Thanks all - lots to read! A few quick responses:

Paul Matthews - yes indeed, I was unaware of the papers you mention, as palaeo is not my field - thanks for responding to my question. Particularly interesting to see the two versions of the Loehle paper. And thanks to Philip too. I also found Mann et al 2008.

BBD - I hadn't seen the Fang et al either. Seems like a strange and confused paper - for example, figure 7b seems odd and pointless (why plot annual increments in CO2 against temperature anomalies - nobody "in the consensus" thinks that temperature responds to CO2 on an annual basis, but it is recognised that there is a relationship the other way round (ie: interannual variability in CO2 concentration rise varying in response to interannual variability in global mean temperature - my colleague Chris D Jones has published on this, and it's one of the reasons to think that feedbacks between climate change and the carbon cycle may begin to increase the airborne fraction of CO2 emissions - but that's off topic because my original post was about evidence for past change and its causes, not projections of future change). Also, oddly, Fang et al cite 3 papers (their refs 85-87) in support of their statement on estimates of "a lower radiative forcing for CO2" when in fact the papers they cite discuss climate sensitivity (not RF). IMHO I don't think it really makes a particularly strong case for a new "consensus".

Justice4Rinka - I completely agree with you on economic "predictions". Your point (correct me if I'm wrong) seems to be that a pro-active approach to emissions reductions is not needed because such reductions may happen anyway for other reasons. Again, socio-economic scenarios are not my field, but given that emissions are generally continuing to rise, I think it is a reasonable thing to do to try to estimate the consequences of a continuation of this. The question of whether the ongoing rise in emissions will halt and reverse with or without a pro-active approach is a different question. Actually the emissions scenarios used in IPCC AR4 do include scenarios of a peak and decline in emissions even without specific climate policy (with the peak occurring several decades into the future), and we do estimate the consequences of those scenarios with climate models.

Dec 1, 2011 at 10:20 PM | Dung

Thanks - although (slightly pedantic point) my understanding was that McIntyre and McKitrick didn't claim to have actually generate their own alternative reconstructions, they just commented on existing ones.

Richard

Please correct any and every error you find in the following:

You appear to say that the Hockey Stick "stands" because there is no other paleo climate reconstruction currently available. I suggest that is a totally non scientific statement, it is also incorrect.

Prior to the Hockey Stick (Mann, Bradley and Hughes), the IPCC had agreed that the true paleo climate reconstruction included the Little Ice Age and the Medieval Warm Period.
McIntyre and McKitrick showed that tree ring data, short centering of data and relacement of proxy data by actual data (when the proxy data stopped agreeing with actual temperatures) showed the Hockey Stick to be a farce.
Other scientists that you listed produced other hockey stick reconstructions but they used the same flawed data that MBH had used.
Since all these Hockey Sticks are discredited then what "stands" is the original IPCC reconstruction which showed the Little Ice Age and the Medieval Warm Period. This reconstruction is also supported by historical records which The Hockey Stick is not.

I think the point that Fang et al. are making is that there is no strong case for a firmly held consensus which extends our knowledge far beyond what Richard had already agreed was probably the current consensus.

There is always a consensus set of knowledge, the minimal set that almost every scientist can sign up to. Fang et al. are pointing out that there is generally accepted to be a lot more uncertainty in most of climate science than had previously been acknowledged in e.g. AR4.

Compare my previosuly stated understanding of the consensus as I outlined to Richard Betts (who agreed with it)

1) global warming is real (whatever that means)
2) man is having an impact
3) at least some of the impact is likely caused by CO2 emissions

to the summary contained in Fang et al.'s abstract

(1) climate warming occurs with great uncertainty in the magnitude of the temperature increase;
(2) both human activities and natural forces contribute to climate change, but their relative contributions are difficult to quantify; and
(3) the dominant role of the increase in the atmospheric concentration of greenhouse gases (including CO2) in the global warming claimed by the Intergovernmental Panel on Climate Change (IPCC) is questioned by the scientific communities because of large uncertainties in the mechanisms of natural factors and anthropogenic activities and in the sources of the increased atmospheric CO2 concentration.

