Brand and Noon • SHRUB-NESTING BIRDS IN A RIPARIAN CORRIDOR 
49 
productivity measures and draws information 
from the literature to assess source-sink status. 
The San Pedro River is generally considered one 
of the best remaining examples of lowland 
riparian cottonwood-willow woodland and forests 
in the region. We hypothesized that native 
cottonwood-willow and mesquite riparian forests 
would maintain higher productivity and serve as 
population sources, and saltcedar would serve as a 
sink. 
METHODS 
We used estimates from field data collection 
supplemented with information from the literature 
to assess reproductive parameters and source-sink 
status for three shrub-nesting species with ade¬ 
quate data: Yellow-breasted Chat, Abert’s Tow- 
hee, and Arizona Bell's Vireo. We collected field 
data at 23 sites on the San Pedro River during the 
1999-2001 field seasons. Study sites included 16 
areas within the San Pedro Riparian National 
Conservation Area (SPRNCA) managed by the 
Bureau of Land Management (BLM) and seven 
sites on privately-owned land north of the 
SPRNCA (Fig. 1). Cottonwood-willow forests 
are the predominant woody vegetation type along 
the San Pedro River floodplain with saltcedar 
occurring in the drier stretches. A second zone of 
riparian vegetation occurs beyond the flood- 
plain—river terraces on the San Pedro are 
vegetated mainly by mesquite forests. 
We searched for and monitored nests in 
cottonwood, saltcedar, and mesquite stands every 
2-5 days during the avian breeding season (10 
May to 20 Jul) at each site following BBIRD 
protocol (Martin et al. 1997, Brand et al. 2010). 
We recorded egg or nestling age, adult behavior, 
and nest status at each nest check. Ages were 
based on observed laying dates, hatch dates, and 
nestling size and development. We estimated 
clutch size by counting the number of host eggs 
following onset of incubation. We considered 
nests failed if they showed clear signs of failure 
(e.g., torn or fallen nest, broken eggs, dead 
nestlings) or if the nest was intact but eggs or 
nestlings disappeared >2 days before the expect¬ 
ed fledging date. We estimated the number of 
young fledged per nest by counting the number of 
nestlings observed within 2 days of the expected 
fledge date, and where there was direct observa¬ 
tion of fledging or indirect evidence such as 
flattened nest rim and fecal material on the rim or 
below the nest. 
The finite rate of population increase ( X ) is a 
key parameter used to assess population status 
and, in particular, whether a population is 
functioning as a population source or sink (Pull¬ 
iam 1988, Battin 2004). We used the finite rate of 
population increase (X) in the absence of immi¬ 
gration or emigration to assess the status of each 
species within each vegetation type and across 
vegetation types. We calculated X = Sa +S/-p by 
vegetation type where Sa is annual survival of 
adults, Sj is annual survival of juveniles, and (3 is 
seasonal fecundity, or the number of females 
produced per breeding female (Pulliam 1988). We 
estimated seasonal fecundity by vegetation type as 
(3 = nsf-a, where ns = % nest survival by vege¬ 
tation type,/= the mean number of female young 
produced per successful nest by vegetation type, 
and a = average number of nesting attempts per 
season (Pulliam 1988, Fauth 2000, Grzybowski 
and Pease 2005). We used estimates of ns for each 
species in each vegetation type, and across all 
vegetation types, from a companion study of the 
same populations (Brand et al. 2010; Table 1). 
That study used the method of Stanley (2000) to 
assess daily nest survival probability ( p ) and then 
calculated nest survival as the percentage of 
clutches that resulted in >1 fledged offspring 
equal to TOO, where dj and d 2 = average 
days in the incubation and nestling periods, 
respectively (Jehle et al. 2004). We estimated / 
from field data as the total number of fledglings 
per successful nest divided by two, assuming an 
equal sex ratio at hatching. We considered 
species- and vegetation-specific estimates of nest 
survival and number of female young fledged per 
successful nest to represent 3-year breeding- 
season averages since nests were monitored 
throughout the breeding season and we combined 
data across years. 
We did not measure the average number of 
nesting attempts by species per season on the San 
Pedro. Instead, we adopted estimates for each 
species from previous studies that followed all 
nesting attempts by a cohort of females though a 
breeding season. Based on previous studies, we 
assumed that Yellow-breasted Chat and Abert’s 
Towhee adult females nested 1.4 ± 0.1 (mean ± 
SE) and 3.9 ± 0.5 times on average per season, 
respectively (Thompson and Nolan 1973, Finch 
1984). Budnik et al. (2001) reported average 
number of nesting attempts separately for para¬ 
sitized versus non-parasitized pairs in recognition 
that birds parasitized by Brown-headed Cowbirds 
