Fixing the Facts – By Steve McIntyre
Figure 1.4 of the Second Order Draft clearly showed the discrepancy between models and observations, though IPCC’s covering text reported otherwise. I discussed this in a post leading up to the IPCC Report, citing Ross McKitrick’s article in National Post and Reiner Grundmann’s post at Klimazweiberl. Needless to say, this diagram did not survive. Instead, IPCC replaced the damning (but accurate) diagram with a new diagram in which the inconsistency has been disappeared.
Here is Figure 1.4 of the Second Order Draft, showing post-AR4 observations outside the envelope of projections from the earlier IPCC assessment reports (see previous discussion here).
Figure 1. Second Order Draft Figure 1.4. Yellow arrows show digitization of cited Figure 10.26 of AR4.
Now here is the replacement graphic in the Approved Draft: this time, observed values are no longer outside the projection envelopes from the earlier reports. IPCC described it as follows:
Even though the projections from the models were never intended to be predictions over such a short time scale, the observations through 2012 generally fall within the projections made in all past assessments.
Figure 2. Approved Version Figure 1.4
So how’d the observations move from outside the envelope to insider the envelope? It will take a little time to reconstruct the movements of the pea.
Read more: http://climateaudit.org/2013/09/30/ipcc-disappears-the-discrepancy/
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
KR:
Your post at October 1, 2013 at 8:34 am says
The draft Figure 1.4 appears to have been incorrectly baselined – set to the 1990 peak rather than the actual trend at that time.
However, the actual caption for the draft figure (as per the leaked document) states: “The 90% uncertainty estimate due to observational uncertainty and 16 internal variability based on the HadCRUT4 temperature data for 1951-1980 is depicted by the grey shading.” In other words, the grey area shows the range of variability around model projection means, which are shown by the colored bands. Observations fall well within the range of projections – something McIntyre somehow… failed to mention.
The draft Figure 1.4 was CORRECTLY baselined to the empirically observed value at the time of the first projection (i.e. in IPCC AR1 labelled as FAR). The new spaghetti graph is baselined wrongly such that observations fall well within the range of projections – something you somehow… managed to misrepresent.
Richard
Sorry for getting the blockquote formatting wrong in my post to KR. Richard
Mark Hladik says: @ur momisugly October 1, 2013 at 8:48 am
….. The general public is not drinking the Kool-Aid anymore!
>>>>>>>>>>>>>>>>>>
I have noticed that too. I was at a three day festival this weekend and all the people I talked to snickered when I asked them how they liked the “Global Warming’ and then I got an earful.
@richard
“Anomalies can be normalised to any reference value. The correct reference value for a projection is from the empirically observed value at the time of the start of the projection.”
So if they had referenced it to 1992, the observations would have been above the projected range! That doesn’t make any more sense than using 1990 and finding the observations low. The most reasonable reference has to be an average for the period, surely?
It was a simple mistake, subsequently corrected.
John
Rud Istvan
“This will become their Waterloo.”
The serpent has many heads, cut off one and two grow in its place. I am sure many climate “scientists” will be prepared to defend this.
KR — “Observations fall well within the range of projections ”
You’re misunderstanding this. Observational certainty = 3 in, damnit. Obersvational uncertainty = 3 in. give or take a 1/16 of an inch. The grey area represents the assumed error range of the ‘true’ value of the observation, regardless of the measurement received; the little black dots.
So it’s correct to say that the measured value was outside the envelopes. But it is also correct to say that the envelopes are wholly within the estimated ‘true’ value range. The only problem with even paying attention to this is that the measured value lies closer to the cold side of the range. And then trends even colder to the cold side of the error range as it goes on. eg. It’s a confession of a systemic warming bias in the measurement apparatus.
No clue how or if that’s justified, that’s just what the grey and black bits are telling you.
– – – – – – – – –
Good skeptics abound.
Thanks Steve McIntyre, you one of the great skeptics.
In the intellectual discourse on the future of the IPCC based on AR5, a great skeptic needs to be where their pea is going to be by January 2014 when the report is formally final.
I maintain they will use a post-modern redefinition of science where it is: not necessary to have non-contradiction; not important to have precise concepts; necessary to have higher ideals over observed data to inform science.
Academia will flock to the intellectual support of that view as January approaches.
For skeptics to capture their pea and take it away from them, it just takes simple reasoning open to the public. Venues exist to do that.
It is fun.
John
richardscourtney – If you are showing a graph of observations versus trends, you should show those observations against the actual trendline. 1990 was a relatively warm year, perhaps 0.1C over the 1975-1990 or 1975-present trends; the observations should therefore be shifted ~0.1C higher in that figure.
However, that’s a relatively small issue compared to McIntyre ignoring the model variation around the mean projections. That means his claim of observations exceeding model bounds is simply false – observations are well within the 2-sigma variation range, as the original figure captioning clearly states. An issue which you have not addressed.
JohnB:
At October 1, 2013 at 9:21 am you quote my having said
then you say
No, it was not a “mistake” and I can see no reason for anybody to claim it was.
The projections are of the projected time period which starts at 1990 when the AR1 (i.e. FAR) was published. Any other date is a ‘cherry pick’.
And if the projections are for the future then “an average for the period” is not “the most reasonable reference”. That reference would be imaginary if it is the average of what will transpire in future reality. Altenatively, that reference is a function of the projected time period if it is the average of the projected values.
The only “reasonable” reference is the empirical value at the start of the projected time period. With that reference, as time passes any divergence of the projection from reality will then be observed.
The draft Figure used the only “reasonable” reference.
