David W. Willey

Effects of habitat fragmentation on avian nesting success: a review of the evidence at multiple spatial scales

We reviewed published literature to examine the effect of habitat fragmentation on avian nesting success at three spatial scales (i.e., edge, patch, and landscape scales). We identified 86 relevant manuscripts that provided 117 individual tests of hypotheses regarding the effects of habitat fragmentation on nesting success. Most papers and reviews on this topic have been narrow in scope and have not examined multiple spatial scales. However, our results suggest that the scale at which fragmentation is measured (i.e., edge, patch and landscape) and the duration of the studydo influence the probabilitythat a studywill detect a fragmentation effect. Studies that measured fragmentation at landscape scales and studies conducted over several years were more likely to detect effects of fragmentation on nesting success. A recent review of research on nest predators and habitat fragmentation found a verysimilar scale-dependent pattern; predator effects were more prevalent when fragmentation occurs at landscape scales than patch or edge scales. Based on these findings, we recommend future research on the topic should be conducted at the landscape scale, over several years, and incorporate accompanying work on nest-predator ecology. Correspondingly, conservation actions that limit fragmentation at landscape scales should have positive impacts on nesting success rates and bird populations. # 2003 Elsevier Ltd. All rights reserved.

BREEDING SEASON HOME RANGE AND HABITAT USE OF MEXICAN SPOTTED OWLS (STRIX OCCIDENTALIS LUCIDA) BELOW THE SOUTH RIM OF GRAND CANYON NATIONAL PARK

ABSTRACT We studied breeding season home range characteristics and habitat of paired male Mexican Spotted Owls (Strix occidentalis lucida) below the south rim of Grand Canyon National Park from 2004–2005. Adult male owls (n = 5) were captured and radio-tracked using tail-mounted VHF transmitters. We used minimum convex polygons and 90% fixed kernels to estimate breeding season home range size (mean = 355 ha and 372 ha, respectively). We also generated adaptive kernel home range estimates to describe areas of concentrated use within home ranges. Home ranges were located in the upper reaches of relatively narrow rocky canyons, and Spotted Owls showed limited use of adjacent forested plateaus. We conducted an analysis of habitat use and selection at two scales and found that owls selected (i.e., used disproportionate to availability) limestone cliffs present in their home ranges. Home ranges were approximately centered on nest and associated roost sites located on limestone cliffs within canyons. Our results contrasted with observations in Utah where spotted owls nested primarily on sandstone cliffs. In Grand Canyon, both sandstone and limestone cliffs were present in the home range, but limestone appeared to be the preferred substrate. At the landscape level, owls placed home ranges in areas dominated by piñon-juniper (Pinus edulis – Juniperus monosperma) woodland. We delineated 40 ha use-areas around nest sites and found that these conservation zones closely approximated adaptive kernel 30% isopleths, thus supporting core area designation of the Mexican Spotted Owl Recovery Plan.

Roost Habitat of Mexican Spotted Owls (Strix occidentalis lucida) in the Canyonlands of Utah

ABSTRACT In large portions of their geographic range, Mexican Spotted Owls (Strix occidentalis lucida) roost in forest-dominated environments, but in some areas the owls use relatively arid rocky canyonlands. We measured habitat characteristics at 133 male roosts (n = 20 males) during 1992–95, and 56 female roosts (n = 13 females) during 1994–95 in canyon environments of southern Utah. Across all years, 44% of Mexican Spotted Owls’ roosts were located in small stands of mixed-conifer forest, 30% in desert scrub vegetation, 16% in pinyon-juniper woodlands, and 10% of roosts were in riparian habitat. Roost sites were located in canyons composed of cliff-forming geologic formations. The width of canyons measured at roosts averaged 68.6 m (8.2 SE), and ranged from 1–500 m. The mean height of cliffs at roost sites was 77 m (10.9 SE) and ranged from 6–411 m. Roosts were located at caves and ledges (46% of all roosts) or in various tree species (54%). Roost height above ground averaged 9.5 m overall (1.1 SE), with mean tree roost height of 3.7 m (0.2 SE), and cliff roost height equal to 17.2 m (2.2 SE). For both males and females, coniferous trees species were used most frequently (64%), primarily Douglas-fir (Pseudotsuga menziesii), white fir (Abies concolor), and Utah juniper; and 36% of roosts occurred in deciduous trees, including big-tooth maple (Acer grandidentatum), boxelder (A. negundo), and Utah serviceberry (Amelanchier uthaensis). Canopy cover of tress at roosts ranged from 44–71%, mean tree height of trees present was 9.5 m and mean diameter of trees was 25.4 cm. Upland habitats that were not used for roosting were warmer, not as steep, possessed fewer caves and ledges, and trees present were of smaller stature than trees present in roost habitat.

HOME RANGE CHARACTERISTICS OF MEXICAN SPOTTED OWLS IN THE RINCON MOUNTAINS, ARIZONA

Abstract We studied a small isolated population of Mexican Spotted Owls (Strix occidentalis lucida) from 1996–1997 in the Rincon Mountains of Saguaro National Park, southeastern Arizona, USA. All mixed-conifer and pine-oak forest patches in the park were surveyed for Spotted Owls, and we located, captured, and radio-tagged 10 adult birds representing five mated pairs. Using radio-telemetry, we examined owl home range characteristics, roost habitat, and monitored reproduction within these five territories. Breeding season (Mar–Sep) home range size for 10 adult owls (95% adaptive kernel isopleths) averaged 267 ha (±207 SD), and varied widely among owls (range 34–652 ha). Mean home range size for owl pairs was 478 ha (±417 ha SD), and ranged from 70–1,160 ha. Owls that produced young used smaller home ranges than owls that had no young. Six habitat variables differed significantly between roost and random sites, including: percent canopy cover, number of trees, number of vegetation layers, average height of trees, average diameter of trees, and tree basal area. Radio-marked owls remained in their territories following small prescribed management fires within those territories, exhibiting no proximate effects to the presence of prescribed fire.