Land Use Associations with Distributions of Declining Native Fishes in the Upper Colorado River Basin

Abstract

The upper Colorado River basin contains one of the most imperiled fish faunas in North America. Anthropogenic land use and nonnative species impacts are considered among the top reasons for imperilment. We determined the association of anthropogenic land use intensity (road density, percentage of converted land, and oil and gas well density), relative abundance of nonnative white suckers Catostomus commersonii and their hybrids, and natural landscape features with the occurrences of three native fishes in the Colorado River basin, Wyoming, that have declined throughout their ranges: flannelmouth sucker C. latipinnis, bluehead sucker C. discobolus, and roundtail chub Gila robusta. We found that flannelmouth suckers occurred more frequently in large, low-gradient stream reaches, but their occurrence was not associated with land use intensity or the relative abundance of white suckers and hybrid suckers. In contrast, bluehead sucker occurrence decreased with increasing intensity of each land use type within a 0.5-km stream buffer, which suggests that localized land use disturbances have negatively affected the species’ distribution. Roundtail chub occurred more frequently in low-gradient stream segments with more riparian vegetation, but associations with land uses were more complex. Roundtail chub were negatively associated with road density in a 0.5-km buffer, but in contrast to our expectations they were positively associated with road density in the contributing watershed and with local oil and gas well density. We think that rocky substrates and pools—habitats preferred by roundtail chub and sometimes associated with oil and gas infrastructure—have concentrated the remaining individuals within their currently reduced distribution. The associations we identified highlight the potential role of land use in distributional declines and can help managers to determine whether proposed land use changes may affect existing populations. Spatial predictions of occurrence have already been used in regional fish conservation planning efforts in Wyoming.

ACKNOWLEDGMENTS

This project was supported in part under U.S. Bureau of Land Management Cooperative Agreement PAA-08-0008 and National Fish and Wildlife Foundation Agreement 2009-0076-000. Fish distribution surveys were made possible by funding from the U.S. Bureau of Reclamation, U.S. Fish and Wildlife Service (State Wildlife Grant Program), and Wyoming Game and Fish Department. We thank Curtis Gill and Aaron Kern for leading fish survey crews, and we are grateful to Robb Keith and Hilda Sexauer for technical and administrative support of the crews. We thank J. Fore, C. Hoagstrom, P. Martinez, and an anonymous reviewer for providing helpful comments on the manuscript.