Bishop Hill Thou shalt extrapolate
Jun 28, 2011 Climate: Surface John Cook, of Skeptical Science fame, has an article in The Age, in which he is very rude about Bob Carter:
A Yiddish proverb states ''a half truth is a whole lie''. By withholding vital information, it's possible to lead you towards the opposite conclusion to the one you would get from considering the full picture. In Bob Carter's opinion piece on this page yesterday, this technique of cherry-picking half-truths is on full display, with frequent examples of statements that distort climate science.
One bit of the article that stuck out at me was this:
[Carter] has long hung his hat on the proposition the climate has been cooling since 1998. But with 2005 and 2010 being the hottest years on record, he resorts to cherry-picking which dataset to use. Rather than use temperature records that cover the entire globe, he opts for datasets that do not include the Arctic region, where warming is the strongest. These temperature records underestimate recent warming and are the darling of those who wish to deny global warming is happening.
Now this is interesting. As readers here know, there are almost no temperature stations in the Arctic and the gaps are therefore infilled by extrapolation.
Now, I don't know about you, but I'm not sure that it is reasonable to get on one's high horse complaining about somebody who prefers to look at, you know, actual data rather than the outpourings of a mathematical model.
Certainly, to accuse them of half truths seems, well, extreme.
Reader Comments (225)
barry
How do we distinguish between the various forcings and their effects (to establish their relative strength)?
For example, wrt Milankovitch, we've got a major ca 100ky cycle that increases summer insolation at 65N sufficient to trigger an interglacial, eg Eemian, Holocene.
When after several ky the orbital eccentricity causes insolation at 65N to reduce sufficiently the next glacial begins.
So clearly with the major land masses where they are, the climate is sensitive enough to slow variations in insolation to respond by entering/leaving glacials.
With ice albedo, ice sheet dynamics (eg Heinrich), THC and related atmospheric teleconnections (and related changes in cloud cover) all in play, as well as CO2, CH4 and water vapour all in play, how do we quantify?
How do we, if you like, reverse engineer paleoclimate to get at a value for climate sensitivity to CO2 forcing?
How do we avoid misattribution?
@barry
Whilst I very much welcome your eagerness to participate in the proceedings here, that’s now the second of my posts that you have misinterpreted. My “So, is there, in your view, any natural variability element to that 0.6C? How much is CO2 driven and how much is down to other anthropogenic causes?” was aimed at dana who claimed that the 0.6C rise since 1960 was “approximately 100%” or “almost entirely” anthropogenic. So I wasn’t talking about the total rise in temperature or the trend of the last century. I was talking specifically about the rise since 1960 that dana claims is “approximately 100%” or “almost entirely” anthropogenic and whether we should take “anthropogenic“ to mean CO2. But thank you for confirming that the earlier 1910-1940 rise was of a similar rate and magnitude.
LC, yes, I misinterpreted. I thought you were suggesting mainstream climate science doesn't put much focus on natural forcing. I'd just read some excerpts along these lines from Carter's book so maybe I carried that with me when I posted. Sorry.
Excellent questions. It's as if you derived them from the classic papers on the topic. If you are familiar with the literature (I've read several score papers, and some on orbital dynamics and insolation), you'll know that the components you listed are assessed and synthesized in studies of climate sensitivity estimated from ice age shifts.
There's a great resource site for paper lists on climate topics, BTW. Here are some pages with relevant material.
http://agwobserver.wordpress.com/2009/11/05/papers-on-climate-sensitivity-estimates/
http://agwobserver.wordpress.com/2010/01/04/papers-on-the-milankovitch-cycles-and-climate/
http://agwobserver.wordpress.com/2010/01/09/papers-on-glacial-terminations/
http://agwobserver.wordpress.com/2009/08/15/papers-on-ghg-role-in-historical-climate-changes/
I am not remotely educated enough to assess the merit of the methodologies or the results. I can only confirm that the factors you listed are core components in the literature, including discussion on how to avoid misattribution. If your questions mean to imply that this area of study is poorly constrained to be very useful, someone smarter than me will have to take that up.
