Part 3: Bird recording in an acoustic slum

Degradation and Blackbird songs

First listen to a Common Blackbird recorded with a Telinga microphone at a moderate distance in a resonant Polish forest (CD1-45). Thanks to the parabolic dish, the sound is crisp and clear as if it had been recorded from much closer than it really was (50 m). Each song starts with the normal slower, mellow whistled part, which climbs and is followed by a faster, fine, higher-pitched series of notes at the end. This high frequency flourish has been emphasised by the Telinga mic. Then listen to the same bird recorded with the SASS from a similar distance (CD1-46). This is more of a birder’s ear perspective, or indeed what a neighbouring blackbird might hear. In the SASS recording, the mellow whistling at the start of each song carries much more clearly than the flourish at the end. The flourish is more easily degraded by the blurring effects of the forest acoustic, not only because the SASS doesn’t emphasise high sounds, but also because higher, shorter notes are much more easily absorbed and reflected by foliage and branches. The lower, longer notes of the mellow, whistled first part penetrate the forest better, and are not so easily degraded. As we will see later, degradation of sounds plays a very important role in the social relations of birds.

Woodcock roding and mics

Sometimes a parabolic microphone’s emphasis on higher frequencies can go beyond simply bringing a bird closer and it changes the picture. A good recordist using a parabolic microphone like the Telinga can compensate for this by pointing slightly away from the bird, or by altering the distance till the bird sounds more natural. But there are some cases where even these techniques are not enough. The ‘roding’ sound of a Eurasian Woodcock Scolopax rusticola, as it patrols over its territory at dusk, is a case in point. Roding combines two very different kinds of sound. Every three seconds or so, a series of deep grunts is followed by a very high-pitched squeak. The frequency of the grunts lies between about 0.6 and 1.5 kHz, perhaps not as deep as it seems, but much lower than the squeaks, which go from about 3 kHz to an amazing 14 kHz and back again in a 10th of a second! Pointing with a parabolic microphone, it is impossible to get the balance right between grunts and squeaks: inevitably the squeaks sound far too loud in the recording. Compare a recording made using a parabolic microphone (CD1-47) with a recording made using the SASS (CD1-48). With the SASS, not only does the balance of grunt and squeak sound more natural, at least from a ‘birder’s ear’ perspective, but the three-dimensional movement of the bird is also apparent to the listener, especially when using earphones. Recording with a stereo Telinga, the temptation is to swing the dish round and follow the bird, which stays more or less in the middle of the stereo picture. With a SASS you hear the total environment and so the beautiful reverberation of the European Nightjar Caprimulgus europaeus in the SASS recording also demonstrates another advantage of an omnidirectional microphone. It captures echoes and reflections from the sides as well as the signal coming straight from the bird, giving a more realistic picture of environmental acoustics.

So if they give more realistic-sounding recordings, why don’t we always use omnidirectional microphones like the SASS? The reason is that most of us live in an acoustic slum and very few places in Europe are as quiet as Białowieza forest and wind, planes, traffic and other low-frequency sounds get in the way of the bird. Ludwig Koch is considered to be one of the first bird sound recordists and reading his biography Memoirs of a birdman (1956) it was reassuring to hear him complain about background noise. In 1940 he was the first nature recordist to describe what has since become familiar: “We were just about to record this badly wanted ‘noise’ (the first ever recording of a bittern’s boom) when a squadron of fighter-bombers roared overhead and as they chose to circle round and round over ‘our studio’ all further efforts were useless… I was able to hear such longed for notes as the grunting of the Water Rail and Lapwings and Redshanks performing quite close to my microphone. It was quite tantalising not to be able to pick up those calls.” His Encyclopedia of British birds (1957), again a first, is one of  my treasures as it has little known facts about bird sound in it. For example it was the first place I read of magpies having a “soft musical song which carries only ten or twenty yards” and as soon as I read about it I desperately wanted to hear it. Not knowing the sound, it was a bit of a surprise when I realised that they sang in my garden for weeks on end. This Eurasian Magpie Pica pica was recorded in Italy but sounds just like the chaps in my garden (CD1-49). 

