Conservation Evidence strives to be as useful to conservationists as possible. Please take our survey to help the team improve our resource.

Providing evidence to improve practice

Individual study: Black bear Ursus americanus supplemental feeding reduces conifer damage on the Olympic Peninsula, Washington State, USA

Published source details

Ziegltrum G. I. (2004) Efficacy of black bear supplemental feeding to reduce conifer damage in western Washington. Journal of Wildlife Management, 68, 470-474


The black bear Ursus americanus population in Washington State, USA, is estimated to number between 25,000 to 50,000 individuals, many of which are present within 400,000 ha of commercial forest. During the spring months (mid-April-June), one bear may girdle 60-70 conifer trees in a day when other food is scarce. The bears prefer trees in 15-25 year old stands, with trees about 20-40 cm diameter at breast height. Tree-bark stripping and subsequent feeding on sapwood can either damage or kill trees, resulting in significant economic losses. In 1989, the Washington Forest Protection Association’s (WFPA) Animal Damage Control Program implemented a supplemental black bear feeding programme as a non-lethal alternative to black bear removal by hunting. The objective was to provide bears with an alternative source of food during the spring, thus reducing damage to tree stands.

Study area: The study took place in mixed Douglas fir Pseudotsuga menziesii and hemlock Tsuga spp. stands on the Olympic Peninsula, western Washington. There were 14 test sites within areas with trees in the age range (18-26 years) preferred by bears. Eight were located on the north side of the peninsula in Clalam County (along State Route 112); six sites were located on the west side of the peninsula in Jefferson County (along US Route 101).

Pre-treatment survey: Pre-treatment surveys began in March 1999 to assess the existing characteristics and damages levels of each stand within the study sites. All stands had been thinned prior to the study. The stands (16 – 20 ha), had similar timber (Douglas fir and hemlock) stock rates of approximately 1,000 trees/ha. Each stand was divided into four sections and a 10 m wide belt transect was extended into the stand perpendicular to the edge in each. The first 250 live trees within a transect were examined and bear damage recorded. A total of 14,000 trees were examined on the  sites (1,000 trees/site). Trees with bear damage were marked with red tree paint and those undamaged with blue.

Post-treatment survey: Over the next four years, annual post-treatment surveys examined the same marked trees. Stands with similar levels of damage were paired for analysis (7 pairs in total). At seven sites, two feeding stations (see below) were established in April 1999, while no supplemental feeding (as a control) was carried out at the other seven.

The first surveys were conducted in July of 1999. The 1,000 marked trees were examined for bear damage. A second survey was conducted in July 2000 and noted new damage from spring 2000. After the bear damage surveys in July 2000, two treatment sites were randomly selected and supplemental feeding was stopped. The final surveys took place in July 2002, when bear damage from the spring of 2001 and 2002 was recorded.

Feeding stations: In the' treatment sites' (i.e. supplemental food sites), beaver Castor Canadensis carcasses were hung on trees in the first year to attract the bears to the supplemental food and shorten the period of habituation. Feeding stations comprised plastic drums capable of holding about 90 kg of food pellets; two cables, at the top and bottom of the drum, attached it to a tree. Inside the drum, a slanted plywood sheet formed a self-replenishment device. The feeding stations were stocked weekly with pellets through the spring at a rate of 100 kg each. The palpability of the pellets was increased through the addition of fat, chicken protein, vitamins, blood concentrates, sugar beet and minerals.

During 1996-2002, between 240 and 250 metric tonnes were distributed from April to June annually throughout 900 feeding stations situated in commercially valuable timber stands.

Pre-treatment surveys: There was no statistically significant differences between the number of trees damaged by bears on treatment and control sites. Between 1999 and 2002, bears damaged significantly more trees on control sites than on treatment sites. To validate these initial results, feeding stations were removed from two of the seven feeding sites in July 2000. Bears proceeded to damage six trees in the first and second seasons (July 1999/July 2000); an additional 40 trees were recorded as damaged in the last survey (July 2002). Damage had increased nearly 7-times at one feeding site. Removing feeding stations from established bear feeding areas appeared to cause the increase in bear tree-girdling.

At least 25% of trees were damaged by bears; furthermore, this damage was compounded by tree-thinning management practices (to 1,000 stems/ha).

Post-treatment surveys: Survey results extrapolated to a 20 ha stand suggested that 769 out of 20,000 trees on stands without feeding stations would suffer bear damage each year. Over a 15 year period, this would equate to 11,535 trees (57.7%) damaged. In contrast, damage estimates within stands with feeding stations during the same period, would be only 2,111 trees (10.6%). One site was omitted from the analysis due to a feeding station not being filled for one week; the ratio of damaged trees between treatment and control sites would have been even greater if this error had not been made.

Feeding stations: A problem identified concerned the maintenance of the feeding stations, especially restocking and pellets getting wet. In 2000, the Animal Damage Control Program maintained 900 feeding stations in western Washington. Based on the figures they provided, one feeding station costs $110, while the price for pellets provided to land managers last year was $495 per tonne.

Conclusions: This study supports earlier anecdotal experience of foresters in Washington, and proves that a supplemental bear feeding programme constitutes a viable, non-lethal method to reduce damage to trees in spring.

Note: If using or referring to this published study, please read and quote the original paper, this can be viewed at: