AMERICAN DIPPERS NESTING NEAR JUNEAU, ALASKA 
Table 3 Vacancy of American Dipper Territories near Juneau, 
Alaska, in Relation to Survival over the Previous Winter and Nest 
Success 
Breeding 
season 
Apparent 
annual survival from 
previous year (n) a 
Probable 
nest success fa 
Vacant 
territories (n) c 
2004 
67% (12) 
91% 
13% (19) 
2005 
53% (40) 
94% 
19% (31) 
2006 
39% (56) d 
74% 
19% (31) 
2007 
37% (57) d 
92% 
48% (31) 
2008 
42% (19) d 
— 
32% (31) 
°n = number of banded birds alive the previous year. 
b Percent of nests fledging chicks, from Willson and Hocker (2008b). 
c n = number of territories whose occupancy was determined. 
d Survival following very cold winters 
next best model (Table 4), estimated that survival declined as a function of 
the number of cold (<-12 °C) days (Figure 4); this relationship was almost 
linear within the range of winters we observed and had a common pattern 
for both sexes. The second-best model estimated separate, parallel (on the 
logit scale) slopes for each sex; the third-best model had separate slopes 
for each sex. The estimated relationship between annual survival and the 
number of cold days is <{> = e 1 - 396 -o.0948x/(]_ + e i.396-o.0948x) ? where x is the 
number of cold days. The estimates (values of b) range from 0.654 (95% 
confidence interval 0.501-0.781) for years with 8 cold days to 0.324 (95% 
Table 4 Statistics for Selection among the Cormack-Jolley- 
Seber Models Fit to the Mark-Resighting Data for the American 
Dipper near Juneau, Alaska 
Model 0 
Parameters 
A QAICc b 
Model weight 0 
<(>(days), p(.) 
3 
0.00 
0.610 
<(>(days + sex), p{.) 
4 
1.98 
0.227 
<(>(days x sex), p(.) 
5 
4.03 
0.081 
<|>(year), p(.) 
6 
4.89 
0.053 
<t>(year + sex), p(.) 
7 
6.94 
0.019 
<j)(sex), p(.) 
3 
8.93 
0.007 
<j)(year + sex + days x sex), p(.) 
9 
11.27 
0.002 
<j)(year x sex), p(.) 
11 
15.13 
0.000 
<j)(year x sex + days x sex), p(.) 
13 
19.71 
0.000 
°<|) = survival probability; p = probability of a dipper being resighted; (days) indicates 
that the number of days <-12 °C is a factor in the model, (year) indicates that varia- 
tion in survival by year is a factor in the model; (sex) or (+ sex) or (x sex) indicates 
that variation by sex is a factor in the model; (.) indicates the parameter is constant 
across years and sexes. Statistics are based on c = 1.25. 
b Change in QAICc (a measure of model fit) among models. 
c Relative support for a model relative to other models considered (sum to 1). 
200 
