations (1850 or 1860) matches observations well until the early 2000s.”
Generally, CMIP5 is only a prediction from 2005 and a hindcast before that. Further uncertainties in forcings are according to Schmidt not insignificant. I personally am really tired of looking at global mean temperature charts comparing with climate models. I just don’t see much point since there is probably conscious or unconscious tuning involved.
My earlier argument based on fundamental numerical analysis is that since truncation (RANS simulations use orders of magnitude finer grids) and sub grid model errors are larger than the deltas in the fluxes we are modeling, there is little reason to expect skill on a measure that was not known to the modelers and/or used for tuning. These have begun to emerge as climate scientists searched for the reason why model ECS is higher than the historical observations would indicate. This has led to documenting of things like the pattern of warming being wrong. Since the patterns of forcings and warming are critical to predicting future states, this I think pretty much shows that the models are not really scientific evidence of much.
More interesting are measures that are not so well known to everyone in the field as they may give a better idea of real skill.
ATTP, I think Nic’s argument is that the feedback buildup you want is very uncertain with regard to clouds ad the many other uncertainties. This would suggest to me that such analyses have large uncertainties.
One can argue that Lewis seems to make a lot of choices that influence the result and people like Dessler actually say they really don’t understand all those choices. However, since Nic entered the field, most of the papers by other people using EBM’s don’t disagree that much with Lewis and Curry.
Merry Christmas!
The “no hiatus” papers seem to be popular for 2018 …
Causes of irregularities in trends of global mean surface temperature since the late 19th century
http://advances.sciencemag.org/content/4/6/eaao5297.full
(open access, discusses 3 slowdowns and 2 speedups, the entire GMST record, all explained)
Abstract
“The time series of monthly global mean surface temperature (GST) since 1891 is successfully reconstructed from known natural and anthropogenic forcing factors, including internal climate variability, using a multiple regression technique. Comparisons are made with the performance of 40 CMIP5 models in predicting GST. The relative contributions of the various forcing factors to GST changes vary in time, but most of the warming since 1891 is found to be attributable to the net influence of increasing greenhouse gases and anthropogenic aerosols. Separate statistically independent analyses are also carried out for three periods of GST slowdown (1896–1910, 1941–1975, and 1998–2013 and subperiods); two periods of strong warming (1911–1940 and 1976–1997) are also analyzed. A reduction in total incident solar radiation forcing played a significant cooling role over 2001–2010. The only serious disagreements between the reconstructions and observations occur during the Second World War, especially in the period 1944–1945, when observed near-worldwide sea surface temperatures (SSTs) may be significantly warm-biased. In contrast, reconstructions of near-worldwide SSTs were rather warmer than those observed between about 1907 and 1910. However, the generally high reconstruction accuracy shows that known external and internal forcing factors explain all the main variations in GST between 1891 and 2015, allowing for our current understanding of their uncertainties. Accordingly, no important additional factors are needed to explain the two main warming and three main slowdown periods during this epoch.”
The Life and Death of the Recent Global Surface Warming Hiatus Parsimoniously Explained
https://www.mdpi.com/2225-1154/6/3/64/pdf
(open access)
Abstract
“The main features of the instrumental global mean surface temperature (GMST) are reasonably well described by a simple linear response model driven by anthropogenic, volcanic and solar forcing. This model acts as a linear long-memory filter of the forcing signal. The physical interpretation of this filtering is the delayed response due to the thermal inertia of the ocean. This description is considerably more accurate if El Niño Southern Oscillation (ENSO) and the Atlantic Multidecadal Oscillation (AMO) are regarded as additional forcings of the global temperature and hence subject to the same filtering as the other forcing components. By considering these as predictors in a linear regression scheme, more than 92% of the variance in the instrumental GMST over the period 1870–2017 is explained by this model, in particular, all features of the 1998–2015 hiatus, including its death. While the more prominent pauses during 1870–1915 and 1940–1970 can be attributed to clustering in time of strong volcanic eruptions, the recent hiatus is an unremarkable phenomenon that is attributed to ENSO with a small contribution from solar activity.”
@Willard
Re: “I predict that global average 2018 surface temperatures won’t be ‘top five’, i.e. cooler than the last few years.”
