Junkscience.com response to New York Times on climate modeling

“Computers Add Sophistication, but Don’t Resolve Climate Debate” – “When the Bush administration issued an update last week on federal climate research, it was criticized with equal vigor by environmentalists and by industry-backed groups.

The update featured new computer simulations showing that the sharp rise in global temperatures since 1970 could only be explained by human influences, mainly rising levels of greenhouse gases.” (New York Times)

Oddly, Meehl’s graphic, reproduced here from the NYT, is truncated at 1999, just post-peak of the powerful 1997/98 El Nio-induced temperature spike evident in both MSU and GISS datasets. MSU data indicates a peak in April of 1998 at +0.746C (annual mean +0.472C) and GISTEMP peaked in February of that year at +0.97C (annual mean +0.711C) – by March ’99 both had fallen significantly, to -0.088C (annual mean -0.022C) and +0.3C (annual mean +0.437C) respectively.

We’re sure the resultant impression of runaway warming in Meehl’s graph is purely accidental. Basing his anomalies graphic on the 1890-1919 average is also a rather novel approach, other items here based on the climatological mean (1951-1980 average).

Regardless, Meehl’s graphic sure differs greatly from this one derived from one of the best financed and arguably best maintained near-surface datasets in the world – the continental United States of America. Kind of odd, considering they’re depicting the same period, that one indicates significant and quite rapid warming while the other shows no increase in 7 decades. Even more strangely, the GISSTEMP near-surface global mean temperature anomaly graph below does not appear to support Meehl’s version either.

So, which ‘reality’ is being modeled then?

The thumbnail to the left links to a graphic of lower troposphere temperature anomalies determined from data captured by NOAA satellite-mounted MSUs. July, 2004 global mean -0.213.

The thumbnail on the right is linked to a graphic of temperature anomalies as suggested by the NASA GISS surface temperature analysis (GISTEMP), a near-surface temperature amalgam – July, 2004 global mean +0.3.

Plotted together – the increasing disconnect between these datasets is obvious. The question is: how does the near-surface amalgam produce a resulting anomaly >0.5C warmer than so-called satellite temps? This does not accord with the enhanced greenhouse hypothesis. Under that hypothesis the troposphere should warm and some of that increase should be reflected subsequently in near-surface measures – diametrically opposite to what has supposedly been measured.

This leaves us with several possibilities: the enhanced greenhouse effects works nothing like we suppose; the lower troposphere measures are incorrect; the near-surface amalgam is incorrect or; some combination of the above. Although there are many uncertainties regarding climate we think we have a fair understanding of the greenhouse effect – if not then the entire argument is moot. That leaves the temperature records. Of these, the satellite data has been validated against balloon-sonde measures while the near-surface amalgam is “odd man out.” Satellite data gives near-complete global coverage while near-surface records increasingly reflect temperatures in cities and at airports, an urbanization of the record accelerated by closure of rural recording stations and urban development.

So, what are these computers modeling? Is it enhanced greenhouse effect (EGE) or urban heat island effect (UHIE)?