Sometimes in July when Arnoud watches the Tour de France on his television he notices that each time the cyclists passed through a French countryside there were nightingales singing. He knows France well and he’s sure Common Nightingale L megarhynchos stops singing around the 15th of June. So this would suggest that the TV sound engineers have done a little judicious engineering to brighten up this French countryside for their Dutch fans. Sound engineering is used in bird sound publications too. At its simplest it is a little filtering to remove low pitched sounds like distant engine noise. Next would come the removal of clicks, bumps and an annoying airplane sound half way through. A good example of this type of editing is the original recording I made of a Ural Owl Strix uralensis for CD1-02b, this recording being one of my first for the project. It was made in deep snow in bitterly cold weather, and days were spent in editing out the sounds of Dick, Killian, Pekka Pouttu and I shuffling in the snow to keep warm. This year we managed to replace it with a better recording and now you have a practically unaltered recording on this CD.

The Ural Owl is a good example because it has long gaps and on poor commercial CDs long gaps are often edited (shortened) to save space and keep the listeners’ interest. Sometimes sounds are looped so that, to make a poor recording of a song work, the same phrase is repeated with suitable gaps. Very occasionally bird sounds are recreated when there has been no known recording and the bird has become extinct or extremely rare. An example can be found for Slender-billed Curlew N tenuirostris in Morocco on the CD series on African bird sounds by Claude Chappuis (2000). 

Scenes are often created where the publication wants to convey a more wonderful scene than was captured. These normally include commentary and typically wish to show for example a Dartford Warbler S undata alarming at a Hobby Falco subbuteo which calls while a Wood Lark Lullula arborea sings in the background. For examples, see the publications by Geoff Sample (1996). We don’t indulge in much of this… well only once so far. Listen again to the Black-eared Kite calls. Notice the Steller’s Sea Eagle calling in the background? We edited that into the recording. There were Steller’s calling in the vicinity but not in that recording so, just as an example (not to grip the listener you understand), we slipped it in. 

Bird sound recordists are a suspicious lot and the final kind of editing may not even happen in real life. It is rumoured that this happens when a collection is being compiled and the authors don’t have a complete set. Then perhaps for completeness another sound recordist’s material is used without permission and edited in an attempt to avoid detection. Those collections where the date, place and recordist are not individually acknowledged are suspected of this. 

In The Sound Approach, editing is minimalised. To let you know how, when and how much we have edited a sound we have devised a numbering sequence for each recording. The unique catalogue numbers at the bottom of each track listing refer to the year, the number of the tape for that year, the recordist’s initials (AB for Arnoud van den Berg, DF for Dick Forsman, KM for Killian Mullarney, MC for Mark Constantine, MR for Magnus Robb), the place on the tape and finally the amount of editing, nought meaning no editing and three significantly edited. And the final number is the amount of filtering again from nought to three. 

Forget the bird, listen to the scenery

Fifteen years ago I was fortunate enough to be sitting on a hotel veranda in Key Largo, Florida, USA. It was early in the morning and a Grey Plover (cf CD1-01a) called as it flew over the still water, beautifully audible at 500 m. Later, in the heat of the day, the same bird called as it flew at about the same distance over mudflats, and I could hardly hear it at all. This was because the warm afternoon air rising from the mud had muffled the sound, while in the morning the cold air and calm water had carried it.

A cool, still night is perfect for carrying sound and also for concentrating the mind on listening: owls and other nocturnal birds vocalise much less on rainy or windy nights. Night sounds can carry even further across the surface of a calm, cold lake. Listen to the Great Northern Loon Gavia immer recorded in perfect conditions in the twilight an hour after midnight at Myvatn, Iceland. It had been sleeping close to the shore, and when Magnus crept up for a closer look, the loon woke up and it gave this beautiful series of wails (CD1-50).

Sounds recorded with the SASS, like all the recordings in this section, are perfect for gauging how the prevalent acoustics affects the way a bird sounds. Not only distance, but also temperature, wind, vegetation and topographical features can change a sound by the time it reaches our ears. Listen to the wails of the Great Northern Loon, each of which is answered by a faint echo slightly to the left, 2.8 seconds after the original wail (by listening carefully you can hear these at 0:04, 0:30, and 0:49). Just for fun, given that the speed of sound would have been about 333 m/sec (at around 3°C), the rock reflecting the echo must have been about 466 m away. This means that the sound has travelled just under a kilometre by the time we hear the echo. More fun? Listen again and see if you can also hear an answer (0:20), which is not an echo, but came from another bird in a neighbouring territory further around the lake to the right. 

