Factors affecting nest box use by Southern flying squirrels (Glaucomys volans) and gray squirrels (Sciurus carolinensis).
Southern flying squirrels (Glaucomys volans) inhabit a variety of forested habitats throughout eastern North America, occurring most frequently in mixed pine (Pinus sp.)-hardwood (Goertz et al., 1975; Heidt, 1977; Sonenshine and Levy, 1981; Fridell and Litvaitis, 1991) and hardwood stands (Weigl, 1978; Gilmore and Gates, 1985; Bendel and Gates, 1987). Many studies have used nest boxes to examine the ecology of this species (e.g., Goertz et al., 1975; Heidt, 1977; Sonenshine and Levy, 1981; Gilmore and Gates, 1985). High usage of nest boxes by flying squirrels has been reported; the majority of boxes are used for nesting or feeding stations (Heidt, 1977; Sonenshine et al., 1979; Gilmore and Gates, 1985). Our study examined variables that might influence the use of nest boxes by southern flying squirrels and associated gray squirrels (Sciurus carolinensis).
The study was conducted in the U. S. Forest Service Camp Clearfork and Charlton Recreation Area within the Ouachita National Forest, 3 km W of Crystal Springs; Garland Co., Arkansas. The composition of this 70-yr-old forest is approximately 70% shortleaf pine (Pinus echinata) and 30% hardwood. The hardwoods are dominated by oaks (Quercus spp.), with sweetgum (Liquidambar styraciflua), blackgum (Nyssa sylvatica) and hickory (Carya spp.) interspersed.
Three grids, each consisting of 30 nest boxes placed 35 m apart in three rows (35 m apart) of 10, were established during March 1990. Boxes were placed on pines and hardwoods at heights ranging from 2.7-4.6 m. Box design was modified from Heidt (1977), with internal dimensions measuring approximately 25 x 23 x 23 cm. Box openings were 3.8 (small) or 6.4 (large) cm in diameter and were alternated throughout the grids. The grids included small-holed nest boxes located on 31, 27 and 25 hardwoods and 14, 13 and 13 pines, and large-holed nest boxes on 25, 29 and 30 hardwoods and 20, 21 and 21 pines for [TABULAR DATA FOR TABLE 1 OMITTED] the 1st, 2nd and 3rd yr of the study, respectively. Discrepancies in numbers among years are explained by the fact that a few small entrances were enlarged during the study. Also, one small-holed box was destroyed, during the 3rd yr of the study, leaving only 89 boxes.
Boxes were checked during daylight hours at 2-wk intervals from March 1990 to March 1993. Exceptions were during June, July and August of each year when, due to decreased usage by flying squirrels, they were checked every 3 wk (similar decreases have been noted by others; e.g., Sawyer and Rose, 1985). During these months, the contents of each box were removed. Thus box usages for each year are considered to be independent of the previous year.
We classified each box as: (1) a primary nesting site of flying squirrels; (2) a feeding station of flying squirrels; (3) a nesting site of gray squirrels, or as (4) "other." Primary nesting sites of flying squirrels had large amounts of nesting material (shredded bark, moss or lichens) and were consistently used by several squirrels simultaneously. Feeding stations of flying squirrels were characterized by middens (with gnawings characteristic of flying squirrels) that included remains of acorns, pine cones, bracts or insects. Nests of gray squirrels were primarily composed of leaves. The "other" category included miscellaneous usages that occurred with low frequency.
Nest box use declined during summer months, so our classification was for October 1990 through March 1991 (yr 1), October 1991-March 1992 (yr 2) and October 1992-March 1993 (yr 3). Thirteen cavity and microhabitat variables (Table 1) were measured in each plot, during February-March 1991 and 1992, using a 700-[m.sup.2] (radius = 15 m) circular plot established around each nest box.