I think it goes without saying, but Curry was wrong in predicting that. For instance, as of December 21, 2018, the Japanese Meteorological Association has 2018 ranked as the 4th warmest year in their instrumental record:
https://ds.data.jma.go.jp/tcc/tcc/products/gwp/temp/ann_wld.html
Of course, I don’t think Curry’s false claim will cause her legion of fans to change their mind, anymore than her numerous previous mistakes, distortions, paranoid claims, etc. caused her fans to change their minds. It’s amazing how Curry can run a business based on making predictions, while continually making false predictions:
“Continue to build my business Climate Forecast Applications Network, with research focused on sub seasonal, seasonal and interannual prediction. Working with new clients on decadal and century scale climate issues.”
http://archive.is/WWP98#selection-255.0-259.149
“These have begun to emerge as climate scientists searched for the reason why model ECS is higher than the historical observations would indicate.”
References that don’t include NL/JC (We’ve been there done that many a time already) please. TIA
Somehow this too reminds of the faux pause, AR5 dropped from 2C to 1.5C (ECS lower likely limit), I fully expect AR6 to go back to AR4 (or stay the same but with more them 66% between 1.5C and 4.5C).
No matter how many papers NL writes, he still counts as one single technical opinion to the IPCC authors (hint get more credible coauthors than JC).
No Steven, he knows how to avoid the more important issues raised. Instead, he fixes on a minor point (in this case Annan’s LGM ECS number) and then refuses to discuss anything else, all the while hurling ad hominems. He of course doesn’t know if I have or have not read Lewis’ paper. I linked to Nic Lewis’ blog article which is more complete than the paper in any case.
Bottom line: Nic is fully honest about and explains why Annan’s estimate is higher than a simple ratio of forcing to temperature change. Sanakan was trying to cast doubt on Lewis’ body of work by misrepresenting it on a minor point. EBM estimates of TCR and LGM estimates are two different lines of evidence.
And the quoting of single sentences from papers as if they were all correct is really insulting. There are lots of papers out there saying many different things. If you lack understanding of modeling or science, you use the proof texts you select that support your points.
This is not science or even an honest discussion.
dpy
I’m as sure that feedbacs net positive as I am sure that I live in an interglacial.
Beyond equilibrium climate sensitivity
Click to access knutti17a.pdf
Figure 4 | Illustration of feedbacks changing in result to various boundary conditions. a, Top-of-atmosphere radiative imbalance as a function of global mean surface temperature (annual mean values) for CESM (dark grey and different timescales highlighted in colours) and CMIP5 models (light grey), illustrating the change in the global feedback over different time periods and the implications on equilibrium climate sensitivity. b, Conceptual illustration of the different processes, boundary conditions and forcings that can cause such changes in the global feedback parameter and climate sensitivity (slope and intercept of the line, respectively).
Figure 5 | Illustrative example of combining multiple constraints for climate sensitivity. Overall PD Fsare the products of three PDFs based on the historical warming, climatological constraints on mostly GCMs, and palaeoclimate. Grey ranges at the top indicate a ‘likely’ (66%) and ‘very likely’ (90%) combined range. a,b, Constraint based on an optimistic interpretation of uncertainty ranges and assuming full independence (a) and on inflated ranges (b) to account for structural uncertainties, and with historical estimates inflated and scaled up to account for observation biases, and feedbacks varying from historical to future and across forcings. See Methods for details.
Andrews, T, Gregory, J M and Webb, M J, 2015. The dependence of radiative forcing and feedback on evolving patterns of surface temperature change in climate models. J. Climate 28, 1630–1648
“Experiments with CO2 instantaneously quadrupled and then held constant are used to show that the relationship between the global-mean net heat input to the climate system and the global-mean surface air temperature change is nonlinear in phase 5 of the Coupled Model Intercomparison Project (CMIP5) atmosphere–ocean general circulation models (AOGCMs). The nonlinearity is shown to arise from a change in strength of climate feedbacks driven by an evolving pattern of surface warming. In 23 out of the 27 AOGCMs examined, the climate feedback parameter becomes significantly (95% confidence) less negative (i.e., the effective climate sensitivity increases) as time passes. Cloud feedback parameters show the largest changes. In the AOGCM mean, approximately 60% of the change in feedback parameter comes from the tropics (30°N–30°S). An important region involved is the tropical Pacific, where the surface warming intensifies in the east after a few decades. The dependence of climate feedbacks on an evolving pattern of surface warming is confirmed using the HadGEM2 and HadCM3 atmosphere GCMs (AGCMs). With monthly evolving sea surface temperatures and sea ice prescribed from its AOGCM counterpart, each AGCM reproduces the time-varying feedbacks, but when a fixed pattern of warming is prescribed the radiative response is linear with global temperature change or nearly so. It is also demonstrated that the regression and fixed-SST methods for evaluating effective radiative forcing are in principle different, because rapid SST adjustment when CO2 is changed can produce a pattern of surface temperature change with zero global mean but nonzero change in net radiation at the top of the atmosphere (~−0.5 W m−2 in HadCM3).”
