Organic Geochemistry

Strengthening of North American dust sources during the late Pliocene (2.7 Ma)

by David Naafs

Published in Earth and Planetary Science Letters

Here we present orbitally-resolved records of terrestrial higher plant leaf wax input to the North Atlantic over the... more Here we present orbitally-resolved records of terrestrial higher plant leaf wax input to the North Atlantic over the last 3.5 Ma, based on the accumulation of long-chain n-alkanes and n-alkanl-1-ols at IODP Site U1313. These lipids are a major component of dust, even in remote ocean areas, and have a predominantly aeolian origin in distal marine sediments. Our results demonstrate that around 2.7 million years ago (Ma), coinciding with the intensification of the Northern Hemisphere glaciation (NHG), the aeolian input of terrestrial material to the North Atlantic increased drastically. Since then, during every glacial the aeolian input of higher plant material was up to 30 times higher than during interglacials. The close correspondence between aeolian input to the North Atlantic and other dust records indicates a globally uniform response of dust
sources to Quaternary climate variability, although the amplitude of variation differs among areas. We argue that the increased aeolian input at Site U1313 during glacials is predominantly related to the episodic appearance of continental ice sheets in North America and the associated strengthening of glaciogenic dust sources. Evolutional spectral analyses of the n-alkane records were therefore used to determine the dominant
astronomical forcing in North American ice sheet advances. These results demonstrate that during the early Pleistocene North American ice sheet dynamics responded predominantly to variations in obliquity (41 ka), which argues against previous suggestions of precession-related variations in Northern Hemisphere ice sheets during the early Pleistocene.

Sea surface temperatures did not control the first occurrence of Hudson Strait Heinrich Events during MIS 16

by David Naafs

Hudson Strait (HS) Heinrich Events, ice‐rafting events in the North Atlantic originating from the Laurentide ice sheet... more Hudson Strait (HS) Heinrich Events, ice‐rafting events in the North Atlantic originating from the Laurentide ice sheet (LIS), are among the most dramatic examples of millennial‐scale climate variability and have a large influence on global climate. However, it is debated as to whether the occurrence of HS Heinrich Events in the (eastern) North Atlantic in the geological record depends on greater ice discharge, or simply from the longer survival of icebergs in cold waters. Using sediments from Integrated Ocean Drilling Program (IODP) Site U1313 in the North Atlantic spanning the period between 960 and 320 ka, we show that sea surface temperatures (SSTs) did not control the first occurrence of HS Heinrich(‐like) Events in the sedimentary record. Using mineralogy and organic geochemistry to determine the characteristics of ice‐rafting debris (IRD), we detect the first HS Heinrich(‐like) Event in our record around 643 ka (Marine Isotope Stage (MIS) 16), which is similar as previously reported for Site U1308. However, the accompanying high‐resolution alkenone‐based SST record demonstrates that the first HS Heinrich(‐like) Event did not coincide with low SSTs. Thus, the HS Heinrich(‐like) Events do indicate enhanced ice discharge from the LIS at the end of the Mid‐Pleistocene Transition, not simply the survivability of icebergs due to cold
conditions in the North Atlantic.

Late Pliocene changes in the North Atlantic Current

by David Naafs

During the late Pliocene global climate changed drastically as the Northern Hemisphere glaciation (NHG intensified. It... more During the late Pliocene global climate changed drastically as the Northern Hemisphere glaciation (NHG intensified. It remains poorly understood how the North Atlantic Current (NAC) changed in strength and position during this time interval. Such changes may alter the amount of northward heat transport and therefore have a large impact on climate in the circum-North Atlantic region and the growth of Northern Hemisphere ice sheets. Using the alkenone biomarker we reconstructed orbitally resolved sea surface temperature (SST) and productivity records at Integrated Ocean Drilling Project (IODP) Expedition 306 Site U1313 during the late Pliocene and early Pleistocene, 3.68–2.45 million years ago (Ma). Before 3.1 Ma, SSTs in the mid-latitude North Atlantic were up to 6 °C higher than the present and surface water productivity was low, indicating that an intense NAC transported warm, nutrient-poor surface waters northwards. Starting at 3.1 Ma, surface water characteristics changed drastically as the NHG intensified. During glacial periods at the end of the late Pliocene and beginning of the Pleistocene, SSTs decreased and surface water productivity in the mid-latitude North Atlantic increased, reflecting a weakened influence of the NAC at our site. At the same time the increase in surface productivity suggests that the Arctic Front (AF) reached down into the mid-latitudes. We propose that during the intensification of the NHG the NAC had an almost pure west to east flow direction in glacials and did not penetrate into the higher latitudes. The diminished northward heat transport would have led to a cooling of the higher latitudes, which may have encouraged the growth of large continental ice sheets in the Northern Hemisphere.

