Are long-term fish assemblage changes in a large US river related to the Asian Carp invasion? Test of the hostile take-over and opportunistic dispersal hypotheses.

Gizzard shad is a dominant planktivore/detritivore in the Wabash River, and populations crashed in the early 1990s. Previous work (1974–2008) identified a substantial shift in body-size structure and functional trait composition in the Wabash River fish assemblage between 1989 and 1996. Invasive Asian Carp appeared in the Ohio River basin including the Wabash River in the 1990s. Our goal was to test for temporal changes in assemblage composition and trophic structure relative to the invasion of Asian carp from the early 1990s. We hypothesized that establishment of Asian Carp was a contributor to the assemblage composition shift, and that their presence altered the trophic pathways and food sources of native fishes including Gizzard Shad (hostile takeover hypothesis). Alternatively, Asian Carp may have found success through capitalizing on an empty niche, likely left vacant by the decline in Gizzard Shad, or abundance changes in other trophic groups (opportunistic hypothesis). We utilized archival fish and mussel collections to test for trophic changes in the ecosystem using δ ¹³C and δ ¹⁵N stable isotope analyses. We examined stomach contents of the dominant planktivore/detritivore consumer, Gizzard Shad, from archival and recent collections to test for changes in the phytoplankton community. Stable isotopes indicated a community reliance on other, more deplete, carbon sources than indicated by the algae, and a slight increase between the δ ¹³C time periods. Although all functional feeding groups of fishes indicated some reduction in δ ¹⁵N, the differences were only significant for omnivores, mussels, and planktivore/detritivores. Although Asian Carp may have contributed to the collapse of Gizzard Shad populations, other stressors were likely more important drivers of their decline. This is the first, albeit indirect, evidence of opportunistic “invasion” as opposed to the historically presumed hostile takeover model.