The observed trajectory of F for 1990-2004 was compared in Figure 1 of R2007 with the average trajectory of F for each of the six SRES scenario families. The observed growth rate r(F) over the period 2000-2005 exceeds the average growth rate in any scenario family for the decade 2000-2010. We took this approach to test how the alternative hypotheses represented by SRES scenario families compare with reality in the decade since their development, focussing only on emissions. Since no individual scenario is more likely than any other (Nakicenovic et al. 2000), comparison of actual emissions with model ensembles in scenario families is the only viable approach.
Here we provide further detail by examining (1) the variability among individual scenario models within each family, and (2) possible uncertainty in the data.

Perhaps you should ask Dr. Raupach about these probabilities. I’m not really following you.

Pardon me for persisting, but this is important.

The SRES was explicit that “No judgment is offered in this report as to the preference for any of the scenarios and they are not assigned probabilities of occurrence…” (quoted in AR4, Report of WGI, chapter 10, p. 803). There is therefore no warrant for Dr Raupach’s statement that all of the scenarios are equally likely – i.e., have an equal probability of occurrence.

Perhaps he has been misled by the claim in Whetton et al, 2005, ‘Australian climate change projections for impact assessment and policy application: a review’ (CSIRO Marine and Atmospheric Research Paper 001) that the IPCC scenarios had been ‘deliberately constructed to be equally plausible’ (p. 32).

I advised CSIRO’s Kevin Hennessy of this error in March 2006. He replied promptly, with copies to eight of his CSIRO colleagues, agreeing with me and saying that ‘This error will be corrected.’ When the error had remained uncorrected for six months I raised the matter with Dr Roger Jones, then of CSIRO, in a post to economist John Quiggin’s blog (“Drying Out” thread). Jones replied as follows:

‘Your [Castles’] point about the error in our paper made earlier this year was noted. A request was immediately sent by a colleague through to the relevant people to correct and repost the document after you first pointed this out but it was not done (a production task). On the basis of your recent post pointing out that the phrase had still not been corrected (quelle horreur) we have asked that it be followed through. There is no conspiracy – it was a breakdown in process.”

The faulty statement remains uncorrected after three years. Although Roger Jones, Kevin Hennessy and Penny Whetton are all IPCC Lead Authors, and also co-authors of CSIRO Marine and Atmospheric Research Paper 001, they have apparently been unable to persuade CSIRO to correct the error in the Paper. This may be because the Paper has been cited in AR4.

Professor Field’s claim that “fossil emissions have proceeded much more rapidly than anticipated in ANY of the scenarios that were characterised in detail” (EMPHASIS added) cannot be sustained. The A1 preliminary marker scenario was one of the four that were used in 1998 to solicit comments during the IPCC’s “open process” and as input for climate models. The projected level of fossil emissions in 2010 was 9.7 PgC/yr in the preliminary version and also in the final version published in the SRES (see the A1p and A1B columns in Table II.1.1 of the TAR scientific report and the A1B-AIM scenario in the SRES: p. 381).

With world primary energy growth in 2008 falling to its lowest rate of growth since 2001 and the expectation of a decline in global outlook this year, it is now unlikely that fossil CO2 emissions will reach the projected level of the A1B marker.

The SRES states explicitly that the marker scenarios “have received the closest scrutiny of the entire writing team and via the SRES open process compared to other scenario quantifications”, and that “The marker scenarios are also the SRES scenarios that have been most intensively tested in terms of reproducibility” (SRES, Technical Summary, section 7, p. 31).

Thus there is no basis for the assertion that “we are basically entering a domain of climate change that has not been explored by the models” as a consequence of the trajectory of actual emissions. Far from being “much more rapid than was anticipated in any of the scenarios” the actual trajectory of emissions in the first decade of the 21st century will probably fall somewhat short of the projections in the most widely used and tested of the A1B scenarios.

Mike Raupach is Co-Chair of the GCP, but I don’t think he is or has been a ‘Prof’.

OK the GCP authors calculated growth rates compounded continuously, and took the average of the models for each of the six ’scenarios’. But these are in fact six scenario GROUPS, and it doesn’t seem sensible to characterise all of the ‘models’ in the group as relating to the same scenario – e.g., between 1990 and 2010 the projected primary use of oil increases by only 10% for A1B MiniCAM, but more than doubles for A1B ASF (SRES, pps. 406, 386). The caption is misleading in stating that ‘the ENVELOPE of SRES projections are shown for comparison’ when, for example, A1B ASF fossil fuel emissions are projected to exceed 10 GtC in 2010. It is most unlikely that these emissions will reach this level.

In an interview on Radio ABC Australia on 17 February 2009, Chris Field (a member of the GCP team who is now Co-Chair of IPCC Working Group II) said ‘Fossil emissions have proceeded much more rapidly than anticipated in any of the scenarios that were characterised in detail. The consequence of that is that we are basically entering a domain of climate change that has not been explored by the models. We’re on a different trajectory of emissions and therefore an unknown trajectory of warming’ ( http://www.radioaustralia.net......93602.htm). This was highly misleading. The A1B marker scenario HAS been considered in detail – it’s the average of all of the A1B scenarios that hasn’t been.

I’ll get them plotted. I’ll see if I did some weird mis-match on the baselines. (It’s a bit of a pain. The AR4 baseline is well defined. The TAR one… not so much. It’s “relative to 1990″, but doesn’t specify a type of smoothing. So, Rahmstorf picked one. But I need to figure out the relative zeros.)

But the AR4 does have more warming than the TAR both over the first 30 years of this century, and over the recent period. AR4 has a highly non-linear variation during the 90s due to pinatubo.

Agreed that it’s a pain. I’m still trying to figure out why the AR4 projections would be so much higher than TAR’s.

The caption to the Figure on p.11 of the Copenhagen Synthesis Report refers to “Observations of anthropogenic CO2 emissions”, but this is clearly an error. Although the vertical scale is not labelled, the observations align fairly closely with those that have been published by the US EIA and CDIAC in GtC. However, these series relate to fossil fuel and cement emissions only and do NOT include emissions from land use change (estimated by the Global Carbon Project (GCP) Team at 1.5 PgC annually.

According to the latest update posted on the website of the GCP (authored inter alia by Mike Raupach and Pep Canadell of CSIRO, two of the three authors of this section of the Copenhagen report), emissions from fossil fuel and cement “increased from 6.2 PgC per year in 1990 to 8.5 PgC in 2007, a 38% increase from the Kyoto reference year 1990.” If emissions from land use changes were to be added in, the total would be literally “off the chart”.

I can’t get the numbers to add up either, and I don’t see the point in checking whether they may relate to some other measure such as fossil fuel usage or global warming potentials: the caption to the figure refers specifically to emissions.

First of all, the source for the chart appears to be Raupach 2007 in PNAS, and the whole report is available for free.

The ordering of the scenarios is (B2<A1T<B1<A2<A1B<A1F1)
which does not match the carbon emissions.

The emissions scenarios actually start in 1990, but the 20 year rates don’t match up either. They also didn’t use 100 year rates.

I was able to get 2.43 for A1B looking at fossil fuel use, but none of the other scenarios work.

One possibility is that the SRES scenarios have changed. Looking at the breakdowns in TAR, there are changes in subgroups, but the total emissions numbers are the same for 2000.
I don’t think this is likely, as the ordering wouldn’t change.

Maybe global warming potentials are included. This explains the ordering as A1F1 shows an increase in N2O over A1B, however I haven’t gotten the numbers to add up.

It is more than ridiculous that such an important world-wide issue is demonstrated in “crayon”.