The Master Set has been in the works for a number of years now. It’s intended to be a distillation of the finest audio clips in the collection of the Macaulay Library, the world’s greatest avian audio archive, comprising some 150,000 separate recordings, about 30,000 of which are from North American birds. Remember the gargantuan government warehouse in the final scene of Raiders of the Lost Ark? That’s how I’ve always pictured Macaulay – a trove of hidden treasures, a giant haystack made of needles. The process of creating the Master Set involved reviewing the majority of these recordings in an attempt to find examples for every vocalization type of each species, including individual and regional variations whenever possible.  “This is a celebration of what the recording community has given us,” says Macaulay audio production engineer Matt Young.

Extending across many years (and the tenures of multiple Macaulay Library employees), this review process has resulted in many inconsistencies and a few errors in the set. As a result, Macaulay is describing the Master Set as “a work in progress,” one which they intend to correct, update, and expand over time. To purchase the first complete draft of this work in progress, you’ll pay $49.99 – a special introductory price that will rise by $10 at some point soon, according to the website. The Master Set is neither perfect nor complete, but it is the most comprehensive audio collection of its kind to date.

The sound files are accompanied by photos that pop up when you play the audio on many devices – a nice touch. The recordings are accompanied by a short booklet with some introductory text and a list of audio and photo credits, but no additional information about the individual sound files. Some of this information, including the Macaulay catalog number of the source recording, can be found in the MP3 data tags for each file.  As you might expect, not all species are covered in the same detail. There are 39 cuts of Carolina Wren, but “only” four of California Thrasher. (I put “only” in quotes because most commercially available audio collections don’t even give you four cuts per species.) This gives you some idea of the scope of the work.

For those who don’t want to wade through 22 Tufted Titmouse cuts simply labeled “song,” there’s a smaller, more manageable version of the collection: the Essential Set, comprising 1376 cuts of 727 species, for $12.99 (that price to rise by $7 at the end of the introductory offer period).

If you find errors while exploring the collection, send an email to MLproductions@cornell.edu.  I presume you can use the same address to write with kudos for getting this massive undertaking off the ground — it’s been a long time coming!

"Olive-backed" Swainson's Thrush, Bismarck, North Dakota, 5/8/2009. Discovered in a storage shed, it was photographed before being released. Photo by Matt Reinbold (CC-by-sa).

The April 2013 issue of Colorado Birds recently hit my mailbox. It’s an excellent issue of a top-flight regional birding journal, and I’m particularly excited about the article about identifying the two distinctive subspecies groups of Swainson’s Thrushes: “Russet-backed” birds from the Pacific Northwest, and “Olive-backed” birds from elsewhere in the range.

(In case you’re wondering about the source of my enthusiasm, I should mention that I not only edit the journal, but also co-authored the article, with the eminent Steve Mlodinow and Tony Leukering.)

I grew up seeing hundreds of Swainson’s Thrushes in South Dakota each spring — all birds of the “Olive-backed” persuasion, though I had no idea of that at the time.  When I encountered my first “Russet-backed” Thrush in northern California in 2000, I was certain I was looking at a Veery.  The uniform, rich chestnut-brown upperparts ruled out all other thrush species, or so I thought at the time, and it wasn’t until many years later that I realized my mistake.

Over the past decade, the work of Kristen Ruegg and her colleagues has shown that the two forms of Swainson’s Thrush not only look different, but migrate on different schedules to markedly different wintering grounds. They hybridize in a contact zone in British Columbia, but that contact zone is quite narrow, prompting occasional rumors and rumblings of a potential future species split.

One of the proposed lines of evidence concerns differences in vocalizations.  And, with some cribbing  from the Colorado Birds article, that’s what I’ll be writing about today.

Differences in song?

A 2006 study by Ruegg et al. claimed that the songs of “Russet-backed” and “Olive-backed” Thrushes differ in statistically significant ways.  The spectrograms in the study seem to indicate some nice obvious differences:

Song spectrograms of "Russet-backed" (left) and "Olive-backed" (right) Swainson's Thrushes, from Ruegg et al. 2006, Fig. 3. Each top-bottom pair (e.g., 1.1 and 1.2) represents two songtypes of one individual.

When I first laid eyes on this figure, I thought this ID should be a cinch.  The Russet-backed songs in the figure are much longer on average (song 1.1 is just under 2.5 seconds long, while song 4.2 is under a second), and they finish with high, fluting, polyphonic phrases like those in a Veery song, while the Olive-backed songs stay relatively low and simple.

But when I looked at a scattering of songs from across the country, it quickly became clear that these differences did NOT hold up across the species’ entire range.

