534 
THE WILSON JOURNAL OF ORNITHOLOGY • Vol 123, No. 3. September 2011 
or longitude. Areas of both high and low adult: 
juvenile ratios occur in the northern and southern 
extents of our analyses (Fig. 7), The lowest adult: 
juvenile ratios were in the two northernmost 
latitude bars, but the highest ratios were in the 
next three adjacent bars (Fig. 6). By-year regres¬ 
sion analysis showed no consistent relationship 
between age ratio and latitude. Only 3 of 10 years 
tested had a significant trend, and these were not 
in a uniform direction (Table 2). Tests of mean 
banding latitude by age were not significant for all 
years, and they were not in a uniform direction 
(Table 3). The direction or significance of these 
tests is not explained by irruption or non-irruption 
years. The small mean difference in migration 
timing between adults and juveniles at each 
latitude bar does not suggest differential migration 
timing by latitude. 
We did not detect movement patterns explained 
by latitude, but our interpolations indicate that 
adult versus juvenile migration is non-uniform 
across eastern North America (Fig. 7). Juveniles 
in Idaho had lower body condition scores than 
adults (Stock et al. 2006). Along the Atlantic 
Coast, juveniles may arrive on the Delmarva 
Peninsula almost 2 weeks earlier than adults 
(Paxton and Watts 2000). Juveniles also benefit 
less from site familiarity than their parents, and 
may be less inclined to remain near breeding areas 
(Cote et al. 2007). These studies imply age-related 
differences in the ability to cope with challenging 
conditions, and suggest that juveniles may migrate 
differently than adults in some areas as a result. 
The predicted areas of high and low adult- 
juvenile ratios were similar in both irruption and 
non-irruption years and support the hypothesis of 
age-specific preferences for migration routes or 
wintering sites (Fig. 7B. C). Regionally variable 
forest structure, pmy availability, or climate may 
influence these preferences, and may explain the 
patchiness observed in our interpolations. It is 
possible that resident populations of Northern 
Saw-whet Owls at high latitudes or high eleva¬ 
tions in the Appalachian Mountains are influenc¬ 
ing these interpolations (Rasmussen et al. 2008). 
However, the migration timing interpolation 
(Fig. 3) and directionality analyses (Fig. 5) 
showed no clear evidence of resident owls 
obscurmg the overall observed migration pattern 
suggesting the impact of residents on the overall' 
data set is minimal. 
iJt qUali,y °f any imcr P° ,ati °n is limited by 
accuracy and distribution of sampling point's 
across the surface. Thus, we refrain from 
interpreting interpolation results in areas with 
low station density or areas influenced heavily by 
one data point. Increased banding efforts in 
regions with low station density will greatly 
improve our understating of large-scale migration 
patterns in this species. 
This study is an example for assessing the 
strength and versatility of using the BBL's large 
banding data base to understand bird migration. 
The kinds of information that can be gleaned from 
banding studies may be limited compared to other 
techniques, but banding is one of the only tools 
available for studying cryptic or nocturnal species 
like Northern Saw-whet Owls. We 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 by assessing movement patterns beyond 
published regional trends. 
ACKNOWLEDGMENTS 
Wc thank M. A. Cunningham of the Department of Earth 
Science and Geography at Vussar College for support in 
designing the analyses. This project would not have been 
possible without data generated by hundreds of Northern 
Saw-whet Owl banders who submit their information to the 
USGS Bird Banding Laboratory: information provided b> 
D. I-. Brinker of Project Owlnct, David Okines of Prince 
Edward Point Bird Observatory. David Evans of the Hawk 
Ridge Nature Reserve. Eugene Jacobs of Linwood Springs 
Research Station. Jon McCracken of Thunder Cape Bird 
Observatory, Bruce Murphy of Hilliardton Marsh, and Scott 
Wcidcnsaul of The Ned Smith Center for Nature and Art 
account lor 47^. of all banding data used in this study. We 
thank Jennifer Pontius of the University of Vermont for 
support in statistical analysis. Logistical suppon was 
provided by Daniel Bystrak of the Patuxent Wildlife 
Research Center and associates of Project Owlnet. We 
thank the Vassar College Environmental Research Institute 
for financial support of the project. 
literature cited 
Alerstam. T.. M. Hake, and N. KjellEn. 2006. Temporal 
and spatial patterns of repeated migratory journeys by 
Ospreys. Animal Behaviour 71 ;555-566. 
Bowman. J.. D. S. Baozdcski, and R. J. Brooks. 2010. 
The numerical response of breeding Northern Saw- 
whet Owls Aetfolius acadicus suggests nomadism 
Journal of Ornithology 151:439-506. 
Brinker, d. F„ k. E. Duffy, D. m. Whalen, B. D. 
Watts, and K. M. Dodge. 1997. Autumn migration 
of Northern-Saw-whet Owls [Ae^olius acadicus) in the 
middle Atlantic and northeastern United States: what 
observations Ironi 1995 suggest. Pages 74-89 in 
Biology and conservation of owls of the Northern 
Hemisphere: Second International Symposium (J- R- 
