Arguments
Recent Comments
Prev 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 Next
Comments 34351 to 34400:
-
Lionel A at 23:21 PM on 11 September 201497 hours of consensus: caricatures and quotes from 97 scientists
Lars, I wonder if Pat and Judy will be annoyed or relieved at being absent?
-
Tom Curtis at 23:19 PM on 11 September 20142014 SkS Weekly Digest #36
Dikran Marsupial @28, the Open Journal of Statistics has a Google Scholar h5 ranking of 5, with a h5-median of 4. That is, in the last five complete years, the journal has just 5 articles which have been cited 5 or more times, and for those five articles, the median number of citations is 7. That compares the mean h-5 ranking of the top 20 statistics journals of 41.45, with a mean h5-median of 65.5.
Such a low h5 index could well be achieved by self citations alone.
It is no wonder that the first thing the website for McKittrick's paper informs you is that it is "An academic publisher". Without that flag, it is dubious that you would notice.
-
Lars Karlsson at 23:15 PM on 11 September 201497 hours of consensus: caricatures and quotes from 97 scientists
And Spencer is the last one of the three against!
-
Dikran Marsupial at 22:56 PM on 11 September 20142014 SkS Weekly Digest #36
Philip Shehan, just to add to the already excellent answers on this, the ridiculousness of McKitricks' conclusion is shown by just plotting the trend for the 26 year hiatus in the RSS data along with the trend for the entire dataset:
You can immediately see that the OLS trend during the hiatus is almost exactly the same as the trend over the whole period. Also there is no obvious evidence of any change in the rate of warming in 1988 (1998 perhaps!), which is what you would expect at the start of a hiatus.
McKitrick's error is in calling a period where the trend is not statistically significant "trendless". A Failure to reject the null hypothesis in a statistical hypothesis test does not mean that the null hypothesis is true, and never has done. To see why this is true, we can consider flipping a coin (which is much easier to analyse).
Say I have (unknown to you) a double headed coin, and you want to determine whether the coin is biased. The first thing to do is to state the null and alternative hypotheses:
H0: The coin is unbiased - p(head) = p(tail) = 0.5
H1: The coin is biased, p(head) != p(tail)
We then observe some flips of the coin, and evaluate the probability (known as the p-value) of a result at least as extreme as the one we observe assuming that H0 is true. This is an important point, you suspect me of cheating and have a biased coin, but to provide self-skepticism, the test by starts by assuming you are wrong and that the coin is unbiased. If the p-value is sufficiently small, often 1 -0.95 = 0.05, then we consider the observations sufficiently unlikely and we "reject the null hypothesis" and can continue with the alternative hypothesis. Otherwise we "fail to reject the null hypothesis" and should keep quiet about H1 for the time being.
Note the purpose of the test is to avoid getting carried away by our enthusiasm for H1, so we have the hurdle of at least being able to reject H0.
Right, say we flip the coin four times, and get heads four times in a row (unsurprising as it has a head on both sides). The p-value (assuming an unbiased coin - which is H0) is 0.5*0.5*0.5*0.5 = 0.0625. This is more than 0.625, so we "fail to reject the null hypothesis", but does that mean that H0 is true? Definitely not, the coin is as biased as it could possibly be!
So what went wrong? The problem is that if you only flip the coin four times, we will never be able to reject the null hypothesis as the p-value can never be smaller than 0.0625. This is because if we don't have enough observations, the test lack statistical power. The power of a test is the probability that the H0 would be rejected if it actually is false. If the test has very high power, then not being able to reject the null hypothesis is evidence that the null hypothesis is true. If the test has low statistical power, then a failure to reject the null hypothesis may be because it is true, or because we don't have enough data to confidently show that it is false.
The bottom line is that a failure to reject the null hypothesis does not imply the null hypothesis is true, unless the test has high statistical power. McKitrick does not evaluate statistical power, but he really should have done! In saying that the period is trendless, he is asserting that the null hypothesis is true. Just plotting the diagram shows this clearly isn't reasonable in this case.
The other key problem is that he is looking (cherry picking) the longest period possible to get a trend that just fails to achieve statistical significance. This means that the observed trend would be a rather unlikely thing to see even if there were a hiatus. Essentially if we had a large number of Earths in parallel universes that were undegound a hiatus (and H0 were true) then about 95% of them would have a trend that is lower than the one we have observed.
This really is a STATS101 issue, and it is dissapointing that this wasn't picked up by the reviewers of a statistics journal.
-
Tom Curtis at 22:12 PM on 11 September 20142014 SkS Weekly Digest #36
Philip Shehan @24, this topic came up at And then there's physics. The key point is that the V-F method used by McKittrick indicates the pause starts at the earliest year such that for that year, and for each following year, the trend to the terminal year is not significant. That permits, however, that there be trends from the initial year to years prior to the terminal year that are significant. In such cases (and it is the case with every temperature set examined by McKittrick) it clearly begs the question to assume that there is a "pause" over the entire period, including over that subperiod in which there is a statistically significant warming trend. That, however, is just what McKittrick does. Indeed, I would go further and say they assymitrical treatment of trends to the terminal year and other trends shows the statistical analysis to be rubbish.
