Untitled

Abstract

Background. The North American prairie biome is considered one of the most endangered ecoregions. Prairie fishes have been affected by many anthropogenic disturbances, including the construction of bridges and culverts as road crossings over streams. The objective of our study was to test fish assemblage characteristics upstream and downstream of single-and double-barreled culverts and compare them with assemblages at bridges within a single prairie stream watershed. Methods. Eight sites located near public or private roads were selected on the South Loup River, Nebraska, USA. Fish were sampled monthly from April through October 2013 using backpack electrofishing. Sampling occurred upstream and downstream of each road crossing structure. Fish collections from all months were combined to calculate species richness; species diversity; the relative abundance of two species of conservation concern in North American Great Plains streams (brassy minnow Hybognatus hankinsoni and plains topminnow Fundulus sciadicus); the relative abundance of two introduced predators (largemouth bass Micropterus salmoides and northern pike Esox lucius), and the relative abundance of both introduced predators combined. A two-way analysis of variance (ANOVA) test was used to determine whether these seven fish assemblage characteristics were different between upstream and downstream reaches, among road crossing structures, and between the interaction of these two factors followed by pairwise comparisons by either road crossing type, direction, or the interaction of both using a Tukey's honest significant difference (HSD) test. Results. Only two fish assemblage characteristics appeared to be related to road crossing type or direction: species richness and relative abundance of brassy minnows. Species richness was significantly higher at bridges compared to single-barreled culverts. The relative abundance of brassy minnows was significantly different between bridges and both types of culverts but was not significantly different between the two culvert types. Discussion. Several reasons could explain the overall results of our study. First, road crossings on prairie streams may not have much effect on fish passage or fish habitat due to the low gradient of these streams. Secondly, because of bank stabilization, bridges could affect prairie fish communities and their habitats as severely as culverts within this watershed. Third, the fish communities upstream and downstream of all crossing types may represent fishes that are mostly tolerant of disturbances in general. Our study indicates that prairie streams and their fish communities could be at least somewhat tolerant and resilient to disturbances associated with road crossings, even though the small-bodied nature of many native species potentially pose higher risks of disconnection within the population.