Bird Sense (11 page)

Subsequent observations by others revealed that great grey owls use the same technique to capture rodents beneath the snow – that is, entirely by sound:

Watching and listening, the owl turns its head from side to side, occasionally peering intently towards the ground. Once the prey animal is detected, the owl dives down, appearing to hit the snow with its head but in fact at the last moment the feet are brought forward beneath the chin to grasp the prey.
28

To be able to hunt by sound alone, great grey owls must have extraordinarily acute hearing, but they also need to be able to pinpoint the source of sound very accurately in both the horizontal and vertical plane. They do this through a remarkable suite of auditory adaptations which include the facial discs, each of which acts as a large pinna funnelling sound towards the inconspicuous ear openings. Early naturalists including John Ray and Francis Willughby commented in the
1670
s on the fact that the barn owl’s eyes were: ‘sunk in the middle of [the facial feathers], as it were in the bottom of a pit or valley’. What Ray and Willughby did not appreciate was that the valleys on either side of the face, created by the facial disc, increase both the effectiveness with which sound is ‘gathered’ and the bird’s ability to localise sound. Three centuries on, and with the benefit of much greater knowledge, Masakazu Konishi, studying the hearing of owls, wrote: ‘When one sees the whole design of the facial disc, one cannot help thinking of a sound-collecting device.’
29

A second adaptation – known since the Middle Ages – is the relatively enormous ear openings of species like the great grey owl. The term ‘ear’ is potentially confusing – some owls are ‘eared’ – the great horned, long-eared and short-eared owls have feathers on top of their heads that look superficially like ears, but have nothing to do with hearing. What I’m referring to are the genuine ear openings which, as in the great grey owl, are also asymmetrical, one being higher than the other. Many owl species possess asymmetric ear openings and in most instances this affects only the soft tissues of the external ear, but in the boreal (or Tengmalm’s), saw-whet, Ural and great grey owl, the skull itself is also asymmetric, although the internal structure of each ear is identical.

The significance of this was recognised in the
1940
s when Jerry Pumphrey pointed out that asymmetric ears would make it much easier for the owl to pinpoint the source of sound. In the
1960
s, Roger Payne of the New York Zoological Society (and later famous for his studies of whale songs) conducted an ingenious experiment on a captive barn owl in a completely dark room, to demonstrate this. When the light was reduced over several successive days, the owl – which was observed by infra-red light (invisible to owls) – was able to catch mice in total darkness simply by homing in on the sound of the mice rustling leaves that covered the floor. As a test of what the owl was homing in on, Payne conducted an experiment in a room whose floor was covered with foam rubber, tying a dry, rustling leaf to the tail of a mouse. The owl swooped on the leaf (the source of the sound) rather than the mouse itself, dispelling the idea proposed earlier than owls might have infra-red vision or some other sense, confirming that sound alone was the cue.
30

Interestingly, the owl was able to capture prey in total darkness only if it was thoroughly familiar with the layout of the room; a bird moved to a new room was reluctant to hunt in total darkness. This makes sense: swooping around with no light whatsoever is potentially extremely dangerous, unless, of course – like the oilbird, which we’ll discuss in a moment – it has some additional sensory mechanism. It is also striking that, on capturing prey in total darkness, the owl instantly turned on itself in order to return directly to its perch, thereby avoiding any unnecessary flying around in the dark. The need to be familiar with the topography before hunting in total darkness explains why certain nocturnal owls remain in the same territory for most of their lives. Nights with no light whatsoever are few, but when they occur (for example, when there is heavy cloud cover and no moon) detailed topographical knowledge must determine whether or not an owl secures a meal without injuring itself.
31

One of the most intriguing features of owls is their extremely quiet flight; their wingbeats are almost imperceptible. When Masakazu Konishi analysed the sound of one of his barn owl’s wingbeats, he was surprised by its low frequency – around one kilohertz. The beauty of this is that, even when the owl is flying, it does not interfere with its ability to hear its prey. The rustling of mice in the undergrowth has a much higher frequency of between six and nine kilohertz. What’s more, since mice are relatively insensitive to sounds lower than about three kilohertz, they are unable to hear an approaching owl.
32

I return to Skomer Island each summer to continue the study of guillemots I started during the
1970
s. The high spot of the season is climbing on to the breeding ledges to catch several hundred chicks to ring (band) them so that we can later assess how old they are when they start breeding and how long they live. Ringing involves climbing on to the breeding ledges and using a carbon-fibre fishing pole tipped with a shepherd’s crook to catch the chicks. This is a social event, involving one catcher; one taker (who removes the chicks from the crook and places them in a net bag, prior to ringing); one ringer; and a ‘scribe’ (the note-taker, who records which ring is placed on which bird). It is also a very noisy affair, for the parent guillemots temporarily deprived of the chicks call loudly, and, deprived of their parents, the offspring reply in a much higher register. It is sometimes so noisy on the ledges that we have to shout the ring numbers to the scribe. By the end of a day’s ringing, our ears are often ringing too.

