Counts vs Pollard Transects

Michael Gochfeld gochfeld at eohsi.rutgers.edu
Wed Jul 16 10:23:10 EDT 2003


Patrick Foley wrote:

> Rather than a fourth of July count, I think American lepsters should
> consider some approach similar to the British Butterfly Monitoring
> Scheme. Pollard and Yates 1993 describe the Scheme in a book called
> Monitoring Butterflies for Ecology and Conservation.
>
> The basic idea as I understand it is to run frequent (weekly) repeated
> transects, attempting to account for weather variation by picking only
> favorable days. The transects are walkable strip transects in locations
> representing a diversity of habitats across the country. Weekly and
> annual information results, which should allow easy comparison (of the
> visual catch per unit effort sort).

===================================================
We (Joanna and I) agree with Patrick about the utility of the Pollard
Transect approach.  We have covered a transect along woodland edge
behind our
house for more than a decade.  It provides very useful and interesting
information, quite different from the 4JC, but one doesn't preclude the
other
(except that our time is limited).

So we believe in transects.  But there are significant limitations to
transects that need to be understood.  For one thing, the habitat on or
adjacent to the transect may change.  Transects need to be managed and
management changes what you study (Hawthorn Effect).

Joanna and I have been engaged in a dialogue with the State of New
Jersey
which is very forward-looking in developing a butterfly survey.  The
original
plan was to rely on Pollard Transects, but we provided arguments that
suggest
that transects are valuable but not sufficient. We advocated
supplementing transects with a maximizing approach over the entire
Wildlife Management Area, seeking out all representative examples of all
habitats. 

 We will be pleased to share the document (6 pages), but provide the
conclusions below:
========================================================================================

ANY METHOD(S) CHOSEN SHOULD BE ABLE TO ADDRESS THE QUESTIONS THAT THE
AGENCY
HAS----mainly
1) Are there or could there be rare, threatened or endangered species on
the
Wildlife Management Area
2) What management strategies should be used to enhance or maintain
habitat
for these species and for butterflies in general.
3) Are there significant changes in the populations of these species
(and
butterflies in general).

No single method can address all biodiversity/conservation concerns for
all
species in all habitats.

The methods chosen should have sensitivity, specificity, and predictive
value
defined as follows:

Sensitivity =  TP/(TP+FN)    How often real change can actually be
detected

Specificity=  TN/(TN+FP)    How often can no-change actually be verified

Positive Predictive Value = TP/(TP+FP)   How often when change is
detected is
it real.

Negative Predictive Value =TN/(TN+FN)  How often when no change is
detected
is it r true that no change has occurred.

IN CONCLUSION
We applaud the formal effort of NJDEP to document the species and
numbers of
butterflies present on the set of Wildlife Management Areas (WMAs) and
hope
that this effort can be expanded to other public lands and private
reserves
that can be managed to maintain and enhance New Jersey’s biodiversity.

It is particularly timely that in June 2004, the Society for
Conservation
Biology will hold its annual meeting in New York with the topic of
“Conservation in Urbanizing Areas”.

We recommend that a paper comparing transect vs maximization approach,
be
prepared for the 2004 meeting of the Society for Conservation Biology.

We recommend that a matched transect/maximization approach be used to
assure
that the butterfly fauna is appropriately represented.

We recommend that nectar sources (and host plants) be recorded as well
as
incidents of predation.

We recommend that the impact of invasive exotic plants be assessed.

 
 ------------------------------------------------------------ 

   For subscription and related information about LEPS-L visit:

   http://www.peabody.yale.edu/other/lepsl 
 


More information about the Leps-l mailing list