Oops, Correct Figure 5 …
[Playing the ref. -W]
Accounting for Changing Temperature Patterns Increases Historical Estimates of Climate Sensitivity
https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2018GL078887
Timothy Andrews Jonathan M. Gregory David Paynter Levi G. Silvers Chen Zhou Thorsten Mauritsen Mark J. Webb Kyle C. Armour Piers M. Forster Holly Titchner
First published: 30 July 2018
“Abstract
Eight atmospheric general circulation models (AGCMs) are forced with observed historical (1871–2010) monthly sea surface temperature and sea ice variations using the Atmospheric Model Intercomparison Project II data set. The AGCMs therefore have a similar temperature pattern and trend to that of observed historical climate change. The AGCMs simulate a spread in climate feedback similar to that seen in coupled simulations of the response to CO2 quadrupling. However, the feedbacks are robustly more stabilizing and the effective climate sensitivity (EffCS) smaller. This is due to a pattern effect, whereby the pattern of observed historical sea surface temperature change gives rise to more negative cloud and longwave clear‐sky feedbacks. Assuming the patterns of long‐term temperature change simulated by models, and the radiative response to them, are credible; this implies that existing constraints on EffCS from historical energy budget variations give values that are too low and overly constrained, particularly at the upper end. For example, the pattern effect increases the long‐term Otto et al. (2013, https://doi.org/10.1038/ngeo1836) EffCS median and 5–95% confidence interval from 1.9 K (0.9–5.0 K) to 3.2 K (1.5–8.1 K).
Plain Language Summary
Recent decades have seen cooling over the eastern tropical Pacific and Southern Oceans while temperatures rise globally. Climate models indicate that these regional features, and others, are not expected to continue into the future under sustained forcing from atmospheric carbon dioxide increases. This matters because climate sensitivity depends on the pattern of warming, so if the past has warmed differently from what we expect in the future, then climate sensitivity estimated from the historical record may not apply to the future. We investigate this with a suite of climate models and show that climate sensitivity simulated for observed historical climate change is smaller than for long‐term carbon dioxide increases. The results imply that historical energy budget changes only weakly constrain climate sensitivity.”
Say bye-bye to 1.5C
[Please chill. -W]
the questions of the “pause” also show up in questions of emissions. as in, nigel at Real Climate says:
“I think we are in a difficult situation because the possible small decrease / flattening in emissions over about 2015 – 2017 looks too small to me to show up in the atmospheric levels yet. So right now we just don’t know with 100% certainty if we are making a difference.
I also asked the question of anyone who knows how much would we have to decrease emissions for it to register above the noise in atmospheric levels, and what would the time delay be?
I have had a brief look at how emissions are measured and the information collated and assessed and my gut feeling is the emissions trends over say a three year plus period are reliable, certainly not 100% proven but certain enough to have some confidence, but the exact numbers dubious because of politics (as you say), and just normal errors measuring such a complex thing. However it’s interesting that 2015 – 2017 reported a flattening of emissions and the 2018 an increase so the increase wasn’t “hidden” by politics.”
I take nigel at his word and assume that he is asking in good faith. I think the answers are:
1. The small decrease/flattening of emissions from 2015-2017 will never show up in the atmospheric levels because the amount of atmospheric change that arose from the “pause” in emissions is less than the noise level in the number set generally.
2. As to when/how fast a change in atmospheric concentration would be detected in response to a reduction in emissions that is well above the noise level, the answer would be that the related change in atmospheric concentration would be almost immediate, as in, a month or less.
I think nigel is demonstrating the scaling problem that folks have with the amount of our emissions, changes in the carbon cycle and the relative size of changes related to our fluctuating emission numbers.
Tamino, you are good/great with numbers. Is there a way to help folks like nigel understand the relative scale of these number sets?