For the record, here are the longest statistically significant warming trends from the year the "pause" started according to McKittrick's method, as determined on the SkS trend calculator:
"RSS: 1990-2010: significant; -2013: not significant (24 years)
UAH: 1993-2010: significant; -2013: significant (21 years)
HadCRUT4: 1995-2011: significant; -2013 not significant (19 years)
NCDC: 1995-2007 significant; 1995-2013 not significant (19 years)
GISS: 1996-2010 significant; -2013 not significant (18 years)"Finally, here is HadCRUT4 running 12 month means from 1970 showing various trend lines. It is plain from the data that it has warmed since 1995 in that data set, and indeed that it warmed through to about 2007. The apparent "hiatus" is almost entirely an artefact of the strong El Nino in 1998 coupled with the record breaking La Nina in 2011-12:
-
Tom Curtis at 21:54 PM on 11 September 20142014 SkS Weekly Digest #36
MA Rodger @16, a light correction. Fig 1a (shown by you) is the regression against ENSO, volcanic record, and solar only. Fig 1b is further regressed against the AMO defined as the linearly detrended NA temperature:
Jim, As I commented above, the result is near linear which is unsurpising given that the AMO used is just the North Atlantic temperatures minus a linear trend. Had the AMO been defined as the North Atlantic temperatures minus total CO2 forcing, the result would have mirrorred GHG forcing over the twentieth century. A paper doing that would have been rejected by AGW deniers as being circular (for good reason), but the same is equally true of Zhou and Tung. (Note, it is not strictly circular. To be strictly circular they would have needed to define the AMO as the linearly detrended global temperatures.)
In fact, the only sensible definition of the AMO from twentieth century data I am aware of is that from Ting et al (2009) (third panel), which is discussed in reference to Tung and Zhou by KR here:
Leaving aside, however, which is the correct AMO signal of the three (in order, linearly detrened NA, NA-Global Oceanic Mean, and NA-NA forcings), it is plain that very different results would be obtained by using either the second or third definition to that obtained by Zhou and Tung.
And finally, seeing as I'm showing figures, here is the CET series at 10, 15 and 50 year running averages. If anybody can show me the consistent approximately 60 year oscillaiton in that series I will be very impressed. Absent that persistent oscillation, there is no reason to think the AMO is a genuine mode of internal variability in global temperatures, rather than an artifact of the forcing history of the twentieth century.
-
MA Rodger at 19:21 PM on 11 September 20142014 SkS Weekly Digest #36
Philip Shehan @24.
This may or may not help but...
I always see four different flavours of "hiatus" being discussed. The first is the David Rose of the Daily Rail version. This version simply shows that the present global temperature is still below the temperature of some cherry-picked temperature of time gone-by. Thus the 'hiatus' is, to be as silly as possible, using HadCRUT4 24 years and 4 months long. The silliness becomes plain when average temperatures over a year or a half decade are compared, rather than single months.
The second flavour is courtesy of Dickie Lindzen who fed the question to Prof Jones via the BBC's Roger Harrabin. Has gobal temperature rise been statistically insignificant over the last X years? The answer given was "Yes, but only just," which I consider to be correct but the wrong answer. The answer should be 'Yes, just like the period of Y years previous to that, and the previous period of Z years before that again. No statistically-significant rise does not mean no significant rise.'' The actual value of X is a function of the rise as well as the size of the wobbles. (Stat significance is not proportional to X but to X^1.5 (or is it X^2.5?) which is why X and Y and Z don't change much.). For the same rate of rise, bigger wobbles means X will be larger. X is usually ~16 years for the likes of HadCRUT (although HadCRUT(T&W) would perhaps be more appropriate to use). Using satellite data UAH or RSS gives you bigger wobbles and so gives you larger X.
The third flavour is the answer to the question "When did the 'hiatus' in global temperature begin?" and I would suggest that happened in 2007. Doing what Ross the Mac did but forwards not backwards so you've got more data and significance is maintained. This method shows that the global temperature record modelled as a linear trend was accelerating (the slope getting steeper) up to that point. See graph of GISStemp regressions two clicks down here. The slope peaks so the rise stops accelerating only in 2007. To say an accelerating linear trend is subject to a 'hiatus' would be an difficult statement to substantiate (although that will not stop fools saying it).
The fourth flavour leads to a more grown-up discussion and it is about when the symptoms that lie behind the 'hiatus' first appear, which is during the half-decade prior to the impact on global temperatures.