The chicks themselves are very good at distinguishing their parents, and vice versa. In fact, they have learned each other’s calls from before the chick even emerged from the shell: as soon as the first hole appears in the egg, the chick and adult start calling to each other. Under normal circumstances a guillemot colony is fairly noisy, but chicks stick close to their parents and there’s usually no great need to maintain continuous vocal contact. But if a gull or other predator causes adults to temporarily abandon their chick, it is vital that parents and chicks can find each other as soon as they return. This is especially important when it is time for the young guillemots to leave the colony, which they do en masse at about three weeks of age, at dusk. The chick, which is still flightless, usually jumps off the ledge into the sea below, either to join its father waiting on the sea or to be swiftly followed by its father. Staying together is vital. In very large colonies, like the one on Funk Island off the coast of Newfoundland, there may be tens of thousands of young birds leaving on the same night, and it can be difficult for fathers and chicks to remain in contact. They do so by means of their individually distinct calls. The departure of guillemot chicks is a cacophony of calling: a high-pitched
weelow weelow weelow
from the chick and a harsh, guttural growl from the adult. Remarkably, the vast majority of adults and chicks find each other on the water and swim off together out to sea where they remain, still together, for a further few weeks.

The guillemots’ hearing is so good that they can cut through the wall of noise, and single out those calls that really matter. This is literally a life and death situation, for unaccompanied chicks die. Natural selection has produced a hearing system that enables both adult and young guillemots not just to hear each others’ cries, but to distinguish them from all the extraneous cries around them. The way birds do this is by filtering and ignoring irrelevant noise, focusing instead only on those sounds that matter both for identifying their own species, and also for identifying specific individuals.

The ability to focus on one particular voice or song against a hubbub of background noise is known as the cocktail-party effect. This is a common problem for birds that live in a noisy world. Just think of the dawn chorus: in pristine habitats there may be as many as thirty different songbird species – with several individuals of each – singing at once, and the effect can be almost deafening. Each bird has to distinguish not only its own species, but also different individuals. In much the same way, starlings coming in to roost in city centres often stop off on a church tower or other high structure and start singing in their hundreds. Can they really distinguish each other in such swarms? The answer is probably ‘yes’. In some experiments (using relatively modest numbers compared with the huge roosts one often sees), captive starlings were able to discriminate between individuals on the basis of song even when the songs of several other starlings were played at the same time.
33

As well as having to deal with the sounds of other birds, the physical environment has a huge effect on what birds can hear. For marine birds it is the sound of the sea crashing against the cliffs at the colony; for birds breeding in reedbeds it is the rustling of a profusion of reeds; for those in rainforest it is rain falling on a million leaves.

It is obvious – and has been known for a long time – that sound gets fainter with distance. The term used to describe this degradation is ‘attenuation’ and it is also well known that attenuation differs between habitats. In a flat, open habitat sound travels further than in a forest or a reedbed. The first studies of the effect of attenuation on birdsong in different habitats were conducted in the
1970
s. Their results had been anticipated, albeit unconsciously, by the makers of the
1940
s Tarzan movies whose soundtracks often featured very specific bird calls – calls that we still associate with rainforest habitat: low-frequency, extended flute-like whistles. Gene Morton, based at a biological station, the Smithsonian Tropical Research Institute, in Panama, noticed the same thing, and wondered whether such calls had been shaped by natural selection for better transmission in dense habitats. The key to establishing whether the quality of the sound affects how well it can be heard at a distance was first to measure the attenuation of different-quality sounds in different habitats. Morton did this by playing sounds from a tape recorder and measuring their quality at different distances and in different habitats. Having shown that low-frequency, pure sounds travelled further in rainforest than other types of calls, Morton then recorded birds from both forest and adjacent open habitat and compared their calls. As he predicted, the forest-dwelling birds had lower frequency calls. In general, low-pitch calls travel further than high-pitch ones, which is why foghorns use deep sounds, and why bitterns and kakapos are among the record-holders for sound transmission.
34

Morton’s study was based on a comparison of different bird species, but other ornithologists wondered whether it might also be true within a single species living in different habitats. One of the first single-species studies was undertaken by Fernando Nottebohm on a very common and widespread Central and South American bird, the rufous-collared sparrow, known locally at as the chingolo. As predicted by Morton’s cross-species comparison, the chingolos’ song contained more long, slow whistles in forested habitat and more trills in open habitat.
35
Similar results were later obtained for Eurasian great tits breeding in dense forest compared to more open woodland habitat.
36

Dramatic evidence that birds respond appropriately to background noise comes from recent studies of birds in urban environments. Nightingales in Berlin sing louder (by a huge
14
dB) than their rural counterparts, and sing more loudly on weekday mornings during rush hour, when the traffic noise is loudest. Great tits, on the other hand, do not change the volume of their songs, but change their frequency or pitch to cope with urban noise. In both species the birds adjust their singing behaviour to ensure that they can still be heard, despite the background noise.
37

Increasing the volume of sounds uttered in a noisy environment is actually a reflex known as the Lombard effect, named after Etienne Lombard, a French ear, nose and throat specialist who discovered it in humans in the early
1900
s. The Lombard effect is most obvious when somebody is talking to you when, for example, you have your iPod headphones on and in response you – unwittingly – increase the volume of your reply, and they say: ‘No need to shout!’