Obviously the Olive-backed songs from elsewhere on the continent seem just as likely to go on longer and end in high, fluting flourishes.  I’ve listened to a large number of songs of both species, and I can’t find any way to distinguish them by ear.

To be fair, Ruegg et al. weren’t just looking for differences obvious to the eye and ear — they were performing quantitative analyses based on spectrographic measurements.  Even so, they recorded songs from just two coastal populations and two inland populations (plus a fifth population in the hybrid zone).  Their results may be due to the differences between the local dialects of these few regions, rather than any larger or more consistent difference between the subspecies groups.

If anybody knows of a way to identify the songs of these two groups by ear, I’d love to hear about it.

Differences in calls

I am unaware of any consistent differences between the “wee?” flight calls of the two subspecies groups, which is a mellow rising whistle rather like the call of the Spring Peeper frog. The “quit”-type contact call, however, does seem to differ slightly on average.  Olive-backed tends to give a quick, sharp “quit” that may recall the sound of a dripping faucet or the “whit” of an Empidonax flycatcher. In Russet-backed, this call is often slightly longer and more musical, an obviously upslurred whistle that might be transliterated as “wee” or “pwee,” like a shorter, sharper version of the flight call.

These examples are illustrative:

Even this slight difference, however, cannot always be trusted.  One recording I found documents a “Russet-backed” from northern coastal California who gives, in between his songs,

• a series of “typical” contact calls of Olive-backed Thrush (1:25 – 1:50);
• a series of “typical” contact calls of Russet-backed Thrush (2:20 – 2:35); and
• a wide variety of “flight calls” (0:37 – 0:43, 2:49, and 3:43 – 3:48).

At least some individual Olive-backed Thrushes give equally variable calls, so caution is in order.

Perhaps the most distinctive vocalization is the alarm call, which is a two-part sound ending in a low, loud, semi-musical purr or chatter. Olive-backed tends to introduce the chatter with a sound like the contact call, “quit-BRRR,” while Russet-backed tends to begin with a much longer and more musical note that is actually more like the flight call: “weee-BRRR.”

On the whole, the differences in voice between these two subspecies groups are pretty subtle, and almost never diagnostic, due to extensive variation within each group. I recommend using voice as a supporting character in the field, to be used in conjunction with visual field marks.

Now, new information has come to light that calls my earlier conclusions into question.  First of all, Empid guru Arch McCallum told me he wasn’t convinced that the “whit-beert” was equivalent to “dew-hic” and “peer-pewit”, due to the different spectrographic shape of the “whit” and the fact that it was never repeated.  Instead, he pointed out the similarity of “whit-beert” to certain calls of Willow and Alder Flycatchers.

He may have hit the nail on the head.  I recently discovered a second recording of the mysterious “whit-beert,” from western Maine, labeled as an Alder Flycatcher.

“Whit-beert” call, labeled as Yellow-bellied Flycatcher. Woodhull Lake, New York, 5/30/1998. ML catalog #106901	“Whit-beert” call, labeled as Alder Flycatcher. Upton, Maine, 6/2/1962. ML catalog #7546

The new recording was taken in 1962, before “Traill’s” Flycatcher was split into Willow and Alder. The recordist was Dr. Robert C. Stein, who had already realized that “Traill’s” sorted into two vocal groups, and was in the process of collecting data for his classic 1963 publication that provided the first strong evidence for the eventual species split.  The notes on the recording indicate that the recording was made during a “hostile response to playback,” but it doesn’t say what sound was used for playback, nor does it explain how the recording was identified as an Alder Flycatcher.

So: which species is it that gives the “whit-beert” call?  Personally, I agree with Arch that Yellow-bellied is probably the least likely culprit.  Alder is the leading contender at the moment, followed by Willow.  But to answer the question definitively, I think we’ll need some more recordings.

If you’re interested in helping to solve the mystery, you might try some playback experiments in northeastern North America this spring.  Does playing “whit-beerts” to an Empid elicit more “whit-beerts”?  What species says “whit-beert,” and what is the behavioral context of the sound?

I don’t think anybody knows the answer to the questions I just asked.  I don’t think anybody has ever known.  But they could be easily answered by anyone with an audio recorder and a couple of spare hours in New England this summer.

This is what I love most about bird sounds — the tantalizingly short distance to the frontier of knowledge.

Lesser Yellowlegs, Jamaica Bay National Wildlife Refuge, NY, 9/30/2007. Photo by Wolfgang Wander (GFDL).

Can you tell Greater and Lesser Yellowlegs apart by voice?

Can anybody?

