Rodney B. Siegel

Habitat use and selection by California spotted owls in a postfire landscape.

Abstract Forest fire is often considered a primary threat to California spotted owls (Strix occidentalis occidentalis) because fire has the potential to rapidly alter owl habitat. We examined effects of fire on 7 radiomarked California spotted owls from 4 territories by quantifying use of habitat for nesting, roosting, and foraging according to severity of burn in and near a 610-km2 fire in the southern Sierra Nevada, California, USA, 4 years after fire. Three nests were located in mixed-conifer forests, 2 in areas of moderate-severity burn, and one in an area of low-severity burn, and one nest was located in an unburned area of mixed-conifer–hardwood forest. For roosting during the breeding season, spotted owls selected low-severity burned forest and avoided moderate- and high-severity burned areas; unburned forest was used in proportion with availability. Within 1 km of the center of their foraging areas, spotted owls selected all severities of burned forest and avoided unburned forest. Beyond 1.5 km, there were no discernable differences in use patterns among burn severities. Most owls foraged in high-severity burned forest more than in all other burn categories; high-severity burned forests had greater basal area of snags and higher shrub and herbaceous cover, parameters thought to be associated with increased abundance or accessibility of prey. We recommend that burned forests within 1.5 km of nests or roosts of California spotted owls not be salvage-logged until long-term effects of fire on spotted owls and their prey are understood more fully.

Mapping migration in a songbird using high-resolution genetic markers

Neotropic migratory birds are declining across the Western Hemisphere, but conservation efforts have been hampered by the inability to assess where migrants are most limited—the breeding grounds, migratory stopover sites or wintering areas. A major challenge has been the lack of an efficient, reliable and broadly applicable method for measuring the strength of migratory connections between populations across the annual cycle. Here, we show how high‐resolution genetic markers can be used to identify genetically distinct groups of a migratory bird, the Wilsons warbler (Cardellina pusilla), at fine enough spatial scales to facilitate assessing regional drivers of demographic trends. By screening 1626 samples using 96 highly divergent single nucleotide polymorphisms selected from a large pool of candidates (~450 000), we identify novel region‐specific migratory routes and timetables of migration along the Pacific Flyway. Our results illustrate that high‐resolution genetic markers are more reliable, precise and amenable to high throughput screening than previously described intrinsic marking techniques, making them broadly applicable to large‐scale monitoring and conservation of migratory organisms.

Novel statistical methods for integrating genetic and stable isotope data to infer individual-level migratory connectivity

Methods for determining patterns of migratory connectivity in animal ecology have historically been limited due to logistical challenges. Recent progress in studying migratory bird connectivity has been made using genetic and stable‐isotope markers to assign migratory individuals to their breeding grounds. Here, we present a novel Bayesian approach to jointly leverage genetic and isotopic markers and we test its utility on two migratory passerine bird species. Our approach represents a principled model‐based combination of genetic and isotope data from samples collected on the breeding grounds and is able to achieve levels of assignment accuracy that exceed those of either method alone. When applied at large scale the method can reveal specific migratory connectivity patterns. In Wilsons warblers (Wilsonia pusilla), we detect a subgroup of birds wintering in Baja that uniquely migrate preferentially from the coastal Pacific Northwest. Our approach is implemented in a way that is easily extended to accommodate additional sources of information (e.g. bi‐allelic markers, species distribution models, etc.) or adapted to other species or assignment problems.

DYNAMICS OF BREEDING-SEASON SITE OCCUPANCY OF THE CALIFORNIA SPOTTED OWL IN BURNED FORESTS

Abstract. Understanding how habitat disturbances such as forest fire affect local extinction and probability of colonization—the processes that determine site occupancy—is critical for developing forest management appropriate to conserving the California Spotted Owl (Strix occidentalis occidentalis), a subspecies of management concern. We used 11 years of breeding-season survey data from 41 California Spotted Owl sites burned in six forest fires and 145 sites in unburned areas throughout the Sierra Nevada, California, to compare probabilities of local extinction and colonization at burned and unburned sites while accounting for annual and site-specific variation in detectability. We found no significant effects of fire on these probabilities, suggesting that fire, even fire that burns on average 32% of suitable habitat at high severity within a California Spotted Owl site, does not threaten the persistence of the subspecies on the landscape. We used simulations to examine how different allocations of survey effort over 3 years affect estimability and bias of parameters and power to detect differences in colonization and local extinction between groups of sites. Simulations suggest that to determine whether and how habitat disturbance affects California Spotted Owl occupancy within 3 years, managers should strive to annually survey ≥200 affected and ≥200 unaffected historical owl sites throughout the Sierra Nevada 5 times per year. Given the low probability of detection in one year, we recommend more than one year of surveys be used to determine site occupancy before management that could be detrimental to the Spotted Owl is undertaken in potentially occupied habitat.

