Don Baccus wrote:

> Dan Varland:
> > I am banding and color marking Peregrine Falcons along
> > Washington's outer coast (Columbia River to Pt. Grenville)and in
> > Grays Harbor. The color markers are blue and black and are coded
> > with two numbers or one number and one letter. They may be read
> > from a distance of up to 100 meters with a spotting scope.
>
> This guy has obviously never met Stuart!
>

Ouch, that hurt :-)

Then Gene Hunn wrote:

> Two _moffitti_ type Canada Geese at the Montlake Fill this morning
> (Nov.12) had neck bands with the following numbers: 042J and 064T.
> Please pass this along if you know who might be interested.

064T usually hangs out with 042T. While never one to dismiss reports out of
hand immediately - well not that often :-) It might be worth checking the
collar inscription of 042J again. I have seen 064T and 042T together (less
than 5m apart) every time I have checked (with one exception) over the last 8
months.

The 'T's on the bands are particularly badly inscribed - tall and the
cross-piece is not very well defined. If part of the inscription was hidden by
neck feathers then it is possible to mis-read the number. Some dirt on the
collar would also be a problem. It took several checks when i first read the
collar to make sure the 'T' wasn't and 'I'. The 'J' on the collars I have seen
so far at UW do not have the top crossed.

Then again I might be wrong .....


The pitfalls in reading bands, collars and color sequences are considerably
higher than one might imagine. The episode over the Michigan Canada geese is a
good example of operator error - though that was an inability to read my own
hundwroting :-)

Trying to read dunlin color bands in the field - both here and in Norway (on
the breeding grounds) is really hard. Typically we found that on the tundra
the ability to correctly read a sequence is based on:

time x distance = constant

ie the closer you are the quicker a sequence can be read. It still takes
several, roughy 15-20 seconds to read and verify a sequence on a dunlin at a
distance of less than 20m. The Tay Ringing Group were getting an error rate of
over 20% for color-banded sanderling reported away from their study site. It
brings up a few interesting points in designing color codes such as Hamming
Distance and other information theory stuff.

Daniel Varland should not have much trouble with the peregrines as they are
likely to sit still for quite a period of time - though it's not that easy to
get close. Some of the birds I have seen sitting on the beach at Ocean Shores
have been pretty cooperative though. Still I would think it would take a
couple of minutes to be sure the sequence has been read correctly.

Stuart