A Comparison of Methods for Studying Life History in Columbian Ground Squirrels

Abstract

We reviewed data from three separate studies that examined the life-history char- acteristics of Columbian ground squirrels, Spermophtlus columbianus, at different elevations in the Rocky Mountains of southwestern Alberta, Canada. One study used a stitic (vertical) life- table methodology, and the other two used a cohort (horizontal) life-table methodology. One of the latter studies also applied an experimental supplementation of food resources to two popu- lations. Absolute values of characteristics measured by static and cohort methodologies were not directly comparable, but the pattern of change with elevation of litter size, adult survival, and body weight were similar. At high, relative to low elevations, litter size was lower, adult survival was higher, and body weight of adult females was lower. The two methodologies produced conflicting results regarding the proportion of females reproducing and juvenile survival at dif- ferent elevations. The static methodology does not reveal the changes in life history that may occur in different years. An experimental method was required to demonstrate phenotypic plas- ticity of life-history characteristics. In two populations that received supplemental food, litter size, juvenile survival, and the body weight of adult females increased. Extensive information on characteristics of life history is available for only a few mammalian species (Millar and Zammuto, 1983; Stearns 1977). Several methods of amassing life-history data exist (Caughley, 1977; Deevey, 1947), and because such data are difficult to gather, different methods have seldom been applied to single species. Thus, one rarely knows if disparities in results for different species studied by different methods reflected real differences, or resulted from the application of different methods. Our purpose is to indicate whether different methods can be combined by population ecologists to strengthen empirical conclusions about natural patterns of variation in life history, and hence provide some indication of comparability of different methods. We briefly describe three independent research programs that examined variation of life- history characteristics in mountain-dwelling populations of Columbian ground squirrels, Sper- mophilus columbianus, at different elevations. One of the studies produced a static (vertical) life table, and the other two produced cohort (horizontal) life tables. One of the latter studies examined changes in life history under experimental supplementation of food resources. By comparing the data obtained by alternative means of life-table construction and alternative study approaches (comparative and experimental), we were able to identify similarities and differences among studies. The differences in particular might be of interest to those designing studies of mammalian life history who need to choose between alternative means of life-table construction or need to apply experimental techniques. METHODS Methods of data collection varied among the research programs. Zammuto (1983; Zammuto and Millar, 1985) collected static life-table data by kill-sampling six ground squirrel populations during pregnancy and early lactation. Three populations were collected in 1980, and three were collected in 1981. Males were designated adult if they had scrotal testes or a darkly pigmented scrotum, and females were designated adult if they contained embryos or placental scars. Litter size was estimated by counting embryos or placental scars in utero, and reproductive females were identified by the presence of embryos or recent placental scars. In each population, juvenile survival was estimated from the number of yearling ground squirrels divided by the total number of embryos and placental scars dissected from females. The ages of