Climate Models: How Much Difference is Too Much Difference?

Does anyboy know the answer?

I re-post here a comment I just left at RealClimate (one never knows what gets published over there, and what doesn’t…)

Re: #101 As a matter of fact if you search on PubMed http://www.ncbi.nlm.nih.gov/pubmed/ there are four articles with my name. In two of them I appear as first author. And no, they are not first-rated earth-shattering Science or Nature articles about climate science.

But as we all agree now, that’s beside the point.

Let’s me start again from a simple question. Hansen et al did compare model results to observations.

“Climate simulations for 1880–2003 with GISS modelE”, Clim Dyn (2007) 29:661–696 – DOI 10.1007/s00382-007-0255-8

For example, consider fig. 9 (the PDF of the article is on the internet, apologies but I do not have time to search for it right now):

“Fig. 9 Global maps of temperature change in observations (top row) and in the model runs of Fig. 8, for 1880–2003 and several subperiods. […]”

Observations there are shown in periods respectively of 124 years (1880-2003), 54 years, 61 years, 40 years and finally 25 years (1979-2003).

Presumably, this provides a first approximation of what time spans are needed to talk about climate (around 25 years). The actual shortest period may be 40 years or longer, as 1979-2003 has been chosen primarily as “the era of extensive satellite observations”. Please correct me if am wrong.

Let’s take now a clear-cut example. The authors write “All forcings together yield a global mean warming ~0.1C less than observed for the full period 1880–2003.”. And that’s a remarkable result.

But…may I ask this rather elementary question: say, if the global mean warming yielded by all forcings together had been much less, or much more than observed, what would have been the (absolute) threshold above which the climate simulations would have been declared a failure?

Or has this question no meaning either? If not, why not?

Once again, I am consciously simplifying things here but this is a blog…more a brainstorming session than a week-long workshop.