A recent thread on the Xeno-Canto forum started me asking questions about how to identify yellowlegs by voice.  Conventional wisdom says the two species can often be separated by their calls, at least with some experience.  But as I was comparing the many online recordings, I came to an uncomfortable conclusion.  Either an awfully high percentage of the recordings were misidentified, or my own identification criteria were wrong.

Today, I invite you to either share my confusion, or help me sort it out.

History of the problem

In the old days (i.e., the 20th century), identifying yellowlegs by voice was simpler.  Back then, according to the field guides, all you had to do was count the notes:

Although all three guides suggest differences in pitch and/or tone quality, most people took away a simple rule: three or more notes is a mark for Greater Yellowlegs, less than three notes is a mark for Lesser. Everybody understood that this was a rule of thumb and not a universal law, but still, it was a very popular mental shortcut.

Then Sibley came along, and started complicating matters:

Here the numbers are tempered with the words “typically” and “usually,” and pitch and quality get a little more attention.  Most importantly, the yellowlegs are now presented as birds with large vocal repertoires — you have to be sure you’re listening to the “flight call” before you start counting.

So… how do you tell what you’re listening to?

The “classic” calls

Here’s a pair of recordings from the same location on the same date, by the same recordist, that fit all the classic descriptions of the two species’ “flight calls”:

Besides the number of notes, check out the huge difference in the shape of each note on the spectrogram.  Lesser gives a pretty simple downslurred whistle, while Greater has a much more complex pattern.  If you look carefully, you can see that each note of the Greater’s call contains an upward voice break.  This means each note is 2-parted, the second part suddenly jumping to a much higher pitch.  This happens so fast in each note that we still tend to hear the two parts as one, but the overall impression is very different from Lesser, more of a “klee-klee-klee” than a “pew-pew-pew.”  The higher second part of each note is what Sibley refers to as the “strident overtones” of Greater.

The alarm series

Both yellowlegs give loud, incessant calls in series when they are upset, year-round. In Greater these notes are noticeably rougher than the typical “flight call,” due to a brief burry or grating sound in the middle of each note. In Lesser, the notes of the alarm call strongly resemble the notes of the “flight call,” but marginally higher. They are perfectly clear, without any trace of roughness:

Where it gets complicated

Here’s the bad news. Based on a set of positively-identified recordings (mostly those that also contain examples of each species’ diagnostic song, in the Macaulay Library collection), it’s pretty clear that

• the notes in Greater Yellowlegs “flight calls” don’t always break.
• the notes in Lesser Yellowlegs “flight calls” sometimes do.
• in both species, the number of notes in a “flight call” depends on the agitation level of the bird.

I don’t have any reason to believe these four recordings are misidentified. In fact, I was present for the recording of Andrew’s Dove Creek bird, and as I recall, it was seen well and its identification was uncontroversial.  All these examples sound like they’ve got Lesser’s tone quality and Greater’s note pattern, and without visual clues, I’m not sure they’re identifiable, even on the spectrogram.  Here’s another pair of examples:

Greater Yellowlegs, from Macaulay Library 27042 (click to listen)	Lesser Yellowlegs, from Macaulay Library 26256 (click to listen; this vocalization is from the eighth minute)

There are some differences in pitch and inflection here, especially towards the end, but the opening notes are nearly identical, and match my expectations for the “classic” Greater Yellowlegs tone quality.

A gallery of confusing yellowlegs

I’m relatively confident about the ID of all the recordings I’ve posted so far, but I’m not at all sure about the ones that follow. The only thing I’m sure of is that an awful lot of yellowlegs recordings thwart my original expectations of both species.  Perhaps these next few recordings are all wrong, and yellowlegs identification isn’t as hard as I’m making it out to be! Perhaps they’re all or mostly right, and yellowlegs identification is very tricky indeed.

Any comments on the identification of these recordings, and yellowlegs vocal ID in general, would be greatly appreciated.

Bay-breasted Warbler, Rondeau Provincial Park, Ontario, 5/5/2007. Photo by Mdf (CC 3.0).

Not too long ago, it was commonplace for birders to make casual references to THE call or THE song of a bird species, as though every bird had only two modes of communication.  Now we know better.

These days, I’m more likely to hear casual references to THE flight call and THE alarm call.  After all, when we notice that, for example, many species of warblers and sparrows give similar high “seets” on the wing in migration and sharp “chips” in alarm when perched, it makes sense to group those calls into categories and name them after similarities in their form or function.

But there’s an odd, insidious trick of psychology at work in all of us.  If a sound doesn’t match our expectations — if it doesn’t fit one of our pre-existing mental categories — we’re much less likely to hear it.  If we do hear it, we’re likely to dismiss it as a quirk of the individual, a bird doing “something weird.”  Or else we’ll just cram it into a category we recognize: “must be the flight call, I guess.”

Something like this happened to me as I researched the “call notes” of various warblers.  I regularly ran into sounds that didn’t fit my mental categories, and after a brief period of confusion, I tended to dismiss them as atypical versions of the normal call.  Beware that word “atypical” — it strongly suggests that what you’re hearing is best ignored, that it’s an outlier at risk of messing up your dataset.

A few months ago, an online chat with my co-blogger Andrew Spencer started unraveling some of my preconceptions:

Andrew: are you going to cover alarm calls in warblers?

me: What kinds of alarm calls?

Andrew: those high pitched tink calls that many warblers give

like http://www.xeno-canto.org/13861

I think every warbler just about does tink calls when alarmed

I heard Connecticut Warbler give one

and I’ve heard it from a number of other species

but I don’t know how many recordings there are

me: Just higher-pitched versions of the normal call? Or something distinct?

Andrew: the one I linked of the Orange-crowned Warbler is distinct I think

all the ones I’ve heard sound damn near the same

me: I remember when I was researching Louisiana Waterthrush calls I came upon a recording in the Macaulay collection of a bird giving calls much higher than all the other recordings — more like your tinks. I just figured it was a crazy variation on the normal call.

Andrew: I have calls like that from Black-throated Blue Warbler as well: http://www.xeno-canto.org/30783

me: And I have a recording of an agitated Kentucky Warbler switching back and forth from a chip-like to a tink-like note.

Andrew: and I got a few from Golden-winged Warbler this past trip

me: These “tink” notes appear to be poorly described in field guides and the literature.

Andrew: doesn’t surprise me

me: This calls for a blog post.

Andrew: haha I was about to say the same thing

As soon as I started digging, I realized that these “tink” calls hadn’t gone unnoticed by everyone.  Paul Driver featured them on his blog back in 2009, under the name “high chip calls”:

A number of warblers (perhaps most?) have high chip alarm calls different to the typical chip call, sounding more like titmice or Golden-crowned Kinglet. They seem to be heard most often on breeding grounds and are often given by birds that are highly agitated; in this way they seem analogous to the high chip call of the Song Sparrow.

Funny he should mention Song Sparrow.  Recordings show that a great many species of sparrow give a “tink” note that sounds much like the “tinks” of warblers, and is given in similar situations of high alarm.  (This warbler-sparrow similarity is no coincidence – but that’s a subject for another day.)

These “tink” notes seem worth of their own category, separate from the “call” and the “flight call.”  They tend to indicate a higher level of alarm than “typical” calls, but not as high as the buzzes and shrieks that these birds give when they’re really upset (e.g., when a predator is attacking a nest, or when the bird is caught in a mist net).  Sometimes one hears calls intermediate between the high “tinks” and the lower “chips,” especially from sparrows.

“Tinks” can raise some interesting questions.  In some species, like Cape May Warbler, the “typical” call sounds much like a tink, and there may be little distinction — or is it perhaps that the tink is just more frequently given, so that we think of it as typical?  At the very least, if you happen to be in the habit of identifying Cape May Warblers by call alone, the existence of similar sounds in all these other species should give you pause.

An online catalog of “tink” recordings

Not every warbler and sparrow has a “tink,” apparently. For example, I haven’t been able to find any convincing examples from the genus Spizella (except for American Tree Sparrow, which many researchers agree is misplaced in Spizella and overdue for a move to its own genus).

I’ve attempted to assemble a collection of “tink” recordings that are available online, in order to document their existence in as many species as possible.  No doubt there are other “tinking” species not currently represented below.

• Ovenbird (maybe these are tinks)
• Worm-eating Warbler (also on Paul Driver’s site)
• Louisiana Waterthrush (mostly chip notes with a few tinks thrown in)
• Blue-winged Warbler
• Golden-winged Warbler
• Lucy’s Warbler
• Orange-crowned Warbler
• Kentucky Warbler (on Paul Driver’s site)
• Common Yellowthroat (on Paul Driver’s site)
• Yellow Warbler (on Paul Driver’s site)
• Black-throated Blue Warbler
• Cape May Warbler (this species may not give lower-pitched calls)
• Blackburnian Warbler (also on Paul Driver’s site)
• “Audubon’s” Yellow-rumped Warbler
• Pine Warbler (on Paul Driver’s site)
• Black-throated Gray Warbler
• Golden-cheeked Warbler
• Black-throated Green Warbler
• Bay-breasted Warbler
• Canada Warbler (maybe these are tinks)
• Painted Redstart
• Spotted Towhee
• Eastern Towhee
• Green-tailed Towhee
• Song Sparrow (on Paul Driver’s site)
• Swamp Sparrow
• Vesper Sparrow
• Fox Sparrow (Slate-colored)
• White-throated Sparrow
• White-crowned Sparrow 
• American Tree Sparrow
• Savannah Sparrow
• Le Conte’s Sparrow

If you can add to the list above, or have other useful observations of “tink”-like notes, let me know!

