The importance of location and operational fishing factors in estimating and reducing loggerhead turtle (Caretta caretta) interactions in U.S. bottom trawl gear

Abstract

Mortalities due to interactions between loggerhead (Caretta caretta) turtles and commercial bottom trawl fisheries impede the recovery of loggerhead populations worldwide. In the U.S. Mid-Atlantic, several hundred loggerheads interact with commercial bottom trawl gear each year despite the implementation of temporal and spatial conservation measures. In this analysis a Generalized Additive Model (GAM) is developed using fisheries observer data to estimate the magnitude of loggerhead interactions and mortalities in U.S. Mid-Atlantic bottom trawl gear from 2009 to 2013. Based on the results, the potential conservation benefits of hypothetical spatial closures or turtle excluder device (TED) requirements in times and areas of high estimated interactions is then evaluated. Loggerhead interaction rates were modeled as a function of retained catch, depth, latitude, and sea surface temperature. From 2009–2013, a total of 1156 (CV=0.13, 908–1488) loggerheads were estimated to have interacted with bottom trawl gear in the U.S. Mid-Atlantic, of which 479 resulted in mortality. The total number of estimated interactions was equivalent to 166 adults, of which 68 resulted in mortality. The trawl fishery targeting Atlantic croaker (Micropogonias undulatus) in the southern Mid-Atlantic had the highest turtle interactions among fisheries investigated; this may be due to larger mesh sizes in the mouth of the trawl and high headline height of the gear. The potential conservation benefit of hypothetical spatial closures or TEDs differs depending on the metric used to define “benefit”, and further depends on factors such as the spatial and temporal design of the closure, the magnitude and distribution of effort displacement, the spatial distribution of observed loggerhead life stages, and the assumed survival rate of animals passing through TEDs.