Echolocation

Seeing with sound

Imagine being dropped into a cave so dark that your eyes are useless, and being told to catch a moth in mid-flight. A bat does exactly this thousands of times a night. It does not see the moth — it hears it, by shouting and listening to the echo. That is echolocation: an animal builds a three-dimensional picture of a pitch-black world by emitting sound and decoding the reflections.

The underlying idea is the same as a ship's sonar or a police speed gun: send out a known pulse and measure what comes back. The genius of biological echolocation is that the entire system — emitter, receiver, and a brain that does the maths in real time — is packed into an animal that may weigh only a few grams, and it runs fast enough to intercept an insect that is itself dodging and weaving. The bat is not passively waiting for sound, like an ear; it is an active sensor that creates the signal it measures, which is why we call it biosonar.

Three numbers come out of every echo. Delay (how long the sound took to return) gives distance. Direction (which ear heard it first and loudest, and how the ear's shape coloured it) gives angle. And change (shifts in pitch, loudness, and the echo's fine structure across successive clicks) gives the target's speed, size, and even surface texture — enough for a bat to tell a moth from a falling leaf.

From click to distance, step by step

Echolocation is a tight loop of production, reflection, reception, and computation. Here is the full chain for an insect-hunting bat:

1. Generation. The bat drives air past its vocal folds in the larynx using superfast laryngeal muscles to produce an ultrasonic pulse, then shapes the beam through its mouth or, in horseshoe bats, its nose-leaf. Typical frequencies are 20–120 kHz — well above the 20 kHz ceiling of human hearing.
2. Emission. The pulse leaves at up to roughly 140 dB SPL at the mouth, among the loudest sounds any animal makes for its size, and spreads outward at 343 m/s (in air at 20 °C). High frequency means short wavelength: a 50 kHz click has a wavelength of about 6.9 mm, fine enough to reflect off a mosquito.
3. Reflection. When the wavefront hits an object, part of the energy bounces back as an echo. A small insect returns only a tiny fraction of the incident energy, so the echo may be a millionth (–60 dB) as intense as the call.
4. Reception. The bat's large, mobile ears (pinnae) collect the faint echo. The cochlea — exquisitely tuned to the bat's own frequencies — transduces it into nerve impulses. A precisely timed middle-ear muscle reflex attenuates the outgoing call so the bat does not deafen itself, then relaxes in time to catch the return.
5. Computation. Neurons in the inferior colliculus and auditory cortex measure the delay between call and echo. Because range = sound speed × delay / 2 (the ÷2 because the sound travels out and back), a 12 ms delay means a target 2 m away, and a 1 ms delay means about 17 cm. Direction is computed from interaural time and intensity differences plus the pinna's spectral filtering. The brain fuses these into a 3-D fix and updates it click by click.

The precision is staggering. Big brown bats (Eptesicus fuscus) can resolve echo-delay differences of a few microseconds, corresponding to range differences under a millimetre — finer than their own wavelength. They achieve this by effectively cross-correlating the echo against a stored neural copy of the emitted pulse, extracting timing from the sound's fine structure rather than just its onset.

The players: who echolocates and how

• Microbats (laryngeal echolocators). Roughly 1,000 species of insect- and vertebrate-eating bats produce clicks in the larynx. They split into frequency-modulated (FM) bats that sweep a broadband downchirp (great for ranging clutter) and constant-frequency (CF) bats like horseshoe bats that hold a narrow tone and exploit Doppler shifts to pick out fluttering insect wings.
• Toothed whales (odontocetes). Dolphins, porpoises, sperm whales, and beaked whales generate clicks with phonic lips in the nasal passages and focus them through a fatty melon. Echoes enter through the fat-filled lower jaw. Sperm-whale clicks are the loudest, exceeding 230 dB re 1 µPa.
• Some swiftlets and oilbirds. These cave-nesting birds use audible-range clicks (1–15 kHz) for coarse navigation in the dark, not for catching insects — a far simpler system than the bat's.
• Trained humans. Blind echolocators such as Daniel Kish use mouth clicks around 2–4 kHz to detect walls and obstacles; expert clickers recruit the visual cortex to process the echoes.
• The prey, as co-evolved players. Echolocation is half of an arms race. Tiger moths (Arctiinae) hear bat ultrasound with simple tympanal ears and respond with evasive dives or by emitting their own clicks that jam the bat's sonar. Many moths and lacewings have ears tuned precisely to the frequencies their local bats use.

Echolocation across species — by the numbers

Property	FM microbat (e.g. Eptesicus)	CF horseshoe bat (Rhinolophus)	Bottlenose dolphin
Medium	Air (343 m/s)	Air (343 m/s)	Seawater (1480 m/s)
Peak frequency	25–60 kHz sweep	~83 kHz constant tone	40–130 kHz broadband
Call type	Short FM downsweep (1–5 ms)	Long CF tone + FM tail (10–60 ms)	Broadband click (~50–100 µs)
Source level	120–140 dB SPL	~110–120 dB SPL	up to ~220 dB re 1 µPa
Search pulse rate	~5–10 /s	~5–10 /s	variable, click trains
Terminal buzz rate	160–200 /s	up to ~100 /s	hundreds /s (buzz)
Detection range	~5 m (insect)	~5–10 m (fluttering insect)	>100 m (fish-sized target)
Special trick	Cross-correlation ranging	Doppler-shift compensation	Melon beam-forming, jaw hearing

