Greater Sage-grouse brood-rearing female habitat data
Data files
Sep 28, 2023 version files 13.01 MB
-
Data_wlb3.01111_Kirol_Fedy_2023.csv
-
README.md
Abstract
Habitat selection analyses conducted at an individual level may reveal patterns in selection not apparent when individuals are pooled in population-level approaches. Using GPS transmitters that gather high-resolution location data, we explored fine-scale habitat selection and space use within home ranges of female greater sage-grouse (Centrocercus urophasianus) that raised young (brood-rearing sage-grouse) in an oil and gas development area. To evaluate fine-scale habitat selection of brood-rearing sage-grouse, we used a two-stage approach. First, we developed models for each individual (i.e., individual-level modeling) and evaluated individual-level responses to modified habitats and infrastructure. Second, we averaged individual-level estimates using a bootstrap approach to make population-level inference. The average home range size during brood-rearing, from nest hatch to six weeks, in our study were 0.85 ± 0.21 km2. Individual- and population-level results indicated that brood-rearing females consistently selected for natural vegetation and avoided disturbed surfaces at a fine spatial scale. Our study area included substantial areas of recent (≤10 years) habitat reclamation which females also avoided. Visible power lines consistently led to avoidance behavior. In addition to consistent patterns of habitat selection, our individual models demonstrated variability and contrasting behaviors in how brood-rearing females responded to specific infrastructure features and anthropogenic water bodies. At the population-level, anthropogenic water bodies were avoided, but at the individual-level, the intensity of avoidance was variable among individuals. Individual variability was often explained by the age of the brood-rearing female (first year or adult). First year females were more likely than adults to use habitats close to infrastructure and consistently established home ranges in areas with more surface disturbance and infrastructure when compared to adults. Our results provide new insights into fine scale habitat-selection strategies used by female sage-grouse with broods in an area were oil and gas infrastructure is widespread and cannot be avoided.
README
Data_Kirol_Fedy 2023_Brood_rearing_sage_grouse_anthropogenic_landscape
This folder contains the data for the article:
Kirol and Fedy 2023 Using individual-based habitat selection analyses to understand the nuances of habitat use in an anthropogenic landscape: A case study using greater sage-grouse trying to raise young in an oil and gas field.
Comments and requests should be addressed to Chris Kirol: chriskirol@gmail.com. Please let me know if you intend to use these data.
The data for this article is a .csv file titled "Data_wlb3.01111_Kirol_Fedy_2023"
We followed a use versus availability design. We used weighted Generalized Linear Models (GLM) to compare use locations with pseudo-absence locations for individual-level models for each brood-rearing sage-grouse. We generated available points (i.e., pseudo-absence) at a ratio of 20:1 to used points for each individual. We established available points within a specific availability domain for each individual. The availability domain was based on the maximum linear distance across that individual’s home range which corresponds directly to that individual’s movements during the brood-rearing period.
- "Bird_name" is the bird id for that female's use locations and corresponding available locations.
- "ID" is the unique identifier for each point, use and available, for each female. "step_id" identifies the use or available point sequentially for each female.
- "case_" and "type_" identify the used locations as case_ = "TRUE" or type_ = "1" and available locations case_ = "FALSE" or type_ = "0". In our GLMM models "case_" is the response variable (y).
- "t" is the date and time of the location which is NA for available locations.
- "year" is the year of the location and corresponding available locations.
- "hr_c_dist" was not assessed as a predictor in our models but is a descriptive value describing the distance of that point to the center of that individual's home range.
- The remaining columns are all predictor variables (x) assessed in our models. For a detailed description of each variable please see Table 1 in the manuscript. Environmental variables include:
- "VRM" is vector roughness measure
- "Slope" that is the slope (%)
- "Sage" that is sagebrush cover (%)
- "ShrubHgt" that is the shrub height (cm)
- "Herb" that is herbaceous cover (%)
- "Bare" that is bare ground (%)
- "NDVI2017" or "NDVI2018" or "NDVI2019" is the Normalized Difference Vegetation Index for 2017 or 2018 or 2019 which corresponds to the "year" for the females' use and available locations.
- Anthropogenic variables include:
- "HabFactor" (Landcover factor in the manuscript) is a categorical variable describing undisturbed natural vegetation (coded as 0) or disturbed (active + reclamation; coded as 1)
- "InfraDist" is the linear distance (m) to the nearest infrastructure feature
- "ResDist" is the distance (m) to the closest reservoir (Pond distance in the manuscript)
- "Structures500Vis" is a count, which can be partial visibility, of the visible structures within 500 meters
- "Powerline500Vis" is a count, which can be partial visibility, of the visible power poles within 500 meters
Methods
Location and movement data came from GPS-transmitted female sage-grouse that were with chicks (brood-rearing). Spatial layers were created and/or processed using ArcGIS 10.7.0 – 10.7.1 (http://www.esri.com) and R statistical software (R Core Team 2020). Spatial varibles were extrated for use and pseudo-absence location using R statistical software (R Core Team 2020).