I think some folks get confused and think everything is fine if we drop emissions by about 2% because the CO2 rose by about 2 ppm. So, 2% should knock out the 2 ppm, right? You divide by two, and voila! everything good.
thank you for a solid year of dumbing down the numbers for those of us who can’t deal with anything above a number squared or cubed. In the land of the blind, the one-eyed person…
Cheers
Mike
Yes Everett, that last paper you found is what I mean. Climate models get the pattern of SST warming wrong over the historical period. The obvious conclusion is that they lack skill on multi-decadal time scales.
And then they assert that if we assume that models are right in the long term, observational estimates of ECS are biased low. Most people would find that strange to say the least. I would think a more valuable exercise is to determine why the models lack skill.
The obvious conclusion is we only have one representation of the Earth system, out of an infinite number of actual possible past realizations. The mean properties of ESM’s are actually quite skillful for the bulk properties of the Earth system.
The paper itself prescribe the SST pattern to eliminate SST spatial and temporal uncertainties. They have been doing these types of exercises for several decades now, they are called AMIP’s (heck it is stated rather clearly in the abstract). With good reason. One less variable to worry about.
“(1) When AGCMs are forced with historical SST and sea-ice changes the models agree on an effective climate sensitivity (EffCS) of ~2K, in line with best estimates from historical energy budget variations (e.g. Otto et al., 2013) but significantly lower than the EffCS of the corresponding parent AOGCMs when forced with abrupt-4xCO2 (~2.4 – 4.6K for the corresponding set of models).
(2) The lower historical EffCS relative to abrupt-4xCO2 is predominantly because LW clear-sky and cloud radiative feedbacks are less positive in response to historical SST and sea-ice variations than in long-term climate sensitivity simulations. This is an example of what is called a ‘pattern effect’ (Stevens et al., 2016), and is consistent with process understanding that suggests lapse-rate and low-cloud feedbacks vary most with SST patterns, especially those in the tropical Pacific ascent/descent regions which have large impacts on atmospheric stability (Zhou et al., 2016; Ceppi and Gregory, 2017, Andrews and Webb, 2018).
(3) The models agree that the most recent decades (e.g. 1980-2010) generally give rise to the most negative feedbacks (lowest EffCS). Hence the pattern effect will be largest for estimates of feedbacks and EffCS based on the satellite era. This is a period when the eastern tropical Pacific and Southern Ocean, regions important for the pattern effect, have been cooling, but are not expected to continue to do so in the long-term response to increased CO2 (e.g. Zhou et
al., 2016).”
Just another nail in the EBM coffin which has no predictive skill whatsoever. 🙂
The climate debate is very unbalanced.
Where are all the Alarmists ??
A putative pause, or slowdown in the trend gets all the attention from Lukewarmers, and even initiates research, which reveals it is statistically insignificant and a result of how energy is distributed in the oceans with a small contribution from reduced solar.
Where are the Alarmists raising the issue of the very rapid jumps in temperature just before and after the ‘Paws’ ?
Climate sensitivity attracts much attention, but only at the lower end of the range. A credible if outlier scientist argues for the low end, but despite a clique of Lukewarmer fanbouys has only managed to get the low end to be included as credible, not any more likely than the rest of the estimates from different methods and sources.
Where are the Alarmists pushing for 4.5C as the likely value and advocating the range should be extended to 6degC/2xCO2 ?
Lukewarmers point to modulz showing a different pattern of ocean heat distribution that leads to faster warming. They claim this ‘error’ refutes any chance that warming could be any faster.
Where are the Alarmists claiming that modulz are predicting a future shift in climate, driven by changing patterns of ocean uptake, that may be underestimating the future rate and magnitude of warming ?
Lukewarmers still assert that SOME of the observed warming may be the result of ‘Natural Variation’ and therefore the observed warming cannot ALL be attributed to CO2 levels.
Where are the Alarmists pointing out that the best evidence is that any Natural Variations have reduced the warming from rising CO2, the only dispute should be how much of AGW warming was negated, not if any warming could be attributed to other causes.
In every field where AGW could have an impact the preponderance of the attention is to the Lukewarmer options of slower warming, less warming and smaller impacts.
Where are the Alarmists ?!
“Where are all the Alarmists ??”
in practice they are lukewarmers. walks like a duck, quacks like a duck. it’s a duck for all practical purposes.
