“Dolphins are phenomenally good at using echolocation, much better than man-made devices”
Dolphins echolocate with two-part acoustic beams. &T ,QUGƂP 5VCTMJCOOCT of Lund University explains how this could help us improve ultrasound technology
Why do dolphins need echolocation?
They use it for navigation, hunting for prey and possibly in social contexts. Dolphins always use acoustics as their primary sense. They generate short sound pulses, which bounce off surfaces and come back as echoes – the time it takes to return is a measure of how far away an object is. The beam is quite focused so the dolphins turn their heads to scan their environment.
How do they produce acoustic beams?
They have a structure below the blowhole called the phonic lips. Sounds come out through the melon, the rounded forehead – it’s one of the tissues responsible for the shape and formation of the beam. It’s basically an acoustic lens: the speed of sound is faster along the edges compared to the core of the melon, so the beam ends up cone-shaped. Dolphins have extremely short signals, usually much shorter than bats.
How did you study dolphin signals?
You need one dolphin echolocating in a specific direction and hydrophones – microphones for underwater use. I built my measurement system as a PhD student with 47 hydrophones. If we use all of them and record the cross-section of the beam, we can see finer details. I wanted to learn more about how they use these small details to solve tasks, because dolphins are phenomenally good at using echolocation, much better than manmade devices. I was looking at the signals and realised that regular methods couldn’t give me the information, so I talked with colleagues working with mathematical statistics and we developed a signal-processing algorithm that helps us look at signals in a much more detailed way.
What did you discover about the beam?
Even though the signal from the dolphins is very short, about 70 microseconds long, my previous research found that it actually consists of two intertwined beam components. The algorithm helped us decipher this, and we’ve discovered that parts of the beam consist of overlapping pulses. You get two slightly time-separated echoes from the upper part of the beam first as a low-pitched note and then a high-pitched one. From the lower part of the beam, you only hear a low-pitched echo. Dolphins have a frequency gradient across the beam, so in theory they could use this
information to locate objects more precisely: if prey is moving upwards in the beam, the pitch will get higher and higher in frequency.
What are the practical applications?
With the algorithm, we hope it can be applicable to non-destructive testing methods for diagnostics, such as if we want to measure the thickness of a very thin layer in the body. So improved image resolution. The other aspect is what we can learn from dolphins. For instance, the concept of using two intertwined sound beams with a frequency gradient cross-section might improve ultrasound machine performance. We could also make boat sonars more effective by employing these principles. After all, millions of years of evolution have formed dolphin echolocation to perfection, so there are definitely things we can learn from them.