An organic geochemical approach to problems of glacial-interglacial climatic variability

Abstract

The concentration and carbon isotopic composition (δ13C) of sedimentary organic carbon (Corg), N/C ratios, and terrigenous and marine δ13C-Corg end-members form a basis from which to address problems of Late Quaternary glacial-interglacial climatic variability in a 208.7 m hydraulic piston core (DSDP 619) from the Pigmy Basin in the northern Gulf of Mexico. Paired analyses of δ13C-Corg and N/C are consistent with the hypothesis that the sedimentary organic carbon in the Pigmy Basin is a climatically-determined mixture of C3-photosynthetic terrigenous and marine organic matter, confirming the model of Sackett (1964). A high resolution (~1.4-2.7 ky/sample) δ13C-Corg record shows that sedimentary organic carbon in interglacial oxygen isotope (sub)stages 1 and 5a-b are enriched in 13C (average ± 10 values are -24.2 ± 1.2% 0 and -23.0 ± 0.8% relative to PDB, respectively) while glacial isotope stage values 2 are relatively depleted (-25.6 ± 0.5%). Concentrations of terrigenous and marine sedimentary organic carbon are calculated using δ13C-Corg and Corg measurements, and terrigenous and marine δ13C-Corg end-members. The net accumulation rate of terrigenous organic carbon is 3.7±3.1 times higher in isotope stages 2-4 than in (sub)stages 1 and 5a-b, recording higher erosion rates of terrigenous organic material in glacial periods than interglacial periods. The concentration and net accumulation rates of marine and terrigenous Corg suggest that the nutrient-bearing plume of the Mississippi River may have advanced and retreated across the Pigmy Basin as sea level fell and rose in response to glacial-interglacial sea level change. A study of selected organic biomarker compounds which could serve as tracers of terrigenous and marine sedimentary organic matter sources was performed by comparison with contemporaneous sedimentary organic carbon isotopic composition (δ13C-Corg). Organic carbon-normalized concentrations of total long chain (C37-C39) unsaturated alkenones and individual C27-C29 desmethyl sterols were determined to be useful proportional indicators of preserved marine and terrigenous organic carbon, respectively. The a1kenones, whose source is marine phytoplankton of the class Prymnesiophyceae, generally occurred in higher concentrations in interglacial isotope stages 1 and 5a-b than in the intervening stages, including glacial stages 2 and 4. Sterols (C27-C29) of a dominantly terrigenous origin had lower concentrations during interglacial stages than in glacial stages. The sedimentary records of both terrigenous and marine organic carbon-normalized biomarker compound concentrations appear to be systematically altered by the remineralization of sedimentary organic carbon, as indicated by a simple, first-order organic carbon decay model. The sedimentary deposition of some terrigenous 4-desmethy1stero1s may be affected by differential hydraulic particle sorting as they are transported from river deltas across the continental shelf and slope to the hemipelagic Pigmy Basin. The marine phytoplanktonic alkenones which originate in the surface ocean and sink through the water column would not be subject to comparable particle sorting. The lack of any 4-desmethyl- or 4-α-methy1sterol which was linearly related to the proportion of marine sedimentary organic matter (as scaled by δ13C-Corg) indicated that either (1) sedimentary diagenesis had obscured the biomarker/Corg versus δ13C-Corg record, or (2) the selected compounds were not proportional indicators of preserved marine organic carbon input. The diagenetic alteration of the sedimentary sterol concentration records in which marine sterols were apparently more susceptible to degradation than terrigenous sterols was consistent with present-day sediment trap and recent (10-1-102y) sediment core observations. Preferential preservation of terrigenous sterols may result in a biased sedimentary record of sterol input which could be misinterpreted as indicating solely terrigenous sterol sources. The value and limitations of a simple model which characterizes the effects of sedimentary diagenesis and source input changes on the relationship between organic carbon-normalized biomarker compounds and sedimentary organic matter carbon isotopic composition are discussed. The potential occurrence of sterol double bond hydrogenations (Δ5',Δ22) in three classes of C27-29-4-desmethy1stero1s was evaluated by examining the time series of expected product/precursor relationships with sterol data from the ~2-100kybp DSDP 619 record. Only the Δ5-hydrogenations of the C29 sterols (24-ethy1cholest-5-en- 3β-01, 24-ethylcholesta-5,22-dien-3β-01) showed significant temporally-increasing trends. The 24-ethy1cho1estan-3β-01/24-ethy1cho1est- 5-en-3β-ol (C29Δ°/C29Δ5) ratio also positively correlated with paired sedimentary organic carbon isotopic composition (δ13C-Corg) values. This may be due to increased susceptibility to diagenetic transformation reactions by the organic matter accompanying finer grain-sized terrigenous sediment particles. A long-term source change of 24-ethylcholestan-3β-01 relative to 24-ethylcholest-5-en-3β-01 to explain the correlation with δ13C-Corg seems less likely since both compounds are predominantly of a terrigenous origin in the Pigmy Basin. A comparison of histograms of stanol/stenol (ΔO/Δ5) ratios for the C27-29 -4-desmethylsterols indicates the following sequence in the relative degree of tranformation: C27 > C28 > C29. The C27 - and C28-sterols appear to have attained their respective degrees of transformation before -2kybp, perhaps prior to deposition in the Pigmy Basin. However, differential rates of competing reactions of both the precursor and products may have obscured these simple transformation ratio records. The sedimentary record of a ratio (Uk37) of long chain (C37) unsaturated alkenones is a useful indicator of glacial-interglacial climatic change in the Late Quaternary northern Gulf of Mexico where a planktonic foraminiferal δ18O-CaCO3 record is complicated by meltwater and/or fluvial events (Williams and Kohl, 1986). Using laboratory temperature calibration data of the Uk37 ratio (Prahl and Wakeham, 1987), it is suggested that the minimum glacial surface mixed layer (SML) temperature was 8±1°C colder than the Holocene high SML temperature of 25.6±O.5°C in a Pigmy Basin hydraulic piston core (DSDP 619). However, this glacial-interglacial Uk37-temperature difference was significantly larger than the differences predicted by either the foraminiferal δ18O or foraminiferal assemblage temperature methods (O.8-2.0°C). A possible cause for this large difference is that the Prymnesiophyte assemblages in this area may vary in response to climatically-induced hydrographic changes. Interglacial periods may be dominated by pelagic Prymnesiophyte assemblages, while glacial periods may be dominated by neritic assemblages. Correlation of the Uk37 ratio with the sedimentary organic carbon composition (δ13C-Corg) is consistent with the predominance of preserved input of erosive terrigenous over marine organic carbon during glacial stages in the northern Gulf of Mexico when sea level was as much as 150m lower than in the present interglacial stage. Marine organic carbon burial dominated in warmer interglacial stages 1 and 5a-b.