Bird Diversity & Management

Contributed Oral Presentations

Contributed paper sessions will be available on-demand for the duration of the conference, then again at the conclusion of the conference.

 

Satellite Image Textures Capture Multiple Features of Vegetation Heterogeneity Important for Predicting Patterns of Bird Richness
Laura S. Farwell; David Gudex-Cross; Ilianna Anise; Michael J. Bosch; Ashley M. Olah; Volker C. Radeloff; Elena Razenkova; Natalia Rogova; Eduarda M. O. Silveira; Matthew M. Smith; Anna M. Pidgeon
Spatial heterogeneity of vegetation has a strong influence on patterns of species diversity, but is challenging to measure across broad spatial extents using traditional field-based methods or costly lidar imaging. Satellite image texture analysis offers a promising and cost-effective alternative for capturing large-scale, continuous measures of vegetation heterogeneity. Our objectives were to identify features of vegetation heterogeneity that are well-captured by image textures, and to compare the performance of textures vs. lidar- and field-based vegetation metrics in models of bird richness. We calculated a suite of textures from 30-m (Landsat 8) and 10-m (Sentinel 2) resolution satellite imagery from 2017-2019. We compared textures with vegetation metrics and bird data provided by the National Ecological Observatory Network (NEON), from 27 field sites representing 16 ecoclimatic domains across the continental USA. Both 30-m and 10-m resolution textures were most strongly correlated with lidar-based canopy height variability (|r| = 0.64 and 0.80, respectively), and were also moderately correlated with multiple field-based metrics, including standard deviations of vegetation height and tree stem diameter, and foliage height diversity (range |r| = 0.31-0.52). Overall, 10-m resolution textures showed stronger correlations with lidar- and field-based metrics than 30-m resolution textures. In univariate linear models of total bird richness, 10-m resolution textures also held the most explanatory power (up to R2adj = 0.45), followed by 30-m textures (up to R2adj = 0.31). In global linear models including top-ranked predictors, both 30-m and 10-m resolution textures outperformed lidar- and field-based metrics, and contributed to explain 42% and 55% of the variance in total bird richness, respectively. In summary, satellite image textures effectively capture multiple elements of vegetation structural complexity, and are powerful predictors of bird richness. Image textures have exciting potential for improving biodiversity models, especially across broad spatial extents and in areas lacking lidar- or field-based vegetation data.
Effects of Grazing Management on Great Plains Bird Communities
Cameron A. Duquette; Torre J. Hovick; Ryan F. Limb; Jason P. Harmon; Benjamin Geaumont; Kevin K. Sedivec; Devan A. McGranahan
Conservation of a diverse Great Plains bird community depends not only on the preservation of remaining grassland area, but also on generating heterogeneous vegetation structure. Grassland birds have a variety of structural requirements for breeding, and grazing management can maximize bird diversity through the creation of diverse vegetation conditions while still maintaining cattle production goals in working landscapes. To determine the effect of grazing management on grassland breeding bird density and community composition, we conducted four years of line transect distance sampling from 2017 to 2020 across four grazing treatments (65 ha pastures) in Streeter, ND. Two were patch-burn grazing at a four-year fire return interval. We varied the size of burn patches by burning the entire 1/4 pasture in spring in one treatment, and 1/8 pasture in each of the spring and summer seasons in the other. We included a modified twice over rotational grazing treatment, designed to generate vegetation heterogeneity without fire. These were compared to a season-long grazing treatment, where cattle had free access to the entire pasture. All treatments were moderately stocked. NMDS ordination showed a high degree of community overlap between treatments, but patch burn communities were more diverse than season long grazing pastures, which were more diverse than modified twice-over pastures. Species-specific responses were variable. Generally, species with a preference for dense vegetation achieved highest densities in the rotational treatment, while species with a preference for short-statured vegetation or with variable needs achieved maximum density the patch-burn treatment. Cattle weight gain data from each treatment showed highest gains in the patch-burn treatments, followed by season-long and then rotational treatments. This suggests that varied management strategies are needed to benefit the full bird community and that heterogeneity-creating patterns of patch-burn grazing are compatible with management for grassland bird diversity and production goals in working landscapes.
Improving Habitat Suitability along Rights-Of-Way Corridors for Young Forest Avifauna
Eric L. Margenau; Petra B. Wood
