Importance of Understanding Landscape Biases in USGS Gage Locations: Implications and Solutions for Managers

ABSTRACT

Flow and water temperature are fundamental properties of stream ecosystems upon which many freshwater resource management decisions are based. U.S. Geological Survey (USGS) gages are the most important source of streamflow and water temperature data available nationwide, but the degree to which gages represent landscape attributes of the larger population of streams has not been thoroughly evaluated. We identified substantial biases for seven landscape attributes in one or more regions across the conterminous United States. Streams with small watersheds (<10 km²) and at high elevations were often underrepresented, and biases were greater for water temperature gages and in arid regions. Biases can fundamentally alter management decisions and at a minimum this potential for error must be acknowledged accurately and transparently. We highlight three strategies that seek to reduce bias or limit errors arising from bias and illustrate how one strategy, supplementing USGS data, can greatly reduce bias.

RESUMEN

la temperatura y flujo de agua son propiedades fundamentales de los ecosistemas fluviales, sobre los cuales se toman diversas decisiones de manejo en cuanto a recursos dulceacuícolos. Los sensores del Estudio Geológico de los Estados Unidos de Norte América (EGEU) son la fuente disponible más importante de datos de flujo de agua y temperatura a nivel nacional, pero el grado al cual los sensores son representativos de los atributos paisajísticos de una población más grande de ríos, no ha sido analizado a profundidad. Se identificaron sesgos sustanciales en siete atributos paisajísticos en una o más regiones a lo largo de las zonas limítrofes de los Estados Unidos de Norte América. Los ríos de cauce pequeño (<10 km²) y aquellos localizados en regiones elevadas no estuvieron adecuadamente representados, y los mayores sesgos se observaron en los sensores que miden la temperatura del agua y en las regiones áridas. Los sesgos tienen el potencial de alterar de manera fundamental las decisiones de manejo, y como mínimo este error tiene que reconocerse de forma precisa y transparente. Se plantean tres estrategias que buscan tanto reducir el sesgo o limitar los errores que surgen de dicho sesgo, como ilustrar cómo una estrategia, suplementando los datos EGEU, puede reducir el sesgo de manera importante.

ACKNOWLEDGMENTS

We thank James McKenna; Robert Carline; all members of the Fish Habitat, Climate, and Landuse Change project team; two anonymous reviewers; and the Science Editor for helpful comments on earlier versions of this work. Supplemental water temperature data were generously provided by Andrew Dollof, Anthony Raburn of the Georgia Departement of Natural Resources, Joan Trial of the Maine Department of Marine Resources, John Sweka of the U.S. Fish and Wildlife Service Northeast Fishery Center, Mark Hudy, the Maryland Department of Natural Resources Monitoring and Non-tidal Assessment Division, Neal Hagstrom of the Connecticut Department of Energy and Environmental Protection, the New Hamphsire Fish and Game Commission, Rich Kern of the Vermont Fish and Wildlife Department, Roy Martin, Steve Means of the Pennsylvania Department of Environmental Protection, the Susquehana River Basin Commission, Tamara Smith, and the Wood-Pawcatuck Watershed Association.

FUNDING

Funding for this research was provided by the U.S. Geological Survey National Climate Change and Wildlife Science Center. Use of trade names does not imply endorsement by the United States Government.