Browsed by
Author: Nathan Pieplow

“Least” Bell’s Vireo

“Least” Bell’s Vireo

Bell's Vireo (subspecies arizonae), 3/28/2009. Photo by Dominic Sherony (Creative Commons 2.0).

I’ve always been mystified by the breeding range of Bell’s Vireo.  It inhabits three seemingly very different places — the dense deciduous tangles of the Great Plains, the mesquite thornscrub of southeastern Arizona, and the riparian chaparral of southern coastal California.  Sure, they’ve all got lots of dense thickets, but so do lots of places in between.

As far as I can determine, the only other birds that occupy Bell’s Vireo’s entire range are continent-crawling generalists like House Finch and Mourning Dove that can be found practically everywhere.  How can Bell’s Vireo be so forgiving of the differences between, say, a streamside plum thicket in South Dakota and a tangle of mesquite in the Arizona desert, without being able to tolerate the deciduous scrub of the Colorado foothills?

Part of the answer, of course, is that each of these habitats boasts a unique population of Bell’s Vireo with unique habitat preferences.  Great Plains birds are the yellowish, greenish nominate subspecies; Arizona is home to the much less colorful subspecies arizonae; and coastal California hosts the endangered “Least” Bell’s Vireo,V. b. pusillus, the grayest of them all.

Recently, Elisabeth Ammon of the Great Basin Bird Observatory asked me for help in determining whether there are any vocal differences between pusillus and arizonae.  During the upcoming breeding season, GBBO will be surveying sites in Death Valley National Park for Bell’s Vireo, including an area where the species was found last summer.  If it is found to breed in Death Valley, the national park’s management plan may depend on whether the birds are determined to be of the federally endangered “Least” subspecies or the commoner arizonae.

The Sibley Guide to Birds says no vocal differences are known between the subspecies, but the recently revised Birds of North America account suggests otherwise:

Geographic Variation

Little known. A comparison of the samples taken from California and Arizona show slight differences in repertoire size, song length and number of notes per song…. Field researchers subjectively report qualitative differences in songs in different regions.

Unfortunately, BNA’s claims of vocal differences are backed up by very little quantitative information — only the observation that “Least” Bell’s Vireo has an average repertoire size of 9.2 songtypes per bird, whereas arizonae averages 10.6 songtypes.  That’s a small difference indeed, one that means the repertoire sizes must necessarily overlap, and one that could even fall entirely within the margin of error.  Even if further investigation confirms this average difference, it would be of zero use in field identification.  And BNA says nothing further about differences in song length and number of notes per song.  Thus, we’re left to investigate the most slippery category:

Qualitative differences

The initial prognosis for this blog post was grim, since the “Big Three” internet repositories of bird sounds – the Macaulay Library, the Borror Lab, and Xeno-Canto – contain precious few recordings of “Least” Bell’s Vireo. As of this writing, Macaulay and Borror lack them completely, and XC has only a few, all from Baja California Norte:

(There are two more possible Leasts on XC, from Baja California Sur and the Yolo Bypass in California, but I couldn’t confirm the subspecies in either case.)

YouTube to the rescue

Strange as it may seem, at the moment, YouTube actually contains more minutes of “Least” Bell’s Vireo song than the “Big Three” audio websites combined.  Here are three definite “Leasts” (YouTube has at least one other possible/probable Least as well):

  • Update 4/2/2011: Matt Medler informs me that Macaulay actually does have several recordings of “Least” Bell’s Vireo [1 2 3 4 5]!  For some reason they were not originally visible to mortal eyes, but Matt has worked his magic, and now they appear when one searches the archive with the common name (“Bell’s Vireo”).

“Arizona” Bell’s Vireo

Now that you’ve wrapped your head around what the “Least” subspecies sounds like, check out these recordings of arizonae:

I’ve listened to all these recordings several times through, pored over spectrograms in Raven, and looked through Peter Beck’s 1996 thesis on the songs of “Least” Bell’s Vireo.  Maybe there are some diagnosable differences in song, but I’ll be darned if I can find them.  On current knowledge, the subspecies are indistinguishable by ear, and that’s the way it’ll stay for now.

Interestingly, there are a few differences in the calls from the different subspecies posted on Xeno-Canto, but I strongly suspect those are due to differences in the state of agitation of the individual birds, and not indicative of their genetic makeup.  Sorry, Elisabeth — I got nothin’.

(If you got somethin’, please leave it in the comments!)

A Pygmy-Owl Challenge

A Pygmy-Owl Challenge

The Northern Pygmy-Owl is a fascinating bird for those of us interested in vocalizations and taxonomy.  Many people think that what we call “Northern Pygmy-Owl” may contain somewhere between two and four species, based on regional differences in vocalizations.  Here’s a brief overview of the differences, according to The Sibley Guide to Birds (2000), with a typical spectrogram and sound of each:

Pacific birds

According to Sibley, birds along the Pacific Coast of North America “give very slow single toots (1 note every 2 or more sec).”  The example below is even slower than most; 2.5 seconds between notes seems pretty standard.  Although one might expect birds in Montana to be part of the Interior West group, the sole recording available seems to fit better in this group.

