Little is known about deep-water predatory attacks and behavior of white sharks (Carcharodon carcharias). Revealing how ocean predators operate and forage within their environment is fundamental to the protection of the species and their ecosystem. It is very difficult to quantify habitat use and behavior of large marine animals that may range widely and are not easy to observe directly, such as sharks. Studies of shark foraging ecology and feeding behavior are extremely difficult as predation events are rarely witnessed. Also, these studies often cannot identify the habitat in which a shark feeds, making definition of critical habitats difficult [1]. Currently, satellite and acoustic tags are used to follow the migration of white sharks, however this method limits information acquisition about detailed behaviors therefore leaving gaps in scientists understanding of dynamic movement of marine animals. Recent developments in fine-scale spatial 3D tracking and imaging of large sharks with an Autonomous Underwater Vehicle (AUV) have given scientists a never before seen view into hunting and foraging behaviors of these animals in the wild [2]. While tracking pelagic predators is no longer a novel idea, improved imaging sensors and navigation capabilities continue to evolve thus making observations of behaviors in deep water even more possible. Significant improvements have been made in hardware and software capabilities from lessons learned while tracking basking sharks and white sharks in Cape Cod in 2012 using a specially modified Remote Environmental Monitoring UnitS REMUS AUV known as SharkCam developed in the Oceanographic Systems Lab (OSL) at the Woods Hole Oceanographic Institution (WHOI. The capabilities and field results from a second expedition taken place near Guadalupe Island, Mexico from November 2013 are presented in this paper. The REMUS SharkCam system consists of a 100-meter depth rated vehicle outfitted with a circular Ultra Short BaseLine (USBL) receiver array for omni-directional tracking of a tagged animal. The vehicle interrogates the tag, and the round trip travel time of the response is then used to determine the range to the animal. This response is then beam-formed to determine the bearing relative to the vehicle, and the vehicle's compass is used to transform this into an absolute bearing. From this, the location in earth coordinates (latitude/longitude) can be determined. A second response from the tag is time delayed proportional to depth. The time delta between the two responses is used to determine the depth of the animal. This combination allows precise location of the tagged animal in three-dimensional space never before possible from an underwater vehicle. In the 2012 field trials we conducted to track great white sharks off of Chatham, MA, it was found that maintaining a solid track on the shark proved to be challenging. There were two hypotheses as to why this was the case. First, when the shark towed the transponder, the orientation of the transponder changed from its normal upright orientation to horizontal, and the torroidal beam pattern of the bottom mounted transducer become vertical, not horizontal. The vehicle was in an acoustic dead zone. The second hypothesis was that when the shark moved quickly, the Doppler shift of the signal caused the beamformer signal match to fail. A significant improvement to real-time tracking was the development of a shipboard tracking system (STS) that enabled operators to track the tagged shark and the REMUS AUV separately, allowing for long term tracking of the animal if the AUV needed to be recovered due to battery depletion or mechanical faults. It also enabled operators to know how well the system was working in real time by providing depth, range and bearing back to the ship.
