Arguments
The 97% v the 3% – just how much global warming are humans causing?
Posted on 15 September 2014 by dana1981
A pair of climate scientists recently had a dispute regarding how much global warming humans are responsible for. Gavin Schmidt from Nasa represented the consensus of 96–97% of climate experts in arguing that humans have been the dominant cause of global warming since 1950, while Judith Curry from Georgia Tech represented the opinions of 2–4% of climate experts that we could be responsible for less than half of that warming.
Curry is to be the featured speaker on this subject at a National Press Club event tomorrow hosted by the Marshall Institute; a right-wing thinktank that has spread misinformation about the dangers of smoking, ozone depletion, acid rain, DDT, and now climate change. She may also discuss the subject at an event next week hosted by the fossil fuel-funded right-wing think tank Texas Public Policy Foundation (TPPF).
The exchange between Schmidt and Curry can be read on RealClimate – a blog run by climate scientists. The discrepancy in both the quantity and quality of the supporting evidence used by each scientist was one of the most telling aspects of their debate.
For his part, Schmidt referenced the most recent IPCC report. The IPCC summarises the latest and greatest climate science research, so there is no better single source. The figure below from the IPCC report illustrates why 96–97% of climate science experts and peer-reviewed research agree that humans are the main cause of global warming.
The black bar indicates the amount of global surface warming observed from 1951 to 2010. The green bar shows the amount of warming caused by human greenhouse gas emissions during that time. The yellow is the influence from other human effects (mainly cooling from human sulfate aerosol emissions, which scatter sunlight), and the orange is the combined human effect. Below those are the contributions from external natural factors (mainly the sun and volcanoes) and from natural internal variability (mainly ocean cycles), while the whiskers show the uncertainty range for each.
IPCC AR5 figure 10.5: Likely ranges (whiskers) and their mid-points (bars) for attributable warming trends over the 1951–2010 period due to greenhouse gases, other anthropogenic forcings (OA), natural forcings (NAT), combined anthropogenic forcings (ANT) and internal variability. The HadCRUT4 observations are shown in black.
Notice that the green and orange bars are both bigger than the black bar. This shows that greenhouse gases have caused more warming than has been observed over the past six decades, but some of that was offset by cooling from human aerosol pollution. And the best estimate from the body of peer-reviewed climate science research is that humans are responsible for more than 100% of the global surface warming since 1950, with natural factors probably offsetting a little bit of that with a slight cooling influence.
Schmidt illustrated this key point in the figure below, which is called a probability distribution of the warming caused by humans since 1950. The curve is centered at about 110% – the most likely value for the human contribution to global warming, while the probability of the human contribution being less than 50% is almost nil.
The probability density function for the fraction of warming attributable to human activity (derived from figure 10.5 in IPCC AR5). The bulk of the probability is far to the right of the ‘50%’ line, and the peak is around 110%. Source: RealClimate
Again it’s important to remember that the IPCC report is just a summary of the latest and greatest climate science research. The figures above are supported by the papers that have specifically investigated the attribution of recent global warming. This isn’t just one study; it’s based on many studies that are all in strong agreement. As the IPCC report concluded,
It is extremely likely that human activities caused more than half of the observed increase in GMST [global mean surface temperature] from 1951 to 2010.This assessment is supported by robust evidence from multiple studies using different methods.
It’s not just “more than half,” it’s also most likely close to 100%. In fact it’s just as likely that humans are responsible for about 160% of the global surface warming since 1950 as it is that we’re only responsible for 50%.
Curry disagrees with the expert consensus on this issue, but her arguments are rather muddled and “confused,” as Schmidt puts it. Her main argument is that there is uncertainty regarding the contribution of internal variability. The problem with that argument is that over long periods of time (like the six decades since 1950), positive and negative phases of ocean cycles tend to cancel each other out, and thus internal variability doesn’t have a large influence on long-term temperatures. As the first figure above shows, the IPCC estimates the temperature influence of internal variability since 1950 at ±0.1°C, during which time we’ve seen about 0.65°C global surface warming.
Curry also references a report written by Nic Lewis for the anti-climate policy think tank Global Warming Policy Foundation (GWPF), which I wrote about here. The GWPF report argues that the climate sensitivity is toward the lower end of the IPCC estimated range. However, the report is biased towards Lewis’ preferred approach, finding poor excuses to reject the many other methods that arrive at higher climate sensitivity estimates. Moreover, recent research has identified flaws in Lewis’ approach that explain why it incorrectly yields the lowest climate sensitivity estimates. In any case, even if the GWPF were correct, it wouldn’t disprove that most of the warming since 1950 is human-caused.
Curry’s other reference is to a single paper written by Zhou & Tung at the University of Washington in 2013, which concluded that roughly half of the global surface warming over the past 32 or 50 years could be explained by ocean cycles (specifically, the Atlantic Multi-decadal Oscillation). Matt Ridley also recently referenced this paper in an error-riddled Wall Street Journal editorial (debunked here and here and here and here). However, as Schmidt points out,
Tung and Zhou assumed that all multi-decadal variability was associated with the Atlantic Multi-decadal Oscillation (AMO) and did not assess whether anthropogenic forcings could project onto this variability. It is circular reasoning to then use this paper to conclude that all multi-decadal variability is associated with the AMO.
As far as I know, the Cook 2014 study was replicated by others, using the same papers. If you're interested you can find them. I am not. Why? Because, as I have argued, the Cook 2014 study isn't really that germane to the Cook/Schmidt debate at all. Nor would any other literature surveys be. The direct opinion study of Verheggen 2014 would be more so, but also not completely. I don't know about a skeptic, but If I did want to answer the question of where climate scientists would put attribution, I would go about it in different ways than both papers. I would not replicate either. That's just me.
Why? Because unquestionably, Curry's beliefs would fall under the 97% and not the 3% under the Cook 2014 methodology. And in the Verheggen methodology, we're not sure. At 50/50 or "middle tercile" it's unclear whether Curry would have put herself at 26-50% or 51%-75% under that method. You'd have to ask her.
But that's not even the major reason. Neither Schmidt or Curry were making opinion arguments. They were stating their own opinions based (arguably) on science. Attacking either position should be done in a similar manner. Neither one referenced Cook 2014, or any other poll, in their arguments that I read.
And attribution is merely one element of climate change. And a complicated and still uncertain one at that. Magnitude being a more important one, in my opinion. I'd be more worried about warming of 6 degrees C. even if only 50% due to anthropogenic causes, than I would be worried about 1.5 degrees, 120% due to anthro causes.
jwalsh @51, you continue to pursue off topic discussion; and continue to push the deceptive misinterpretation of the Cook et all classification as indicating only some of recent warming is anthropogenic. I have discussed why that is deceptive, and a misinterpretation here, where it is on topic. I suggest you take the discussion there.
