RELATIONSHIPS BETWEEN HABITAT FEATURES AND BROWN TROUTS POPULATIONS (SALMO-TRUTTA L) IN NESTE-DAURE VALLEY

The relationships between habitat features and biomass and densitiy of brown trout (Salmo trutta L.) were studied in 33 sections of the Neste d'Aure stream and three of its tributaries: the Neste du Louron, the Neste du Rioumajou and the Espiaube stream in the Hautes-Pyrenees region. The Habitat Quality Index (HQI) model I (BINNS and EISERMAN, 1979), based on 10 environmental variables, was tested. The biomass predicted by the model were not linearly related with biomass of trout observed by electrofishing on the 33 sections. The best linear model was obtained after logarithmic transformations of the two variables. However, the slope of the regression line was significantly different from 1 (t=2.53 (p<0.01)).The HQI model did not appear to be a good method of assessing the biomass of brown trout in the Neste d'Aure valley. Correlations between habitat variables and brown trout biomass and density were investigated. Total biomass was significantly related to elevation (between 1350 and 600 m), cover, maximum summer temperature (between 10 and 16-degrees-C), conductivity, mean bottom velocity, mean depth and width/depth ratio. Total density was significantly related to the same variables, with the exception of mean depth, and in addition to water gradient and stream width. The study by age-class showed that the abundance of young-of-the-year trout is related to elevation, temperature and conductivity. Only width was the habitat feature related to abundance of young-of-the-year. No correlations were found with other physical habitat features. Abundance of one-year-old trout (1+) was related with the same variables of elevation, temperature and conductivity in addition ; their density and biomass were also related to cover. Density of catchable trouts (length greater than 180 mm) was correlated with cover, depth, temperature, elevation and conductivity. The stepwise regression analyses produced combinations of variables that explained 86% of the variations in biomass (with 5 variables). This type of work can be very useful in the management of fishing and of trout populations.