Demography of a population collapse: The Northern Idaho ground squirrel (Spermophilus brunneus brunneus)

We studied the demography of a population of Northern Idaho ground squirrels (Spermophilus brunneus brunneus) in Adams Co., Idaho. The population was completely censused yearly from 1987 to 1999, during which time it declined from 272 to 10 animals. The finite population growth rate, based on a Leslie matrix model of average life-history parameters, was only 0.72 (i.e., significantly <1.0). Growth rate was more sensitive to proportional changes in juvenile female survival than to any other single life-history parameter. Comparisons with self-sustaining populations of closely related ground squirrel species revealed that juvenile survival and breeding rates of yearling females were anomalously low. We believe that the ultimate cause of the population's collapse was inadequacy of food resources, particularly seeds, due to drying of the habitat and changes in plant species composition, likely the result of fire suppression and grazing. No 'rescue' by immigration occurred, probably because S. b. brunneus seldom disperse long distances and fire suppression has allowed conifers to encroach on inhabited meadows, shrinking them and closing dispersal routes. The proximate cause of the population's collapse was mortality of older breeding females, which reduced the mean age of breeders. Younger females had lower average pregnancy rates and litter sizes. To place our results in context we developed a new, general classification of anthropogenic population declines, based on whether they are caused by changes in the means of the life-history parameters (blatant disturbances), their variances (inappropriate variations), or the correlations among them (evolutionary traps). Many S. b. brunneus populations have disappeared in recent years, apparently due to blatant disturbances, especially loss of habitat and changes in food-plant composition, resulting in inadequate prehibernation nutrition and starvation overwinter. In addition, our study population may have been caught in an evolutionary trap, because the vegetational cues that could potentially enable the animals to adjust reproduction to the anticipated food supply no longer correlate with availability of fat-laden seeds.