Nest box entrance diameter and nest box tree type, both categorical variables, were tested for significance using chi-square contingency tables. Significance was set at P [less than or equal to] 0.1 for all comparisons to decrease the probability of a type 2 error (Askins et al., 1990). Entrance diameter for nests of flying squirrels was compared to entrance diameter for the other three categories for each year. The same procedure was used for entrance diameter of feeding stations and for nests of gray squirrels. These comparisons established differences between species regarding nests; subsequent analyses of nests of flying squirrels were confined to small-holed boxes and those of gray squirrels to large-holed boxes. Feeding stations of flying squirrels were compared to all other categories because they were in both large- and small-holed boxes.
Significance among 11 continuous variables was tested with Wilcoxon two-sample tests for each of the 3 yr (P-values reported for yr 1, 2 and 3, respectively). Univariate and multivariate statistics were calculated using Statistical Analysis Systems (SAS; SAS Institute, 1989). Because multivariate statistics did not clarify our analyses, only univariate results are reported.
Nest box entrance diameter (see Table 1 for means and standard deviations for cavity and microhabitat variables) was significantly associated with nests of flying squirrels (P = 0.000, 0.000 and 0.001, respectively) for all 3 yr. With one exception, nests of flying squirrels (n = 43) were established in boxes with small openings. Nest box entrance diameter was also significantly associated with nests of gray squirrels for all 3 yr (P = 0.000, 0.040 and 0.002, respectively). Nests of gray squirrels (n = 34) were established exclusively in boxes with large openings. No significant association was present between nest box entrance diameter and feeding stations; P = 0.809, 0.107 and 0.214 for all 3 yr, respectively.
Chi-square tests, which took into account the proportion of boxes placed on hardwoods vs. pines, showed no significant association between nest box tree type (hardwood vs. pine) and nests of flying squirrels (P = 0.693, 0.305 and 0.744). Significant associations were shown between nest box tree type and feeding stations (n = 87) for all 3 yr (P = 0.019, 0.014 and 0.078), and tree type and nests of gray squirrels for 2 of the 3 yr (P = 0.066, 0.070 and 0.309). Feeding stations of flying squirrels were established significantly more often in boxes on hardwoods than pines, and gray squirrels established nests in boxes on pines significantly more frequently than hardwoods. Remaining variables that were sometimes significant were not consistently informative over the 3-yr study (Table 2).
Previous investigators (i.e., Gilmore and Gates, 1985; Bendel and Gates, 1987; Boardman, 1991) determined that variables such as total shrub-layer stem density, amount of upper- and lower-understory cover, density of tree cavities, bark texture and densities of medium snags and large stumps are important indicators of suitable habitat for flying squirrels. However, data from these studies were collected over relatively brief time periods (approximately 1-yr). In our study, some of these variables differed significantly among used and unused boxes, but not during all 3 yr. Variables that are useful indicators of habitat suitability should exhibit consistently significant differences over time provided the habitat is not appreciably altered. Because most variables were not significant for all 3 yr, other factors (e.g., behavior of the species, site fidelity, microhabitat characters not measured, climate or synergistic effects) probably influenced nest box use by flying squirrels.
Southern flying squirrels and gray squirrels segregated with regards to nest box use. As suggested by Dolan and Carter (1977), flying squirrels minimize nest site overlap by establishing nests in cavities with openings that exclude gray squirrels which are, on average, much larger than flying squirrels as adults. In our study, gray squirrels rarely (n = 6) enlarged smaller holes; however, Heidt (1977) reported gray squirrels do enlarge openings [TABULAR DATA FOR TABLE 2 OMITTED] when only boxes with small holes are provided. Perhaps the availability of boxes with large holes on pine trees precluded this activity. Although not a variable in our study (because all boxes had the same internal dimensions), cavity volume may also be important in nest site selection by these species.
The only other variable seemingly selected for by flying squirrels was nest box tree type (hardwood vs. pine) when establishing feeding stations. Hard mast constitutes a majority of the diet of flying squirrels (Dolan and Carter, 1977; Harlow and Doyle, 1990) and establishing feeding stations in boxes on hardwoods provided easier access to hard mast.
