Identifying bats on the wing relies on a combination of visual cues you can pick up with the naked eye and acoustic signatures you can capture with an ultrasonic detector. Neither method alone is foolproof. A bat’s silhouette, wing shape, flight path, foraging style, and the time and place it appears all narrow down the possibilities before you ever press record on a detector. When you do record, the structure of its echolocation calls can push an identification further, sometimes to genus and occasionally to species. But the process is messier than field guides suggest, and the tools have real limits worth understanding before you trust a software readout.
What a Bat’s Silhouette Tells You
Most bat identification guides start with plumage or pelage details you will never see on a fast-moving animal at dusk. In practice, the first thing you register about a flying bat is its size and the shape of its wings. Wing shape is described by aspect ratio, which is the relationship between wingspan and wing width. Bats with long, narrow wings tend to fly fast in straight lines and forage in open air. Bats with short, broad wings are more maneuverable and tend to stick to cluttered environments like forest interiors. That said, the actual range of wing proportions across bat species is narrower than you might expect, with most species falling into a fairly tight band, so wing shape alone rarely clinches an identification.
Tail membranes are another visual feature worth watching for. The membrane stretched between a bat’s hind legs, called the uropatagium, varies in size across species and plays a measurable role in flight control. Wind tunnel tests on a model based on the long-eared bat showed that adjusting the angle of the tail membrane changed lift, drag, and pitching moment, with the membrane acting as an extra control surface during slow or complex maneuvers.1PubMed Central. A potential role for bat tail membranes in flight control In the field, a large, visible tail membrane fluttering behind the body is a useful starting point for separating broad groups. Free-tailed bats, for instance, have a tail that extends beyond the membrane, giving them a distinctive trailing silhouette.
Flight Style as a Field Mark
How a bat moves through the air is sometimes more diagnostic than what it looks like. Different species have evolved to exploit different foraging niches, and those niches produce recognizable flight patterns. Aerial hawkers chase insects in open air with fast, swooping turns. Trawling bats skim over water, dragging their feet across the surface to snag prey. The large-footed bat Myotis adversus, for example, was documented both hawking insects in the air and trawling prey from water surfaces with its feet.2Journal of Zoology. Flight performance, foraging tactics and echolocation in the trawling insectivorous bat Myotis adversus (Chiroptera: Vespertilionidae) Gleaning bats, by contrast, pluck insects off leaves, bark, or the ground, and their flight often includes distinctive hovering phases before an attack. The long-eared bat Myotis evotis was found to alternate between gleaning moths off surfaces and catching them in the air, with a clear search-hover-attack sequence visible during gleaning attempts.3PubMed. Substrate-gleaning versus aerial-hawking: plasticity in the foraging and echolocation behaviour of the long-eared bat, Myotis evotis
If you see a bat repeatedly circling a streetlight and making tight turns, it is almost certainly an aerial hawker working an insect swarm. If one is cruising just above the surface of a pond in long passes, it is likely trawling. A bat fluttering in place near a tree trunk before darting in is probably gleaning. These behavioral signatures are not species-level identifications, but they can immediately narrow your search to a handful of candidates in most regions.
When and Where the Bat Appears
Context does a lot of the identification work before you even see the bat clearly. The time of emergence after sunset varies reliably among species, and some researchers have used this as a sorting tool. In a study of neotropical bats, the first species to leave their roosts appeared while ambient light was still relatively bright, around 30 to 300 lux, while the latest emergers waited until light had dropped below 5 lux.4PubMed. Sunset-related timing of flight activity in neotropical bats This pattern holds broadly across temperate regions too. Pipistrelles and noctules, for example, tend to be among the earliest fliers in European skies, sometimes appearing while swifts and swallows are still active. Horseshoe bats and long-eared bats typically emerge later, in deeper darkness.
Where a bat is flying also constrains the possibilities. Vertical stratification in forests is remarkably consistent. In both tropical and temperate forests, open-space foragers concentrate their activity near or above the canopy, while clutter-adapted species like many Myotis stay closer to the ground. Research in a neotropical rainforest found that bat activity increased roughly sevenfold from the understory to the canopy, and species diversity doubled.5Ecography. Patterns in the use of rainforest vertical space by Neotropical aerial insectivorous bats: all the action is up in the canopy A parallel pattern showed up in European old-growth forest, where open-space foragers and some edge-adapted species like the soprano pipistrelle used the upper canopy, while Myotis species were most active near the ground in forest gaps.6PubMed Central. Three-dimensional stratification pattern in an old-growth lowland forest: How does height in canopy and season influence temperate bat activity? If you are watching bats at a forest edge, the ones zipping along above the tree line are a different pool of candidates from the ones weaving between trunks at head height.
