An improved Bayesian model for periodic signal detection has been used to ascertain whether the presence of sinusoids in the IntCal04 calibration data is plausible. Probability distributions for the number of frequencies will be given both for calibration model output and raw tree-ring data. It is shown that while the presence of continuous signals in the model data is in doubt, there is evidence for short high-frequency oscillations in the raw data with periods in the 2 to 20 year range. Some general comments on Bayesian model selection will also be made.
See more of Poster Session I
See more of The 10th International Conference on Accelerator Mass Spectrometry (September 5-10, 2005)