Our results provide useful insight for the placement of nest boxes in future studies of flying squirrels and gray squirrels. Because there was no significant difference in nest box height between used and unused boxes (heights ranging from 2.7 to 4.6 m), they can be placed at heights convenient to the researcher. We recommend not placing nest boxes on pine trees if reducing interactions between flying squirrels and gray squirrels is an objective of the study (use more pines if interaction is desirable). Nest box hole size is an important consideration in that smaller openings will generally exclude gray squirrels. To deter hole enlargement, supplemental boxes with large holes should be placed on pines or the use of restrictors might be considered. We demonstrated that year-to-year variation in usage is a likely possibility and thus inferences based on data from only 1 yr should be viewed with caution.
Acknowledgments. - We extend our appreciation to W. S. Hargis and U.S. Forest Service personnel, including D. A. Saugey, D. M. Harris, L. J. Hogeboom, E. D. Sharpe and R. A. Raines for their assistance with the project, and two anonymous reviewers for their editorial comments. This study was partially funded by the U.S. Forest Service, Ouachita National Forest, Womble District, Arkansas.
ASKINS, R. A., J. F. LYNCH AND R. GREENBERG. 1990. Population declines in migratory birds in eastern North America. Curr. Ornithol., 7:1-57.
BENDEL, P. R. AND J. E. GATES. 1987. Home range and microhabitat partitioning of the southern flying squirrel (Glaucomys volans). J. Mammal., 68:243-255.
BOARDMAN, J. L. 1991. Nest site selection by southern flying squirrels (Glaucomys volans) in north-eastern Missouri. M. S. Thesis, Northeast Missouri State Univ., Kirksville. 67 p.
DOLAN, P. G. AND D. C. CARTER. 1977. Glaucomys volans. Mamm. Species, 78:1-6.
FRIDELL, R. A. AND J. A. LITVAITIS. 1991. Influence of resource distribution and abundance on home-range characteristics of southern flying squirrels. Can. J. Zool., 69:2589-2593.
GILMORE, R. M. AND J. E. GATES. 1985. Habitat use by the southern flying squirrel at a hemlock northern hardwood ecotone. J. Wildl. Manage., 49:703-710.
GOERTZ, J. W., R. M. DAWSON AND E. E. MOWBRAY. 1975. Response to nest boxes and reproduction by Glaucomys volans in northern Louisiana. J. Mammal., 56:933-939.
HARLOW, R. F. AND A. T. DOYLE. 1990. Food habits of southern flying squirrels (Glaucomys volans) collected from red-cockaded woodpecker (Picoides borealis) colonies in South Carolina. Am. Midl. Nat., 124:187-191.
HEIDT, G. A. 1977. Utilization of nest boxes by the southern flying squirrel, Glaucomys volans, in central Arkansas. Proc. Arkansas Acad. Sci., 31:55-57.
SAS INSTITUTE INC. 1989. SAS/STAT user's guide. Version 6 ed. SAS Inst. Inc., Cary, NC. 1686 p.
SAWYER, S. L. AND R. K. ROSE. 1985. Homing in and ecology of the southern flying squirrel (Glaucomys volans) in southeastern Virginia. Am. Midl. Nat., 113:238-244.
SONENSHINE, D. E., D. M. LAUER, T. C. WALKER AND B. L. ELISBERG. 1979. The ecology of Glaucomys volans (Linnaeus, 1758) in Virginia. Acta Theriol., 24, 26:363-377.
----- AND G. F. LEVY. 1981. Vegetative associations affecting Glaucomys volans in central Virginia. Acta Theriol., 26, 23:359-371.
WEIGL, P. D. 1978. Resource overlap, interspecific interactions and the distribution of the flying squirrels, Glaucomys volans and Glaucomys sabrinus. Am. Midl. Nat., 100:83-96.
|Printer friendly Cite/link Email Feedback|
|Author:||Stone, K.D.; Heidt, G.A.; Baltosser, W.H.; Caster, P.T.|
|Publication:||The American Midland Naturalist|
|Date:||Jan 1, 1996|
|Previous Article:||Female aggression in Microtus pennsylvanicus: arena trials in the field.|
|Next Article:||Invasion of the meadow vole (Microtus pennsylvanicus) in southeastern Kentucky and its possible impact on the southern bog lemming (Synaptomys...|