Do I? Curry as representing the "3%" seems very much on topic. Let me see after having read that link that I now understand it correctly then. It sounds like your suggesting that Curry should be rated a "7", and part of the 3% based her belief about it being approximately 50/50. If Curry had simply gone that extra 1%, and quantified anthropogenic as at least 51%, she'd be a "1"? Curious, since the vast majority of those taking a position as endorsing, were rated as either the 2 or 3 part of the 97% simply because they didn't sign up to a numerical value that could be inferred from the abstract. It was never my intent to get stuck into the paper that hard, since, as I said, I would do it differently. I have a science degree, but I don't think I could confidently rate a whole lot of papers not in my field based on that criteria by abstract alone. Would be a bit of a crap-shoot to be honest.
However, we don't need to do that at all. Turns out Judith Curry is actually in the 97%. Two of her papers were rated "3" in Cook 2014. And a couple were "4's". So she either speaks for some aspect of the 97%, or speaks for the vast majority of papers taking no position, or she speaks for herself! :)
[RH] You and Russ have had this explained to you several times now. Curry continually argues minimal influence <50% from man-made sources, which is consistent with rating levels 5, 6 and 7.
As stated to Russ, you and others, if you believe Cook13 results are not robust, you are more than encouraged to produce your own research and submit to peer review. Posting comments of the same ilk as this one will be deemed as excessive repetition and deleted, per policy.
jwalsh @53:
Yes you do. In fact you wrote:
That clearly indicates that you consider the Cook et al criteria to be >0% anthropogenic contribution rather than the 50%+ also used by Verheggen et al. Further, you are simply rehashing a trivial point with regard to Curry. As I have already noted:
How much a challenger is seen by the fact that she assigns a 50% probability to a possibility (anthropogenic contribution <50%) that the IPCC assigns a 5% or less probability to in its exectutive summary, and a <0.1% probability in its formal attribution (Fig 10.5). Even using the purely empirical data from Fig 10.6 (which you steadfastly refuse to discuss), and inflating the (Fig 10.5) uncertainty by 50%, there is still a less than 5% (3.7%) probability of an attribution less than 50%. If we consider the probability of less than 40%, or less than 30% the comparison would likely be still worse for Curry, who, however gives us no estimate of uncertainty dispite he many criticisms of the IPCC for inadequately stating uncertainty.
Given this vast discrepancy, the detail of whether or not Curry just scrapes into the 97% or the 3% is trivial, and your obssession with looks exactly like evasion of more substantive discussion. Indeed, having been shown that you were wrong about Fig 10.5 being purely model based; and being directly shown that purely empirical methods of attribution yield a mean result scarcely distinguishable from that for Fig 10.5 you suddenly went completely silent on the substantive points of the post, and started of on misrepresenting Cook et al 2013 and obsessing about whether Curry was just in, or just out of Verheggen's consensus category.
People who have nothing to of substance to say about the main points of the article often obsess about the details, especially when they don't want people to pay attention to the substance of the article.
Which reminds me:
But they sure can tell when your a troll:
I have responded to a couple of jwalsh's comments about Cook et al on an appropriate thread.
jwalsh @56:
1) I have twice explicitly stated that Curry just falls inside the 97%. There is no basis for your misrepresenting me as saying otherwise. I do share the moderator's (RH) belief that she argues as though she thought the value was lower, and suspect she states her position as 50/50 merely to position herself as "in the middle" and by implication "fair and balanced" when neither of those are in fact true.
2) I left several discussions in my preceding comment to which you are responding explicitly dealing with the actual attribution studies. True to form you totally ignore them, concentrating solely on the trivial (whether or not Curry falls just inside or just outside the 97% when it is blindingly obvious that she does not fall within the IPCC concensus), and the off topic.
3) The two Curry papers endorsing the consensus were both written prior to 1996. It is well known that she has had a major shift in position since then, and it follows that they are irrelevant to determining her current position. I find it difficult to believe that you did not know this, and if you did you have deliberately and knowingly presented irrelevant evidence in the hope that it will be mistaken as relevant. Perhaps, however, it was a mere accident. Doubly so because you present that as an argument for removing those papers from the consensus when by her own statement she accepted the consensus at the time.
4) There is no "much larger third side of the coin" either as determined by Cook et al (2013) or Verheggen et al (2014). My comments under (3) above apply.
5) One of the features of trolling is the repeated concentration on trivial or off topic points, a feature that describes your behaviour perfectly. I stand by my assessment.
I suspect that the moderator will remove your post on the grounds of excessive repetition, and given the warning you have recieved they would be quite justified in doing so. That will not preclude your posting on this thread on topic - ie, attempting to show why the IPCC assessment is wrong beyond your mere say-so (your only evidence todate); or why Curry is right, or why some third explicitly stated position is right. If they remove your post they will, of course, remove mine as well, as responding to a deleted post (and are welcome to do so). Just remember when you go of to other boards, if your posts are deleted, you clearly brought it on yourself.
[RH] The moderator is in agreement.
I'll try not to mischarecterize your arguments, but please do me the courtesy of the same. I no more think the IPCC assessment is fully "wrong" any more than I think it's fully "right". On balance, I think the assessment is much more right than wrong though. And even if I did, why should my opinion matter a whole lot? To do so is improperly pretending that the complexites are simple. And I would not do that. I merely made the observation that the 10.5 graph was primarily derivative of CMIP5 models. The CMIP5 model assumptions (educated ones) about variables may indeed be correct or very close to it. I was a bit questionable about placing an inordinate amount of focus on models. Reliance on models and their outputs is one of the major criticisms of climate science in general. But I fully understand why they are used. We can hardly experiment with altering the variables of the planet (although some would argue that we are in an unguided way). Therefore models are used as a proxy. And they get better and better with time, and increased computational power.
FYI: I think Curry is right about some things and wrong about others. My quibble was with deciding her opinion (alone) was representative of a larger group.
You keep complaining that I haven't responded to things like the 10.6 graph. There's a reason for that. I looked at it again (not the first time I have read the entire chapter), but I have not yet found or read, any of the referenced papers. So I didn't comment on it (yet). I am not going to flat out be forced to make an uneducated opinion. When I get a chance to I'll probably take a look.
As a side observation, I think the denigration of a "side", any side (doesn't matter), online or otherwise, does not serve a useful purpose. It's more about the human tendency to form "tribes". As such, probably of interest to anthropologists, but not climate science.
[RH] This is continuing to drift far off topic. Please bring it to a close or locate a more appropriate thread.
jwalsh, " And I would not do that. I merely made the observation that the 10.5 graph was primarily derivative of CMIP5 models."
The repeated issue here would appear to be some confusion between how the models are used for attribution as opposed to forecasting long term climate. Do you think that model failure to have any skill at short term prediction, especially ESNO/PDO variability affects the way in which models are used in attribution studies?
I am going to assume this is on topic. It's getting hard to tell. :) I don't know that it does affect the attribution studies, but I see no reason why it can't affect the attribution breakdown picture. Could be anything at issue, stratospheric aerosols, transient carbon dioxide sensitivity estimates, etc.. We might not really know without model enhancements, better computers, and a couple of decades of new (or better) data. I have seen some commentary about employing better and more modern statistical analysis techniques to get a better handle on uncertainties as well.