Forest disturbance regimes have shifted over the past century reducing the amount of habitat available for young forest wildlife. Powerline and pipeline (hereafter “ROWs”) corridors, prominent features through the central Appalachian region, can offset this loss by providing habitat for young forest avifauna if they are managed to optimize these areas for young forest avifauna. We assessed linear tree cutting treatments (hereafter “cut-back borders”) adjacent to ROWs for managing young forest avifauna. We evaluated combinations of three harvest sizes (15 m [0.5 ha], 30 m [0.9 ha], and 45 m [1.4 ha] into the forest, perpendicular to forest edge) and two harvest intensities (14 m2/ha and 4.5 m2/ha basal area retention) throughout West Virginia. We monitored species- (i.e., abundances) and community-level (i.e., species richness) responses over three years (pre-treatment through two-year post-treatment) to assess which cut-back border combinations were most effective in managing young forest avifauna. Abundances and species richness were higher in most cut-back border treatments than the control at one-year and two-year post-treatment periods. Significant positive responses were primarily in the 15-m wide cut-back borders (both 14 m2/ha and 4.5 m2/ha harvest intensities), where American goldfinch (Spinus tritis), Carolina wren (Thryothorus ludovicianus), eastern towhee (Pipilo erythrophthalmus), indigo bunting (Passerina cyanea), and the young forest community all had higher abundances/richness at one-year and/or two-year post-treatment relative to pre-treatment levels. Additionally, abundance of indigo bunting was higher in the 45 m, 14 m2/ha treatment at one-year and two-year post-treatment. Common yellowthroat (Geothlypis trichas) was the only species we analyzed that did not exhibit significant short-term responses to cut-back border treatments. No species responded negatively to cut-back borders. Based on our results, cut-back borders appear to increase short-term habitat suitability along ROWs for young forest avifauna and provide another management tool for conserving young forest avifauna within ROW landscapes.
Assessment of Long-Term Breeding Density Responses of Cerulean Warblers to Timber Harvests in the Appalachian Region
Douglas Raybuck; David Buehler; Patrick Keyser; Than Boves; Jeffery Larkin; Amanda Rodewald
Even-aged, closed-canopy forests are prevalent across the Appalachian region today because of widespread logging during the first half of the 20th Century. Forest management prescriptions designed to diversify stand structural conditions are increasingly being recommended and applied with the goal of creating, maintaining, or improving habitat for declining, disturbance-adapted avian species. One declining disturbance-adapted species considered a species of conservation concern is the Cerulean Warbler (Setophaga cerulea; CERW). Millions of dollars have been allocated across the Appalachians to create or enhance CERW habitat. Although a few studies have examined short-term (1-3 years post-treatment) responses of CERW densities to partial timber harvest implementations and have documented positive short-term effects, longer-term effects are unclear. Our objective was to assess longer-term density responses by CERW to forest management prescriptions that mimic natural disturbances. In the 2019 and 2020 breeding seasons, we re-surveyed three sites in Ohio, Kentucky, and Tennessee (~20-ha study plots) in which basal area and overstory canopy were reduced in 2006 by 0%, ~20%, ~40% and ~75% in control, light, intermediate, and heavy harvests respectively. The plots had been surveyed for two years pre-harvest (2005-2006) and 4-years post-harvest (2007-2010). We followed the same spot-mapping methods as done previously and compared territory density estimates in harvests with those in adjacent buffers and with unmanaged control plots. In 2019, densities (territories/ha ± SE) were lesser or similar in the heavy treatment (0.22 ± 0.13) compared with adjacent buffer (0.38 ± 0.28), intermediate treatment (0.68 ± 0.08) compared with adjacent buffer (1.04 ± 0.12), and light treatment (0.96 ± 0.32) compared with adjacent buffer (0.94 ± 0.34). Control plots (0.79 ± 0.15) held densities similar with 2005-2010 estimates (0.83 ± 0.07). Results from May-June 2020 will be incorporated into the analysis and forest management recommendations for Cerulean Warblers will be presented.
Monitoring the Response of the Cerulean Warbler and Associated Species to Shelterwood Harvests in the Central Appalachian Mountains
Dakotah Shaffer; Darin J. McNeil; Luke Gray; Halie Parker; Jeffery T. Larkin; Joseph Duchamp; Michael Tyree; Jeffery L. Larkin
Cerulean Warbler (Setophaga cerulea) populations have experienced a steady annual decline over the past several decades. Part of this decline has been linked to reduced availability of quality breeding season habitat. Cerulean Warblers nest in mature deciduous forests with late- successional characteristics including structurally complex canopies comprised of preferred tree species used for foraging and nesting such as large white oak, hickories, and sugar maple. Since late-successional conditions are rare in eastern forests, silviculture (e.g. timber harvests) can be used to emulate conditions Cerulean Warblers require. In 2013, species-specific habitat guidelines were developed to reverse Cerulean Warbler declines. Studies on public lands have documented an immediate response by Cerulean Warblers to habitat management. Recognizing that eastern forests are largely privately-owned, a Regional Conservation Partnership Project (RCPP) was initiated through the Natural Resource Conservation Service (NRCS). In four years, >3,900 ha of private forestland in Pennsylvania and western Maryland were enrolled in this