Premium
A Predictive Index of Biotic Integrity Model for Aquatic‐Vertebrate Assemblages of Western U.S. Streams
Author(s) -
Pont Didier,
Hughes Robert M.,
Whittier Thomas R.,
Schmutz Stefan
Publication year - 2009
Publication title -
transactions of the american fisheries society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.696
H-Index - 86
eISSN - 1548-8659
pISSN - 0002-8487
DOI - 10.1577/t07-277.1
Subject(s) - index of biological integrity , ibis , species richness , environmental science , benthic zone , streams , ecology , habitat , geography , biology , computer science , computer network
Because of natural environmental and faunal differences and scientific perspectives, numerous indices of biological integrity (IBIs) have been developed at local, state, and regional scales in the USA. These multiple IBIs, plus different criteria for judging impairment, hinder rigorous national and multistate assessments. Many IBI metrics are calibrated for water body size, but none are calibrated explicitly for other equally important natural variables such as air temperature, channel gradient, or geology. We developed a predictive aquatic‐vertebrate IBI model using a total of 871 stream sites (including 162 least‐disturbed and 163 most‐disturbed sites) sampled as part of the U.S. Environmental Protection Agency's Environmental Monitoring and Assessment Program survey of 12 conterminous western U.S. states. The selected IBI metrics (calculated from both fish and aquatic amphibians) were vertebrate species richness, benthic native species richness, assemblage tolerance index, proportion of invertivore‐piscivore species, and proportion of lithophilic‐reproducing species. Mean model IBI scores differed significantly between least‐disturbed and most‐disturbed sites as well as among ecoregions. Based on a model IBI impairment criterion of 0.44 (risks of type I and II errors balanced), an estimated 34.7% of stream kilometers in the western USA were deemed impaired, compared with 18% for a set of traditional IBIs. Also, the model IBI usually displayed less variability than the traditional IBIs, presumably because it was better calibrated for natural variability.