-
Dikran Marsupial at 18:42 PM on 11 September 201497 hours of consensus: caricatures and quotes from 97 scientists
dhf it is dissapointing that you appeared to have ignored the responses to your initial post on this thread.
Do you think it is rational for those unable to understand the science of climate change to base their views on the balance of expert opinion on the topic, yes or no? If no, please explain why.
-
Philip Shehan at 16:41 PM on 11 September 20142014 SkS Weekly Digest #36
In the last week the “skeptic" blogs have been agog with Ross Mc Itrick’s statistical analysis that the haiatus goes back 16, 19 or 26 years depending on the data set.
file:///C:/Documents%20and%20Settings/user/My%20Documents/Downloads/OJS_2014082814175187.pdf
Reliably, Watts puts up a graph of the Hadcrut4 data set (1995 to April 2014) from WFT but decides not to put in the WFT trend line which like the data in Mcitrick’s paper indicates a warming trend of 0.0925 °C/decade (the skeptical science trend calculator gives Trend: 0.093 ±0.100 °C/decade (2σ)) but substitutes a cooling slope.
http://wattsupwiththat.com/2014/09/01/new-paper-on-the-pause-says-it-is-19-years-at-surface-and-16-26-years-at-the-lower-troposphere/
My problem with McItrick’s paper is that he appears to be claiming that unlike the skeptical science trend calculator, his data shows that the hiatus is statistically significant. The stats is a bit beyond me , Does anyone have any ideas?
-
Leto at 15:50 PM on 11 September 201497 hours of consensus: caricatures and quotes from 97 scientists
dhf @40... Do you have a point? If so, make it.
-
97 hours of consensus: caricatures and quotes from 97 scientists
Very many are trained through education and profession to recognize and disregard arguments containing logical fallacies. The reputation of individuals and organizations are at risk when they use logical fallacies in their argumentation.
Moderator Response:[JH] You are already skating on the thin ice of sloganeering which is prohibited by the SkS Comments Policy.
Please note that posting comments here at SkS is a privilege, not a right. This privilege can be rescinded if the posting individual treats adherence to the Comments Policy as optional, rather than the mandatory condition of participating in this online forum.
Please take the time to review the policy and ensure future comments are in full compliance with it. Thanks for your understanding and compliance in this matter.
-
97 hours of consensus: caricatures and quotes from 97 scientists
On the other hand, the consensus of evidence . . .
-
Dikran Marsupial at 04:56 AM on 11 September 201497 hours of consensus: caricatures and quotes from 97 scientists
dhf Ironically it is the skeptics that initiated this with their petitions suggesting that there isn't a consensus and that the basic science is seriously questioned amongst scientists. Pointing out that this is not true is not a logical fallacy.
The point is that nobody is seriously suggesting that the existence of a consensus actually has any bearing on the science (actually the causal arrow is in the opposite direction). Those who understand the basic science are unlikely to care about the consensus. How should those who don't understand the science evaluate the issue? I would suggest that following the consensus view of the experts is a fully rational thing to do. That is the point.
-
97 hours of consensus: caricatures and quotes from 97 scientists
dhf - Please be careful with the alternative 'Galileo gambit' fallacy frequently offered by the pseudo-skeptics, that they are iconoclastic geniuses who see everything more clearly than the vast majority of scientists studying the data.
From RationalWiki:
In reality, taking up the mantle of Galileo requires not just that you are scorned by the establishment but also that you are correct — that is, that the evidence supports your position. There is no necessary link between being perceived as wrong and actually being correct; if people perceive you to be wrong, you usually are wrong.
There are far far more cranks than Galileos - the consensus generally reflects the most reasonable interpretations of the data. And it doesn't help 'skeptic' arguments to see that every self-identified climate Galileo holds different, contradictory hypotheses, sometimes with a single pseudo-skeptic expressing multiple contradictions simultaneously.
I will note that the last time the scientific consensus shifted on this topic was mid-20th century, the '50s and '60s, when accumulating evidence began to convince climate scientists that we affected climate change - the shift to acceptance of AGW. Reversing that consensus would take a completely new and convincing set of evidence - and none is apparent.
-
97 hours of consensus: caricatures and quotes from 97 scientists
Please be careful with the consensus argument.
"In argumentation theory, an argumentum ad populum (Latin for “appeal to the people”) is a fallacious argument that concludes a proposition to be true because many or most people believe it. In other words, the basic idea of the argument is: “If many believe so, it is so.”
This type of argument is known by several names,[1] including appeal to the masses, appeal to belief, appeal to the majority, appeal to democracy, appeal to popularity, argument by consensus, consensus fallacy, authority of the many, and bandwagon fallacy, and in Latin as argumentum ad numerum (“appeal to the number”), and consensus gentium (“agreement of the clans”)."
http://en.wikipedia.org/wiki/List_of_fallacies
-
jim10940 at 03:09 AM on 11 September 20142014 SkS Weekly Digest #36
I agree that the contrarian blogs are very bad nonsense. But the 2013 and 2014 papers by Tung were somthing i was not able to work through on my own. The references and discussion KR and Tom provided helped very much! Now back to thesis writing!
