212 
THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 123. No. 2. June 2011 
TABLE 1. Microhabitat characteristic comparisons for Cerulean Warbler study plots in northern Alabama during th; 
2005 and 2006 breeding seasons. Values are means (± SE) for /-tests and median (interquartile range) for Mann-Whitne 
(/-tests of untransformed variables. Bold type denotes significant differences. P < 0.1. 
Micruhabitat characteristic 
Variable code Variable description 
Used (n = 52) Unused (n - 47) 
BA 
NTREE" 
RBATR" 
NSNAG 
SNAGBA 
RBASNAG 
DECBA 
TRDBH“ 
LOWHT* 
MIDHT 
UPHT 
CCVR 
CSTRC'' 
GAPDIST 
GAPSZ" 
UNDSTRY" 
SLOPE 
ASPECT 11 
Basal area of live trees (nr/ha) 
Number of live trees (ha) 
BArNTREE 
Number of snags (ha) 
Basal area of snags (nr/ha) 
SNAGBA:NSNAG 
Deciduous tree BA (%) 
Diameter nearest tree (cm) 
Lower canopy height (m) 
Mid canopy height (m) 
Upper canopy height (m) 
Canopy cover (%) 
Canopy structure 
Distance to nearest gap 
Size of nearest gap 
Understory density (stems/ha) 
Slope (degree) 
Aspect 
27.8 ± 1.5 
840.4 ± 45.6 
0.04 ± 0.003 
25.0 (25.0-68.8) 
0.6 (0.1—],9) 
0.01 (0.01-0.03) 
100.0 ( 100 . 0 - 100 . 0 ) 
22.0 ± 1.6 
5.8 ± 0.2 
17.0 ± 0.4 
29.4 (24.9-31.6) 
90.0 (83.1-95.0) 
0.72 ± 0.02 
2.0 (1.0-3.0) 
3.0 (1.0-4.0) 
771.6 ± 90.4 
8.0 (0.0-21.4) 
1.0 ( 1 . 0 - 1 . 3 ) 
“ Logarithmic or square root Iranstormalion. 
Shannon-Wiener diversity index. 
: as r: 5 *'-»>■ • <>» ->• 
Transformed: A’ = cost45 - A) + I (Beers et al. 1%6). 
25.6 ± 1.2 
1.305.3 ± 121.1 
0.03 ± 0.003 
50.0 (25.0-100.0) 
1.0 (0.3-2.4) 
0.02 (0.03) 
96.2 (77.8-100.0) 
13.5 ± 1.1 
5.2 ± 0.2 
15.5 ± 0.5 
26.2 (22.4-31.2) 
82.5 (77.5-90.0) 
0.75 ± 0.01 
2.0(1.0-5.0) 
1.0 (1.0-4.0) 
1.132.4 ± 92.4 
8.5 (0.0-14.0) 
1.0 (0.9-1.5) 
t-tesUU-ita 
df 
tfl 
r 
97 
1.13(f) 
0.26 
97 
-4.01 (/) 
0.01 
97 
3.56 (f) 
0.01 
97 
-1.51 (Z) 
0.13 
97 
—112 (Z) 
0.26 
97 
-0.51 (Z) 
0.61 
97 
-4.50 (Z) 
0.01 
97 
5.49 i/i 
0.01 
97 
1.69(f) 
0.10 
97 
2.58 (f) 
0.01 
97 
-1.68 (Z) 
0.09 
97 
-3.48 (Z) 
0.01 
97 
-1.64 (/) 
0.10 
97 
-0.71 (Z) 
0.48 
97 
-1.69 (Z) 
0.09 
97 
-3.23 (/) 
0.01 
97 
-0.64 (Z) 
0.52 
97 
-0.19 (Z) 
0.84 
TABLE 2. Principal components analysis of microhabi- 
tat characteristics from plots used and unused by Cerulean 
Warblers in northern Alabama during the 2005 and 2006 
breeding seasons. Factor loadings < 0.25 not displayed. 
Original variables 
RBATR 
BA 
TRDBH 
UPHT 
LOWHT 
MIDHT 
GAPSZ 
GAPDIST 
SNAGBA 
RBASNAG 
NSNAG 
DECBA 
NTREE 
UNDSTRY 
CSTRC 
CCVR 
ASPECT 
SLOPE 
Cumulative 
Rotated factor loadings' 
l7 - 2 2S.9 40.4 50.3 59.1 
“ Rotation ir 
varimax with Kaiser normalization. 
—3 cm diameter, and .sparse understory (Fig. 4). 
Positive yet weaker affiliations existed with 
components describing canopy cover and struc¬ 
ture, tree size (i.e., height, DBH, basal area) and 
density (i.e., basal area: number of stems), and 
snag density. The components derived from gap 
size and proximity to gaps, as well as aspect and 
slope, had strong negative associations with 
Cerulean Warbler presence. 
The best-supported model relating Cerulean 
arbler occurrence to microhabitat variables had 
an Akaike weight (w f ) of 0.87 (Table 3). Cerulean 
Warblers preferred flat bottomlands containing 
large, well-spaced deciduous trees, a moderately 
complex canopy structure, closer canopy gaps, 
and many smaller snags. 
DISCUSSION 
The closest associates of Cerulean Warblers 
were neotropical migratory species that breed near 
streams (Louisiana Waterthrush and Northern 
Parula. but also Belted Kingfisher) in moist 
woodlands and deciduous bottomland forests 
(Kentucky Warbler. American Redstart, and 
Acadian Flycatcher). Cerulean Warblers wem 
not related to species that typically favor xeric 
