Variability in the distribution and abundance of stream salmonids, and the associated use of habitat models

Abstract

The effective management of salmonid fisheries requires that the factors influencing variation in the abundance of stream populations are understood. The use of habitat models to explain the spatial component of population variance offers potential for management, but has not previously been set in the context of long term variation in population abundance because of the lack of suitable data sets. This paper examines contributions of spatial and temporal factors lo fish density variance using a 10‐year data set from five tributaries of the River Conwy, North Wales. Recently developed habitat models were applied to the data to test their ability to explain nominal spatial variance. Spatial variance accounted for between 21 and 62% of the overall variance of salmonid abundance, and habitat models explained up to 95% of the spatial variance component. Synchrony in population variation amongst sites within and between tributaries is described, and some of the factors that may influence this are discussed.

Variabilité des populations de salmonidés en eau courante : étude de la variance liée à différentes échelles spatiales ou temporelles et implications pour les modèles d'habitat

“…The methods outlined here refer to two différent exercises : a habitat modelling and catchment scale variance partitioning on the Conwy (MILNER et al, 1993) and a more récent development using improved techniques in a national scale habitat modelling project (WYATT etal., 1995).…”

“…Model parameters have been reported elsewhere (MILNER et al, 1993). Performance as measured by the percentage of population density (independent data set) variance (R^, adj %) explained by the models were 72.4, 66.2, 48.3 and 26.2 % for trout 0*, trout >0"^ and salmon >0* respectively.…”

“…Most (88 %) Conwy sea trout and salmon migrate to sea at âge 2 years (NRA, 1992). This is reflected in the âge composition of trout populations such that >2* fish are absent or infrequent (<1.5 %) in streams supporting migratory populations (MILNER et al, 1993). Because of the very low incidence of >2* fish the data following are considered in terms of late summer O-i-(fry) and >0* (parr).…”

Le développement des modèles d'habitat pour les salmonidés d'eau courante et leur application à la gestion piscicole

“…Earlier work (MILNER et al, 1993) demonstrated the need to examine the contributions made by the différent potential sources of variation to the overall variablity of the data-set. Techniques were devised to effectively 'partition' the variance in a data-set to allow the true variation due to spatial factors to be derived (see, for example, MILNER et al, 1993).…”

“…Techniques were devised to effectively 'partition' the variance in a data-set to allow the true variation due to spatial factors to be derived (see, for example, MILNER et al, 1993). In gênerai terms, the spatial variance component is all that can be hoped to be accounted for by habitat models.…”