Master Thesis Defense by Mattis Ring Westergaard

Title: Extending AstroComb: Determining varying sedimentation rates and Milankovitch periods from geological data. 

Abstract:
Climate proxies from sediment cores needs to be investigated in the context of time, but direct dating is depth specific and instantaneous. Astrochronological models aim to establish a relationship between depth and time by detecting the time-domain Milankovitch frequencies in the depth-domain spectra through a sedimentation rate. The result depend on the model used, the sedimentation rate estimation method and how accurately the Milankovitch periods are known. AstroComb transforms the windowed data into spectra and creates probability distributions across sedimentation rates. By introducing several methods to estimate the sedimentation rate in the presence of multimodal probability distributions, the uncertainties can be more accurately determined. Performing model validation through the use of synthetic datasets, the importance of a reasonable window size is revealed, and the choice should depend on the sampling rate and investigated sedimentation rate ranges to ensure all target Milankovitch cycles can be accurately resolved. A model was also developed to determine the Milankovitch periods present in the data, however the implementation showed unstable results, and highlighted the sedimentation rate degeneracy that appears with resolution and smears high sedimentation rates over larger regions. 

Supervisor: Klaus Mosegaard