Ecological correlates of mortality of roe deer fawns in a predator-free environment

The mortality of roe deer (Capreolus capreolus) fawns in summer was studied on Storfosna Island (10 km(2)), a predator-free island 2 km off the west-central coast of Norway, during a period of rapid population increase. Between 1991 and 1994 the population density increased from about 10 to 40 roe deer/km(2). During this period 285 live and 36 stillborn fawns were found and the live fawns were all subsequently radio-collared. All observed mortality occurred within the first 35 days of life. The main causes of death were stillbirth, starvation/hypothermia, drowning, car accidents, and falls. The average annual total mortality for all 4 years was 18%. Although there was a large increase in mortality between 1991 and 1992 (from 10 to 18%), there was little subsequent increase in the last 3 years, resulting in an overall weak density dependence in total mortality. However, the proportion of stillbirths (5% in 1991 to 15% in 1994) showed a clear and significant trend toward an increase with increasing density. Low temperatures in April, which strongly affect the timing of the spring flush of high-quality forage during late gestation, also tended to contribute to higher mortality. Fawns that died were found to have reduced individual growth rates prior to death compared with fawns that survived. Fawns born to relatively light mothers or in triplet litters had higher mortality rates than those born to heavy mothers or in smaller litters. Maternal age and date of birth or sex of fawn had no effect on mortality.