The Social Biology of the Olympic Marmot

Abstract

This report presents the results of a 3-year study of the Olympic marmot in Olympic National Park. The animals inhabited sub-alpine and alpine meadows and formed closelyorganized colonies with extensive burrow systems. Southern exposures were preferred and no correlation of seasonal emergence with snow depth was apparent. Marmots ate nearly all meadow plants plus occasional animal matter early in the season. A variety of natural predators were present, and instances of predation by a cougar and coyote are reported. Visual and accoustic communication is described and sonograms of different vocalizations are presented. Activity patterns paralleled those reported for other marmot species, with early-morning and late afternoon peaks. Detailed activity budgets are presented, indicating behavioural differences among sex and age classes, and with changes in season. Records of weights are presented for all classes, indicating steady seasonal increases and sharp declines during hibernation. A black moult began in mid-summer and is apparently unique to the Olympic marmot.

Most colonies were composed of one adult male, two adult females, a litter of yearlings and a litter of infants. Occasionally one or more 2-year-olds and one or two ‘satellite’ adult males were present. Colony members generally hibernated together but parous females and satellite males moved away to separate burrows shortly after emergence. Varying combinations of living arrangements are described.

Greeting behaviour was analysed and quantified as a major component of marmot sociality. Greeting frequency varied significantly with age and sex class, season and time of day: it was especially high in the morning, among adult males in the spring, and among infants. Dominance relations were generally diffuse, except those between resident and satellite males (resident dominated) and between pregnant or lactating females and any other animal (female dominated). Individual territories and home ranges did not exist within colonies. Quantification of inter-individual distances revealed a tendency for greater aggregation early in the season among most animals but greater earlyseason repulsion between resident and satellite males.

The ontogeny of infant activity is quantitatively described, demonstrating progressively more time spent above ground and increased distance travelled from the home burrow. The decline in association between females and their litters is also described. Upright, ‘play-fighting’ was common and showed age, sex and seasonal differences with high levels among adult males early in the season and among infants. ‘Playfulness’ was prominent in marmot social behaviour. Burrow visiting, greeting, social grouping and playfighting all reflected intense social interaction early in the season with gradual declines later. Flow charts of sexual activity are used to compare the behaviour patterns of parous and nonparous females and to analyse changes in pattern with time following emergence. Distinct oestrus behaviour was apparent for the parous females, peaking during the second week post-emergence. Interactions between colonies are also described: they generally evidenced lower frequencies and greater intensities of agonism than did withincolony behaviours.

The social system of M. olympus was shown to be more closely integrated and less aggresively-organized than that of the yellow-bellied marmot, M. flaviventris, which in turn is more highly social than the woodchuck, M. monax.

Survivorship data indicate that mortality rates were highest among infants and during the winter, especially during winters with scanty snowfall. Two-year-olds dispersed upon seasonal emergence in colonies where winter mortality was low; however, following high winter mortality and consequent small colony size at emergence, dispersal did not occur. Thus, colony population size was maintained essentially constant and relatively independent of yearly fluctuations in mortality or recruitment. Dispersal from large colonies was correlated with behavioural responses of the 2-year-olds to a relatively high level of social interaction in these colonies.

Olympic marmot females mature at 3 years and bear young in alternate years only. The evolutionary significance of biennial breeding is considered, and this population system is interpreted as providing maximum effective reproduction under the conditions of extremely short growing season prevailing in the marmot meadows.

Comparison with other North American marmots reveals a progressive decrease in length of growing season experienced by the wood-chuck, yellow-bellied and Olympic marmots. This in turn correlates with decreases in proportional growth rates and breeding frequency, and increases in age at dispersal, age at sexual maturity and in the degree of ‘sociality’ of these marmot species. Implications for the evolution of marmot behaviour are discussed, including the suggestion that exposure to progressively shortened growing seasons favours the development of less aggressively-organized social systems.