Date: 10/7/19 10:02 pm
From: <clearwater...>
Subject: [obol] Re: Oregon eBird Tropical Kingbird statistics
Thanks Bob, for this careful compilation.

Now if you can work through the even larger number of records of Tropical Kingbirds all along the coast of Washington, up to BC, all or nearly showing a similar pattern of timing of arrival in late fall (in contrast to the spring occurrence of the very small number of Couch's Kingbirds that have strayed anywhere west/northwest of Texas, as I tabulated earlier), you'll have a more representative data set.

Then you can go back to the concept of probability. Even if 2/3 of the Tropical-type kingbirds seen in this region were not reported as vocalizing, what are the odds of any given percentage being Couch's? I've already demonstrated that this is below 10%, with very high confidence. But working through more of the data will allow justification of setting that percentage even lower ... it's a straightforward numbers game.

Then we could do the same thing with American Three-toed Woodpeckers -- how many reports in eBird conclusively rule out its Eurasian cousin? Or Black-billed Magpie vs. Eurasian? Or for an example less far-fetched, how about Western Meadowlark vs. Eastern Meadowlark?

From: "Robert O'Brien" <baro...>
To: "clearwater" <clearwater...>
Cc: "Oregon Birders OnLine" <obol...>
Sent: Tuesday, October 8, 2019 7:41:29 AM
Subject: Re: Oregon eBird Tropical Kingbird statistics

Joel & OBOL
Some years ago I said on OBOL that there is more to birding than identifying individual birds.
Well, there is also more to birding than eBird, as several OBOLers have lately mentioned.
But, in this case, both are involved, which makes it either doubly interesting or doubly dull.
And I imagine that those in the latter camp aren't even reading this email. There could be 100's of non-readers. 1000's?

I agree with Joel's statement that calling all kingbirds Tropical is unlikely to have a large effect on the Tropical distribution maps
Even though there are only 10 calling eBird records from Oregon up through 2016 (see below). Especially since there are also records from Washington and British Columbia; but which I have not analysed.

I do concede to Joel that you only need to identify an individual bird once. Then it stays identified so long as it stays put.

So I've gone back to the eBird data to see how many of the non-calling reports were of a bird that was identified by
at least one person hearing the call. There a significant number of such cases, as Joel intimidated, but the last one in the table below wins hands down.
There were 55 reports of this bird and only 1 person reported it calling. Twenty eight eBirders reported that kingbird on 11/20/2016 (Lincoln City) and three reported it at from the exact time as the single mention of calling. Clearly a close call. One would have to guess that more than one person heard the call, but didn't think to mention it. Ergo my previous suggestion that with this species one should always mention vocalization: yes or no . And most especially if you hear Kip, kip.....................

As the table below shows, there were 101 reports of Tropicals of which only a single person reported the call. There were two reports of a bird that 2 out of 5 observers independently reported calls two days apart.

Now with the data as reported (and neglecting possible failure to report heard calls) you can subtract 101 from 302, leaving 201 allegedly 'silent' reports, 2/3 of the total reports, which were technically-unsubstantiated reports of the 'Oregon' expectation. Any of these could have been a silent Couch's or even a calling Couch's that the observer failed to report on. A calling Couch's has a very high probability of NOT being reported as it is just assumed to be a Tropicall, except by an alert, experienced birder. There are 201 Kingbird reports that fall into this category in Oregon. I have to agree with the original OBOL recommendation that such observations should appear as Tropical/Couch's, a classification available in eBird. Otherwise what is the point of this classification. If you wish to confirm the species, get there early, sit there all day if necessary, hoping for a call. Then report that you observed the bird for 10 hours and it was silent the entire time! As Tropical/Couch's.

Bob OBrien Carver OR See updated table below. Individual reports are total reports of that bird at that locations and time period.. Days gives the period over which the bird was observed at the same location, not necessarily every day. This assumes that it wasn't a Cassin's or a Western Kingbird, unlikely at this location and time of year, but so is a Couch's.