[RH] Please move any further discussion of climate modeling to a more appropriate thread.
jwalsh, to reduce the probability of anthropogenic factors contributing 50% or less of observed warming from 1951-2010 to 4.99%, you have to increase the uncertainty shown in Fig 10.5 by 208%, and the uncertainty relative to model uncertainty by 419%.
What evidence beyond hand waving do you have that the uncertainty is understate by such a large margin? What evidence beyond handwaving do you have that the IPCC's original increase of uncertainty by 201% relative to the innate uncertainty of the models is insufficient?
jwalsh @57:
Thankyou for your quibble (=def "small complaint or criticism usually about something unimportant") that Curry, being a borderline member of the 97% cannot be taken as representative of the 3%. The corrollary, that the position of the 3%, is even less rational or evidence based than hers is duely noted.
jwalsh - then try for something less noisy in the models. The attribution statement equally well for total OHC. Not much in way of pesky natural cycles operating there. Do you accept that pdf is accurate representation of cause of increase in OHC - or do you have some other source of energy that might do?
First, If you water-board me to come up with a figure for AGW since 1950, I'd probably say 66-75%, so I am arguing from a position "within" the 97% (if I had published something on climate). If I had strong evidence that the figure 10.5 uncertainties were off in a publishable manner, I'd probably point you to my peer-reviewed paper on that! Heh.
I don't know if the CMIP5 models inputs or outputs correctly characterize uncertainties or not. However I do know that the outputs regarding temperature have been "running hot" for a couple decades now. Something must be wonky, without knowing precisely "what". And yes, I know there is no shortage of potential explanations. But it may not be the uncertainties that are wrong, but the underlying assumptions. Figure 10.5 shows natural variation as effectively "nil" with a small uncertainty. I don't believe the evidence points that way.
When I read Gavin Schmidt's statement from the realclimate discussion, I did a bit of a double-take.
"It is worth pointing out that there can be no assumption that natural contributions must be positive – indeed for any random time period of any length, one would expect natural contributions to be cooling half the time. " - Schmidt
He's right of course. You can't assume that the net natural contributions must be positive. But the same argument can be made about assuming them to be "zero" or negative over a short time frame. And I think the models make that assumption (and 10.5). I'd dearly love to be able to play with a super-computer for a while and test out various things myself, but I can't be certain I wouldn't use the computer time to mine Bitcoins or pick stocks instead...
But if you look at longer-term trends and paleo-climate studies, most (but not all) show cooling or warming over centuries long periods. So to me, it seems at least plausible that we're still (since 1600 or so) in an upward natural trend that would shift the value for the natural component up in figure 10.5.
I don't think those arguing for 50% or less are completely off-base, or unscientific, for thinking as they do.
[RH] "I don't think those arguing for 50% or less are completely off-base, or unscientific, for thinking as they do." But they are putting forth a position that holds the same likelihood as >160% man-made contribution, and completely ignoring that fact as well.
While I appreciate you staying on topic here, the discussion doesn't seem to be moving forward, and is only circling back on to previous themes. Please find a way to advance the conversation so that it doesn't become repetitive.
jwalsh @57 & before.
You seem to be struggling with the classification of Juduth Curry. I would suggest this may be because there is more than one Judith Curry.
Judith Curry-Scientist tends not to express wild theories which would not easily pass scientific review. For instance take Wyatt & Curry (2013) (PDF here). This is rather tame, saying nothing very controversial. Rather it proposes a hypothesised mechanism that allows "numerous observations of climate behaviour" to be seen as part of a larger mechanism, a hypothesis called the Stadium Wave. Although the writing is bad, you will see that nowhere does this work say NHT wobbles as a result of this Stadium Wave mechanism. It only shows that NHT wobbles in step with it given present data and that there is a mechanism for warming certain NH lands. Although a coded message could be perceived by those looking for it, the paper actually goes no further than to say about "numerous observations of climate behaviour":-
You will find Judith Curry-Hypothesist synthesises her talk at the APS using language that is much stronger. "The stadium wave hypothesis provides a plausible explanation for the hiatus in warming and helps explain why climate models did not predict this hiatus."
However, the fully-powered message is only apparent when presented by Judith Curry-Blog-Mom. "The stadium wave and Chen and Tung papers, among others, are consistent with the idea that the multidecadal oscillations, when superimposed on an overall warming trend, can account for the overall staircase pattern."
jwalsh @63
It would improve your message immeasurably if you could in some way indicate what "longer-term trends and paleo-climate studies, most (but not all) show cooling or warming over centuries long periods" are you asking us to "look at"?
Oh, as I said, there are many. But Mann, M.E., Zhang, Z., Hughes, M.K., Bradley, R.S., Miller, S.K., Rutherford, S., 2008 contains as good a graphical representation as any. http://www.meteo.psu.edu/~mann/shared/articles/MannetalPNAS08.pdf
At least visually, a better way to look at it, I think is to remove the instrumental portion. I am not a fan of stapling temperature records to data that is considerably smoother as a result of method. But it doesn't really matter. (snip) Depends on which you pick, but I think most people agree that it was cooler in the LIA than in the 19th century or early 20th.
[RH] If it "doesn't really matter" then you won't mind if I remove thinly veiled and inaccurate accusations against Dr Mann's work.
jwalsh @63, like MA Rodger, I am curious as to which longer term trend you have found in the Paleostudies. The Paleostudies I am aware of show virtually no trend between 1730 and the commencement of the (GISS) instrumental period in 1880:
The overall NH temperature trend over that period amounts to 0.003 C per decade. It is likely that the global trend is less. Assuming that global trend is the same, and that it represents a natural cycle of internal variability rather than a consequence of forcing (which is already accounted for) gives a 0.018 C of 0.65 C temperature increase over 2.7% of the 1951-2010 warming. Both assumptions (ie, that global temperatures increased at the same rate, and that the increase is a consequence of internal variability rather than forcing) stretch credulity.
I suspect you want to include the period from the greatest temperature resonse associated with the maunder minimum (approx 1700). That, however, sill only gives a NH trend of 0.018 C per decade, for a total 1951-2010 warming of 0.11 C, or 16.8% of warming. Further, the trough in temperatures at 1700 is known to be a forced response both to solar variations (from sun spots) and especially to the volcanic record. Both factors are already included in the IPCC attribution, such that counting them again would be double dipping.
The data in that image can be discussed here.
All of this furhter begs the point as to why the long term cooling trend visible in the paleo record only (slightly) reversed itself in the early eighteenth century, ie, after the invention of the steam engine and the widespread use of coal for domestic heating:
Given that jwalsh has confirmed a reliance of Mann et al, 2008, I have determined the global CRU-EIV trends from 1700 and 1730 to 1850 (ie, commencement of the CRU instrumental record) as being 0.011 C per decade, and 0.007 C per decades respectively. To 1880 they are 0.013 and 0.011 C per decade respectively. The CRU and EIV reconstructions were chosen because EIV shows a larger trend than the CPS, and CRU is the land only, and again shows greater trends.
