Northern Goshawk Breeding Habitat in Conifer Stands with Natural Tree Mortality in Eastern Oregon

Publication Date


Type of Culminating Activity


Degree Title

Master of Science in Raptor Biology



Major Advisor

Marc J. Bechard


This study was conducted on the Bear Valley Ranger District, Malheur National Forest, in eastern Oregon. In the 1980's, a western spruce bud worm (Choristoneura occidentalis) epidemic began, and some weakened trees were killed by insect pests and diseases. By the mid-1980's, 21% of spruce budworm host trees, such as Douglas-fir (Pseudotsuga menziesiii, white fir (Abies grandis and A. concolor), and western larch (Larix occidentalis), had died in affected areas. Concern about the possible loss of goshawk breeding habitat prompted this study, because the U.S. Forest Service proposed to salvage dead and dying trees.

The study area was located in areas of natural tree mortality (damaged stands). Calling surveys for breeding goshawks were conducted in 1993 and 1994. I collected vegetative data in 2 nest stands in the study area and 18 nest stands in healthy areas elsewhere on the Bear Valley Ranger District. Also, I collected vegetative data from 21 randomly selected stands in the damaged study area. When compared on an area basis, goshawk sightings declined 42% from 1993 to 1994, and sightings of other raptors declined 23%. The reason for this decline is unknown.

Of the habitat variables measured, density of live and dead trees, tree species composition, % slope, and % canopy closure differed between nest and damaged stands. Tree density did not vary significantly between nest and damaged stands. When placed in diameter (dbh) classes, differences were significant only in the > 53.3 cm class. Nest stands had more live trees in all dbh classes, and this was significant except in the 23.0-53.3 cm class. Also, damaged stands had more dead trees in all dbh classes and differences were significant except in the > 53.3 cm class. Differences in tree density appeared to be due to natural conditions, not logging.

The most common tree species in nest stands were ponderosa pine (Pinus ponderosa) (53%), Douglas-fir (18%), and white firs (17%). The most common species in damaged stands were ponderosa pine (19%), Douglas-fir (24%), and white fir (45%). Ponderosa pine was the only species that varied significantly.

Slope varied significantly between nest stands (18%) and damaged stands (33%). Also, canopy closure varied significantly between nest stands (55%) and damaged stands (46%). Because there was no clear relationship between tree density and canopy closure, it was not possible to determine if the difference in canopy closure between nest and damaged stands was due to insect- and disease-induced tree mortality or due to differences in crown diameters of different tree species.

Although insects and diseases affected damaged stands, the degree and severity could not be determined from available data. It was likely that other factors, such as prey density, density of large, live trees, or overall habitat quality throughout the home range, might have been more important than the presence of tree mortality in determining if goshawks nested in an area. I recommend that nests in damaged and healthy stands be equally protected, and that damaged stands without goshawk nests should be managed to produce goshawk nesting habitat in the future.

Files over 30MB may be slow to open. For best results, right-click and select "save as..."