Evidence for bias in C/N, δ13C and δ15N values of bulk organic matter, and on environmental interpretation, from a lake sedimentary sequence by pre-analysis acid treatment methods.

by Chris Brodie

In Press - Quaternary Science Reviews

Lead Author: Dr. Chris Brodie (Hong Kong University, Hong Kong)

Co-Authors: Dr. James Casford (Durham University); Dr. Jeremy Lloyd (Durham University, UK); Prof. Melanie Leng (NIGL, UK); Dr. Timothy Heaton (NIGL, UK); Christopher Kendrick (NIGL, UK); Dr. Zong Yongqiang (Hong Kong University, Hong Kong).

ABSTRACT:
There is a known bias in C/N, d13C and d15N values of organic matter (OM) due to pre-analysis acid... more ABSTRACT:
There is a known bias in C/N, d13C and d15N values of organic matter (OM) due to pre-analysis acid treatment methods. We report here, for the first time, the results of a pre-analysis acid treatment method comparison of measured C/N, d13C and d15N values in bulk OM from a sedimentary sequence of samples to illustrate this bias. Here we show that acid treatment significantly reduces the accuracy (between method biases) and precision (within method bias) of C/N, d13C and d15N values of OM, suggesting a differential response of sample OM between methods and sample horizons, and in some cases inefficient removal of inorganic carbon. We show that different methods can significantly influence environmental interpretation in some of our sample horizons (i.e. interpretation of aquatic vs. terrestrial OM source; C3 vs. C4 vegetation). Specifically, there are unpredictable and non-linear differences between methods for C/N values in the range of ~ 1 – 100; d13C values in the range of 0.2 – 6.8 ‰ and; d15Nvalues in the range of 0.3 – 0.7 ‰. Importantly, these ranges are mostly much greater than the instrument precision (defined as the standard deviation of replicate analysis of standard reference materials; for this study, ± 0.5 for C/N values, ± 0.1 ‰ for d13C values and; ± 0.1 ‰ for d15N). The accuracy and precision of measured C/N, d13C and d15N values of bulk OM is not just dependent upon environmental variability, but on acid pre-treatment, residual inorganic carbon and organic matter state and composition. Collectively, this makes the correlation between samples prepared in different ways, including those from down core reconstructions, highly questionable.

Keywords: C/N ratios, δ13C, δ15N, organic matter; pre-analysis acid treatment methods, environmental interpretation, palaeoclimate.

Evidence for bias in measured δ15N values of terrestrial and aquatic organic materials due to pre-analysis acid treatment methods.

by Chris Brodie

Rapid Communications in Mass Spectrometry. 25, 1089 - 1099.

Lead Author: Dr. Chris Brodie (Durham University, UK; Hong Kong University, Hong Kong)

Co-Authored with Dr. Tim Heaton (NIGL, UK); Prof. Melanie Leng (NIGL, UK) Christopher Kendrick (NIGL, UK); Dr. James Casford (Durham University, UK) and; Dr. Jeremy Lloyd (Durham University, UK)

ABSTRACT:
We investigate the effect of acid treatment methods on δ15N from a range of environmental organic... more ABSTRACT:
We investigate the effect of acid treatment methods on δ15N from a range of environmental organic materials in the context of the increased application of “dual-mode” isotope analysis (the simultaneous measurement of δ13C and δ15N from the same acid treated sample). Three common methods are compared; (i) untreated samples; (ii) acidification followed by sequential water rinse (rinse method); (iii) acidification in silver capsules (capsule method). The influence of capsule type (silver and tin) on δ15N is also independently assessed (as the capsule and rinse methods combust samples in different capsule; silver and tin respectively). We find significant differences in δ15N values between methods and the precision of any one method varies significantly between sample materials and above instrument precision (> 0.3 ‰). δ15N of untreated samples did not produce the most consistent data on all sample materials. In addition, the capsule type appears to influence the measured δ15N value of some materials, particularly those combusted only in silver capsules. We also compare the new δ15N data with previously published δ13C on the same materials. The response of δ13C and δ15N within and between methods and sample materials to acidification appears to be relatively disproportionate, which can influence environmental interpretation of the measured data. In addition, statistical methods used to estimate inorganic nitrogen are shown to be seriously flawed.