So, even with these exagerated trends we have a "trend" warming from 1951-2010 of between 6.4% and 12%. With no evidence that the trend in question is not forced, or even not anthropogenic, jwalsh concludes from that the IPCC attribution should be reduced by 38%, from 108 to 70%. That is, he exagerates the influence of his basis for his non-expert attribution by at least a factor of 3, without bothering to have showed that it is even a basis for a change in the IPCC estimate.
jwalsh @63:
I did not ask about where you estimated the anthropogenic contribution to be, but what evidence you had in support of inflating the IPCC's uncertainty by a factor of two or more. Never-the-less it is interesting to note that your "estimate" of the anthropogenic contribution has a probability of 2.3% based on Fig 10.5, and that even if we inflate IPCC uncertainty so that there is a 5% chance of a 50% contribution or less, that probability rises to only 5.8%. Further, it is interesting to note that, as discussed above, your evidence for your preffered contribution is weak, ambiguous (at best) and does not in fact support your preferred value. However, if I ever decide to determine the anthropogenic contribution by inexpert ellicitation, I'll be sure to keep your wild assed guess in mind.
This argument misses the true beauty of the AR5 assessment. The GMST record shows a gradual fall from 1880-1910, a sharp rise from 1910-1940, a gradual fall from 1940-1970, and a sharp rise from 1970-2000, with a gradual rise thereafter. That pattern is the basis for claims that internal variability is a significant contributor. It sets the phase and period of the multi-decadal internal variability in global temperature, if it exists. But that being the case, the period from 1951-2010 is exactly in phase. It goes from the early declining phase of the first cycle to the exactly equivalent year of the second cycle. Ergo any contribution from that cycle to the rise in temperature over that period comes entirely from any purported change in amplitude of the internal variability - which change of magnitude is not evident on any grounds, and on best evidence is a decrease in magnitude.
Nor is it any good to claim that the periods of the internal variation is not coordinated with that pattern. If it is not, then the effect of internal variability has been small such that it does not impact the overall pattern of warming through the twentieth century. So, either 2010 is in phase with 1950 and the internal variability between the two is consequently small, or the internal variability is small in any event, having little effect on the overall temperature pattern.
The logic here is very simple, and leaves no room for a large non-forced component of the temperature change from 1951-2010 unless it comes from multi-century patterns of internal variability, and as seen above those patterns do not have sufficient trend to be relevant, and are in any event most probably forced.
So, my questions remain unanswered. I asked for evidence as to why the IPCC uncertainty should be greatly inflated and you have merely provided unsupported speculation about the actual contribution, and ill thought out meanderings about Schmidt's comment that paid no attention to the actual relevant durations.
jwalsh @63 & @65.
Being as generous as I can , Figure 3 from Mann et al (2008) here shows perhaps 0.1ºC per century rise for the NH "since 1600 or so", about 7% of the 20th century rise. (Being less generous, note that some of the reconstructions are flat with zero warming.) I'm not sure how much of the warming is attributable to the Maunder Minimum/Dalton Minimum - probably the lion's share. When solar forcing is considered for the post-1950 contribution, it will be small but cooling.
So using Mann et al (2008) to argue for a significant natural contribution to the global surface warming since 1950 (33% to 25% is quoted @63) doesn't make any sense to me.
MA Rodger @69, from an extended discussion of the issue by K.a.r.S.t.e.N:
As he concludes, "From a NH point of view, the AMO plays a minor role as far as the temperature evolution is concerned."
Hasn't William F. Ruddiman (old article, first of many on GOOGLE) published studies that show the human contribution to AGW started 5,000 years ago? The increase in temperature from 1500 to 1900 is just the slower increase from preindustrial human caused AGW, not a natural variation.
The last I saw on Rudiman's work the claim was made that most scientists supported his theory but it was not yet accepted as a consensus position.
It strikes me that Russ needs more than his unsupported asertion to show that any purproted increase in temperature over that past 5,000 years is natural. It is at least as likely that it was preindustrial AGW.
michael sweet @71, Ruddiman mounts a very convincing case that human generated GHG emissions from land use change have prevented a significant decline in global temperatures following the "Holocene Climactic Optimum" 8000 years ago.
However, the conventional end date of the pre-industrial era is 1750, not 1900. It is 1750 for a reason, specifically that from 1750 human emissions commence their near exponential growth, although the resulting increase would not emerge from the noise caused by changes in CO2 concentratrion due to changes in global temperatures for about 100 years after that:
Tom,
It seems to me that if jwalsh (sorry for the previous reference to Russ) wants to claim that there may be significant increase in temperature recently due to natural causes he has to show that humans have not been responsible for the increase in CO2 and temperature you document over the past 8000 years.
He claims "it seems at least plausible that we're still (since 1600 or so) in an upward natural trend that would shift the value for the natural component up in figure 10.5."
If Ruddiman is correct, and he has peer reviewed papers to support his claims, than any increase since 1600 was AGW, not natural. The natural forcing since 1600 was negative, not positive. Vague natural cycles are hand waving.
The long term natural forcing for the past 8,000 years is cooling, not warming. All warming, plus the additional cooling that would have occured naturaly, is AGW. Perhaps jwalsh can provide data to support his claim that natural forcings have been positive. I doubt he can provide such data for any time period greater than 100 years in the past 8000, because the natural forcing is cooling.
michael sweet @70, I don't think your argument holds. Specifically, while anthropogenic forcing has been a factor since 6000 BC (as can be seen by the rise in CO2 from that time in the first figure @72), that influence has been small and gradual such that natural variations have dominated the trends on a centenial scale up till 1850 at least. That can be seen clearly from the forcings of the industrial era (1750 onwards):
Clearly the volcanic forcing dominates in the early 1800s (and as it happens, the early 1700s) so that the rise in temperature over the 19th century is primarilly due to the reduced volcanism. As late as 1940, the strength of volcanic forcing effectively equals that from GHG, so that for the rise in temperature in the early twentieth century, anthropogenic factors only contributed about a third.
For earlier periods, volcanic forcings are also dominant on century scale trends. The temperature contribution of various forcings is shown by Cowley 2000 (click here for clearer image):
Blue is volcanic, pink, green and gold are three proxies for solar activity, red is GHG and blue are tropospheric aerosols. A climate sensitivity of two is assumed, but the fit is to MBH 98 which is known to understate climate variability.
Here are more recent treatments, from AR4:
And from a 2014 comparison of NH forcings with the AMO:
The abstract of the later (Knudson et al) is worth reading:
So, while I agree that the increase since 1600 has an anthropogenic component, and that the dominant cause of that increase has been changes in forcing rather than internal variability (as jwalsh requires), I think it misreads to the data to think that it was primarilly anthropogenic. Of course, as I have argued above, just the dominant forced component is sufficient refute jwalsh, who requires not just that the increase be natural (for natural forcings have been accounted for in the assessment), but that it be predominantly due to internal variability.
Well if you want to get schooled in the data on a subject just make a comment to Tom!
Thank you for the data on long term forcing. It appears that over centuary spans natural forcings exceed human forcings.