Some bats operate at surprising altitudes. Mexican free-tailed bats have been recorded foraging at up to 860 meters above ground level, with peak activity between 400 and 600 meters, coinciding with layers of migrating insects in the atmospheric boundary layer.7Fascinating Life Sciences. Bats Flying at High Altitudes – Section: 12.5 Bats Feeding at High Altitudes Migrating bats can fly even higher. Tracking of migratory Nathusius’ pipistrelles over the Baltic Sea recorded bats at altitudes of several hundred meters, possibly up to 2,500 meters, selecting heights with favorable tailwinds.8PubMed Central. Migratory movements of bats are shaped by barrier effects, sex-biased timing and the adaptive use of winds These high-altitude fliers are effectively invisible to anyone standing on the ground with binoculars, but they can still be picked up acoustically if detector range allows.
Listening With a Bat Detector
Most bat echolocation calls are ultrasonic, well above the roughly 20 kilohertz ceiling of human hearing. To hear them, you need a bat detector that translates those ultrasonic frequencies into something audible. The three main detector types each give you a different experience. Heterodyne detectors let you tune into a narrow frequency band and hear calls in real time, which is useful for quick field identification of species whose calls have a known peak frequency. Time-expansion detectors record a snippet of calls and play them back at a fraction of the speed, preserving all the frequency detail for later analysis. Full-spectrum detectors record everything across a wide frequency range simultaneously, producing files you can analyze with software afterward.
For casual field identification, heterodyne detectors are the most intuitive. You dial through frequencies until you hear the strongest signal, and the frequency at which the bat sounds loudest gives you a starting clue. This approach has been used effectively in large-scale citizen science programs. The UK’s National Bat Monitoring Programme, for example, trained volunteers to conduct standardized surveys using heterodyne detectors and produced statistically robust population trends for a large proportion of Britain’s bat species over many years.9Biological Conservation. Citizen science reveals trends in bat populations: The National Bat Monitoring Programme in Great Britain You do not need a lab or expensive software to get useful information. But heterodyne detectors sacrifice detail. For anything beyond the most distinctive species, you will want recordings you can examine on a spectrogram.
How Echolocation Calls Differ Among Species
Bat echolocation calls are not all created equal. The two broad categories are constant-frequency (CF) calls, where the bat holds a steady pitch for much of the call, and frequency-modulated (FM) calls, where the pitch sweeps rapidly from high to low. Many species produce calls that blend both elements in varying proportions. Three small bat species studied in Central and South America illustrate the range. Myotis riparius produced rapid FM sweeps dropping from around 120 kilohertz to just over 50 kilohertz in about two milliseconds. Rhynchonycteris naso used multiharmonic calls with both narrowband and broadband components, with most energy near 100 kilohertz. Thyroptera tricolor produced longer calls dominated by a single steady frequency near 45 kilohertz.10Canadian Journal of Zoology. Constant-frequency and frequency-modulated components in the echolocation calls of three species of small bats (Emballonuridae, Thyropteridae, and Vespertilionidae)
Some species are easy to separate acoustically. Among 22 bat species recorded in Italy, four could be identified by measuring a single feature: the frequency carrying the most energy. Adding call duration, start frequency, and end frequency into a classification model raised overall accuracy to about 82 percent. When the analysis was done at the genus level, with a few more parameters included, accuracy climbed to 94 percent.11Journal of Zoology. Identification of twenty‐two bat species (Mammalia: Chiroptera) from Italy by analysis of time‐expanded recordings of echolocation calls A similar approach in Australia’s Little Sandy Desert successfully separated all nine bat species present using the peak frequency and the sharpness of the frequency peak in their search-mode calls.12Australian Mammalogy. Identifying Little Sandy Desert bat species from their echolocation calls In regions with relatively few species, acoustic identification can work well.