Interior West group

Very few recordings of this group are available online (or anywhere else) — just two or three from Colorado [1 2] and one from Utah.   They all seem to give single notes at very regular intervals, just over 1 second apart, totalling about 50 “toots” per minute when they’re going full-bore.

Mexican group (“Mountain” Pygmy-Owl)

Sibley says these birds “give mainly paired notes more rapidly (about 1 pair every sec).”  Paired and single notes are usually mixed together, as on the recording below, and the paired notes are only slightly closer together than the single ones:

"Mountain" Pygmy-Owl (Glaucidium gnoma gnoma), Big Bend National Park, TX, 3/28/2008.

However, “Mountain” Pygmy-Owls also sometimes forgo the paired notes in favor of a rapid-fire string of single hoots almost identical to the song of the Northern Saw-whet Owl:

What we don’t know

What exactly is this Northern Pygmy-Owl saying? You could help us find out. Photo taken 11/4/2008 in Mission, BC by NechakoRiver (Creative Commons 2.0)

Nobody knows exactly where the changes between these songtypes occur, or how abrupt they are, because we just don’t have enough data.  Most recordings of Northern Pygmy-Owl are of the highly vocal Mexican birds.  As I mentioned above, very few recordings exist of the Interior West birds.  There are none from potential areas of transition, like Idaho, Wyoming, northern Arizona, or New Mexico.

Now, my friend Arch McCallum is setting out to get to the bottom of this tricky situation — and you can help.

If you have access to Northern Pygmy-owls anywhere in their range this spring and summer, please do one of the following:

  1. Find a singing pygmy-owl.
  2. Get out a stopwatch and count how many “toots” the bird makes in one minute.
  3. Send this information, along with location, date, and time of day, in an email to Arch (mccalluma   AT   appliedbioacoustics.com) or post it in the comments below.

If you wish, you can also make a one-minute audio recording.  (Just take a video with your digital camera, or get a cheap voice recorder if you don’t already have the means.)  Actually, if you wish, you’re welcome to record (or listen to) the bird for longer than a minute!  The more data, the better.

Hope to see a lot of data points roll in this spring!  Here’s to good owling.

To Stereotype or Not

To Stereotype or Not

Lilac-crowned Parrot, a species with relatively unstereotyped vocalizations. Arizona-Sonora Desert Museum, Tucson, AZ, 5/13/2009. Photo by DrStarbuck (Creative Commons 2.0).

Lately I’ve been thinking a lot about stereotype. Not the racial or ethnic kind, but the degree to which a bird’s song is the same each time it’s sung. In a couple of recent blog posts, I showed that Violet-green Swallows and Evening Grosbeaks produce stereotyped vocalizations in complex strings, and therefore I argued that they were exhibiting true singing behavior. Shortly afterwards, Andrew Spencer asked me an intriguing question: why does a vocalization have to be stereotyped in order to qualify as a song? Are there any birds that sing non-stereotyped songs?

I can’t find an answer in the scientific literature.  Among researchers, the definition of bird “song” (as opposed to “call”) has been pretty controversial over the years, but the question of stereotypy isn’t the controversial part.  Most authors simply take for granted that bird songs should be stereotyped — or “crystallized,” in the terminology often used for learned songs.

And perhaps for good reason: I haven’t been able to find any clear-cut examples of unstereotyped courtship song.  Sure, there are the “subsongs” of young, inexperienced males, and birds may sing variable songs at different times of year — but it seems that when an adult male really needs to impress a female, in almost any species, stereotype becomes important.

This may be because females, at least in some species, prefer stereotyped songs.  In the Zebra Finch, for example, the degree of stereotypy varies depending on what the male is doing. When he’s by himself, each song is different from the last, albeit in very subtle and minor ways. But when a female shows up nearby and the male’s got courtship on his mind, he begins to sing directly to her, and those subtle differences go away: each song now becomes a precise copy of the last. Woolley and Doupe (2008) showed that females prefer this stereotyped courtship song over the more variable undirected kind — meaning that females are listening not just to how males produce their songs, but how they reproduce them.

Is stereotypy easier or harder?

Which demands more of the singer—stereotyped or non-stereotyped vocalizations?  My guess would be the former; it seems intuitive that having to produce a sound multiple times in precisely the same way, without any tolerance for variation, should be the more difficult task. But when it comes to demands on the listener, things might be the other way around. Recognizing a sound is likely to be easier when the sound only takes one form. If the singer is trying to send the same message in a slightly different way each time, it actually requires the listener to recognize a class of similar sounds rather than a single sound – presumably a more difficult cognitive challenge.

Let’s consider an example from human speech.  A common word like “birthday” is not going to sound exactly the same every time you say it. Sometimes you might say it quickly, sometimes slowly; sometimes while laughing or when your nose is stuffed up; sometimes with your voice rising toward the end of a question. Sometimes you might even sing it. In all these cases, the spectrogram of your voice saying “birthday” is going to look different—and yet, every single time, our brains reliably classify these sounds as renditions of the the same word. That takes some mental firepower—it’s difficult for computers to do, which is why speech transcription software remains imperfect.