If temperature and rising currents can affect the transmission of a sound, imagine what wind can do. For downwind listeners, a slight breeze can greatly improve the transmission of a sound. Meanwhile, upwind of the source much less is heard, and gusts of wind can obscure it completely. Rain or snowfall adds noise, and in poor weather most birds will save their energy; for others, like Rock Ptarmigan Lagopus muta, such conditions are normal. When male Rock Ptarmigans perform display flights, they take off with a call that has the rhythm of the famous wedding march HERE comes the bride. Then, after they have flown some distance, they land with …Comes the Bride…dededededede…. A Rock Ptarmigan at close range can sound very different from one heard at a distance. 

Writing about this is much easier than recording it. It was 02:00 am, and Killian and I were in a desolate part of northern Norway. He was away with the parabola, recording Eurasian Dotterel C morinellus courtship. After struggling through several flat batteries and loose cables in the freezing cold and snow, the wind buffeting my ears via the mics was finally too much for me. I picked up the equipment, threw it in the back of the car, wound up the windows and screamed obscenities. I now blush to think that Killian has a recording of this slight wobbly. Anyway, calm returned and at 03:00 am I made a series of recordings. They still weren’t quite right, and we have edited the start of one recording to the end of another. However, I promise this is exactly as it sounded in real life. So listen now to the single performance as the bird sings and flies over 100 m away and sings again. You can hear the snow on the mics and how both distance and the snow shower ‘weather’ away the higher frequencies of the call given on landing (CD1-51). The timbre as it lands in the distance is really quite different, a much more rounded, less dry sound, whereas in reality the bird is producing a more or less identical timbre both on taking off and landing.

In some cases, low parts of the sound can be the first to disappear at a distance. It depends on how powerfully the different frequencies are produced at the source, and also on other sounds in the environment, which may have a masking effect. Thanks to Marek Borkowski, who maintains large parts of the famous Biebrza marshes in Poland for conservation, we were able to make recordings of displaying Great Snipe Gallinago media. First, a stunning ‘bird’s ear’ recording of male Great Snipe making some amazing popping and whizzing sounds as they defend their patch at a traditional, communal display site or ‘lek’ (CD1-52). At close range, a deep and rather quiet tup-tup-tup… which Marek calls the ‘tractor’ sound can be heard, along with the sound made by their wings as they leap in display. 

Now listen to a full atmospheric recording that I made two years earlier (CD1-53). Under a sky completely full of the ‘drumming’ sounds of displaying Common Snipe G gallinago, a Great Snipe lek can just be heard. At a distance of about 60 m, which puts it firmly in the ‘birder’s ear’ category, only the high frequency popping sounds carry well. Besides the effects of distance, there is also competition from louder, closer sounds at similar frequencies: the throbbing wings of a displaying Northern Lapwing Vanellus vanellus, and the little barking sounds of Moorfrogs Rana arvalis.

Over a couple of years the three of us spent hours along the ‘Bison’s Ribs’ trail in the Białowieza forest in Poland, recording woodpeckers. It’s a magical place and you get the distinct feeling that forest gods are somehow in charge of proceedings. Each of us has his own style. Arnoud is the first out, hungrily prowling round with his parabola, arm outstretched, boots deep in the water, striding out for miles hunting down his target species. Magnus has an ill-deserved reputation for lying in, then needing two cups of coffee before he can function, then he also goes into prowl mode and will stay up all night at the drop of a hat. My own style is sedentary, I carry the SASS and cables in a wicker basket to a quiet spot, unfold my stool, sit down and sink into reverie as I wait for something to turn up. 

On the first visit, I recorded many woodpeckers drumming, and desperately wanted them to be the ever elusive White-backed Woodpecker D leucotos, but each time Arnoud returned from his travels he would tell me that he had only seen Great-spotted Woodpeckers. After several days, Arnoud was back asleep in his van when the forest gods gave me a little gift. A tour bus full of birdwatchers arrived some way away but visible from where I was sitting in the woods. The leader then set his scope on a nest hole. Having seen what was peeping out I couldn’t wait to get back to tell Arnoud, without bird bus detail, that I had at last found the White-backed’s nest hole. 

