Re: [CALBIRDS] eBird: A Call for Participation
- At 12:56 PM 4/4/2003 -0500, you wrote:
>What process in eBird insures the quality or validity of the reports thatBirders:
>are entered? This has been a primary concern I have had. Similarly, with
>the Great Backyard Bird Count, the info seems largely unvetted and other
>than selected rarities, I can't see how such a thing could be controlled
>for quality. Any thoughts on this re eBird?
David Suddjian raises an important point about eBird that I did not
elaborate upon in my initial message.
For each state (and, in California, for the northern and southern regions)
a set of "Filters" was developed by the eBird staff and local authorities
(usually an NAB regional editor). These filters allow eBird to flag
unusual observations. For example, if you were to report five
Yellow-breasted Chats in southern California in May, there is no
problem. But a report of a hundred chats in May, or of even a single chat
in December, would trigger a flag when such records are entered. The
observer is alerted that such a sighting is unusual, and is asked to
"confirm" it. Obviously a bogus sighting can still be "confirmed" by the
observer, but it is not entered into the eBird database until a regional
authority (NAB editors, again, will likely play an important role here) has
vetted the record (either through familiarity with the circumstances or
through requesting, receiving and approving documentation). For CBRC
review species, a flag will obviously pop up for any claim in any month of
Yes, incorrect sightings will infiltrate the eBird database. Especially
troubling are records that don't trigger flags given the existing filters
(for example, an Abert's Towhee could be entered from Santa Barbara because
the filters only go down to the state or regional level). But there are
ways of assuring as clean a database as possible, and the eBird staff, NAB
staff (over time) and other regional authorities will be enlisted to
continually improve mechanisms for keeping the records "clean." And on the
positive side, this could be a great vehicle for bringing important records
to the attention of regional editors.
Ultimately I would like to see eBird filters in California written at the
County level, if we can get the participation of all county subregional
In summary, I can assure birders that quality control in eBird will be
orders of magnitude better than that in place for the Great Backyard Bird
Count, but I recognize that there will be some recurring issues that we
will all have to work on to improve.
Thanks, David, for your excellent query.
Kimball L. Garrett
Ornithology Collections Manager
Natural History Museum of Los Angeles County
900 Exposition Blvd.
Los Angeles, CA 90007 USA