-
2014 SkS Weekly Digest #36
jim - Regarding Curry contradicting her own work, see Rabett Run, Curry vs. Curry, where he describes the situation far better than I could.
Christy is, if anything, worse - he habitually presents irrelevant, uncorrected, cherry-picked (Sierra Nevada winter snow extent _only_ while ignoring yearly anomalies and the greater Rockies), or simply deceptive information (graph from Christy), and claims from those that climate change either isn't happening or isn't having an impact. I consider his impact on public policy to be strictly destructive.
I would suggest reading a wider variety of work.
-
2014 SkS Weekly Digest #36
jim - Ah, that explains the Duarte mention. Happens to everyone, I'm really bad at remembering names myself.
We are facing the first clear evidence of a dangerous climate change. However, some of the researchers and some of the media are plunged into a semantic debate about whether the Arctic Sea-Ice has reached a tipping point or not. This all is distracting the attention on the need to develop indicators that warn about the proximity of abrupt changes in the future, as well as on the policymaking to prevent them.
I would consider this a well-reasoned statement against hysteria. I would not consider him to be optimistic regarding climate change, however, as his work has also stated:
Loss of Arctic ice due to anthropogenic climate change is accelerating, with the extent of Arctic sea ice displaying increased variance at present, a leading indicator of the proximity of a possible tipping point. Reduced ice extent is expected, in turn, to trigger a number of additional tipping elements, physical, chemical, and biological, in motion, with potentially large impacts on the Arctic marine ecosystem.
Hardly an optimistic outlook, IMO.
Side note: your mention of "I keep getting exposed to" Curry, Duarte (Jose), and Tung give me the impression that you've been reading denialist nonsense blogs that emphasize those particular positions. You should avoid those - they rot the brain.
-
saileshrao at 02:05 AM on 11 September 2014When their research has social implications, how should climate scientists get involved?
Sapient Fridge@36: Good point, sorry for not cutting those out.
Moderator@31: I'm unable to access the Comments Policy link as it is behind an authentication firewall.Moderator Response:[JH] Click here to access the SkS Comments Policy.
-
jim10940 at 02:02 AM on 11 September 20142014 SkS Weekly Digest #36
KR @ 17 I am in an Agricultural Engineering department which does a lot of work with extention and I attend the Southeast Climate Conference of which John Christy and Judith Curry are prominent. I think in my department, which does a lot of extension work, most members don't want to emphasise climate because frankly our clients (farmers, water managers) don't want to hear about it. It seems adaptive response to current climate variability is an acceptable argument for the extension work but don't talk about climate change.
-
jim10940 at 01:44 AM on 11 September 20142014 SkS Weekly Digest #36
KR @17 this is very embarrasing. I am confuseing my Duarte's. I meet Carlos Duarte and he seemed to downplay the effect the severity of the climate threat, giving reasons for optimism. But I dont even whant to attribute that position to him maybe he was just making the argument.
-
Tom Curtis at 01:38 AM on 11 September 20142014 SkS Weekly Digest #36
jim @13, I am not claiming that is what Tung did. I am claiming it is what you can do with Zhou and Tung's data, and that when you do it falsifies the claim that 40% of warming over the last 50 years was due to internal variability. Further, using Zhou and Tung's data if you accept the 40% claim, you are then required to accept that for a 30 year period from 1974 onward where natural variability accounts for 100% of warming (indeed, more than 100% once ENSO is included). That, however, is strictly inconsistent with the known relative strengths of anthropogenic and natural forcings, and the idea that anthropogenic forcings effect temperature. In other words, Zhou and Tung require the physics of climates to change on a decadal time scale for their regression to make sense.
In Zhou and Tung 2013, they do infact determine a residual after removing solar, volcanic, and ENSO components. The result is close to a straight line, which is no surprise given that they defined the AMO as the North Atlantic temperature minus a linear trend. To the extent that NA temperatures follow global temperatures, it follows that the residual of the regression will be a straight line and overstate the AMO component. The result is close to circular, and of little interest as a result.
-
2014 SkS Weekly Digest #36
jim - WRT Tung, not even the most prominent and respected people can be expected to bat 1000. And he's hardly alone in misidentifying the textbook definition for AMO (linearly detrended N. Atlantic SSTs) as correctly depicting internal variability in the Atlantic; internal Atlantic variation is change over and above what occurs due to the non-linear forcing changes (natural and anthropogenic) over the period under consideration.
