The Wilson Journal of Ornithology 123 ( 3 ): 521 - 535 , 2011 
LARGE-SCALE MOVEMENT AND MIGRATION OF NORTHERN 
SAW-WHET OWLS IN EASTERN NORTH AMERICA 
SEAN R. BECKETT 1 AND GLENN A. PROUDFOOT 1 2 
ABSTRACT—We used information compiled by the U.S. Geological Survey’s Bird Banding Laboratory and 
geographic information systems (GIS) analysis to identify trends in annual Northern Saw-whet Owl {Aegolius acadicus) 
movement across eastern North America. .Analysis of 81,584 Northern Saw-whet Owl banding events revealed a 
southbound annual fall migration front with peak banding activity occurring progressively later in the season as latitude 
decreases. Northbound owls comprised C9% of owls banded and recaptured elsewhere in the same season, and <5% were 
recaptured northbound >100 km from banding location. There was no relationship between banding latitude and adult-to- 
juvenile ratio. However, the proportion of adults versus juveniles banded was not uniform among banding stations, 
suggesting age-differentiated migration patterns may exist. Information from multiyear foreign recaptures revealed that 
1- a c of owls banded and subsequently recaptured at the same latitude in different years were recaptured <100 km from 
banding location. A similar trend was found in the Appalachian Mountains, the Great Lakes Basin, and the Atlantic 
seaboard. This indicates that Northern Saw-whet Owls may exhibit high migration route fidelity. These findings expand the 
Northern Saw-whet Owl information portfolio and illustrate the versatility of aggregate data sets as a tool for answering 
large-scale questions regarding migration. Received 22 August 2010. Accepted S February 2011. 
The Northern Saw-whet Owl (Aegolius acadi¬ 
cus) is a common but poorly-understood member 
of the North American forest fauna. Researchers 
first learned this species exhibited migratory 
behavior in 1906 when many washed up on the 
shores of Lake Huron after an autumnal storm 
(Saunders 1907). It is now widely recognized that 
large numbers of Northern Saw-whet Owls move 
south from breeding areas each fall, traveling as 
far south as Alabama, Louisiana, and northern 
Florida (Rasmussen et al. 2008). This autumn 
exodus is presumably undertaken to escape 
challenging winter conditions and to find a more 
stable resource base (Cheveau el al. 2004). 
D. F. Brinker and colleagues created Project 
Owlnet (www.projectowlnet.org) to network a 
small group of banding stations in eastern North 
America. Project Owlnet has grown into a 
nationwide organization for coordinating and 
standardizing Northern Saw-whet Owl (NSWO) 
banding methodology. Currently, >125 NSWO 
banding stations allied with Project Owlnet mon¬ 
itor this species’ migration annually (Huy 2010). 
These banding stations report time windows 
during which the majority of Northern Saw-whet 
Owls are caught in a season. These windows tend 
to occur later at southern stations than northern 
stations (Holroyd and Woods 1975, Weir et al. 
1980. Brinker et al. 1997). Banding efforts have 
also revealed that some Northern Saw-whet Owl 
' Department of Biology, Vassar College, Poughkeepsie, 
N Y 12604, USA. 
Corresponding author; e-mail: glproudfoot@vassar.edu 
populations have cyclical migration irruptions 
about every 4 years. These irruptions are likely 
due to periods when prey abundance is followed 
by scarcity, implied by exceptionally high num¬ 
bers of Northern Saw-whet Owls captured in the 
fall compared to ‘normal 1 years (Davis 1966, 
Brinker et al. 1997, Whalen and Watts 2002, 
Brittain et al. 2009). Banding information has 
begun to illuminate age-differentiated migration 
trends in Northern Saw-whet Owls. Juvenile owls 
may migrate earlier than adults in some areas, and 
die age ratio of banded owls varies greatly among 
years and locations (Paxton and Watts 2000, 
Stock et al. 2006, Brittain et al. 2009). 
Our knowledge of Northern Saw-whet Owl 
migration is clearly limited by the scale of 
previous research. Virtually all publications have 
been local or regional, often analyzing data from 
one or two banding stations. The only study in 
eastern North America using data from >six 
stations is 36 years old and limited by the number 
and distribution of banding stations available at 
that time (Holroyd and Woods 1975). Over 
160,000 Northern Saw-whet Owls have been 
banded since Holroyd and Woods (1975) pub¬ 
lished their findings. This rigorous banding effort 
has generated an extensive data base archived at 
the U.S. Geological Survey’s Bird Banding 
Laboratory (BBL) that has remained unexplored 
in eastern North America. 
Our objectives were to use the BBL data base to 
explore multiple questions. (1) Does the timing 
and direction of the migration front reported in 
regional studies exist across eastern North Amer- 
521 