I would say that these two views are compatible.

Dung

Or it could be ruined by stupid, provocative behaviour like yours.

Think.

Dung

To be clear, the above was in response to this, from the previous page:

This very interesting thread could be ruined by the appearance of BBD.

Richard Betts

Many thanks for taking the time to look over Fang et al.

I have to say that my view is a jaundiced one. I see the study as politically positioned (or directed, if you prefer), and timed to coincide with Durban.

Others here do not agree, and there has been some nasty language thrown about, but I stand by this. Especially as you confirm my own view of this paper: it is weak, even actively misleading.

Hi Dung

I didn't use the term "Hockey Stick" and I mentioned MBH99 in with a number of other papers, all as context for my citation of the AR4 conclusion that “It is likely (66%) that [the second half of the 20th Century] was the warmest Northern Hemisphere period in the last 1,300 years”.

AR4 does say that the Medieval Warm Period existed, with "warm" being relative to the last 2000 years but (probably) not the last few decades:

The evidence currently available indicates that NH mean temperatures during medieval times (950–1100) were indeed warm in a 2-kyr context and even warmer in relation to the less sparse but still limited evidence of widespread average cool conditions in the 17th century (Osborn and Briffa, 2006). However, the evidence is not sufficient to support a conclusion that hemispheric mean temperatures were as warm, or the extent of warm regions as expansive, as those in the 20th century as a whole, during any period in medieval times (Jones et al., 2001; Bradley et al., 2003a,b; Osborn and Briffa, 2006).

This is preceeded by:

The uncertainty associated with present palaeoclimate estimates of NH mean temperatures is significant, especially for the period prior to 1600 when data are scarce (Mann et al., 1999; Briffa and Osborn, 2002; Cook et al., 2004a). However, Figure 6.10 shows that the warmest period prior to the 20th century very likely occurred between 950 and 1100, but temperatures were probably between 0.1°C and 0.2°C below the 1961 to 1990 mean and significantly below the level shown by instrumental data after 1980.

In order to reduce the uncertainty, further work is necessary to update existing records, many of which were assembled up to 20 years ago, and to produce many more, especially early, palaeoclimate series with much wider geographic coverage. There are far from sufficient data to make any meaningful estimates of global medieval warmth (Figure 6.11). There are very few long records with high temporal resolution data from the oceans, the tropics or the SH.

The full discussion is here

The reason I asked for pointers to other reconstructions was to help me (as something of an outsider to the field) see whether there appears to be any credible arguments against the AR4 conclusion I cited above. Since I'm not involved in paleaoclimate reconstructions myself, I'm not in a position to make pronouncements either way - I'm just interested because it provides useful context for my own work.

It was interesting to see the Loehle paper and also McShane and Wyner. I see the latter has been the subject of a huge amount of discussion which will take me a while to get through (as will the paper itself, as I've only skimmed it so far). I see that Loehle was re-written after apparently having had comments from Gavin Schmidt, Steve McIntyre and CA readers, and the conclusion is that the warmest 3 decades of the MWP were "warmer than" the last 3 decades "but not significantly so". I also found a comparison of several reconstructions including Loehle, Ljungqvist and Mann 2008 & 2009 at skeptical science, which notes that all the reconstructions show the MWP, LIA and 20th century warming. (Sorry if this is going over old ground for many people here!)

It will be interesting to see the AR5 conclusions on the last 2000 years. Incidentally, the First Order Draft of the Working Group 1 report will be available for review in 2 weeks, and you can sign up to be a reviewer here.

Dec 2, 2011 at 6:37 AM | matthu

I think the Fang et al conclusions do have some differences to your 3 points on the "consensus"

The most important difference is between the two points (3). You say:

at least some of the impact is likely caused by CO2 emissions

and Fang et al say:

the dominant role of the increase in the atmospheric concentration of greenhouse gases (including CO2) in the global warming claimed by the Intergovernmental Panel on Climate Change (IPCC) is questioned by the scientific communities because of large uncertainties in the mechanisms of natural factors and anthropogenic activities and in the sources of the increased atmospheric CO2 concentration.

