Date: 11/9/17 9:45 pm
From: Marshall Iliff <miliff...>
Subject: Re: [MASSBIRD] ebird reports

One of my roles on the eBird team is managing our taxonomy and also helping
to define our subspecies policy. Birders have the option to select or
unselect "show subspecies" on their data entry via the web portal (
We will soon be adding this as a "Preference" in the eBird app for data
entry too.

I should state at the outset that we manage subspecies in eBird as
subspecies GROUPS. A subspecies group is a group of subspecies (or a single
subspecies) which is distinctive, and in some cases, may represent a
species. In these cases we are OK with birders differentiating the
subspecies group, but the subspecies within that group are so similar that
they are very very hard to ID. Eastern and Western Willets are each
monotypic subspecies groups (consisting of one subspecies) while Song
Sparrow (melodia/atlantica) is one that can be separated safely from some
or all other groups, but for which separating the widespread melodia
subspecies from the Atlantic coast breeding atlantica (NJ south) is much
more risky.

I always find it odd when people talk about "subspecies" as though they are
something totally discrete from species. In theory I guess they are, but
that would be assuming that we had perfect information and our taxonomists
had it all worked out already. In truth, our perceptions are ever-changing
and just because something is treated as a subspecies vs. a species does
not change the ability of birders to recognize them or track useful
information on their status, distribution, and identification traits.
Yesterday's subspecies include Nelson's Sparrow, Saltmarsh Sparrow, Pacific
Loon, Arctic Loon, Short-billed Dowitcher, Long-billed Dowitcher, Willow
Flycatcher, Alder Flycatcher, American Golden-Plover, Pacific
Golden-Plover, Clapper Rail, Ridgway's Rail, and many more. And yesterday's
species include American Coot, Caribbean Coot, Iceland Gull and Thayer's
Gull, which appear in eBird now as American Coot (Red-shielded), American
Coot (White-shielded), Iceland Gull (Iceland) and Iceland Gull (Thayer's).
So any birder that used to report American cot could report American Coot
(Red-shielded) and that is a 1:1 match with our understanding of the
taxonomic concept (think of it as a range+set of characteristics that
describe the taxon) both before the split and after the split.

So as birders, maybe it is worth trying to understand the variation, worth
trying to be careful with our reports of species *and* subspecies (and
genera, such as Pterodroma sp. or Semipalmated/Western Sandpiper), and
focus on that?

As to whether people are identifying their subspecies, you'd have to ask
them. Van Remsen, Tim Spahr, Peter Trimble and I had a nice day at Race
Point on Tuesday. Here's our list -- I was able to check most
(not all) eiders for subspecies to eliminate S. m. borealis, almost all
Cory's Shearwaters to eliminate C. d. diomedea, checked the Herring Gulls
for Vega Gull or European Herring (as best I could), and we were thrilled
to see and document an Ipswich, I mean, the princeps
subspecies of Savannah Sparrow...which was a lifer for one of the members
of our group. Can a subspecies be a lifer? (Of course it can!).

The Northern Harrier is one I would not have been comfortable eliminating
Circus cyaneus cyaneus on, since identifying females is really hard, but
the AOU recently split them, so that the USA bird is now known as Circus
hudsonius. I guess if I was really feeling honest I coulda reported it as
Hen/Northern Harrier.

I guess my semi tongue-in-cheek point is that reporting and studying
subspecies enriches the birding experience for many of us. If you want to
ignore, them, then that is fine.

If you want to learn about them, then David Sibley has this primer that
aligns very well with the eBird taxonomy:

For those in to real esoterica, the official guideines for what subspecies
are both distinctive and worth including on eBird filters (and thus
especially worth tracking around the country) is available here:
Download the link at the bottom of the page for the excel file of all
subspecies of interest in the USA/Canada.

Here's a bit more:

Red-tailed Hawk (borealis) -- See
Also see field guides that depict Harlan's, Krider's, and calurus

American Coot (Red-shielded) -- Really a synonym for American Coot
pre-2016. Nobody ever though Caribbean coot occurred here, and in the
end, the aOU decision was that those two species were just morphs
(i.e., did not even deserve subspecies status). Either way, most MA
birds have deep reddish frontal shields (vs. white in much of the

Song Sparrow (melodia/atlantica) -- Lots of variation here that I find
recongizable in some extreme cases and maybe in all cases. Compare: (so far, the only
form known from Massachusetts) (especially distinctive) (also distinctive
and one of my favorites)

From my perspective, folks should:

1) Report birds to the finest level that they can CONFIDENTLY identify them to.

2) Avoid use of subspecies if it makes them uncomfortable or if they
don't understand both the field marks and expected status and

3) Do their best to be consistent on using geographic assumption with
both species and subspecies. If I am going to assume that my distant
harriers are Northern Harrier (Circus hudsonius) after the split, I
might as well call them Northern Harrier (American) (Circus cyaneus
hudsonius) before the split. They are the same bird after is
just our perception of them that has changed. The same could apply to
Herring Gulls -- assuming Herring Gull (American) is not a federal
crime in eBird (although we do encourage use of subspecies when you
actually identify them to the subspecies).

4) Enjoy their birding and eBirding however they approach it.


Marshall Iliff

eBird Project Leader

PS - Willets depart early, typically, and in Barnstable County the eBird
filter allows 1 or more from 16 Apr-31 Oct. Eastern Willet (here we go on
subspecies again) is allowed at 1 or more from 16 Apr-25 Sep and Western
Willet from 25 Jun-31 Oct. Marbled Godwit, an easier ID and species that
more regularly lingers to Nov, is allowed at counts of 3-5 from 7 July to
14 Nov. These are judgment calls, of course, but based on both regularity
of misidentification and status and distribution in the area. If you want
to find out where Marbled Godwit is being seen, you could: 1) go to the
eBird map; 2) sign up for a "year needs alert" to hear about al reports
during that year until you see your personal first. The information is
there for those that want it -- in fact, that is what eBird is all about.

On Thu, Nov 9, 2017 at 7:56 PM, George W Gove <gwgove...> wrote:

> Are people really IDing birds to subspecies? For example in ebird reports:
> Red-tailed Hawk (borealis)
> American Coot (Red-shielded)
> Song Sparrow (melodia/atlantica)
> Pretty good if the observer can discern subspecies in the field!
> Another question. Glen D reported Western Willets and in the same
> checklist, there were 4 Marbled Godwits but only the Willets showed up as
> alerts. Do the Godwits not rate an alert? I spent a long time in the fog
> and fightin’ an extreme hi tide lookin for them stinkin birds; seems like
> they rate an alert to me.
> Geprge Gove
> Marlboro


Marshall J. Iliff
miliff AT
Westwood, MA
eBird Project Leader
Cornell Lab of Ornithology
Ithaca, NY

Join us on Facebook!