BTW, there were a few calls upthread to cite studies deriving climate sensitivity from 'measurements' (as opposed to pure modeling, I suppose). I cited a couple upthread, but there are more in the links above.
No problem barry. Incidentally, a little while back, BBD and I had a little discussion about scientists we trust and I said then that there were scientists on both sides of the argument whose claims would cause me to spend that little extra time in evaluation. Both Carter and Plimer would definitely be on that list. In my view, in their desire to prove the AGWers wrong, they will sometimes stretch to unwarranted conclusions, thereby making the same mistake they criticise opposing scientists for. Even Lindzen has been guilty of it. Doesn't make them wrong on everything they say though. Just means you have to give what they say a little extra attention ;)
LC -
Over the past 60 years, there may be a small 'natural variability' component to the 0.6°C surface warming.As for other anthropogenic causes, non-CO2 GHGs have caused some warming, and aerosols have caused some cooling. They have roughly canceled eachother out, but due to the aforementioned uncertainty regarding the aerosol forcing, we can't be exactly sure. That's why I say things like "approximately" and "rough calculation". If you go back to the link I provided in which I estimated climate sensitivity based on recent warming, I carried the uncertainties through and ended up with an equilibrium sensitivity of 1.6 to 6.6°C for 2xCO2. A wide range, but it also excludes the very low values proposed by a number of "skeptics" (Lindzen, Spencer, Christy, Easterbrook, Plimer, etc.).
I agree, and frankly I would say Lindzen is frequently guilty of it, though not quite as frequently as Carter and Plimer.barry
That's excellent - more than I could have asked for.
No - I was writing in a rush - sorry. I am simply trying to improve my understanding of derivations of CS from paleoclimate studies.
Have a good weekend, and thanks for an interesting (and lengthy) exchange.
Dominic
Dana
Response to your article on renewables now up at SkS. Sorry for delay but work plus the horrors of formatting etc ...
Thanks BBD. It will take me a while to look through all those links, but I posted a preliminary response. A couple others have responded as well.
Consider the implications of what Mr Dana writes:
Mr Dana is conflating the overall concept of climatic intertia with sensitivity. Nevertheless if there is such a thing as a single 'sensitivity' of climate, his argument goes like this:
"If the hockey stick is broken and the climate changes much all the time, it means that the sensitivity is high. This means that things are bad, because humans are emitting CO2 which will send the climate swinging. Just a little nudge is all it takes."
The original analogous case made by Mann (Bradley and Hughes) and the IPCC was the exact opposite:
"As you can see, the otherwise staid climate has been sent careening, in the past 50 years. A millenium worth of change has taken place all in the past century. Human civilization depends on slow change, so rapid change is bad"
Quoting Michael Mann:
In other words, Mann (and Bradley and Hughes) implied that the climate was fairly insensitive to all factors except CO2 which the humans happened to emit. Whereas Mr Dana (and it is not only he, who has made this case) that the climate is very sensitive to all factors including CO2.
It is clear however, that the root cause for alarm in both cases relates to the purported rapidity or abruptness of climatic change, but Mr Dana undercuts his own case by repudiating the Mann hockeystick - which is the only source of the claim that the rate of change is rapid and has reached levels never seen before.
There is no resolution of such conflicts. Instead, the rest of the discussion on sensitivity is just a rehashing of well-rehearsed positions - it has been seen in countless forums in the past 3-4 years worth of time.
Shub there are a bunch of errors in your last comment.
No I'm not. I'm saying you can estimate climate sensitivity based on recent temperature measurements if you can quantify ocean thermal inertia, among other things. There is, except for the slightly different efficacies of the various radiative forcings. Climate sensitivity is basically sensitivity to a certain energy imbalance, regardless of the cause of that imbalance.You have basically summarized my argument correctly though. Large past 'natural' changes means sensitivity is high, which means large CO2 increases will cause large climate changes.
