Back to Theses

Janne Sundell 2002: Vole Population Dynamics: Experiments on Predation. PhD-Thesis.

Northern vole populations exhibit regular fluctuations in numbers. In these so-called cycles, voles reach their peak numbers at intervals of 3-5 years. Many hypotheses have been put forward to explain these regular oscillations. The most popular one is the predation hypothesis, which emphasises the important role of delayed density-dependent predation by small mustelids.

In the thesis, I tested experimentally the predation hypothesis with a large-scale field experiment. The least weasel is assumed to be the most important predator of voles, especially during the decline phase of the vole cycle. I also studied different aspects of the predation of the least weasel (Mustela nivalis nivalis) on vole populations in smaller scale experiments. The thesis work also involved a study of large-scale spatial and temporal patterns of vole populations in Finland.

I showed that the least weasel has potential to shape vole dynamics. The least weasel has a very specialised diet of small mammals, and it shows a type II functional response. Both of these are necessary components for a predator to destabilise vole dynamics. The least weasel has a tendency to kill more voles than is necessary to maintain its current needs, and this further increases weasel's impact on vole populations. Weasel can also rapidly react numerically by reproducing more offspring if voles are abundant. They may reproduce twice during the same season and young born early in the season may have their own young by the end of the summer. Weasels differ from the other important vole predators, owls and raptors, by mating every year despite low vole abundance. I observed that weasels hunt different vole speciesdepending on their relative abundance, even hunting disproportionately more of the more abundant species, i.e. showing predator switching. This may facilitate coexistence of vole species and explain why different vole species exhibit synchronous fluctuations.

The mean size of voles varies during the cycle, so that voles are larger in the peak phase than during the decline and the low phases, a phenomenon called the 'Chitty effect'. I have put forward a new hypothesis, supported by extensive field data and experiments conducted in laboratory, to explain the phenomenon. According to my hypothesis, small size of voles during decline and low phases of the vole cycle is due to disproportionate predation on larger voles, especially by the least weasel.

In a large-scale field experiment, I manipulated the numbers of weasels in large (5-10 km 2 ) islands in lakes by adding weasels to the system in order to eliminate the natural lag in the numerical response of the weasel. The results were ambiguous. In two out of three islands, there was no noticeable effect of weasel additions on vole population dynamics, while in one of the islands the number of field voles decreased or remained the same (depending on the methods of assessing vole numbers) after weasel additions. In a comparable adjacent control island, field vole populations increased. Furthermore, field vole populations were male-biased in experimental island suggesting higher predation on female field voles by the least weasels. The interpretation of the results was complicated by the irregular dynamics and low numbers of field voles on the two islands (and their control islands) that showed no treatment effect. Based on the high breeding success of vole-eating avian predators in areas surrounding the study islands, I conclude, that the lack of cyclic dynamics was part of a more general phenomenon of increased irregularity in the formerly rather regular vole population dynamics in southern Finland.

The investigation of large-scale patterns in vole population dynamics with help of ringing data on vole-eating avian predators showed previously observed geographical gradients in cycle length. Vole population dynamics showed mainly 3-year periodicity in southern Finland and 4-5 year periodicity in the northern Finland. I also observed that cycle length tends to increase from west to east. Vole populations were largely synchronous over large distances, up to several hundred kilometres. The most likely explanations for this large-scale synchrony are predation by mobile avian predators and/or correlative climatic factors.

Full pdf available via etheses