Millennuium Idols Part II
(Go to Part I)
This essay primarily addresses the misconception that prior to the IPCC First Assessment there prevailed in paleoclimatology the picture of a generalized warming that reach an outstanding peak in the high Middle Ages. We have not found this. Indeed, we have not found much interest at all in depicting a generalized temperature trend (whether hemispheric or global) across the last millennium. In 20th century paleoclimatology the interest at this timescale was much more in the patterns of climate variability found to be shifting slowly across vast regions of the globe. Out of this work emerged the various theories of climatic oscillations and of the impact of climate change on human history.
But this is not to say that there was no demand for the depiction of a generalized temperature trend on this time scale. Indeed, during the 1970s and 1980s this demand emerged in the assessments commission by governments in response to the global climate change scares. It was just that, for this demand, the science and the scientists were ill prepared.
Nevertheless, in 1975 one millennium graph was chosen to depict a generalized trend, a version of which is given in fig 5 above. This trend line was repeatedly used though to the 1980s in reports assessing the risk of a human impact on climate. Not all the various assessments during the 1980s depicted a generalized millennium temperature trend (e.g., not this one associated with the Villach ’85 conference (1)). But, where a trend was given, (at least, as we have found so far) only various versions of this graph (fig. 5) were used.
Two other millennium graphs have also been found in use during the 1980s to give the generalized trend, but these were not in assessment reports, nor the primary literature. They are found in textbooks, one by Davis, the other by Tickell. As Steve McIntyre noted in a comment (here) on our Part I, it seems to be from the latter that the IPCC graph is derived. What can we say about the transition to this new and very different looking chart? The IPCC First Assessment authors must surely have known the previous usage in the previous assessments, and so they would have consciously chosen the new graph (whether or not via Tickell) for their new assessment.
Not until the late 1990s did a group of scientists set out to meet the demand of the climate change assessments for a generalized millennium temperature trend. Using mostly the previously much maligned tree ring data, this group produced 3 different trend lines for the northern hemisphere, all published in 1998 (see here). One of these was not yet extended across the millennium until 1999, but, complete with the spliced instrumental data of recent years, it prevailed in the IPCC Third Assessment as the famous Hockey Stick. And so we have the third of our three millennium idols.
Before the Hockey Stick, all the previous generalized millennium trends lines mentioned above (including Davis 1986) were derived from charts by Hubert Lamb. Lamb, who remained skeptical until he died in 1997, never saw the Hockey Stick, but he would undoubtedly have seen his various charts used in various ways to depict the generalized temperature trend. On this misuse of his work, so far no comment from Lamb has been found. But, if this essay in any way brings greater clarity to the true legacy of Hubert Lamb, then it has served some purpose appropriate to this day, the day of Lamb’s centenary.