Quantified figures that anchor the system

• Speed of sound sets the clock. 343 m/s in air at 20 °C; 1480 m/s in seawater. The ÷4.3 ratio means a dolphin's echoes return more than four times faster than a bat's for the same target distance.
• Wavelength sets the resolution. A 50 kHz call has a 6.9 mm wavelength in air; a 120 kHz call, about 2.9 mm. Higher frequencies image finer detail but attenuate faster. (In many FM bats the terminal buzz actually drops in frequency — by roughly an octave, sometimes below 20 kHz — to widen the sonar beam so a dodging insect stays inside it.)
• Atmospheric attenuation is brutal at ultrasound. Air absorbs high frequencies strongly — roughly 1–3 dB per metre at 100 kHz versus a fraction of that at audible frequencies — which caps bat detection range at a few metres regardless of how loud the call is.
• The echo is whisper-quiet. Spherical spreading loses energy as 1/distance² each way, so a small target's echo can be 60–110 dB below the outgoing call — a billionth to a millionth of the emitted power.
• Microsecond timing. Big brown bats discriminate delay differences of ~2–10 µs, equating to sub-millimetre range resolution; their auditory neurons phase-lock with this precision.
• Pulse-rate dynamic range. From ~5 clicks/s in open-air search to ~160–200 clicks/s in the terminal buzz — a 30–40× increase in update rate in the final 100 ms before capture.
• Loudest sounds in nature. Sperm-whale clicks reach ~230 dB re 1 µPa, the loudest sound produced by any animal; bat calls near 140 dB SPL in air are louder, relative to body size, than most aircraft.

Where it shows up — real organisms and arms races

The bat–moth night war. The most studied example of echolocation in the wild is the contest between bats and moths. Eared moths detect approaching bats from up to 30 m using ears with as few as one to four sensory neurons, and respond with power dives or erratic loops. Tiger moths escalate further: they fire trains of ultrasonic clicks from a thoracic organ called the tymbal. Some clicks warn the bat of the moth's toxicity (acoustic aposematism, demonstrated by Jesse Barber and William Conner in 2007), while others — at high enough rates — actively jam the bat's ranging by injecting phantom echoes (shown by Aaron Corcoran, Barber and Conner in 2009). This is a textbook predator–prey coevolutionary arms race playing out in the ultrasonic band we cannot hear.

Doppler-shift compensation in horseshoe bats. CF bats holding an ~83 kHz tone face a problem: their own flight speed Doppler-shifts the returning echo. They actively lower their call frequency in flight so the echo always comes back at the frequency their cochlea is best tuned to — a tiny acoustic "fovea." This lets them detect the rhythmic Doppler flicker of an insect's beating wings against a cluttered leafy background, picking a moth out of foliage that would swamp an FM bat.

Dolphins and the melon lens. Bottlenose dolphins can discriminate a target's material (steel versus brass) and detect a fish behind a screen, build mental object representations they can match across vision and echolocation, and even classify objects they have only "felt" by sound. Their fatty melon focuses the click into a forward beam, and pathological loss of echolocation (from sonar-induced injury) can strand whole pods.

Medical and engineering echoes. Human technology rediscovered the same physics: SONAR (1910s), medical ultrasound imaging, and ultrasonic rangefinders all time an echo to infer distance. Engineers now copy bat tricks — broadband FM sweeps for clutter rejection, beam-forming melons for SONAR arrays — and blind-navigation devices translate ranges into audible tones, an artificial echolocation aid.

Common misconceptions & pitfalls

• "Bats are blind." No — almost all bats have functional eyes and many see quite well; megabats (fruit bats) rely mostly on vision and largely lack laryngeal echolocation. Echolocation supplements vision in the dark; it does not replace eyes that work.
• "Echolocation is just hearing." Hearing is passive reception. Echolocation is an active sense — the animal generates the signal it then measures, like radar. The brain must subtract its own outgoing call, time the return, and ignore irrelevant echoes; that computation is the hard part, not the hearing.
• "The bat measures distance from echo loudness." Loudness is a weak, ambiguous distance cue because target reflectivity varies enormously. Distance is read from delay (timing), which is unambiguous given a fixed sound speed. Loudness and spectrum inform size and direction, not primarily range.
• "Higher frequency is always better." Higher frequency gives finer resolution but is absorbed far faster by air, shrinking range, and it also narrows the sonar beam. Bats trade pitch off against range and beam width — which is exactly why many FM bats lower their frequency in the terminal buzz, widening the beam to keep a swerving insect in view during the final strike.
• "Dolphins and bats inherited echolocation from a common ancestor." They did not — it evolved independently (convergent evolution). Strikingly, both lineages converged on the same changes in the hearing gene Prestin and other genes, a famous case of molecular convergence.
• "Sound speed is constant, so the maths is trivial." Sound speed depends on temperature, humidity, and pressure in air, and on depth, temperature, and salinity in water. A bat or dolphin must implicitly calibrate to its conditions; a 10 °C error in air shifts sound speed by ~2%, enough to matter at the millimetre resolutions bats achieve.