In case A you have actual lukewarmers, go slow types
In case B you have alarmists arguing with other alarmists , the outcome of which is ?? go slow.
I mentioned that keeping track of multiple reflections can be a signal of interest.
Dave said:
Nope, two different beasts.
Consider that a standing wave stretching across the equatorial Pacific is conceptually due to multiple interfering reflected waves. Deconvolving that is the goal to a physical understanding. So they aren’t different beasts at all when considered in a climate context.
Izen asks, “where are all the alarmists”? We are the ones who make up 97% of the scientific community and more probably like 99%, and who are the ones literally begging governments to take immediate action to mitigate GHG emissions and have been doing so for the past 20 years or more while Rome continues to burn and we pass one tipping point after another. We are the ones pointing out repeatedly that we are pushing complex adaptive systems towards collapse, with human civilization in its wake. We are the ones exposing the fraudulent ‘science’ of deniers, and debunking each and every new meme they switch to in a never-ending attempt to do nothing about AGW.
There is no ‘debate’ among the vast majority of scientists over the looming calamity that awaits us as we continue to procrastinate. I really hate it when people refer to the ‘climate debate’ without being more specific. For the scientific community as a whole the ‘debate’ is over what should be done to deal with the greatest threat to face civilization in human history; for deniers, most of whom are laymen, the ‘debate’ is whether it is warming at all, and if so, if humans are the primary driver and if it represents any kind of threat that requires a response (most of whom will say no’ camouflaging their ideological biases with ‘science’). And whereas the actual scientific and response ‘debate’ among scientists is taking place in universities, research institutions and through the peer-reviewed literature, the pseudo-scientific/business-as-usual ‘debate’ among deniers is taking place primarily through the mainstream and social media. Two very uneven playing fields.
The word ‘debate’ therefore is one laden with multiple interpretations, much like the term ‘sustainable development’. Both are used in very different ways depending on the perspective of the person using them.
dpy,
Yes, I realise that one way in which the ECS could be less than 2K is if the cloud feedback is negative. Most people who work on this, however, regard this as unlikely. As far as the pattern effect is concerned, it seems possible that the pattern of sea surface warming has lead to slightly reduced feedbacks compared to what might have been expected. There’s no guarantee that this will continue. Hence we might expect warming to accelerate as we approach equilibrium. It might not, but I’ve seen no compelling argument that demonstrates that this is likely.
Given that we’ve probably gone over this ground a number of times before, maybe we could reflect on whether or not it is actually worth doing so again?
dpy:
“Chubbs, if you read carefully, I think Nic’s Figure 5 is for GHG forcing only model runs. This section of his writeup is in my view not the strongest argument he makes. He also says: “It is well known that, in the mean, the evolution of GMST in CMIP5 models from the start of their historical simulations (1850 or 1860) matches observations well until the early 2000s.””
The choice of GHG forcing for Figure 5 is puzzling, and makes the graph misleading. NL’s quoted statement hasn’t aged very well. With the end of the so-called hiatus models and observations are now in good agreement. That is the whole point of Lewandowsky et al and this blog.
Finally your point regarding tuning is not supported by observations, the CMIP5 forward projections are getting better with time, not worse, as natural variability evens out.
Well, I did say “in seismic processing” Paul. And multiples and diffractions remain stubbornly different beasts. But whatever.
Anyway, a belated merry Christmas to one and all!
Ah, I see the confusion. Not multiple interfering waves. Multiple reflections of the same wave (down-up-down-up, or down-up-a-bit-down-a-bit-up, as opposed to down-up).
If cloud feedback was negative we’d never have got out of the last glacial. Thus perishes the negative cloud feedback meme.
[Enough piling on. -W]
Sure ATTP, no need to keep repeating. I don’t have much of a strong opinion on TCR anyway. I just note that Lewis has a good track record of correcting issues with papers claiming he is lowballing it.
My problem with a lot of the cloud work including Dessler’s is the strong dependence on GCM’s where the cloud and convection models are known to be weak. Models miss pretty badly the latitudinal mean cloud fraction for example. The experts may be right, but the evidence seems weak to me.
Chubbs, It is indeed true that CMIP5 has lower TCR than CMIP3 and thus replicates the Global mean temperature record better. I already explained though why I think that’s a pretty weak measure of skill. On most regional measures, the skill is a lot worse and that’s becoming well documented in the literature. (See the papers cited above on the pattern of warming0.