Keywords: acid treatment, capsule method, rinse method, acid method comparison, δ15N, inorganic nitrogen.

Carr, A.S., Boom, A., Dunajko, A., Bateman, M.D., Holmes, P.J., Berrio, J.C., (2010) New Evidence of the Age and Palaeoecology of the Knysna Formation, South Africa. South African Journal of Geology, Volume 113 (3) p241-256

by Andy Carr

The existence of lignitic deposits (Knysna Formation) on the South African south coast, near to the town of Knysna has... more The existence of lignitic deposits (Knysna Formation) on the South African south coast, near to the town of Knysna has been recognised for more than a century. However, a combination of limited study and few natural exposures has meant that the age and stratigraphic position of the Knysna Formation are unclear, despite its potential as a palaeoenvironmental archive. We present a new suite of chronological, geochemical and palynological data obtained from a recently identified lignite exposure in this area. The lignite pollen assemblage is dominated by palms (Arecaceae), which are now locally extinct, and contains additional palynomorphs of tropical affinity, along with (moist-temperate) Podocarpus-type pollen, grasses, and herbaceous pollen types (e.g. Cliffortia-type, Asteraceae). Overall, the assemblage shows some commonalities with the Miocene Elandsfontein Formation in the Western Cape. The lignites are dominated by a diverse range of higher plant biomarkers, including abundant leaf wax lipids, as well as lignin monomers and leaf cuticle-derived macromolecular organic matter. All strongly indicate a terrestrial depositional setting, perhaps akin to contemporary palm swamps. A number of sesquiterpenoids imply the presence of gymnosperms, supporting observations from the pollen data and previously reported macro-fossil finds.

The application of isothermal thermoluminescence techniques to coversands overlying the lignite produced a minimum age of ~1.7 Ma. Additional clues as to the likely age of the lignite are provided by compound-specific stable carbon isotope analyses of the leaf wax lipids. These are approximately 8 enriched relative to typical C3 vegetation leaf waxes, and imply a potential contribution from C4 vegetation. From this, an age post-dating the Oligocene may be inferred, and in conjunction with the site’s geomorphic setting, an age post dating the middle Miocene is considered plausible. This is markedly younger than previous (Eocene) age estimates for the Knynsa Formation.

Carr, A.S. Boom, A., Chase B.M. (2010) The potential of plant biomarker evidence derived from rock hyrax middens as an indicator of palaeoenvironmental change. Palaeogeography Palaeoclimatology Palaeoecology 285, 321–330

by Andy Carr

Hyrax middens are unique environmental archives with the potential to provide unprecedented high-resolution... more Hyrax middens are unique environmental archives with the potential to provide unprecedented high-resolution palaeoenvironmental records, particularly in the arid regions of southern Africa. This study provides the first detailed characterisation of the organic matter composition of hyraceum and aims to identify biomarker evidence capable of providing new or supplementary palaeoenvironmental data from these novel archives. Pyrolysis gas chromatography mass spectrometry reveals hyraceum to be dominated by nitrogen-containing aromatic compounds, notably benzamide. This is almost certainly derived directly from the hyrax urine and is probably the main source of nitrogen as measured in bulk δ15N measurements. Solvent-extractable lipids comprise homologous suites of long-chain n-alkanes (C24–C34) and n-alkanols (C16–C26), characteristic of higher plant leaf waxes, along with an abundance of animal-derived sterols, higher plant sterols and terpenoids; as well as the ubiquitous benzamide.

n-alkane distributions and compound specific δ13C clearly differentiate samples from the C3 vegetation dominated Cape Floristic Region, and the more arid, C4 grass-rich savannas of central Namibia (Klein Spitzkoppe). Distinct changes in n-alkane distribution and δ13C are observed within the Spitzkoppe midden; most notably the mid to late Holocene period (c. 6000–2000 cal yr BP), which records a progressive reduction in grass-derived organic matter inputs. Based on the animal's feeding behaviour and contemporary site ecology, a phase of increasing aridity is inferred, which is consistent with other proxy data from this site (e.g. bulk δ15N). The excellent preservation of plant and animal biomarkers suggests that there is significant potential for midden-derived biomarkers to provide long-term palaeoenvironmental proxies.