Perhaps it’s no coincidence that the best examples I could find of birds that rarely give stereotyped vocalizations are also the most intelligent birds: parrots and ravens.  Nothing that these birds say in the wild meets the classic definitions of bird “song,” but they are apparently lifelong vocal learners — ravens, like parrots, can be taught to speak in captivity — and they apparently recognize other individuals of their species by voice, even though no two utterances are likely to be precisely the same.  Compare these calls from a Lilac-crowned Parrot in Mexico:

Four examples of calls from one Lilac-crowned Parrot; they sound identical but vary slightly on the spectrogram. Rancho Santa Barbara, Sonora, Mexico, 6/24/2010.

Here’s an unedited sample of consecutive calls from the same bird, showing how different types of calls grade into one another:

Sequence of Lilac-crowned Parrot calls, showing variation. Same bird as above.

Even these parrots have the ability, and apparently the need, to reproduce almost exactly the same call over and over, even if they don’t do it with quite the same precision as most other birds.  Is this lack of perfectionism a sign of intelligence?  Or merely a sign that they’re not in an amorous mood?  Are there any species in which the females are attracted to vocal innovators instead of virtuoso self-imitators?

At the moment, nobody really knows.

Do Evening Grosbeaks Sing?

Do Evening Grosbeaks Sing?

A while back, I asked in a blog post whether Violet-green Swallows sing — and I answered that they do, because they produce complex repeating strings of stereotyped syllables, even if those syllables don’t sound like much to the human ear.  Now it’s time to ask the same question of the poor, misunderstood Evening Grosbeak, whose vocalizations have often been vastly underappreciated, even by the authors of the BNA account:

Unlike most of their fellow oscines, Evening Grosbeaks do not make much use of the longer, more complex, learned vocalizations (i.e., songs) that characterize the vocal behavior of most songbirds. The Evening Grosbeak seems to be a songbird that doesn’t regularly use songs.

The BNA authors discuss the possibility of Evening Grosbeak song at some length, and by and large I think they do a thorough job of it.  However, much of their analysis rests on a fundamental assumption that song in songbirds must be musical, preferably with trills and warbles attached — and as we’ve already seen many times, that just isn’t true. Not until the end of the article do they really strike pay dirt:

Some observers (L. Elliott pers. comm.; G. Budney pers. comm.) have recorded long series of flight calls, sometimes intermixed with trills, from Evening Grosbeaks. During these sequences, calls (or pairs of calls) are repeated rhythmically at intervals of about 1 s during bouts that can last as long as twenty minutes. Budney reports that long bouts are a regular occurrence at dawn in the Sierra Nevada of California. Perhaps these aggregated, rapid-fire calls act as the functional equivalent of songs during the dawn chorus.

I too have heard song-like strings of flight calls from Evening Grosbeaks (for example, this bird), but I actually feel the best candidate for song in this species is its long strings of trill calls, not flight calls.  In the spring of 2008, Evening Grosbeaks had come out of the higher elevations to invade a number of towns in southern Colorado, and when I stopped in the town of Norwood, I found the place infested with them (along with Pine Siskins and Cassin’s Finches galore).  One male was sitting up atop a small aspen tree, broadcasting his trills to the world:

Evening Grosbeak song, Norwood, CO, 5/13/2008.

These trills are variable, but not randomly so.  In fact, they sort into three well-defined types that we’ll call A, B, and C.  They sound almost identical to the ear, but not quite; if you listen carefully, you can distinguish the order of the male’s calls on this 20-second cut: ABC, ABC, ABC, ABC, ACB, CAC.  When we zoom the spectrogram in, we can see that the differences are subtle, but distinct:

First three calls of the above recording, zoomed and cropped.

The “A” and “B” calls are quite similar at first glance, composed of backwards-L-shaped upslurs, while “C” is obviously distinctive, composed of zigzag backwards-N-shaped notes.  The key distinction between “A” and “B” is that “A” is polyphonic, while “B” is not.  It’s nearly impossible to determine this by looking at the fundamental (that is, the lowest and darkest of the three vertically stacked sounds on the spectrogram), but it becomes obvious when looking at the harmonics (the upper two versions) — the horizontal parts of the call are doubled in “A” and single in “B.”

These differences are not random.  All the “A,” “B,” and “C” calls are stereotyped — that is, they are perfect copies of one another, reproduced with exquisite precision.  Compare the first five renditions of each call on the recording:

The first five "A" calls on the above recording.
The first five "B" calls on the above recording.
The first five "C" calls on the above recording.

The Norwood bird isn’t the only Evening Grosbeak who’s been caught singing on tape.  The Macaulay Library has a couple of recordings [1 2] of singing males recorded a few days apart in central Oregon by Thomas Sander.  The first cut is rather brief, including only sixteen individual trills, but they too fit nicely into “A,” “B,” and “C” categories, much like those of the Norwood bird (click here to see a labeled spectrogram).  The second cut is more extensive, and more impressive — for most of its first three minutes it features an apparently solo singer who incorporates six different trill types instead of three (labeled spectrogram here).

Obviously, more study of Evening Grosbeaks is needed to determine the actual function of these suspiciously song-like vocalizations, but I strongly suspect that they function as male advertising calls.  Hopefully Aaron Haiman, who is currently studying Evening Grosbeaks, can get to the bottom of some of this!

