New insights into the global carbon cycle from measurements of CO2 stable isotopes: methodological improvements and interpretation of a new EPICA Dome C ice core d13C record

Elsig, Joachim (2009). New insights into the global carbon cycle from measurements of CO2 stable isotopes: methodological improvements and interpretation of a new EPICA Dome C ice core d13C record (Unpublished). (Dissertation, Universität Bern, Philosophisch–naturwissenschaftliche Fakultät, Physikalisches Institut, Abteilung für Klima– und Umweltphysik)

Text elsig09phd.pdf - Other Restricted to registered users only Available under License BORIS Standard License. Download (6MB)

In this work, the isotopic composition of atmospheric carbon dioxide during the last 22’000 years is analyzed. For that purpose, measurements of δ13 C and δ18 O on CO2 extracted from air bubbles enclosed in Antarctic ice were performed. The ice core was drilled at Concordia Station (Dome C) in the framework of the EPICA project (European Project for Ice Coring in Antarctica). Other ice core analyses revealed past atmospheric CO2 variations from 180 ppmv during glacial to 300 ppmv during interglacial periods respectively (1-3). The mechanisms behind these changes however, are still not fully understood. Since the reservoirs of the carbon cycle (atmosphere, biosphere, ocean and sediments) have different carbon isotope signatures, measurements of δ13 C on atmospheric CO 2 have the potential to assess past variations in the carbon cycle and its portioning to reservoir exchange fluxes. The thesis is divided in five main chapters as follows.
Chapter 1
In Chapter 1 a general introduction on ice cores, isotopes and the global carbon cycle is given. A more detailed approach to the respective topics will be given in the individual chapters.
Chapter 2
This chapter is dedicated to technical and methodological approaches for measuring carbon isotopes with continuous flow gas chromatography-isotope ratio mass spectrometry (GC-IRMS). Various problems had to be overcome and considerable improvements were made to achieve a reproducibility of 0.07‰ for δ13 C measurements on sample sizes of ~ 0.1 μl CO2 STP. Firstly the modified and improved Cracker system is presented, based on the PhD thesis of Marc Eyer (4). The converted Cracker can be flushed entirely with helium and the sample can be transported continuously to the mass spectrometer. Contaminations coming from the Cracker could be segregated from the sample by connecting a gas chromatography column ahead of the mass spectrometer. This has the additional advantage of separating N2 O and other possible contaminants such as drilling fluid from CO2 . The amount of contaminant generated during the cracking process which nonetheless remained was identified as being adsorbed CO 2 of fairly constant δ13 C. Measurements could therefore be corrected accordingly. Additional adsorption of CO2 in the mass spectrometer was reduced by establishing a conditioning background flux of CO gas (5). For carbon isotopes of CO2 we found a fractionation factor of 0.982 ± 0.005 for the adsorption process on metal surfaces )5). Moreover, adsorption processes in the gas chromatography column are most probably causing a small fractionation as well and are changing the impact of adsorption in the mass spectrometer on isotope measurements. Fractionations in the open split were eliminated by adjusting fluxes and positions of the capillaries (5). In order to test the reliability of our system we measured 19 consecutive ice samples at a resolution of about 2.2 cm. While this test revealed a reproducibility of 0.07‰ for δ13 C, δ18 O measurements indicated a periodic variation of about 11 year. Furthermore, a detailed description of the measuring process and the applied corrections are given as a guideline for successful measurements of δ13 C.
Chapter 3
In the third chapter, all δ13 C, δ18 O and CO2 measurements performed during this study are presented. In total, we processed 265 single samples corresponding to 81 different depths and covering the last 22’000 years. After a short introduction about the last glacial termination we present our new δ13 C record, interpret and compare it with existing records. By means of our δ13 C data in comparison with other proxy for the same period we come up with a plausible explanation for the CO2 evolution over the last glacial termination. The cause for the first increase in atmospheric CO2 commencing about 18’000 years before present is most probably the upwelling of an old water mass from the abyssal South Ocean. This implies that the Southern Ocean was affected by reduced deep-water ventilation during the termination of the last glacial maximum. This further serves as an explanation for the reduced atmospheric CO2 concentrations during glacial times: CO2 was stored in the deep, barely ventilated abyssal ocean. The CO2 increase at the beginning of the Bølling-Allerød is most probably attributable to a temperature increase whereas the weak decline in CO2 during the Bølling-Allerød may be the result of a small uptake by the land biosphere. During the Younger Dryas, the increase in CO2 is most probably explicable by combined temperature and ocean circulation changes.
Chapter 4
The evolution of atmospheric CO2 during the Holocene is the subject of chapter 4. By combining CO2 measurements with measurements of δ13 C performed with the sublimation method (6) and the δ13 C data from this study we are able to significantly constrain the dominant pathways of the carbon cycle evolution during the Holocene (7). Based on inverse model calculations of the mass-balance we show that the decrease in atmospheric CO2 of about 5 ppmv during the early Holocene is most probably the result of a combination of land biosphere carbon uptake of about 290 GtC (-15 ppmv) and oceanic carbon release in response to carbonate compensation of the terrestrial uptake during the Termination (+10 ppmv). The 20 ppmv increase in atmospheric CO2 during the later Holocene is mainly attributable to the release of CO2 by the ocean as result of carbonate compensation of the land biosphere uptake during the Transition (+5.3 ppmv) as well as during the early Holocene (+7.7 ppmv). This is completed by a contribution of about 50 GtC (+1.8 ppmv) caused by a small decrease of the land biosphere, and by a contribution from coral reef formation. Natural or anthropogenic changes in land biosphere storage cannot solely explain the CO2 variations in the second half of the Holocene. Furthermore, possible millennial oscillation in the δ13 C record during the Holocene will be discussed briefly.
Chapter 5
The focus of chapter 5 is on the isotopes of oxygen, which has, in contrast to carbon, three stable isotopes. Therefore, effects occurring in a three-isotope system can be explored. We give a short introduction into the 17 O excess (Δ17 O) and the application of the equilibrium method to Δ17 O measurements. The equilibration method is the present-day standard method used for measuring δ18 O in water samples. Since water cannot be directly measured with a mass spectrometer it has to be equilibrated with CO2. The isotopic composition of the water is thereby transferred to CO2, which can then be processed. 17 O is thus masked by the 14 times more abundant 13 C, which is fractionated during the equilibration process depending on water temperature and pH. Notwithstanding we could show, that by applying a chemical buffer the fractionations can be kept constant which leads to a Δ17 O precision of higher than 0.1‰ for the equilibration method (8).

