In any article by Mann and coauthors, it is always prudent to assume that even seemingly innocent choices use up a researcher degree of freedom – to put it nicely. For example, Rahmstorf et al focus on their “AMOC index” in the period ending 1995 and show their AMOC index up to as shown below.
Figure 1. Annotated Rahmstorf et al 2015 Figure 3b. Annotated to show 1995 endpoint. In their running text, Rahmstorf et al say: “The most striking feature of the AMOC index is the extremely low index values reached from 1975 to 1995 … The significance of the 1975-1995 AMOC index reduction was estimated using a Monte Carlo method.”
However, their AMOC reconstruction is defined as the average temperature in 15 gyre gridcells in the Mann et al 2009 reconstruction, which continues to 2006. This raises the obvious questions: why didn’t Rahmstorf show values between 1995 and 2006? Does the withheld portion of the reconstruction between 1995 and 2006 continue to decline?
Once the question is posed, you can guess the answer, which is shown in the figure below, in which the Mann et al 2009 gyre average (blue) is overplotted onto the Rahmstorf et al 2015 figure shown above. Over most of its history, the two series match closely (more or less confirming the provenance). However, instead of the reconstruction, based on Mann et al 2009 gridded data, closing at relatively low 1995 values as shown in the Rahmstorf figure, the reconstruction based on Mann et al 2009 gridded data ends in 2006 at a record high. (By showing this, I absolutely do not imply that this data has any meaning, since I do not agree that a reconstruction of Atlantic ocean currents can be made from contaminated Finnish lake sediments, strip bark bristlecone chronologies etc
Figure 2. Annotation of Rahmstorf et al 2015 Figure 3b, showing Mann et al 2009 gyre reconstruction (blue), highlighting its closing 2006 value.
The truncation of the subpolar gyre series can be seen in more detail in the zoom below The Rahmstorf series (the muddy “orange” series) ends in 1995, while the calculated gyre series from M09 gridded data goes to 2006. The red series is the GISS instrumental average for the gyre gridcells. The Mann et al 2009 RegEM series is spliced instrumental data in the calibration period – the discrepancy to GISS data appears to arise from differences in instrumental data.
Figure 3. Detail of Rahmstorf et al 2015 Figure 3b, with annotation. The “orange” reconstruction ends ~1995 as shown by the red point. The gyre using Mann et al 2009 gridded data is shown in blue, ending in 2006. The “instrumental” NASA GISS version ends in ~2014-2015. Mann’s RegEM methodology splices instrumental data, but it used a different target instrumental data set.
In Rahmstorf’s Figure 5, he plotted the difference between the gyre series and NH: his pseudo-AMOC index, described as follows:
Figure 5 illustrates corroborating evidence in support of a twentieth-century AMOC weakening. The [light=]blue curve depicts the AMOC index from Fig. 3.
Rahmstorf’s Figure 5 also shows a coral dN15,series: this is a very novel proxy, which had been applied by the coauthors in a 2009 article to assess distance of the coral reef from sewage sources (see here) – the evidence of this article, showing, if nothing else, that the article itself is proximate to a sewage source, if not an actual emitter of sewage. The article cited by Rahmstorf et al (see here) reported results from a Nova Scotia location, saying that d15N values are higher in Labrador current waters and lower in Gult Stream waters. Curiously, we recently discussed alkenone proxies in this region – a proxy which, unlike contaminated Tiljander sediments, might actually shed light on the topic. Needless to say, this was ignored by Rahmstorf in favor of contaminated sediments and stripbark bristlecones.
Once again, I’ve overplotted the difference between the gyre and NH series (the pseudo-AMOC index) using Mann et al 2009 gridded data (blue), once again showing its elevated 2006 closing value. As I understand it, the Rahmstorf 2015 “AMOC index” series was calculated using the gyre series from Mann et al 2009 gridded data – so how it extends to 2014 or 2015 is a mystery for which I did not see any explanation in the article.
Figure 4. Rahmstorf 2015 Figure 5, showing gyre using Mann et al 2009 gridded data (blue.) Red circle-dot shows closing 1995 value in Figure 3b. Original caption: The blue curve shows our temperature-based AMOC index also shown in Fig. 3b. The dark red curve shows the same index based on NASA GISS temperature data48 (scale on left). The green curve with uncertainty range shows coral proxy data25 (scale on right). The data are decadally smoothed. Orange dots show the analyses of data from hydrographic sections across the Atlantic at 25 N, where a 1 K change in the AMOC index corresponds to a 2.3 Sv change in AMOC transport, as in Fig. 2 based on the model simulation. Other estimates from oceanographic data similarly suggest relatively strong AMOC in the 1950s and 1960s, weak AMOC in the 1970s and 1980s and stronger again in the 1990s (refs 41,51).
Obviously, the most important point about the Rahmstorf reconstruction is its phrenology – i.e. there is no plausible mechanism by which the difference between two linear combinations of unrelated proxies (contamined sediments, bristlecone chronologies, truncated MXD chronologies and hundreds of nondescript tree ring series) can plausibly yield a reconstruction of Atlantic ocean currents as well as separate reconstructions of NH and SH temperature.
But over and above that, there are technical puzzles over the closing values of this data. The Mann et al 2009 gridded data goes to 2006, but the Rahmstorf gyre reconstruction only goes to 1995 in Figure 3. Figure 5 has the opposite problem: the Rahmstorf version goes to 2014/2015, closing on a downtick, whereas the corresponding series calculated from Mann et al 2009 gridded data only goes to 2006 and ends on a uptick.
In my next post, I’ll examine why the Rahmstorf data begins in AD900, though the EIV reconstructions of Mann et al 2009 being considerably earlier.