Assessing the Capacity of Pelagic Sharks to Serve as Ocean Observing Platforms in the Mid-Atlantic Bight and Beyond

The use of marine animals as ocean observing platforms (OOPs) has been identified by scientists and managers as having the ability to significantly improve existing oceanographic models and data resources. As OOPs, animals are fit with oceanographic sensors with the capacity to measure in situ temperature and conductivity with an accuracy comparable to ship-based conductivity-temperature- depth (CTD) casts or autonomous drifters. While initial applications of this approach focused on regularly surfacing marine mammals, there is significant unrealized potential for non- air-breathing species that still surface regularly (specifically sharks) to sample a wider range of global habitats. Sharks occupy habitats across the globe, in particular habitats that are coincident with hurricanes. One region where sharks have enormous untapped potential to serve as OOPs and directly benefit coastal communities is the Mid-Atlantic Bight (MAB). Using pelagic shark species that occur in the MAB to generate high- quality water column CTD profiles has the potential to greatly improve regional oceanographic and hurricane models by providing real-time oceanographic data to the Integrated Oceanographic Observation System (IOOS) and the Global Telecommunication Service (GTS) which are directly accessed by ocean and atmospheric models. The objective of this paper is to develop an algorithm quantifying the effectiveness of different pelagic shark species to serve as OOPs using a new CTD tag (Sea Mammal Research Unit [SMRU], University of St. Andrews) to collect real-time oceanographic data in the MAB for hurricane prediction applications. To quantify the effectiveness of pelagic species to serve as OOPs, we analyzed Argos smart position and temperature (SPOT) tag data acquired for 14 shark species from 2002 to 2017, as well as transmitted and recovered archival data from pop-up satellite archival (PSAT) tags acquired for four species over the same time period. We assessed the quality of candidate species based on the quantity and type of Argos location classes a species was able to transmit (i.e., time the animal spent at the surface), the frequency of surfacing over time (a proxy for data transmission rates), as well as the range of depths following a surfacing event that a shark species was able to attain in a given time window (i.e. a proxy for the likelihood of receiving transmitted CTD profiles from a given depth under a specified tag battery duty cycle). Results indicate significant cross-species differences in time at surfacing and surfacing rates, with highly migratory and endothermic sharks surfacing more frequently (approximately between once every 6-20 hours) and for longer suggested periods of time (evinced by higher proportions of location classes with lower spatial error) than residential and ectothermic species. Comparisons of the distribution of depths a given species dove to following a surfacing event identified threshold depths for each species, whereafter receiving a CTD cast beyond the threshold depth would be less common and likely require a longer tag duty cycle and greater battery consumption.