Although the two statements are not inconsistent because the Fang view falls within the wider range of views expressed in your statement "at least some of the impact is caused by CO2", the two statements are not identical. Your statement is something that I think pretty much everyone agrees with, but I think the Fang claim that "the dominance of CO2 is questioned by the scientific communities" is (IMHO) rather over-stating things. Also, the Fang claim that the cause of CO2 rise is questionable is bizarre - as they themselves show in their Figure 6, the addition of CO2 to the atmosphere from fossil fuel emissions and deforestation is twice as large as the actual increase in CO2 in the atmosphere. There is therefore more than enough anthropogenic CO2 emissions to account for the change seen in the atmosphere (the other half is being taken up by a net sink in the land biosphere and oceans).

So I am with BBD in not buying the Fang paper!

Richard,

Thank you for your response.

I am not ready to let go of question 1. Chambers dictionary defines "most" as "the greatest part, amount or number (of)". The 2007 IPCC AR4 report means to say: "51 % or 60% or perhaps it is 70 %, maybe 55% or some other figure above 50% of the observed increase in global average temperatures since the mid-20th century is very likely due to the observed increase in anthropogenic greenhouse gas concentrations." "Very likely" means a confidence level of over 90%. Do you wish to amend your initial response?

You asked about D'Arrigo. Well, according to her testimony to the NAS hearing 2006 she may be willing to "cherrypick" data.

There is a post by McIntyre that you might find interesting, perhaps helpful.:"D'Arrigo et al on Bristlecone calibration" Feb 11 2006. This post compares two papers, Briffa and Osborne 2006 with D'Arrigo et al 2006. The discussion is about the use of bristlecone and foxtail proxies. McIntyre was unable to replicate the claimed B&O correlation of (decadal) 0.56 between the MBH 1998 PC1 ( with its weight on bristlecone data) and the relevant gridcell temperature. He quoted B&O: "We only used proxy records that are positively correlated with their local temperature observations...". Support for McIntyre's conclusions came from D'Arrigo et al.McIntyre pointed to the fact that they discarded bristlecone data and foxtails. They said:" The regional chronologies were also screened by comparisons with instrumental(local and larger scale) temperature data.... In so doing some potential data were discarded due to ambiguous signals..."

I have copied and pasted some of McIntyre's comments as IPCC reviewer which might interest you.

"Without acknowledging a source, IPCC AR4 conceded that the reconstructions shown in its new spaghetti graph were “not entirely independent” [carefully not mentioning bristlecones or Yamal]:

As with the original TAR series, these new records are not entirely independent reconstructions inasmuch as there are some predictors (most often tree ring data and particularly in the early centuries) that are common between them, but in general, they represent some expansion in the length and geographical coverage of the previously available data (Figures 6.10 and 6.11)

Citing IPCC AR4, Jones et al 2009 now concedes that “several chronologies” have been utilized in “virtually all published studies”, again carefully not mentioning bristlecones or Yamal:

Several chronologies extending over a longer time span, with variability displaying a strong and direct association with changing local temperatures, have been utilized in virtually all published studies aimed at reconstructing Northern Hemisphere (NH) or global average surface temperature changes during the millennium leading up to the present (Jansen et al., 2007[IPCC AR4]).

Indeed, nowhere in their entire review dp Jones et al 2009 mention the word “bristlecone”.

Note that Jones et al 2009 substantially embellished the actual IPCC AR4 wording. IPCC AR4 did not state that these repetitively used chronologies had “a strong and direct association with changing local temperatures”. Indeed, Review Comments are against this claim. As an IPCC AR4 reviewer, I contested the use of bristlecone/foxtails in a spaghetti graph of proxies supposedly qualified against local temperature, observing (See SOD Review Comment 6-1144) in respect of a series from Osborn and Briffa 2006 (which claimed a correlation of 0.19 – itself hardly “strong and direct”:

I checked the correlation of this data to HadCRU2 gridcell temperature and only obtained an insignificant correlation of 0.04. The authors said that they had cited the temperature data incorrectly, that they had actually used CRUTEM2 yielding a correlation of 0.19 and that HadCRU2 data was spurious in its early portion (1870-1887) because there was no station data. However there is station data at GHCN going back to the data in HadCRU2. D’Arrigo et al 2006 considered using foxtails and rejected the use of this data because it did not meet standards of being correlated to gridcell temperature, expressed in very similar terms to Osborn and Briffa 2006. The contrasting views of D’Arrigo et al 2006 certainly establish that the relationship is “ambiguous” and that this proxy should not be used on multiple grounds.