Splitting Mountain Chickadee

Splitting Mountain Chickadee

The AOU’s North American Checklist Committee has posted a set of proposals currently under consideration.  The biggest surprise is a proposed split of Mountain Chickadee into two new species:

  1. Gambel’s Chickadee (Poecile gambeli) in the Rocky Mountains and the Great Basin, including populations in eastern Washington and Oregon;
  2. Bailey’s Chickadee (Poecile baileyae) in the Sierra Nevada, the Cascades, and the California coastal mountains.

Until the AOU proposal appeared, a potential split of Mountain Chickadee was not even on my radar screen.  However, two different molecular studies have found evidence that the two groups of chickadees are genetically quite distinct, and they apparently differ slightly in appearance, with “Gambel’s” Chickadees having a slightly longer tail, slightly more white above the eyes, and a faint buffy tinge to the underparts and back.

"Gambel's" Mountain Chickadee, Sandia Crest, NM, 3/26/2008. Note the buffy underparts and broad eyestripe. Photo by J.N. Stuart (Creative Commons 2.0)."Bailey's" Mountain Chickadee, Yosemite National Park, CA, 11/24/2007. Note the gray underparts and narrow eyestripe. Photo by Yathin (Creative Commons 2.0).

In addition, the proposal mentions song differences.  Sadly, it provides scant evidence for this claim.  Here’s the entire discussion of vocalizations:

Miller (1934:163) reported on song differences that he detected between Mountain Chickadees from southern Utah (wasatchensis) and from California (abbreviatus or baileyae): “I note repeatedly that the songs of this chickadee [wasatchensis] consists of two groups of notes separated by three or more half tones of pitch. In contrast to this type of song are those of the races P. g. baileyae and abbreviatus in which the greatest interval of pitch with rare exceptions is no larger than one whole tone.”

This is flimsy evidence indeed.  The difference in pitch interval that Miller noted could potentially be significant, but it’s only one metric by which to measure the complex matrix of geographic variation in Mountain Chickadee vocalizations.  Mountain Chickadee’s vast geographic range comprises a balkanized patchwork of dozens of different dialect regions, as one would expect in a bird that learns its song.  Miller was a single naturalist noting a single difference between just two or three of these dialects, in an era before sound recording and spectrographic analysis.

Furthermore, the evidence I’ve found so far doesn’t even corroborate his original observation.  Here’s the most common dialect variant of “Bailey’s” Chickadee, the version of Mountain Chickadee song most likely to be heard all throughout the Sierra Nevada:

"Bailey's" Mountain Chickadee song, Lava Beds National Monument, California, 5/29/2002. Recording by Geoff Keller (LNS 120257).

(Click here to listen to the recording at the Macaulay Library.)

Many observers in the region transliterate this (and similar songs) as “cheese-burger,” although there’s actually a short extra note in front of the “cheese” in most parts of the Sierra.  The “cheese” and “burger” parts of the song are separated by about a full step on average, with slight variations from place to place [1 2 3 4 5].  So far, so good — these recordings are mostly in line with Miller’s observations.

But the wide pitch intervals that Miller reported from southern Utah are certainly not representative of most “Gambel’s” Chickadees.  In this Borror Lab recording from northern Utah, the notes are quite close to one another in pitch, each about a half step lower than the last.  And most Mountain Chickadees in Colorado sing nearly monotone songs, like in this typical example:

Mountain Chickadee song, Larimer County, CO, 5/28/2008.

Here’s an example of a “Gambel’s” Chickadee from British Columbia that sings a songtype not unlike the Sierra Nevada “Bailey’s” song:

Meanwhile, here’s a “Bailey’s” from the heart of the Sierra Nevada that barely changes pitch at all, and here’s a “Gambel’s” from Wyoming that apparently sings two songtypes, one with a large pitch change and another that’s nearly monotone.

Perhaps the genetic data is clear enough to warrant a split of Mountain Chickadee, and perhaps vocalizations do differ systematically — they may even act as an isolating mechanism between the two groups.  But a far more in-depth study would be needed to demonstrate this.  On the basis of the evidence presented in the AOU proposal, I can see no reason at this time to add “vocalizations” to the list of reasons for the split.

Trumpeter and Tundra Swans

Trumpeter and Tundra Swans

What kind of swan is this? Are you sure? Ridgeway NWR, WA, 2/1/2009. Photo by SigmaEye (Creative Commons 2.0).

Trumpeter and Tundra Swans present a consistently underrated identification challenge in North America.  Field guides and websites present a number of visual field marks, often with clear-cut illustrations or photos to show the differences — David Sibley presents a particularly good summary — but many of the differences are variable (like bill shape and color), change with the angle of viewing (like head shape), or are difficult to judge in the field (like size).  Some of the best field marks, like the shape of the border between the facial feathers and the bill, may not help much in separating immature birds.

Many people think of swans as silent — ask them to imagine a swan and they will picture a species from Europe, the (somewhat) aptly-named Mute Swan, gliding serenely around a garden pond.  It’s easy to forget that both native North American species were named for their vocalizations:  “Trumpeter” and “Whistling” (as the New World subspecies of the Tundra Swan was known before it was lumped with the “Bewick’s” Swans of Siberia).  As those very different appellations suggest, the two species sound quite distinct, and voice can be a good way to identify them.

