Assessing the Viability of Scandinavian Brown Bear, Ursus arctos, Populations: The Effects of Uncertain Parameter Estimates

Abstract

Based on data from radio-collared individuals, we present an analysis of the viability of two small populations of the Scandinavian brown bear, Ursus arctos. The northern and southern populations had different demographic characteristics, even though the population growth rate r and the demographic variance s' were high in both populations (r= 0.13 and sd= 0.180 in the north, and r=0.15 and s'=0.155 in the south). In the northern population the environmental variance s5 was not significantly different from 0, whereas in the south s5 = 0.003. In the south, this was related to high environmental stochasticity in the survival rate of the youngest animals, which resulted in an increase in survival with age in this population. In contrast, in the north, the probability of survival showed a slight decrease with age. Uncertainties were obtained from the joint distribution of bootstrap replications of r, s' and S5. Although the uncertainty in these estimates is quite large, it is unlikely that even relatively small populations ( > 10 females ? 1 year old) will decline to size less than I after 100 years. Analysis of the distribution of the critical population size (i.e. the population size where the population's logarithmic growth rate is zero) shows that these brown bear populations must be larger than 3-4 females 1 year or older to secure a positive growth rate. Similarly, if we define a viable population as the population size where the chance of survival is greater than 90% during a period of 100 years, 8 females ? 1 year old must be present in the north and 6 females in the south. This high viability of even small brown bear populations is due to high reproductive and survival rates. A relatively small increase in the mortality rate will strongly reduce the viability of even relatively large brown bear populations.