Cognitive bias certainly plays a part in interpreting science, particularly in the public policy arena, but it really doesn't have any significant effect on the science, the data, itself - reality is a very harsh critic. Unless you're a fan of Stephen Colbert, who's noted, "Reality has a well-known liberal bias." :)
Judith Curry is a curious case, someone who overemphasizes uncertainties to the point of contradicting her own prior works. Just keep in mind that there are very very few scientists unconvinced by the evidence for AGW, for our role in climate change, and that those very few get overamplified by the blogosphere and media, overrepresented, because they are the only ones to ask for contrary opinions and false (i.e., unsupported) 'balance'. That overrepresentation doesn't make them right, and in fact is a useful heuristic for identifying extremely minority positions.
[ If you're actually a grad student in climate related work I'm very puzzled by your mis-identification of the recently blogging Duarte as an ocean ecologist - to the best of my knowledge he's actually a PhD candidate in Social Psychology, one who appears to have ideological axes to grind, personal issues seemingly disconnected from the work he then criticizes. ]
-
MA Rodger at 01:15 AM on 11 September 20142014 SkS Weekly Digest #36
jim @13.
The MLR analysis carried out by Zhou & Tung (2013) is a single process which takes a data series (D) and a set of functions (Fa Fb Fc etc) and calculates the best fit such that D= xFa +yFb +z Fc + residuals. Zh&T did this without AMO and produced their Figure 1a.
This is all non-controversial. It is when ZH&T repeat the exercise but with AMO (shown in the inset with the de-trended residuals) added to Sol Vol & ENSO and various profiles for Anthro that it all starts to go wrong. But attributing AMO is their whole purpose, so without it there is no MLR analysis.
The strength of AMO's input into HadCRUT (where they get their 40% from) is a simple output from the MLR. You can fire off a whole pile of objections as to why using AMO in the MLR is simple curve-fitting but even when we had KKTung here last year (here and here) to present his version of it, answers were not forthcoming.
-
2014 SkS Weekly Digest #36
You have elevated Jose Duarte to a position he has yet to earn. The kid's loud, confident and wrong.
-
jim10940 at 00:13 AM on 11 September 20142014 SkS Weekly Digest #36
The reason I am pursuing the climate literature so closely is that I am troubled by the role cognitive bias plays in climate science. It seems that those that those that support global warming as severe and imminent threat are classic liberal while those who doubt its severity are libertarian or republican. Now I personally think that global warming is a severe threat but then I'm a classic liberal. If only I were libertarian and believed global warming was a severe threat this would be so much easier!
I am currently a graduate student doing climate related work but i keep getting exposed to prominent scientists that seem to downplay the severity of global warming. i.e. Judith Curry (climate scientist), Jose Duarte (ocean ecologist) and now I suppose in the literature Tung. My explanation is that they may be biased by their social beliefs but than this makes me question my own biases.
-
jim10940 at 23:57 PM on 10 September 20142014 SkS Weekly Digest #36
Tom @11 That was my thought at first was well, that it seemed ridiculous just to plot the increase over the last upward swing of the AMO cycle. But i dont think thats is what Tung is doing. I think he is using the residual after removing the oscillation. I could be wrong, but if he is doing what you claim that would truly be ridiculous. I don't think that would have gotten through the review process.
-
jim10940 at 23:52 PM on 10 September 20142014 SkS Weekly Digest #36
KR @10 the link to posts where Tung laid out his defence were very helpful. I think i now understand what the problem is. The AMO has an 80yr oscillation and the completes 2 cycles in the modern record since 1850. But it also is undergoing a nonlinear increase. I take it that Tung is calling this increasing temperature in the AMO in the recent decades as partially explaining global temperature rather than a response to global temperature. I am not able to get much deeper into it but I think its more likely long term warming (longer than the AMO cycle) is likely to be a response not a driver.
This is so puzzeling to me because Tung does great work. Its very technically adept and his latest paper published in Science is a very important document of the response of the Atlantic to global warming.
I get the sense that his 2013 paper is good but the review process failed in letting him make a such a strong claim. I get the impression that he wanted to make similar strong claims in his 2014 paper but here the review process did not allow that.
-
DerekfromBerwick at 22:48 PM on 10 September 2014Cutting Emissions Pays for Itself
This is not a surprise but what scale of activity is required to see a payback of this order? Would a small town of a few thousand see an impact in terms of health benefits or does the unit size need to be larger? What would be useful to know is how the benefits breakdown as this would allow for a better understanding of how policy changes would impact on a population.
-
Lionel A at 21:27 PM on 10 September 201497 hours of consensus: caricatures and quotes from 97 scientists
I see Richard Lindzen has indeed joined the group, between David Karoly, Keith Shine and Veerabhadran Ramanathan.
-
Sapient Fridge at 17:37 PM on 10 September 2014When their research has social implications, how should climate scientists get involved?
Saileshrao, it is considered bad netiquette to publically post people's E-mail addresses on web sites without their permission as it increases the amount of spam they get. Unfortunately many spammers have programs to "harvest" E-mail addresses from web pages.