At the same time, swans are vocally quite versatile (especially Tundra).  I didn’t expect so much variety when I began my research, but then I never do.  Variability aside, the bottom line is that vocalizations of Trumpeter and Tundra Swans consistently differ in pitch and tone quality with almost no overlap, and should thus be an excellent way to tell the species apart in the field.

Trumpeter Swan

Trumpeters sound remarkably low-pitched and nasal compared to Tundras — their call can easily be compared to the sound of a Red-breasted Nuthatch or (believe it or not) an Ivory-billed Woodpecker call — with many authors drawing comparisions to tin trumpets and taxi horns.  The length and pattern of Trumpeter vocalizations is quite variable, and occasionally they can sound noisy (like in the last two calls on this recording).  Typically, however, they sound much like this:

Note that on Tayler’s recording above, several of the calls include voice breaks, which are characteristic of large birds like waterfowl, hawks, and gulls, but would never be heard from the likes of a Red-breasted Nuthatch.  The voice breaks contribute strongly to the tendency to describe these sounds as “trumpeting” or “bugling,” since voice breaks in bird sounds follow the same basic principle as changes between notes on a bugle (see this page for more information).  Often, the voice breaks are lacking from Trumpeter calls.

Tundra Swan

Tundra’s voice is variable in pitch and pattern, but virtually always higher than Trumpeter, and much less nasal.  Some authors compare it to the sound of Canada Goose, but in my experience the familiar honks of “Giant” Canadas are actually closer to Trumpeter calls.  Tundra typically sounds much more like a Snow Goose, fairly high-pitched and rather mellow:

Tundras, however, make a wide variety of calls, including short high “barks” that remind me of something an American Coot might say:

Tundra Swan in flight, Alaska, 6/14/1972. Recording by William Gunn (LNS 62668).

(Listen to the above sound at the Macaulay Library)

Sometimes they sound distinctly burry and strikingly reminiscent of Sandhill Crane:

Burry, crane-like Tundra Swan calls, Alaska, 6/2/2007. Recording by Gerrit Vyn (LNS 137568).

(Listen to the above starting at 3:40 on the recording)

Tundra Swans also apparently make some low moaning sounds, as attested by this recording at the Borror Library (probably a captive bird, given the location and the date):

Tundra Swan moan, Utah, 6/16/1973. Borror Laboratory of Bioacoustics #12361.

(Listen to the above recording at the Borror Lab)

But the only sound I’ve found that might cause identification confusion is this occasional low-pitched, rough, nasal braying:

Tundra Swan braying, Alaska, 6/12/2006. Recording by Michael Andersen (LNS 133225).

(Listen to the above at the Macaulay Library, starting about 2:20 in the recording)

Of all the Tundra Swan vocalizations I’ve heard, this one sounds the most like a Trumpeter Swan.  However, it may be given primarily on the Arctic breeding grounds, and usually appears interspersed with higher-pitched, more typical Tundra calls, so hopefully it shouldn’t cause too much confusion in the field.

For an excellent introduction to the variety of Tundra Swan calls, check out this recording by Gerrit Vyn of winter flocks approaching an evening roost in North Carolina.  It’s nearly 50 minutes long, but it really gives a good sense of what you can expect Tundras to sound like in the field.  (Note that some of the higher, hoarser calls on the recording may come from juveniles — but I don’t know enough about juvenile calls to say so with certainty.  As always, there remains more to learn!)

The Vowels of Birds

The Vowels of Birds

In my 2007 article in Birding magazine, I distinguished three methods of describing a bird sound in words: transliteration, analogy, and analytic description.  But the article was so focused on promoting the third of those methods that it gave the other two short shrift.  I was particularly scathing in my condemnation of phonetic transcription, bemoaning its “limited capacity to carry information,” with “little or no regard to pitch, tone quality, variation, or any other crucial components of birdsong.”

However, the more I study phonetic transcriptions, the more I become convinced that they tend to be more informative than I thought.  In fact, even though the people writing the transcriptions may be completely unaware of it, their choice of vowels almost always follows a consistent set of rules for indicating the pitch and inflection of the bird sound.  Consonants are another story, but the consistency of the vowel rules is extraordinary once you learn to see it.

The first principle is this: different vowels represent different pitches in bird sounds.  The following list arranges common (American) English vowels — plus the semivowels r, w, and y — from highest transcription pitch (at the top of the list) to lowest transcription pitch (at the bottom of the list).