> no need to keep repeating
Then follows more repetition that evades all the counters.
That’ll be a wrap on “but Nic.” This is a Da Paws thread.
Paul,
Google “seismic coherent noise”. There are several types.
Chubbs “With the end of the so-called hiatus models and observations are now in good agreement. That is the whole point of Lewandowsky et al and this blog.”
As posts Infinitum here in the past have said a temporary surge back touching a boundary hardly qualify as now in good agreement. The current status is, as pre El Niño, not in good agreement at all.
No recurrence of the pause and going out on a very long plank some signs that the current baby El Niño may dissapear.
2019 a boundary forming year.
Sounds like side-effects from the main signal, somewhat akin to an artifact of your measurement. Check the section on characterizing radar clutter in prev cited book.
izen:
Heh. Sheldon’s got your alarmists for you.
Nope, side-effects from things in the ground additional to the things you’re interested in imaging, or the things you were interested in imaging higher in the section that come back to haunt you lower in the section, or things you’re interested in imaging several shot-lines over, that appear on this shot-line.
We’ve drifted a long way from the original topic of signal vs. noise Paul. If you won’t read freely available material on the Web, including from authoritative sources like the SEG or AGU, major contractors like Schlumberger and peer-reviewed publications, all on the first page of Google hits, I’m not going to the expense and distraction of reading a book about radar.
Perhaps it would have been better to stick to a term like “internal variability”, which is agnostic on the question of random vs. chaotic vs. known-but-unwantedly-interfering vs. unknown-unknown.
@-Mal Adapted
“Heh. Sheldon’s got your alarmists for you.”
Somewhere up-thread kenh(?) suggested that the scientists were the Alarmists, because they keep finding out inconvenient truths.
It seems the US administration agrees. The statements in response to the 4th National Climate Assessment dismiss all the scientists involved of having produced something,’not based on facts, ‘not based on data, ‘lacks transparency’, and is just based on modelling the worst case scenario.
Which was on the orders of Obama.
All of this is false, the NCA4 documents many impacts already happening from climate change and all the research and data used is referenced, and often the reliability discussed. The report uses RCP8.5 as the ‘no mitigation’ BAU scenario to estimate damages from climate impacts.
In dismissing the NCA4 as Alarmist for using the worst case scenario, the US executive at least tacitly admits that ANY mitigation by reduction in CO2 emissions is of benefit, because BAU, the path we are on, is the worst case!
current baby El Niño may dissapear.
A thing, should it happen, of complete irrelevancy.
Let’s bring back ozone depleting gases. After all, the 1990 IPCC projections that included them were wrong because they made yet another miserable mistake by including them.
Angech wrote
Evidence required.
dpy – CMIP3 predictions have stood the test of time as have Hanson’s in 1988 and 1981. Must have been good tuning. I think you are making this way too complicated. You can make all kinds of errors in the details and still get the right climate answer. That is one nice feature of modeling chaotic systems. There is no memory of past mistakes.
Dave, Yup, my original comment was “Many researchers are treating the ENSO part as separable and thus compensating the ENSO fraction out, with the result showing a much smoother trend. This is less a statistical approach than elementary signal processing.”, which seems to coincide with the pause-in-warming topic, and nothing to do with seismic processing.
It”s not in a book about radar, but a book with one section on the statistics on EMI clutter and another section that demonstrates how to isolate and remove the ENSO signal. It’s partly a treatise on how to deal with signals and how to deal with noise as appropriate for the analysis context
I think a lot of people forget that there is a whole range of CMIP5 models with a corresponding range of model TCRs and ECRs. Looking at the CMIP5 model mean and comparing it to observations is meaningless. You need to consider the whole range.
The laws of statistics still apply, even if you don’t call it statistics (arguably signal processing is a form of applied statistics).
Izen:
“Somewhere up-thread kenh(?) suggested that the scientists were the Alarmists …”.
Nope. Must have been someone else.
KenH
@-KenH
sorry 😉
Should’a looked.
jeffh
Well Chubbs, you can look at RealClimate’s page on model comparisons to data. They show Hansen’s predictions vs. later observations. The long term trends are also given there. The long term trend of the observations is only a little higher than scenario C and a lot less that scenario B. Don’t get distracted by the “appearance” of the graph.
