Science – Feb. 11, 2005

There’s a combined discusison of MM05 (GRL) and MàƒÆ’à‚ⵢerg et al [2005] in tomorrow’s Science, together with a new spaghetti diagram that differs from the MàƒÆ’à‚ⵢerg rendering of proxy data a few posts ago.
Kerr in Science
Figure 1. Spaghetti Diagram, Kerr, Science, Feb. 11, 2005

Kerr says:

Even as greenhouse skeptics revel in what they presume is the downfall of one of global warming’s most prominent supports, paleoclimatologists have come up with yet another analysis. In a paper published this week in Nature, Swedish and Russian researchers present their first entry in the millennial climate sweepstakes.

It’s interesting to see the spin: I thought that MàƒÆ’à‚ⵢerg was a kick in the groin for the Hockey Team, but Phil Jones is still gamely arguing:

"The past couple of decades are still the warmest of the past 1000 years.”

I’ve been looking through MàƒÆ’à‚ⵢerg and don’t exactly see how one can draw this conclusion from MàƒÆ’à‚ⵢerg other than gross splicing. The $64 question for the splices is surely this: the proxies should be going off the charts in the 1990s. I’ve not tried to look through recent proxy collections systematically and it’s a big job, but I don’t get the sense that they are going off the charts or the Hockey Team would have used them. If the proxies don’t pick up the loud 1990s, how do we know that they didn’t pick up some other loud period in the MWP. I’m probably going to spend some time on this topic.


10 Comments

  1. Hans Erren
    Posted Feb 10, 2005 at 6:21 PM | Permalink | Reply

    ALL proxy reconstructions unanimously agree with each other and disagree with the instrumental record between 1940 an 2000. Makes you think about the instrumental record….

  2. Tom Rees
    Posted Feb 11, 2005 at 11:24 AM | Permalink | Reply

    Your readers may be interested to know that not only Jones but also Moberg et al come to this conclusion. From Moberg et al 2005:

    “We find no evidence for any earlier periods in the last two millennia with warmer conditions than the post-1990 period”¢’‚¬?in agreement with previous similar studies”

    It’s interesting to see the spin, as you say. But whose spin is more interesting?

  3. Steve McIntyre
    Posted Feb 11, 2005 at 12:01 PM | Permalink | Reply

    Tom, you’ve made a very fair point about what Moberg says.

    Equally, I think that it is self-evident that the unvarnished Moberg graphic would not have been an effective sales tool for IPCC. (“unvarnished” = without the instrumental extrapolation). The unvarnished proxy graphic shows 20th century warming returning to MWP levels – it would have begged too many questions for the public as to why MWP levels were so high, what caused them to be so high, how we could be so sure that the same thing wasn’t happening now and what was so bad about the MWP anyway. one of the Where’s Waldo links has a picture of John Houghton in front of the hockey stick diagram; now picture him in front of unvarnished Moberg – there wouldn’t have been a sale.

    While climate science, having made the sale, now wants to move one, I’m not sure that some members of the public won’t feel a little tricked if IPCC tries to disown the Mann hockeystick. In business, if you put something in an offering to the public without adequate due diligence, you can’t just “move on” because you want to. Steve

  4. Louis Hissink
    Posted Feb 12, 2005 at 1:29 AM | Permalink | Reply

    A range from -.8 to +0.2 Kelvin as the difference between “hot” and “cold” is not a large variation – one could make that much error from reading a thermometer!

  5. Ferdinand Engelbeen
    Posted Feb 12, 2005 at 7:54 AM | Permalink | Reply

    From Briffa e.a. http://www.nature.com/cgi-taf/DynaPage.taf?file=/nature/journal/v391/n6668/abs/391678a0_fs.html

    Reduced sensitivity of recent tree-growth to temperature at high northern latitudes

    “During the second half of the twentieth century, the decadal-scale trends in wood density and summer temperatures have increasingly diverged as wood density has progressively fallen. The cause of this increasing insensitivity of wood density to temperature changes is not known, but if it is not taken into account in dendroclimatic reconstructions, past temperatures could be overestimated.”

    I did send the opposite view to RealClimate that extra tree ring growth by CO2/nitrogen/… fertilizing might have been the cause in the past and that this is halted now, thus that surface temperatures might be in error and/or that tree rings are underestimating the real temperature changes. That was not published on RealClimate…

  6. Steve McIntyre
    Posted Feb 12, 2005 at 8:20 AM | Permalink | Reply

    Ferdinand, thanks for stopping in. This comment of Briffa’s is one that has stick in my craw a long time. But there’s another possibility that I think the more likely: that the proxies are no good. In our E&E article, we refer to a recent article by D’Arrigo et al about declining ring widths at TTHH despite rising temperatures. They posit the possibility of a quadratic (upside down U) response to temperature – which is completely consistent (actually more consistent) with botanical theory than a linear relationship. I don’t think that people have fully comprehended how subversive this is to efforts to reconstruct past temperatures. How can you tell on present methods whether a narrow ring in the MWP is from hot-dry weather or cold weather? The narrower the ring, the bigger the uncertainty. Steve

  7. Jeff Norman
    Posted Feb 12, 2005 at 12:04 PM | Permalink | Reply

    I have not seen the recent Màƒⵢerg et al paper. Does it present error margins? I recall other proxy reconstructions including error margins.

    The graph you show above appears to be smoothed. Do you know the period used for averaging?

  8. Douglas Hoyt
    Posted Feb 12, 2005 at 3:22 PM | Permalink | Reply

    Of all the reconstructions shown, it seems like MBH98 is the only one that has lower temperatures around 1900 compared to 1600. That would seem to make MBH98 an outlier among the reconstructions. Are there any single proxies that actually show a cooling trend from 1600 to 1900? I can’t recall any.

  9. Steve McIntyre
    Posted Feb 12, 2005 at 5:20 PM | Permalink | Reply

    Hi, Doug, Flattered that you’ve dropped in. For other readers, Doug Hoyt is a very eminent author on the Sun. I’ve posted up a list of the AD1600 period proxies together with the difference between the 1636-1645 mean and the 1896-1905 mean (after standardizing to 1902-1980 mean and sd per MBH). This is at http://www.climate2003.com/blog/hoyt.txt. I’ve also posted up graphs of the 58 proxies used in this period (the PC series are from different rosters than in the AD1400 network that I’ve discussed.) These are at http://www.climate2003.com/blog/hoyt.htm. I did these quickly so they are not very pretty. Steve

  10. Tom Rees
    Posted Feb 14, 2005 at 7:19 AM | Permalink | Reply

    Steve: “While climate science, having made the sale, now wants to move one, I’m not sure that some members of the public won’t feel a little tricked if IPCC tries to disown the Mann hockeystick. In business, if you put something in an offering to the public without adequate due diligence, you can’t just “move on” because you want to.”

    Steve, I’m sure you’re right from a political perspective, but not from a scientific one. When the facts change, as they often do in science, so should opinions. As Keynes is alleged to have said: “When the facts change, I change my mind. What do you do, sir?”. I think it’s great that the paleo community is adopting new reconstructions, no matter that they don’t fall within the previous paradigm. It shows that the data are the key, after all.

    Of course, the new results do fit within another paradigm – climate models of paleoclimate work better with the new data than with the older.

    BTW, you know that Moberg has published extensively with other paleoclimatologists whom you label the ‘Hockey Team’? Jones, Briffa, Osborne.

Post a Comment

Required fields are marked *

*
*

Follow

Get every new post delivered to your Inbox.

Join 2,882 other followers

%d bloggers like this: