Date: 3/14/19 12:39 am
From: David Irons <LLSDIRONS...>
Subject: [obol] Reporting birds to subspecies and how your local eBird reviewer will deal (for now) with reports of non-local Song Sparrows
Greetings All,

This post mostly pertains to users of eBird, which seems to be a good majority of us these days. If you are not an eBirder, you may find this post of minimal interest.

Over the past couple months three different non-local Song Sparrows have appeared in western Oregon, with single birds at a Brookings feeder, along Bond Lane nw. of Eugene and one along Rentenaar Road on Sauvie Island. That latter bird has been seen and photographed by dozens of observers, some of whom became convinced early-on that it was an "Eastern" type Song Sparrow. In general appearance this bird closely resembles some of the migratory subspecies that breed east of the Rockies and at least one sedentary subspecies that is widespread in California. After considering the likelihood of three different individuals wandering north from a sedentary population in the same season (not very likely) there is now at least a general consensus that these birds originated from a more northerly and migratory breeding population.

Somewhere along the line some folks began reporting the Sauvie Island bird under the designation "Song Sparrow (melodia/atlantica), which at this point is really the only way to indicate in an eBird checklist that you believe you saw one of the eastern forms. I and others have taken issue with this designation, as it suggests the bird was one of these two subspecies, neither of which occurs regularly any where near Oregon. In fact, Melospiza melodia atlantica is a salt marsh subspecies that is resident along the immediate coast from roughly Massachusetts down to the Carolinas. The chances of this non-migratory Atlantic Coastal subspecies making it to Oregon are incredibly remote. Then we get to M. m. melodia the nominate eastern form of Song Sparrow. Here is how the Birds of North America Online account describes its taxonomic history:

"Melospiza melodia melodia (Wilson, 1810). Includes as synonyms M. m. juddi Bishop, 1896; M. m. acadica Thayer and Bangs, 1914; M. m. beata (not of Bangs) Todd, 1930; M. m. euphonia Wetmore, 1936; M. m. callima Oberholser, Oberholser 1974c<>; and M. m. melanchra Oberholser, Oberholser 1974c<> (see American Ornithologists' Union 1957<>, Todd 1963a<>, Browning 1978<>, Patten 2001<>). Type locality restricted to Philadelphia, PA. Distributed as a breeder throughout eastern part of range, except parts of mid-Atlantic Coast, west through the Great Plains and Prairie Provinces; winters in se. U.S. south to Florida. Characterized by white underparts with well-defined black streaking, reddish-brown dorsal streaking on brown background with buff-gray fringes to feathers, short bill, and long wings. Recognition of subspecies through much of the east “extend[s] the trinomial system to the limits of utility” (Mengel 1965b<>: 511). Shortly after its description, Ridgway (Ridgway 1901<>: 358) synonymized M. m. juddi with M. m. melodia, remarking that specimens from Atlantic Coast and Great Plains “average slightly grayer than those from the intermediate region, but the difference is so slight and inconstant that subspecific separation seems to me unjustifiable.” Bull (Bull 1974<>: 600) considered neither M. m. euphonia nor M. m. juddi valid, noting the “poorly differentiated” M. m. euphonia “is merely part of an east-west cline extending from New England and New York . . . through several slightly differentiated populations including the more western . . . juddi (itself part of the cline).”

As you might infer from reading the paragraph above, sorting out whether a subspecies is "valid" or not is no day at the beach, even for professional ornithologists who are examining museum specimens in the hand. The northwestern most birds currently assigned to M. m. melodia (historically and still sometimes referred to as M. m. juddi) seem to be a pretty good match for the Brookings, Eugene and Sauvie Island trio of birds, which are nearly identical in appearance. But can we assign these birds to this subspecies with any degree of certainty? The answer to this question is currently no and is likely to remain so.

Herein lies the issue. These birds look and are distinctly different from any of the populations of Song Sparrows that regularly occur across Oregon and the Pacific Northwest. Thus when we report them, particularly in eBird, there is a natural desire to make it clear that what we saw were non-locals. Your local eBird reviewers all understand this and we want to be able to endorse your well-documented sightings for inclusion in the public database. Unfortunately, in our collective opinion the current eBird taxonomy does not offer a suitable designation for these birds. We have been discussing this issue for about a week or so in an effort to come up with a solution. None of us is comfortable assigning these birds to the "melodia/atlantica" slash pile, as that would imply that these individuals were accepted as being one of these two subspecies, a conclusion that would be at best speculative.

In the coming days, I will be reaching out the eBird project managers in hopes of convincing them to create a new taxonomic sub-heading for the reporting of birds like this. I'm hoping they will see the utility in creating a "Song Sparrow (melodia group)" sub-heading, sort of like "Song Sparrow (rufina group)" which is a sub-heading for the suite of larger, darker and more reddish Song Sparrow forms found from Alaska down the Pacific Coast through B.C., Washington, Oregon and into Northern California. This has been done with a number of polytypic species, most notably Fox Sparrows, for which there are four subspecies group headings -- Sooty, Red, Slate-colored and Thick-billed. There are multiple subspecies that fall under each of these very useful sub-headings, not that most of us can name those subspecies. It occurred to me the other day that most Oregon birders have a good understanding of what a "Sooty-type" Fox Sparrow looks like, but how many of us can name any of the subspecies that come under that heading? With Song Sparrows we face much the same issue. There are at least 20 recognized subspecies of Song Sparrow, many of which are not readily separable in the field from other similar subspecies.

In the absence of clarity about which subspecies these three birds are and without a good way (currently) to define them in an eBird checklist, the statewide eBird review team has decided that we will for now table the review of birds submitted as "Song Sparrow (melodia/atlantica). We will neither validate nor invalidate these reports. Please do not take this personally. It reflects on our lack of certainty about how to treat these birds rather than an issue with your reporting of them. As stated above, we want to add these birds into public eBird database and in time I believe that we'll find a way to do it. At the point that we are offered a reporting option that makes sense for these birds, your local reviewer will send you a query asking you to edit your checklist and move these birds to the desired sub-heading and they will be processed at that point.

One final thought on the reporting of birds to subspecies. Unless you yourself can make the case for a bird being of a particular subspecies or subspecies group, it is best that you report the bird only to species. Too often eBird checklists have birds reported to subspecies by observers who are making presumptive subspecific assignments that are often based on geography and not field marks. It may feel like you are making a more important contribution to science with higher resolution reporting (denoting subspecies rather than just species), but in essence this type of reporting only muddies the waters with low quality data. We should only report what we can identify with some reasonable degree of certainty. With most subspecies, certainty is fleeting and subject to debate if you have more than one ornithologist in the room.

Please know that as a group Oregon's eBird review team understands the confusion and frustration that some of you may be feeling upon seeing that your report of one of these birds has not been accepted into the public database. Please also understand that just because someone is an eBird reviewer, that doesn't mean that they are a professional ornithologist. We are not necessarily any more prepared than you are to identify a bird to subspecies by looking at it in the field or seeing a photo of it. We would be doing a disservice to the eBird community as a whole if we were to make a habit of accepting subspecific assignments without some solid reason for doing so.

Dave Irons
Statewide eBird Review Coordinator

Join us on Facebook!