Streamflow, trophic interactions, and stream fish assemblages

Background/Question/Methods

The response of aquatic species to streamflow varies due to strength of trophic interactions. For example, fish populations can exhibit positive associations with discharge due to increased input of food resources, yet deplete food resources at various degrees under different flow conditions.  The goals of this study are to assess: 1) how trophic components (epibenthic algae, macroinvertebrates, and fishes) are correlated with flow at the drainage-basin scale?; and, 2) whether fish assemblages, in terms of population density and species richness, can be explained by flow and density of macroinvertebrates?  We hypothesize that in addition to assemblage-flow correlations, there will be a positive (resource driven processes), or alternately negative (top-down processes) associations between related trophic levels.

We studied ten streams, representing a range of flow characteristics, within the Meramec River drainage basin in eastern Missouri.  During the summer of 2012, data were collected for epibenthic algal biomass, assemblages of benthic macroinvertebrates and fishes, as well as flow and other habitat variables.  We conducted multiple regressions to assess the relationships among flow and the three trophic components using discharge or current velocity and population density (or biomass) of related trophic levels as predictors in various models.

Results/Conclusions

There was a curvelinear relationship between population density of benthic fishes and discharge, with highest population densities associated with the mid-range of discharge (R² = 0.75, p = 0.003).  The most abundant benthic fish species [Rainbow Darter (Etheostoma caeruleum)] was positively correlated with current velocity (R² = 0.70, p = 0.003).  Incorporating the density of macroinvertebrates as an additional predictor variable did not improve these two models.  Density of macroinvertebrates and fishes were not correlated with current velocity or discharge, while variation in density of both macroinvertebrates and fishes tends to show a curvelinear relationship with discharge and a negative association with each other (R² = 0.42, p = 0.112, ΔAIC = -4.06).  Moreover, epibenthic algal biomass is positively correlated with current velocity (R² = 0.72, p = 0.002), while accounting for the density of benthic macroinvertebrates improved the prediction (R² = 0.84, p< 0.001, ΔAIC = -10.44).

These results do not suggest the importance of resource-driven processes due to the negative association among related trophic components, but alternatively reflect the effects of top-down processes.  In particular, population densities of lower trophic levels are controlled by members from adjacent higher trophic levels in the context of suitable flow regimes.