Subject: Re: BTPI
Date: Mar 9 16:24:06 1995
From: Dennis Paulson - dpaulson at ups.edu


>Re Jennifer Seavey's BTPI discussion:
>
>Yeow, she wasn't kidding. Here are the BBS trends for WA and OR:
>
>State Period Trend n sig.
>OR 1982-1991 -14.7 19 p<.01
>OR 1966-1991 -3.2 21 p<.01
>WA 1982-1991 +.7 21 none
>WA 1966-1991 -3.5 23 p<.1
>
>The trend is expressed as % annual change using a formula similar to that
>for calculating compounding interest. n is the number of BBS routes from
>which the trend was calculated. The authors warn against using data for
>species where n<14. p is the significance of the trend, from testing the
>hypothesis that the trend is actually 0. The source is an untitled,
>unpublished report from Bruce Peterjohn for USFWS/NBS biologists.
>
>____________________________
>Mike Smith
>Univ. of Washington, Seattle
>whimbrel at u.washington.edu
>http://salmo.cqs.washington.edu/~wagap/mike.html

Hmmm, maybe a few more words are needed here. Yes, it's clear that pigeons
have declined in Oregon, but it's anything but clear that they've done so
in Washington. A p of <0.1 is not considered particularly significant in
science (people prefer 'p's <0.05 or less for most statistical tests), so
the 1966-1991 decline is "not particularly significant." The decline in
Washington in the last decade--presumably what should concern most of us on
this list--is *not* a decline.

A little hypothesis testing was in order: I just checked BTPI totals for
the Seattle Christmas bird count for two 10-year periods, 1963-72 vs.
1983-92. Average per count 78.7 (range 2-182) vs. 145.0 (20-495).
Hypothesis of decrease *rejected.* Party hours increased between the two
intervals, but not that much (138 in 1972, 180 in 1992). If anyone wants
to do a similar analysis for another Washington CBC, be my guest. We can't
do the same thing for Oregon, because, as I wrote before, they don't seem
to hang around for Oregon CBCs!

I have already seen several reports in which BBS data were used to make
generalizations about massive bird declines when in fact further assessment
casts doubt on such conclusions. We should carefully monitor populations,
and even err on the side of caution when interpreting data, but let's stay
credible.

No offense to Jennifer or Mike intended by my response, and thanks for
giving us some quantification of what is otherwise a bunch of subjective
impressions. But my subjective impression is still that BTPIs haven't
declined obviously in western WA since I've lived here.

Another hypothesis: we're feeding pigeons more in Washington, and the
Oregon ones have heard about it.....

Dennis Paulson, Director phone: (206) 756-3798
Slater Museum of Natural History fax: (206) 756-3352
University of Puget Sound e-mail: dpaulson at ups.edu
Tacoma, WA 98416