Lapland Longspur, Kodiak National Wildlife Refuge, Alaska. USFWS image in the pubic domain.

In wintertime, huge numbers of Lapland Longspurs come down to the northern United States from their Arctic breeding grounds — sometimes gathering in enormous single-species flocks, but often mixing with other longspurs, Horned Larks, and Snow Buntings. At a distance, their cryptic winter plumage pattern can make them hard to pick out from these other birds. That’s why many people “look” for longspurs with their ears.

Away from the Great Plains, Lapland is usually the only expected longspur species, and one can usually detect and identify it by its characteristic rattle call, which generally resembles the rattles of other longspurs.  Although the number of notes is highly variable, this call is pretty much the same across the Lapland Longspur’s entire (worldwide) range:

Lapland Longspur rattle, Weld County, CO, 11/19/2007.

But this isn’t all that longspurs say.  They give a variety of other calls, some of which are more immediately distinctive than the rattle.  Here in Colorado, I recorded a longspur giving “chewlup” and “terlee” calls in addition to rattles:

As I started listening to recordings of Lapland Longspurs from other parts of North America, I started to hear other types of calls.  Individual birds give at least 5 or 6 different whistled calls, especially on the breeding grounds, and individual repertoires seem to differ, especially from one geographic region to the next.  In trying to catalog Lapland Longspur calls, I ended up making a map of variation.  First, I found recordings from seven distinct locations where this species breeds in the North American Arctic:

1. Seward Peninsula, western Alaska (LNS 49598 and LNS 141100)
2. Denali National Park, central Alaska (LNS 50024)
3. Colville River Delta, northern Alaska (LNS 131257)
4. Babbage River, northern Yukon (LNS 61441)
5. Bathurst Island, Nunavut (LNS 137341)
6. Devon Island, Nunavut (LNS 61444)
7. Baffin Island, Nunavut (LNS 61426)

Locations of the Lapland Longspur recordings

Then I went through the recordings and made a table showing the geographic similarities and differences in calls:

In the table above, all the spectrograms in a given row represent different calls from the same individual bird (except for the first row — the “chewlup” and “ter-lee” are from a different bird than the other three).  Here are the takeaway lessons:

1. Some calls are clearly the same across wide portions of the range. For example, the call that sounds like “ti-turtle” is clearly the same in western Alaska, northern Alaska, and the northern Yukon.  But it does not appear in recordings from other regions — at least not in a recognizable form.
2. This graph isn’t wide enough to show all the types of calls. In particular, the recordings from northern Alaska (#3) and Bathhurst Island (#6) contain multiple call types that I didn’t include because they didn’t seem to fit into any of the existing columns.
3. Few, if any, of the whistled calls are truly universal. The only whistled call type that comes close to appearing in all regions is the “few,” and even then I’m not certain that the “fews” I illustrate above for Devon and Baffin Islands are really equivalent to the “fews” from elsewhere.  It’s possible that more recordings would change this, but note that the extensive recordings from northern Alaska and Bathurst Island show an almost complete lack of shared calls — and that includes the ones I left out for lack of space.
4. Most, if not all, of the calls can be heard in winter as well as summer.  Though rather few recordings are available from wintering birds, it appears that the calls of longspurs wintering in western North America and on the Great Plains tend to resemble the those from Alaska (locations 1-3), while the calls of longspurs wintering in eastern North America tend to resemble those from Arctic Canada (locations 5-7).

What can we conclude from this information?  Well, for starters, it’s pretty clear that these calls are learned, not innate.  That would explain why birds from a given region tend to share call types, while birds from far away tend to sound quite different.  It fits a pattern of regional dialects that are typical of learned vocalizations.  Each bird learns not just one call typical of its region, but an entire set of regional calls — much like the Red-winged Blackbirds I discussed a while back.  In longspurs, like in Red-winged Blackbirds, dissimilar calls likely fulfill similar functions in different regions of the Arctic.  For example, longspurs expressing agitation at a human near a nest tend to cycle through 3-4 different call types — but the birds on Bathurst cycle through a totally different set of calls than the birds on the north slope of Alaska.

Obviously, people attempting to identify Lapland Longspurs by their calls in winter have their work cut out for them.  My general impression is that longspurs wintering in Colorado sound the same year after year, consistently giving “few,” “chewlup” and “ter-lee” calls like the northern Alaska birds above.  But if I went to, say, Iowa, the longspurs would be likely to sound rather different.