I note that you omitted your own E-mail address...Moderator Response:[Dikran Marsupial] Good point, I have redacted the email addresses.
-
Rob Honeycutt at 10:47 AM on 10 September 2014When their research has social implications, how should climate scientists get involved?
Agreed.
-
Tom Curtis at 10:38 AM on 10 September 2014When their research has social implications, how should climate scientists get involved?
Rob, given that saileshrao has been asked to cease and desist, for the sake of fairness, so ought we.
Moderator Response:[JH] Spot on.
-
Rob Honeycutt at 10:31 AM on 10 September 2014When their research has social implications, how should climate scientists get involved?
saileshrao @30... Your post there is little more than speculation used to justify your position. Skeptical Science is a website that expects people to substantiate their positions with data and research. AGU people would, presumably, also be far more compelled to listen to what you have to say if you were to present substantive data rather than just a quote that you interpret to validate what you say.
Having been vegetarian (ovo-lacto) for some 35 years, I'm not completely against your position, but I would expect a far more substantive presentation of facts before supporting what you're doing.
As it stands, AGU does a great deal many other things that justify their stance on "core values" already. If you want to see them add veganism to that list, present hard data on how much of an impact changing their catering would have. And part of that would always entail being prepared to prove to yourself that your position may be weaker than you want to believe.
-
Tom Curtis at 10:22 AM on 10 September 20142014 SkS Weekly Digest #36
jim @7, you quote Tung and Zhao as saying:
"Quantitatively, the recurrent multidecadal internal variability, often underestimated in attribution studies, accounts for 40% of the observed recent 50-y warming trend."
Given the mean 60 year period of the AMO, that is an astonishing claim. The temperature impact of the AMO over 60 years (ie, one full cycle) should approximate to zero, and over 5/6's of a cycle it must be a small fraction of its impact over half a cycle.
To illustrate, consider Zhao and Tung (2013). They show a 0.1 C impact from the AMO from 1961-2010 (50 yrs), compared to a 0.49 C impact from 1974-2005 (ie, the most recent trought to peak interval). (See fig 1a inset). Their 50 year trend to 2010 in the regressed temperature data (fig1b) is 0.083 C per decade, or an increase of 0.415 C. That leads to a total increase due to natural variability of only 19.4% ignoring other factors (and in fact less than that considering the total actual temperature increase). Even then, that requires that the contribution from the AMO be 100% of the 0.49 C increase from 1974-2005, and hence that the change natural forcings be equal but opposite in sign to the change in anthropogenic forcings over that period.
So, when you actually look at 20th century data, their 40% claim is absurd on its face, even using their own data from a different paper. It requires far overestimating the peak to trough temperature impact of the AMO (which they consider the primary source of decadal variability), and ignoring the duration of the cycles to which they appeal. Even then we are ignoring the multiple additional problems with their claim, some of which are mentioned by KR.
-
Rob Honeycutt at 05:23 AM on 10 September 2014When their research has social implications, how should climate scientists get involved?
saileshrao... I have to say, at this point I'm rather disappointed in this whole exchange.
-
saileshrao at 03:51 AM on 10 September 2014When their research has social implications, how should climate scientists get involved?
Here's the letter I sent today to all members of the AGU 2014 Fall Meeting Planning Committee...
Subject: Core Values at the AGU 2014 Fall Meeting
Dear Dr. Rousseau and Distinguished Members of the AGU Fall Meeting Planning Committee,
Prof. Will Steffen of ANU wrote recently,
"Climate change is one of many global changes that are destabilising our planetary life support system. It is ultimately a question of core values. Can we change our core values rapidly enough – and decisively enough – to halt our slide towards collapse? That is humanity’s most important question in the 21st century!"
Fortunately, we can begin to exemplify and promote the decisive changes in core values that Prof. Steffen is referring to, during the AGU Fall Meeting this December. The caterers at the Moscone Center, Savor Catering, have assured me that they can certainly provide all locally-sourced, organic, plant-based, Vegan foods during the AGU 2014 Fall Meeting.
Please have the appropriate staff at AGU contact Robert Duncan ([email address redacted]) or Denise Roque ([email address redacted]) to work out the details. And please let them know that I would be happy to assist them in any way possible to make this happen.
Thanking you for your kind consideration,
Yours truly,
Sailesh Rao.
Executive Director, Climate Healers Initiative for Transformation
3145 E Chandler Blvd #110-233
Phoenix, AZ 85048
http://www.climatehealers.orgModerator Response:[JH] Please cease and desist from conducting your campaign re the food to be served at the AGU Fall meeting on the SkS website. Any further posts by you on this matter will be summarily deleted.
Please note that posting comments here at SkS is a privilege, not a right. This privilege can be rescinded if the posting individual treats adherence to the Comments Policy as optional, rather than the mandatory condition of participating in this online forum.
Please take the time to review the policy and ensure future comments are in full compliance with it. Thanks for your understanding and compliance in this matter.