RCPP. From 2016-2019, we monitored sites (n=127) enrolled in the RCPP that were either shelterwood harvests (post-treatment) or scheduled shelterwood harvests (pre-treatment). Cerulean Warblers were only detected at 14 sites, of which 62% were post-treatment sites. Despite low occupancy, habitat management was successful in meeting recommended guidelines for basal area (14.5 m2/ha, 95% CI[13.5, 15.6]) and increasing the proportion of preferred trees from 37% (95% CI[33, 42]) on pre-treatment sites to 49% (95% CI[44, 54]) on post-treatment sites. However, average DBH of preferred trees (33.9 cm, 95% CI[33.2, 34.5]) in treated sites fell well below the recommended >40 cm threshold. Considering Cerulean Warblers display strong preference for large trees, it is not surprising that our sites had low occupancy. Ultimately, the habitat management actions taken on these sites have improved forest health, and with more time, the residual trees should reach the larger sizes preferred by Cerulean Warblers.
Assessing Priority Bird Response to Open Pine Management in Eastern Mississippi
Holly M. Todaro; Kristine O. Evans; Steven E. Lewis
Fire suppression, combined with lack of forest thinning and short-rotation monodominant management, has drastically altered the landscape in the southeastern U.S., leading to the loss of open pine ecosystems and associated bird species. Restoration of open pine ecosystems through forest thinning, herbicide application, and prescribed fire is a common practice; yet, there remains uncertainty regarding how vegetation structure resulting from management practices impacts priority open pine species. The objective of this research is to determine the range of tolerances for specific forest structural characteristics that are most advantageous to these declining species. To determine this, yearly point-transect surveys were conducted in five separately managed pine habitat units at Sam D. Hamilton Noxubee National Wildlife Refuge in eastern Mississippi. Using a combined distance and time removal sampling approach we assessed avian communities in each management unit from 2016-2019. We used a Bayesian hierarchical modeling framework to assess species associations with vegetation structural characteristics while accounting for availability and perceptibility of individuals. We determined that priority bird species abundance is associated closely with pine basal area and percent pine overstory canopy cover. Additionally, Bachman’s sparrow (Peucaea aestivalis), prairie warbler (Setophaga discolor), and northern bobwhite (Colinus virginianus) abundance was positively associated with percent cover of native grasses. Hardwood midstory cover was negatively associated with both priority and non-priority species. These results suggest that open pine management with forest thinning, followed by prescribed fire to reduce midstory and increased herbaceous ground cover enhances abundance of priority bird species as well as non-priority bird species by providing structural conditions preferred by open pine species.
Assessing Effects of Habitat Amount versus Configuration on Avian Diversity in Managed Pine Landscapes
Craig Andrew Sklarczyk
The large spatial footprint of pine forests managed for timber production in the Southeastern Coastal Plain ecoregion provides resources for resident and migratory bird species across a range of habitat. Typically, these working forests are managed as a matrix of different stand age classes, but little is known about how spatial patterning of those stands influence bird diversity and abundance. Given their extensive coverage and a growing demand for forest products, it is critical to enhance our understanding of how forest management activities influence avian diversity at the landscape scale to achieve conservation objectives. To determine optimal amount and proximity of habitat conditions, we examined breeding-season avian communities in regenerating and mid-late succession managed loblolly pine stands in east-central Mississippi from 2019-2020. We assessed species associations with local and landscape characteristics using a Bayesian hierarchical modeling framework which combined distance-sampling and time-removal methods to address availability and perceptibility. Our focal species showed varying associations with amount and proximity of habitat. Prairie Warbler and Northern Bobwhite exhibited a quadratic relationship with the amount of regenerating pine on the landscape, whereas Field Sparrow and Red-headed Woodpecker exhibited strong positive associations. 70% of our focal species showed slight to strong positive association with proximity of neighboring patches, although Northern Bobwhite and Field Sparrow exhibited a quadratic relationship with proximity of neighboring patches. Across bird guilds, proximity of habitat patches seemed to be the most influential on predicted abundance within our working pine forests. However, proportion of that habitat on the landscape should not be ignored for area-sensitive species. We suggest associations with habitat amount and configuration are species-specific, but necessarily community specific. By improving our understanding of how spatial patterning influences birds in working pine forests, managers can better ingrate conservation measures, particularly for at-risk species, while simultaneously achieving land use objectives.

 

Contributed Oral Presentations
Location: Virtual Date: Time: -