At the SciLogs website, Stefan Rahmstorf of the über-alarmist Potsdam Institute for Climate Impact Research presents Part III of his Mann-hockey-stick-resurrection series.
Here he explains to his German readers how the temperature really developed over the Holocene and why the 20th century is “unique”.
To do this, Rahmstorf uses the findings of the already refuted Shaun Marcott et al paper, reminding us that Marcott et al “tried out 20 different averaging methods and carried out 1000 Monte Carlo-simulations ” in order “to confirm the robustness of their results“. Rahmstorf writes:
To put the main results in a nutshell, it appears as follows:
Fig. 1. Blue curve: global temperature reconstruction from proxy data from Marcott et al., Science 2013. Shown is the RegEM version – noteworthy differences between the variants with different averaging methods appear only at the end, where the number of proxy series decreases, but where the temperature course is well-known from the instrumental measurements. Red curve: global temperature for (HadCRU). Graphic: Klaus Bittermann.”
Next Rahmstorf explains how there had been a long term general cooling over 5000 years, and that that this was interrupted suddenly by a skyrocketing warming over the last 100 years, a warming that completely cancelled all the cooling that had taken 5000 years to accomplish.
Rahmstorf then compares the Marcott reconstruction to the PAGES 2k reconstruction, as shown in his Fig. 3:
Fig. 3. The last two thousand years taken from Fig. 1 compared to the PAGES reconstruction (green), which is explained in Part 1. Graphic: Klaus Bittermann.”
On Fig. 3 Rahmstorf comments:
What we see is that both reconstructions produce consistent results. The course of the last 1000 years is practically identical and is, by the way, one more confirmation of the ‘hockey stick’ from Mann et al. 1999 – see part 2 of this series, The hockey stick debate.”
So now in Rahmstorf’s mind, the hockey stick is back to life (again).
Next Rahmstorf asks if the modern warming is unique, and explains the methodology used for the 20th century temperature reconstruction:
Because of the mentioned limitations of sediment cores, the new temperature reconstruction does not extend to the present, rather only until 1940, and the number of data curves used decreases significantly before that. (Therefore the shown range of uncertainty becomes increasingly wider near the end and reconstructions with various averaging methods diverge here – here we show the RegEM method because it works the best with the decreasing data coverage – see the blog from Grant Foster for more details, who is a professional statistician). The warming of the 20th century is therefore only captured to some extent, which is not a big problem because the warming is optimally documented by weather stations. There’s no doubt about the climate development of the 20th century.”
True there is no doubt about the 20th century. All the doubt surrounds the reconstructions over the course of the Holocene, and there’s very much in dispute. Moreover, splicing datasets coming from totally different measurement methods is dubious in the least, and is thus nothing less than creative curve designing. On the splicing, Rahmstorf comments.
There is a certain amount of room for play on how one splices the proxy data (blue) to the thermometer data (red) – Here I made it so that the mean value of the Marcott curve is the same as the mean value of the PAGES 2k reconstruction for the time period of 1000 to 1940 AD.”
Rahmstorf does admit that there are problems with the Marcott paper, yet feels that in principle it is correct and that the 20th century warming is unique, citing three reasons: 1) other proxies have never shown a warming like that seen in the 20th century during the entire Holocene, 2) statistics expert Grant Foster says so, and 3) such a rash warming must have physical reasons, and thus would have to disappear just as quickly.
Rahmstorf summarizes by saying that although the Marcott curve can be improved on, he believes that the course of the curve is “for the most part correct” and claims that a gradual cooling over the last several thousand years was abruptly interrupted in the 20th century.
To drive the point across to his readers, Rahmstorf then reaches ever deeper into his creative curve-designing toolbox and next splices on a scenario produced by model projections that use the worst-case climate-sensitivities (which have been shown to have been completely exaggerated) to the already dubiously diced and spliced proxy/instrumental reconstruction. By now the curve looks worse than Evel Knievel did after his worst motorcycle crash:
Fig. 4. Global temperature since the last ice age 20,000 years ago and extended until the year 2100 for a intermediate emissions scenario with about 3°C of warming. Graphic: Jos Hagelaars.”
This is not above finger-painting of curves in climate-kindergarten.
Rahmstorf’s Fig. 4 tells us what will happen if we “do not rashly reduce our emissions of greenhouse gases”. He writes that “we are in the process of catapulting ourselves out of the Holocene.” and that if we continue on, “we will know longer recognize our Earth by the end of the century.”