Horseshoe bats and their relatives are among the most acoustically distinctive. These bats produce long CF calls and have evolved a remarkable trick called Doppler-shift compensation. As a horseshoe bat flies forward, the echoes bouncing back from objects ahead are shifted to a higher frequency by the Doppler effect. The bat counteracts this by lowering the frequency of its outgoing call so that the returning echo stays at a precise, preferred frequency tuned to its hearing.13PubMed Central. Precise Doppler shift compensation in the hipposiderid bat, Hipposideros armiger This behavior has been confirmed in multiple species across different families, including the Taiwanese leaf-nosed bat, which adjusted its call frequency depending on flight speed to keep the returning echo constant.14PubMed. Doppler-shift compensation in the Taiwanese leaf-nosed bat (Hipposideros terasensis) recorded with a telemetry microphone system during flight If you hear a long, steady tone on your detector from a bat flying in a temperate or tropical area, a horseshoe bat or a close relative is a strong guess.
Why Calls Change With the Surroundings
One of the biggest complications in acoustic identification is that bats do not always make the same call. They adjust their echolocation on the fly depending on what they are doing and where they are doing it. A bat cruising in open air produces longer, lower-frequency search calls designed to detect distant targets. The same bat entering a cluttered forest switches to shorter, higher-frequency calls that give better resolution at close range. Research on two North American Myotis species showed that clutter and species both had significant effects on the structure of search-phase calls, and that a neural network could correctly classify more than 94 percent of calls to species in high-clutter conditions.15Oxford Academic (Journal of Mammalogy). Effects of Clutter on Echolocation Call Structure of Myotis septentrionalis and M. lucifugus The flip side of this finding is that calls recorded in open habitat may look quite different from calls of the same species recorded in forest, potentially leading to misidentification if you only have reference recordings from one setting.
Weather also affects what you hear, or whether you hear anything at all. Atmospheric attenuation, the loss of sound energy as it travels through air, hits high-frequency sounds much harder than low-frequency ones. At 20 kilohertz, attenuation maxes out around 0.9 decibels per meter under the most absorptive conditions. At 110 kilohertz, that figure exceeds 6 decibels per meter.16PubMed Central. Weather conditions determine attenuation and speed of sound: Environmental limitations for monitoring and analyzing bat echolocation Temperature and humidity both influence attenuation in complex, nonlinear ways. The practical upshot is that a bat calling at 100 kilohertz is effectively inaudible to your detector beyond a few meters, while one calling at 25 kilohertz can be picked up from much farther away. This means surveys conducted on a hot, dry night will have a different detection range than those on a cool, humid one, and high-frequency species will always be underrepresented in acoustic datasets relative to low-frequency ones.
Measurements of atmospheric attenuation on bat-like signals confirmed attenuation in the range of 1 to 10 decibels per meter across the 20 to 200 kilohertz range, with the heaviest losses at the highest frequencies.17The Journal of the Acoustical Society of America. Measurement of atmospheric attenuation for bat sonar signals This physical constraint is so severe that it likely shaped the evolution of bat echolocation itself, limiting how high-frequency a bat’s calls can practically be before the detection advantage of short wavelengths is wiped out by atmospheric absorption.
How Reliable Is Automated Call Classification
If you record bat calls with a full-spectrum detector, you will probably want to run the files through identification software. Several commercial and open-source programs exist, and their accuracy varies considerably. A comparison of multiple versions of widely used classification software found that overall agreement between different versions ranged from 28 to 82 percent, depending on the species, the recording location, and which software versions were being compared. Newer versions tended to be more conservative, assigning fewer species-level identifications, but were not substantially more accurate than older ones.18Wildlife Society Bulletin. Testing Accuracy and Agreement among Multiple Versions of Automated Bat Call Classification Software
A European evaluation of automated classifiers found they could always distinguish rain noise from bat calls and could usually assign bats to broad groups like Nyctalus/Eptesicus, Pipistrellus, Myotis, or Plecotus with 68 to 100 percent accuracy. Species with structurally distinctive calls, such as the soprano pipistrelle and the barbastelle, were identified correctly 83 to 99 percent of the time. But species with less distinctive calls were another story. Identification of Vespertilio murinus succeeded only 16 to 26 percent of the time, Myotis species ranged from 4 to 93 percent, and Plecotus auritus ranged from 0 to 89 percent.19Ecological Indicators. Testing the performances of automated identification of bat echolocation calls: A request for prudence Similar patterns emerged in the northeastern United States, where two of three tested programs exceeded acceptable accuracy thresholds for big brown bats, hoary bats, and tri-colored bats but performed poorly for several Myotis species, including the endangered Indiana bat.20PLOS ONE. Automated echolocation classifiers vary in accuracy for northeastern U.S. bat species
The general pattern is clear: software is good at separating broad groups and picking out species with unusual call shapes, but struggles with the many species whose calls overlap in frequency and structure. The genus Myotis, which contains over a hundred species worldwide, is a persistent headache for automated classifiers. If a program tells you it detected a Myotis but will not pin it to species, that is an honest answer. If it tells you confidently that a given call belongs to an endangered Myotis species, treat the identification with skepticism.
