Responses of rodent populations to spatial heterogeneity and successional changes within Sitka spruce (Picea sitchensis) plantations at Hamsterley Forest, County Durham
Populations of woodmice (Apodemus sylvaticus),
bank voles (Clethrionomys glareolus) and field voles (Microtus agrestis) were studied by
five-trapping in Sitka spruce plantations at Hamsterley Forest, northeast England, from
February 1990 to June 1992. The study was carried out at two distinct spatial scales.
At the coarser spatial scale, the effects of successional change and of spatial (intersite)
heterogeneity within the forest were compared by censusing rodent commmunities in
sites representing three successional stages: mature plantations (about 40 years after
planting), clear-fellings and young plantations (5-8 years after planting). In, mature
plantations woodmice and bank voles were both abundant,
in clear-fellings the former
were usually dominant and in young plantations field voles were also abundant alongside'
the other two species. Pooling all sites, in young plantations rodent communities had
highest diversity and evenness, mostly because of an inter-site component @-diversity).
On a site-by-site basis, in young plantations communities had neither higher a-diversity
nor evenness than in mature plantations. Clear-fellings showed least diversity and
evenness. Multivariate analyses revealed marked taxonomic and structural changes in
vegetation during succession and how the rodents responded to such changes. Within
young plantations bank voles were associated with dense ground cover, provided mostly
by heather. Field voles. were associated with the non-palatable grass Deschwnpsia
flýxuosa and to palatable grasses as well. Woodmice were habitat generalists, but their
abundance was negatively correlated with that of field voles. Spatial heterogeneity in soils
explained much of the inter-site variation in ground vegetation which in turn explained
much of the P-diversity in rodent communities in young plantations.
At the finer spatial scale, populations of woodmice and bank voles were studied by
monthly trapping in five 0.8 1 ha grids within an habitat mosaic produced by the felling of a
plantation of mature Sitka spruce. Inter-grid movements were frequent for both species,
but especially woodmice, which also had larger home ranges. Population dynamics of
woodmice in the whole mosaic were apparently similar to patterns described previously in
other habitats, except that density-dependent reduction of survival and reproduction by
late autumn was more severe than usual. Woodmice were more abundant in 1991 than in
1990, apparently due to increased seed supply. Bank voles, in contrast, did not show
regular annual fluctuations in numbers. Although breeding stopped in both winters,
population densities increased steadly during 1991 and remained high until spring 1992.
Clear-felling of a part of the study area had little immediate effect on populations of either
species; responses to clear-felfing were gradual rather than sudden. The experimental
removal of tree brashings from recent clear-felfings scarcely affected the populations of
woodmice, but made the clear-feUings unsuitable for bank voles, apparently due to the
reduced availability of shelter. Habitat selection within the whole habitat mosaic was
density-independent in bank voles, but density-dependent in woodmice. In the later
species, demographic differences among local subpopulations accounted for most of the
observed density-dependent changes in spatial distribution, although inter-grid movements
also played an important part.
The responses of each rodent species to spatial and temporal heterogeneity and a
possible role of density-dependent habitat selection in population regulation are discussed.