Competitive urges to make the ultimate recording are perfect for mischievous forest gods. Two years later as I lay-a-bed, Magnus, out early, had at last found and recorded (with the Telinga) the elusive Three-toed Woodpecker Picoides tridactylus drumming several kilometres along Bison’s Rib trail. He phoned me, I gathered my basket and SASS (perfect for this kind of recording), walked to his spot and settled down. Magnus returned for his breakfast, and while he was away, I made the following recordings (CD1-54a). I hope he has forgiven me.

As I said, the ancient forest ambience is mystical, almost sacred, but it all comes down to sound degradation. Fallen trees, grasses and more obviously the boughs, twigs and foliage of trees and bushes can block, absorb and reflect bird sounds. Listen to the drumming of this male Three-toed Woodpecker when he was still at a distance (CD1-54a). Don’t turn up the volume: listen to it as part of the forest ambience, taking its place among Wood Warblers P sibilatrix, Common Chaffinches, and a host of other forest birds. Three minutes later the recording was made, the mics at exactly the same spot, when the male flew in and started drumming just above them (CD1-54b). 

Listen to both examples again. In the more distant recording, the drumming sound reflects and echoes off hundreds of tree trunks and millions of spruce needles, twigs and sprouting alder leaves. Due to the varying distances travelled, the sound is arriving along each pathway at a slightly different time. It’s more difficult to hear the individual taps, and they lack definition. You can clearly hear that the sound is degraded. Nearly everything you can hear of the drum in the more distant recording is reverberation; very little sound is coming directly from the source. In the second cut the drumming sound, besides being much louder, is now quite different in timbre. At this close range, most of the sound is coming straight to the mics, with only a little reverberation heard after each drum. 

Acoustics change with the seasons, and the effects of vegetation growth and withering, prevailing weather conditions and the degree to which other species are also vocalising all have an effect. Birds may time their songs and displays to coincide with the most favourable acoustics. Woodpeckers do most of their drumming in the last couple of weeks before foliage appears. Three-toed Woodpecker, although we recorded it at this time, actually drums for a longer period, being an inhabitant of evergreen spruce forests.

How birds adapt to acoustics

I was in the northern forests of New Jersey, listening to the dawn chorus, when I first noticed that birds seem to take their turn to perform. An American Robin T migratorius first, then the stunningly beautiful song of a Veery Catharus fuscescens, followed by the Black-capped Chickadee Poecile atricapilla, Grey Catbirds Dumetella carolinensis, and then Baltimore Oriole Icterus galbula got started. By six o’clock, Chestnut-sided Warbler Setophaga pensylvanica, then at 06:15 Downy Woodpecker P pubescens, Hermit Thrush C guttatus, Yellow-throated Vireo Vireo flavifrons and on and on. It’s the same in Europe although such a diverse forest avifauna is hard to come by. Birds are adept at taking advantage of good acoustics and, at dawn, the atmosphere is at its most suitable for carrying sounds, while wind and convection currents are at their weakest. While it is too dark to forage and predators are not yet up and about, birds like to check whether all the neighbouring territories are still occupied, and answering each other in song is a quick and efficient way to do this. Most species join the dawn chorus in a predictable order each day, linked primarily with light intensity. It is thought that they get a little earlier in relation to dawn until the peak of breeding and then the pattern reverses. Birds make less sound straight after dawn when they forage, then there is a little surge. Things decrease through the day as the acoustics deteriorate, and increase again in late afternoon as they improve. 

A reed bed presents a formidable obstacle to bird sounds; the reed stems and leaves are a physical barrier, but they also rustle in the slightest breeze, masking many sounds. Despite this, the booming of Great Bittern (CD1-02a) is so powerful and deep that it can be heard several kilometres away. This type of low-frequency sound travels round and even through solid objects. For the tiny Cetti’s Warbler Cettia cetti to be heard it has had to develop a tremendously powerful song in order to penetrate the reed beds. On the other hand, an Aquatic Warbler A paludicola has to use very different tactics if its weaker, higher-frequency song is to be heard from any distance. Waiting for ideal acoustics in the late afternoon or just before dusk, it perches on one of the highest stems in its sedge field habitat. Listen to this SASS recording, made at about 10-20 m, as it then makes a series of song flights, crossing its territory from left to right during the peak of the evening chorus (CD1-55). In the distance (over 500 m), you can hear a Great Bittern booming, the reeds having degraded the sound so much that now it sounds like someone blowing over the mouth of a bottle. Some of the Common Cranes Grus grus trumpeting in the distance sound more like wind in the rigging of a boat as the distance distorts the sound. 

