Microhabitat use by fish in the upper Rio Matarraña, Spain, 1984–1987

Abstract

Abstract– We evaluated the microhabitat use and abundance of Barbus graellsii, Bardus haasi, Chondrostoma toxostoma, Noemacheilus barbatulus, Oncorhynchus mykiss and Rutilus arcasii between 1984 and 1987 in the upper Rio Matarraña, Spain. The mean abundance classes for all species ranged from 1.8 to 3.0 (where 1=1–6, 2=6–10, 3=11–20 and 4=>20 specimens, respectively). Mean abundance was consistently higher in 1984–1985 (2.4–3.0) than during 1986–1987 (1.8–2.8). The decreases noted in 1986–1987 were attributable to declines in abundance of O. mykiss (introduced in winter 1984) and B. haasi. Barbus graellsii and Ch. toxostoma, however, remained abundant throughout the entire 4‐year study. We only observed N. barbatulus and R. arcasii irregularly in the study site. Analyses of microhabitat availability data indicated that the study site contained more silt and less algae/debris during spring 1985 and early and late summer 1986 than in the majority of the remaining season. The converse was true for late summer 1985 (i. e., less silt and more algae/debris than the majority of seasons). Principal component analyses showed that B. graellsii and Ch. toxostoma generally occupied deeper microhabitats with low to average velocities, higher amounts of depositional substrata and lower quantities of erosional substrata. B. haasi tended to avoid microhabitats with rubble substrata and occurred in those with higher amounts of algae/debris. O. mykiss occupied shallower areas with slightly higher velocities and a heterogeneous substratum. With the exception of B. haasi, microhabitat use by assemblage members was similar from 1984 to 1987. B. haasi, however, was not as strongly affected by depth in the latter two years of the study as it was during 1984–1985. Seasonal and annual analyses of intraspecific microhabitat use showed that most changes were due to variations in microhabitat availability. Nonetheless, all species exhibited minor seasonal shifts in microhabitat use. Size‐related analyses indicated that both smaller B. graellsii and Ch. toxostoma occupied shallower microhabitats with slower velocities than larger specimens. Intersite differences in microhabitat use for B. graellsii and Ch. toxostoma showed that most differences in substratum use were attributable to disparities in substratum availability. Both species occurred closer to the substratum in the site with higher velocities (i. e., upper reach), although this response was more pronounced in Ch. toxostoma. Interspecific analyses indicated that B. graellsii and Ch. toxostoma did not consistently occupy statistically differentiable microhabitats, although both species occurred farther from the substratum and refuges than did B. haasi. O. mykiss occupied shallower microhabitats with more gravel than either B. graellsii or Ch. toxostoma. The lack of microhabitat shifts by native species during the study period indicates that interactions with either B. haasi or O. mykiss did not play a strong role in microhabitat use by the remaining assemblage members.