Premium
Effects of longitudinal variations in stream habitat structure on fish abundance: an analysis based on subunit‐scale habitat classification
Author(s) -
Inoue Mikio,
Nunokawa Masanori
Publication year - 2002
Publication title -
freshwater biology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.297
H-Index - 156
eISSN - 1365-2427
pISSN - 0046-5070
DOI - 10.1046/j.1365-2427.2002.00898.x
Subject(s) - habitat , abundance (ecology) , oncorhynchus , tributary , ecology , leuciscus , upstream and downstream (dna) , community structure , biology , environmental science , fishery , fish <actinopterygii> , geography , upstream (networking) , computer network , cartography , computer science
SUMMARY 1. Stream reaches contain assortments of various habitat types that can be defined at different spatial scales, such as channel unit (e.g. pools, riffles) and subunit (patches within channel units). We described longitudinal (upstream–downstream) patterns of stream habitat structure by considering subunits as structural elements, and examined their effects on the abundance of masu salmon ( Oncorhynchus masou ) and rosyface dace ( Leuciscus ezoe ) in a third‐order tributary of the Teshio River in northern Hokkaido, Japan. 2. Nine subunit types were determined on the basis of water depth, current velocity and substrate, using 0.5 × 0.5 m grids. Although both masu salmon and rosyface dace used pools as a major habitat, the former preferred a subunit type occurring at pool heads (PH subunit) while the latter preferred a slow‐current edge type (SE‐2 subunit). 3. Along the course of the stream, slow‐edge subunits (SE‐1, 2 and 3) increased in frequency downstream while fast‐edge subunits (FE‐1 and 2) decreased, suggesting a downstream development of slow‐current edges. Regression analyses indicated that longitudinal variation in masu salmon abundance was explained by the area of PH, rather than pools. Masu salmon density increased with the area of PH. Rosyface dace abundance was explained by a combination of water depth and the area of SE‐2, both effects being positive. 4. Longitudinal variations in the abundance of both species were related to the abundance of their preferred habitat at the subunit scale, rather than channel‐unit scale. The results emphasise the importance of fine‐scale patchiness when examining stream fish habitats.