Vital rates of two small populations of brown bears in Canada and range-wide relationship between population size and trend

Abstract

Identifying mechanisms of population change is fundamental for conserving small and declining populations and determining effective management strategies. Few studies, however, have measured the demographic components of population change for small populations of mammals (<50 individuals). We estimated vital rates and trends in two adjacent but genetically distinct, threatened brown bear (Ursus arctos) populations in British Columbia, Canada, following the cessation of hunting. One population had approximately 45 resident bears but had some genetic and geographic connectivity to neighboring populations, while the other population had <25 individuals and was isolated. We estimated population-specific vital rates by monitoring survival and reproduction of telemetered female bears and their dependent offspring from 2005 to 2018. In the larger, connected population, independent female survival was 1.00 (95% CI: 0.96-1.00) and the survival of cubs in their first year was 0.85 (95% CI: 0.62-0.95). In the smaller, isolated population, independent female survival was 0.81 (95% CI: 0.64-0.93) and first-year cub survival was 0.33 (95% CI: 0.11-0.67). Reproductive rates did not differ between populations. The large differences in age-specific survival estimates resulted in a projected population increase in the larger population (λ = 1.09; 95% CI: 1.04-1.13) and population decrease in the smaller population (λ = 0.84; 95% CI: 0.72-0.95). Low female survival in the smaller population was the result of both continued human-caused mortality and an unusually high rate of natural mortality. Low cub survival may have been due to inbreeding and the loss of genetic diversity common in small populations, or to limited resources. In a systematic literature review, we compared our population trend estimates with those reported for other small populations (<300 individuals) of brown bears. Results suggest that once brown bear populations become small and isolated, populations rarely increase and, even with intensive management, recovery remains challenging.

Keywords: Ursus arctos; brown bear; carnivore conservation; grizzly bear; population recovery; small population; survival.

FIGURE 1

Mother brown bear and offspring in a waterhole. These bears are part of the growing McGillvary Mountains population in southwest British Columbia, Canada

FIGURE 2

Extant (green) and vagrant (yellow) brown bear distribution in North America (inset) and southwest British Columbia, Canada. The geographic boundary between the genetically distinct McGillvary Mountains (MM) and North Stein‐Nahatlatch (NSN) brown bear populations bisects the study area (blue dash) centered at 50.579181, −122.400917. Highways (red lines), major rivers (blue lines), international border (bottom solid black line)

FIGURE 3

Density plots of (a) the stable reproductive state distribution estimated using in multi‐state transition models on reproductive data from collared adult female brown bears (≥6 years). (b) Bootstrapped estimates of the stable age distribution estimated from vital rates for grizzly bears in the McGillvary Mountains and North Stein‐Nahatlatch populations in southwestern British Columbia, Canada

FIGURE 4

The relationship between brown bear population growth rate (λ) and population size from this research and other unhunted populations including isolated populations (●), population with some connectivity to neighboring populations (○), and augmented populations (+). Populations are identified by letters: AI: Apennines, Italy (Gervasi et al., 2017); SA: Southern Alps, Europe (Tosi et al., 2015); BA: Banff, Canada (Garshelis et al., 2005); CU: Cabinet Mountains, USA (Kasworm et al., 2014; Kendall et al., 2016; Proctor et al., 2012; Wakkinen & Kasworm, 2004); CE: Cantabrian East, Spain (Clevenger et al., 1987; Palomero et al., 2007; Pérez et al., 2014); CW: Cantabrian West, Spain (Clevenger et al., 1987; Palomero et al., 2007; Pérez et al., 2014); CP: Central Pyrenees, Spain‐France (Chapron et al., 2009); DP: Deosai, Pakistan (Nawaz et al., 2008); MM: McGillvary Mountains, Canada; NSN: North Stein‐Nahatlatch (This research, McLellan et al., 2019); PSY: Purcell‐South Yaak (Proctor et al., 2012); WP: West Pyrenees, Spain‐France (Chapron et al., 2009); GYE: Yellowstone, USA (Schwartz et al., 2006; van Manen et al., 2015)