Monday, October 30, 2006

Writing the Song



This past spring on April 20th, Charles Naeseth found a singing Tennessee Warbler along the stream corridor of Pheasant Branch Conservancy. Walking the path toward him that morning, I heard the song and saw Charles grinning as he pointed up at the bird. We figured it was pretty early in the spring for a Tennessee. We watched and listened to the warbler for a few minutes; eventually Charles decided to keep walking down the path to look at other birds. I stayed behind to try to get a picture of the warbler, but was unable to. Still, there was no doubt in our minds regarding the identification of the bird.

I reported the sighting to the Wisconsin Birding Network and Bob Domagalski replied that April 20th would tie for record-early (Wisconsin) going back to 1980. Supposing Charles would submit his sighting, I put together supporting documentation of the observation into WSO's on-line form. For whatever personal reasons he might have, I later found out Charles wasn't interested in submitting his discovery to WSO (he's sort of a quirky character, but a great birder with 40 years experience), leaving my brief report to stand on its own.

Recently I received a letter in the mail from Jim Frank, WSO Records Committee chair, indicating that they were unable to accept the Tennessee Warbler record citing Orange-crowned Warbler has a similar sounding "trill song" and was more expected for that time of year. Jim also noted, as I stated in my report, that my view of the bird wasn't unobstructed or in great light and subsequently failed to note pertinent field marks. Jim concluded, with appropriate skepticism, that I might have gotten it wrong.

My first reaction was "no way!" and that the song totally gave the bird's identification away. This quickly changed to apathy, but the more I thought about it I strongly felt the record should be included in the context of other unusual sightings at Pheasant Branch last April. There was a record-early Golden-winged Warbler (April 8th), a Hooded Warbler (April 15th) and this record-early Tennessee ought to be included in WSO's "citizen science" dataset. Is it an important observation even if an anomaly, else why keep track of these things?

I emailed Jim and was adamant about my field identification by song. He thoughtfully explained that written song descriptions are very problematic for records committees due to the uncertainty of how they audibly translate - which syllables to accent, inflect, etc. I admitted that I hadn't really thought about it in that context before.

I think a Tennessee Warbler has a very distinct song, but how should I write it out? Making the identification is relatively effortless upon hearing one – I either think to myself or say aloud, "Tennessee Warbler over there!" Certainly, I'm not infallible when it comes to bird song identification, but there are those species you just know the moment you hear them – it's a diagnostic Tennessee Warbler song, Blue-winged Warbler or Louisiana Waterthrush. Such songs stand out as being unique in a way that rules everything else out in an instant. In my documentation, I described the Tennessee's song as "see-bit, see-bit, see-bit, see-bit, seet seet seet seet seet" with slight acceleration. Jim's point is well taken especially when comparing various field guides.

Tennessee Warbler song:

Sibley: tip tip tip tip teepit teepit teepit teepit ti ti ti ti ti ti ti ti

Dunne: chitsi chitsi chitsi see see see chchchchchchch

Cornell's BNA on-line: ticka ticka ticka, swit swit, sit-sit-sit-sit-sit-sit

Almost makes you wonder if they're even talking about the same species! I can't say for certain that if someone independently sent me above text asking for song ID that I would have absolutely responded with Tennessee Warbler in every case. To further complicate the issue, Dunn/Garrett cite geographical variances in song in Peterson's Warblers field guide. Identifying a bird by song might be enough for a year list or pointing a species out during a field trip (because everyone can hear it), but trying to express it in words is problematic, especially when submitting it on a record.


Tennessee Warbler


Orange-crowned Warbler

Believe it or not, I eventually won a written appeal to WSO's records committee and got the record accepted by putting together a short presentation with sonagram of Tennessee and Orange-crowned and explaining structural differences of their songs and how unlikely it was that I could have been mistaken - detail I should have included in the original submission. Naturally, the observation was now a half a year ago and I've listened to many Tennessee Warblers since then. How can they be certain I've described a bird (by song) that was actually there?

I'm guessing they wanted to make sure I knew the difference between the two songs. In other words, I think they sincerely believed Charles and I heard a Tennessee Warbler last April, but believing isn't enough - how can you prove it? Records committees look at hundreds of reports in a given year, but the question is should citizen scientists be held to such standards and scrutiny? If veritable sightings are being dismissed for lack of adequate documentation, should we still keep them on a gradation list of some sort, so that shifts and population trends might have an element of fuzziness, but nevertheless indicate something?

Though I was extremely confident about the Tennessee Warbler song, from a scientific standard of peer-review, I provided no evidence other than my say-so. Without a corroborating report or additional evidence, one person's written interpretation of song is just as effective (or ineffective) as stating, "I heard it sing and it was a Tennessee Warbler."

Tennessee Warlber image © 2006 Mike McDowell