BIRD COMMUNITIES IN THINNED VERSUS UNTHINNED SIERRAN MIXED CONIFER STANDS

Abstract We used point counts and nest monitoring to compare avian community composition and nesting success in thinned and unthinned stands of commercially managed Sierran mixed conifer forest. We conducted point counts and monitored 537 active nests of 37 species on 10 study plots during three consecutive breeding seasons in the northern Sierra Nevada. All 10 study plots had experienced a similar long term management history that included fire suppression and single-tree selection logging, but five of the plots also underwent a protocol of combined commercial and biomass thinning 5–8 years prior to the beginning of the study. Pooling species by nest substrate, we found that detections of ground-nesting bird species were similar on thinned and unthinned plots, but we detected canopy-, cavity-, and especially shrub-nesting species much more frequently on the thinned plots. Nest success rates were not statistically different between thinned and unthinned plots for ground-, shrub-, canopy-, or cavity-nesting species. Thinned stands were characterized by significantly less canopy cover, significantly lower density of small and medium conifers, and significantly greater understory cover and deer brush (Ceanothus integerrimus) cover than the unthinned stands. We surmise that the thinning protocol stimulated vigorous shrub growth, and conclude that forest conditions associated with a relatively open canopy and a well-developed shrub understory are highly beneficial to numerous breeding bird species in the Sierran mixed conifer community, including many species that may not nest or forage in the understory. Forest thinning that promotes vigorous shrub growth may correlate with an increased abundance of nesting birds, at least within stands affected by historical fire suppression and single-tree selection logging.

USING POINT COUNTS TO ESTABLISH CONSERVATION PRIORITIES: HOW MANY VISITS ARE OPTIMAL?

Abstract We conducted point counts three times during the 1994 breeding season at 48 stations across the northwestern United States, and used cumulative totals from the three visits to rank the sites by two potential indices of conservation value: species richness and overall abundance of birds. We then recalculated each of the indices (1) using data from only a single visit to each site and (2) using data from only two visits. Rankings based on only one or two visits revealed that eliminating one, and even two of the visits had relatively minor effects on species richness rankings but affected rankings based on overall abundance more substantially. We also evaluated how effectively one or two visits to each site detected particular species of management concern. We conclude that when resources are limited, species richness based on point counts conducted during just one or two visits to potential conservation sites may provide a reliable index for prioritizing conservation efforts. When the primary objective is to determine the presence or absence of a particular species, however, at least two visits may be warranted. Finally, we conclude that, in general, researchers must be careful when using overall abundance as an index for establishing conservation priorities, as values may fluctuate substantially throughout the season.

Occupancy modeling of Black-backed Woodpeckers on burned Sierra Nevada forests

The Black-backed Woodpecker (Picoides arcticus) has been designated by the USDA Forest Service as a management indicator species for snags in burned conifer forests of the Sierra Nevada of California, USA. However, little is known about the characteristics that affect between-fire and within-fire habitat selection by the species in the region. Here we report on the first broad-scale survey of Black-backed Woodpeckers on wildfire-affected forests of the Sierra Nevada. We implemented a Bayesian hierarchical model to: 1) estimate Black-backed Woodpecker occupancy probability in fire areas burned within a time window of 1–10 years; 2) identify relationships between occupancy probability and habitat covariates (fire age, change in canopy cover pre-to-post fire, and snag basal area), elevation, and latitude; and 3) estimate detection probability and relate it to survey interval length and survey type (passive v. broadcast). We included random fire-area effects in our model of occupancy probability to accommodate clusters of non-independent points surveyed within the larger set of fire areas. Mean occupancy probability was estimated to be 0.097. Elevation (after controlling for latitude) had the strongest effect on occupancy probability (higher occupancy at higher elevation) followed by latitude (higher occupancy at northerly sites). Fire age was also important; occupancy probability was about 4× higher on the youngest compared to oldest fires. Although the direction of regression coefficients were in the expected direction (positive), snag basal area and canopy cover change were of minor importance in affecting occupancy probability. There was some indication, however, that the importance of snag basal area increased with fire age. Weak links between occupancy and canopy cover change suggested the species uses a range of burn severities, and such heterogeneity may promote habitat longevity. Our estimate of overall detection probability (across all survey intervals) was 0.772. We found strong effects of survey interval length (higher for longer interval) and, especially survey type (higher for broadcast survey) on detection probability. Our modeling framework and implementation illustrates the flexibility of the Bayesian hierarchical approach for handling complexities such as estimating derived parameters (and variances) and modeling random effects, and should prove generally useful for occupancy studies.

