Title | Modeling environmental factors affecting assimilation of bomb-produced Δ14C in the North Pacific Ocean: Implications for age validation studies |
Publication Type | Journal Article |
Year of Publication | 2014 |
Authors | Helser, TE, Kastelle, CR, Lai, H-lin |
Journal | Ecological Modelling |
Volume | 277 |
Pagination | 108-118 |
ISSN | 0304-3800 |
Abstract | The bomb-produced radiocarbon (14C) chronometer has become the gold standard for assessing the accuracy of otolith growth ring based fish age estimates. In the northeast Pacific Ocean, nearly a dozen age validation studies have been conducted, ranging from California to Alaska, most of which have relied on a single reference chronology from the Gulf of Alaska. We developed a Bayesian hierarchical model using data sets of bomb-produced radiocarbon in the northeast Pacific Ocean and investigated whether latitude and upwelling exerts an influence on the parameters that describe the rapid Δ14C increase in marine calcium carbonates. Models incorporating both latitude and upwelling as linear covariates of a 4-parameter logistic model were favored based on ΔDIC statistics. There was substantial evidence to support that the timing of the Δ14C pulse was advanced and that total Δ14C uptake increased with increasing latitude. In contrast, increased oceanographic upwelling resulted in lower total radiocarbon input as well as a delay in the timing of the pulse curve, as was demonstrated in the upwelling dominated California Current System. Within the observed latitudinal and upwelling range of the data sets examined in this study the predicted timing of the bomb pulse curve varied by as much as 3 years, which could be misinterpreted as aging error. Our results suggest that new reference chronologies may be needed for regions of the North Pacific Ocean differing in latitude, seasonal upwelling strength and other mixing factors that can potentially change the functional form of the Δ14C curve. |
URL | http://www.sciencedirect.com/science/article/pii/S0304380014000453 |
DOI | 10.1016/j.ecolmodel.2014.01.011 |