Evidently, I must be wearing those butterfly wings and using them to propel myself on a flight of fancy inspired by the recent announcement that the official sea surface temperatures. You know, the little snafu over switching from buckets and engine inlets during and after WWII? Visiting I read:
Blogger Steve Bloom said…
Ha, well, RP Jr. and Lucia have taken off on the inevitable flights of hopeful fancy. Also I see that Thompson et al specify (in their abstract, anyway) that the effect is limited to mid-century.
Hhmmm… let’s see… what flights of hopeful fancy did I embark on?
We’re putting in flooring, so I haven’t read the paper in any detail. It will be interesting to see what various people suggest for the new temperatures during the 40-60s era. Right now, it appears the mid-forties data quality has been officially put in question, not just by Climate Audit, but by a paper published in Nature.
This new finding will be spun all kinds of ways. Certainly, losing the empirical observations of low temperatures during the forties will tend to do just what you’ve shown in your figure: Reduce the empirical estimates of warming since the 50s.
It also likely will reduce the apparent amount of “weather noise”, which would make the remaining upward trend stand out with greater certainty. If so, we may conclude that a) warming isn’t quite so rapid overall but b) we have even greater certainty the uptrend is not due to “weather noise”.
Still, we won’t really know which conclusions people will make until we know what data adjustment are supported by the evidence about transitions between buckets/ inlets and bouys!
You can read more of comments at Roger’s blog.
Wow, that sounds almost as fanciful as what James Annan says. You can read his response to Steve here:
30/5/08 3:30 PM
Blogger James Annan said…Well, by “recent” I meant O(50y) or so, I would be surprised if it affected the last 30y much. They may take the opportunity to look at a number of assumptions again. You can bet that the work is well under way with a slew of temperature reconstructions and detection and attribution work, but us plebs will have to wait until the tablets of stone are handed down through the pages of Nature…so Lucia and Roger can enjoy themselves for a bit. I look forward to the howls when people find out that the new analysis matches the models even better than the old one did 🙂
Hhmmm… So how do James’s and my views compare?
- I said we don’t yet know how people will adjust the data. Seems to be what James thinks.
- I said this will affect data in the 40s-60 era. James thinks the last 50 or so. (That would be 2008-50 = … Hhmm… Let me think… does that overlap 40s-60s? )
- If the temperature authorities raise the temperature during the mid-40s or so the absolute estimate for the trend in GMST will decrease as shown in Roger’s graph..
Well, James didn’t comment on Roger’s graph. But should think this “if/then” statement is obviously true. James also does speculate the changes would lower the estimate of climate sensitivity.
- Raising the 50s change would make the century long uptrend stand out better from weather noise and thus make us more certain of overall warming.
James also didn’t comment on this effect. But do you suppose James think removing false “weather noise” will make the trend in the empirical data less certain?
As for the James’s odd intimations of howling: If the models end up looking decent I will not be howling, or gnashing my teeth.
With regard to models, what I think is:
b) Ideally, they should be tested against data not available when runs were performed and
c) When predictions don’t match the existing data, that should be admitted.
It may be news to James Annan or Steve Bloom, but this position doesn’t mean I want climate models to be wrong. I engineering models to the same standards. The reason I hold these standards is, I want them to be right and I want the level of skill to be evident to the public.
What of the current data?
I would like to comment on an data quality issue related to testing the current projections.
The scientists writing the recent Nature paper are also dropping hints that recent SST data may also be flawed. The article included this:
And further corrections may come. For example, the gradual shift since the 1970s from (warm-biased) ship-based measurements to (cold-biased) drifting buoys has probably led to a slight underestimate of SST warming, says Richard Reynolds of NOAA’s National Climatic Data Center in Asheville, North Carolina. “More data problems will undoubtedly come along, if likely of smaller magnitude,” he says.
That’s suitably vague. Are Reynolds suggesting there is a creeping cold bias now? Who can tell?
Let me make a recommendation: If the current flat trends in GMST is even partially due to a creeping bias in measurements, it behooves those creating the official datasets for GMST to investigate and correct the problem now rather than twenty years from now. They should also advise the public which years are in doubt quite soon. That advice should not be tucked away in journal articles, but indicated by expanding uncertainty intervals published with their data sets.
Remember, the time will soon draw near when those working on the 5th assessment report for the IPCC must decide whether they think the AR4 projections matched the subsequent data. If the current mismatch between projections for the AR4 and data turns out to be due to bad data, that’s fine with me. Show the work explaining what was wrong with the data. Explain the correction. Make the process transparent.
Do it soon: preferably long before people are meeting to write the Fifth Assessment Report for the IPCC.
They also need to look at the step change in the HADSST data around the 97/98 El Nino. It appears in a graph of the difference between HADSST and ERSST.v2 data (or ERSST.v3: not shown).