Carr, A.S., Boom, A., Chase, B.M., Roberts, D.L., Roberts, Z.E. (2010) Molecular fingerprinting of wetland organic matter using pyrolysis-GC/MS: an example from the southern Cape coastline of South Africa. Journal of Paleolimnology 44 (4), 947-961

by Andy Carr

Pyrolysis–gas chromatography mass spectrometry (py-GC/MS) allows the characterisation of complex macromolecular... more Pyrolysis–gas chromatography mass spectrometry (py-GC/MS) allows the characterisation of complex macromolecular organic matter. In lakes and wetlands this can potentially be used to assess the preservation/diagenesis and provenance of sediment organic matter. It can complement palaeoenvironmental investigations utilising ‘bulk’ sediment variables such as total organic carbon (TOC) and TOC/total nitrogen ratios. We applied py-GC/MS analyses to a ~32,000-year sediment record from the southern Cape coastline of South Africa. We used the results to evaluate the sources and extent of degradation of organic matter in this semi-arid environment. Marked down-core changes in the relative abundance of multiple pyrolysis products were observed. Correspondence analysis revealed that the major driver of this down-core variability in OM composition was selective preservation/degradation. Samples comprising highly degraded OM are primarily confined to the lower half of the core, older than ~12,000 years, and are characterised by suites of low-molecular-weight aromatic pyrolysis products. Samples rich in organic matter, e.g. surface sediments, are characterised by products derived from fresh emergent or terrestrial vegetation, which include lignin monomers, plant-derived fatty acids and long-chain n-alkanes. Pyrolysates from the late glacial-early Holocene period, approximately mid-way down the core are characterised by distinct suites of long-chain n-alkene/n-alkane doublets, which may reflect the selective preservation of recalcitrant aliphatic macromolecules and/or enhanced inputs of the algal macromolecule algaenan/polymerised algal lipids. Increased TOC, lower δ13C and increased abundance of more labile lignin and fatty acid products at the same depths suggest this period was associated with increased lake primary productivity and enhanced inputs of terrestrial OM. TOC is the only ‘bulk’ parameter correlated with the correspondence analysis axes extracted from the py-GC/MS data. Distinct fluctuations in TOC/total nitrogen ratio are not explained by variation in organo-nitrogen pyrolysis products. Notwithstanding, the study suggests that py-GC/MS has potential to complement palaeolimnological investigations, particularly in regions such as southern Africa, where other paleoenvironmental proxy variables in sediments may be lacking or equivocal.

Evidence for bias in C and N concentrations and δ13C composition of terrestrial and aquatic organic materials due to pre-analysis acid preparation techniques

by Chris Brodie

Chemical Geology. 282, 67 - 83.

Lead Author: Dr. Chris Brodie (Durham University, UK; Hong Kong University, Hong Kong)

Co-Authors: Prof. Melanie Leng (NIGL, UK), Dr. James Casford (Durham University, UK), Christopher Kendrick (NIGL, UK), Dr. Jeremy Lloyd (Durham University, UK), Dr. Zong Yongqiang (Hong Kong University, Hong Kong), Prof. Michael Bird (James Cook University, Australia).

ABSTRACT:
This study is the first systematic comparison of the effect of acid treatment methods on the... more ABSTRACT:
This study is the first systematic comparison of the effect of acid treatment methods on the reliability of organic carbon [C] and nitrogen [N], and carbon isotope (δ13C) values on a range of terrestrial and aquatic, modern and geological environmental materials. We investigated the 3 most common methods; (i) acidification followed by sequential deionised water rinses (“rinse method”); (ii) acidification in silver capsules (“capsule method”); and (iii) acidification by exposure to an acid vapour (“fumigation method”). We also investigated the effect of sample size and capsule type (silver and tin) on C/N ratio and δ13C values. We find (i) %C, %N, C/N and δ13C showed significant within and between method variability; (ii) disproportionate and non-linear offsets of %C, %N and C/N values after acidification within and between methods and within and between sample materials; (iii) that alterations in %C did not necessarily manifest themselves in shifts in δ13C, and vice-versa; (iv) small (~ 90 μg C) sample sizes showed consistent overestimations and inaccuracies after acidification; (v) The effect of capsule type was not significant on most samples, but did show a notable effect on our aquatic materials, generally increasing %C and %N, and producing depleted δ13C values. These findings raise cause for concern on the interpretative nature of C/N ratios and their support for carbon isotope values. The comparability between laboratories (different preparation methods) and environmental settings (amount, type and nature of OM) are also likely to be problematic. We conclude that the response of C and N concentrations in organic matter to acid treatment in environmental materials is neither negligible nor systematic.

Keywords: δ13C; C/N ratio; Method comparison; Rinse method; Capsule method; Fumigation method; environmental reconstruction