  • ee and y (as in feet and yes)
  • ih (as in pit)
  • eh (as in pet)
  • ah (as in father)
  • oh (as in pole)
  • er and r (as in herd and rip)
  • oo and w (as in boot and win)

The lowest-pitched bird sounds in North America are those of doves and owls — and it’s common knowledge that they coo and hoot, respectively.  Meanwhile, if I tell you to listen for a “seet” or a “peep,” you’ll expect something high-pitched.  Only medium-pitched sounds get filled in with other vowels, like the nasal notes from this Cooper’s Hawk, which Sibley’s guide describes as “pek-pek” notes and the old Audubon Society guides transliterate as “cack-cack-cack” (using vowels from the middle of the chart):

But what we’ve just seen isn’t the cool part.  The cool part is how people use diphthongs — that is, vowel combinations — to consistently, systematically represent changes in the pitch of bird sounds.  The correspondence between vowels and bird sounds isn’t random — it’s based on the acoustic properties of the human voice and the way it produces vowels and semivowels.  The rules are simple:

  • Monotone sounds are transliterated by single vowels, never diphthongs: for example, “tseet,” “peep,” “hoo.”
  • Upslurred sounds are transliterated by two consecutive vowels, the first one lower on the chart, the second one higher.  For example: w + ee = “whee.”
  • Downslurred sounds are transliterated by two consecutive vowels, the second one lower on the chart than the first.  For example: ee + r = “eer.”
  • Overslurred sounds are transliterated by three consecutive vowels, the middle one highest on the chart.  For example: w + ee + oo = “wheeoo.”
  • Underslurred sounds are transliterated by three consecutive vowels, the middle one lowest on the chart.  For example: ee + oo + ee = “eeyoowee.”

Believe it or not, despite the huge variety of ways to transcribe bird sounds, most transcriptions follow these principles — and for good reason, as we shall see.

Upslurred sounds

The commonest call of the Great Crested Flycatcher is a classic upslur, and like all upslurred sounds, it rises on the spectrogram from lower left to upper right:

The Sibley, National Geographic, and Audubon field guides all transliterate this sound as “wheep,” which starts with an “oo” sound (w) and changes to an “ee” sound — thus, spanning the entire range of the pitch table from bottom to top.  This oo + ee combination virtually always denotes an upslurred sound, as it does in the “whit” calls of Empidonax flycatchers, the “squeet” call of Sprague’s Pipit, and the “kwit” flight call of Type 4 Red Crossbill.  A look at the human voice on the spectrogram shows the reason:

My lovely assistant Molly saying "wee, ree, oy."

Whenever the human voice pronounces a diphthong that begins low on the chart and ends high, the resulting spectrogram shows a distinct dark band that runs from lower left to upper right — in other words, an upslur.  However, this dark band does not represent the pitch of the person’s voice — note that the darkness moves independently of the harmonics (the underlying horizontal lines), which change with the voice’s pitch.  The dark bands are called formants, and they differentiate vowel sounds.  (For a detailed explanation of spectrograms as they relate to the human voice, see this page.)

Downslurred sounds

Like all downslurred sounds, the call of the Olive Warbler traces from upper left to lower right on the spectrogram:

Sibley transliterates this “teew” or “tewp,” National Geographic “phew,” and Audubon “kew.”  In all cases, the vowel combination is ee + oo = “ew,” as it is in Sibley’s transcriptions of the American Robin’s “tseeew” alarm call and the “kewp” flight call of Type 2 Red Crossbill.  Another classic vowel combination for downslurred sounds is ee + r = “eer,” like in the “pdeeer” call of Say’s Phoebe, the “veer” call of Veery, and the “cheer” call of Carolina Wren.  As you might expect, spectrograms of the human voice saying “eer,” “eew,” and “yo” show downward-sweeping formants:

My lovely assistant Molly saying "eer," "eew," and "yo."

Overslurred sounds

The burry overslurred call of the Couch’s Kingbird can be seen three times on the spectrogram below, mixed with shorter calls:

Sibley transliterates this as “kweeeerz” and Audubon as “queer” — in both cases, the vowel combination is oo + ee + er = “weer.”  National Geographic goes almost the same route, with er + ee + er = “breeeer.”  Similar patterns occur in Sibley’s “hweeeeeew” for Dusky-capped Flycatcher and “urrREEErrr” of Common Pauraque.  And, of course, a similar pattern can be seen in spectrograms of the human voice pronouncing these combinations:

My lovely assistant Molly saying "reer" and "weew."

The possibility of standardization

So far, I have merely tried to describe the way people do describe bird sounds, not the way that they should do it.  However, it may be possible to go one step further and create a standardized system by which transcriptions could communicate the basic properties of a sound to any audience unambiguously, and two people hearing the same sound would transcribe it the same way.  Much more work would need to be done, particularly on consonants, but it might be well worth doing.  I expect to explore some of the possibilities in future posts.

George the Sparrow

George the Sparrow

Ian Cruickshank of Victoria, BC sent me a remarkable recording of a very confused Song Sparrow, which seems to be incorporating the complete song of a Northern Waterthrush into its own singing.  Here’s the recording on Xeno-Canto:

When I first heard the recording, I thought the bird could be a juvenile using some imitations in its subsong — a decent possibility, given the late September date  — but the comments Ian sent me about the recording seem to rule that out:

I first heard this Song Sparrow giving this song phrase in April of this year; I didn’t manage to record it at the time and it was a stroke of luck that I came across it again, engaged in a territorial match with another male Song Sparrow, belting out this song in the exact same location, in September this year. Obviously it’s a resident bird.

