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

Abstract

Trout (Salmo trutta, L.) and salmon (Salmo salar, L.) populations in streams exhibit temporal and spatial variation. However, the ability of habitat models (empirical models relating fish abundance to spatial features) to explain overall variance in abundance is restricted just to the spatial component. It is therefore important to be able to quantity the contribution from the spatial component to total variance. This allows assessment of both the potential maximum performance of models as well as their actual performance in relation to the maximum. Furthermore, such variance partitioning offers insight into the relative rôles of spatial and temporal (synchronous) factors in influencing population abundance and how thèse vary according to the geographical scale of sampling. Habitat models (HABSCORE), recently developed for Welsh streams, were used to explain variance in a ten year data set for which temporal and spatial variance could be estimated. Spatial factors explained between 46 and 62 % of overall vahance within the Conwy System. This identifies the maximum limits for the performance of such models working at this scale. With the exception of poor performance for salmon parr, the habitat models accounted for 60-95 % of the spatial component, corresponding to 38-46 % of overall variance. In addition, variance structure was compared at four différent levels of analysis : within tributaries on the Conwy, within nine différent large separate river Systems, within three areas and within the région of Wales. Spatial variance increased from 22-42 % (means) at within-tributary level to 42-65 % at régional level. In contrast, temporal variance (a measure of synchrony in population variability) decreased from 24-39 % within tributaries to 0.7 - 9.0 % at régional level. At within-rivers and larger scale the temporal variance displayed in 0-H abundance was consistently lower than that for >0-i- fish. Some of the factors influencing variability at the différent geographical scales are briefly discussed.