Evaluation of Within-stand Differences in Small Mammal Populations and Habitat

Tim Marvin

A trapping station next to a rock.

Background
Many factors influence how small mammals select habitat. Small mammal habitat has been defined on a broad stand-level scale, and investigated by forest types, stand age, and managed versus unmanaged stands, while other stand level studies have looked at food sources as a primary correlation with small mammals habitat. Past studies indicate there is little difference in density of small mammals between different forest stand types. However, it has been noted that mammals can respond to variables at a smaller, within-stand, scale, since stand conditions can be diverse and support a variety of small mammal habitats at any stage of its development. My study investigated microhabitats within a stand and focused upon variables such as percent rock, coarse woody debris (CWD), vegetation cover and open ground. I also estimated density of large trees and density of small woody stems within the stand. Species composition and density data for small mammals were taken from a long term monitoring project of small mammals in the Bureau Brothers Turn (BBT) stand at the Huntington Wildlife Forest. The objectives of this study were to 1) compare species composition and density of small mammal populations between 2 sites within a forested stand, 2) compare important microhabitat variables between 2 sites within a forested stand, and 3) to compare the study sites based on the small mammal and microhabitat data collected to find a possible correlation between mammal species and microhabitat variables.

Methods

Small mammal trapping
Two grids were set up in the BBT stand to sample for small mammal density and microhabitat variables. Small mammals were trapped in the BBT stand during June and September of 2001 and June 2002. Each grid was 120 m X 210 m consisting of 28 trapping stations, 30 m apart with 2 Sherman live traps at each station. The animals that were captured were evaluated, tagged to identify re-captures and released.

Peromyscus spp. in capture bag.

Microhabitat sampling
Microhabitat variables were sampled on both of the small mammal trapping grids. Using a measuring tape, a 22.6 m transect line was set bisecting the 0.04 ha plot in the cardinal directions, centered upon each trapping station to estimate ground cover. At 2 m intervals along the measuring tape, the presence of rock, CWD, vegetation, open ground (i.e. leaf litter, bare soil), or "other" (i.e. water) was noted. Coarse woody debris was defined as a dead woody object > 7.6 cm diameter and > 30.4 cm in length; rock cover was defined as any rock > 7.6 cm diameter and not covered by leaf litter or vegetation; vegetation cover was defined as any live vegetation < 0.5 m tall; open ground was defined as leaf litter or soil and no other variables.

Coarse woody debris on the grid.

At every other trapping station a small stem tally was made for small woody stem density. Stems < 4 cm DBH intersecting a 2 m pole held horizontally at breast height while walking each transect were tallied as hardwood, conifer or shrub.

At 5 trapping stations on each grid I tallied trees > 20 cm DBH by species within the 0.04 ha plot. This was done to estimate tree density and also to make inferences about the potential for seed production on each grid. I restricted tallying trees to > 20 cm DBH because it is the approximate size of trees at seed bearing age.

Tim measuring the DBH of a tree.

Results

Population data
Small mammals were trapped in June and September 2001 and June 2002 for a total of 1,680 trap nights. I found a statistical difference between new captures of B. brevicauda, (P= 0.004), C. gapperi (P= 0.276), M. chrotorrhinus (P= 0.057), N. insignis, (P= < 0.001), S. cinereus (P= 0.07) and T. striatus (P= <0.001) captured in grids A and B. There was no statistical difference in the number of new captures of Peromyscus spp. (p= 0.439) between grids A and B (Table 1).

Table 1. Results of T-tests comparing means of first captures over 1,680 trap nights on 2 grids in the BBT stand. Captures during June-September, 2001, and June 2002.

BLBR = Blarina brevicauda, CLGA = Clethrionomys gapperi, MICH = Microtus chrotorrhinus, NAIN = Napaeozapus insignis, PEMA = Peromyscus spp, SOCI = Sorex cinereus, TAST = Tamias striatus

Microhabitat data
Microhabitat variables were samples in June-July 2002 in the 2 trapping grids at BBT. A statistical difference was found between percent CWD (P= 0.008), rock (P= < 0.001), open ground (P= 0.006) on grids A and B. Percent vegetation cover and "other" did not have a statistical difference between the 2 sites (Table 2).

Table 2. Results of T-tests and means of habitat variables in 2 grids in the BBT stand. Data collected June 2002.

CWD = Coarse Woody Debris, Veg = Vegetation

Spearman correlations
A positive correlation (rs= 0.46, P= 0.015) was found between B. brevicada and percent open ground. There was a negative correlation (rs = -0.89, P= 0.041) between B. brevicauda and density of A. sacharrum. N. insignis showed a positive correlation (rs =0.35, P= 0.068) with percent rock cover. T. striatus showed a positive correlation (rs =0.61, P= 0.023) with the shrub small stem density, and a negative correlation (rs = -0.884, P= 0.046) with density of F. grandifolia. No other correlations were significant (Table 3).

Table 3. Results of Spearman correlation statistics (rs) of small mammal abundance by species and habitat variables at BBT stand.

H.S.S. = Hardwood small stems, C.S.S. = Conifer small stem, S.S.S. = Shrub small stems,
ACSA = Acer sacharrum, FAGR = Fagus grandifolia

Conclusions
This study began with the objective of defining small mammal habitat on a microhabitat level and to find a correlation between mammal species and microhabitat variables within a stand. The diversity between the microhabitat variables in this study site shows that variation can occur on a smaller within-stand level and not just a broad, between stand scale. Upon completion of this study I found many correlations that were unexpected and though significant, did not appear to make biological sense.

In conclusion, I found that there is no one specific microhabitat variable that can be associated with a specific small mammal species. However, there may be other variables influencing the mammals. Future investigation of other variables could prove to be beneficial to understanding the relationship between these animals and their habitat.

Captured redback vole.

Contact info:
For further information on any of these projects, you can contact:
Adirondack Ecological Center- aechwf@esf.edu
Tim Marvin- tmarvin@syr.edu

About the author:
I am currently an undergraduate at The SUNY College of Environmental Science and Forestry. I am studying wildlife biology and hope to pursue a graduate program on carnivores after my bachelors. I hope to get a job with the Fish and Wildlife Service after graduation. Other hobbies include hiking, mountain climbing and generally anything outdoors.