Notwithstanding the fact that the actual correlation was 0.04 (as could be easily calculated), IPCC rejected my call that they not be shown in a spaghetti graph of temperature proxies as follows:

Some of what the reviewer says may be true, but is as yet unpublished and the current review is based on multiple strands of evidence, among which the results of Mann and colleagues remains relevant.

Whether or not I had published the incorrectness of the claimed Osborn and Briffa correlation is irrelevant. My calculation was correct and the authors either knew it or should have known it. (And, needless to say, IPCC did not require that some point be published if the opinion was adverse to us e.g. Ammann’s secret emails to Briffa, withheld from the IPCC Review Process.)

From this weak and contested gruel – an actual correlation of 0.04 to local temperature – Jones and the extended Team have ratcheted the argument in favor of the non-independent proxies to a claim of a “strong and direct association with changing local climates” for these non-independent proxies."

The list of studies you cited, Richard, contains a number that use bristlecone and foxtails as proxies as well as the D'Arrigo study. McIntyre and D'Arrigo are right about bristlecone and foxtails. I don't think you can cite both D'Arrigo and other studies that include foxtails and bristlecones.

According to McIntyre, D'Arrigo et al has its own problems. I have copied and pasted below.

"While there’s lots to complain about in D’Arrigo et al [2006], it is both a better and a more candid study than Briffa and Osborn [2006]. I had started a post on this a couple of weeks ago, but it got overtaken by getting ready for NAS. You remember about “bring the proxies up to date! Well, D’Arrigo et al. carries proxies up to 1995 in a decent sized network. After 1985, their verification fails, so they go through some contortions to re-calibrate up to 1978 and make some (IMO) unsupportable statements about the relationship between medieval and modern periods, given the out-of-sample failure. But they at least showed the divergence in their graphic (rather than snipping it off a la Briffa). Here’s a graphic from DWJ06 (which D’Arrigo showed.)

The discrepancy between the forecast and the actual caught Cuffey’s eye and he asked D’Arrigo about it. She said “Oh that’s the “Divergence Problem”‘?. Cuffey wanted to know exactly how you could rely on tree ring proxies to register past warm periods if they weren’t picking up modern warmth “questions dear to the heart of any climateaudit reader. D’Arrigo explained that it had all been discussed by Briffa et al. I think that D’Arrigo said that the "divergence problem" only applied to a few sites.

I’ve discussed Briffa’s approch to the "Divergence Problem" before – see for example here – and we modified our presentation to respond to concerns about the "Divergence Problem". First, we showed the following graphic, reporting that the "Divergence Problem" was not limited to a few sites, but applied to an entire network of 387 sites selected to be temperature-sensitive."

I have not looked much for Cook's research. I have noted a comment he is supposed to have made, suggesting that the divergence problem applied only to the 20th century. How would he know it was not happening in the MWP? Also he was co-author with Esper 2002 where there are foxtails present and a failure to disclose data.

You ask whether there are reliable reconstructions. From memory I think McIntyre has found only three reconstructions that do not contain foxtails or bristlecones. One of the studies is Jones 1998 which has major issues. According tot McIntyre, "the hockey-stickness of J98 depends on the extension of the Polar Urals data beyond its range of confidence ( and then relying on three incorrectly dated cores which i discussed earlier)". There is much more that McIntyre has to say about Jones 1998, which provided one of the graphics for the WMO and IPCC TAR (the "hide the decline" graphic) if I remember rightly.

I will look for the other two reconstructions when I can discover again McIntyre's refernce to them.

