Effects of box culverts on stream habitat, channel morphology, and fish and macroinvertebrate communities at selected sites in South Carolina, 2016–18

Abstract

Much attention has been placed on the role that under-roadway culverts may have in inhibiting upstream fish movement because of altered hydrology and unsuitable conditions for accessing or swimming through the culvert. Other culvert effects related to habitat alterations or disturbance to macroinvertebrate communities have received relatively little attention. Entities responsible for culverts or other stream crossing structures are required to follow the U.S. Army Corps of Engineers guidelines for compensatory mitigation should any disturbance result from an engineering activity. One factor considered in the scoring of mitigation requirements is culvert length. Except for shading a longer length of stream, it is unknown whether longer culverts result in greater disturbance to stream habitat or the biotic communities than shorter culverts. The U.S. Geological Survey, in cooperation with the South Carolina Department of Transportation, evaluated the role of culverts in altering physical habitat and community structure of fish and macroinvertebrates at 20 sites in South Carolina. Culvert sites were categorized by length (either greater than 30.5 meters or less than or equal to 30.5 meters) and physiographic province (Piedmont or upper Coastal Plain). This study design allowed for a regional assessment to determine if culverts may have different effects on habitat and biotic communities in different physical settings. The results indicated considerable variation in physical habitat characteristics within and among the culvert sites from all categories. A consistent finding was that channel cross-sectional area tended to increase in reaches downstream from culverts in the upper Coastal Plain. The primary dimension of change was vertical, that is, incision of the streambed. This change, however, did not seem to coincide with a deleterious effect on the fish community. Increased habitat complexity and greater taxonomic richness were observed at most sites with downstream incision. Macroinvertebrate communities were highly variable and did not tend to cluster along any of the culvert categories, which may reflect the variability of microhabitats within each site. In contrast, fish communities were largely segregated by physiographic province but did not show any other significant clustering on the basis of upstream or downstream reach or culvert length. Given the small within-group sample size, extrapolation of results should be done carefully, acknowledging the physiographic and group characteristics.