Echolocation: How Bats and Dolphins See with Sound

Ranging by time of flight

The foundation of all echolocation is simple: emit a pulse of sound, measure how long the echo takes to return, and convert that delay to distance. Because the sound travels to the target and back, the round trip is twice the range.

d = c·t / 2      c = speed of sound, t = round-trip echo delay

In air:   c ≈ 343 m/s  → a target 1 m away returns an echo in ~5.8 ms
In water: c ≈ 1500 m/s → sound travels ~4× faster, so timing must be 4× finer

This d = ct/2 relation is identical to the equation used by radar (where c is the speed of light) and by engineered SONAR — the bat and the submarine solve exactly the same geometry, only the wave and the medium differ.

live demo · a pulse and its returning echo● LIVE

Chirps and why they beat a single tone

Many echolocating bats emit a chirp — a pulse whose frequency sweeps rapidly, often downward, across a wide band, for example from 100 kHz to 40 kHz in a few milliseconds. A long, constant-frequency pulse carries plenty of energy but blurs range; a short pulse pins down range but carries little energy. The chirp escapes that trade-off because range resolution depends on bandwidth, not pulse length:

Δd ≈ c / (2B)     — range resolution depends on bandwidth B, not pulse length

A wide-bandwidth chirp gives fine range resolution while still being
long enough to carry plenty of energy for detection.

Matched filtering: hearing the echo in noise

The returning echo is faint and buried in noise. The optimal way to detect a known signal in noise is the matched filter: correlate the incoming sound with a stored copy of the transmitted chirp. A chirp's autocorrelation collapses to a narrow peak — pulse compression — marking the true arrival time precisely, even though the transmitted pulse was long. Behavioural and neural evidence suggests bats perform something functionally equivalent, cross-correlating the returning echo against the call they just made.

Doppler shift and velocity

Motion changes the pitch of the echo: an approaching target returns a higher frequency, a receding one a lower frequency.

Δf ≈ 2·f₀·v / c    (round-trip Doppler shift, target speed v ≪ c)
Approaching target (v > 0) → echo frequency increases

Horseshoe bats take this further with Doppler-shift compensation: they lower their outgoing call frequency precisely so the returning echo always lands at the same "acoustic fovea" their cochlea is most finely tuned to — an elegant closed-loop control system that cancels their own flight Doppler.

Microsecond timing and jamming avoidance

Behavioural experiments show some bats can resolve changes in echo delay on the order of a microsecond or less — sub-millimetre differences in target range. When many bats fly together, they exhibit a jamming avoidance response, shifting the frequency, timing or structure of their calls to keep their echoes distinguishable from the crowd; some even fall silent or jam rivals to steal prey. Dolphins take a different approach, emitting broadband clicks rather than FM sweeps, produced in nasal structures and focused forward by a fatty "melon" that acts as an acoustic lens, timing each next click to the returning echo.

Biosonar vs engineered SONAR

Engineered SONAR rests on the same principles — d = ct/2, chirps and pulse compression, Doppler for velocity — but differs in the details. Bats use ultrasonic frequencies up to 100+ kHz for millimetre detail at short range, while long-range naval SONAR uses lower frequencies that travel further but resolve less finely. A bat continuously reshapes its calls, shortening them and firing faster as it closes on prey (the "feeding buzz"), whereas most engineered systems use fixed or menu-selected waveforms. Dolphins steer their beam with the melon; SONAR arrays steer electronically by phasing many transducers. And separating prey from background clutter remains a problem engineered systems still find hard — which is why biosonar continues to inspire new sensor designs.

Frequently asked questions

How do bats and dolphins measure distance to a target?

They use time of flight: emit a pulse of sound, measure how long the echo takes to return, and convert that delay to distance with d = c·t/2, where the factor of two accounts for the round trip. In air, sound travels at about 343 m/s, so a target 1 m away returns an echo in roughly 5.8 ms; in water, sound travels about four times faster at 1500 m/s, so timing must be four times finer.

Why do bats use frequency-swept chirps instead of a single tone?

A long, constant-frequency pulse carries more energy but blurs range; a short pulse gives good range resolution but little energy. A chirp — a pulse whose frequency sweeps across a wide band, for example 100 kHz to 40 kHz in a few milliseconds — escapes this trade-off, since range resolution depends on bandwidth (Δd ≈ c/2B) rather than pulse length, so it can be long enough to carry energy yet still resolve range finely once processed with a matched filter.

What is Doppler-shift compensation and which bats use it?

Motion shifts the pitch of a returning echo (Δf ≈ 2f₀v/c); an approaching target raises the echo frequency. Horseshoe bats actively lower the frequency of their outgoing call to precisely counteract their own flight-induced Doppler shift, so the returning echo always lands at the same "acoustic fovea" frequency their cochlea is most finely tuned to — an elegant closed-loop control system.

Try it live

Everything above runs in your browser — open Echolocation and emit chirps, time the echoes, and build a sound picture of hidden targets, just like a bat or a dolphin. Nothing is installed, nothing is uploaded.

▶ Open Echolocation simulation