There are major problems with temperature reconstructions as they are done now. The divergence problem has not been explained. It is not limited but ubiquitous. There is interaction between temperature and water that affects growth and makes impossible the task of teasing out the effect of temperature without detailed information about rainfall that does not exist. It seems possible if not likely that what the reconstructions show are local effects not global.

There are or should be questions about the competence and honesty of many of those undertaking temperature reconstructions. I say that based on my industrial relations experience not scientific knowledge and can post on it if you are interested. Meanwhile, I do not think you have produced sufficient evidence to show that temperatures today fall outside the range of natural variability. Do you think the temperature reconstructions you have cited match historic evidence?

I notice that Ljungquist also affirms the agreement between the more recent reconstructions:

The minor differences compared to, for example, the reconstructions by Moberg et al. (2005) and Mann et al. (2008) can easily be attributed to differences in geographical data coverage and other statistical treatment of the data, although the substantially larger variability c. AD 300-800 in our reconstruction probably reflects a genuine signal originating from the use of more data.

The paper also affirms that "the post-1990 warming in the instrumental record still seems to be outside the range of natural late-Holocene temperature variability" but tempers this with the following comment:

Since AD 1990, though, average temperatures in the extra-tropical Northern Hemisphere exceed those of any other warm decades the last two millennia, even the peak of the Medieval Warm Period, if we look at the instrumental temperature data spliced to the proxy reconstruction. However, this sharp rise in temperature compared to the magnitude of warmth in previous warm periods should be cautiously interpreted since it is not visible in the proxy reconstruction itself.

I'm tempted to suggest a determination here to avoid hiding a decline in the incline, but I don't think I'll bother. I think the point is that the millennial variations seen in the reconstruction may still be too small. Lovejoy and Schertzer pick up on this as well. Their figure 9 shows the scaling analysis for the 20th C instrumental record, and for pre- and post-2003 reconstructions. Notice that:-

1/ The instrumental and post-2003 reconstructions exhibit similar scaling, in distinction to the pre-2003 reconstructions. This supports the conclusion that the pre-2003 reconstructions do not show adequate variability.

2/ The instrumental curve is displaced to the left of the post-2003 reconstruction. Lovejoy and Schertzer draw attention to this but conclude that "the instrumental and reconstruction discrepancy in fig. 9 thus remains unexplained." However, it may also be possible that the discrepancy supports the suggestion that the amplitude of variations in reconstructions is too small.

Richard

I think you should consult McIntyre and his readers about Mann 2008. Mann 2008 uses bristlecones. He claims his reconstruction is robust to the exclusion of bristlecones but that has been found to be because of the inclusion in the data of the Tiljander series. This is series of lake sediments. They are not proxies for temperature and are known not to be. Tiljander said so. They arise from the human activity of building ditches. Why any climate scientist should use as a temperature proxy data that is stated in the literature not to be reliable as a proxy or not to be a proxy at all escapes me. Any suggestions?

Thank you for your reply, Richard.

Just to clarify: Previously I stated that

2) man is having an impact
3) at least some of the impact is likely caused by CO2 emissions

To which you replied

Thanks - yes, to my mind that pretty well captures the bits that virtually everyone who knows about the science can agree on.

Now it seems you are reverting to a position that the dominance of CO2 is as yet unquestioned by the scientific communities i.e. CO2 is dominant over other GHG and over all other forces including both natural and manmade forces and both previously considered and unconsidered forces?

Bearing in mind that my statement was a present view and looking forward, are you not revising your stance here?

Hi Sam

I am not ready to let go of question 1. Chambers dictionary defines "most" as "the greatest part, amount or number (of)". The 2007 IPCC AR4 report means to say: "51 % or 60% or perhaps it is 70 %, maybe 55% or some other figure above 50% of the observed increase in global average temperatures since the mid-20th century is very likely due to the observed increase in anthropogenic greenhouse gas concentrations." "Very likely" means a confidence level of over 90%. Do you wish to amend your initial response?

