I’m going to do a detailed post on my diagnosis of the Marcott uptick, but before I do so, I want to comment on the reconstructions for NH and SH extratropics, neither of which have attracted sufficient notice though both are very remarkable. In a substantive sense, because orbital changes have different effects on NH and SH, the difference between NHX and SHX proxies is a source of substantive interest and is what ought to have been reported on.
In the running text, Marcott et al had emphasized that their reconstruction was “indistinguishable within uncertainty”. This was illustrated in their Figure 1E, which showed the Mann08 GLB and NH reconstructions (which of several versions is not denoted in the caption), Moberg 2005 (NH) and the Wahl-Ammann version of MBH98 (NH).
Our global temperature reconstruction for the past 1500 years is indistinguishable within uncertainty from the Mann et al. (2) reconstruction… This similarity confirms that published temperature reconstructions of the past two millennia capture long-term variability, despite their short time span (3, 12, 13).
The zonal (NHX, SHX and tropics) reconstructions were illustrated in Marcott Figure 2I,J,K, but Marcott et al conspicuously did not compare their zonal reconstructions with previous NH and SH reconstructions, instead comparing them to proxies often considered to be precipitation proxies. In their text, they stated:
Trends in regional temperature reconstructions show strong similarities with high-resolution precipitation records, consistently associating greater warmth with greater wetness (Fig. 2, H to J). For example, extratropical Northern Hemisphere mid-to- high–latitude temperature correlates well with records of Asian monsoon intensity (14, 15 -Dongge O18) and the position of the Atlantic intertropical convergence zone (16 – Cariaco) (Fig. 2H), tropical temperatures track precipitation proxies from speleothems in Borneo (17) and Indonesia (18) (Fig. 2I), and extratropical Southern Hemisphere temperatures parallel speleothem proxies of precipitation and temperature from South Africa (19) and South America (20) that are independent of our reconstruction
Their decision not to compare their reconstruction to popular hemispherical versions was a curious one and I’ve remedied this apparent oversight below.
First here is a comparison of the Marcott NH Extratropic reconstruction with the Moberg 2005 reconstruction (which Marcott et al showed in connection with their GLB reconstruction). I suspect that some readers will not find these two series to be “remarkably similar” (TM- climate science). (For greater certainty, Marcott et al did not say that these two series were “remarkably similar”, he said that the Moberg NH reconstruction was similar to his GLB reconstruction, but omitted any direct comparison with his NHX reconstructions.
Figure 1. Moberg NH versus Marcott et al NHX. Red dot shows 1990-2010 average HadCRU temperature. Dotted vertical black line at 1890 is period after which reconstruction already admitted to be “not robust”.
Next here is a comparison of the Marcott NHX reconstruction with two Mann reconstructions: the Mann 2008 NH EIV iHAD variation (which in effect splices temperature from 1850 on) and the MBH NHX version (extracted from Briffa et al 2001 SI). I am not persuaded that Marcott’s NHX reconstruction is “indistinguishable within uncertainty” from the corresponding NH reconstructions, making one wonder why the GLB reconstruction is are supposedly “indistinguighable within uncertainty”.
In the figure below, I’ve first compared the Marcott et al SHX reconstruction to the SH reconstruction of Mann and Jones 2003. After readers recover from their awe at the remarkable similarity (TM-climate science) between the two reconstructions, they will note that the Marcott reconstruction began a pronounced and dramatic increase in the 18th century and reached its maximum in AD1900, which, in their Northern Hemisphere reconstruction, was the coldest year since the LGM. Marcott’s estimated AD1900 temperatures in SHX were much higher than even recent temperatures (1990-2010 HadCRU SH average denoted by red dot.)
Finally, here is a comparison of Mann 2008 SH EIV-CRU and CPS-CRU versus Marcott. There are important inconsistencies between Mannian EIV and CPS results, improving the odds of matching at least one of them. (Why anyone would regard Mannian EIV as an “improvement” on anything is a continuing source of puzzlement to me, but that is another day’s story.) Even doubling the odds of a match, I find the correspondence between the Mannian and Marcottian version to be underwhelming, though perhaps not so underwhelming as to preclude them being “remarkably similar TM-climate science”.
I will return to details of the uptick later. For now, I’ll close with this graphic showing the NHX and SHX reconstructions against the GLB reconstruction shown as the Marcott base. While one expects a difference between NHX and SHX in the Holocene, the remarkable difference between NHX and SHX not just in the 20th century, but in the 19th century is a source of considerable interest. According to Marcott, NHX temperatures increased by 1.9 deg C between 1920 and 1940, a surprising result even for the most zealous activists. But for the rest of us, given the apparent resiliency of our species to this fantastic increase over a mere 20 years, it surely must provide a small measure of hope for resiliency in the future.