Let’s compare spectrograms.  Here are a couple of excellent recordings of Northern Waterthrush songs.  Note that waterthrushes, like Song Sparrows, have numerous song dialects across their range:

Here’s a spectrogram of one strophe of Ian’s weird Song Sparrow.  Because the other birds on the recording make the spectrograms difficult to read, I’ve highlighted the Song Sparrow song in the background by coloring it red (a la The Sound Approach):

The last four (red) notes on the spectrogram are pure Song Sparrow, but boy, the rest of it sure looks like Northern Waterthrush, with the classic pattern of three contiguous series, including the rapidly downslurred whistles at the end. I think it’s highly likely that this Song Sparrow, during the “critical period” in which it was listening to the songs around it and piecing together its repertoire, mistook a Northern Waterthrush for a legitimate Song Sparrow tutor.  This phenomenon is rare in Song Sparrows, but not unprecedented.  Here’s what the Birds of North America account has to say about it:

Species displays innate preference for learning con-specific song and, like other Emberizidae, rarely mimics other species. Song Sparrows exposed to natural Song and Swamp Sparrow song in lab preferred strongly to learn conspecific song, but sometimes sang syllables of Swamp Sparrows (Marler and Peters 1987, 1988). Song Sparrows fostered by canaries (Carduelinae) did not mimic foster parents in one study (Mulligan 1966) but copied some elements in another (Kroodsma 1977). In contrast, Eberhardt and Baptista (1977) suggested Song Sparrows imitated Wrentit (Chamaea fasciata) syllables in wild, and Baptista (1988) recorded a Song Sparrow in San Francisco, CA, that produced White-crowned Sparrow song; another bird at Tioga Pass countersang White-crowned Sparrow song with neighbors of that species (M. Morton pers. comm. in Baptista and Catchpole 1989).

Song acquisition has been pretty well studied in Song Sparrows, and we know a lot about how they put together their repertoires.  When they are raised in the laboratory with adult tutors of their own species, they usually tend to copy whole songs verbatim. They also prefer to copy the songs that are shared by multiple tutors — in other words, they seem predisposed to learn the most popular local tunes.  However, some Song Sparrows act differently; one bird invented all its own songs, and several others tended to recombine elements from multiple tutor songs in creating their repertoire.  Ian’s sparrow most likely did the latter: it was trying to invent a new Song Sparrow song using elements of other Song Sparrow songs it had heard, but it misidentified one of its tutors and ended up with a weird, chimeric melody.

Now that it is an adult bird, it’s likely to sing this hybrid songtype for the rest of its life.  It’s hard to say whether that will disadvantage it.  Various studies have measured Song Sparrows’ responses to abnormal songs (including, among others, artificially constructed songs that arranged Swamp Sparrow syllables according to Song Sparrow syntax, and vice versa), and the findings tend to agree that imitations or corruptions of Song Sparrow songs elicit weaker responses than typical songs, but they still elicit responses.  Thus, this abnormal song likely won’t be as effective in driving away a rival male or attracting a female mate, but it may get the job done.  If, like most Song Sparrows, this individual has between 5 and 13 different songtypes in its repertoire, then the weird waterthrush-song might only be deployed between 8% and 20% of the time.  Assuming it’s the only abnormal songtype in the repertoire, it might not prove a huge disadvantage to the singer.

If this bird maintains a territory over multiple years, there’s a chance that juvenile Song Sparrows moving into nearby territories might even select it as a tutor, adding some or all of the Northern Waterthrush syllables to their own songs second-hand and potentially propelling them into the local Song Sparrow vernacular in the long term.  A similar process might explain why, for example, so many of the “Thick-billed” Fox Sparrows in the Sierra Nevada end their song phrases with what appears to be a straightforward imitation of the “kleer” call of Northern Flicker:

However, I think this unlikely to happen among Vancouver Island Song Sparrows.  Song Sparrows seem much less likely to imitate than Fox Sparrows, which means Songs probably have a stronger (though not ironclad) genetic mechanism to guide young birds to ignore the syllables of other species and incorporate only their own.  My prediction: this wrong-singing sparrow might not be a complete pariah, but in the long run he probably won’t prove a strong competitor for territories and mates either, and his borrowed syllables are unlikely to impress the next generation to follow in his footsteps.  He’s a slightly socially inappropriate, oddball schmo: the George Costanza of Song Sparrows.  We’ll call him George for short.

Do Violet-green Swallows Sing?

Do Violet-green Swallows Sing?

Violet-green Swallow, Nicasio, CA, 6/6/2008. Photo by Len Blumin (Creative Commons 2.0).

The Violet-green Swallow has long been one of my favorite birds.  Perhaps that’s because it is the only one of the seven widespread North American swallow species that I didn’t see regularly when I was growing up in eastern South Dakota.  Perhaps it’s the wonderful way that the colors blend on its iridescent back.  Perhaps it just reminds me of my beloved Rocky Mountains.

At any rate, most of the reasons I like Violet-green Swallows have always had to do with what they looked like, not what they sounded like.  I was fairly familiar with their calls, or so I thought, but I had never really listened to them closely.

Even so, I was astonished to find that the current scientific literature says that Violet-green Swallows don’t sing at all.  “I detected no recognizable, repeating series of syllables which could be interpreted as a song,” wrote Charles Brown in a 1983 paper in the Wilson Bulletin.  An undergraduate at the time of the research, Brown is now legendary as a swallow researcher, having spent two decades studying Cliff Swallow coloniality in Nebraska.  But after writing the BNA account on Violet-green Swallow in 1992, Brown apparently hasn’t published anything further on the species.  Nor does anyone else seem to have studied its vocalizations.  And so, according to the literature, Violet-green Swallows don’t sing.

