U.S. Department of Agriculture: Animal and Plant Health Inspection Service


Date of this Version



Forest Ecology and Management 432 (2019) 599–606


This document is a U.S. government work and is not subject to copyright in the United States.



Black bear (Ursus americanus) damage to trees in the Pacific Northwest is common, although volume and economic losses are unknown. Common measures to quantify bear damage to conifers at large scales rely solely on aerial estimates of red tree crowns (caused by complete girdling) and broad assumptions about stand characteristics. We surveyed 122 vulnerable stands in the Coast Range and western Cascades of Oregon using both aerial surveys and ground surveys. Then, we modeled 4 damage scenarios (Salvage; Total Loss; Root Disease; and Combined Damage) with the Forest Vegetation Simulator (FVS) growth and yield model and the Fuel Reduction Cost Simulator (FRCS). Damage polygons, digitized in real time from aerial surveys identifying red (dead or dying) tree crowns, overestimated bear damage by approximately 5-fold due to misclassification with root disease, and failed to detect partially peeled trees that contributed to economic loss. Damage polygons assessed from the air generally did not include red crowns, and were a mean distance of 58.8m (SE=8.8) from damage polygons’ outer edges to the nearest red crown. We accounted for mortality and volume losses from partially girdled trees that did not show red crowns in our Salvage scenario, whereas we assumed that all bear-peeled trees resulted in complete loss in the Total Loss scenario. At the landscape scale, economic loss was ≤0.35% of net present value under both damage scenarios, while processing bear damage trees (Salvage) was the most efficacious option. At the landscape scale, our worst-case scenario (Total Loss) resulted in an estimated loss of $56/ha to bear damage, 10-fold less than a previously reported estimate of $585/ha. Root disease was a more prevalent damage agent than bear damage but did not affect net present value at harvest. The majority (92%) of bear damage observed in ground surveys was older (> 2 yrs) and existed at a low frequency (1.5 bear damaged trees/ha) and severity across the landscape. Our results suggest that black bear damage is not uniformly distributed and that perceived impact varies with spatial scale. On-the-ground monitoring of the status of bear damage across the western Oregon landscape will identify hot spots of severe peeling and provide an understanding of these changes over time.

Included in

Life Sciences Commons