A. Forsman et Le. Lindell, RESPONSES OF A PREDATOR TO VARIATION IN PREY ABUNDANCE - SURVIVAL ANDEMIGRATION OF ADDERS IN RELATION TO VOLE DENSITY, Canadian journal of zoology, 75(7), 1997, pp. 1099-1108
We tested for effects of temporal and spatial variation in;prey densit
y on survival, growth, body condition, and emigration of a snake speci
es feeding on voles. We carried out a capture-mark-recapture study (19
89-1994) of two subdivided populations of adders, Vipera berus, inhabi
ting two groups of islands (Angskar and In-Fredeln) in the Baltic Sea
to estimate, compare, and model survival and capture probabilities, an
d to test for effects of time, locality, sex, and vole density on surv
ival, using the computer program SURGE. TO quantify variation in prey
availability among years and localities we trapped field voles, Microt
us agrestis, the primary prey of adders, every year from 1989 to 1993
on the islands where we captured snakes. We found that vole densities
showed large fluctuations among years at In-Fredeln but remained relat
ively low throughout the study period at Angskar. Annual survival prob
abilities of adders varied in a nonparallel manner through time betwee
n localities and ranged from 0.36 to 1.00, with a mean of 0.72 across
years and localities. Survival did not differ between males and female
s. Survival of adders increased with population density of voles in on
e of the two populations (In-Fredeln), suggesting that starvation may
have been an important source of snake mortality in some years, or tha
t physiological stress resulting from food shortage may render individ
uals more susceptible to predators and diseases. Adders at In-Fredeln
also grew faster and were in better body condition when vole density w
as high than when it was low. Inter-island movement rates of adders ra
nged from 0 to 29% among years but were not correlated with vole densi
ty, suggesting that the relationship between snake survival and vole d
ensity did not simply reflect increased emigration in years when prey
were scarce.