Those of you reading this who are APS members should have received an email in the last few days pointing to a new "commentary" that is proposed, by the society's public affairs committee (POPA), to be attached to the 2007 climate change statement. While overall I found the commentary reasonably fair and somewhat useful as a clarifying measure, it had several embarrassing errors that I hope will be corrected before it becomes official. Below I've included my response to this commentary - it will probably be meaningful only to those who have been granted access, sorry about that...
In a statement by a scientific organization the science has to be right. There are several areas in which this proposed commentary is unduly weak, factually wrong, or adds to confusion, with errors both in terminology and in misrepresenting the state of the science. Details below; I strongly recommend study of these major reviews cited:
(4) IPCC AR4 WG2
* the statement on "no credible natural mechanisms" is far too weak. Modeling natural mechanisms over the past century leads to at most 1/5 of the observed warming. In most models it leads to cooling; human forcings account for at least the vast majority of recent warming (see Ref. 3 Fig. 6.14 for example)
* surface warming is far from the only indicator of climate change. GHG's have *decreased* the temperature of the stratosphere and raised tropopause height. Changes in seasonal signals, biological responses, precipitation, wind and glaciers have also been observed, almost all in the direction expected (See Ch. 3 and 4 of Ref. 3, and Ref. 4)
* the word "forcing" implies causation. Water vapor and cloud effects are not causal, they are a consequence of changing climate (water vapor's atmospheric half-life is about 9 days and humans have not directly increased the planet's active water). These are "climate feedbacks", not "forcing functions" (Ref. 1)
* the statement overstates the role of models. Many independent lines of analysis of Earth's climate response to perturbations don't require detailed physical modeling. Ref. 3 covers these; for a more concise overview see Ref. 2. Figure 3 of Ref. 2 needs to be understood in detail on sensitivity - the evidence comes from far more than models.
Overstatement of model roles persists to the definition of CO2-equivalent - in reality this is a matter of radiative forcing and the approximation that climate response depends largely on that one factor: see Ref. 3 Sec. 2.8 (p. 195ff)
* besides the confusion of forcing and feedback, the statement claims "uncertainties" on water vapor. But that uncertainty is minimal. Changing water vapor has both a positive GHG-radiative feedback and a negative lapse rate feedback, but all physical models show that the combined water vapor + lapse rate feedback is strongly positive, on the order of 20 sigmas from zero (Ref. 1). Water vapor plus lapse rate results in a 40-50% increase in the bare response of surface temperature to any radiative forcing.
* The vast majority of models also show a net positive feedback from clouds (Ref. 1), but only about 2 sigmas from zero. There is also observational evidence to support positive cloud feedback - see "Observational and Model Evidence for Positive Low-Level Cloud Feedback," by Clement, Burgman, and Norris, Science v. 325, p. 460 (2009).
* the statement on sensitivity is numerically misleading. Ref. 3 found it highly unlikely that sensitivity would be less than 1.5 C, with a likely range of 2 to 4.5 C. This is strongly supported by Ref. 2. To be as low as 1 C requires additional negative feedbacks to overwhelm the large water + lapse rate number; such claims are not supported by any model or observational evidence at this point.