Communications of the ACM
'Backpack' Computers Help Scientists Track Wild Animals in Hard-to-Reach Habitats
Vampire bats tagged with wireless network sensors.
Credit: Jineth Berrio Martinez
To truly understand an animal species is to observe its behavior and social networks in the wild. With new wireless biologging network technology, researchers are able to track tiny animals that divide their time between flying around in the sky and huddling together in caves and hollow trees — by attaching little backpacks to them with glue.
The researchers describe their work in "Thinking Small: Next-Generation Sensor Networks Close the Size Gap in Vertebrate Biologging," published in the journal PLOS Biology.
The high-tech backpacks, which can communicate with each other and ground-based receivers, provided data for an earlier study published in Current Biology in 2019 showing that vampire bats developed social bonds in captivity that they maintained in the wild.
The wireless network developed by a team of engineers, computer scientists, and biologists contains functions similar to what are found on smartphones — such as motion detection and Bluetooth-style connectivity — at a fraction of the weight and energy consumption.
Keeping the system automated and lightweight was critical to the success of the network to track adult vampire bats, which weigh between 1 and 1.5 ounces and grow to 3.5 inches in length. Using devices that can track larger animals, such as those incorporated into harnesses or necklaces, wouldn't work for bats or other small species.
"Using backpacks on bats saves weight and it also makes sure the sensors fall off easily," says Simon Ripperger, lead author of the PLOS Biology paper and a postdoctoral scholar in evolution, ecology, and organismal biology at The Ohio State University. "We don't really want the bats to have that burden of additional weight for extended time periods."
The sensors often get scratched off in the bats' roost within about two weeks. The researchers retrieve them if they can to recycle the backpacks and recharge and reuse the batteries.
Though the study describes the complexity of constructing the network and testing its effectiveness on bats, the system would work for other animals, such as birds, rodents, reptiles, and amphibians, Ripperger says. Hence the team's name for the network: Broadly Applicable Tracking System, or BATS.
When Ripperger was pursuing his Ph.D. a decade ago, he used a much more primitive system to study bats. He relied on radio-telemetry, sometimes running after the animals, tracking their flight path with an antenna in hand. At best, he might be able to calculate where they were every two minutes over a 30-meter area.
"It was a time-consuming, exhausting, and inaccurate method," he says.
With funding from the German Research Foundation, the grant's principal investigators at the Museum of Natural History in Berlin, and a number of German universities assembled an interdisciplinary team and set out to make a better system. Ripperger was a postdoc at the time at the natural history museum, and is still a visiting scientist there.
The work took about seven years, with computer scientists writing code from scratch to come up with the highest-performance network possible using ultra-low levels of energy. The capacity of each battery that powers the network amounts to about 5 percent of a AAA battery's capacity.
The network consists of the tiny computers — accelerometers producing data when the bats are moving and proximity sensors to show when they are close to each other — all encased in each 3D-printed plastic backpack that weighs less than a penny. A series of basestations on the ground pick up signals and record data on bats' social activities and flight trajectories. The components are asleep most of the time, waking up when they receive a signal from another bat and then broadcasting every two seconds.
"One key advantage of our system is these wake-up receivers. They are in energy-saving mode and only wake up when they receive a signal from another bat, and then they are shouting, 'I'm here, I'm here!' and there's another receiver that comes into full consciousness and exchanges data," Ripperger says. "That's one way we conserve power consumption."
Despite the low power, the network produced robust results in various studies of different species of bats. A two-week test in which 50 vampire bats were tagged produced data on almost 400,000 individual meetings. Researchers can download all the data from the system onto their phones in the field.
Ripperger describes BATS and GPS, the most commonly used method for tracking animals on a grander scale, as highly complementary systems, with BATS able to collect signals in places where GPS cannot.
"If you want to study social behavior, once a bat enters a cave or tree trunk, a GPS logger doesn't give us information because the signal from the satellite gets interrupted. But inside the roost is where all the social behavior is happening," he says. "These are really two different approaches to studying animal behavior."
Using the system to tag vampire bats and the cattle they feed on can also help scientists better understand the spread of rabies, Ripperger says. He and colleagues are currently designing a conservation study to tag protected sand lizards living near railway lines in Germany to determine how track maintenance affects their movement.
This work was supported by the German Research Foundation, a Smithsonian Institution Scholarly Studies Grant, and a National Geographic Society Research Grant.
Gerald Carter of Ohio State co-authored the study. Additional co-authors are from the Smithsonian Tropical Research Institute in Panama and the Museum of Natural History, Friedrich-Alexander University Erlangen–Nürnberg, Brandenburg University of Technology, Technische Universität Braunschweig, Paderborn University and the Berlin-Brandenburg Institute of Advanced Biodiversity Research in Germany.
No entries found