Jones and White * NEST SURVIVAL OF SPRAGUE'S PIPIT 
311 
TABLE 1. Mean distance (m) from Sprague Pipits nests to linear anthropogenic and natural features at Bowdoin 
National Wildlife Refuge, Montana, USA. 
Fealure 
Mean 
Range 
Primary road 
Secondary’ roads 
Agriculture field 
Railroad rights-of-way 
Shoreline 
313.6 
62.1-654.1 
201.7 
14.5-503.5 
251.3 
69.1-401.8 
357.4 
242.9-569.8 
387.5 
28.6-805.2 
158.64 
40 
124.43 
125 
94.91 
32 
102.70 
17 
175.10 
125 
addressed the effect on DSR of the distance from 
the nest to the habitat edges created by the linear 
features. 
METHODS 
Study Area.—This study was conducted during 
1997-2007 at Bowdoin National Wildlife Refuge 
iNWR) in Phillips County, near the town of Malta 
in north-central Montana. USA (48 24’ N. 107 
39’ W; - 750 m asl). The study area consisted of 
five permanent sites (26-59 ha) totaling 218 ha 
largely consisting of native mixed-grass prairie 
with flat to gently rolling terrain. The climate was 
continental and semiarid, and characterized by 
strong winds and high evaporation rates. The plant 
community was dominated by western wheatgrass 
(Puscopyruin smithii). needle-and-thread (Hesper- 
usttpa comma), blue grama (Boutehua gracilis), 
and spikemoss (Selaginella densa). Low shrubs 
(Sarcohalus vermiculatus, Artemisia catui, Cer- 
atoides lanaia) were widespread but sparse and 
trees were largely absent (Jones et al. 2010). 
Nest Searching .—Sites were searched for nests 
^-5 times per week from 10 May through - 15 
August in all years in an attempt to locate all 
active nests. Search techniques were primarily 
rope dragging (Davis 2003. Jones et al. 2010), but 
also included behavioral observations (Martin 
and Geupel 1993) and opportunistic fool flushes; 
details of the nest-searching techniques are 
reported in Jones et al. (2010). Nesting outcomes 
wer e: (I) successfully fledged at least one young 
°i the parental species, (2) depredated or aban¬ 
doned (j.e., eggs or nestlings left permanently 
unattended), or (3) unknown. Observations of 
fledglings in the natal territory within 3 days ot 
expected fledging, presence ol leces and leather 
scales in the nest, fledglings near the nest, and 
adults uttering alarm calls nearby or feeding new 
fledglings within 50 m of the nest were evidence 
of reproductive success. Predation was assumed 
when eggs or nestlings too young to fledge 
disappeared from the nest or the nest was 
destroyed. 
The locations of all Sprague's Pipit nests (n = 
125) were mapped. We used Arclnfo/GIS (ESRI 
2007) to map the non-grassland features of 
interest in and around Bowdoin NWR: (1) 
secondary road. (2) tertiary roads and trails, (3) 
agriculture field. (4) active railroad rights-of-way. 
and (5) the shoreline of Lake Bowdoin. We used 
Arclnfo/GIS (ESRI 2007) to measure the shortest 
distance from each nest to each ot these attributes. 
The secondary paved toad was a small, raised-bed 
two-lane farm road with low to moderate traliic 
volume. The vegetation along the side ot this road 
included u higher density of trees and shrubs than 
present in the surrounding grasslands. Also, along 
this road were ditches with moving water, fences, 
and rural farm development. The tertiary dirt 
roads or trails were administrative maintenance 
roads with dikes and low ditches on each side, and 
u public birding trail with a low volume of 
seasonal traffic. Ditches associated with roads and 
the railroad rights-of-way held water for an 
extended period, resulting in taller and denser 
vegetation than the adjacent native prairie. Exotic 
vegetation was more prevalent immediately 
adjacent to the secondary toad, mainly brome 
(liminus spp.) and thistle (Cirsium spp.). The 
railroad rights-of-way included a built-up rocky 
bed and exotic vegetation moving out from the 
tracks. The agriculture field was fenced and 
irrigated. The shoreline of Lake Bowdoin was 
an alkaline salt flat of crusted mud with little or no 
vegetation. The secondary road, agriculture field, 
and railroad rights-of-way were adjacent to one 
site; the tertiary roads and trails and shoreline of 
Lake Bowdoin were adjacent to all five sites 
(Table 1). 
Data Analysis .—We estimated the relationship 
of roads and other edges on Sprague's Pipits DSR 
using the nest survival program in Program 
MARK (White and Burnham 1999, Dinsmore 
