Impacts of Multiple Scales of Variation in Atlantic Menhaden Otolith Chemistry

In widely ranging marine fishes, recruitment often originates from multiple sources with widely varying proportions.  However, estimating the proportion of recruits from these sources has proven difficult.  Historically, the two most common approaches to address this problem were the application of artificial tags and genetic based approaches.  Both of these approaches have limitations such as high tag associated mortality for small individuals or low differentiation among source locations.  However, the use of natural tag methods such as otolith chemistry has proven to be powerful tracers of fish origin in many systems.   

Otolith chemistry has been applied to numerous marine, estuarine, and freshwater species to investigate stock structure and source locations, but classification success has varied widely.  In some cases researchers have reported complete separation among areas, whereas in others the classification success was not different than random chance.  Still, relatively few studies have examined spatial differences in otolith chemistry across a species range.  In these few examples, researchers have often noted misclassifications to nonadjacent locations.

Using juvenile Atlantic menhaden collected from Connecticut to Florida, we will compare several strategies for minimizing classification error associated with multiple scales of variation present through multiple nursery locations.  By directly addressing the causes and implications of misclassification, we offer a more accurate picture of recruitment sources along the entire eastern seaboard.