Do you know if data supports the claim that the decrease in CO2 around 1550 (from your post 72) was caused by the collapse of Native American Indian populations due to the introduction of European diseases (and subsequent reforestation of their farmlands)? How much did this decrease in CO2 change the temperature (if the sensitivity is 3.0) over the next 250 years before the CO2 concentration rose again?
michael sweet @75, the change in CO2 is a drop from approx 282 ppmv to 274 ppmv, which would entail a negative forcing of 0.15 W/m^2, which would correspond to a 0.06-0.18 C temperature influence over centenial scales, with a 0.12 C influence for a climate sensitivity of 3, taking about 200 years to approach the full impact. That in turn is about a fifth of the total temperature change from the MWP to the LIA.
It is a matter of scientific dispute as to the cause of the decrease in CO2. I lean to the theory that it is temperature related. That is, as global temperatures fall, oceans absorb more CO2 thereby lowering global atmospheric CO2 concentration. Such a process would make little change to the C13/C12 ratio in atmospheric CO2. I have only read the abstract, but this article appears to support such a cause for the fall in CO2 levels. However, other causes have been proposed, including increases in wetland areas, and, as you mention, depopulation in the Americas due to disease brought by European colonists (for which I cannot find a suitable reference).
Against the last proposal, I would note that the Americas are only one part of the world and deforestation was occuring in Europe and potentially other places in the world at the time; and further, the lost farmlands had not denuded the forest as with modern agriculture, but formed a patchwork within the forest so that the impact may have been less than suggested.
However, I cannot claim to be well informed on this topic so beyond noting that there are alternative theories, I do not think I can really help you.
Tom Curtis @72
Some interesting points being made in recent comments here (I always enjoy a good AMO discussion) but I was surprised to read "Ruddiman mounts a very convincing case..." I didn't think his case has established itself beyond being an interesting controversy.
I'm not averse to the idea that a few hundred thousand humans in pre-history could have been responsible for CO2 & CH4 levels by some means as Ruddiman suggests they did and that the Holocene ice cores yield data that looks a lot different to the Eemian and other interglacials.
But surely the "means" suggested do remain speculative and interglacials are not expected to be carbon-copies of each other. Or have I missed something?
MA Rodger @77, if his case were established "beyond ... controversy" I would have called it an overwhelming case. IMO, the chief points in Ruddiman's favour are that CO2 levels have risen while global temperatures have fallen from the HCO. Absent significant emissions into the atmosphere from a terrestial source, that is not possible (as colder water absorbs more CO2). Further, at the same time deltaC13 has also risen indicating the the excess CO2 has come from plants:
The mechanisms whereby humans can cause such a change, deforestation, increased rice agriculture (leading to increased methane production from C12 enriched sources), and increased cattle grazing (same as rice production) definitely have that effect. The question is whether they have been sufficient to account for a significant proportion of the effect, or whether natural deforestation (as with the Sahara) or increases in wetlands are to blame. I think the case for the former is significantly stronger than the later, in part because while interglacials can vary, it would be a remarkable coincidence for this interglacial to vary on just this point immediately following the invention of agriculture.
Tom Curtise @78.
Perhaps we use the word "controversy" differently. When a theory has a whole list of stuff to overcome (like Rudiman's theory gets in AR4 Section 6.1.5.2), that is when I would consider aspousing such a theory "controversial."
I was of the understanding that the isotope data (eg from Elsig et al 2009) was a problem for Ruddiman's case was as yet unresolved. Elsig et al. Figure 1. of data from Dome C cores shows no fall in C13 as required for biosphere emissions. Ruddiman in his 2011 paper did respond with a peatland argument but that draws the whole early Holocene carbon cycle into the equasion rather than put the matter to rest. I note Ruddiman's latest presentation of his theory (Ruddiman et al 2014) remains silent on any isotope issue.
My sympathy for the Ruddiman theory perhaps rests on not knowing how agriculture developed. I tend to find the human population (gu)estimates are quoted with more credence than they deserve and feel a low figure is worthy of more consideration. A small population is one of the things leveled at Rudimman. How can so few illiterate prehistoric humans armed with only a stone axe be so much more effective at CO2 emissions than the illiterate 4x4 owners today? I think due consideration would show it is far from impossible for a small population using fire and perhaps even hunting to make a big impact on CO2 emissions (averaging 30 MtC over the period) and which appear to be up to speed very quickly as sedentary agriculture had only just begun.
MA Rodger @79, mindful of the context of the discussion, I will note that by 1700 there is already a significant ongoing decline in δC13, and hence already significant evidence of increasing anthropogenic forcings:
Indeed, the rate of decline of δC13 does not change significantly until at least 1900 although I very strongly suspect that conceals a switch from LLUC to fossil fuels as the primary source of emissions over that period.
With regard to Ruddiman's hypothesis, knocking out the declining δC13 does knock out a major evidentiary support for his theory making it far more controversial than I had supposed. He does, as you note, have an auxilliary hypothesis. That auxilliary hypothesis is plausible, and has some evidence in support in that Elsig et al 2009 does appear to use a low estimate of peat burial. Ruddiman et al (2011) write:
Taken at face value, that means human preindustrial emissions from the Holocene climactic optimum are in the order of 260 GtC, or approximately half of their emissions in the industrial era. That is consistent with an anthropogenic increase in CO2 of 20 ppmv over the last 8000 years to 1750. (For those not familiar with the carbon cycle, over periods of thousands of years, only 10-30% of emissions remain in the atmosphere depending on the amount of emissions.)
Returning to the context of the discussion above, my take on Ruddiman's hypothesis is that the basic idea must be true in principle, in that humans did start about 10,000 years ago undertaking activities which increased CO2 and CH4 emissions such that they increased atmospheric concentrations relative to what they would otherwise have been. What is being argued about is whether that increase amounts to 5% or 100% of the increase in CO2 over that time. Therefore the assumption that preindustrial changes in temperature are entirely natural is unwarrented. However, I must revise down my current estimate of how much anthropogenic factors reduce the millenial scale cooling trends since about 8000 years ago.
I get roughly 0.02 per decade for Moburg. Which, while the outlier for that paper, there are others showing 0.03. Arguing over it would probably start a big paleo-climate discussion.
And going to Greenland ice core data, there have been strong periodic shifts of roughly +- 1.0 deg C per millennia or so, which is more consistent with 0.02 than 0.01. And while the core trend shows an overall downward trend, that slope is much shallower than the millennial swings. So (presumably solar?) effects could be on the order of 0.12 degrees for 1950-2010. And they seem to be either strongly positive or strongly negative, not normally distributed around a mean on decadal scales.
Tom, what you seem to keep forgetting is that someone who disagrees with the logic of the attribution assignment numbers and estimated uncertainties derived from models and summarized in 10.5 is that they think the IPCC got that wrong. So arguing probability estimates from 10.5 is a circular argument. They think those probability estimates are wrong as well.
And I like to think my opinion is at least an educated one. I derive my own thoughts on attribution from a scientific education, reading peer-reviewed papers on it since 1992 or so, and every IPCC report ever published. If there's a better way to come by climate information, I'd appreciate any tips. No, putting a lot of stock in the opinion of a random person on a blog, would certainly be an inadvisable approach. But I don't think the policy types would shift to that mode after ignoring the IPCC.... I mentioned it for context of my thoughts. Where do you think attribution percentage is? And did you come by your climate education guiding your own opinion any differently than I did?
