Hi Tweeters,

Anyone know which species may be responsible for the recent roost
signatures on regional NEXRAD weather radar ? See message below.

Stan Kostka
lynnandstan at earthlink.net
Arlington WA


Assistance of birders requested in location and confirmation of
Purple Martin pre-migratory roosts in WA and OR.

Background
Purple Martins, a once fairly abundant colonial nesting swallow
species in decline on the west coast after the mid-1900s and rare by
the mid-1970s, have generally responded well to nest box based
recovery efforts in OR, WA and BC over the past 2-3 decades and with
recent major increases, populations in the PNW are now higher than
they have been in the past 40-50 years.

Martins have the characteristic behaviour of gathering in communal
pre-migratory night roosts at the end of the nesting season, which
serve as staging areas prior to migration. Roosting birds assemble at
dusk and disperse to forage at dawn. They migrate individually and in
small groups rather than as one large flock, so roosts may persist
for several weeks or more as the post-breeding population builds to a
peak in mid-late August and then declines rapidly as birds begin to
leave.

With the small recovering west coast martin population in recent
years, no significant roosts have been documented in the PNW since
the mid-1940s, prior to the overall decline, when they still nested
in city buildings and a roost of ~10,000 birds was reported in
Seattle. Any that may have occurred recently were likely small and
short lived, perhaps mixed with large numbers of starlings,
blackbirds or other songbirds, so easily overlooked.

Given the recent population recovery success (perhaps 10,000 - 15,000
birds including fledglings in the PNW at the end of the current
breeding season), martin recovery researchers are now beginning to
look for new pre-migratory roost sites, which are of interest for
study and documentation of movement of colour-banded birds and may be
important as target sites for conservation initiatives for the
continued recovery of this desirable at-risk native songbird species.
Limited ground searches so far have been unsuccessful, but martin
roosts also have a characteristic ephemeral "expanding ring"
signature on WSR-88D (NEXRAD) weather surveillance radar
(<www.intellicast.com>) as the roosting birds disperse in all
directions at dawn, detection of which has been well documented and
used for wide area survey, location and study of larger eastern
martin pre-migratory roosts.

Survey
We have thus been monitoring images from US west coast NEXRAD weather
radar sites in the dawn hours (6-7 am local time) recently, looking
for telltale daily-recurring expanding ring and ring segment
signatures of dispersing martin pre-migratory roosts, as well as
checking imaging of some known eastern martin roost sites at local
dawn (3-4 am here) for visual comparison and timing reference. The
PNW weather radar sites we are monitoring most closely west of the
Cascades are Seattle, Portland and Medford. These 3 stations provide
coverage of the southern Strait of Georgia, Puget Sound, Puget
Trough, lower Columbia R. Basin and western OR, where most west coast
martins nest and any pre-migratory roosts are most likely to occur.

We had no problem finding many known large eastern roost sites
easily, mostly at more northerly lattitudes near the Can/US border,
similar to the PNW and with correspondingly later seasonal nesting
timing, so roosts here should also be visible now in clear weather IF
they are large enough. We don't know the minimum roost size for a
radar-detectable signature, or if there are enough martins in the PNW
population yet to produce one. On the other hand, there are more
martins nesting in the PNW now than at any time in recent history (or
since WSR-88D weather radar became available), so it's a good time to
start looking.

Results (preliminary)
So far we have found 2-3 possible roost sites in WA and OR that
repeatedly showed complete or partial roost signatures (see attached
animation loop; subtract 7 hrs from Z/GMT for local time):

1): South of Portland near I-5, just north of Salem, OR. This site
consistently showed a characteristic expanding ring signature in the
6-7 am time window on 4 consecutive days, strongly suggesting a
martin roost.

2): South of Olympia, east of Centralia and Chehalis, WA, on or just
south of the border of Thurston and Lewis Counties. This site showed
a smaller partial ring signature expanding to the SE with the same
timing, also suggestive of a martin roost.

3): East of Kelso, WA, west of Mt. St. Helens. This site showed some
typical roost signature characteristics but was too obscured by
ground clutter to show clear expanding ring images in successive
radar frames, so may or may not be a roost.

There's no certainty yet these are martin roosts - they could contain
blackbirds, starlings or a mixture of both, with or without martins
(they may use mixed roosts when numbers are low, which might be why
we haven't noticed them earlier), or perhaps even a large crow or
pigeon roost. However, they display most/all of the radar
characteristics and timing of a martin roost and involve birds that
climb to sufficient altitude during dawn dispersal to appear up to 50
+ mi away on +0.5? elevation radar.

Confirmation
Therefore, we are requesting assistance from local birders for
location and/or confirmation of any such roost sites:

Is anyone aware of a known large communal roost of martins,
blackbirds, starlings, crows (or recent Purple Martin sightings) in
these areas, or the location of a known martin roost elsewhere? If
not, do you have contacts in any of these areas who can provide such
information? If so, please let me know via e-mail.

Martin night roosts are likely to be in trees or large wetland reed/
cattail stands or in a large man-made structure with lots of perching
space, such as a bridge or tower, possibly in a remote area and
usually close to a body of water. If martins are involved, they
should be visible and audible heading in to the roost site at low
altitude (literally below weather radar so not detected) in the
evening just before/at dusk, as well as leaving at first light, ~6:00
- 6:30 am, a little later on cloudy mornings.

Please note that sites where martins may perch for a time, feed or
bathe during the day but do not spend the night are also of some
interest but are NOT roost sites.

Any information or assistance with identifying martin communal roost
sites in these areas or elsewhere is much appreciated.

Colour leg band sightings
As well, if a martin roost site or mixed species roost including
martins is located or martins are observed perched elsewhere, some of
the birds may have coloured numbered leg bands from BC (gold, blue)
and/or WA (orange, green) which are easily visible with a spotting
scope and readable at close range. It would be of great interest to
know if any of these are present and if the colour band code is
obtained we can provide the origin location where the bird was banded
and (usually) fledged. Thanks.

Cheers, ...Bruce
______________________________________________________
Bruce Cousens, B.Sc., M.Sc., R.P.Bio., Senior Biologist,
BC Purple Martin Stewardship & Recovery Program Coordinator,
Georgia Basin Ecological Assessment and Restoration Society
#133 - 4176A Departure Bay Rd., Nanaimo, BC, Canada V9T 4V7
Phone/msg./fax: (250) 758-2922; E-mail: pmartins at island.net