Palynology and carbon

Palynologists have been using δ13C of organic matter and palynomorphs (δ13Corg) - as opposed to δ13C from carbonates (δ13Ccarb) - for several decades. I’ve used it quite a lot in academic and commercial contexts for paleoenvironmental reconstruction and correlation. Here’s my non-exhaustive summary of the uses and pitfalls of δ13Corg in palynology.

From a palynologist’s point of view, the main point is that carbon has two main isotope components, 12C and 13C, and their ratio is expressed as δ13C. Put crudely, the two isotopes are processed differently by plant photosynthesis when CO2 is taken in. Plants ‘prefer’ 12C for their tissues but take in both isotopes, ‘freezing’ the ratio in their tissues. The δ13C is therefore a record of the atmosphere at the time that the plants were growing. But also when plant growth is associated with large amounts of carbon burial—as in the Carboniferous coal swamps—the burial alters the balance of δ13C in the atmosphere of the time. Lots of burial of carbon causes atmospheric δ13C to increase and these increases are passed on to new plants.

The changes in δ13Corg through the stratigraphy can be interpreted as local or more secular variations in the environment and palaeoclimate. Where a change in δ13Corg is seen as worldwide, theoretically it can be used to correlate successions.

The way that δ13Corg is used and interpreted depends on the organic material that’s being analysed. δ13Corg can mean the δ13C signature of the complete complement of organic matter in the sedimentary rock (δ13Cbulk or δ13CTOC), or some subset of the organic matter, for example only spores and pollen(e.g. δ13Cpollen), or only palynological residue material with size >100 µm (e.g. δ13Cres>100µ).

My first use of δ13Corg was to analyse a known post-glacial succession in the Early Permian of Oman (Stephenson et al. 2005). The carbon isotope composition of organic material and the palynology of split samples, were studied to pinpoint any changes in palaeoatmospheric CO2 during a post-glacial period, and to allow links to be made between floral and palaeoatmospheric changes. Split samples provided the ability to track any δ13Corg change precisely against palynological change.

Well sections from the Thuleilat Field, southern Oman, were selected because closely sampled core material was available. The study did reveal some systematic change through the sections which had an interesting link to the development of palynological assemblages and probably to climate change, but the most important finding for me was that residual hydrocarbons in the rock – that were not obviously present - had a powerful influence on δ13Cbulk.

This is shown below (Fig. 1). After careful study, it was found that residual hydrocarbons (‘bitumen’) was present in both the arenaceous and argillaceous parts of the section. All samples were treated to remove and collect any hydrocarbons and δ13C analysis was done again. Where a lot of hydrocarbon had been present, δ13C values were found to have markedly deflected the curve to the left exaggerating the extent of the ‘excursion’. Analysis of the separated hydrocarbons gave very negative δ13C values. These values were similar to those of ‘Huqf oil’ (δ13C ~ 36‰) that is known to charge southern Oman reservoirs. The negative deflection of δ13C was found to be roughly proportional to the amount of extracted hydrocarbon.

Fig. 1. Gross quantitative palynological character, ranges of selected palynomorphs, palynofacies, and δ13Cbulk values for the Thuleilat-16 well section. Note the deflection to the left caused by migrated hydrocarbons (probably ‘Huqf oil’; δ13C = 36‰) that are known to charge southern Oman reservoirs. From Stephenson et al. (2005).

So for me, this is the first rule concerning the use of δ13Cbulk or δ13CTOC – make sure that your samples are free of allochthonous hydrocarbon material, otherwise many of your peaks and troughs may simply relate to the δ13C of a deeper hydrocarbons source rather than a true palaeoenvironmental signal.

A later study of the Palaeocene-Eocene Thermal Maximum in 2012 revealed just how much δ13Cbulk is influenced by its constituent, more autochthonous material (Kender et al. 2012). Like the earlier Permian study, this PETM study was aimed at aligning palynology with a highly expanded δ13Cbulk curve from a well in the North Sea, 22/10a-4. Again split samples allowed pollen and spore assemblages to be forensically analysed alongside δ13C.

After lots of fiddly separation of organic matter types, several fractions of the AOM were studied including δ13CTOC (all the organic matter in the sample), the δ13C of the amorphous organic matter only (δ13CAOM), δ13C of material with <30% wood or plant tissue, and δ13C of material with >30% wood or plant tissue (Fig. 2).

Fig. 2. 22/10a-4 well with δ13CTOC (all the organic matter in the sample), the δ13C of the amorphous organic matter only (δ13CAOM), δ13C of material with <30% wood or plant tissue, and δ13C of material with >30% wood or plant tissue. From Kender et al. (2012).

What these different traces show is the disaggregated ‘spectrum’ within the δ13CTOC curve. The trace of δ13C of material with >30% wood or plant tissue is over to the right because terrestrial organic matter (at least at this particular time in Earth history) has a higher δ13C than marine organic matter (the amorphous organic matter in the samples). These latter samples (δ13CAOM ) predictably record the lowest δ13C values irrespective of the undeniable changes that were going on at the time of the PETM.

So the second rule for me is to remember that δ13Corg is complex and it can change simply because the spectrum of organic matter in your sample is changing. It doesn’t have to be caused by a big palaeoclimatic change!

I’m far from an expert in δ13C. Its value is undeniable when used alongside palynology, because pollen and spores and palynofacies elements are so closely linked to changes in the character of sedimentary organic matter, but also because pollen and spores are such good indicators of paleoenvironmental and paleoclimate change. But changes in δ13C may be more subtle than first meet the eye, and in my experience, at least, not always linked to palaeoclimatic change.

References

Kender, S, Stephenson, M H, Riding J B, Leng, M J, Knox, R W O’B., Vane, C H, Peck, V L, Kendrick, C P, Ellis, M A, and Jamieson, R. 2012. Marine and terrestrial environmental changes in NW Europe preceding carbon release at the Paleocene–Eocene transition. Earth and Planetary Science Letters, 353-354, 108-120.

Stephenson, M H, Leng, M J, Vane, C H, Osterloff, P. L & Arrowsmith C. 2005. Investigating the record of Permian climate change from argillaceous sediments, Oman. Journal of the Geological Society, London, 162, 1-11.

Prof Mike Stephenson is available for consulting.

Email: mikepalyno@me.com