Not really! I think my initial response is supported by the points (i) to (iv) I made following it. We've had a lot of discussion relating to point (iv) but even then I remain unconvinced that there is anything wrong with my statement that "Changes over the last century appear to be unusual in the context of the last millennium". And that's only one point out of 4.

But on the subject of (iv) (the rest of your post, and the following one) I will indeed continue to read up on both sides of the argument here.

Hi matthu

No, I don't think I'm changing my position here. Personally I still stand by the conclusions of my AR4 chapter that CO2 RF is the largest, and I think the majority of the community accepts this too, but as for "something everyone can agree on, your statement still holds.

The difference is between "majority" and "everyone".

Fang are correct to say that some people question the CO2 dominance, but I think they give a false impression of how widely that view is held.

So like I say, Fang's statement is compatible with yours since the former view represents a subset within the latter.

Thank again Richard.

My original statement was an attempt to define the current consensus opinion.

If it is your view that AR4 still defines the consensus opinion: i.e. that there is an unchanged consensus opinion that has not advanced in the last 4 years, then I think we no longer agree.

If on the other hand if you are saying that you still adhere to AR4 even though there is no longer a consensus, then I can accept that.

matthu

Have you looked at Lacis et al. (2010)?

Science brief here.

Discussion with Lacis at Judith Curry's blog here.

More on Dot Earth here (two-part article).

Full paper here (1.8MB PDF).

@ Richard

Your point (correct me if I'm wrong) seems to be that a pro-active approach to emissions reductions is not needed because such reductions may happen anyway for other reasons.

Not entirely - my point is kinda twofold. First, the uncertainty in the values of these three particular inputs (and all the others too most likely) is such that nobody knows what they may be. Thus the usefulness, in CO2 abatement terms, of applying taxes to fossil energy now simply cannot be computed - beyond saying that its effects will be trivial, upon which I think everyone is agreed. Will a tax of £10 a gallon deter people from driving in 2061? If it costs £1 otherwise, maybe. If it costs £1,000 already, not so much. If it costs £1 because everyone has stopped driving cars, again, not so much. So what's the point?

The second point is the obverse, which is whether anyone has worked outwards from the assumptions used to test the plausibility of what they imply. The example I gave above was of the newsagent who extrapolates a future demand of 4 billion copies of the Radio Times magazine a week based on three weeks' sales. Nothing wrong with the arithmetic, nothing wrong with the apparent trend, but his mistake was not to consider what his assumptions implied. A print run of 4 billion, for example, of a magazine that's useful in a country of 60 million. An unmeetable need for extra trucks, roads and paperboys.

So what energy price is implied by emissions being the same in 100 years' time as now? About the same? Hmmm. That graph suggested that has always been a bad assumption. Does anyone knowledgeable agree with this implied price? What population level is implied? Are we good at predicting population over 100-year timescales? If not, how do we check the implied population inference for plausibility?

We already recognise the weight of this sort of argument when considering the misnamed sustainable fuels. An acre of arable land yields a tonne of biodiesel. The UK uses 90 million tonnes of liquid oil a year, so to replace this with biodiesel, we need to convert 180 million acres to biodiesel use. The trouble is, the UK has only 66 million acres, and only about a quarter are arable land, and we'll need to replace the food grown - so we need 180 million across plus the 16 we've taken out of food use. About 200 million acres, not the 16 we have.

It is thus clear that a move to rely on biodiesel can't happen. If one projected that in 100 years' time we will do so, the inference is either GM crops or war with Ukraine.

So it's a simple question really - do these 100-year models imply anything patently stupid, or indeed anything that proves so when studied? Who's in charge of fact-checking the plausibility of energy prices in 100 years' time? Because where I come from this is not an intellectually respectable pursuit.

Much discussion of paleoclimate.

A point that's often made, but bears repeating, is that the causes of the MWP are not those of modern warming. The current hypothesis is that increased TSI and reduced volcanism might have triggered the MWP, and a persistent, teleconnected La Nina and positive NAO may have stabilised and maintained it.

Observed changes in TSI and volcanism in the C20th do not explain the pattern of climate change post-1950.

We aren't having a re-run of the MWP. So something else must be causing energy to accumulate in the climate system.

