Microtine Rodent Dynamics in Northern Europe: Parameterized Models for the Predator‐Prey Interaction

We added seasonality to a general predator—prey model to make it a more realistic description of the interaction between microtine rodents (Microtus) and their mustelid predators (Mustela nivalis) in northern Europe. The model parameters were estimated from field data, with the exception of three parameters, for which a range of plausible values was selected. Most parameter combinations generated high—amplitude, chaotic oscillations with a distinct periodic component; the median parameter values predicted dynamics closely resembling the observed 3—5 yr rodent oscillations in northern Fennoscandia. The results were not sensitive to a number of structural changes in the model, suggested by empirical observations: stochastic variation in the prey—carrying capacity, predator refuge at low density, predation on mustelids by large generalist predators in the years when rodents decline, and an alternative assumption about winter breeding in rodents. The model—predicted period and amplitude of oscillations agreed most closely with observations when the model included a low—density refuge for the mustelids. Another predator—prey model with ratio—dependent functional response did not predict dynamics resembling the observed dynamics. The present results provide further support to the hypothesis that the rodent oscillations are maintained by delayed density dependence imposed by specialist predators. Our results also suggest that the dynamics may be chaotic, with a periodic component, rather than a limit cycle with environmental stochasticity.