MULTIVARIATE-ANALYSIS OF GOLDEN MARMOT MAXIMUM RUNNING SPEED - A NEW METHOD TO STUDY MRS IN THE FIELD

I develop a multivariate technique permitting the study of variation in the maximum velocity of animals. The method was applied to study maximum running speed (MRS) of adult golden marmots (Marmota caudata aurea). Subjects were encouraged to run to home burrows after being livetrapped, and were timed while running over a randomly selected distance between 1.7 and 36.0 m. Body mass, sex, distance run, substrate, and incline the individual ran across for 52 different subjects were entered into a linear model, which implicitly modeled MRS as a function of the running time. Body mass did not significantly explain variation in running time and was deleted. A new model fitted to the remaining variables significantly explained variability in time, and therefore MRS. The model did not violate any of the assumptions of linear models, and appeared to be robust. Intrayear repeatability of individual MRS was small, suggesting that environmental factors may account for much of the variation in MRS in golden marmots. The model predicts that the four significant variables (sex, distance run, substrate, incline) should influence space use and patch selection in golden marmots since they influence the speed at which an alarmed marmot can reach a refuge. The general method developed here can be applied to other species and should shed light on patch use and foraging behavior of refuging species.