Bishop Hill The calm after the storm
Dec 15, 2012 Climate: WG2 After the leak of the AR5 Second Order Draft, initial attention was directed at the IPCC's partial retreat on the solar influence on climate. However, the focus seems to have shifted overnight to weather extremes, with Pielke Jr tweeting up a storm.
1/2 Draft IPCC Ch2 bottom line on extremes: "generally low confidence that there have been discernable changes over the observed record"
2/2 con't ...on lack of trends in extremes, exceptions are trends seen in temperature extremes and regional precipitation (but not floods)
As Ryan Maue responded, the IPCC might have better expressed this as "we have high confidence that natural variability dominates any AGW influence in observed/historical TC records". Pielke continues...
IPCC AR5 draft shows almost complete reversal from AR4 on trends in drought, hurricanes, floods and is now consistent with scientific lit
Of course IPCC AR5 is a draft but the scientific literature it is reporting is available for all to see, and AR5 has it presented accurately
I have been critical of past IPCC assessments on the science of extremes. But after SREX and now IPCC AR5 SOD the IPCC is 2 for 2. Nice job!
So what happens now to those whose views on extremes run counter to IPCC? Are they now the new "deniers"? Somehow I doubt it ;-)
However, it's not all good news. Richard Tol notes that the WG2 first order draft was "unreconstituted alarmism". We will be kept busy.
Reader Comments (65)
Lance Wallace (9:20 PM) -
Figure 5.12. The text says "A 1000-year temperature reconstruction for land and ocean representing Australasia provides evidence that summer temperatures in the post-1950 CE period were warmest in the last ~580 years (PAGES 2k Consortium, submitted). Prior to this, the reconstruction is based on only three or fewer proxy series and less confident conclusions can be supported."
Remarkably, the reconstruction shows an error interval of ~ +/-0.3K even during the periods of three or fewer proxies, not appreciably different from the error in more modern times when presumably many more proxies are available. It will be interesting to see the derivation of error intervals when that article is published.
"The calm after the storm"
I wonder why certain usually calm tea cups now no longer need stirring?
You may also want to check on page 14-62 of AR5 re arctic ice: There is
44 high confidence that Arctic sea ice anomalies exhibit substantial interannual variability, so that ice loss or
45 gain in any particular year cannot be taken as an indication or absence of a long-term trend due to
46 anthropogenic forcing.
Does that mean that all the huffing and puffing about sea ice loss was not necessary?
Just asking.
Michael Hart (Dec 15, 2012 at 1:19 PM) asks
Try Ch2 page 34 Section 2.4.4 Upper Air Temperature
What is the point of the science, if the process is broken/never straight to begin with?
The IPCC's process is: 2nd round of reviewer comments->review->intervening steps by select few->final draft->press release->summary release->report release.
It should be 2nd round->review>lockdown->report release->summary release->press release
Point made by Jo Nova
The purpose of reviewer comments and banging-on about transparency etc, is to add to the credibility of the process. A credibility that is undeserved, as the same party line from the previous report is mostly intact in several places. Which means, the IAC recommendations have had no effect either.
This is borne by the IPCC's own release:
This is a well-known 'feature' of the IPCC system where authors exert final say, over the word of reviewers.
Point made by Donna
Thirdly, while the overall contours of the AR4 WG1 argument remain the same, 'confidence', especially for attribution, has markedly increased. How?
Point made by Judy Curry
It seems to me that the ladies the whole thing figured out here.
Can we break the present one a trifle faster, please?
"It was one of the most enduring hoaxes in history and fooled scientists into believing a crucial evolutionary 'missing link' had been found in England.
Now 100 years after the discovery of the Piltdown Man, a team of archeologists and anthropologists will finally be able to expose the truth behind the scam, and pinpoint who was responsible.
Dec 15, 2012 at 4:58 PM | Nullius in Verba
I don't see in that list of possible reasons for the discrepancy between model and observation that the theory might be plain wrong. Odd that they'd omit the most obvious possibility :-)
And I think Dr Pielke has it right: SOD the IPCC.
So, I think the interpretation goes something like this:
1) errors in observations: we are trying to model things for which we have inadequate data, hence an inadequate understanding of it's current/past state.
2)specific manifestation of natural variability in observed coupled atmosphere-ocean system: Our models are incomplete and the things we don't know about are important.
