Abstract

Territorial behaviour is thought to play a role in limiting population density. Consequently the factors that affect territory size would also influence population density. I examined the relationship between visibility and territory size in young-of-the year (YOY) rainbow trout ( Oncorhynchus mykiss ) to test the hypothesis that increasing habitat heterogeneity results in a reduction in territory size and consequently in higher population density. As predicted, the territory size decreased with decreasing visibility. However, the decrease in territory size did not produce an increase in density, perhaps because few individuals defended territories or because the experiment was too short for population density to reach an equilibrium. The optimal size of a contiguous territory is predicted (1) to remain unchanged at low food abundance levels and (2) to decrease only when food abundance is high enough to reduce territory size below the contiguous optimum. I manipulated food abundance to test this model, using YOY steelhead trout ( Oncorhynchus mykiss ). Increasing competition resulted in increasing mortality, higher propensity to emigrate, higher variance in body mass, lower growth, lower population density, lower biomass and lower percent habitat saturation. Territory size did not change with food abundance. Increasing levels of intraspecific competition in stream-dwelling salmonid populations often lead to density dependent mortality and emigration. However, density dependent growth is less frequently detected. I examined the relationship between average fork length and density of YOY Atlantic salmon ( Salmo salar ), to investigate (1) whether there is evidence for density dependent growth, (2) the shape of the relationship, and (3) the effect of spatial and temporal scale on the ability to detect density dependent growth. There was a negative relationship, described by a power curve, between the average body size and density of YOY Atlantic salmon. Most of the variation in body size was explained by YOY density, with year, location and older salmon density accounting for a minor proportion of the variation. Density dependent growth was equally well detected within and across years. Spatial scale did not affect the ability to detect density dependent growth. My analysis suggested that YOY Atlantic salmon populations are regulated by two different mechanisms: density dependent growth at low densities and density dependent mortality at high densities.