Abstract.-We examined distribution and abundance patterns of 69 commonly occurring fishes at several hundred sites in Lower Michigan streams. We used cluster analysis to group fishes that commonly occurred together at stream sites. These seventeen clusters explained about 39% of the variation in species abundances among the stream sites, providing a reasonable, albeit simplified picture of general associations of fishes in Lower Michigan streams. Known ecological differences among species and further analyses suggested that a single measure of cluster abundance should not be used to predict abundances of its constituent species.
We used measures of stream size and hydrology as landscape-scale, habitat axes (a "macro-template") for comparing streams. We identified catchment area (CA) and low-flow yield (90% exceedence flow divided by catchment area) as key driving variables that linked features of the landscape to multiple, site-scale characteristics of stream habitat (e.g. temperature, velocity, and depth) important to fishes. As a measure of groundwater loading to streams, low-flow yield (LFY) integrates geology, landform, and soil characteristics of catchments, reaching its highest values in basins with highly permeable soils and relatively steep topography. In Lower Michigan streams, high LFY values were generally associated with: greater portions of coarse-textured glacial deposits in catchments; higher stream gradients; coarser stream substrates; and cooler and less variable predicted July weekly temperatures. High CA values were generally associated with lower stream gradients, and warmer and less variable predicted July weekly temperatures.
Ordinations of fish clusters and species' abundances on LFY-CA axes provided insight into the structure of fish assemblages in Lower Michigan streams. The seventeen fish clusters spread out in a meaningful pattern when plotted on LFY-CA axes, reflecting stream size and temperature preferences of constituent species. Plots of abundances of individual species on LFY-CA axes showed differences among fishes in LFY and CA conditions where species occurred and were most abundant. These patterns supported the notion that stream fishes respond in an individualistic manner to stream conditions, and that species-specific models are needed to describe fish assemblage structure in streams. We used relations between LFY, CA, and fish abundances to describe longitudinal changes in stream conditions and fish assemblages both within streams, and among hydrologically different streams. These relations have also been used to characterize potential fish assemblages of stream valley segments. Relations between LFY, CA, and fish abundances that we described are specific to Lower Michigan streams, because relationships between LFY, CA, and stream temperature vary regionally. However, our approach could be used to develop similar models specific to other regions.