Landscape features affect occupancy probability of Antrostomus vociferus (Eastern Whip-poor-will) in West Virginia managed forests
Data files
Dec 09, 2025 version files 22.20 KB
-
EWPW_occu.csv
10.55 KB
-
EWPW_veg_occu.csv
9 KB
-
README.md
2.64 KB
Abstract
Antrostomus vociferus (Eastern Whip-poor-will) have experienced a greater population decrease range wide than any other caprimulgid in North America. Hypotheses for this population reduction include conversion of open pasture to forest, increasing levels of urbanization, and the reduction of naturally occurring forest fires. Antrostomus vociferous forage over low forest canopies within early successional areas such as farmland intermixed with woody vegetation, forest patches recently cleared by fires and timber harvesting, and large artificial canopy breaks created by roads and power-line cuts. On institutional timber-harvest land in south-central West Virginia, we examined what site-specific and landscape features were positively associated with A. vociferus detection and occupancy probability. In June 2021 and 2022, we used autonomous recording units (ARUs) to estimate occupancy on a ~104,000 ha working forest. To estimate occupancy, we targeted surveys in regenerating clear-cut forest stands with ages ranging from recently harvested (2021) to ~20 years since harvest. Antrostomus vociferus detection probability was positively related to lunar illumination, increasing by 10% for every 20% increase in illumination. Occupancy at sites above 900 m a.s.l. elevation was below 40% and decreased with distance from riparian zones. Antrostomus vociferus occupancy decreased as tree diameter increased and occupancy was positively related to the amount of grass ground cover. Thus, in a timber-harvested landscape with forest heterogeneity, A. vociferus were negatively associated with increasing elevation but occurred near riparian areas with open grassy areas and smaller trees. Implications of this study suggest that the management of mosaic forests at certain elevational thresholds (approximately 800 m) will increase the probability of A. vociferus occupancy.
Dataset DOI: 10.5061/dryad.mpg4f4rc1
Description of the data and file structure
We have submitted our raw occupancy data (EWPW_occu.csv) which include presence/detection and landscape covariates. We have also included (EWPW_veg_occu.csv) which includes all vegetation-related covariates for all the same sampled sites.
Descriptions
Data set 1 - Detection/Occupancy
ARU_30(and 5) - data not used in the models but represent total (undivided) detection throughout the whole 30 minute recording periods.
interval(s)_1 and 2 - two possible intervals of detection for EWPW. Two intervals of time used for occupancy probability models. Intervals 1 and 2 represent the first and last 5 minutes respectively of the 30 minute recording periods.
ID - Forest stand ID
Rip - presence or absence of riparian within stipulated distance
age_years - stand age
basal - basal area of the given site
dist_rip - distance from riparian zone (transitional zone)
cor_date - Julian date
lab_date - labeled date
moon_perc - percent moon illumination
perc_esf - percent early successional forest
dom_type - dominant forest type considered by GIS files
dom_type_num - numerical equivalent of dom_type
min_sunset - minutes since sunset on that given day
obs_date - Date
Data set 2 - Vegetation
all vegetation covariates begin with "perc", representative of percentage cover of that particular vegetation cover type. All other covariates included in this dataset are also precent in Data set 1 (listed above).
perc_canopy - the total percent canopy cover at a given site
perc_grass - the total percent for grass ground cover at a given site
perc_forb - the total percent for forb ground cover at a given site
perc_solidago - the total percent for Solidago ground cover at a given site
perc_vine - the total percent for vine ground cover at a given site
perc_rubus - the total percent for Rubus ground cover at a given site
perc_shrub - the total percent for shrub ground cover at a given site
perc_sapling - the total percent for sapling ground cover at a given site ( < 8cm diameter at breast height)
elevation_m - the elevation (in meters) at a particular site
dist_edge_m - distance to a forest edge (where mature forest met either early-successional or recently cleared/logged) from the point and recording unit placement
Software used
Package "unmarked" (Fiske and Chandler 2011) in R version 4.1.1