1. Monnin, E. et al. Atmospheric CO 2 concentrations over the last glacial termination. Science 291, 112-114 (2001).
2. Siegenthaler, U. et al. Stable carbon cycle-climate relationship during the late Pleistocene. Science 310, 1313-1317 (2005).
3. Lüthi, D. et al. High-resolution carbon dioxide concentration record 650,000-800,000 years before present. Nature 453, 379-382 (2008).
4. Eyer, M. Highly resolved δ13 C measurements on CO 2 in air from Antarctic ice cores. PhD Thesis Climate and Environmental Physics, Physics Institute, University of Bern, 1-113 (2004).
5. Elsig, J. & Leuenberger, M. Implications of Fractionations in Continuous CO2 Isotope Ratio Analysis by Mass Spectrometry. Submitted to Rapid Communications in Mass Spectrometry (2009).
6. Schmitt, J. A sublimation technique for high-precision δ13 C on CO2 and CO2 mixing ratio from air trapped in deep ice cored. PhD Thesis Fachbereich Geowissenschaften, Universität Bremen (2006).
7. Elsig, J. et al. Global carbon cycle changes during the Holocene based on ice-core stable carbon isotope measurements. Submitted to Nature (2009).
8. Elsig, J. & Leuenberger, M. Measurements of the 17 O excess in water with the equilibration method. Analytical Chemistry 80, 3244-3253 (2008).

Interest & Impact

Downloads

1 since deposited on 07 Mar 2024
1 in the past 12 months

Citations

5 Citations in Web of Science ®
6 Citations in Scopus

Search

in Google Scholar™

Services

Actions (login required)

Edit item

Actions (login required)

Edit item