2005-11-14

Testing the Fidelity of Methods Used in Proxy-Based Reconstructions of Past Climate

There's an intersting new paper just out in J Climate, Testing the Fidelity of Methods Used in Proxy-Based Reconstructions of Past Climate by Michael E. Mann, Scott Rutherford, Eugene Wahl & Caspar Ammann (hat tip to John Fleck).

[Update: the actual article is now available: thanks John & Mike]

Two widely used statistical approaches to reconstructing past climate histories from climate 'proxy' data such as tree-rings, corals, and ice cores, are investigated using synthetic 'pseudoproxy' data derived from a simulation of forced climate changes over the past 1200 years. Our experiments suggest that both statistical approaches should yield reliable reconstructions of the true climate history within estimated uncertainties, given estimates of the signal and noise attributes of actual proxy data networks.


This is similar to (but I think there is more than... I really should finish reading it before I post...) von S's Science thing of last year, of which it sayeth:

One study by Von Storch et al. (2004--henceforth 'VS04'), however, concludes that a substantial bias may arise in proxy-based estimates of long-term temperature changes using CFR methods. VS04 based this conclusion on experiments using a simulation of the GKSS coupled model (similar experiments described by VS04 using an alternative simulation of the HadCM3 coupled model showed little such bias). The GKSS simulation was forced with unusually large changes in natural radiative forcing in past centuries [the peak-to-peak solar forcing changes on centennial timescales (~1 W/m2) were about twice that used in other studies (e.g. Crowley, 2000) and much larger than the most recent estimates (~0.15 W/m2--see Lean et al., 2002; Foukal et al., 2004)]. A substantial component of the low-frequency variability in the GKSS simulation, furthermore, appears to have been a 'spin-up' artifact: the simulation was initialized from a very warm 20th century state at AD 1000, prior to the application of preanthropogenic radiative forcing, leading to a long-term drift in mean temperature (Goosse et al., 2005).... These arguably unrealistic features in the GKSS simulation make the simulation potentially inappropriate for use in testing climate reconstruction methods.


We shall see.

3 comments:

  1. On JF's blog Eli commented:

    'I thought the most important point in this new paper was

    '“CFR methods are known to perform poorly in capturing patterns of variability that are entirely or largely missing during the calibration period”

    'In other words if variability differs during the calibration period and the period of the proxy reconstructions the method fails or at best does not work well. This is a hard hurdle for proxy reconstructions to meet over long periods of time if variability is not constant over the entire period and is quite close to what McKitrick and McIntyre have been saying about anomolous growth in the bristlecome pine series. I don’t think Mann et al, have really fully explored the implications of this for their work.'

    I have assiduously avoided making the considerable (for me) effort needed to understand this issue to the point of being able to form an independent opinion, but of course I'm very interested in the outcome of the debate. Somehow I suspect that Mike doesn't think this creates the kind of problem Eli postulates, but then again science moves on and maybe he does.

    Content aside, as an outsider I find it fascinating to watch all of this apparent maneuvering to capture the high ground in the AR4. Can we expect more netween now and the end of the year?

    Speaking of which, I have joined the AGU (associate member) and paid the outrageous single day attendance fee so I can show up at the paleo session at which "Dr." McIntyre is slated to attack the Hockey Team. May I expect any fireworks?

    ReplyDelete
  2. All the methods need some way to tag the proxy record onto the instrumental record (its most obvious in the Moberg case: see my and RC's post on this).

    ReplyDelete
  3. That much I do understand (it's the statistics I'm weak on). Hopefully Eli can explain what he meant in a bit more detail.

    ReplyDelete