It has an obvious impact on the HADCRUT global temperature. It doesn’t drop back to or near to its pre-97/98 El Nino value as the other temperature indices do.
Lucia,
In my opinion, any evidence that the models could be wrong should be viewed in a positive light because the consequences if they happen to be right will be very bad from a lot of people. The only people that lose if the models are wrong are a few climate scientists that made the mistake of overstating the certainty provided by their models.
I also find James’s comment about how the eventual adjustments will likely improve the fit to then models to be quite telling. He is more or less admitting that the primary object of these exercises is to choose data adjustments that make the models look better. I am surprised that he does not realize that such an outcome would further undermine the credibility of the models amoung people who have not drunk the AGW kool-aid.
Odd. Hadley is showing more negative trend in temperature than NOAA though.
Raven–
To clarify: I don’t mean to suggest I necessarily want the current prediction of models to be true.
What I’d prefer is for models to skillfully predict. I’d also prefer this to be demonstrated clearly, and beyond dispute. If this happened, the public and policy makers could rely on them for planning purposes.
But, like it or not, the current official data does not agree with the current model predictions. The models never agreed in the hindcast. The AR4 forcasts don’t look so hot right now either.
If the apparent lack of skill is due to poor data, fine. Show it. Explain it.
But in the meantime, I think ‘some’ should stop acting as if it was or is irrational to notice the discrepancy between the data and the models.
It used to be pointing out the poor hindcasts was the sign of “denialism”. Interestingly, everyone seems to agree that relative to the “official” data, the hindcasts look like cr*p! Who knows, soon maybe everyone will agree the forecasts don’t agree with the current data either. 🙂
But somehow, the authors of the Nature article want to spin this as a victory for those who insisted the hindcasts looked Grrrr8! 🙂
There’s about a +1 degree C per century rise in the 1800-1900 data also though. So reducing the “weather noise” and seeing the climate trend more clearly still doesn’t indicate AGW. Just GW.
This would seem to force the model’s expected sensitivity to CO2 way, way down.
Annan would seem to be saying “down”, but marginally — or at least, that’s my interpretation of this:
How, he or steve bloom imagine data revisions that support what I’ve repeatedly said I believe would make me howl is odd. After all, I’ve repeatedly said, I think warming is real, caused by people, but that the current evidence indicates the IPCC projections in the AR4 are, likely, too high.
I think its much worse than you expected
http://www.climateaudit.org/?p=3053#comment-255421
and la nina may be coming back?
With around 150 data points, what is the likelihood that every single day this year has been cooler than the same day last year?
Next, what is the meaning of such a fabulously unlikely outcome?
==============================================
Kim–
I noticed that in the graphic Anthony Watts showed. Here are the things I can think of:
1) A big bias was introduced in a measurement device.
2) This year is distinctly cooler than last year
3) A small bias was introduced and this year is still distinctly cooler or
4) GMST is really, really, really correlated over more than 5 months.
Weather is correlated. But still, it seems remarkable that there is not one crossover in that particular series. Someone would have to look at the amount of weather noise and correlation in other years to figure out if this sort of thing (in either direction) has happened before or whether it happens frequently.
Other theories solicited!
Waste no time on Bloom. You are slowly feeling your way through the evidence, one step at a time, to a rational answer. This is what is valuable about this and a couple of other blogs. Bloom will never accept there can be any rational answer which is in any way different from the IPCC. Worse, nor will he ever accept that there can be any legitimate motivation for any enquiry into the evidence or arguments which is open ended in where it is going. You’ll just get into a chorus of flames. Your time is worth more than that and better spent on the evidence. This is your real strength. Just ignore all the personal stuff – of which there will be increasing amounts, the clearer your view of the evidence gets.
What is happening is that every time you get into something in the case in detail, it turns to mush. We saw this happen with the hockey stick, with urban heat effects, now we see it with sea surface temperatures, we are also seeing it with the tropical troposphere trends. You have shown it happening with the IPCC predictions from 2000. There is enough of a pattern that we can expect more.
Of course, it still could be warming, and it still could be all man made. Its just that one by one the previously cited reasons for believing this are being patiently knocked over.
Thanks, lucia, and yup, fred; you point out what is horrifyingly insidious. The blurring of knowledge such that anything can be made of it. This is a real challenge for science, under withering fire, now. And all in the name of money, power, and fame. We should be surprised?
=====================
Bloom is a rabid
attack dogpuppy who doesn’t ever know what he’s talking about-sometimes willfully. Anyone who used to lurk On Roger Pielke Sr’s blog in the comments days knows this. What Icould never understand is why he was so dogmatic. Come to think of it, I’ll never understand why so many people are so dogmatic-but in Bloom’s case, I finally just decided that it was political bias. I am hesitant to pass this judgement on to everyone like him, but it becomes hard not to.