Flying Squirrel Population Densities And Habitat Assessment In Hardwood, Mixed, And Coniferous Stands At Huntington Wildlife Forest In The Adirondacks

Madeleine Fairbairn

Releasing a flying squirrel into the weighing bag (AEC image files).

Background
Recent studies have investigated the importance of different habitat variables to northern flying squirrel (Glaucomys sabrinus) populations. Some studies have found that G. sabrinus are habitat specialists whose survival depends largely on the suitability of their surrounding habitat. Others have found that they are capable of surviving in a wide variety of woodland habitat types. G. sabrinus are nocturnal sciurid rodents which inhabit forested regions throughout Canada as well as parts of the northern U.S. and occasional high altitude forests in the southern U.S. Although several recent studies have examined G. sabrinus habitat in the Pacific Northwest due to their role as principle prey species for the endangered northern spotted owl (Strix occidentalis caurina) fewer studies examine habitat in the eastern end of their range. The purpose of my study was to examine G. sabrinus habitat variables and population densities in the Adirondacks. The objectives of my study were to 1) estimate G. sabrinus populations in hardwood, mixed, and coniferous stands 2) document the physical characteristics of the 3 stands, and 3) assess the relationship between habitat type and G. sabrinus density.

A northern flying squirrel (AEC image files).

Study Area
The study was conducted in the Huntington Wildlife Forest (HWF) in Newcomb, New York in the center of the Adirondack Park. The 6,070 ha HWF property is owned and maintained by the State University of New York College of Environmental Science and Forestry at Syracuse (SUNY-ESF). The 3 live-trapping grid locations were chosen to represent hardwood, mixed, and coniferous stands. The hardwood trapping grid was located in a stand dominated by sugar maple (Acer sacharum) and American beech (Fagus grandifolia). American beech and red spruce (Picea rubens) were the dominant species in the mixed trapping grid. Finally, the coniferous trapping grid contained primarily red spruce and balsam fir (Abies balsamia).

Hardwood site.
Mixed site.
Conifer site.

Methods

Trapping
Each of the 3 grids contained 42 trapping stations (arrayed in a 6x7 trap configuration). Traps were spaced 33 m apart and each grid was 3.27 ha in area. Two live traps were located at each trapping station: one trap was wired1.5 m high on the closest large tree (< 8 m from the grid point) and the second trap was placed on the ground at the base of the tree. Trapping occurred for 4 days from June 24, 2002 through June 28, 2002, and again for five days from August 5, 2002 to August 9, 2002. The bait was a mixture of peanut butter, oats, and paraffin. The traps were baited each evening and checked early the following morning. When flying squirrels were caught in the traps they were identified, weighed, sexed, and ear tagged.

A trapping station in the hardwood area.

Habitat Assessment
Vegetation sampling was conducted on each of the 3 grids. At each of the 42 stations on a grid, the proportion of hardwood to conifer canopy cover was assessed. At every other station, the density of both hardwood and conifer small stems (<8cm dbh) was estimated. At every fourth plot all snags > 10.2 cm dbh and trees >8 cm dbh were measured and their species were recorded. We also tallied decomposing logs with dbh > 10.2cm.

Estimating cover type and percentage.

Results
During the two weeks of trapping, we caught only 2 flying squirrels. They were both caught on the hardwood grid, resulting in an average of 0.265 squirrels per 100 trap nights for that grid.

Habitat sampling was conducted in June and July of 2002. We found many significant differences between the vegetation of the 3 stands In both percent conifer and percent hardwood cover, the hardwood and mixed grids did not differ from one another but they did differ (P<0.0001) from the conifer grid. The density of both hardwood and conifer small stems was different (P<0.0001) between all three grids. The density of small trees (8-16.7cm dbh) was different (P<0.0001) between the conifer grid, which had a very high density of small conifers, and the other 2 grids. The density of medium sized trees (>16.7cm dbh) was also higher (P=0.0117) in the conifer grid. The density of large trees (>48.3cm dbh), which would be most likely to provide suitable flying squirrel dens, was higher (P<0.0001) in the hardwood grid than in either of the other 2 grids. The density of snags within the 3 grids were not significantly different, but the density of decomposing logs in the mixed grid was higher (P=.0456) than in the other 2 grids.

One of the decomposing logs tallied in the study.

Discussion
Of the 3 areas assessed in my study, the hardwood area probably provided the most suitable G. sabrinus habitat. Although G. sabrinus are known to prefer mature conifer stands to mature mixed or hardwood stands, the relatively small size of the conifers in the conifer area made it unsuitable denning habitat. Because it contained the highest density of trees >48.3 cm dbh, the hardwood area contained the best G. sabrinus den sites. Studies have found that older, larger trees contain a greater number of suitable den sites. G. sabrinus populations are therefore often higher in older, more mature stands. The hardwood area also contained a higher density of small stems than the other two sites. Several studies have found that the presence of deciduous shrubs in the forest understory is correlated with higher flying squirrel abundance because they provide cover from predators while foraging. Although the correlation between understory density and G. sabrinus density is still being investigated, the hardwood area may have provided an advantage in this respect. The one advantage of the mixed area was that it contained more (P=.0456) decomposing logs than the other two areas and may therefore have provided more fungus for the flying squirrels to eat. The fact that both of the flying squirrels captured in this study were captured on the hardwood grid supports the idea that the presence of large trees and abundant understory makes the hardwood area the most suitable flying squirrel habitat despite its species composition.

I captured 0.265 squirrels/ 100 trap nights in the H grid, which is relatively low compared to many other studies. This low capture rate is probably due in part to the fact that none of the 3 areas provided ideal flying squirrel habitat.

Contact info:
For further information on any of these projects, you can contact:
Adirondack Ecological Center: aechwf@esf.edu
Madeleine Fairbairn: mpf9@cornell.edu