Assessing the effects of changes in reproductive condition on the survival of anadromous Dolly Varden (Salvelinus malma) using Bayesian multistate capture-recapture modelling
Data files
Sep 22, 2025 version files 1.56 MB
-
Babbage_Breeding_Data.csv
117.14 KB
-
Babbage_Breeding_Sex_Data.csv
128.92 KB
-
Babbage_Capture_Data.csv
117.14 KB
-
Big_Breeding_Data.csv
129.02 KB
-
Big_Breeding_Sex_Data.csv
140.61 KB
-
Big_Capture_Data.csv
129.02 KB
-
Dolly_Varden_Codes.zip
68.93 KB
-
Firth_Breeding_Data.csv
51.51 KB
-
Firth_Breeding_Sex_Data.csv
59.27 KB
-
Firth_Capture_Data.csv
51.51 KB
-
Joe_Breeding_Data.csv
46.75 KB
-
Joe_Breeding_Sex_Data.csv
53.04 KB
-
Joe_Capture_Data.csv
46.75 KB
-
Rat_Breeding_Data.csv
134.85 KB
-
Rat_Breeding_Sex_Data.csv
145.86 KB
-
Rat_Capture_Data.csv
134.85 KB
-
README.md
6.26 KB
Abstract
Understanding the interrelationship between the survival and reproductive states of salmonids is important for both management and conservation purposes; however, their complex life history introduces challenges. The survival probabilities of anadromous Dolly Varden (Salvelinus malma) are influenced by their reproductive cycle, while predicting their reproductive state in successive years is difficult, as it can vary based on an individual's sex. We developed a Bayesian multi-state capture-recapture model that estimates reproductive state transition probabilities, survival probabilities, and the effect of sex on state transitions using the Cormack-Jolly-Seber model. We applied the model to Dolly Varden data collected from five river systems in the western Canadian Arctic. We demonstrate sex-specific state transition probability differences, with females exhibiting higher transition rates into reproductive states. Moreover, survival probabilities were influenced by both sex and reproductive status. Across all rivers, survival probabilities for both sexes decreased by approximately 50% while spawning compared to non-spawning. This study provides important insights into how reproduction and sex affect survival among stocks, which improves their assessment to ensure management and conservation goals.
Dataset DOI: 10.5061/dryad.bzkh189pw
Description of the data and file structure
The data were collected through long-term monitoring efforts conducted by Fisheries and Oceans Canada (DFO) on Northern Dolly Varden (Salvelinus malma malma) populations in the western Canadian Arctic rivers of the Yukon and Northwest Territories. Sampling years varied by river:
- Firth River (2014–2020)
- Joe Creek (2014–2020)
- Babbage River (2010–2020)
- Big Fish River (2009–2020)
- Rat River (2007–2020)
Fish were captured annually using seine nets, tagged, and released, with sex (if identified) and reproductive state recorded at the time of capture. These capture–recapture data were collected to study survival and reproductive patterns of Dolly Varden, a species of cultural and subsistence importance to local communities.
Files and variables
All variables are Fall_YYYY, and in some files, the sex of the individual.
File: Big_Breeding_Data.csv
Description: Each column represents a different sampling year, where 1 indicates a capture as a spawner, 2 indicates a capture as a non-spawner, and 0 indicates not captured, for the Big Fish River.
File: Big_Breeding_Sex_Data.csv
Description: Each column represents a different sampling year, where 1 indicates a capture as a spawner, 2 indicates a capture as a non-spawner, and 0 indicates not captured, for the Big Fish River. The sex column contains 1 for males, 0 for females, and NA for unknowns.
File: Big_Capture_Data.csv
Description: Each column represents a different sampling year, where 1 indicates captured and 0 indicates not captured, for the Big Fish River.
File: Babbage_Capture_Data.csv
Description: Each column represents a different sampling year, where 1 indicates captured and 0 indicates not captured, for the Babbage River.
File: Babbage_Breeding_Data.csv
Description: Each column represents a different sampling year, where 1 indicates a capture as a spawner, 2 indicates a capture as a non-spawner, and 0 indicates not captured, for the Babbage River.
File: Babbage_Breeding_Sex_Data.csv
Description: Each column represents a different sampling year, where 1 indicates a capture as a spawner, 2 indicates a capture as a non-spawner, and 0 indicates not captured, for the Babbage River. The sex column contains 1 for males, 0 for females, and NA for unknowns.
File: Firth_Breeding_Data.csv
Description: Each column represents a different sampling year, where 1 indicates a capture as a spawner, 2 indicates a capture as a non-spawner, and 0 indicates not captured, for the Firth River.
File: Firth_Breeding_Sex_Data.csv
Description: Each column represents a different sampling year, where 1 indicates a capture as a spawner, 2 indicates a capture as a non-spawner, and 0 indicates not captured, for the Firth River. The sex column contains 1 for males, 0 for females, and NA for unknowns.
File: Firth_Capture_Data.csv
Description: Each column represents a different sampling year, where 1 indicates captured and 0 indicates not captured, for the Firth River.
File: Joe_Capture_Data.csv
Description: Each column represents a different sampling year, where 1 indicates captured and 0 indicates not captured, for Joe Creek.
File: Joe_Breeding_Data.csv
Description: Each column represents a different sampling year, where 1 indicates a capture as a spawner, 2 indicates a capture as a non-spawner, and 0 indicates not captured, for Joe Creek.
File: Joe_Breeding_Sex_Data.csv
Description: Each column represents a different sampling year, where 1 indicates a capture as a spawner, 2 indicates a capture as a non-spawner, and 0 indicates not captured, for Joe Creek. The sex column contains 1 for males, 0 for females, and NA for unknowns.
File: Rat_Capture_Data.csv
Description: Each column represents a different sampling year, where 1 indicates captured and 0 indicates not captured, for the Rat River.
File: Rat_Breeding_Sex_Data.csv
Description: Each column represents a different sampling year, where 1 indicates a capture as a spawner, 2 indicates a capture as a non-spawner, and 0 indicates not captured, for the Rat River. The sex column contains 1 for males, 0 for females, and NA for unknowns.
File: Rat_Breeding_Data.csv
Description: Each column represents a different sampling year, where 1 indicates a capture as a spawner, 2 indicates a capture as a non-spawner, and 0 indicates not captured, for the Rat River.
File: Dolly_Varden_Codes.zip
Description: This zip file contains the R codes for each river and each model presented in the manuscript, organized separately. It also includes a folder named "Prior Sensitivity", which contains two subfolders: "Model 2 Codes" and "Model 3 Codes". Each of these subfolders provides R codes for analyses using alternative priors (Beta(1,2) and Beta(2,1)) explored in this study. In addition, the zip file includes the R codes for the simulation studies conducted for Models 2 and 3.
Code/software
All data files are provided in CSV format, which can be opened with Microsoft Excel or any open-source spreadsheet software (e.g., LibreOffice Calc). All scripts used for data processing and analysis are included in the “Dolly_Varden_Codes.zip” folder. Analyses were conducted in R (version 4.5.1). The primary R packages used were:
- nimble – for Bayesian modeling and MCMC simulations
- tidyverse – for data cleaning and manipulation
- MCMCvis – for summarizing and visualizing MCMC outputs
The workflow involves data cleaning with tidyverse, model fitting with nimble, and post-modeling summaries and diagnostics with MCMCvis. The scripts are organized to clearly show the sequence of steps from raw data to final results.
Access information
Other publicly accessible locations of the data:
- NA
Data was derived from the following sources:
- NA