One of the things about IPCC reports is that they sometimes lack self-consistency. The attribution experts within the IPCC seem to have issues with the model estimates for temperature and the under-lying assumptions as well, and in fact say so.
Figure 11.25b in Chapter 11 of AR5 (having issues with the new Firefox and Imgur), shows an expert-assessment of the IPCC of the next two decades. It shows "likely" warming of as low as 0.15 deg. C (this is below the 95% confidence range of the models) with a midpoint around 0.30 degrees per decade. The stated reasons for this are possible under-estimation of solar effects, other natural variabilities like ocean currents, and possible over-estimation of CO2 sensitivity and aerosol effects. What is unstated from a backing away from model predictions due to backing away from model assumptions, is that prior attribution of warming might be in need of adjustment. And this is even forgetting that the amount of warming is a little bit contentious. Taking the satellite data instead of CRUT4 for example gives an observed 1950-2010 warming of closer to 0.5, rather than 0.7.
But I think this is focusing on the wrong "enemies" in any case. Regardless of where attribution is now, it doesn't take a genius to follow the trend in carbon emmissions to realize that it's going to be most of a much higher number in short order in any case. And that's even forgetting that replacing a finite energy source is going to be necessary.
My 2 cents. Spent the weekend burning very little carbon, and no computer access up north. No electricity, roads, or land-line. :)
jwalsh @81:
Running through some numbers:
Moberg 2005 trend from 1600-1850 - 0.08 C / Century.
Moberg 2005 mean trend from start years between 1600 to 1700 inclusive through to 1850 - 0.09 C / Century
Moberg 2005 maximum trend from a start year between 1600-1700 inclusive to 1850 - 0.11 C / Century.
jwalsh overestimation factor 175-250%
Percentage of trend from "1600 or so" that is unforced - no estimate by jwalsh, assumed to be 100% by his implicit argument.
Discussion of forcing history from "1600 or so" by jwalsh - zero.
Invocation of higher trends from unspecified data sets so that the numbers can't be checked - One to date.
Standard troll attempt to mistake regional (Greenland) temperatures for global or NH temperatures by jwalsh - One to date.
jwalsh @81:
Some more numbers, with data from the IPCC AR5 reconstruction of TSI as detailed here.
1951-2010_|_Forcing_|_7 Year mean_|_Units
Difference__|_-0.033__|___-0.064____|_W/m^2
Trend______|_-0.008__|___-0.0082___|_W/m^2/Decade
So, a negative solar contribution magically becomes a positive contribution of 0.12 C (ie, with an effective Transient Climate Response of -13.45 Degrees K/W/m^2 all by the magic of an arbitrary and irrelevant invocation of GISP2.
No wonder jwalsh thinks the IPCC underestimates uncertainty. He is able to find contrary certainties whereever he looks.
jwalsh @81.
Down this comment thread, you have indeed been picking at the basis of the IPCC's Figure 10.5 but you have not until now reputed it in its entirety.
Yet if you were to get an answer to your questioning ("Where do you think attribution percentage is?"), that answer should be a lot more detailed than the likes of "Oh I see it at 110% ±10%." and such a fuller acount would very likely be a verbal Figure 10.5; something like this:-
Just like in Figure 10.5, things that affect global temperature can be placed into 4 catagories, all zeroed at AD1750.
Firstly there are positive anthropogenic forcings of which CO2 is the biggest, and scariest because it is very long-lasting. The force of this first group can be evaluated with some accuracy.
Secondly is negative anthropogenic forcings. These are not so easily evaluated. If their recent effect is very large, that is scary as it means climate is highly sensitive to forcings and potentially we could see some very large temperature changes if the negative forcings were to reduce. That is possible as they are not as long-lived as the positive ones. It is very unlikely that the negative ones are very small (and they cannot be negative) which means sensitivity cannot be very low so there is every reason to be worried by the size of the positive forcings.
The third category is natural forcings which can be evaluated with fair accuracy. There is no evidence to suggest they are very large. There is no evidence to suggest they are at present a positive forcing.
The fourth category is unforced internal variability of the climate system. There is no reasonable evidence to suggest this is a large effect.
If somebody does wish to overturn Figure 10.5, they should really be indicating why - what within this description here is seen as wrong.
Judith Curry for instance, the subject of the post, believes the third category could contain "known unknowns" and even "unknown unknowns." However she does so on the unscientific-basis of zero evidence so jwalsh may well disageree with her give the acknowledgement of a "scientific education." Curry also has objections with the category 4 assessment that are a little better defined. Here she advocates the Stadium Wave, a highly dodgy use of a poorly defined hypothesis.
And on this unlikely basis Curry refutes Figure 10.5 with her unswerving 50:50 attribution alongside assertions of a low ECS. The reasoning for her position and how they link to her objections are not at all clear.
Of course, while Curry has so obviously lost the plot, she actually does better than those whose argument gets no further than "No no no!!" which is my assessment of the position presented thus far by jwalsh.
Tom Curtis @80.
The graph you show from Dean et al (2014) gets me scurrying for the Houghton land-use emissions data which shows a pronounced kink in the emissions data, changing from 3.5Mt/yr/yr to 23Mt/yr/yr, at, you guessed it, 1950. How robust the Houghton data actually is, I know not, but the shift is quite profound (evidently, being x7) and well defined. I note Dean et al. make no comment on changing land-use emissions at that time, rather they stress the effect of FF emissions which of course have turned the biosphere from an increasing net CO2 source to a smaller CO2 net source/sink (which it had become by 1970). However the dramatic fall from an increasing source to a minor player occurred in the early 1950s.
Ahh. I see your error straight away. Yes, picking an end still at what is considered to be the end of the little ice age would indeed give a low estimate, and wrong just by inspection. The IPCC considers about 1950 as being the threshold period when anthropogenic causes start to be detectable in the record. Before then anthropogenic forcings just too small.
1600-1950 Moburg 2005 by my quick math : ABS(0.9-0.2 (deg. C approx))/35 decades = 0.02 deg C./decade .....exactly as I said. I should have specified a range. But it honestly didn't occur to me that someone familiar with climate would decide that 1850 at the end of the LIA was a sensible choice.
As for the troll discussion? I prefer to keep things on a mature and civil level or not at all. I'm funny that way. I think you'll find that it's not that easy to get a rise out of me though. I'm not so thin-skinned. Perhaps it's a relative age thing.
Yes, there's a tricky limitation with ice cores. The ones at the equator don't last nearly as long. I didn't say they were a perfect match to NH temps (or global). Evidence that the Greenland temperature swings were localized for some reason? None provided. Evidence of the Minoan, Roman, and Medieval warm periods from either historical records and other proxies? Hell yes. But sure, might not be as extreme in swing. Do you have a good explanation for the approximately 1200 year cycles?
Agree with that. Especially for the present and future.
I kind of agree with Lord Monckton that this appears to be somewhat of a universal "fudge-factor", varying wildly. I think it's over-estimated. And there's evidence that it was declining into the late 20th century.
