Potential sources of red snapper (Lutjanus campechanus) recruits estimated with Markov Chain Monte Carlo analysis of otolith chemical signatures

Abstract

Otolith chemical signatures were used to estimate the number of likely nursery sources that contributed recruits to a suite of red snapper (Lutjanus campechanus) year-classes sampled in 2012 in US Atlantic Ocean waters from southern Florida (28°N) to North Carolina (34°N). Otoliths from subadult and adult fish (n≤139; ages 2-5 years) were cored and their chemical constituents analysed for δ13C, δ18O, as well as the elemental ratios of Ba:Ca, Mg:Ca, Mn:Ca and Sr:Ca. Results from multiple linear regression analyses indicated that there was significant latitudinal variation for δ13C, Ba:Ca, Mg:Ca and Mn:Ca. Therefore, these variables were used to parameterise Markov Chain Monte Carlo (MCMC) models computed to estimate the most likely number of nursery sources to each age class. Results from MCMC models indicated that between two and seven nursery sources were equally plausible among the four age classes examined, but the likely number of nursery sources declined for fish aged 4 and 5 years because of apparent mixing between more northern and more southern signatures. Overall, there is evidence to reject the null hypothesis that a single nursery source contributed recruits among the age classes examined, but increased sample size from a broader geographic range may be required to refine estimates of the likely number of nursery sources.