Incorporation of bomb-produced 14C into fish otoliths: A novel approach for evaluating age validation and bias with an application to yellowfin sole and northern rockfish

TitleIncorporation of bomb-produced 14C into fish otoliths: A novel approach for evaluating age validation and bias with an application to yellowfin sole and northern rockfish
Publication TypeJournal Article
Year of Publication2016
AuthorsKastelle, CR, Helser, TE, Wischniowski, SG, Loher, T, Goetz, BJ, Kautzi, LA
JournalEcological Modelling
Volume320
Pagination79-91
ISSN0304-3800
Abstract

Fish age validation with bomb-produced radiocarbon (14C) requires a known-age Δ14C reference chronology spanning the era of a marine increase in bomb-produced 14C (1950s to 1960s). Concordance between otolith Δ14C in a validation sample and the reference chronology indicates accurate test ages. Here, we present an information-theoretic approach to hypothesis testing and use Bayesian data analysis with Markov chain Monte Carlo simulation as a probabilistic framework to quantitatively estimate age determination bias and its uncertainty. Using this approach, we compare a new Δ14C reference chronology from the eastern Bering Sea and a previously established reference from the Gulf of Alaska with otolith Δ14C in two validation species, eastern Bering Sea yellowfin sole (Limanda aspera) and Gulf of Alaska northern rockfish (Sebastes polyspinis). Our goals were twofold: (i) to validate the age determination methods for northern rockfish and yellowfin sole using comparisons within oceanic basins, and (ii) to explore the outcome of making naïve comparisons of these validation data sets to reference chronologies across oceanic basins. Based on within-basin comparisons we concluded that estimated ages for eastern Bering Sea yellowfin sole and Gulf of Alaska northern rockfish were accurate. We further concluded that there were important differences in otolith 14C uptake between fish from the two ocean basins.

URLhttp://www.sciencedirect.com/science/article/pii/S0304380015004354
DOI10.1016/j.ecolmodel.2015.09.013