Laserfiche WebLink
<br />identify positive associations. <br />At a smaller scale, our methods were too gross to detect <br /> <br />microhabitat partitioning, such as vertical position in the water <br /> <br />column. The typically high turbidity of the San Juan River <br /> <br />precluded methods that would enable characterization of <br /> <br />microhabitat partitioning. Thus, our results only represent <br /> <br />habitat use and associations on the pool/riffle level (sensu <br /> <br />Frissel et ale 1986). <br /> <br />Considering the spatial and temporal scale of our sampling, <br /> <br />a number of general patterns emerge. The different age-classes <br /> <br />of native fish tended to differ in degree of overlap with <br /> <br />nonnative species. Adult/sub-adult native fish had the least <br /> <br />overlap in habitat use with nonnativesi a consequence of native <br /> <br />species using habitats with deeper and/or greater current <br /> <br />velocities. For example, adult R. osculus, a small-bodied fish <br /> <br />(<90mm), occupied shallow habitats with high current velocities. <br /> <br />Currently, no nonnative species that is primarily a riffle <br /> <br />occupant has become established in the San Juan River. The two <br /> <br />native catostomids are large (>300mm SL) as adults and typically <br /> <br />occupy deeper habitats with moderate to rapid current velocity. <br /> <br />This provides a partial explanation for the absence of sub-adults <br /> <br />in habitats <O.3m which predominate in secondary channels during <br /> <br />low flow periods. Once these large-bodied native species move to <br /> <br />deep habitats (mainly available in the primary channel), there is <br /> <br />a suite of similar-sized nonnative species (e.g., ~. punctatus <br /> <br />and C. carpio) to potentially contend with. We did not <br /> <br />16 <br />