A combination of the two of these seem to be completely off-setting anthropogenic warming for the last decade and a half, and may have accounted for a good piece of the 1980-1998 warming. I think this is the IPCC's current biggest challenge. I have yet to see convincing a explanation for the 1910-1940 warming, and the above two reasons seem as likely as any other.
I see you missed my bit about the IPCC currently (and quietly) estimating temperatures at the the bottom range of model estimates (and even below). This appears to be an expert determination that the models are simply over-projecting by the IPCC. Perhaps you disagree with the IPCC on this. Your prerogative.
The bolded portion is entirely incorrect.
Ah, I clicked "Submit" before finishing, for which I apologise.
It is an excerpt from jwalsh's comment #86 that I have called out as incorrect in #87.
jwalsh - Several comments here:
1950-1960 is not when anthropogenic contributions become detectable in the climate record, but rather when they become dominant over natural forcings.
GISP2 is a local record, not a global one, recording temps at a single point on the Greenland ice cap. There is no evidence that I am aware of for 1200 year cycles, incidentally - that claim of yours appears to have materialized out of left field.
Negative anthropogenic forcings have a fairly high uncertainty - but the best estimate is for a climate sensitivity around 3C/doubling of CO2. Claiming that they are small and that correspondingly ECS is low (as you appear to) is a cherry-pick of but one low-likelyhood end of the PDF, and that isn't justified by anything other than wishful thinking.
Temps have been running below (averaged) model projections for ~15 years - a statistically insignificant time period, while remaining in the 2-sigma model range. That means there has been no invalidation of the models to date. Add in more accurate forcings (better than the ones used for CMIP5) such as discussed by Schmidt et al 2014, which clarify that short term internal variation is indeed negative right now, and there is every indication that the models are right on track.
---
The gist of your various Gish Gallops here seems to be that you disagree with the IPCC estimates of natural forcings, of indirect aerosols, of internal climate variability, and that you don't like the climate models. All because you "think the IPCC got that wrong". IMO your gut feeling simply doesn't measure up to the evidence.
jwalsh @86.
You write:-
I suppose I should be grateful that you implicitly agree to there being negative anthropogenic forcings. I know not what "kind of agree" is meant to men except when the difference between "kind of agree" and "agree" are made plain, something you fail to manage. But then it is dangerous ground being associated intellectually with the Viscount of Brenchley.
I would suggest that there is a contradiction hiding within something appering to be "...somewhat of a universal "fudge-factor", varying wildly." Of course, if it is "universal" I would assume that the Viscount of Brenchley has some use of it. And me. And you.
As negative anthropogenic forcings are not easily evaluated, their impact could easily be over-estimated as they could be under-estimated. And you say there is "evidence" of their decline. Where is that evidence? It appears to apply to a time "...into the late 20th century" which is a bit ambiguous but my interpretation suggests a period I haven't seen the slightest evidence for a decline, rather evidence for rapid rise.
And with that, I have dealt with less than 10% of your comment @86. I point out this proportion to demonstrate why some of the assertions within your comments will go without comment despite their complete lack of veracity.
jwalsh,
It is worth pointing out now that the IPCC positions are a consensus low amount of warming and/or damage. The majority of scientists in many fields think the IPCC projections are too low. None of the IPCC projections are thought to be too high. We use the IPCC projections here for debate because it gives us a reasonable starting point. Claiming without any data that they are much too high is a waste of time (trolling).
For example, sea level rise has always run at the very top of IPCC projections. Alternate methods of estimation of sea level rise are double the IPCC projections. Arctic sea ice has run ahead of almost all projections and is currently 50 years ahead of AR4.
Your suggestion that the IPCC is too alarmist without any data to support your claims is simply uninformed. Your posts have become longer and more disjoint. At the same time your claims have become even more extreme. Perhaps you need to rethink where you are getting your ideas and see if they have any data to suport your wild claims.
The LIA was a local event, not a global event. There is no trace of it in the reconstructions of global temeprature. You are incorrectly applying European and North American temperature records to the globe. You look uninformed when the only support you have for your claims is incorrectly applying a local event to a global discussion.
Claims made without data are easily dismissed. You are fortunate that Tom and MA Roger are so patient with you.
jwalsh @86:
It goes tiresome correcting the errors, lack of evidence and outright falsehoods on which you base your "expert opinion". Never-the-less, here are the results of six near equatorial ice cores from high altitudes:
Here are three of the tropical or subtropical icecores along with three polar icecores:
And here the equivalent ice core (in blue, dO18) from Mount Kilimanjaro, which at 3 degrees, 3.5 minutes south, I think counts as being "at the equator":
You will notice that only Sajama has, what might be considered to be, your 1,200 year cycles. You will further notice the distinct hockey stick in the 6 ice core composite.
Further, I refer you again to the Marcott et al (2013) reconstruction of holocene temperatures, as displayed above along with eight temperature proxies and their arithmetic mean as constructed by Robert Rohde for wikipedia:
Again, the Roman Warm Period and the Minoan Warm Period, not to mention the 1,200 year cycles are only present in GISP2, and is distinctly not present in the global reconstructions.
The RWP and MWP are distinctly North Atlantic phenomenon, and have significant impact over European temperatures. That they do not have any discernible impact on global temperatures is a spear in the side of any theory that modulation of North Atlantic Temperatures is a significant, let alone a major cause of variance in global temperatures.
So:
Evidence the sky is blue? None provided either, and none needed because it is assumed to be well known by anyone well informed on the topic as you claim to be.
But exclusively restricted to NA (and immediately neigbouring land) proxies showing beyond doubt that they are regional, not global variations in temperature. As we are discussing impacts on global temperatures, your introduction of a known regional temperature proxy with poor correlation with other regional temperature proxies counts as a red herring at best - and is either proof that you are not well informed on the topic, or that you are intent on deception (if you are indeed well informed).
michael sweet @91, that is not a fair statement of the consensus. There are a significant core of climate scientists who do think that the IPCC, in general underplays the problem. There are also a substantial majority that think it gets it about right, and a large number who think it overestimates the problem by a small degree. Some of those (such as James Annan, and John Nielson-Gammon) are clearly very competent scientists who are following the science as best they understand it, as indeed are those on the other side of the coin. Further, some IPCC projections are clearly high including global temperature increase, which runs about 15% below projections even after accounting for ENSO. For other observations, IPCC projections are low (as with reduction of Acrtic Sea Ice extent, and sea level rise).
My problem with jwalsh is not that he thinks the IPCC has overestimated the problem, and more specifically the attribution level. His position is a consensus position. My problem is that he does so based on either no, or clearly misrepresented evidence. If you are going to disagree with the IPCC, you should do so scientifically, which clearly he does not.
This would be a distinction without a difference for the purposes.
That the climate has varied wildly in the past is not "out of left field". It is considered to be more established scientific fact than most IPCC statements. The Minoan, Roman, and Medieval warm periods occurred at roughly 1200 year intervals. I acknowledge that the GISP2 is a local record. It is not, however, the only record. And a person would need to describe some proposed mechanism of extreme arctic warming and cooling cycles independent of the rest of planet earth to speculate that it was local. Could there be such a thing? I am not sure. I would be curious to hear one.