From Realclimate: “Trends from 1984: GISTEMP: 0.19ºC/dec, Scenarios A, B, C: 0.33, 0.28, 0.17ºC respectively (all 95% CI ~±0.02). Last updated: 20 Jan 2018.”
Chaotic systems almost certainly do not always “forget” previous numerical errors as you assert. This is a question about which very little is known, but what is know is not comforting. There are in fact often many attractors and these have bifurcations and indeed well known regimes where predictability (even for long term statistics) is poor. I may have mentioned subcritical transition to turbulence here previously. An “attractor” may or may not be strongly attractive and can have very high dimension. It’s an area of deep ignorance. Paul Williams has a great talk where he shows that even for the Lorentz “butterfly” if the time step is too large, you get quite different long term statistics from your simulation. There are also some other alarming results about sensitivity of results to time steps that I don’t have at the tip of my fingers.
“The laws of statistics still apply, even if you don’t call it statistics (arguably signal processing is a form of applied statistics).”
That’s an interesting point. If I were to perform a signal processing algorithm to compute tidal tables based on recent tidal gauge data, I could accurately predict the times over the short term. Whatever small error left may be statistically analyzed, but it may not be worth it.
“ Evidence required.“
Roy Spencer UAH
Even before our December numbers are in, we can now say that 2018 will be the 6th warmest year in the UAH satellite measurements of global-average lower atmospheric temperatures, at +0.23 deg. C (+0.41 deg. F) above the thirty-year (1981-2010) average.
2018 is also the 40th year of satellite data for monitoring global atmospheric temperatures.
We are currently working on Version 6.1 of the dataset, which will have new diurnal drift corrections. Preliminary results suggest that the resulting linear warming trend over the 40 years (+0.13 C/decade) will not change substantially, and thus will remain considerably cooler than the average rate of warming across the IPCC climate models used for energy policy, CO2 emissions reductions, and the Paris Agreement.
All the caveats apply, cherry picking etc, but is support for an assertation?
No, not really. Is it reproducible? No. Does it give the range of expected warming of the models for that period (rather than an average centennial warming rate)? No.
PP you are missing the point (as usual). Obviously there are benign tasks where statistical pitfalls (such as overfitting) are likely to be easily avoided. There are others where that is not the case. Most of the good signal processing people I have known/worked with tended to be pretty good at stats as well, and with good reason.
Yet the removal of the ENSO component from the global temperature anomaly by a straightforward differencing of signals is able to effectively remove the pause. That’s pretty impressive for some simple math!
dpy – real-world forcing since 1988 is between Hanson scenarios B and C, reasonably close to the observations. Its not distracting if you use the real-world forcing to evaluate the scenarios.
http://www.realclimate.org/index.php/archives/2018/06/30-years-after-hansens-testimony/
So angech, “pauses” are real and of interest, even though we know they’re statistically insignificant although they do have some real internal-variability causes like La Nina. “But “temporary surges” are ?not real? and not of interest, even though we know they’re statistically insignificant although they do have some real internal-variability causes like recovery from La Nina or from neutral to El Nino. And I presume that the long-term warming trend, which has remained statistically unchanged all through the “pause”, is going to end any time now, even though we’ve been promised that for at least two decades and it hasn’t happened yet. And the latest RSS is right, with the working unpublished as yet, not even un-reviewed for the latest I think – how long is it now since they’ve published their working in the peer-reviewed literature? But the other dozen or so datasets are wrong.
Interesting. I think I see a pattern emerging here – wonder if it’s statistically significant?
PP you do know that is a statistical (regression) analysis, don’t you?
Doesn’t have to be. That’s why I used tidal analysis as an example. I can do an FFT on the tidal gauge data and scale the spectral amplitude coefficients of the first several known constituent factors directly. Or can try a direct linear matrix inversion using 2N data points for the N primary tidal constituent factors. These methods are taught without reference to a statistical analysis.
The removal of a strong daily cycle or annual cycle can also be accomplished by FFT, but instead deleting those spectral components and then reconstructing the waveform. No statistics there either.
I do agree that least-squares is equated with statistical analysis but in some ways that is a quirk of convenience, as it allows one to solve systems of equations with more data (or equations) than unknowns. That is the math of an overdetermined system. Since the Wijkipedia page for https://en.wikipedia.org/wiki/Overdetermined_system does not directly call out a statistical basis but does refer to ordinary least squares, I suggest you edit the page if you feel strongly about it.
