[obol] Re: seawatch gull counts
- From: Mike Patterson <celata@xxxxxxxxxxxx>
- To: 'OBOL' <obol@xxxxxxxxxxxxx>
- Date: Mon, 27 Oct 2014 17:32:17 -0700
We shorted many of our numbers, because they seemed so overwhelmingly
large. In my notes for Sunday, wrote rates rather than try to keep
an actual count. We counted for 2.5 hours, so the math is:
(Count/min)*60*2.5 = number encountered during watch
From that I got:
California Gulls 100*60*2.5 = 15000 (5000 reported)
Northern Fulmar 10*60*2.5 = 1500 (300 reported)
Pacific Loons 80*60*2.5 = 12000 (8000 reported)
I always hesitate reporting math based values, in part because the
rate is not uniform over the morning and making enough rate counts
over a morning to be statistically robust requires not paying attention
to all the lower frequency action which is often way more interesting.
But I've also had folks argue with the big numbers, usually folks
who've not witnessed these kinds of events, who think I'm being over
dramatic. So I habitually trim them.
But given all the supporting data from other locations, I may go back
into the eBird list and fix these (eBird filters don't like these big
numbers either).
--
Mike Patterson
Astoria, OR
String Theory
http://www.surfbirds.com/community-blogs/northcoastdiaries/?p=2182
OBOL archives: www.freelists.org/archive/obol
Manage your account or unsubscribe: //www.freelists.org/list/obol
Contact moderators: obol-moderators@xxxxxxxxxxxxx
Other related posts:
