Down near the end of a long and ultimately tedious* comment thread over on Prometheus, Gavin Schmidt offered an incredibly useful little gem today.
As I’ve written before, “paleoclimate and future projections on a regional scale are critical.” This is one of the reasons I’ve been so interested in the argument Roger Pielke Sr. makes about the inadequacy of globally averaged surface temperature as a metric for climate change. I wanna know what has happened, and what we can say about what might happen in the future, right here. (And by “right here,” I actually mean in large part the oceans – ENSO and the like – which are the big regulators of what happens here.) Gavin, one of the RealClimate guys, suggests in his Prometheus comment that also is the real important piece of the hockey stick climate reconstruction:
The actual part of the reconstruction that is most interesting are the *spatial* patterns of change. This is the cutting edge for the GCMs – whether modes of variability like ENSO or the NAO have forced (and potentially predictable) components. Our own work has used the MBH reconstruction to look for patterns related to solar and volcanic forcing with some success (Schmidt et al, 2004; Shindell et al; 2001, 2004) as have others (Adams et al, 2004 for instance).
It is one of those odd ironies that the figure that everyone keeps talking about is actually the least important from the point of view of understanding climate responses to forcing.
* Sorry William and Roger, but I gave up trying to parse who’s saying what regarding how the SPM did or did not use the hockey stick to support its attribution claims.
Hmmm. That’s a very unfortunate title, only partially redeemed by the word “sorta”. I won’t speak for Gavin, but I doubt that he wants to “vote against” global-average surface temperature.
There’s no doubt that this metric doesn’t tell the whole story, but it’s quite another thing to suggest that it tells us nothing, or very little. Now I’m not suggesting that Roger (Senior) is suggesting such a thing, but the trouble is that on this issue, as on many others, I find it hard to work out what Roger *is* suggesting.
Now, in the post you link to, Rogger says “the concept of basing climate policy on such an ambiguously measured climate metric as a globally-averaged surface temperature change is inadequate with respect to actual human- and natural-caused climate change. We cannot actually measure such an average directly.” (No, but we can make some pretty damn good estimates of the *temporal changes* in the average.) Then he says “A more appropriate metric for policymakers, for global warming, for instance, would be the global-averaged and regional-averaged patterns of changes in heat content”. First: that’s not one metric, that’s a whole family of them. Second: given the difficulties in measuring (temporal changes in) temperatures above the surface with radiosondes and satellites, we cannot actually measure it directly (to borrow a phrase).
Roger is also in favour of monitoring ocean heat content. So am I. But measuring that accurately is quite a challenge (but becoming more feasible with satellite altimeters and Argo). And it doesn’t *replace* atmospheric temperature measurements, because the ocean has its own thermal dynamics which are quite different to those of the other components in the climate system.
“Insufficiency” might be a better word than “inadequacy” to describe my feelings about the globally averaged temperature number. The issue for me is, “what’s the most important thing for me to spend my time thinking about,” and the global air temperature number isn’t it. For my purposes it’s the spatial stuff that Gavin was talking about, and that Mann and others are working on. As I mentioned over on RPSr’s blog a few weeks back, I think that global number is just fine for a first order explanation for the broad public, but it also can become something of a distraction to serious discussion. I was happy to see Gavin say something quite similar.
Mark – I’d missed the “ambiguously measured climate metric” bit of Rogers argument. That bit really *is* nonsense: whatever the (near) sfc T might be, its the least ambiguous of all the metrics.
John – I’m glad you gave up parsing the tedious arguments. That was pretty well my point: you shouldn’t bother!
Mark – you wouldn’t be at Dunedin Dec 5th, would you?
Pingback: jfleck at inkstain » Blog Archive » More on Globally Averaged Surface Temperature