MSc: Probability and Statistics, CIMAT (Mexico) 2014
BSc Mathematical Engineering, ESFM-IPN (Mexico) 2011
School of Geography, Archaeology and Palaeoecology (GAP)
Queen’s University Belfast
Belfast, BT7 1NN
Northern Ireland, UK
+44 (0)28 9097 3828
Bayesian improvements to radiocarbon calibration and palaeoecological compilations
The goal with this research project is to contribute to the statistical techniques used in palaeoecology, radiocarbon-based techniques specifically, and at the same time their possibilities and limitations. The main interest in this research is to explore a Bayesian solution to problems in palaeoecology/archaeology, which have not been solved satisfactorily. For example;
Statistical improvements to radiocarbon calibration will be analysed and their impacts tested. Specifically, my MSc thesis will further develop on which I investigated calibration using F14C instead of the usual 14C scale (which although universally used has asymmetric errors and involves additional steps in calculations). The F14C scale is more directly related to the AMS measurements and to past atmospheric 14C concentrations.
For deposits covering the most recent centuries, both 210Pb dating and 14C dating (be it postbomb or pre-bomb) are very commonly used, often together with additional known-age markers such as Cesium spikes or human-induced pollen markers. However, often the 210Pb dates do not agree with the 14C dates, and no methods yet exist to combine the two strands of chronological information. Statistical techniques will be explored with the aim to combine 210Pb dates with 14C and other dates into integrated age-depth (or age-mass) models. The new methods will be applied to a range of existing datasets.
Recent promising work at CIMAT on combining multiple palaeoecological records through common time markers (Kuschinski 2014) will be further developed and tested on a range of existing key Holocene records. Specifically, enhanced age estimates will be produced for volcanic ash layers (tephra) that are used to align records across wide regions (e.g., Armit et al. 2014, Jensen et al. 2014.
Dr Maarten Blaauw
Prof Paula Reimer