This origin of this story does go back a bit further than my origin story in part 1. While it was reading that paper that was the push I needed to start this project, the idea behind it goes back to a conference I attended in graduate school with co-author Jianghui. One poster grabbed both of our attention, it was looking at neodymium isotope shifts in the North Atlantic over glacial-interglacial cycles in fish teeth. What was striking to us however, and quite atypical for these type presentations, the authors also showed photos of the core corresponding to the various sampling intervals. Just like the isotopes varied glacial to interglacial so did the sediment – and very visibly so… what else was changing?
With the support of the Australian New Zealand International Ocean Drilling Program Consortium (ANZIC), we decided to find out! We requested samples from two sites in the North Atlantic with good coverage since the Last Glacial Maximum and at least some published neodymium records. We hired then-undergraduate-students now-co-authors Annabel and Hannah as summer research interns and each took a site to describe the sediments (XRD, SEM-EDS) and measure the neodymium isotope compositions (column chromatography, TIMS). The first thing to know is that while both sites (1308 and 1063) are in the North Atlantic, they’ve experienced the last 25,000 years quite differently. 1308, further north, was strongly impacted by the Heinrich Events (times of massive iceberg discharge during the last glacial period) but doesn’t show much response to the deglaciation. 1063 on the other hand, doesn’t really ‘see’ the Heinrich Events but responds to the deglaciation.
Ok, so what about the records are we looking at to say they respond to or ‘see’ different climate events? Let’s start with the mineralogy and sediment composition. We use two methods to characterize this: X-ray diffraction (XRD) and scanning electron microscopy (SEM). X-ray diffraction is the workhorse of mineralogy, probably the most common approach it relies on the unique crystal structure of different minerals. Different structure = influences how light reflects differently. XRD hits a powdered sample with x-ray beams and measures the angle at which that beam comes back to tell which minerals are present. In terms of the scanning electron microscope (SEM), think a very powerful microscope-you can see very tiny features smaller than the diameter of a human hair. Average human hair is about 100 microns in thickness, a lot of the features we’re looking at are less than 10 microns across. The SEM we used is coupled with energy dispersive spectroscopy (EDS) = very close look at the sediment with elemental data. The chemistry of each spot analysed is then compared to a database of mineral compositions to identify the minerals present.
I briefly discuss using mineral maps from scanning electron microscope EDS analyses in several posts because its an excellent way to better understand the context from which we’re getting geochemical records. For a detailed look at why SEM-EDS is such a cool tool in sediment studies and how it is quantitatively comparable to XRD check out Mehrnoush’s paper (Quantitative petrographic differentiation of detrital vs diagenetic clay minerals in marine sedimentary sequences) and stay tuned for Shujun’s upcoming paper. For the purpose of this discussion, the important part is that SEM-EDS based mineral identification offers three major advantages relative to XRD analysis in that it allows for: 1) identification of trace phases (XRD doesn’t ‘see’ very minor components but these can still be isotopically important) 2) establishment of spatial context/intergranular relationships and 3) unambiguous distinction of sedimentary constituents of authigenic versus detrital origin, none of which are possible using XRD results.
For example, when we look at the sediments from site 1308 (above) we can see changes in sediment composition between normal glacial (cold) times and Heinrich events (still cold, but large release of icebergs into the Atlantic). Notably, there is a large influx of ice rafted debris (larger detrital grains that could have only reached the site if carried by ice) as seen in the lower image above. In contrast, ‘normal’ glacial sedimentation is very fine sediments (scale bar is 100 um) with the largest things present being foraminifera (calcareous microfossils). Corresponding to the Heinrich Events there is also a large negative excursion in the authigenic neodymium isotope signatures and a muted, but still present, negative excursion in the detrital neodymium isotope signatures.
At site 1063, we don’t see any noticable changes in mineralogy or isotopes associated with the iceberg discharge events. Instead, we see a change from carbonate lean muds during the last glacial to calcareous mud after the deglacial. This compositional shift corresponds to the beginning of the isotope shift at the Bølling–Allerød around 14 kyr (BA in image below, an abrupt but short warm period towards the end of the last glacial period).
Combined, these records support the idea that the detrital sediments are at least to a degree determining the authigenic signature as both sites demonstrate coupled shifts in authigenic and detrital neodymium isotope records (fit within the range set by the global compilation) and those shifts align with changes in sediment composition either through scenario 2 (neodymium released from the sediments alters the bottom water via a ‘benthic flux’ or scenario 3 (neodymium of the authigenic phases is controlled by the detrital sediments) – for the full description of each scenario see part 2. Any given location is likely a combination of both scenario 2 and 3 because the reactive sediment component driving the benthic flux in 2 could also alter the authigenic phases in the sediment.
Well, that’s all folks! At least for here— plenty more in the paper itself if you want the more technical run down. Read it here or let me know if you don’t have access and I can send you a pdf 🙂
aprilabbott 
2 thoughts on “Circulation or Sedimentation? P3”