This summer, a new underwater robot will start tracking some of the ocean’s top predators — including great white sharks — to learn more about their habits.
Sharks receive more media attention than the average fish, but marine scientists still don’t know much about how they live. Researchers often rely on anecdotal reports from commercial fishermen to understand the range and behavior of fish. But the shark industry is relatively small, so anecdotal information is scarce.
Biologist Chris Lowe from California State University Long Beach and engineer Chris Clark from Harvey Mudd College in Claremont, CA have been developing a shark-tracking robot for the past three years to learn more about the fishes’ habits.
“Working with computer scientists to answer these questions is a game changer,” Lowe told Wired. “It’s going to let us answer questions in biology that we have never been able to answer before.”
Marine scientists have used similar autonomous underwater vehicle (AUVs) to collect oceanographic data for decades, but the technology to simultaneously track a moving animal has only developed within the past three years. Other robots now follow penguins, fish, and marine mammals, but this is the first designed specifically to track sharks, Lowe said. His team studies a variety of species, but plans to focus their robot work on great whites.
(Another group from the Woods Hole Oceanographic Institute is close on their heels — more on that below).The team first uses hooks, lines and nets to catch a shark in the wild, embeds a transmitter tag on its dorsal fin, and releases it back into the wild. The AUV then follows along at up to 4 miles per hour, always lurking between 300 to 500 meters behind so as not to alarm the shark.
“Any predator is not going to be as good a predator if it knows its being followed,” Lowe said. “We have programmed the robot to not disturb the shark’s behavior.”
Two hydrophone receivers, situated on the front and rear of the AUV, pick up acoustic signals from the shark’s tag every second, and send the signals back to a computer on land. The computer converts these signals into a map of the track. Physical and chemical sensors on the AUV send off a suite of environmental data — including dissolved oxygen, temperature, and salinity — to the computers, while an image sonar and a high definition camera collect snapshots of the path.
Together, this information helps characterize the preferred habitats of sharks, and could inform future marine management plans.
“We know where [shark] hotspots are, but we don’t know what makes these spots special,” Lowe said.
The team successfully tracked a leopard shark off the coast of Los Angeles last summer. Since then, Clark and Lowe have redesigned their prototype to travel more efficiently through the water, and have been tweaking the software to allow two robots to work together. These adjustments will help conserve battery power — one of the main limitations of the project at this point. Currently, their longest track is six hours, but they hope to reach at least 20 hours in the future.
This summer, the team plans to use their revamped robot to track a juvenile white shark off the coast of Southern California.
Meanwhile, engineers at the Woods Hole Oceanographic Institute in MA are currently designing a similar shark-tracking robot for use on the East Coast, and hope to have it ready for fieldwork this summer as well.
Greg Skomal, a shark biologist at the Massachusetts Division of Marine Fisheries working with WHOI on their robot, is optimistic about both the East and West Coast projects. “This is great, new, innovative technology that will allow us to do unmanned tracks,” Skomal said. “I think it’s healthy to pursue two different lines of development, see which works, and then ultimately take the best results from both.”
External link: http://www.wired.com/wiredscience/2013/04/sharkbot/