[Dikran Marsupial] email addresses redacted
-
saileshrao at 03:49 AM on 10 September 2014When their research has social implications, how should climate scientists get involved?
Tom Curtis @29:
I was also under the impression that eggs and milk are low-impact foods and was lacto-ovo-vegetarian until 6 years ago before I realized that those are "low-impact" foods only because of how we did the accounting. The chickens that lay the eggs and the male chicks that are ground up as by-products of the egg industry are all being accounted for by the chicken nugget consumers and the cows that produce the milk are all being accounted for by the hamburger eaters. India has a proliferation of cattle (320 million heads of cattle vs. 90 million in the US) which are literally eating up the forests, because a lot of Indians drink milk, but not many consume beef and averse to the killing of cattle.
The egg and dairy consumers depend upon chicken and beef consumers to clean up after them or else their foods have higher impact than chicken and beef consumption itself. -
Tom Curtis at 03:08 AM on 10 September 2014When their research has social implications, how should climate scientists get involved?
saileshrao @26:
"...an entirely locally sourced, plant-based, Vegan Fall meeting..."
(My emphasis)
Apparently not even eggs, a low carbon, low ecological impact food if ever there was one are acceptable. That underlines the point that saileshao is pursuing this agenda for ideological rather than scientific or ethical reasons.
-
saileshrao at 02:54 AM on 10 September 2014When their research has social implications, how should climate scientists get involved?
Rob Honeycutt @27:
Will do, I'll send the AGU Fall Meeting planning committee another reminder today. The caterers actually gave me the impression that I was the first one from the AGU to contact them and wanted to know details on what we're looking for, breakfasts, breaks, lunch, dinner, etc.
Thank you all for your encouragement, support and stimulating discussions on Skeptical Science. We'll get through this together...
-
Rob Honeycutt at 01:32 AM on 10 September 2014When their research has social implications, how should climate scientists get involved?
Give them a chance, saileshrao. Your issue is likely number 228 of 1000 other things they have to get done in preparation for the December meetings. And I wouldn't be surprised if the decision on catering was made months ago, and this year can't be changed one way or another.
You need to have patience and tenacity for the things you believe are important.
-
saileshrao at 01:14 AM on 10 September 2014When their research has social implications, how should climate scientists get involved?
Rob Honeycutt @25:
The Moscone Center caterer, SAVOR Catering, has already assured me that if the AGU Fall Meeting planning committee wishes to organize an entirely locally sourced, plant-based, Vegan Fall meeting, they are ready and able to do that.
The decision is squarely with the AGU planning committee now. I have already written to every member of the AGU Fall Meeting planning committee and have received no replies whatsoever from any of them. I have reluctantly concluded that there seems to be some kind of blockage in our mental processes when it comes to connecting our personal habits with the environmental catastrophes that we study. -
2014 SkS Weekly Digest #36
jim - Yes,Tung and Zhou is part of the literature considered in the IPCC reports. But note that it is only one piece of the literature, that there are other works using similar and quite different methods that find the anthropogenic contribution to be much higher. In fact, there are papers referenced in AR5 showing much lower and much higher percentages.
But in looking at all of the literature, and the uncertainties expressed in those works, the IPCC has reported a PDF as shown above. It would be completely inappropriate to ignore the body of evidence in favor of a single paper.
[ Side note: IMO most of the estimates of AMO attribution are flawed. In particular, a linear detrending of North Atlantic temps to identify AMO is wholly inappropriate in the face of other factors/forcings being non-linear over the periods of interest, linear detrending doesn't correctly remove forcings. If you are looking for variability you must first accurately subtract knowns and only then look at unknown contributions, or you will misattribute the remaining forcings to the unknowns. T&Z 2013 fails on this point. ]
-
Lionel A at 00:44 AM on 10 September 201497 hours of consensus: caricatures and quotes from 97 scientists
Congratulations team on another world class utility that shows up the opposition.
-
Lionel A at 00:43 AM on 10 September 201497 hours of consensus: caricatures and quotes from 97 scientists
Yes Lars, I spotted him a few hours back, before noon UK time, looking grim with arms folded. Will we see Lindzen, Spencer or Michaels soon, oh and Soon too?
BTW I first started looking at the graphic with Firefox on a Linux box, Ubuntu, and have more difficulty getting the name up and arm wave. It does work but not quite as slick as on a Win box, with Firefox.
I was wondering about those rotate buttons too, they didn't show up on Linux, the monitor here is at the limits for brightness and contrast, but now I what I am looking for I find them if hovering over the patch where they are, having just located them on the Win box.
-
Tom Curtis at 00:39 AM on 10 September 20142014 SkS Weekly Digest #36
herrhund @6,
"DSL, well isn't that what sciense is about? Ongoing discussion has to be allowed and is a demand of science."
