Bats have been on Earth for more than 50 million years. With over 1,400 species, they inhabit almost every corner of the world. Bats provide vital ecosystem services: they control insect pests, they pollinate plants and disperse seeds, thus playing a decisive role in reforestation. Today, bats are under unprecedented threat from human persecution and culling, widespread habitat destruction, pesticide use, invasive species, a general decline in biodiversity and the consequences of climate change. In order to protect bats, and thereby our ecoystems, we need to understand the way they live and forage.
Living their lives in darkness, bats use echolocation (biosonar) as their main remote sense. They emit ultrasonic calls that are reflected off an object in their surroundings and return to the bat as echoes carrying information about the object. There exists a wealth of knowledge from empirical studies, mostly in the laboratory, as well as from observational studies in the wild. However, the research has focused almost exclusively on isolated situations with one target object at a time. What remains severely understudied to this day is echolocation behaviour in natural, complex situations. Very few studies have addressed echolocation where the animal is tracking multiple targets, but they provide evidence to suggest that echo processing is dynamic and differs for situations with multiple targets compared to single targets. Furthermore, even more recent studies suggest that in some foraging situations the bat’s own movement is necessary for the echolocation task.
The crux of the matter lies in the nature of empirical research: we need controlled environments to come to conclusions, and any treatment affects the results. If we put a microphone in a bat’s natural flightpath to record its echolocating behaviour, that behaviour is affected the moment the microphone sends the first echo back to the bat. If we put the array off-axis, we cannot record the echolocation calls in a meaningful way from the perspective of the bat. The solution to this dilemma has now arrived in the form of backpack microphones, keeping the influence of our treatment on the animal to a minimum. The eTag used here is the first to record both the outgoing calls and incoming weak echoes. This offers unmatched fine-scale sampling of the entire acoustic scene perceived by the flying bat in addition to the fine-scale sampling of the bat’s movements.
The research goal is to quantify biosonar dynamics as a function of natural behaviour and habitats. What is the bat’s sensory percept in terms of sensory volume and resolution? How do bats adapt their sensory behaviour in the wild? Which role do external biotic and abiotic factors play, such as ambient light, ambient sounds and prey-induced sounds? Which role do behavioural contexts play, such as active-acoustic foraging with biosonar, passive-acoustic foraging and navigating?
My model species is the well-studied Neotropical fringe-lipped bat, also known as the frog-eating bat. It forages both in dense understory and above water puddles, and roosts in caves or hollow trees, from where it navigates to its foraging grounds in the rainforest via open flight paths. It is an opportunistic carnivore, hunting insects and small vertebrates such as frogs or fish. It uses varying cues and weighs passive and active acoustic modes depending on a series of factors such as availability of cues and background noise.
In Soberanía National Park, Panamá, I equipped 20 fringe-lipped bats with the eTags. I retrieved them from the bats’ day-roosts. Recordings generally started with the bats’ activity period, therefore all recording intervals covered the onset of the foraging period. The eTags recorded audio data via an ultrasonic microphone, and movement data with triaxial accelerometers and magnetometers. To ground-truth the wild data, I used the data recorded by the eTag in combination with video data from prey-capture trials that I had collected from bats in the flight cage at the laboratory.
I have found that Trachops cirrhosus uses echolocation continuously when in flight and sporadically when resting. The absence of audible prey cues right before a prey attack suggests that prey is not only detected by passive listening but also by means of echolocation. Further, I found that these bats catch and handle prey that almost reaches the bat’s own body size, judging by the duration of the chewing sounds after each kill. In contrast to large predators, frog-eating bats can forego high preemptive energy investments: These bats often spend less time searching for their prey than eating it and only fly about 10% of their time. They achieve success rates of >50%, which is rare for carnivorous animals, maybe because they overpower their prey out of flight. Since this hunting strategy critically relies on rich ecosystems with high prey densities, dwindling ecosystems give severe cause for concern about the fate of these miniature predators.
