Data and code from: Abundance and occupancy trends of sooty grouse in western Oregon: Determining best modeling practices by comparing observed and simulated data
Data files
Mar 17, 2026 version files 1.06 MB
-
abund_trend_dynamic_JAGS.R
11.35 KB
-
abund_trend_PLR_JAGS.R
11.33 KB
-
abund_trend_PLR_ubms.R
10 KB
-
abundance_simulation_functions.R
19.10 KB
-
jags_data_abund_pts.rds
75.01 KB
-
jags_data_dynamic.rds
47.09 KB
-
jags_data_occu_pts.rds
74.63 KB
-
occu_trend_model_JAGS.R
14.12 KB
-
occu_trend_model_ubms.R
9.48 KB
-
occupancy_simulation_functions.R
12.74 KB
-
README.md
3.72 KB
-
SOGR_input_data_summarized_by_centroid_trend_models_2011-2025.csv
480.26 KB
-
umf_sogr_abund.rds
145.54 KB
-
umf_sogr_occu.rds
145.89 KB
Abstract
We estimated sooty grouse population trends using hierarchical models that account for imperfect detection when estimating abundance or occupancy and their dynamics. We used the survey point along a route as the sampling unit. We included route random effects to account for the non-independence of points along a route.
We compared population trend estimates of abundance and occupancy and 95% credible intervals (CIs) from the following modeling frameworks:
1) Binomial N-mixture model with Poisson linear regression (PLR): We ran this model in both JAGS and the 'ubms' package frameworks.
2) Occupancy trend model with logistic regression (OTM): We ran this model in both JAGS and the 'ubms' package frameworks.
3) The Dail-Madsen model with exponential growth (EGM): We ran this model in the JAGS framework.
Finally, we assessed model fit to select one abundance and one occupancy trend model to use in simulation tests to determine which model provides the most accurate trend estimates for sooty grouse. All statistical code is in the R programming language.
Dataset DOI: 10.5061/dryad.djh9w0wbb
Description of the data and file structure
We estimated Sooty Grouse population trends using hierarchical models that account for imperfect detection when estimating abundance or occupancy dynamics. We compared trend estimates (and 95% CIs), measured as lambda, the finite rate of increase per unit time (in this case per year) from the following modeling frameworks: 1) The Dail-Madsen model with exponential growth (EGM) in JAGS framework, 2) Binomial N-mixture model with Poisson linear regression (PLR) in ubms and JAGS frameworks, 3) Occupancy trend model with logistic regression (OTM) in ubms and JAGS frameworks.
Files and variables
Files and variables
File: abund_trend_PLR_ubms.R
Description: Abundance trend model using the ‘ubms’ R package.
File: umf_sogr_abund.rds
Description: Abundance input data formatted as an unmarked frame.
File: SOGR_input_data_summarized_by_centroid_trend_models_2011-2025.csv
Description: Raw data file containing max counts of sooty grouse by centroid used as the input for all models.
Variables
- Route.ID: Unique route ID
- Year: Year of survey
- cent_nn: Unique centroid ID
- Cent_Yr: Centoid ID and year of survey
- Date_max: Date of the max count within survey area (400-m radius around centroid)
- maxSOGR: Maximum sooty grouse count within survey area (400-m radius around centroid)
- stps_per_upt: Number of survey stops summarized with a 400-m radius around the centroid
File: abund_trend_PLR_JAGS.R
Description: Abundance trend model run in the JAGS framework.
File: jags_data_abund_pts.rds
Description: Input data
File: occu_trend_model_JAGS.R
Description: Occupancy trend model run in the JAGS framework.
File: jags_data_occu_pts.rds
Description: Input data
File: occu_trend_model_ubms.R
Description: Occupancy trend model using the ‘ubms’ R package.
File: umf_sogr_occu.rds
Description: Occupancy input data formatted as an unmarked frame.
File: abund_trend_dynamic_JAGS.R
Description: Abundance trend dynamic model run in the JAGS framework.
File: jags_data_dynamic.rds
Description: input data
File: abundance_simulation_functions.R
Description: Functions used to simulate abundance data at different trend levels and run trend models on simulated data.
File: occupancy_simulation_functions.R
Description: Functions used to simulate occupancy data at different trend levels and run trend models on simulated data.
Code/software
We ran the Poisson linear regression abundance trend model, the exponential growth model, and the occupancy trend models using logistic regression in two modeling frameworks that employ Bayesian estimation methods. The first was the R package ‘ubms’ (unmarked Bayesian models; Kellner et al. 2021), which was developed to be similar to the ‘unmarked’ package (Fiske et al. 2000) but provides a user-friendly way to use Bayesian methods for estimating abundance and occupancy models. The second was the JAGS framework, where models were specified in the JAGS language and then run in conjunction with the R packages ‘runjags’ (Denwood 2016), ‘rjags’ (Plummer 2024), and ‘coda’ (Plummer et al. 2006).
Access information
Other publicly accessible locations of the data:
- NA
Data was derived from the following sources:
- Oregon Department of Fish and Wildlife annual Sooty Grouse surveys