Really?  No song?

It struck me as unlikely that Violet-green Swallows should lack a song, first because I thought I had heard complex song-like vocalizations from them, and second because the Tree Swallow, a close relative of the Violet-green, definitely sings.  It’s rare for one member of a species pair to sing and not the other.  (However, it’s not unprecedented.  Eastern Bluebirds sing a complex, melodious song; Western Bluebirds reportedly just string call notes together.)

So I went looking through my collection of Violet-green Swallow recordings.  Interestingly, most of my recordings matched Brown’s original findings: during the day, the species tends to give three kinds of calls, and even though some of them sound “complex” and I had taken them for song, it appears that none of them are stereotyped–meaning they’re not carbon copies of one another.  Instead, each rendition of a call varies slightly from all the others, like in the calls of House Finches or Rosy-Finches.  As Brown noted, if these calls were going to be strung together in some kind of song, one would expect to see stereotyped versions, possibly repeated in some kind of predictable pattern.

However, note the caveat in what I said above: during the day.  Some swallow species (including Tree Swallow and Purple Martin) sing dawnsongs that apparently differ significantly from anything they say when the sun’s up.  Brown’s work did not reveal the existence of a dawnsong in Violet-green Swallow, but it did note the passing mention of “predawn song-flights” in The Birder’s Handbook and “apparent chee-chee calls” starting as early as 2:20 a.m., according to Bent’s Life Histories.

And that led me to a most interesting recording in my collection.  At 4:20 in the morning on the 2nd of  July, 2008, when the sun was only barely lighting the eastern horizon, I was hiking with a friend up the Loch Vale trail in Rocky Mountain National Park, hoping to reach a Black Swift nesting site before dawn.  Just as we reached the Loch, I heard chirping from far overhead, and I immediately turned on my recorder, hoping to catch a Black Swift coming off its night roost.  But no: the chirping was too regular, too relentless, not high-pitched or clear enough.  The bird was completely invisible as it circled, singing endlessly against a backdrop of stars, but I was pretty sure it was a Violet-green Swallow.  Here is a snippet of the recording:

The bird went on like this for at least several minutes without pausing.  When I first looked at the spectrogram of the above snippet, it looked like the daytime calls had: a jumble of similar notes, each slightly different than the last, showing no apparent rhyme or reason.  But then I zoomed in and started looking very carefully, and sure enough, a pattern emerged:

The calls are stereotyped: all the “A” phrases are basically identical to one another, as are the “B” phrases, etcetera.  Furthermore, although the order is variable, there’s a strong tendency to stick to a couple of basic patterns.  In other words, this is classic song, crystallized and rule-bound, no matter how disordered it sounds to our slow human ears.

You may not want to give them any prizes for their melody, and you may not even hear them unless you rise long before the sun.  But make no mistake: Violet-green Swallows do have a song–one that seems to have gone almost unnoticed until now.

I Saw a Vampire!

I Saw a Vampire!

Although it’s a bit of a stretch from what this blog is normally about (and late for Halloween), this is too good not to post.  I just got the photos from my friend Carol Beardmore that she took this summer on the Western Field Ornithologists / Sonoran Joint Venture expedition to the Sierra de Alamos / Rio Cuchujaqui Protected Area near Alamos, Sonora, Mexico.  Carol and I were stationed for a week at El Cajon, a mid-elevation site in the tropical deciduous forest, where temperatures reached 115 degrees on several days when we were there.  To beat the heat in the afternoons, we headed to an amazing swimming hole about a mile from camp, where a large, deep, permanent pool of sparkling blue water stretched back into a flooded slot canyon.  Swimming back into the shade of the canyon, we would find the water getting quite cold in a hurry, and before long the canyon got so narrow that we could touch both walls at once — but not the bottom, because the water was far too deep.  A little farther on, the canyon widened out again and a beach appeared on one side below a shallow cave at the bottom of the cliff.  Getting out of the water, we saw the hallmark of the inhabitants of the cave:

Vampire bat guano, Sonora, Mexico, 7/1/2010. Photo by Carol Beardmore.

Vampire bat guano is unmistakable: it’s the red-black color of dried blood, and it’s viscous, sticky, and pungent.  The walls of the cave were liberally stained with this stuff, so we knew that vampires slept here.  And when we approached the cave (really only a shallow grotto, no more than a couple of meters deep), a few bats flew out.  It took multiple visits over a couple of days — and an inner-tube float trip with the camera equipment in a plastic garbage bag — but eventually, with Carol operating the camera and me holding the flashlight, we managed to get one of the cave’s residents to pose for a photo:

Common Vampire Bat (Desmodus rotundus), Sonora, Mexico, 7/1/2010. Photo by Carol Beardmore (click to enlarge).

Of the three species of vampire bat, this one is the most widespread and occurs closest to the United States.  It has been increasing in numbers and apparently expanding its range north in recent years, especially since humans have brought in large numbers of cattle, a favorite vampire food (beverage?).

For the record, this is the primary reason that I slept under a net while we were in Sonora.