THE BAYESIAN APPROACH TO RADIOCARBON CALIBRATION CURVE ESTIMATION: THE INTCAL13, MARINE13, AND SHCAL13 METHODOLOGIES
THE BAYESIAN APPROACH TO RADIOCARBON CALIBRATION CURVE ESTIMATION: THE INTCAL13, MARINE13, AND SHCAL13 METHODOLOGIES
复制标题
DOI：
10.2458/azu_js_rc.55.17222
复制
发表时间：
2013-01-01
期刊：
影响因子：
8.3
通讯作者：
Buck, C. E.
中科院分区：
文献类型：
--
作者：
Niu, M.;Heaton, T. J.;Buck, C. E.
本文概述了贝叶斯模型和方法，用于促进2013年国际商定的放射性碳定标曲线的构建，这些曲线称为IntCal13、Marine13和SHCal13。这些模型建立在2004年和2009年曲线估计所使用的模型基础上，并与2009年一样，使用马尔科夫链蒙特卡罗抽样，特别是吉布斯大都会采样器。除了2004年和2009年模型中考虑到的数据结构外，这里概述的方法还允许：存在数据提供者无法量化的额外不确定性；通过摆动匹配得出日历年龄的树轮数据(除了年轮计数外)；显示两个或更多连续层之间日历年龄误差零增长的可变计数数据；以及我们具有依赖日历年龄不确定性的任何数据源。
This article outlines the Bayesian models and methods used to facilitate construction of the 2013 internationally agreed radiocarbon calibration curves known as IntCal13, Marine13, and SHCal13. The models build on those used for the 2004 and 2009 estimates of the curves and, as in 2009, are implemented using Markov chain Monte Carlo sampling, specifically a Metropolis-within-Gibbs sampler. In addition to the data structures accounted for within the 2004 and 2009 models, the approach outlined here also allows for: the presence of additional uncertainty that the data providers have been unable to quantify; tree-ring data that derive their calendar age from wiggle-matching (in addition to ring counting); varve-counted data that exhibit zero increase in calendar age error between 2 or more consecutive layers; and any data source for which we have dependent calendar age uncertainties.