Richard
KR says [emphasis in original]:
October 1, 2013 at 8:34 am
Your nasty snark at Steven just reveals that you haven’t been following the story. Not only that, but you’ve been suckered by the IPCC. As McIntyre stated:
So the grey area was specially put in by the IPCC to sucker the fools who weren’t paying attention. Hey, I wasn’t either, I didn’t notice that bit of sneakiness … but then I didn’t attack McIntyre based on my gullibility and ignorance, either.
That would be you … and no, the observations do not fall well within the range of projections, they fall within the bogus grey area.
w.
Let’s call the current IPCC graphic the Spaghetti Monster chart.
@richard,
Imagine for a second that there had been an AR1.5 in 1992, and they had referenced using the 1992 value. It would show that observations were lower that the range of projections. Are you saying you would NOT be accusing them of cherry-picking?
Come on, 1990 is an outlier in a noisy data set (as is 1992). They corrected the graph to reference against the trend line at the time they created it. Perfectly reasonable!
John
A few commenters here, KR, JohnB, Alistair Ahs, seem to be missing one basic point. Establishing “settled science” requires a string of successful predictions. When the published version of your prediction fails, you can go back and argue that you made a mistake, and your prediction should have been different. If everyone agrees that you meant something different than what you showed, you may get credit for a successful prediction. If you instead argue post hoc that you can display the data in such a way that makes your prediction look right, don’t expect credit for making a correct prediction!
AGW Cotwology
So if I can use an analogy to see if I’m understanding the justification correctly.
I predict/project at 7AM the temperature will rise 10 degrees from 20C to 30C at noon. Then at noon the temperature is 25C so I say my prediction was incorrectly baselined and the correct baseline would be an average of the temperature between midnight and 7AM. Lets say the respective temps were 10C and 20C for an average of 15. I can then claim my prediction was correct because it was for a 10C rise and there was a 10C rise from the average of a certain period. Is this their justification?
KR: Think about this for a minute. The model ranges already include the 95% certainty range. The gray area cannot be the two sigma range for the model output that already has the 2 sigma range. This makes no statistical sense.
The gray area is simply a gray area (pun intended) inserted by the authors to imply that the measured temps are still inside model ranges. The gray area has no statistical meaning, and no provenance.
JohnB:
Your post at October 1, 2013 at 9:56 am begins by saying to me
No, I will not join you in imagining anything. I am discussing reality and not your dreams.
I would not be accusing them of cherry-picking if they were not doing it.
Richard
shenanigans24:
re your post at October 1, 2013 at 10:06 am.
http://wattsupwiththat.com/2013/10/01/mcintyre-on-ipccs-switching-the-pea-under-the-thimble/#comment-1432996
No, they merely adjusted the baseline to give a false impression.
However, your description is the excuse for the adjustment that John B has attempted to use on their behalf.
Richard
An interesting overlay (as an animated .gif) has been assembled here:
http://climateaudit.org/2013/09/30/ipcc-disappears-the-discrepancy/#comment-441622
WHAT LYSENKO SPAWNED.
A clear case of “Lowering the Bar”
The OLD graph indicates the Base predicted Temperature Anomoly for the TAR @ur momisugly + 0.46 by 2015
The NEW graph indicates the Base predicted Temperature Anomoly for the TAR @ur momisugly +0.36 by 2015
They made the uncooperative imperical data fit the modeled projections by hiding the decline through lowering the boundries of the predicted trend lines
richardscourtney says:
October 1, 2013 at 9:40 am
“No, it was not a “mistake” and I can see no reason for anybody to claim it was.
The projections are of the projected time period which starts at 1990 when the AR1 (i.e. FAR) was published. Any other date is a ‘cherry pick’.”
Bazinga! Double Bazinga on “Any other date is a ‘cherry pick’.”
“And if the projections are for the future then “an average for the period” is not “the most reasonable reference”. That reference would be imaginary if it is the average of what will transpire in future reality. Altenatively, that reference is a function of the projected time period if it is the average of the projected values.”.
Bazinga! Double Bazinga on “That reference would be imaginary.”
“The only “reasonable” reference is the empirical value at the start of the projected time period. With that reference, as time passes any divergence of the projection from reality will then be observed.”
Bazinga! The alternative is that, in the case of models, the IPCC gets to just make up stuff.
Wonderful work, Richard.
I continue to find RSC unbecomingly rude to strangers, but at least correct.
The forward-projected error window must be referenced to the point selected at initial publication (and they could have picked any reasonable value they wanted to at that time). Adjusting the old prediction based on new information constitutes a new prediction. They have thrown out their old work because it embarrassed them.
If the counter argument boils down to “academic nitpicking”, then why change the figure at all? Let it stand, then claim vindication later when the truth wins out.
Willis Eschenbach says:
October 1, 2013 at 9:44 am
“Your nasty snark at Steven just reveals that you haven’t been following the story. Not only that, but you’ve been suckered by the IPCC. As McIntyre stated:
‘Observed values have been outside the AR4 envelope for all but one year since publication of AR4. IPCC authors added a grey envelope around the AR4 envelope, presumably to give rhetorical support for their false claim about models and observations; however, this envelope did not occur in AR4 or any peer reviewed literature.’ ”
The IPCC has a set of rules. In the quotation above, Mcintyre points out that the IPCC violated its own rules by publishing a chart that did not occur in AR4 or any peer reviewed literature. I will add that they presented the chart as if it had occurred in AR4. In the world of peer review, you do not get to change items that have been published.
KevinM:
I find your personal remark at October 1, 2013 at 11:06 am extremely rude and not correct in any way. But so what?
Richard