The above passage seems to be combining (a) the authors’ conclusions from all 3 regions included in the 1990 study (eastern China, eastern Australia and European parts of Soviet Union) with (b) results from earlier studies of the contiguous USA and (c) data from other yet to be sampled land areas. The 0.05C seems to refer to this combination of (a), (b) and (c). But maybe I have misunderstood it. Anybody else like to venture an opinion?

Another point which is worth bearing in mind is that although the data on eastern China came from 42 pairs of geographically associated ‘rural’(in fact small town) and ‘urban’ (big city)sites, it seems that Jones et al (1990) did not make comparisons between the two individual members of each pair. Instead “From the 42-station rural network, we formed an average (RCHI, Fig. 2c) for eastern China using the inverse-weighting gridding scheme. The 42 station urban network (UCHI) was averaged in the same way.” (p 171)
(Can anyone tell me what ‘inverse-weighting gridding scheme’ means?)

Later on in the paper (top of p 172) is the following passage: “The warming in UCHI is 0.39C, considerably higher than that in RCHI. For this region, UCHI is the only series for which warming is statistically significant.” It should be noted that the 0.39C warming referred to here was over a period of 3 decades (1954-1983). Warming of 0.23C was found in the corresponding ‘rural’ set over the same 30 year period, but not considered to be statistically significant.

This interesting result is however apparently disregarded in the conclusion (penultimate para of p 172, see above) and in the summary (p 169) which reads “The results show that the urbanization influence in two of the most widely used hemispheric data sets (refs to 3 earlier papers) is, at most, an order of magnitude less than the warming seen on a century timescale”.

The later paper by Wang et al (1990) does however treat the same 42 station pairs as individual pairs and comes up with different conclusions as to the magnitude of the urban effect on warming trends in eastern China. I find it interesting that Wang and Karl, who had both been co-authors of Jones et al (1990), found it necessary to publicly contradict in Wang et al (1990) conclusions which they themselves had signed up to less than 3 months previously.

Chronology of the two papers:
Jones et al: published 13 September 1990, Nature.
Wang et al: received 24 August, revised 12 October, accepted 15 October, published December 1990, GRL.

When you consider everything that is blatantly obvious about UHI over time, including the growth of air conditioning and other energy use, as well as pavement and asphalt (as pointed out above by others), the veracity of Jones 1990 is not really in question. It is very quickly obvious that you are faced with two options, either they are extremely ignorant or they are lying. Given their educations and positions the former is extremely unlikely, and that leaves us with Keenan’s position. And that’s really the best way to put it.

Onward to part 2.

(I presume you and others must have noted these problems sometime in the past, but you haven’t mentioned them yet in this series of posts.)

Thanks for the S.D. observation.