Considering that there is strong observational evidence to support an ECS estimate of below 3, I am not alone. In fact, I suspect that a FAR greater percentage than 3% of the IPCC themselves would agree there. One need only pick through the IPCC expert reviewer comments to easily demonstrate a lack of clear agreement. I think that the PDF cobbled together from the 10.5 figure is getting an inordinate amount of attention. It would be similar to deciding that any particular PDF for climate sensitivity itself was the "correct" one. As observed my many, AR5 seemed to weight model methods of determination of ECS over observational ones with little justification. This would be, incidentally, in contrast to AR4 with no well-documented reasoning.
If you'd like to discuss what is a statistically valid time period, that would be an interesting discussion. Certainly a great deal of ink has been made over the statistically much more "valid" 18 year period of 1980-1998. Do we really need to wait just a few years to consider an 18, 20, or 25 year trend? The IPCC thinks not. How many people have drawn trend lines over CRUT4 data or UAH satellite data? However, if 15 years is too short of a time period to consider making model adjustments, those actually doing so, such as those running them, are being hasty. And the IPCC themselves would be being "hasty" in using short term temperature data to alter their own 20-year temperature projections well below that of the model outputs. And they did exactly that in AR5.
It's worth noting here, that I don't think the IPCC attribution is "way off". Just minorly so. And for the record, I don't rule out climate sensitivity of "4" either, or indeed attribution of AGW of 120%. How would I know? As I said, greater effort would probably spent worrying about convincing the much smaller fraction of scientists who put it well below 50%.
Then don't? You made the choice to reply to me at all or not. But if it's your job to do so, there are worse things to be doing, and every job can get tiresome at times. The six ice-cores mentioned come from Lonnie Thompson's paper. Interestingly, it's referenced as often by those wishing to show evidence of a LIA and MWP as often as it is to try to show a lack of both. Most that have looked at that the data have deemed it too noisy to say anything about either, which allows people to interpret it however they like. It's also trotted out as corroboration of a particular set of reconstructions, that seem to be moderated if mentioned, or critiqued in any way. I mentioned that I didn't want to delve too deeply into paleo reconstructions for the same reason. The existence or not of the LIA/WMP is a point of contention. Due to inconvenience, there are many trying to suggest that they were only European, or Northern Hemisphere, or didn't happen at all... etc. etc. There is evidence that both were global. And there is evidence that it wasn't. Part of the problem is that most studying it over the years seem to do so in the north. The NCDC of NOAA thinks the Greenland, Antarctic and other Arctic cores suggets both well enough. I wish we truly did have a great source of data spanning the globe and time with high precision.
As for Marcott 2013? I think Gavin Schmidt summed it up well.
http://www.realclimate.org/index.php/archives/2013/03/response-by-marcott-et-al/
"This is not the paper you should be interested in to discuss the details of medieval/modern differences. Given the resolution and smoothing implied by the age model uncertainties, you are only going to get an approximation."
And the vaunted academic source for all things climate "Wikipedia" (does Wm. Connolly still babysit it like a hawk on methamphetamines?) .... Hmmm. Not sure what the scientific utility is of averaging multi-proxy studies together. It gives rise to interesting features though. Such as it being cooler in 2004 than it was 8,000 years ago. I'd check the math myself, but the premise itself is too absurd to bother.
However, back to attribution. Nothing to say about the IPCC experts on attribution downshifting estimates of future temperatures, setting aside temperatures at the mean or above from climate model projections as being unlikely? That would seem to be pretty relevant.
Jwalsh, I note that you do not comment on any of Tom Curtis's arguments at #92, starting with the equatorial ice cores canard. I am unimpressed with your contribution so far.
Well, it seemed you did, kinda, or not, I'm not sure. I find it rather amsing that you'd adopt such snark toward Wikipedia but maintain deference for a buffoon like Monckton.
http://www.parliament.uk/business/news/2011/july/letter-to-viscount-monckton/
jwalsh - " Nothing to say about the IPCC experts on attribution downshifting estimates of future temperatures, setting aside temperatures at the mean or above from climate model projections as being unlikely? That would seem to be pretty relevant."
Perhaps you might point us to what you are precisely what you are implying but "relevant"? It seems you are yet again confusing sensitivity with attribution. Why do you continue to ignore the points the about OHC? Attribution is about sorting which cause created a particular change. Show us your evidence for another cause or stop trolling.
jwalsh @95:
1) I criticized you for using a regional proxy (GISP2) as though it were a global proxy. You implied your use was justified on the basis that tropical ice cores did not exist "they don't last so long". You now claim that you knew about them all along, which makes your original use of GISP2 simply dishonest.
2) I do not discuss the MWP or LIA with reference to those ice cores. Rather, I discuss the absense of evidence of evidence for a RWP or Minoan WP, which you claimed exist on the basis of GISP2. Clearly from the ice core data they were not global events - and changing the topic does not make them so.
3) From Marcott et al, we learn:
If approx 50% of variability is preserved at 1000-year periods, >50% of a 1,200 year cycle would show up in the reconstruction. It is, however, not there. No amount of quoting Gavin Schmidt out of context will change that.
4) I criticized you for (in effect) taking the average of just one proxy as an indicator of changes in global mean surface temperature. You now respond by arguing that taking the average of eight such proxies is of dubious "scientific utility" and that it is a premise that is itself " itself is too absurd to bother" checking the maths. That you so argue in order to maintain that the data from the once proxy is a reliable guide to the timing and direction of trends in GMST (if not their magnitude) shows how absurd your position is, and how completely lacking in scientific merit.
As an aside, I did not credit the graph to wikipedia regardless of your misrepresentation. I sourced from wikipedia, and acknowledged the source as is required by copyright law (as they wave their copyright on condition that you credit the source). However, I cited Robert Rhode, the author of the graph. That you ignore that to play your empty rhetorical games is only to be expected from a troll.
I am showing "deference" to Monckton by agreeing with a single thing the man said? I simply attributed the comment to him. I've read very little about, or by him, but happened to have done so recently enough. Should I? As for Wikipedia. Surely you know what it is. Wikipedia can be convenient for some things, but you almost always better check the references. The quality of information on it varies wildly. I usually avoid it for anything other than a starting point when possible. YMMV.
I've not once confused sensitivity with attribution. But if you change how much warming is assessed by CO2, it follows that the relative attribution of the warming is altered. I mostly ignored the points on OHC, because I am trying to stay on topic (attribution). OHC seems to me to be about where the heat is going, not what is causing it in the first place. Some are speculating issues with assumptions. And these "some" are many in the IPCC. Over-estimates of CO2 sensitivity could be one of those reasons. If there's nothing wrong with the attribution percentages, but only that the heat is all going into the ocean. That's fine, then the models have to be adjusted accordingly for the policy-critical estimates of surface temperatures. But I don't think there's complete agreement that that is indeed what we are seeing.
As for trolling. I am trying not to. If you don't understand the "why's" of those thinking somewhat differently, you stand little hope of convincing them, should you think that important.