3) forcing errors included in the historical simulations: Not only can we not measure the variables that are model outputs, it turns out that we aren't able to quantify the inputs
4) model response errors: We needed to bury this in a list, so it's the last and the least. It's a simple way of saying that the models might be wrong (which we already knew, hence the list, but circular logic is fun, right?)
Circular logic? How about this for circular logic:
1. Conclude that [rising]* CO2 will cause rising temperatures.
2. Incorporate CO2 rises in a GCM not by adding CO2 but by adding an instantaneous drop in heat emitted from the model atmosphere to space
3. The model heats up, temperatures rise, since now more heat is coming in than out.
4. Conclude that rising CO2 will cause rising temperatures and proffer the model output as evidence.
Note that the wheeze in step 2 converts an internal process in the real climate system - the ground-level emission of CO2 which is spatially and temporally inhomogeneous - into an external step-change in the virtual one, a change which is usually, I believe, modeled as homogeneous over the spherical surface at the top of the model atmosphere. This is ‘climate forcing’, and I think it is beloved by computer programmers who can use it to incorporate no end of ‘factors’ by means of their presumed effect. But I fail to see that it ought to be beloved by physicists, or indeed by anyone who might have imagined that factors such as CO2 were actually modeled as participants in GCMs rather than as ‘external forcings’. The conclusions have been included in the models as inputs.
[* inserted 6.35pm BH]
Regarding the missing hotpot, I commented extensively in my review of Chapter 2 the First Order draft that a key table just happened to describe temperature changes of all atmospheric regions except the upper troposphere. In my review of the Second Order draft, I drew particular attention to the fact that, among others, they had failed to cite Singer, S Fred, (2011). Lack of Consistency Between Modeled and Observed Temperature Trends, Energy & Environment, 22, 375-406 or Douglass, D. H., Christy, J. R., Pearson, B. D. and Singer, S. F. (2008), A comparison of tropical temperature trends with model predictions. Int. J. Climatol., 28: 1693–1701. I felt both these papers should have been cited in Section 2.4.4. Also, in view of the firm prediction of the models quoted in AR4, that the upper troposphere would warm faster than the surface or lower troposphere, Table 2.8 had to included some estimate of upper troposphere trends, not be silent on the topic. The only real mention of the upper troposphere occurs in the conclusion to this section:
"Based upon multiple independent analyses from weather balloons and satellites it is virtually certain that globally the troposphere has warmed since the mid 20th Century. However, there is only medium to low confidence in the rate and vertical structure. There is medium confidence in the rate of change and its vertical structure in the NH extra-tropics, while elsewhere confidence is low, particularly in the tropical upper troposphere and over the shorter period since 1979. Through construction of several additional data sets, detailed intercomparisons among data sets, and better understanding of uncertainties in pre-existing data sets, the large uncertainty has become much more apparent since AR4."
Note that there is no reference to the modelled faster rate of heating of the upper troposphere - the topic is airbrushed away by referring to the differences between the data sets. Any reasonable analysis suggests that the differences between the data sets are insignificant compared to the differences between all the data sets and the model predictions of some 0.6 deg C per decade.
-Werner Krauss (via Tom Nelson)
Elegance lost. Just back from a walk by the river, only to notice that the word 'rising' is missing from my step 1. above (12:37 PM). Readers are kindly requested to insert it, as they read, just before 'CO2'. [Done. BH]
Dec 16, 2012 at 3:11 AM | shub
One of the (many) IAC recommendations that appears to have been ignored is the one which derived from their observation that:
As a consequence, the IAC had recommended (inter alia) that:
Has anyone spotted any examples of "traceable accounts" in this Draft?! And I wonder what happened to the evidence which would substantiate former UNFCCC head honcho, Yvo de Boer's Nov. 7, 2012 assertion to the effect that AR5 would "scare the wits out of everyone".
Not to mention the IPCC's deafening silence on the heels of de Boer's prediction! Then, again, perhaps de Boer knows something about the as yet unreleased final version to which the reviewers of the SOD were not privy. Which has led me to ask ...
Where’s the scare in AR5?
OTOH, I suppose it's also possible that the IPCC - in its infinite wisdom - has redefined both "transparency" and "traceable".
Ta BH
Yes.
In a book review, I wrote:
The evidence for CO2 being a threat is non-existent. The only "evidence" is computer models, which have been programmed by people with a strong desire to produce evidence for a strong relation between CO2 and global temperatures. But, as someone said, a computer model is an illustration of a hypothesis, it is not evidence.