No, the exact opposite would be to say that climate sensitivity is low, therefore large CO2 increases are of no concern. This is the "skeptic" argument, not the Mann argument.The Mann argument you quote is basically "abrupt climate change is bad", which I certainly agree is true. Based on more modern 'hockey sticks', the change isn't quite as abrupt as compared to Mann et al. 1998 and 1999, but it's still abrupt, and the abruptness of it is increasing.
No, Mann is talking about temperature changes, not sensitivity in that quote. To compare the two, you need to compare radiative forcings, which Mann is not doing. We didn't know the magnitude of the MWP radiative forcing in 1999. Perhaps you should consider that it's continually "rehashed" because it's correct.The phrase 'CO2 sensitivity' pops up in a number of skeptical blogs here and there. There's no such thing. Climate sensitivity is as dana describes it, the response of the biosphere to a change in the radiative balance of the planet, regardless of the cause.
There are slight differences between responses to different radiative forcings (called "efficacies", as I mentioned above), but they're generally not more than 20-30% different from the efficacy of CO2. So it's a fair approximation to talk about a single climate sensitivity.
Our reading for today
Dana
"the magnitude of the MWP radiative forcing in 1999"
I know 1999 was a while back, but I wouldn't describe it as mediaeval.. :-)
What was it and what caused it, then?
dana1981: as the IPCC AR4 WG1 estimates of 'sensitivity' have been a significant discussion point within this thread, I'd be interested in your initial reaction to the following post...
http://judithcurry.com/2011/07/05/the-ipccs-alteration-of-forster-gregorys-model-independent-climate-sensitivity-results/
My initial reaction is that it's a nitpick. Forster and Gregory's results were within the IPCC's stated climate sensitivity range. Apparently the IPCC used a different statistical approach which changed the study's distribution a bit, but it doesn't change the fact that their results were within the IPCC range. I'd be interested to hear Forster and Gregory's take on it - they haven't said anything over the past 4 years, so I assume they don't think it's a big deal.
dana1981: Having had a quick read-through of that post, I'd say it's more than just a nitpick. Fig. 4 seems to show a quite significant shift and, given that this is the only model-independent assessment and that it used real-world data with known error distributions, indicates a range that peaks far lower and is more tightly constrained than the 'consensus' value.
Perhaps discussion of this issue should occur on the website it originated. I've commented there.
http://judithcurry.com/2011/07/05/the-ipccs-alteration-of-forster-gregorys-model-independent-climate-sensitivity-results/#comment-83676
My biggest complaint about GISS's Arctic temperatures is the fact that they delete SST data in order to extend the land surface temperature data out over areas where there's seasonal sea ice. And GISS does that same lame thing over the Southern Ocean. I discussed that in the following post:
http://bobtisdale.wordpress.com/2010/05/31/giss-deletes-arctic-and-southern-ocean-sea-surface-temperature-data/
Regards
Since the SkS forum leak, thought it would be worth adding John Cook's behind the scenes narrative on this Bishop Hill blog post and Carter story
Cook:
"My narrative is Carter is a liar who uses devious techniques to mislead people"
Oh how times have changed, Cook and Dana noticed this Bishop Hill article:
Dana:
The Bishop Hill crowd is interesting. A few reasonably intelligent commenters. Several who attack John, SkS, and myself. Bishop Hill has now twice asked people to stay on topic.
2011-06-29 06:26:42
John Cook
Thanks Dana for stepping in - deflected some of the hate :-)
I don't read the site (apart from posts where he criticises SkS) but my impression is he's a pretty reasonable, civil guy. His critiques of SkS were all civil and some of the criticisms were reasonable. What's his story?
Dana Nuccitelli:
Don't know his story, other than he wrote The Hockey Stick Illusion, which I hear is quite inaccurate. That's all I know about him though. Compared to other 'skeptics' he seems okay.
I didn't so much deflect the hate as absorb it though :-)
----------------------------------
http://www.hi-izuru.org/forum/Carter%20Confusion/2011-06-28-My%20response%20to%20Carter%20published%20in%20The%20Age%20this%20morning.html
Thank you for nice information. Please visit our web:
Wahyu Stiyawan
Wahyu Stiyawan
How does the accuracy of extrapolation compare to other predictive methods in different contexts?