Geomorphology, habitat, and spatial location influences on fish and macroinvertebrate communities in modified channels of an agriculturally-dominated watershed in Ohio, USA

Abstract

We evaluated relationships between in-stream habitat, spatial distribution, and geomorphic features at 28 study sites within a predominantly agricultural watershed in Ohio. Objectives were to: (1) measure and compare the physical structure and biotic communities of highly modified drainage channels to those of minimally impacted channels, (2) identify significant environmental factors influencing fish and invertebrate assemblages in modified channels, and (3) relate them to biotic communities in a multivariate statistical model and then to compare fish species and macroinvertebrate taxa models to common multi-metric bioassessment index models. We used canonical correspondence analysis (CCA) and variance partitioning to relate environmental variables to fish and macroinvertebrate community attributes. Geomorphically, minimally impacted sites were statistically different than bench and trapezoidal sites. Bench sites were statistically different than trapezoidal sites in floodplain width and depth ratios indicating that bench formation provided some amount of attached floodplain for the inset channel and functioned more similar to how natural streams would function. Key ecological drivers for macroinvertebrate communities were stream size, gradient and connectivity to a floodplain. Key ecological drivers for fish communities were quality of in-stream habitat variables; however, stream size and connectivity to a floodplain also were important. Larger, perennial sites tended to support more aquatic biota and more diverse assemblages either as primary habitat or as conduits between higher quality upstream or downstream locations. The latter can be critical to sustaining biota in highly modified agricultural watersheds. In smaller systems, in-stream habitat was a limiting factor but, more importantly, they experienced smaller discharges and can become intermittent during dry months. We hypothesized that leaving vegetated benches in the agricultural ditch would improve local ecology, but our data do not support this hypothesis as measured benches were either intermittent or too small to be important fluvial features. In the agriculturally-dominated modified headwater systems studied, we hypothesize that the proximity to a nearby patch of high quality habitat might be the main driving factor.