Much remains to be learned about the Lapland Longspur’s complex communication system.  I’m looking forward to knowing more.

Visualizing variety

Ornithologists use the term variety to describe the pattern of delivery of a bird song over time.  Some individual birds sing only a single song over and over (no variety).  A bird that can sing multiple songs might choose to sing one for a while before switching (eventual variety), or it might switch constantly (immediate variety).

In the field, it can take many minutes of listening to determine a bird’s pattern.  Animated GIFs of spectrograms can condense all this listening into just a few seconds of looping video:

Olive-sided Flycatcher	Carolina Chickadee	Hermit Thrush
No variety (all songs identical)	Eventual variety (songtypes repeated several times before switching)	Immediate variety (consecutive songtypes always different)

In each GIF above, the spectrogram of a single bird song appears on the screen for one fifth of a second. It is then replaced by the spectrogram of the next song by the same bird.  After a couple dozen songs, the animation loops back to the beginning.

Seeing song similarities (and differences)

As all naturalists know, the pieces of nature rarely fit into neat categories — and so it’s no surprise that the three categories of variety above are inadequate for describing the more complex patterns of variation found in many bird songs.  A GIF, though, might be up to the task.

Take these 18 songs from a Vesper Sparrow:

Vesper Sparrow, Montezuma County, CO, 5/11/2008.

Note that the level of variety at the beginning of each song is completely different from the level of variety at the end.  Each song starts with the same 3 (rarely 4) downslurred whistles, followed by the same rapid series of vertical notes (the number varying from 5 to 7).  After that, variety increases dramatically.  The middle section irregularly alternates between two different patterns, and the ending switches even more frequently, between at least three different motifs.

This type of cascading variety is typical of Vesper Sparrows.  The opening notes tend to vary little within an individual, but by the end of the song, variation is tremendous.  Perhaps this allows the sparrows to “have it both ways” — that is, to simultaneously send two conflicting messages to the listener.  The stereotyped opening satisfies their need to identify themselves unambiguously to a potential mate or rival (“I am a typical Vesper Sparrow!”), while the jazzy ending allows them to show off their improvisational virtuosity (“I’m not your ordinary Vesper Sparrow!”).

I’m fascinated by the potential for GIF visualization… perhaps you’ll see more animated spectrograms on this blog in the future.

Rusty Blackbird, Cleveland, Ohio, 10/9/2011. Photo by Laura Gooch (CC-by-nc-sa).

The voice of Rusty Blackbird isn’t particularly well known.  Perhaps because the species is uncommon and its breeding grounds rather remote, it’s never been the subject of a formal bioacoustic study.  But Rusty Blackbirds sing quite a bit — both males and females sing, in fact.  They sing on the breeding grounds, on migration in fall and spring, and pretty much all winter long.  The winter flocks I encountered in the swamps of Arkansas could be heard for half a mile.

Several authors have described Rusty Blackbirds as having two types of songs — one more creaking, one more gurgling — and this would make sense given that the closely related Brewer’s Blackbird has also been reported to have two different songs of more or less the same types.  As I went through online recordings of Rusty Blackbirds, however, I came to the conclusion that I was hearing three different types of songs from the species, not two.  Or is that two types of song and a very song-like call?

Three things Rusty Blackbirds say

The gurgle-creak seems like a good candidate for a typical “song.”  It starts with a complex jumble of rapid notes and ends with a high monotone whistle with a metallic quality, like the creak of a rusty hinge:

Another song-like sound is the gurgle.  It’s like the first part of the gurgle-creak, but at least twice as long.

Then there’s the creak.  It’s like the gurgle-creak, but starts with only 1-2 quick notes, usually including a noisy “chuck” note much like the species’ contact call.  On the spectrogram below, two creaks follow a gurgle-creak:

It’s fairly clear that these three song-like sounds are associated with different behaviors.  The gurgle-creaks seem typical of songs — they’re practiced throughout the winter, getting gradually more stereotyped; they’re heard from spring migrants, who seem to have mastered them by the time they head north; and they’re delivered in long bouts on the breeding grounds, where they apparently function in territorial defense and mate attraction.

The gurgles, unlike the gurgle-creaks, are rarely heard on the breeding grounds.  Instead, they mostly seem to be given by birds immediately prior to or during spring migration.  The only recording I know of from the breeding grounds is from early in the season — 4 June in northwest Alaska.  Perhaps they function primarily in pair establishment, and are used for just a few days after arrival on breeding territories.

