[BCNnet] Fw: Birds and Science from Doug Stotz

[John Elliott]

This is Doug Stotz's response (posted on IBET) to Steve Bailey's comments.


I felt I needed to respond to Steve Bailey's interesting and provocative
post
with respect to science, censusing and specifically projects that attempted
to
look at declines of some birds in the Chicago area this summer.
Steve makes a number of good points about how science should be done
ideally--the value of long-term data, some of the weaknesses of short-term
and
limited data, and some of the hazards of interpreting Christmas Count data.
But
I have to say that I think I fundamentally disagree with him in three major
areas: the value of the data that was collected in the crow and chickadee
study, the strength of the tie to West Nile Virus, and the usefulness of
Christmas Count data for investigating such issues.
Before I get started in the specifics, I should just mention my personal
biases. I think I qualify as a scientist. I have a Ph. D. in Evolutionary
Biology, and I have worked at the Field Museum in a scientific position of 8
years. However, I am less concerned with the need for scientific purity than
with the need to try to understand what is going on in the world. Our
desire to
understand what is going on in the world of birds, I think requires us to
accept data from sources that are not perfectly clean and in fact to search
out
those sources. One needs to keep in mind the limitation of those sources,
but
to ignore these data reduces our ability to determine the "truth." Just as
an
example of this, in most taxonomic fields, sight records are completely
ignored
in determining the known distribution of a species. Ornithology is one of
the
few fields that pays any attention to sight records at all. Records
committees
such as IORC, of which I am Secretary, are an attempt to improve the
quality of
the data based on sight records, so that that data approaches the ideal for
most scientists of a well-documented specimen housed in a museum. But the
fact
is that much of what we know about the distribution and abundance of birds,
and
especially the dynamic nature of this comes from sight records of amateur
birders. One of the heartening things of the past couple of decades is the
rise
in formalized systems for collecting and reporting that data. Anyway...
On the data in the chickadee and crow studies, there is not question that
the
amount of data collected is small, smaller than you would like. It is
clearly
not sufficient for a publication in a major scientific journal. I think
though
that it is sufficient, especially the chickadee survey to demonstrate a
major
decline in that species around Chicago. Ideally, some Illinois scientist
would
have designed a carefully worked out and scientifically rigorous study to
evaluate the effects of West Nile last year. But that didn't happen, in part
because I think nobody imagined that West Nile would hit Illinois so hard,
although we did know it was coming. So, BCN put together a project to
attempt
to get at least some information on the topic. Timing was important. If they
waited for established monitoring programs such as the breeding season
censuses
in 2003, or even Christmas Bird Counts, some of the local declines could
have
been reversed I applaud the attempt to get data in the heart of last year's
West Nile outbreak, even as I recognize the limitations of that data, and
look
forward to further attempts to collect data that can be used to track
population changes.
Steve talks about biases, and this is always a problem. In an ideal study,
the
sites would be chosen randomly to avoid biases. However, the fact that this
was
not done does not necessarily invalidate the results. Unless the biases in
how
sites were selected would be expected to influence the results in a fashion
relevant to the question being examined, I think the issue of bias is not
something I would focus on. Frankly, I can't personally see an obvious way
in
which the chickadee study would have had a bias that would invalidate the
results. I personally participated in the chickadee survey, surveying two
areas
in western Cook County. These areas were selected because they were places
where I had comparable data in earlier years from the same time of year,
with
numbers and hours spent observing. So these sites were not selected
randomly,
but they were not selected with reference to anything specific about
chickadee
numbers (except that there had been chickadees there in past years). I don't
think that such selection criteria would be expected to bias the results.
In terms of the tie of the declines found for these species to West Nile,
the
nature of the tie is correlational and circumstantial, not causal. Ideally,
you
would demonstarte directly that the cause of any observed decline was due to
West Nile. That was not done, and frankly I am unable to imagine how one
would
do it. However, I think there are several strong lines of evidence-the
declines
were most severe in areas where human cases are most concentrated, the
declines
occurred during the time period that West Nile was having most impressive
affects in terms of birds being found dead and tested as positive for West
Nile. Finally we know that these species are susceptible to West Nile, and
crows apparently are particularly susceptible. So, yes there is no "proof"
that
West Nile is responsible for the declines we've all observed in crows and
chickadees, and that the studies this summer and fall provided some data
for,
but in my view there is strong evidence. I would just remind people that the
scientific purists who work for tobacco companies still claim that there is
no
proof that cigarette smoking causes long cancer.
Finally, we have just come through the Christms Bird Count season. Exciting
to
me because I love Christmas Counts, but also exciting because it is a huge
dataset with years of data collected from the same areas. As Steve rightly
points out, there are variables that need to be considered when looking at
CBC
data, and because of the highly unstandardized manner in which the data is
collected, you need to look at a lot of data and/or be looking for large
changes. If you go to the CBC home page,
<http://www.audubon.org/bird/cbc>http://www.audubon.org/bird/cbc/ , at the
bottom you can see a variety of studies that use CBC data to track changes
in
bird populations. Fortunately (or perhaps unfortunately for the species
involved) the changes in populations being looked for were massive, at
least in
the Chicago area. This year's CBC season was blessed with normal weather,
and,
at least in the Chicago area, little in the way of inclement weather, so I
think the weather variable is pretty minimal in affecting this year's data.
I
think that the CBC dataset is almost ideal for looking for the details of
the
declines in local bird populations due to West Nile. It provides many hours
of
observation, a great geographic spread, and many years of comparable data.
I am
analyzing that data myself, and I know others are too. Not to steal
anybody's
thunder, but preliminarily it looks clear that crows are way down in the
Chicago area, and are low, but much more variable elsewhere in the state.
So sorry to have gone on for so long, but I thought that a different
perspective on the West Nile story was called for. It will be interesting to
see if the Spring Bird Count and censuses done next summer show the
continued
declines or whether this will have been a one year deal. The summer censuses
will be especially interesting because it will be our first chance to see if
some of the more sensitive migrant species (things like Ovenbirds,
Rose-breasted Grosbeak, etc.) may have been affected. Because loss of birds
due
to mortality last year may get obscured by immigration from other parts of
the
range that were not hard hit by West Nile, we may never know if there was
significant mortality in many of our Neotropical migrants. My personal
expectation is that West Nile will not be a big factor in the Chicago area
next
year and that we'll see complete recovery within a couple of years. I hope
so.

Doug Stotz
Douglas Stotz
Conservation Ecologist/Ornithologist
Environmental and Conservation Programs
Field Museum of Natural History
1400 S. Lake Shore Dr.
Chicago, IL 60605
Phone: (312)-665-7438
Fax: (312)-665-7440e-mail: stotz@fmnh.org