Search

Small mammals have been widely used to further our understanding of spatial and temporal population dynamical patterns, because their dynamics exhibit large variations, both in time (multi-annual cycles vs. seasonal variation only) and space (regional synchrony, travelling waves). Small mammals have therefore been the focus of a large number of empirical and statistical (analysis of time-series) studies, mostly based on trapping indices. These studies did not take into account sampling variability associated with the use of counts or estimates of population size. In this paper, we use our field study focusing on population dynamics and demography of small mammals in North Norway at three spatial scales (0.1, 10 and 100 km) to illustrate some methodological and practical issues. We first investigate the empirical patterns of spatial population dynamics, focusing on correlation among time-series of population abundance at increasing spatial scales. We then assess using simulated data the bias of estimates of spatial correlation induced by using either population indices such as the number of individuals captured (i.e., raw counts) or estimates of population size derived from statistical modeling of capture-recapture data. The problems encountered are similar to those described when assessing density-dependence in time-series -a special case of the consequence of measurement error for estimates of regression coefficients- but are to our knowledge ignored in the ecological literature. We suggest some empirical solutions as well as more rigorous approaches.

The European bison Bison bonasus is an example of nearly extinct ‘charismatic megafauna’. The Białowieża Primeval Forest in Poland is among the few places where they still live in the wild. The management of this free-living herd has to reconcile to the conservation needs of a species and the economic and environmental objectives of their habitat: protected as well as commercial woodlands of Białowieża. Here we present a detailed account of the population development and analyse variation in vital rates based on monitoring that started in 1952 and continued until 2002. The population was allowed to grow freely until 1970, when removal started with the aim to stabilize population size. We found that recruitment rate, but not mortality, was density dependent, suggesting that the population density was not very high relative to resource levels. Winters with much snow and cold temperature had a strong negative effect on survival. May temperature of the previous year positively affected recruitment rates. Masting (oak seed) also positively affected recruitment rates, which provides a rare account of masting affecting the performance of a large ruminant. Sex ratio of offspring was even and was not strongly affected by density or climate. We use an age-structured matrix model to show how this knowledge of intrinsic and extrinsic factors affecting vital rates may help managerial decisions by providing explicit links between given environmental conditions and the population growth rates. [ABSTRACT FROM AUTHOR]

Microtine body weights vary temporally and geographically: they are low before the onset of winter and are highest in northern cyclic populations, particularly at peak densities, as well as in island populations. Adaptive explanations have focused on the relationships between body weight, age-specific reproductive and survival rates as well as on population dynamics. We live-trapped five populations of bank voles for up to five years in the French Alps in subalpine habitats: two spruce forests, a deciduous forest, a hedgerow network, and a boulder field recolonized by birch and spruce. Population dynamics were noncyclic, characterized by seasonal fluctuations. Densities were generally high, with little variability among years. Survival rates were higher (0.80 to 0.90 per month), both during summer and winter, than reported from other populations of bank voles in Europe, or of microtines in general. Survival rates varied little among years, despite large variation in snow depth and duration of snow cover, and were similar between males and females. Sex ratio was generally biased towards males (average proportion of males: 0.59). Body weights were higher than for lowland or northern European populations, both for adults in the spring and for subadults at the onset of winter. The latter difference was particularly large, autumn body weights being 50% higher than in these other populations. Difference between autumn and spring weights was therefore much reduced compared with northern populations.Moreover, sexual dimorphism was the inverse of what is known for other populations in which males are smaller than females, males being larger than females during the breeding season in all populations studied here. We suggest that the overall larger body weight is a consequence of a higher survival rate selecting for lower reproductive effort and a higher somatic allocation, and that the reversed sexual dimorphism is a consequence of more intense sexual selection among males, d [ABSTRACT FROM AUTHOR]