Then there are the creaks, which seem to occupy an odd territory in between traditional “calls” and “songs.”  Like traditional “songs,” they appear to be learned rather than innate — birds spend the winter practicing them, and the finished springtime performances vary greatly from one male to the next.  Furthermore, they’re often deployed along with the gurgle-creaks by territorial birds on the breeding grounds, in response to playback of a rival’s song.  That pretty well matches the classic definition of “song.”

But when else do we hear creaks?  When Rusty Blackbirds mob Common Ravens.  And in response to a tape of Northern Goshawk.  And when a human threatens a nest.  Most birds do not give songs, or anything like a song, in these situations.  Instead we’d expect to hear alarm calls — typically harsh, noisy, simple sounds that carry far and don’t change from one bird to the next.  Rusty Blackbirds have such calls — sharp chucks, long churrs, rattling chatters, and the like.  As one would expect, they readily chuck and churr and chatter at ravens and goshawks and people, but they also give song-like creaks.  What’s up with that?

Longtime readers of this blog will know that I don’t put much stock in the traditional “song/call” distinction, and here’s one more example of why not — a peculiar sound that straddles the traditional boundary in both form and function.  Rusty Blackbird has attracted a fair amount of research attention in recent years due to evidence that its populations may be in steep decline.  But I’m unaware that anyone has attempted to investigate its vocal communication.  Here’s hoping someone will take it on — I think this bird deserves it.

Current taxonomy draws species boundaries despite the hybridization, in part due to differences in vocalizations.  For example, Red-bellied and Hoffmann’s Woodpeckers both give a long, churring rattle call that Golden-fronted lacks.  Vocal differences probably help the woodpeckers keep each other straight in areas where ranges meet, making them more likely than not to choose a mate of their own taxon, keeping the level of gene flow low and maintaining the boundary between the species in the long term.

As I was going through the Golden-fronted Woodpecker recordings in my collection, some of which are from Texas and some of which are from Mexico, I realized that at least one call in the Golden-fronted repertoire varies geographically.  Birds in southern Texas and northern Mexico give a rough, single-syllabled “gaf” call rather like that of the Red-bellied Woodpecker:

But birds from Veracruz south give a distinctive, two-syllabled “CHUCK-a”:

These differences hold up in every recording I’ve managed to find of this call type.  The single-note “gaf” can be heard at least as far north as the Texas Panhandle and as far south as San Luis Potosi.  The double-noted “CHUCK-a” has been recorded not only in Veracruz but also in Guatemala and El Salvador.

It turns out that these differences almost exactly mirror the geographic boundaries between two major subspecies groups.  The “aurifrons group” is the “classic” Golden-fronted Woodpecker of Texas and northern Mexico.  The double-noted birds appear to correspond to the “santacruzi” group, which is found from Veracruz all the way south to Honduras (including the “polygrammus” group of southern Pacific-slope Mexico).  It’s worth noting that the subspecies dubius from the Yucatan Peninsula apparently gives a slightly different version of the two-syllabled call, more of a “chuck-trrr” [1 2], sometimes shortened to a single-noted “chuck” rather like the call of aurifrons [3].

A 2009 genetic analysis by Erick García-Trejo and colleagues found that the “aurifrons” group and the “santacruzi” group were genetically distinct, with northern aurifrons more closely related to the Red-bellied Woodpecker, Melanerpes carolinus, than to southern birds.  As announced in Birding magazine and elsewhere, this set the stage for a formal split of the species by the AOU into Golden-fronted Woodpecker (M. aurifrons) and Velasquez’s Woodpecker (M. santacruzi).  However, nobody has apparently submitted a formal proposal to the North American Checklist Committee yet — so I took it upon myself to do that.  Sometime later this year, we’ll find out what they say!

"Northwestern" Crow. This crow from La Push, Washington (on the outer Olympic coast) is supposedly a Northwestern Crow according to various sources. But is it really? Photo by Andrew Spencer.

There are few species in North America as ambiguous as Northwestern Crow (Corvus caurinus).  Even in a group of birds that are exceedingly similar the differences between American Crow (Corvus brachyrhynchos) and Northwestern Crow are minuscule at best.  The only “surefire” way to tell them apart is by range; however a number of sources also cite vocal differences as a distinguishing characteristic (e.g., the Sibley Guide, some versions of the National Geographic Guide, Birds of North America).

American Crow is a very familiar bird to most people in the US – it occurs throughout the lower 48, except for certain areas near the Mexican border.  Northwestern Crow is typically described as being found from northwestern-most Washington state along the Pacific coast to Alaska, as far west as Kodiak Island and the adjacent mainland.  Outside of a narrow overlap zone in Washington state and southern British Colombia you can typically call the species by range alone.

