Work

Home Range Size and Habitat Selection of Timber Rattlesnakes (Crotalus horridus) in Southwestern Wisconsin

Year: 2009
Type: dissertation
Abstract: The timber rattlesnake (Crotalus horridus) has experienced rangewide declines at \nleast partially as a result of human exploitation and persecution, is listed as a threatened \nspecies in at least 9 states, and has received ―protected wild animal‖ status in Wisconsin. \nWhile some persecution persists, current impacts on the species are thought to be largely \nassociated with habitat alteration, including human development and vegetative \nsuccession. Knowledge of home range size and habitat selection is essential to \nconservation of timber rattlesnakes in Wisconsin, which are now experiencing increased \nconversion of critical habitats for residential use. However, the ecology of this species in \nthe Upper Mississippi region is poorly understood. My objectives were to determine if \nhome range size or habitat selection of timber rattlesnakes (1) differed among groups of \nmales, non-gravid females and gravid females and (2) between protected natural areas \nand a residential development. \nI used radio telemetry to track 36 timber rattlesnakes in southwestern Wisconsin \nduring 2007 and 2008. I generated minimum-convex polygon home ranges for 34 of the \nsnakes, which were tracked for periods of 11 to 20 weeks between May and October of \neach year. I used factorial analyses of variance (ANOVA) to compare total distance \nmoved (TD), average daily distance moved (ADD), and home range area (HR) among \nmale, non-gravid female, and gravid female rattlesnakes and between two protected \nnatural areas and a residential development. Mean movement did not differ significantly \nbetween males (TD: 2981 m [95% CI 2238.8 to 3968.3 m]; ADD: 27.7 m [95% CI 21.0 \nto 36.5 m]; HR: 21.1 ha [95% CI 11.4 to 39.0 ha]) and non-gravid females (TD: 2082 m \n[95% CI 1520 to 2851 m]; ADD: 21.8 m [95% CI 16.1 to 29.5 m]; HR: 13.3 ha [95% CI \n6.7 to 26.1 ha]). However, movement of gravid females (TD: 483 m [95% CI 351 to 663 \nm]; ADD: 4.9 m [95% CI 3.6 to 6.6 m]; HR: 0.7 ha [95% CI 0.3 to 1.3 ha]) was \nsignificantly smaller (p < 0.001 for all parameters) than for males or females. Movement \nand activity of rattlesnakes did not differ significantly between sites. The size of male \ntimber rattlesnake home ranges was positively related to snake length (R^2 = 0.73, p < \n0.001) but not for either gravid (R^2 = 0.01, p = 0.815) or non-gravid females (R^2 = 0.21, p \n= 0.157). \nI used both discriminant function analysis (DFA) and logistic regression with \nAkaike’s Information Criterion (AIC) model selection to model timber rattlesnake habitat \nselection based on 8 habitat variables. Variables were measured at locations used by \nsnakes (n = 523) and random locations assumed to be available to the snakes (n = 519). \nMeasurements from locations were averaged for each snake to create vectors of snake \nhabitat use (n = 36) and habitat available to snakes (n =36) for analysis. The first two \ndiscriminant functions identified percent cover of rock, presence of brush piles, and tree \nstem density as the most important habitat features associated with rattlesnake habitat \nuse. This DFA model was similar to the best approximating logistic regression model \nidentified from AIC model selection. The third discriminant function in the DFA model \nincluded relative predominance of eastern redcedar (Juniperus virginiana) as an \nimportant habitat feature at sites used by timber rattlesnakes, as did the second-best \nsupporting model from logistic regression (ΔAICc = 1.11). Habitat selection of timber \nrattlesnakes was also influenced by reproductive status. Gravid females used areas with \nrelatively high amounts of rock. Males and non-gravid females used areas of higher tree \ndensity and brush piles, particularly in the developed area. Habitat use by gravid females \nalso differed between sites, with snakes in the residential development using areas with \nhigher relative dominance of eastern redcedar than snakes in the natural area treatment, \npresumably because rocky areas without eastern redcedar were scarce in the developed \narea. \nHome ranges and habitat selection of timber rattlesnakes in southwestern \nWisconsin vary as a function of both sex and reproductive status. Therefore, \nmanagement strategies designed to conserve this species in the region should reflect these \ngender- and reproduction-related differences. Gravid females are relatively stationary, \nand small patches of habitat should be suitable as long as the requisite features of surface \nrock and open forest canopies are present. Males, however, range widely and their home \nranges encompass a broader array of habitat features. Therefore, conservation of males \nwill require broader-scale conservation measures with a focus on minimizing mortality \non roads and direct human-rattlesnake conflicts. Inclusion of movement corridors in \nresidential developments with strategic placement of brush piles may facilitate timber \nrattlesnake movements while minimizing risks to humans.
Language: English
Open Access status: green
OpenAlex ID: W3152904117
Cites: 86
Cited by: 5
Related to: 9