Analyzing Statistical Age Models to Determine the Equivalent Dose and Burial Age Using a Markov Chain Monte Carlo Method

Abstract

Abstract In optically stimulated luminescence (OSL) dating, statistical age models for equivalent dose (De) distributions are routinely estimated using the maximum likelihood estimation (MLE) method. In this study, a Markov chain Monte Carlo (MCMC) method was used to analyze statistical age models, including the central age model (CAM), the minimum age model (MAM), the maximum age model (MXAM), etc. This method was first used to obtain sampling distributions on parameters of interest in an age model using De distributions from individual sedimentary samples and subsequently extended to simultaneously extract age estimates from multiple samples with stratigraphic constraints. The MCMC method allows for the use of Bayesian inference to refine chronological sequences from multiple samples, including both fully and partially bleached OSL dates. This study designed easily implemented open-source numeric programs to perform MCMC sampling. Measured and simulated De distributions are used to validate the reliability of dose (age) estimates obtained by this method. Findings from this study demonstrate that estimates obtained by the MCMC method can be used to informatively compare results obtained by the MLE method. The application of statistical age models to multiple OSL dates with stratigraphic orders using the MCMC method may significantly improve both the precision and accuracy of burial ages.