T. Hesthagen et al., Effects of water chemistry and habitat on the density of young brown troutSalmo trutta in acidic streams, WATER A S P, 112(1-2), 1999, pp. 85-106
We examined the relationship between young brown trout ( Salmo trutta) dens
ity in lake tributaries, and water chemistry and habitat variables. The stu
dy was carried out during the autumn in three acidic, softwater river syste
ms in western and southwestern Norway; Gaular and Vikedal (1987-1993) and B
jerkreim (1988-1993). The streams had mean calcium concentrations of 0.35 m
g L-1 (Gaular), 0.52 mg L-1 (Vikedal) and 0.84 mg L-1 (Bjerkreim). The conc
entration of inorganic Al was generally low, with mean values of 8.40 (Gaul
ar), 22.22 (Vikedal) and 43.36 mu g L-1 (Bjerkreim). In multiple regression
s that involved different water chemistry variables, brown trout density co
rrelated best with calcium concentration and with a combination of calcium
and pH; the Ca2+:H+ ratio. In Vikedal and Gaular, calcium explained 51 and
57%, respectively, of the variability in brown trout densities. Althoug alk
alinity exhibited the best correlation with brown trout density in Bjerkrei
m ( r(2)=0.33), it was similar to that of the model that included all major
ions plus pH. The Ca2+:H+ ratio had a larger effect for variability in bro
wn trout density in Gaular (r(2)=0.66) than calcium alone. In Vikedal and B
jerkreim, the Ca2+:H+ ratio also correlated with brown trout density, but c
onsiderably less than in Gaular. The predictive power of habitat variables
was much lower than that of water chemistry; the single most important fact
ors were altitude in Gaular (r(2)=0.22), mean water temperature in Vikedal
(r(2)=0.11) and depth SD (index of heterogeneity) in Bjerkreim (r(2)=0.07).
Models that included both habitat and water chemistry variables showed tha
t the density of young brown trout was predicted primarily by calcium conce
ntrations in Gaular (r(2)=0.75) and Vikedal (r(2)=0.54), as opposed to pH i
n Bjerkreim (r(2)=0.25). Habitat had low effect in all three river systems
(r(2)=0.01-0.04). The final model explained 86, 68 and 32%, respectively, o
f the variability in brown trout density in the three catchments. Thus, wat
er chemistry variables seem to be factors that limit the density of young b
rown trout in acidic softwater streams.