Variation in home-range size of Black-backed Woodpeckers

ABSTRACT The Black-backed Woodpecker (Picoides arcticus) is a species of conservation concern that is strongly associated with recently burned forests. Black-backed Woodpeckers are known to have variable home-range sizes, yet the ecological factors related to this variation have not been adequately explored and may hold insights into the natural history of the species and the management of its habitat. During 2011 and 2012, we radio-tracked Black-backed Woodpeckers nesting in 3 forested areas of California that burned between 2 and 5 years before the initiation of tracking. Among 15 individuals with robust tracking data, we found that home-range size varied by an order of magnitude, from 24.1 to 304.1 ha, as measured by movement-based kernel estimation. Using an information-theoretic approach, we evaluated the functional relationship between snag basal area—an a priori key resource—and home-range size, additionally controlling for sex, age, and years since fire as covariates. We found that snag basal area alone best predicted home-range size, explaining 54–62% of observed variation. As snag basal area increased, home-range sizes exponentially decreased. This relationship held true both with and without the inclusion of 3 individuals that nested in burned forest yet foraged predominantly outside the fire perimeter in unburned forest. Snag basal area, unlike other potential influences on home-range size, is an attribute that forest managers can directly influence. We describe a quantitative relationship between home-range size and snag basal area that forest managers can use to predict Black-backed Woodpecker pair density in burned forests and assess the likely population consequences of specific harvest treatments. Given that the birds in our study, foraging primarily in burned forest, all had home ranges with an average snag basal area ≥17 m2 ha−1, this may represent a benchmark for minimum habitat needs in postfire stands.

Molt patterns and age and sex determination of selected southeastern Cuban landbirds

Abstract Data from a banding station at Guantanamo Bay and the examination of 830 specimens were used to study molt patterns and criteria for determination of age and sex in 15 resident species of Cuban landbirds. All 15 species undergo their prebasic molt in August to November, with a few species commencing as early as May and/or completing as late as December. This timing corresponded with that of related taxa found in North America. With few exceptions, the extent of the first prebasic molts was also comparable to those of related taxa found at northern latitudes, being partial to incomplete in all nine passerines and four of six nonpasserines. Three species showed evidence of eccentric primary molt patterns during the first prebasic molt, to be expected in birds residing in scrubby or exposed environments such as those found at Guantanamo Bay. Determination of age is possible in most Cuban landbirds based on molt limits and the shape and condition of the primary coverts and rectrices, criteria very similar to those found in North American species of related taxa.

An integrated occupancy and space‐use model to predict abundance of imperfectly detected, territorial vertebrates

Summary It is often highly desirable to know not only where species are likely to occur (i.e. occupancy) but also how many individuals are supported by a given habitat (i.e. density). For many animals, occupancy and density may be determined by distinct ecological processes. Here we develop a novel abundance model as the product of landscape-scale occupancy probability and habitat-scale density given occupancy. One can conceptualize our model as fully packing a landscape with home ranges or territories based on habitat quality, and then subtracting territories based on a probabilistic process that accounts for the fact that species rarely exhibit full occupancy across heterogeneous landscapes. The model is designed to predict abundance at fine spatial scales, using resolutions equal to or smaller than a single home range or territory. We demonstrate this model on the Black-backed Woodpecker (Picoides arcticus), a species of management concern linked to post-fire forests. Occupancy is derived from a regional monitoring effort, while density given occupancy comes from a telemetry study of variation in territory size. A Bayesian framework is used to combine independent occupancy and home-range size models and predict abundance of Black-backed Woodpeckers at 4 fires that burned in 2012 or 2013. Predictions are evaluated with independently collected survey data, showing that the model is successful at predicting both absolute abundance at fires as well as relative abundance within and among fires. The conceptual model presents a promising new framework for fine-scale modelling of density and abundance for other territorial yet elusive species. Telemetry and occupancy data are widely collected for many species, but rarely utilized in combination, and the ecological exploration of the factors that determine occurrence versus home-range size may provide useful biological insight. As applied to the Black-backed Woodpecker, the model provides a tool for resource managers to explore trade-offs in retaining burned forest habitat versus managing for other post-fire goals, such as salvage logging or reforestation efforts that require snag removal.