Local density and group size interacts with age and sex to determine direction and rate of social dispersal in a polygynous mammal

Abstract

Movement away from an area or social group in response to increasing density (density-dependent dispersal) is known for most species; why it evolves is fundamental to our understanding of ecology and evolution. However, we have yet to fully appreciate how individuals of varying conditions (e.g., age and sex) might differently consider effects of density (quorum) when deciding to disperse or not, and scale dependence in their sense of quorum. We tracked movements of all individuals of a naturalized population of feral horses (Equus ferus caballus; Sable Island National Park Reserve, Nova Scotia, Canada) during a period of rapid population growth (N increased from 375 to 484 horses from 2008 to 2010). Permanent dispersal from breeding groups (bands) was positively density dependent for all age and sex categories with respect to local density (horses/km(2), bounded by the 99th percentile of individual movements [8000 m]), but was negatively and positively density dependent for males and females, respectively, in relation to group (band) size. Dispersal was generally female biased, with the exception of foals which moved with their mothers (no sex effect), and for yearlings and subadults when band sizes were smaller than average, in which case males dispersed at higher rates than females. Dispersal distance was positively related to local density. We conclude that dispersal rate can be both positively and negatively density dependent for feral horses, contingent on the state of individuals and the scale at which quorum with respect to choosing to disperse or not is assessed. Scale effects and interactions of density-dependent and sex- and age-biased dispersal may have both ecological and evolutionary consequences through effects on resource and mate competition.

Keywords: Density dependence; Sable Island; feral horse; quorum sensing; scale; social dispersal.

Figure 1

A band of Sable Island horses. Photo © Philip D. McLoughlin (2008).

Figure 2

Sable Island National Park Reserve (43° 55′ N; 60° 00′ W), located approximately 275 km east–southeast of Halifax, Nova Scotia, Canada. The island is a crescent-shaped sand bar 49 km long and 1.25 km at its widest (vegetation in green).

Figure 3

Variation in local density of adult horses (ages 4+) in bands (horses per km² of vegetated habitat) versus a band's location (UTM easting [x]) on Sable Island, Canada (2008–2010). Each point represents the annual median x-y (UTM northing and easting) centroid of a band; local density is the number of adult horses in bands within 8000 m of each point.

Figure 4

Interaction plots of dispersal probability versus band size for juvenile (yearling–subadult) and adult horses, Sable Island, Canada (2008–2010), at different local densities (low density in black and high density in gray). Density is measured as the number of adult horses in bands within 8000 m of an individual's centroid of movement (per km² of vegetated habitat), presented as 1 SD below (low density: 5.11 horses/km²) and above (high density: 11.1/km²) the mean local density (8.1/km²).