In mountain habitats, the ‘white noise’ of fast moving streams can overpower many bird sounds, while rocks can reflect and distort the sounds. On the other hand, birds that sing in cliff and mountain habitats use features of the landscape to project their song to incredible distances. Urban singers do the same with buildings: think of a Common Blackbird or a Common Starling Sturnus vulgaris singing from a rooftop. A beautiful example is provided by this recording of Caspian Snowcocks Tetraogallus caspius (CD1-56). Perched on top of a precipitous ridge in the Taurus mountains of southern Turkey, their songs carry over the rocks. These birds are singing in a natural amphitheatre several kilometres across. The sound reflects off several cliff faces, resonating beautifully and giving them all the more ‘presence’ (note for example the echo of the more distant bird at 0:54). Caucasian Snowcock T caucasicus is very sought after, and this is the first available recording (CD1-57). Arnoud and René Pop climbed the 2600 m to the snow on the steep slopes of mount Kuro to record these birds at dusk. Caucasian prefer the edge of the snowline and have to compete with noisy streams full of melt snow. You can hear several birds calling above the mountain streams, and they pitch their calls at frequencies higher than most of this ‘white’ noise. René, so exhausted by the climb, slept and snored, yet his low pitch snoring cannot be heard above the stream in this recording. 

When you look at a sonagram you can see the acoustics as well as the bird 7. For example, you can see a ‘sound tail’ after the notes as the sound echoes off of the surroundings.

One group of species that have to deal with a great variety of conditions is the larks Alaudidae. Listen to a recording of a sky absolutely full of Eurasian Skylarks Alauda arvensis over the vast steppes of Kazakhstan (CD1-58). As you can hear in the recording, the next bird is not far away, so wind and currents won’t damage the song too much at this kind of distance. In the stiff competition for a piece of the steppes to call their own, they have to sing with virtuosity and intricate detail. 

Listen to a lone Greater Hoopoe Lark Alaemon alaudipes (CD1-59), recorded in a huge desert of sand dunes in eastern Morocco. In this habitat, food resources are few and far between, and birds are spaced much further apart, at distances somewhere between 500 m and a kilometre. To avoid the heat of the desert and its unfavourable effects on song this lark sings in the morning when the atmosphere is cooler and calmer. One of the great sights of the desert is when this lark delivers its song while leaping up from a bush or sandy knoll then, still singing, somersaults in the air, flashing the white of its wings. This song is far carrying, perhaps because of the force of delivery but mainly because its simplicity, which allows long distance transmission with a minimum of degradation. 

The Arctic tundra is a cold and windy place. During the short summer, food is not so scarce for the few that are tough enough to cope with the climate, and Horned Larks Eremophila alpestris are able to live at densities somewhere between the extremes of steppe and sand desert. Listen to one struggling against the wind as it sings over the north-eastern Siberian tundra (CD1- 60). Magnus recorded this while he was stranded during a six-day period of easterly gales, and he had to hold the parabolic microphone with both hands to stop it blowing away. Horned Larks are thought to take turns in song flights (Drury 1961) and the song period is short so they don’t waste too much time in this fruitless pursuit. The bird’s main weapon against the wind may simply be repetition. Like many Arctic birds, especially waders, Horned Lark works on the principle that if it repeats a relatively simple kind of song often enough, the message will get through, even if most renditions are obstructed by gusts of wind.

The Wood Lark has a pretty cushy life, at least as far as singing is concerned. It breeds at moderate densities, in comparatively calm, semi-open habitats where it rises in song flight giving this magical song. Often, as in the case of this recording, it will perch in a tree or on a post to deliver its song. I just set up the equipment under its favourite tree, moved away to sit under another almond tree and waited for the performance to start. To conclude this section, listen to this singer of the Spanish dehesas, with accompanying Eurasian Hoopoes Upupa epops, Iberian Azure-winged Magpies Cyanopica cooki and a host of other Extremaduran delights (CD1-61). None of these other birds is using quite the same frequency range as the Wood Lark, so it has this acoustic niche all to itself.