How many bats? Counting bats is a key part of developing effective conservation efforts, especially since White-nose Syndrome began decimating hibernating bat colonies in eastern North America seven years ago. But accurate counts are hard to come by and, for hibernation sites, typically require groups of biologists to take artificial lights or flash photography into caves. Yet disturbing hibernating bats can have serious consequences.
|A LIDAR image, produced in total darkness, of bats in Long Cave. Photo courtesy of Cave Research Foundation.
Virtually all bat counts are informed estimates, and reliability varies sharply. Improving the quality of those estimated numbers is an important goal.
Traditional bat counts involved simply counting the number of individual bats hibernating on cave walls for a few hundred or thousands of bats. When numbers climb higher, researchers count the number of bats in one discrete area, a square meter for example, then multiply by the total area used by bats for a rough calculation. A lot of research has gone into improving existing methods and developing new ones: low-light video, arrays of bat detectors, infrared beams that count the times they are broken by bats and other methods monitor bats emerging from their roosts. But few less-intrusive options exist for counting bats during hibernation.
Now bat biologists are exploring a technology that is more accurate than the human eye and can examine an entire cave in total darkness. That technology is called LIDAR – a combination of "light" and "radar" – and it requires just one researcher with a single, small headlamp. LIDAR sends rapidly pulsed laser beams to a target, then analyzes the reflected lights to measure exact distances and generate images.
Bat Conservation International partnered with the Cave Research Foundation and the U.S. National Park Service to conduct the first LIDAR scan of Long Cave in Kentucky's Mammoth Cave National Park. This cave is an important hibernation site for endangered gray and Indiana myotis, and WNS was confirmed there this year.
LIDAR was developed in the 1960s and has been applied to everything from atmospheric physics to archaeology (and in some police traffic-radar units). It has been used to produce precise, 3-D topographic maps of caves, including some used by bats. Only very recently, however, has LIDAR been tested for conservation biology.
A team of researchers in Malaysia recently published a paper describing their use of LIDAR to successfully count Horsfield's leaf-nosed bats in a Malaysian cave. The counting itself was automated with computer software the team developed. This cave-roosting species forms small colonies of up to about 100 individuals and typically roosts singly within caves. The Malaysian success prompted other researchers to try this new technique with cave-roosting colonies.
The LIDAR scan of Long Cave in Kentucky was conducted during their annual winter cave survey this past February. "The images are mesmerizing," says Katie Gillies, who leads BCI's Imperiled Species Program. "Individual bats are clearly visible, even though the images were taken in complete darkness. But the next challenge is the data analysis."
The Malaysian researchers were able to create an algorithm that automatically differentiated between individual bats and the cave walls, but they were dealing with a relatively small number of bats in a simple cave system. That task is far more complex at Long Cave, since its bats hibernate in large, dense clusters, making it much harder to distinguish between bats and rock walls. The cave is about 1.3 miles long.
"We are working to develop an automated way to count bats using this technology. We aren't there yet," Gillies said, "but the continued use and development of LIDAR will ultimately provide bat researchers with what could become a powerful new tool for our conservation toolbox."