” These methods are taught without reference to a statistical analysis.”
As I said, the laws of statistics still apply even if you don’t know about them. I note that you did not acknowledge that the Tamino article (oddly enough) was a statistical analysis and just start a Gish gallop to ignore the point I was making.
Dave_Geo – yes, they actually think a big La Niña in 2020-21, and associated changes in the Eastern Pacific, will be the end of anthropogenic global warming in the 21st century. That the Eastern Pacific cooled from 1983 until 2015 does not register. According to them, the cold side of natural variation has not yet hit. Most of the warming from 1950 to present is natural, and the big freeze is about to commence because scientists are dreadful socialists.
I entered the conversation because I wanted to point out that it doesn’t take much effort to remove the ENSO confounding factor from the global temperature anomaly. And I also wanted to point out that it shouldn’t take much skill in statistics or that a stats guru is even required to do this. In Tamino’s blog post he doesn’t even describe how that part is done — all he does is say that ENSO and volcanic forcing is statistically significant. Everyone that experiences El Ninos understands that it will bump up the average global temperature by a few tenths of a degree. Trying to make it seem obvious should really be the goal. If I had a pile of jelly beans and wanted to remove the red ones, it really is not that challenging a job.
Again, this is what I originally responded to:
So why can’t we tell Angech to just remove the separable ENSO and other confounding factors out (see below courtesy of Tamino again) and tell us again about the pause that he and his fellow skeptics appear to be seeing?
Because to do that requires statistics (at least to do it properly), which is what Tamino did. As I said “For analysis of noisy data, you need statistics.”; I don’t really understand your reluctance to just accept that.
So dpy is still pretending that a non-existent pause can be used as evidence for low climate sensitivity because chaos and teh muddles… This despite the wholly predictable, repeating climate behaviour in the past, eg Pleistocene glacials.
But it’s really not that noisy of data compared to other data sets. ENSO is not noisy and the isolated volcanic disturbances are all known. To me the degree of difficulty is like removing a 60 Hz hum, but for what ever reason you want me to cry uncle and relent.
PP It is irritating that you are playing rhetorical games here “cry uncle and relent.” If you cannot admit that your error fine, however if you keep trying to argue your case, you should expect me to continue pointing out your error.
The fact that you are trying to argue that statistics is unnecessary and use Tamino’s statistical analysis to illustrate your point should at some point have given you pause for thought that perhaps, just perhaps you might be making yourself look silly.
What I am trying to do is reduce the barrier of acceptance in understanding what is happening with respect to discriminating the signals. When you say that only with statistics that you can properly extract the signal, you are raising the barrier. Even Tamino says this “We can subtract the estimate of how these factors have affected temperature, to get an estimate of how temperature has changed apart from these known factors. “. Nowhere in the blog post does he mention multiple linear regression, even though he may have indeed used that. I could have done it by calibrating the ENSO signal to the 1998 El Nino peak and simply subtracting off the scaled time-series factor, providing a pretty good estimate for a skeptic to evaluate.
And I did admit that least-squares is often used in regard to statistical analysis, but I can also use it to deduce unknown values of components in an electrical circuit where I can make some measurements. It’s a general tool that is taught independent of a statistical context, which is where I first learned about it in my educational progression.
PP Everything should be made as simple as possible, but no simpler. If you ditch statistics, then there is no reason not to argue that there has been a reduction in the trend estimate from cherry picked start and end points (which is what angech was doing).
” Nowhere in the blog post does he mention multiple linear regression, even though he may have indeed used that. ”
IIRC he said he was updating a previous study (he also mentioned that the explanatory variables were statistically significant). Wonder what that could have been? ;o)
“I could have done it by calibrating the ENSO signal to the 1998 El Nino peak and simply subtracting off the scaled time-series factor, providing a pretty good estimate for a skeptic to evaluate.”
If a climate skeptic came up with an argument like that, would you be happy with it? It is exactly what e.g. Scafetta and Loehle do. The world is full of spurious correlations and it is best to engage the minimal self-skepticism provided by significance tests (as Tamino did).
If that could not turn into another Dikran c. Web, that’d be great.
You might both like:
Pingback: The significance of the “pause” | …and Then There's Physics
Pingback: Sample Reblog – Global Warming 9-4
Pingback: The Escalator | …and Then There's Physics