Data from: Annual survival in a dynamic species: Pronghorn survival patterns across their northern range
Data files
Feb 12, 2026 version files 263.02 KB
-
covariates.csv
2.80 KB
-
ehjuv.csv
68.90 KB
-
ehmale.csv
25.41 KB
-
ehmt.csv
86.59 KB
-
ehsd.csv
72.69 KB
-
README.md
6.62 KB
Abstract
Quantifying variation in demographic patterns, such as survival and recruitment, is critical for understanding population dynamics and informing evidence-based and adaptive wildlife management. In this study, we leverage an extensive dataset from over 1,000 GPS collared pronghorn (Antilocapra americana) to provide the first large-scale evaluation of survival patterns for an iconic North American ungulate. This research spans multiple years, regions, and age and sex classes, and includes more than 3,000 animal-years of monitoring, representing the most spatially and temporally extensive survival dataset for the species to date. Using a hierarchical Bayesian analytical framework, we estimate the process distribution of survival, thereby capturing true biological variation in adult females, adult males, and juveniles. Our results reveal that, across pronghorn populations, mean pronghorn survival is variable yet statistically consistent across regions and through time. In contrast, adult female pronghorn show lower and more temporally variable survival than other sympatric ungulates across North America (e.g., deer and elk). Through this extensive survival analysis and robust dataset, we identify demographic mechanisms (e.g., temporally varying adult female survival rates) that contribute to dynamic population trends within the species. Our findings advance understanding of pronghorn demography and highlight species-specific vulnerability to stochastic events. More broadly, this work underscores the value of large-scale demographic monitoring for identifying sources of variation in population trends critical for informing adaptive management in the context of global environmental change.
Dataset DOI: 10.5061/dryad.dv41ns2cd
Description of the data and file structure
Overview
This Dryad archive contains data supporting a Wildlife Biology manuscript examining pronghorn (Antilocapra americana) survival across multiple study areas and years. The dataset includes (1) study-area–level and annual environmental covariates and (2) individual-level encounter histories and metadata for pronghorn monitored over a 36‑month period. Data are organized to support time‑to‑event and capture–recapture / known‑fate survival analyses.
The archive includes encounter history data for female pronghorn over three years (36 months), as well as corresponding datasets for male pronghorn and juvenile pronghorn (6‑month monitoring window).
File inventory
1. covariates.csv
Description: Study area– and year‑specific environmental covariates used as predictors in survival models.
Rows: One record per DAU (Data Analysis Unit) per year.
Columns:
- yr – Study year (integer; 1–3).
- DAU – Data Analysis Unit name.
- dau – Broad DAU grouping (e.g., Central MT, Mountain Valley).
- pdsi – Palmer Drought Severity Index (mean annual or seasonal value).
- ndvi – Normalized Difference Vegetation Index (scaled or summarized as used in analyses).
- snow – Snow metric (e.g., cumulative snowfall or snow water equivalent).
- cold – Cold exposure metric (e.g., number of cold days below a 10 degrees F).
Note: Unnamed: 0 is an index column retained from data export and can be ignored.
2. ehmt.csv
Description: Montana monthly encounter histories and metadata for female pronghorn, monitored over 36 months. Female pronghorn monthly encounter histories for Montana Study Areas; NA = missing data for month (no observation)
Rows: One record per individual.
Columns:
- animalID – Unique individual identifier.
- 1–36 – Monthly encounter history (1 = alive/observed, 0 = not observed or dead, depending on modeling framework).
- Fate – Final fate of the individual (e.g., alive, mortality).
- DateMort – Date of mortality, if applicable.
- DAU – Data Analysis Unit where the individual was monitored.
- Sex – Sex of individual (F).
- agenow – Age class at start of study period and during monitoring (as defined in the associated manuscript).
Note: Unnamed: 0 is an index column retained from data export and can be ignored.
3. ehsd.csv
Description: South Dakota monthly encounter histories and metadata for pronghorn monitored over 36 months, formatted for survival analysis with metadata fields preceding encounter histories. Female pronghorn monthly encounter histories for South Dakota Study Areas NA = missing data for month (no observation)
Rows: One record per individual.
Columns:
- animalID – Unique individual identifier.
- Sex – Sex of individual.
- DAU – Data Analysis Unit.
- agenow – Age class.
- DateMort – Date of mortality, if applicable.
- Fate – Final fate classification.
- 1–36 – Monthly encounter history.
Note: Unnamed: 0 is an index column retained from data export and can be ignored.
4. ehmale.csv
Description: South Dakota monthly encounter histories and metadata for male pronghorn monitored over 36 months, formatted for survival analysis with metadata fields preceding encounter histories. Male pronghorn monthly encounter histories for South Dakota Study Areas; NA = missing data for month (no observation)
Rows: One record per individual.
Columns:
- animalID – Unique individual identifier.
- Sex – Sex of individual.
- DAU – Data Analysis Unit.
- agenow – Age class.
- DateMort – Date of mortality, if applicable.
- Fate – Final fate classification.
- 1–36 – Monthly encounter history.
Note: Unnamed: 0 is an index column retained from data export and can be ignored.
5. ehjuv.csv
Description: South Dakota monthly encounter histories and metadata for juvenile male and female pronghorn monitored over three 6-month periods (January – May), formatted for survival analysis with metadata fields preceding encounter histories. Juvenile pronghorn monthly encounter histories for South Dakota Study Areas; NA = missing data for month (no observation). Note that months from June - December will be all NA's as pronghorn became adults on June 1 and so we only recorded monthly survival for pronghorn from capture dates in January until May 30th. We captured pronghorn in 3 subsequent years.
Rows: One record per individual.
Columns:
- animalID – Unique individual identifier.
- Sex – Sex of individual.
- dau – Data Analysis Unit.
- agenow – Age class at capture.
- DateMort – Date of mortality, if applicable.
- Fate – Final fate classification.
- 1–29 – Monthly encounter history.
Note: Unnamed: 0 is an index column retained from data export and can be ignored.
Code/software
Software and workflow
All data files in this archive are provided in comma-separated values (.csv) format and can be viewed using any standard spreadsheet software or text editor (e.g., Numbers, Microsoft Excel) or imported into statistical software.
All data processing and analyses associated with this manuscript were conducted using R, an open-source statistical computing environment, with RStudio used as an integrated development environment.
- R version: R ≥ 4.2.0
- RStudio: ≥ 2023.06 (optional)
R packages used
The following R packages were used for data processing and analysis:
tidyverse– data wrangling, formatting, and visualizationr2jags– Bayesian survival modeling implemented via JAGS
These packages are freely available and open source.
Workflow overview
- Encounter history and metadata files were imported into R from
.csvformat. - Monthly encounter histories were formatted for survival analyses.
- Individual-level encounter history data were linked to study area– and year-specific environmental covariates using Data Analysis Unit (DAU) and year identifiers.
- Environmental covariates were processed prior to analysis, as described in the Methods section of the manuscript.
- Survival models were fit using Bayesian methods implemented in JAGS via the
r2jagspackage.
No proprietary software is required to view, process, or analyze these data. Analytical methods and model structure are fully described in the associated manuscript.