Growth Rates and Variances of Unexploited Wolf Populations in Dynamic Equilibria: Data, R Code, and Supporting Results

• Main Authors: Fieberg, John R, Mech, David
• Other Authors: [e-mail]
• Format: Dataset
• Language: unknown
• Published: 2014
• Subjects: Canis lupus; wolf; gray wolf; moose; Denali; Isle Royale; Superior National Forest; density; natural population; observation error; population trajectory; population dynamic models; process error; Ricker model; Alaska
• Online Access: http://hdl.handle.net/11299/164048; https://doi.org/10.13020/D6RP4N

This dataset contains four files. PopulationModels.R is an R script defining functions used to fit density-independent and Ricker population models to associated time series data. With these functions, population measurements can be modeled under three different measurement assumptions: i) measured without error; ii) measured with Poisson error; or iii) measured with log-normal error. MechFieberg.R is a R script that will run all analyses supporting the findings in Mech and Fieberg (2014). MechFieberg.html is a summary of the output expected when running the MechFieberg.R script. Wolfdat.csv is the raw data file containing the wolf home range measurements. The four columns in this data correspond to the year of measurement (YR), and the location of measurement: Denali National Park (Denali), Isle Royale (IsleRoyale), and Superior National Forest (SNF). These files contain data and R code (along with associated output from running the code) supporting all results reported in: Mech, D. and J. Fieberg. 2014. Growth Rates and Variances of Unexploited Wolf Populations in Dynamic Equilibria. Wildlife Society Bulletin. In Mech and Fieberg (2014), we analyzed natural, long-term, wolf-population-density trajectories totaling 130 years of data from three areas: Isle Royale National Park in Lake Superior, Michigan; the east-central Superior National Forest in northeastern Minnesota; and Denali National Park, Alaska. We fit density-independent and Ricker models to each time series, allowing for 3 different assumptions regarding observation error (no error, Poisson or Log-normal observation error). We suggest estimates of the population-dynamic parameters can serve as benchmarks for comparison with those calculated from other wolf populations repopulating other areas.