OBSERVATION DATE Calling Vocalizing Individual Reports Days Report Calls SPECIES COMMENTS Observations from earliest reports through 2016

11/06/12 Yes 5 5 2 One in 8' pine about 15 ft from me along fenceline, the other at top of a spruce. Both had bright yellow belly and breast, white throat. Calling between both birds, single note brief trill. Both actively feeding. These birds have been reported by others at this location during the last few days.
11/08/12 Yes same same same Reported on OBOL. Hawking insects from low to middle height perches. Deeply notched dark--brown tail, no white edging. Calls as in Sibley app.
11/01/13 Yes 1 1 1 Photos available upon request (both myself and Tom got some). This is presumably the same bird that many have been seeing at this location. Vocalizing freely with characteristic metallic twitter. Obvious kingbird morphology and behavior, sitting on various treetop perches and flycatching before returning to perch. Forked tail and long heavy bill and yellow breast evident.
11/27/13 Yes 1 1 1 1st thought WEKI, but this bird had a forked tail, yellow up the chest to throat, much yellow underside of wings, no white tail margins, also it vocalized as we were leaving.. in true TK form.. woohoo
11/05/14 Yes 1 1 1 Found today by Janet Lamberson, who told me of it. Thank you, Janet! About 2 minutes after I stepped out of the car, I heard the characteristic chittering call of this species. As I jogged toward the sound, I saw two birds in the air and realized it was the kingbird being chased by a Sharp-shinned Hawk! The Sharp-shin came within a couple of feet and just missed. The kingbird escaped and flew about 200 m to a perch atop a tall conifer and continued chittering. Apparently it was really worked up (understandably so!) -- and likely had been scolding the hawk in the first place when I'd initially heard it.
12/13/14 Yes 13 8 1 Seen by others previously at this location recently. Watched it fly from the east into the trees behind the pizza shop calling.
10/02/15 Yes 4 4 1 Yellow-bellied Kingbird, much larger bill than western. Rattling call different than western. Photo.
11/03/15 Yes 5 11 1 Heard his trilling call across the pond near the parking lot first- I was on the northern side opposite. Used my scope to locate him on roadside wires. Distant view but could detect his yellow breast, large bill and characteristic kingbird perching style. Wasn't able to relocate when I returned to parking lot. Seen here previously a couple weeks ago.
11/29/15 Yes 16 14 1 Photos plus heard calling giving the zippy trill which we are very familiar with. Video can be seen at: [ | ] .
11/20/16 Yes 55 13 1 Heard and seen. Twittery call diagnostic (as opposed to the "pip" of a Couch's). Primaries appeared to be of uneven length which also suggests Tropical rather than Couch's, which tend to have evenly spaced primaries (Couch's).--Seen in shrubbery on the south side bank of river downstream from bridge.

TOTALS 101 58 10

On Fri, Oct 4, 2019 at 5:53 AM < [ mailto:<clearwater...> | <clearwater...> ] > wrote:

P.S. To be a little more accurate, not all of those 17 reports on the Wireless Rd. bird from Nov-Dec 2015 were independent observations. Some were by people birding together who gave identical comments/photos in their reports. From a quick glance through, there were about 11 independent observations in the series, one of which (by Shawneen Finnegan and Dave Irons on 29 Nov 2015) includes a description of typical Tropical Kingbird vocalizations.

If anyone has bottomless energy to dig into this, you might consider making a table like in the attached graphic, to summarize the evidence for each *series* of observations of a single or flock of reported Tropical Kingbirds at a given location. This is rather tedious, and some interpretation is needed to handle cases where people have posted checklists for personal/GPS locations rather than using an eBird "hotspot."

But personally and probabilistically, I feel like I've looked at enough of these reports to establish that there is no reasonable cause to doubt that at least 90% of Tropical Kingbird reports in our region can safely be treated as such, for any reasonable use of these data in an aggregated analysis.

Joel Geier
Greater Tampico, Benton County

Join us on Facebook!