| White-footed mice (Peromyscus leucopus) are one of the most prevalent species in deciduous forest ecosystems in eastern United States and depend on coarse woody debris (CWD) for travel routes, nesting areas, shelter, and a stable thermal environment. In May 2003 an F4 tornado with wind speeds ranging from 330-416 km/h traveled through Mermet Lake Conservation Area (MLCA) in southern Illinois and impacted 166 ha of forest by uprooting and splintering trees. My objective was to compare the population dynamics of white-footed mice in disturbed and undisturbed habitats. Three transects (B-undisturbed, C-disturbed/unsalvaged, and D-disturbed/salvaged) were sampled once a month from October 2004-October 2005. A fourth transect (E-undisturbed/burn) was sampled once a month from December 2004-October 2005. Five-hundred forty-seven individual white-footed mice were captured 1355 times at MLCA. They composed 77.9% of the 702 individual small mammals captured during 8454 trap nights, and were the only species with a large enough sample size to be included in the analyses. Minimum number known alive enumeration method indicated that most of the white-footed mice captured during the 2004-2005 trapping period were located within the tornado-disturbed transects. Overall, population measures on all four transects peaked during November 2004 through January 2005, then began to decrease toward the end of the trapping period. The numbers of white-footed mice captured on disturbed transects were an order of magnitude greater than on the undisturbed transects. Repeated measures ANOVA indicated the undisturbed and disturbed sites were significantly different in their monthly abundance estimates (F=39.25, df=1,2, P=0.0245). There was a highly significant time effect (F=9.13, df=12,22, P<0.0001) in monthly abundance, as well as a highly significant interaction between the treatment and time effects (F=6.57, df=12,22, P<0.0001). Results from a post-hoc Tukey test indicated that abundance was consistently higher on disturbed than undisturbed transects. Apparent survival was analyzed using the Cormack-Jolly-Seber model in program MARK. On transect B (undisturbed), the average monthly rate of survival was 0.38 for males and 0.71 for females. On transect E (undisturbed/burn), the average monthly rate of survival was 0.32 for juveniles and 0.78 for adults. For transect C (disturbed/unsalvaged), the monthly survival rate was 0.75 for sexes and ages combined. For transect D (disturbed/salvaged), the monthly survival rate was 0.82 for sexes and ages combined. More reproductively active males and females were captured on the disturbed transects, although on a percentage basis, transects were similar. During the summer and winter, the proportion of canopy cover and number of standing trees were significantly greater on undisturbed sites (B and E) than disturbed sites (C and D). Additionally, the proportion of ground and midstory cover were significantly greater on disturbed transects than undisturbed transects. Differences were also detected between the two undisturbed transects and two disturbed transects, and were attributed to seasonal variation and different silviculture methods used in these forest sites. When forested areas are impacted by natural or anthropogenic disturbances, managers should strive to leave a mosaic of microhabitats including coarse woody debris and associated downed logs, snags, and dense understory cover. |
