New Data and Upside-Down Moberg

I’ve been re-examining SH proxies for some time now, both in connection with PAGES2K and out of intrinsic relevance.  In today’s post, I’ll report on a new (relatively) high-resolution series from  the Arabian Sea offshore Pakistan (Boll et al 2014, Late Holocene primary productivity and sea surface temperature variations in the northeastern Arabian Sea: implications for winter monsoon variability, pdf).  The series has considerable ex ante interest on a couple of counts. Alkenones yield temperature proxies that have a couple of important advantages relative to nearly all other temperature “proxies”: they are calibrated in absolute temperature (not by anomalies); and they yield glacial-interglacial patterns that make “sense”. No post hoc screening or trying to figure out which way is up.  In the extratropics, their useful information is limited to summer season, but so are nearly all other proxies. Though more or less ignored in IPCC AR5, the development of alkenone series has arguably been one of the most important paleoclimate developments in the past 10 years and is something that I pay attention to.

But there is a big conundrum in trying to use them for 20th century comparisons: all of the very high resolution alkenone series to date are from upwelling zones and show a precipitous decline (downward HS) in 20th century temperatures. See discussion here.  These precipitous declines have been very closely examined by specialists, who conclude, according to my reading, that this is not a “divergence” breakdown of the proxy-temperature relationship, but rather an actual decrease in local SST in the upwelling zone, attributed (plausibly) to increased upwelling.

Because upwelling zones form only a small fraction of the ocean (though an important fraction due to biological productivity),  it is important to obtain corresponding high-resolution alkenone series from non-upwelling zones. The Boll et al 2014 is the first such example that I’ve seen and, in my opinion, it sheds very interesting new light on the vexed issue of two-millennium temperature.

The Boll Arabian Sea proxy is nearby to the “Arabian Sea” proxy in the Moberg (2005) network, the most HS-shaped proxy in the Moberg network and discussed at CA from time to time.   In my original posts on the Moberg reconstruction, I argued that higher values of the Moberg proxy actually demonstrated colder ocean temperatures.  The Moberg proxy was from an upwelling zone. Comparing high-resolution alkenone data from other upwelling zones confirms beyond any reasonable doubt that the Moberg proxy, as a proxy for local ocean temperature, was used upside down.

Offshore Pakistan

See here for a relatively recent CA discussion of the alkenone SST problem in the 20th century.  The decline seen in upwelling zones does not occur in the Boll et al 2014 alkenone series, which it taken from the “Oxygen-Minimum Zone” offshore Pakistan, rather than an upwelling zone. Boll et al compared their results to the Oppo et al 2009 Pacific Warm Pool SST estimate (from Mg in foraminifera) and the Wanxiang (China) speleothem O18 series interpreted as monsoon strength (see Appendix below), but for my analysis today, I’ve compared their results to another recent alkenone series, McGregor’s series from offshore Morocco – see figure 1 below.

At similar latitudes, the alkenone SST for the Atlantic is considerably colder than for the northern Indian Ocean.  In both locations, alkenone SSTs in the Little Ice Age were lower than the average over the past two millennia. However, the Boll series does not have the precipitous decline in the late 20th century of the offshore Morocco series. (The red series shows a very high resolution box core that measures very up-to-date results.)



Figure 1. Alkenone SSTs. top panel – offshore Pakistan: 39KG-275KL splice of Boll et al 2014; bottom panel – McGregor’s GeoB6008-1 (black) and GeoB6008-2 (boxcore -red).


Comparison to Moberg’s Arabian Sea Proxy

The high-resolution alkenone SST results are dramatically inconsistent with Moberg’s “Arabian Sea” proxy, one of only a couple of “low-frequency” proxies in the Moberg network that have a pronounced HS blade (Agassiz ice melt, another interesting proxy, being another.)



Figure 2. Excerpt from Moberg et al 2005 Supplementary Information, showing Arabian Sea proxy in 11th panel.  

In the next figure, I’ve compared the Moberg Arabian Sea proxy (top panel) to the Boll et al 2014 alkenone SST proxy on a consistent time period. The differences are obviously dramatic. Whereas the Moberg proxy (G Bulloides %) has a pronounced HS with unprecedented 20th century levels, the new high-resolution alkenone SST series shows a long-term decline, reaching a minimum in the late 18th/early 19th century, with modern recovery, but not to earlier levels – a pattern that we’ve noticed in some other NH series. It’s worth noting that the original authors of the RC2730-723A bulloides series did not propose the proxy as a temperature proxy, but as an index of wind speed (and thus monsoon strength). Its adoption by Moberg (and some others) as a temperature proxy seems mostly due to its HS-ness, rather than any use of the proxy by specialists (in Moberg orientation) as an SST proxy.

Figure 3. Top panel – G Bulloides % from Moberg et al 2005; bottom panel- alkenone SST reconstruction from Boll et al 2014.


Inverting Moberg’s Arabian Sea Proxy

Moberg’s Arabian Sea proxy (RC2730/723A splice) is from an upwelling zone offshore Oman.  In previous CA posts, I observed that G Bulloides were a subpolar foraminifera and that measurement of G Bulloides percentage was actually measuring increased prevalence of cold water, rather than warm water.  If the next figure, I’ve inverted the Moberg Arabian Sea proxy for comparison to the McGregor alkenone SST proxy, also in an upwelling zone.  While the Moberg Arabian Sea may well have a relationship to local temperature, the relationship is almost certainly that higher G Bulloides percentage is associated with colder local SSTs, not warmer.