But the situation is more complex than just being able to tick a bird for your list based on where you are.  The American Crows found along the Pacific coast (ssp. hesperis) are at the small end of the spectrum for the species, sometimes look slightly “shinier”, and have subtly but noticeably different voices than American Crows further east (as an aside, the official designation of what constitutes hesperis is murky at best, and birds from as far away as Utah or Colorado could be this subspecies depending on some readings of the literature).

For a long time I’ve had a theory that the real “Northwestern” Crow was actually a “Pacific” Crow, since the American Crows along the northern California and Oregon coasts  sounded and looked more like “Northwestern” Crows in Washington and further north, and different from the American Crows further east.  If this was true, I theorized, then the crows west of the Cascades in the more humid areas of Washington, Oregon, and California, were a different species than those east of the Cascades in the drier rainshadow of the mountains.

So on a few recent trips I’ve taken to Washington I’ve made an effort to record crows wherever I found them, including going out to what were supposed to be the most “pure” Northwestern Crows left in the state, on the San Juan Islands and along the outer coast of the Olympic Peninsula.  I also recorded some crows to the east of the Cascades, from the Okanogan Valley and further east.  What I was hoping to find was either that the supposedly pure “Northwestern” Crows had a notably different voice than the taxonomically ambiguous ones around Seattle and along the outer coast in southern Washington, OR that all the crows west of the Cascades had similar voices that were different than crows to the east.

Unfortunately, as often happens with research of this type, what I found was different than either ideal situation.  I’ve been able to hear what I feel are distinguishable differences between birds in Colorado and the ones in western Washington.  But when I compared birds I recorded to the east of the Cascades in Washington I begin to hear what sounds like the start of a cline to the birds in Colorado.  Recordings from further afield, in Oregon, California, and British Colombia seem to continue this cline.  On the other side of the spectrum, though, when comparing the crows at the northwestern end of the range of American Crow to Northwestern Crows from Washington and further north into Alaska I don’t hear any vocal differences.  This would seem to show that the end of the cline in crow vocalizations is not where the species boundaries are currently drawn.

When comparing the sounds from various crows it’s important to have an idea of what vocalizations each species makes.  Anyone who has ever listened to a crow knows they have a remarkably large repertoire, everything from caws to screeches to rattles to song-like warbles.  The most common calls, though, are variations on the “caw” call.  These can typically be lumped into two groups, long and short, but there is a nearly complete continuum between those.  Often within any individual bird the long calls sound lower and more nasal, so care has to be taken when comparing different individual crows and making assumptions about differences in pitch.  Here are two cuts from the same individual bird in Washington showing a long and a short “caw”.

Ok so now to start the comparisons – listen here to what I consider a “typical” American Crow call from most of the continent, first longer calls then shorter:

And here is what I consider a typical American Crow from the Seattle area:

And a few from further east in Washington:

And here are some American Crows from further south along the Pacific coast:

Also listen to ML#13108 from the Willamette Valley of Oregon, ML#118854 from Tulare County, California, and ML#58191 from Hedley, British Colombia, all inland locations.

The ones around Seattle to my ears sound a bit more nasal, and on average slightly lower pitched than the birds from east of Washington state.  It isn’t a huge difference, and there is some overlap, but whenever I arrive in Seattle after spending time in Colorado or further east I hear the difference.  The birds from Okanogan County to the east of the Cascades sound similar to the Seattle birds to my ear, but are starting to having a hint of an eastern “accent”, as are the birds from coastal California. The recording from Oregon, though, sounds more eastern to my ears, and the ones from Tulare County, California, and British Colombia sound eastern to me.

Now listen to the calls of putative “Northwestern” Crows, also from Washington:

Not very different, right?  And here are links to some Northwestern Crows in the Macaulay Library from further north, far enough away from American Crow to be pure, supposedly: ML#132185, ML#136466, ML#58198.

Also not very different from the American Crows in Washington!

There are still some potential wrinkles to work out.  Without knowing the sex of the crow recorded it’s hard to determine what potential differences in male and female voices there are and how that affects the variation.  And it’s hard to get a real handle on the complete variation in each population without more recordings.

Now I can’t comment on the distinctiveness of Northwestern Crow based on genetics, structure or plumage, or physiological differences.  But I would contend that despite what many field guides, and BNA say, there are no compelling vocal differences between that species and American Crows in the Pacific Northwest.  With this lack of a vocal dividing line there seems to be little support for drawing the species boundaries in their current location.  Add to that the apparent evidence of a cline between the crows in the Pacific Northwest and those east of the mountains to the birds even further east, and the evidence for having two species at all starts to look weak indeed.