The Plasticity Problem
Beyond the inherent similarity of some species’ calls, there is a more insidious issue: bats sometimes shift their calls so dramatically that they start sounding like a different species entirely. This has been documented in European pipistrelles, a pair of species that were only recognized as separate based partly on their different peak call frequencies. The common pipistrelle typically calls near 45 kilohertz, while the soprano pipistrelle calls near 55 kilohertz. But researchers studying soprano pipistrelles in a region where common pipistrelles were absent found that the soprano pipistrelles had shifted their calls downward to match the frequency range normally occupied by their missing relative, producing false acoustic records of a species that was not there at all.21Behavioral Ecology and Sociobiology. Bat echolocation plasticity in allopatry: a call for caution in acoustic identification of Pipistrellus sp.
This kind of plasticity means that peak frequency alone cannot always distinguish even well-separated species pairs. The researchers used a combination of DNA barcoding, trapping, roost surveys, and acoustic recording to untangle what was happening, a level of effort that underscores how unreliable a purely acoustic identification can be for certain groups. In megadiverse tropical regions, the problem compounds. Many species lack reference recordings altogether, some have calls too similar to separate, and quantitative classification tools have rarely been tested.22Methods in Ecology and Evolution. Acoustic identification of Mexican bats based on taxonomic and ecological constraints on call design If you are working in a region with dozens of bat species, acoustic records alone should be treated as provisional identifications rather than confirmed ones.
Putting Visual and Acoustic Cues Together
The most reliable field identification combines everything: time of emergence, habitat, flight altitude, foraging behavior, wing silhouette, and call structure. Each layer of evidence on its own has blind spots, but they compensate for each other. A bat that emerges early, flies fast and straight along a river at canopy height, and produces low-frequency calls around 20 to 25 kilohertz is a very different candidate pool than one that appears late, hovers near vegetation, and calls above 50 kilohertz.
In practice, experienced bat surveyors develop regional search images. They learn which three or four species are common in their area, what those species look and sound like under local conditions, and which oddball records to flag rather than trust. The detector output gets you to genus in most cases, the behavior and context narrow it further, and the remaining ambiguity is something you learn to live with. Attempting to force every recording into a species-level identification, whether by ear or by software, creates more false confidence than real knowledge. The most useful skill in bat identification is knowing when to stop and label a record as “unidentified Myotis” or “probable pipistrelle” rather than pushing past what the data actually support.
High-Altitude and Migratory Bats
Bats you will never identify by sight alone are the ones flying hundreds of meters overhead during seasonal migration. Migratory bat species, particularly in Europe and North America, travel long distances between summer breeding areas and winter hibernation sites. Tracking studies have shown that migrating bats are sophisticated about altitude selection, climbing to levels where tailwinds reduce their energy expenditure. Nathusius’ pipistrelles crossing the Baltic Sea flew at ground speeds of roughly 10 to 17 meters per second, with wind assistance cutting their energy cost per unit distance by about 55 percent on average.8PubMed Central. Migratory movements of bats are shaped by barrier effects, sex-biased timing and the adaptive use of winds When headwinds or crosswinds picked up at higher altitudes, the bats dropped to lower levels where conditions were calmer.
Mexican free-tailed bats exploit migrating insect swarms at similar altitudes, foraging at 400 to 600 meters above ground and sometimes higher, with activity peaking on nights when migratory moths were more abundant.7Fascinating Life Sciences. Bats Flying at High Altitudes – Section: 12.5 Bats Feeding at High Altitudes These behaviors are largely invisible to ground-based observers and difficult even for acoustic monitors, since atmospheric attenuation limits detector range. Radar studies and radio-tagged tracking have filled in much of what we know about bats at these altitudes, but for the average observer, these high-flying bats remain a reminder that whatever you see and hear from the ground is only a fraction of the bat activity happening overhead.