Habitat selection and population trends of the Torngat Mountains caribou herd

Abstract

Understanding why species at risk select certain habitat and what components of their life history influence changes in numbers can help mitigate population declines. The Torngat Mountains caribou (Rangifer tarandus) herd in northern Quebec-Labrador, Canada, is declining, and few studies have examined the potential causes of this decline. We fitted 9 Argos and 26 global positioning system (GPS)-collars on 35 adult caribou (25 female, 10 male) between 2011 and 2016 to assess seasonal habitat selection at 2 spatial scales, current and future population trends, and interactions with the neighboring Rivi ere-George migratory caribou herd. The caribou of the Torngat Mountains herd selected areas with abundant food resources in winter and where prevalence of insects was lower in summer. They did not avoid areas where predation risk was high during calving. Spatial overlap with the Rivi ere-George herd range decreased from 1990 to 2015 and was correlated with the size of the Rivi ere-George population, which declined drastically during this period. The decline of the Torngat Mountains population was principally attributed to the low survival of adult females (0.72 annual survival rate) owing to subsistence harvest (9/24) and predation (7/24). Demographic models revealed that the growth rate of the population (l) could vary from 0.83 (current) to 0.94 following a decrease in harvest pressure. Using demographic scenarios, we showed that the Torngat Mountains herd could continue to decrease if no management actions were taken to increase adult female survival. Ó

Effect of wind farms on wintering ducks at an important wintering ground in China along the East Asian–Australasian Flyway

“…(2016), we employed the method of k‐fold cross‐validation to evaluate the model fit. Our data were divided into 10 k‐folds based on the Huberty (1994) rule of thumb, and we reported the average Spearman's correlation for the 10 iterations (Bélanger, Leblond, & Côté, 2019). All statistical analyses were conducted in R 3.6.0 (R Core Team, 2019), with the “glmulti” (Calcagno, 2019) and “MuMIn” (Bartoń, 2019) packages for model selection and averaging.…”

“…Candidate models were selected using Akaike's information criterion (AIC) and ranked using AIC c differences (ΔAIC c ) and AIC weights (W i ) to estimate the relative likelihood of each candidate model (Burnham & Anderson, 2002). Competitive thumb, and we reported the average Spearman's correlation for the 10 iterations (Bélanger, Leblond, & Côté, 2019). All statistical analyses were conducted in R 3.6.0 (R Core Team, 2019), with the "glmulti" (Calcagno, 2019) and "MuMIn" (Bartoń, 2019) packages for model selection and averaging.…”

Impacts of land cover on reptile movement and habitat use in farming landscapes

“…Random activity areas for each individual were the same size and shape as real activity areas. To guide the placement of random activity areas we used the random point tool in QGIS version 3.6, and defined the boundary of available habitat by adding a 500 m buffer around the real activity areas of all individual animals within each site (Bélanger, Leblond & Côté, 2019). We calculated the proportion of farms, tree patches, grassland, water bodies, roads and human dwellings within each real and random activity area using a land cover map that we manually constructed based on satellite imagery.…”

Investigating potential for competition between migratory caribou and introduced muskoxen

“…These models estimate the utilization distribution (UD) as a continuous-time stochastic model of movement (Horne et al 2007). Previous studies have used BBMMs to describe movement patterns of migratory species (Sawyer et al 2009(Sawyer et al , 2019, including caribou (Nicholson et al 2016, Bélanger et al 2019. Prior to this analysis, we filtered all GPS locations and removed locations with a GPS positional dilution of precision >10 (D'Eon and Delparte 2005).…”