Habitat selection of a coastal shark species estimated from an autonomous underwater vehicle

Abstract

Quantifying habitat selection in marine organisms is challenging because it is diffi- cult to obtain species location information with multiple corresponding habitat measurements. In the ocean, habitat conditions vary on many spatiotemporal scales, which have important conse- quences for habitat selection. While macroscale biotic and abiotic features influence seasonal movements (spatial scales of 100�1000 km), selectivity of conditions on mesoscales (1�100 km) reflects an animal's response to the local environment. In this study, we examined habitat selec- tivity by pairing acoustic telemetry with environmental habitat parameters measured by an autonomous underwater vehicle (AUV), and demonstrate that migrating sand tiger sharks Carcharias taurus along the East Coast of the USA did not randomly use the coastal environment. Of the variables examined, we found evidence to suggest that sand tigers were selecting their habitat based on distance to shore, salinity, and colored dissolved organic matter (CDOM). Notably, temperature was not predictive of habitat use in our study. We posit that during their coastal migration, sand tigers select for specific mesoscale coastal habitats that may inform navi- gation or feeding behaviors. To our knowledge, this is the first empirical measure of mesoscale habitat selection by a coastal marine organism using an AUV. The applications of this method extend beyond the habitat selectivity of sand tigers, and will prove useful for future studies com- bining in situ observations of marine habitats and animal observations.