| Summary: | This thesis addresses local population consequences of variability in abundance, habitat use and catchability as expressed by Atlantic salmon (Salmo salar) juvenile and adult populations in a riverine system. Ranges of abundance are extremely broad among different sampling sites, and classical methods of estimating population abundance lack generality, because they do not use all available information and, in combination with low catchability, often yield unreasonable estimates. A method that simultaneously uses all sampling information from several sites or sampling occasions is developed to estimate local population sizes. It is based on the reconstruction of a statistical (Beta) distribution function from observations within the population. A mechanism is proposed of habitat use as a function of population expansion and contraction with changes in overall population abundance. Variation in local densities proves consistent with the hypothesis that response to fluctuation in population abundance occurs mainly in marginal habitats. Preferred habitats also provided the fastest and most stable growth rates. This accords with more general theories of density-dependent habitat use and on ranges of populations. The scale of density-dependent growth is shown to be much greater than implied by local population density of juveniles, and is also affected by overall population abundance. Furthermore, the strength of density-dependent relationships increases downstream, indicating cumulative effects. Variability of density-dependent growth is shown to be habitat specific and offers further support to the population expansion-contraction mechanism. Some general consequences of habitat-specific density-dependence are discussed. Thesis (Ph.D.)--Dalhousie University (Canada), 1994.
|
|---|