Like so many other biologists, we want to put a "critter cam" on our study animal, the Giant Pacific Octopus. A critter cam is an animal-eye-view camera typically mounted on the back or the head of an animal. They have been used quite successfully to gather data on how an animal interacts with its habitat when it is very difficult for an observer to follow an animal in its daily rounds. For example, critter cams have been mounted on Sperm Whales (see the Smithsonian site In Search of the Giant Squid) and on seals in the Antarctic.
Unfortunately, the octopus (and many other animals of interest) are not large enough to carry a camera on their own. So we want to design a mini-robot that will carry a video camera, follow an octopus around under the water, and always keep the animal in the camera's viewfinder. Since the robot will act like a Private Eye and tail the octopus everywhere, we think it should be called Shadow.
The challange for Shadow will be to keep the octopus in view, as well as to recognize when the octopus has crawled under a rock where Shadow cannot follow, and know to stand-by until the octopus re-emerges.
Shadow would allow us to collect data that cannot be obtained in any other way. Octopuses spend a great deal of each day resting, and it is not possible for an observer to sit underwater waiting for an octopus to begin its active cycle. An autonomous platform such as Shadow, however, could be stationed outside an octopus den and 'shadow' the octopus whenever it went out.
Shadow is a unmanned underwater vehicle (UUV) designed to conduct focal-animal sampling by video technique on a sonic tagged marine animal, such as an octopus. The system includes features novel to tracking an animal (signal tracking and standby mode while the animal is inactive). These features are under development in the current (2000-2003) phase of this project funded through NSF; while additional features are part of an intended future proposal (2003 or later). Field operation is envisioned in either semi-autonomous tethered or fully autonomous untethered modes to provide maximum flexibility.
Shadow III has now been built and is a 2 m/s self-propelled platform with an operating depth to 30 m and an estimated underwater active mission life of 9 hours (up to 72 h including stand-by mode). The platform is passively stable in pitch and roll, weighs 64 kg and carries a video-camera payload. This package has been pool tested to 6 m depth with the camera system and is scheduled for field testing in June 2003. This work so far has been completed by a total of nine undergraduate students (as senior projects) and three graduate students under supervising faculty at three different Universities (see Teams). Additional specifications for field-testing in June 2003 include:
This Photo at right show Shadow II, the 2001 prototype designed and
built by the University of Arizona AME Power Control Team.
See the Shadow History link for more details on precursors to Shadow III.
We are trying to understand what limits Enteroctopus dofleini populations in Prince William Sound. An important component of population limitation may be food supply (or energetics). For example, populations in PWS are an order of magnitude lower than those in British Columbia, yet our ability to compare between the two areas is limited because the diet of the octopuses is so different. In BC they eat primarily clams; in PWS primarily crabs. This makes it critical to understand the foraging behavior of octopuses. From surveys and laboratory work, we can get a good handle on diet preferences and energetic content of prey. An incomplete picture of handling time can also be obtained using existing techniques. However, to completely analyze how octopuses obtain their prey, we must collect data on the foraging behavior of octopuses in the field. Such data is unobtainable using existing techniques (telemetry, SCUBA, submersible). Shadow can help.
Differences in local abundance of octopuses could be due to differential mortality, differential recruitment, or habitat selection. In addition to analyses of foraging behavior, focal-animal sampling via AUV video-taping would allow us to experimentally test our understanding of the latter mechanism, habitat selection, through a series of octopus relocations. Octopuses would be captured in the intertidal or at depth using pots, outfitted with sonic tags, and released in different habitats. Their response to relocation and subsequent choice of habitat would help us determine which of several factors play an important role in octopus habitat choice. This in turn will help us understand how habitat characteristics (such as the type and availability of prey, available cover, or numbers of predators) may limit octopus densities in Alaska. These analyses cannot be conducted, however, with tracking data collected from the surface: they require video tape or visual observation to record habitat features such as substrate, kelp cover, prey availability, den availability, etc.