Figure 4. Top panel – Moberg’s Arabian Sea proxy (from upwelling zone offshore Oman) inverted to use by Moberg; bottom panel – McGregor’s alkenone SST series from upwelling zone offshore Morocco.

Boll’s Comparisons

The Boll alkenone series is one of the first very high-resolution modern alkenone series that is not from an upwelling zone and obviously has a different and “more sensible” appearance. The authors compared its appearance to two other prominent recent series, both of which had prominent LIA features, but did not have exceptional 20th centuries: the Oppo 2009 Pacific Warm Pool SST composite from foraminifera Mg and the Wanxiang (China) speleothem d18O series.   I’ve shown their figure below. The data as archived at the Pangaea database (which I’ve used in my above plots) has a different age scale than shown in the print figures. The comparison seems even closer to me using the Pangaea-archive date scale (replotted in bottom panel).  Note right-to-left time scale in the Boll figure below.


boll_2014_comparison_annotatedFigure 5. From Boll et al 2014; Note right-to-left time.  Comparison of Boll alkenone SST (bottom panel) to Wanxiang, China speleothem d18O, interpreted as monsoon strength and Oppo et al 2009 Pacific Warm Pool Mg-Ca composite.


A high-resolution alkenone series from a non-upwelling zone is long overdue (see CA discussion on this point here). It is interesting that the Boll series is so comparable to the Oppo Pacific Warm Pool reconstruction.  In my opinion, the evidence is overwhelming that the Moberg “Arabian Sea” proxy actually measures cold water. In order to claim the G Bulloides series as a NH temperature proxy, Moberg surely ought to have expressly explained why colder SSTs offshore Oman were an appropriate index of Northern Hemisphere warming.

Having said that, the colder SSTs (and increased upwelling) at many upwelling zones in the 20th century seems like an important phenomenon, particularly with the increased interest in ocean heat being transferred to the deep ocean.  I wonder how well these upwelling zones are shown in the GCMs.



  1. Posted Nov 25, 2014 at 1:28 PM | Permalink

    I wonder how well these upwelling zones are shown in the GCMs.

    As well as they model the pause in globally averaged temperature anomaly, surely.

    How long is it going to be for CA to go through all proxies used in PAGES2K? Despite their obvious importance the alkenones have had not a single mention in my personal wiki till now. We learn from every new post and we’re grateful.

  2. Howard
    Posted Nov 25, 2014 at 1:46 PM | Permalink

    I could only find GCM/upwelling papers focused on the Southern Ocean. Some researchers propose that changes in upwelling may be the primary driver of the 100-ppmv glacial-interglacial delta CO2, the increase in sea surface nutrients (productivity) and local lowering of SST. This is, IMO, the most important area of climate research. In this field, there seems to be much less controversy on the older paleoclimate records as compared with the high frequency records that cover the second half of the common era.

    Click to access 2011_Paleo_LCMT.pdf

    Click to access dail_wunsch_jclim_2013.pdf

    Click to access cp-8-545-2012.pdf

  3. Kenneth Fritsch
    Posted Nov 25, 2014 at 2:56 PM | Permalink

    Good stuff SteveM -keep it coming. Also good that you have an open mind to potentially valid temperature proxies.

    I am wondering whether instrumentally derived SST trends can vary this much – or even if we have relatively long SST records at the same or near same locations. I know when we talk about location correlations we usually consider the oceans as being more correlated over time. I also know that high correlations at two locations does not necessarily mean that the trends will be similar. In the bigger picture nearly 70% of the global area is water and most of that is oceans. Most, I think, would judge that we need far fewer spatially separate ocean locations to get a reasonably certain mean temperature measure of that area than we do on land.

  4. TerryS
    Posted Nov 26, 2014 at 4:30 AM | Permalink

    But there is a big conundrum in trying to use them for 20th century comparisons: all of the very high resolution alkenone series to date are from upwelling zones and show a precipitous decline (downward HS) in 20th century temperatures. See discussion here. These precipitous declines have been very closely examined by specialists, who conclude, according to my reading, that this is not a “divergence” breakdown of the proxy-temperature relationship, but rather an actual decrease in local SST in the upwelling zone, attributed (plausibly) to increased upwelling.

    Doesn’t this create a similar problem as the divergence with tree rings?

    Unless they have the timings for when a particular proxy site was an upwelling, downwelling or neutral zone they wont know whether the proxy-temperature is running too cold (upwelling), too hot (downwelling) or just right (neutral).

    Steve: no, it’s very different. tree ring chronologies just give indices of ring width, there is no consistent relationship between ring width and temperature across locations. But there is for alkenones. The same relationship is used everywhere to estimate deg C. Nor is there a divergence between SST and alkenone SST in the 20th century – the divergence is that temperatures in upwelling zones – a small fraction of the ocean – have decreased as more deep water has been upwelled to the surface. A conundrum, but a different conundrum.

  5. mpainter
    Posted Nov 26, 2014 at 2:39 PM | Permalink

    Has the increased upwelling been adequately measured and quantified?
    Because if it has not, it seems speculative to claim such an effect.

  6. Jeff Alberts
    Posted Nov 26, 2014 at 11:20 PM | Permalink

    Who is this G Bulloides person anyway? The nerve!

  7. Posted Nov 27, 2014 at 10:00 AM | Permalink

    Reblogged this on I Didn't Ask To Be a Blog.

%d bloggers like this: