Common vole (Microtus arvalis) population sex ratio:: biases and process variation

Vole population sex ratio varies seasonally. However, population sex ratios have usually been estimated using naive estimators that do not allow for biases owing to the sex difference in capture probabilities and movement distances (i.e., effective areas sampled). Here we aimed to advance the methodological approach, recognizing that there are two different classes of contributing mechanisms to the pattern which are best addressed separately: (1) those mechanisms imposing a systematic error (bias) in population estimates and (2) those generating the true process variation. Analyzing 7-year capture-recapture data in the common vole (Microtus arvalis (Pallas, 1778)), we quantified both types of biases and revealed that the bias owing to differential capture rates was often severe and less predictable, whereas that owing to differential effective areas was smaller and overestimated male numbers for most of the year. We demonstrated unambiguously that the unbiased population sex ratio indeed varies seasonally, with the males usually being more numerous over winter and spring. By testing predictions from two mechanistic hypotheses to explain the process variability, we found evidence for both the differential recruitment hypothesis and the differential survival hypothesis. From April-May to August, it was the females that were recruited more to the population and that had higher survival rates than males. We suggest that the seasonal variation in the population sex ratio is not merely a result of biasing mechanisms but an important population property driven by the joint effect of differential recruitment and differential survival between sexes.