When Detecting Bats, Methods Matter

bat flying at night
An Indiana bat exits a Kentucky cave. White-nose syndrome, habitat loss, and pesticides threaten extinction of the little (quarter-ounce) bat. Photo by Will Seiter, Copperhead Environmental Consulting, Inc., via Flickr.

If you want to record bat calls in summer, go early.

Detectors recorded significantly more high-quality call files during late June and early July than August. USDA Forest Service research ecologist Susan Loeb and colleagues published results from a bat detection survey in Acta Chiropterologica.

The likely reason bats had very high recorded activity in early summer was the presence of Indiana bats (Myotis sodalis) in maternity roosts near the detectors. In August, as the young bats started flying, the maternity colonies may have moved to other roosts or migrated.

However, other studies – of different bat species in different locations – have found more activity in August. One explanation could be that loud late-summer insects can drown out other sounds in southeastern forests.

Length of recording session also matters. Detectors that record from before sunset until after dawn significantly increased the chances of detecting bats.

Loeb and colleagues also investigated whether a microphone’s height or distance from the forest edge affected the quality of bat calls recorded or the probability of detecting a bat species.

“We conducted this study because the U.S. Fish and Wildlife Service has specific guidelines for surveying Indiana bats, an endangered species,” Loeb says. “Prior to our research, they only allowed directional microphones. This was problematic for people with only omnidirectional microphones. They wanted us to figure out the best way to set out omnidirectional microphones and how to minimize the noise.”

a person puts a bat detector into a tree hollow
Susan Loeb places a bat detector inside a hollow tree. USDA Forest Service photo by Adrienne DeBiase.

The omnidirectional microphones Loeb used for the study performed as well as directional microphones had done in an earlier study in similar Kentucky habitats. As a result of this study, the U.S. Fish and Wildlife Service now allows both directional and omnidirectional mikes. An omnidirectional microphone can detect bats in any direction, while a directional microphone only detects bats that are in front of it.

Results may vary with bat community or habitat type. When designing acoustic surveys, researchers need to consider bat behaviors, microphone type and orientation, time of year, and placement of detectors with regard to habitat types, distance to clutter, and reflective surfaces. The size and shape of a microphone’s area of detection should be chosen in relation to the types of bats likely to be recorded (e.g., high-flying bats with high-intensity echolocation calls versus gleaning bats with low-intensity calls).

Loeb participated in another study, led by Katherine Teets, then a graduate student at Clemson University, that compared active to passive monitoring and was also published in Acta Chiropterologica. Active sampling includes a surveyor who actively changes a microphone’s direction to follow the flight path of a passing bat.

By contrast, passive sampling uses automatic or remote recording, where the detector’s microphone is fixed in space: one height, direction, and orientation.

Active sampling can result in higher-quality calls and a longer call sequence, which can make call identification easier. However, passive sampling requires less labor and allows researchers to sample multiple sites simultaneously throughout the night.

two people with a tall acoustic detector stand in a forest
Two team members with an omnidirectional microphone. Photo by Dorothy Wells, U.S. Fish & Wildlife Service.

Active sampling typically for 20–30 minute periods between sunset and 2:00 a.m. The researchers used passive sampling at the same points in South Carolina’s Savannah River Site during the same 20-minute time periods that active data were collected. Each method detected each of five bat species being studied, at least once.

They also continued the passive sampling all night and recommend that practice. Collecting data passively throughout the night yielded detection probabilities that were consistently two to four times higher than active sampling, for all bat species.

These differences in detection probability mean that active sampling datasets can’t be compared to passive sampling findings. You might think a site had a lot more bats present, when really, you detected more simply because you used a better recording method.

Read the article on microphone type.

Read the article on monitoring modes.

For more information, email Susan Loeb at susan.loeb@usda.gov.

Access the latest publications by SRS scientists.