Niklas Kappelt
Doctoral student
Dating Ice Cores with the 36Cl/10Be Ratio
Author
Summary, in English
Ice cores are unique archives of paleo-climate information and require accurate dating for its interpretation. Continuous chronologies are usually based on time markers, stratigraphic matching, and orbital tuning, but often end several meters above bedrock, because extreme thinning and sometimes disturbances in the stratigraphy complicate the identification of climate signals and their alignment with other records in the deepest ice. Independent age estimates can be obtained with the radioactive decay of 36Cl and 10Be, two radionuclides, which are produced in atmospheric spallation reactions initiated by galactic cosmic rays. Since the flux of these rays is modulated by the magnetic fields of the Sun and the Earth, individual concentrations vary over time, but their ratio is theoretically independent of production variations and decays with a half-life of 384 thousand years.
In this thesis, the application of the 36Cl/10Be ratio as a dating tool was tested and developed with a focus on three key challenges. Due to the different chemical properties of 36Cl and 10Be, they are transported and deposited differently, so their concentrations as well as the 36Cl/10Be ratio exhibit a variability in ice, which determines most of the age estimate uncertainty, as the initial 36Cl/10Be ratio at the time of deposition can only be estimated. A deuterium-based climate correction was applied to radionuclide data from a drill site in coastal West Antarctica and a remote drill site in East Antarctica, reducing uncertainties significantly. A second challenge for the dating method is a loss of volatile H36Cl at low accumulation sites. However, we were able to show that this mostly affects ice from interglacial periods, not from glacial periods, and that the loss is captured by the general trend of higher 36Cl/10Be ratios in colder times, which means the initially present 36Cl can be estimated. A decrease of 10Be concentrations with age faster than possible through radioactive decay alone poses a third challenge. It is likely related to an increasing association of 10Be with dust over time. Testing various variations of the standard sample procedure, we found that passing samples through ion exchange columns resulted in systematically lower 10Be concentrations compared to directly precipitated samples, suggesting they prevented the quantitative detection of 10Be in previous analyses.
The improved dating method was tested on ice from the 800 thousand-year-old EPICA Dome C ice core and was in agreement with the established age scale. It was also used to estimate the age of the deepest part of the Skytrain ice core in West Antarctica and revealed that the ice in this location has been around for at least 500 thousand years, whereas it was previously hypothesised that the West Antarctic Ice Sheet melted in the last interglacial period. As several other bottommost ice core sections have not been dated so far, the method will be able to extend other age scales and help understand the history of the Earth's ice sheets better in the future.
In this thesis, the application of the 36Cl/10Be ratio as a dating tool was tested and developed with a focus on three key challenges. Due to the different chemical properties of 36Cl and 10Be, they are transported and deposited differently, so their concentrations as well as the 36Cl/10Be ratio exhibit a variability in ice, which determines most of the age estimate uncertainty, as the initial 36Cl/10Be ratio at the time of deposition can only be estimated. A deuterium-based climate correction was applied to radionuclide data from a drill site in coastal West Antarctica and a remote drill site in East Antarctica, reducing uncertainties significantly. A second challenge for the dating method is a loss of volatile H36Cl at low accumulation sites. However, we were able to show that this mostly affects ice from interglacial periods, not from glacial periods, and that the loss is captured by the general trend of higher 36Cl/10Be ratios in colder times, which means the initially present 36Cl can be estimated. A decrease of 10Be concentrations with age faster than possible through radioactive decay alone poses a third challenge. It is likely related to an increasing association of 10Be with dust over time. Testing various variations of the standard sample procedure, we found that passing samples through ion exchange columns resulted in systematically lower 10Be concentrations compared to directly precipitated samples, suggesting they prevented the quantitative detection of 10Be in previous analyses.
The improved dating method was tested on ice from the 800 thousand-year-old EPICA Dome C ice core and was in agreement with the established age scale. It was also used to estimate the age of the deepest part of the Skytrain ice core in West Antarctica and revealed that the ice in this location has been around for at least 500 thousand years, whereas it was previously hypothesised that the West Antarctic Ice Sheet melted in the last interglacial period. As several other bottommost ice core sections have not been dated so far, the method will be able to extend other age scales and help understand the history of the Earth's ice sheets better in the future.
Department/s
- MERGE: ModElling the Regional and Global Earth system
- Department of Geology
Publishing year
2025-09-22
Language
English
Publication/Series
LUNDQUA THESIS
Issue
99
Full text
Document type
Dissertation
Publisher
Department of Geology, Lund University
Topic
- Geology
Keywords
- Ice cores
- Dating
- Radionuclides
- Antarctica
Status
Published
Project
- Dating Ice Cores with the 36Cl/10Be Ratio
Supervisor
- Raimund Muscheler
- Eric W. Wolff
ISBN/ISSN/Other
- ISSN: 0281-3033
- ISSN: 0281-3033
- ISBN: 978-91-87847-93-6
- ISBN: 978-91-87847-92-9
Defence date
17 October 2025
Defence time
13:00
Defence place
Pangea lecture hall, Sölvegatan 12, Department of Geology, Lund.
Opponent
- Robert G. Bingham (Professor)