Turtles test wireless network
Despite his best efforts to escape the clutches of two scientists from the University of Massachusetts and get back to the swamp he was just lifted from, the 18-kilogram snapping turtle finally gives up and lets Mike Jones and Matt Garber do their jobs.
Knowing where M16 goes could help scientists protect him, and help researchers perfect a new networking technology.
In an experiment taking place along the Deerfield River in western Massachusetts, two otherwise unrelated groups of researchers are working together: Computer engineers like Garber who are testing a new wireless communication network and biologists like Jones who are tracking snapping turtles— a species they worry may be headed for decline as land development shrinks their habitat.
The idea behind the technology is to create a network of constantly moving devices that record and store information, transmit data from one device to another, then relay all the saved information to a central location while running on self-charging batteries.
"A lot of the existing technology works great as long as you're not moving around and you have stable networks and people who could recharge batteries," said Jacob Sorber, a doctoral candidate in computer science who designed the network he calls TurtleNet, a project funded by grants from the U.S. National Science Foundation.
The solar-powered computers are light enough so they don't weigh the turtles down and they don't interrupt their mating habits, Jones said.
How it works
Stuck to the shells of about 15 turtles found in spots near the Deerfield swamp, the gadgets will take periodic readings of the reptiles' location and body temperature.
When one computer-carrying snapper gets within about a 10th of a kilometre of another, the machines swap information. The series of short-distance transmissions allows for long battery life in each computer, and the solar panels attached to the units are expected to constantly keep the batteries charged. Without a relay system, a longer transmission would require a larger battery that would drain too quickly or be too big for a turtle to carry.
The turtle-to-turtle relay ends when one of the snappers passes near a single base station that receives all the accumulated information. While Jones thinks the snappers may roam up to 16 kilometres from the Deerfield swamp they know as home, he says it's in their nature to return to the bog where the base station is.
Working like a cellphone sending a text message, the base station zaps the data to the UMass-Amherst campus about 20 kilometres away, where biologists are charting each turtle's whereabouts.
"We're trying to get a better idea of their range, the routes they take and where they hibernate," said Jones, who is working on a doctoral degree in biology. "If you have that information for a good number of turtles, you can predict what their patterns will be for the next 50 years or so."
By mapping where and how the snappers move, biologists are trying to generate information that could be used to help protect turtle habitats.
"People think they're a nuisance, they're aggressive and they're smelly," he said. "And you see a lot of dead snappers on the side of the road. But most of the turtles that people are running over are mothers trying to get somewhere to nest."
System means easier turtle tracking
Until now, tracking turtles has been a difficult— and messy— business.
Jones has been following turtles around New England by attaching radio receivers to their shells. When he goes looking for them, he has to carry a radio receiver while wading through swamps and bushwhacking through woods hoping to pick up a signal. And the radio batteries are good for only about two years.
If TurtleNet— which was launched in June— works, he'll be able to spend less time hunting for his subjects. The computers should let him know where the turtles are at any time.
Researchers from Princeton University have been using a similar technology during the past five years to track zebras in Kenya. Unlike TurtleNet, the Princeton project uses computers with larger batteries that could be more easily carried on collars attached to the strong, fast-moving zebras. Princeton scientists say their studies have shed new light on the animals' migratory patterns.
"These are early examples of using computer engineering to answer questions about biology," said Margaret Martonosi, a professor of electrical engineering at Princeton. "If you know where these animals are going and how they're moving, you could take steps to better preserve the land and their habitat."