Sorry, you only get to play that card if the paper is published in peer review, or is the official report of an internationally recognized scientific body.
There are too many kooks out there to expect academics to respond to every pet peeve produced by political hacks, politically motivated "think tanks", and other web crazies. If they think they have something of substance, let them publish. If they have not published, and you think they have something of substance, you specify what it is and why it is substantial.
-
2014 SkS Weekly Digest #36
herrhund: "I am sure they have no problem to responding to some of the points in the paper from the global warming policy foundation."
They've responded. You haven't looked. And, no, science isn't required to answer every publicly displayed argument against it. If that were the case, no science would ever get accomplished. GWPF is designed to shape public opinion, not be scientifically accurate. Its writers can write just about anything they want without facing scrutiny. The general public certainly won't scrutinize, and scientists, for the most part, completely understand what the GWPF is all about and treat it accordingly (by ignoring it).
Moderator Response:[PS] It would be better if any further discussion on this was moved to the thread pointed to by DSL. Herrhund - in the interests of a reasonable discussion, it would be best if you picked the point in the GWPF "paper" that you found most compelling (a criticism of the method used that would alter the conclusions of the paper) and that is not already answered.
-
jim10940 at 00:15 AM on 10 September 20142014 SkS Weekly Digest #36
Thank you scaddenp and KR for the references. In reading Chapter 10 of the IPCC report I see that Tung and Zhou (2013) is referenced quite a bit particularly in section 10.3.1.1.1. Here is a quote from the end of that section:
Some studies have suggested that the warming is a response to the AMO (Schlesinger and Ramankutty, 1994; Polyakov et al., 2005; Knight et al., 2006; Tung and Zhou, 2013), or a large but random expression of internal variability (Bengtsson et al., 2006; Wood and Overland, 2010). Knight et al. (2009)
diagnose a shift from the negative to the positive phase of the AMO from 1910 to 1940, a mode of circulation that is estimated to contribute approximately 0.1°C, trough to peak, to GMST (Knight et al., 2005). Nonetheless, these studies do not challenge the AR4 assessment that external forcing very likely made a contribution to the warming over this period. In conclusion, the early 20th century warming is very unlikely to
be due to internal variability alone. It remains difficult to quantify the contribution to this warming from internal variability, natural forcing and anthropogenic forcing, due to forcing and response uncertainties and incomplete observational coverage.Now my reading of the chapter would be that there is anthropogenic warming but the contribution from internal variability and anthropogenic forcing is uncertain. I looked at the Tung and Zhou 2013 paper and they claim that "Quantitatively, the recurrent multidecadal internal variability, often underestimated in attribution studies, accounts for 40% of the observed recent 50-y warming trend."
My question is how to treat that body of literature in the above quote? Its very odd that this literature is included here and supports significant contribution of internal variability to the 20th century warming and yet the chart fig. 10.5 shows mostly a very small contribution from natural and internal variability. So obviously it has been discounted in the IPCC summary. But it is also not explicitly stated that these authors are are wrong in their conclusions. I am uneasy about this. The authors statements should be refuted completely or it is a reasonable opinion to hold that anywhere from 0 to 50 percent of the observed 20th century warming is from internal variability.
I also want to acknowledge that is a very good point that the assumption that the variability is all internal is an open question.
My own position right now is that there is 0 contribution of natural variability to the overall 20th century trend. And yet these above mentioned published authors hold a different opinion. I have to say claiming up to 50% contribution from internal variability is a reasonable position.
-
herrhund at 00:14 AM on 10 September 20142014 SkS Weekly Digest #36
DSL, well isn't that what sciense is about? Ongoing discussion has to be allowed and is a demand of science. There is always a chance that a study is not correct - to act like something is 100% sure is not allowed in scientific methodes, acting like it disqualifies whoever claims to know something for 100%.
If the method of the consensus project is well done according to scientific standards - I am sure they have no problem to responding to some of the points in the paper from the global warming policy foundation.Moderator Response:[JH] You are already skating on the thin ice of sloganeering which is prohibited by the SkS Comments Policy.
Please note that posting comments here at SkS is a privilege, not a right. This privilege can be rescinded if the posting individual treats adherence to the Comments Policy as optional, rather than the mandatory condition of participating in this online forum.
Please take the time to review the policy and ensure future comments are in full compliance with it. Thanks for your understanding and compliance in this matter.
-
Falk at 23:19 PM on 9 September 2014Rising Ocean Temperature: Is the Pacific Ocean Calling the Shots?
In the label of Fig. 1 you write that it shows the Ocean heating rates. Are the rates or the total heat content plottet in the graphs? I took it for the absolute change of heat content relative to a time average.
I find it interesting to see that the HOC is increasing the strongest around antartica in the deep while the see ice area (not volume) shows slight increase and surface temperature is more or less constant. Are there any effects on ocean currents expected from a changing gradient in this region?
Prev 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 Next
