eBird issues in California
[Note: if you don't use or care about eBird, you may not want to read
on, though I encourage everybody to give eBird a try and, ideally,
incorporate it into your regular birding routine]
California birders should be congratulated for going full-speed into the
eBird - our state is routinely among the leading states/provinces in
eBird submissions each month.
In reviewing flagged eBird submissions for southern California (mainly
for Los Angeles County), I've noticed a couple of frequent problems that
arise from eBird submissions that I hope the eBirding community will
take note of. I hesitate to call these "abuses of eBird", since one of
the functions of eBird is its use in personal list-keeping, and each
user should have some leeway in how they use it for that purpose.
However, these problems make the eBird public/scientific database less
useful and make the review of eBird submissions more difficult. These
1. eBird submission should ONLY be a list of bird observations (or even
a single observation) for a PARTICULAR DAY at a PARTICULAR LOCALITY.
The latter part is relative - it is valid to enter in a single
submission a list birds for a large area (say, an IBA) or along a
traveled route of up to several kilometers (mapping the submission at
the starting point of the route) - but obviously the better defined and
more succinct the locality, the more useful the data. But it is clear
that some observers use eBird to enter a cumulative list of all birds
seen at a "locality" (backyard, county, etc.) over a period of days,
months, or years. The problem is that you must specify a date for the
submission, and the database will then treat all of these records as
having been on that date. This is an invalid use of eBird and leads to
nonsense data (such as Common Nighthawks or Willow Flycatchers in the
San Bernardino Mtns. in March). The problem is that the filters only
catch the most outlandish records - many others slip through. I would
suggest to eBird reviewers that when a list clearly contains records
over a long time period, or a cumulative list, that the entire
submission be invalidated (it will still be available to the submitting
observer in his or her "My eBird" database).
2. It is a GREAT help to eBird reviewers if submissions make use of the
"Notes" functionality or the "Comments" section to amplify the
submission of any unusual records. In general, if you are asked to
confirm a flagged record (you know the drill, the "Oops!" message), you
should give some indication of how that record was documented. You're
not required to do this, of course, but it is a tremendous help to those
of us who review flagged records to know things like: "several photos
obtained and posted at.....", or "brilliant red male, only the third
I've seen at this locality", or "present for a week; first found by
Chuck Darwin", or "the earliest record for this site by four days", etc.
Any help you can give the reviewers will help reduce the time that
flagged submission sit in limbo while awaiting validation. I am
perfectly aware that the eBird filters for some parts of California are
crude, at best, and that many perfectly normal observations are being
flagged; filtering is an adaptive process, and your help in suggesting
improvements to county filters is always appreciated.
I know that this listserve is not the proper forum for discussing the
intricacies of eBird, but with eBird use exploding in California and
throughout the continent, it does seem appropriate to try to increase
its efficiency and quality.
Kimball L. Garrett
Ornithology Collections Manager
Natural History Museum of Los Angeles County
900 Exposition Blvd.
Los Angeles CA 90007
(213) 746-2999 FAX
[Non-text portions of this message have been removed]