Date: 1/11/18 7:36 pm
From: Jon King <jonking271...>
Subject: Re: Documenting unusual records in eBirds
In response to Tom's post several things come to mind. The kind of papers
that would examine bird vagrancy patterns are the kind that would appear in
state or regional ornithology publications most likely. The *Kansas
Ornithological Society Bulletin* for example contains many papers on
unusual bird records. A quick scan through publications from neighboring
states shows that syntheses of unusual bird records are routinely
published. Do these count as science? I don't know and that probably
depends who you ask. It's not the kind of thing that appears in the *Auk*
or other prestigious ornithology journals but I think most us can agree
that such work has value. The changing patterns of nature have to be
documented somewhere right? As for what the real scientists do I'm not
qualified to make statements upon the value of the eBird review system. I
did once work on a niche modeling project however, with an ornithologist,
and part of my job was to remove outliers from the BBS dataset we were
using. Apparently the models used were sensitive enough that such outliers
(correctly identified or not) might potentially interfere with the results.

Tom brings up a second point which I like very much. He says we should rank
bird records probabilistically. To that I say absolutely. I wish that all
bird records committee's worked this way. There could be several categories
of records. The highest would be confirmed records substantiated by a
photo, video, audio, or specimen. Next would be probable records supported
by solid descriptions. Ambiguous and possibly correct records would be
ranked lower and unlikely records even lower. As for eBird the only problem
is that I'm not a computer programmer and I can't reconfigure eBird to work
this way. We use the software we have. Also keep in mind that eBird does
not exactly rake in the bucks and therefore they have limited resources to
develop new features. There are many features which have only recently been
released that have been in planning for a really long time. Yes eBird is
very imperfect but the system in place nonetheless allows for understanding
of status and distribution information in a way never before possible. As
Tom says hopefully it will get better :)

This whole conversation once again reminds me of the discussion on the ABA
blog I was referring to earlier on. It's worth a read for those interested
in how unusual bird records are kept. Be sure to read the comments below
Ted's post as there are many valuable points in there.

Jon King
Lawrence, KS

On Thu, Jan 11, 2018 at 9:08 PM, Tom SHANE <tom.shane...> wrote:

> Well said Henry ! !
> I have said that bird records should be scored from 1 to 10 by a records
> committee. Then a researcher would have access to everything, not just what
> a committee considered important. The researcher has a much different ideas
> and needs than what a committee thinks he should have. I remember when the
> boys from KC found the Canyon Wren out in MT County. Sara and I drove out
> to document the bird and we did. Shortly after that, it was posted that the
> first record was found in Halstead in the early forties, but never
> published in any of the books or directories for the following 50 years. A
> researcher of wrens might have considered it a fair record, but other
> people took that away from him. I'm not with Science when it comes to poor
> decisions like that.
> Then, are "Patterns of Vagrancy" really science. What university professor
> in this country has a grant to study and analyze patterns of vagrancy? To
> me it is just a side bar of the enormous energy spent by birders looking
> for rarities. Can someone cite for me a couple papers published on patterns
> of vagrancy?
> Hope and Change, probably a dream
> Tom Shane
> 67846
> For KSBIRD-L archives or to change your subscription options, go to
> For KSBIRD-L guidelines go to
> To contact a listowner, send a message to
> mailto:<ksbird-l-request...>

For KSBIRD-L archives or to change your subscription options, go to
For KSBIRD-L guidelines go to
To contact a listowner, send a message to
Join us on Facebook!