630 
THE WILSON JOURNAL OF ORNITHOLOGY • Voi 124. No. 3. September 2012 
The Wilson Journal of Ornithology 124(3):630-633, 2012 
Eastern Screech-Owl Responses to Suburban Sprawl, Warmer Climate, and 
Additional Avian Food in Central Texas 
Frederick R. Gehlbach 1 
ABS I RACT.—Effects of suburban development 
(sprawl), concurrent climate and increasing avian food 
on a population of Eastern Screech-Owls (Megascops 
asio) were studied for three decades using nest boxes and 
natural tree-cavity nests in residential yards and adjoin¬ 
ing natural lores! in central Texas. The suburban climate 
was warmer by 5.7 C associated with suburbia's heat- 
island effect by the last decade of study. Nesting started 
earlier by an average of 4.5 days annually and fledgling 
productivity increased by 31,4%. Avian prey increased 
and contributed to 93% successful annual nests in a more 
stable population. Bird feeders and bird baths were likely 
enhancing factors at residences, where owls obtained 
prey and used bird-bath water for drinking and bathing. 
Received 13 September 2011. Accepted 6 March 2012. 
Suburban-nesting Eastern Screech-Owls (Mega- 
scops asio ) differ from rural populations in central 
Texas by living in a warmer climate, nesting earlier, 
and producing more fledglings (Gehlbach 2008). 
The effects of warming city climates have not been 
widely considered for raptors nesting in cities and 
suburbs of the United States until quite recently 
(e.g., Meyburg and Chancellor 1994. Bird el aL 
1996. Marzluff 2001, Bildstein 2008. Dunn and 
Winkler 2010). I evaluated the likely ‘heat-island* 
(Landsberg 1981) of increasing suburban develop¬ 
ment and its effects on Eastern Screech-Owls and 
their avian prey in residential and remaining natural 
habitat (cf Marzluff et al. 1998, Gehlbach 2005). 
METHODS 
Study Site .—Data were recorded annually 
during 1980-2009 in a 270-ha study plot at 
180-184 m elevation in Woodway, a suburb of 
Waco, McLennan County. Texas (Gehlbach 1988. 
2008). Suburban housing, other buildings, and 
road development replaced all but 48 ha (18%) of 
natural forest habitat during this study. Wooded 
residential yards with 6-15 mature deciduous 
and evergreen trees, open-canopy woodland, and 
closed-canopy riparian forest were used for 
nesting and toraging by screech-owls. 
7 '^r nt c,rBiolo &y- fUylor University, Waco. TX 
76798, USA: e-mail: Fred Gehlbach@Baylor.edu 
Data Collection and Analyses .—Twenty wood 
nest boxes were scattered randomly in the plot: 
L5-17 annually in wooded front and back yards 
and the rest in natural riparian forest. Boxes were 
7-12 m from the nearest house and street, and 3 m 
high on tree trunks with diameters equal to or 
greater than the 21-cm box width. Annually, (i to 
19 yard nests did not differ from 0 to 3 similar 
natural, tree-cavity nests (Gehlbach 1994). Used 
and potential nest sites were checked weekly or 
more often during the March-May nesting season, 
and all adult and feathered nestling occupants 
were marked with l ISGS leg bands. 
Suburban sprawl was measured as the cumula¬ 
tive number ol annual, city-issued building permiLs 
ol all types in the study plot and adjoining area 
within I km. Air temperature and precipitation were 
recorded weekly in a wooded yard adjoining a 6-ha 
private riparian forest preserve centrally located in 
the study plot. Nesting birds were counted weekly 
in residential yards and the preserve for data on 
avifauna! size, composition, possible competition, 
and potential food for screech-owls (Gehlbach 
2005). Use of avian prey was based on stored 
bodies and their remains in screech-owls' nests. 
Annual data were assembled by single and 
multiple decades for comparisons using factorial 
analyses of variance (F). while 30-year population 
and environmental relationships were evaluated with 
best-fit linear, second-, or third-order polynomial 
regressions. Comparative stability was considered 
with coefficients of variation (CVs). A Principal 
Components (PC) Analysis with orthogonal trans- 
iormalions and varimax solution identified signifi¬ 
cant population and environmental factors ranked by 
their individual sum of PC values. Numerical data 
are given only for significant (P < 0.05) values and 
summarized by means ± standard errors. 
RESULTS 
Population. —Most (71%) of the study's 367 
screech-owl nests were in wooded residential 
yards, and 48% of the 23 natural cavity nests were 
in yards. All sites were within 104 m of riparian 