The only plausible candidate is CO2.

Richard

I know you guys are busy with AR5 and all that but I wondered if you had regular contact with Sarah Ineson and might ask if she would be willing to drop by and give us some thoughts on UV, NAO and ENSO and perhaps explain where the research is heading after the publication in October. How has it been received in the CS community as a whole etc?
Let her know that most of us don't bite and are genuinely interested.

Sam, you are making a valiant effort but it is doomed to failure. Richard was a lead author on AR4 and will be on AR5, so he is never going to admit that one of the key claims of AR4 is unjustified, no matter how convincing your arguments are.

Here's how Tallbloke asks the question:
"If natural variability during it’s negative phase in the C21st is capable of balancing the forcing you claim for the increase in co2 since 1940, then how much did it contribute to the warming during it’s positive phase in the late C20th?"
To put it another way, if 'natural variation' or whatever, is currently compensating for the warming, they must be 50/50. So how can the warming in the 20th C be mostly CO2?

BBD
See if you can spot why I have a problem ...

A point that's often made, but bears repeating, is that the causes of the MWP are not those of modern warming. The current hypothesis is that increased TSI and reduced volcanism might have triggered the MWP, and a persistent, teleconnected La Nina and positive NAO may have stabilised and maintained it.
Observed changes in TSI and volcanism in the C20th do not explain the pattern of climate change post-1950.
We aren't having a re-run of the MWP. So something else must be causing energy to accumulate in the climate system.
The only plausible candidate is CO2.

What you are doing is listing a series of reasons that may have caused the MWP and then definitely dismissed them as possible causes of the late 20th century warming.
I don't necessarily disagree with the assessment but what you are saying is you don't know what caused the MWP but whatever it was it can't have been what caused the late 20th century warming.
That is a leap of faith not a scientific deduction. So either you have missed out a step in your working or you are guessing.
And what caused the early 20th century warming?

Paul Matthews

To put it another way, if 'natural variation' or whatever, is currently compensating for the warming, they must be 50/50.

Or looked at another way, we've got a decade with La Nina dominant and a significant reduction in solar activity and rather more volcanic stratospheric aerosols that we realised* and GAT is... flat. That rather suggests that CO2 is a non-trivial climate forcing.

Surface temperatures are rising, according to BEST, although they have cooled several times in recent decades. Clearly care should be taken not to over-interpret short trends. The bigger picture is more informative.

So how can the warming in the 20th C be mostly CO2?

Because natural variation varies. CO2 forcing is gradually emerging as the dominant upward influence on GAT. There were, are and will be periods of flat GAT or even weak cooling as natural variation offsets warming forced by CO2. But the multi-decadal trend is upward. See the big picture.

*Major influence of tropical volcanic eruptions on the stratospheric aerosol layer during the last decade, Vernier et al. (2011); GEOPHYSICAL RESEARCH LETTERS, VOL. 38, L12807, 8 PP., 2011 doi:10.1029/2011GL047563

http://www.agu.org/pubs/crossref/2011/2011GL047563.shtml

Mike

Well, you know what the mainstream scientific view is: nothing has varied enough to account for the accumulation of energy in the climate system except CO2.

Here's the comparison of radiative forcings from AR4 WG1 SPM. Look at the RF from CO2 relative to other forcings. Look at the net anthropogenic forcing (bottom).

Whatever caused the MWP, and whatever caused the 1910 - 1940 warming, it is either invisible to modern instruments or not happening now. The latter seems more likely because if somehow unobserved, this forcing would be acting in addition to the known RF from CO2. It would be much warmer than it is, and that would bugger up the calculations.

So far from a leap of faith, we have a scientific deduction: based on observations and well-established estimates of RF from CO2, the latter is the most likely cause of the warming post-1950.

BBD - thank you for the reference to Lacis et al. (2010).

While this paper is interesting, I regard it as a distraction from what I was discussing because it throws no light on the subject i.e. how can it have any bearing on whether CO2 is dominant when it does not consider factors other than GHGs and I am not convinced there exists any consensus today that GHGs are dominant.