Long-term fish assemblages of the Ohio River: Altered trophic and life history strategies with hydrologic alterations and land use modifications

Abstract

Long-term monitoring of species assemblages provides a unique opportunity to test hypotheses regarding environmentally induced directional trajectories of freshwater species assemblages. We used 57 years of lockchamber fish rotenone and boat electrofishing survey data (1957-2014) collected by the Ohio River Valley Water Sanitation Commission (ORSANCO) to test for directional trajectories in taxonomy, trophic classifications, and life history strategies of freshwater fish assemblages in the Ohio River Basin. We found significant changes in taxonomic and trophic composition of freshwater fishes in the Ohio River Basin. Annual species richness varied from 31 to 90 species and generally increased with year. Temporal trajectories were present for taxonomic and trophic assemblages. Assemblage structure based on taxonomy was correlated with land use change (decrease in agriculture and increase in forest). Taxonomic assemblage structure was also correlated with altered hydrology variables of increased minimum discharge, decreased fall rate, and increased rise rate. Trophic composition of fish catch correlated with land use change (decrease in agriculture and increase in forest) and altered hydrology. Altered hydrology of increased minimum discharge, increased fall discharge, decreased base flows, and increased number of high pulse events was correlated with increased counts of herbivore-detritivores and decreased counts of piscivores and planktivores. We did not find directional changes in life history composition. We hypothesized a shift occurred from benthic to phytoplankton production throughout the basin that may have decreased secondary production of benthic invertebrates. This may also be responsible for lower trophic position of invertivore and piscivore fishes observed in other studies.

Fig 1. The Ohio River Basin showing locations of navigational dams (bars) and two USGS gaging stations (open circles).

Locations for fish sampling are described by Thomas et al. [11]. Coordinates for the Louisville USGS gaging station are Lat 38°16’49” Long 85°47’57”.

Fig 2

Nonmetric multidimensional scaling ordination of rotenone (left panels) and electrofishing collections (right panels): taxonomic counts (top), trophic counts (middle), and life history strategist counts (bottom). Final stress values for rotenone collection analyses were 0.09, 0.09, and 0.07 from top to bottom. Final stress values for electrofishing collection analyses were 0.06, 0.05, and 0.01 from top to bottom. Lines connecting years represent temporal trajectories. Highest loading variables are listed on axes.

Fig 3. Log-transformed trophic counts by year, for electrofishing collections.

Fig 4

Redundancy analysis using taxonomic counts from electrofishing collections, with significant environmental vectors on bottom (A, B). Axes explained 22.4% and 6.4% of variation. PC axes represent altered hydrology variables (loadings are in Table 2). Species or trophic traits with highest loadings on axes are listed. Redundancy analysis using trophic counts from electrofishing collections, with significant environmental vectors on bottom (C, D). Axes explained 25.6% and 13.2% of variation. PC axes represent altered hydrology variables. Species or trophic traits with highest loadings are listed on axes.