Accounting for non-native Brown Trout in biological assessments: Implications for selecting reference conditions

Project Author
Issue Date
Kirk, Mark A.
Wissinger, Scott A.
Thumbnail Image
This article is available for viewing to the public after September 18, 2020.
First Reader
Additional Readers
reference conditions , non-native species , taxonomic completeness , biological assessment , biological integrity
The efficacy of assessments that evaluate biological integrity can be improved by accounting for the ecological processes that influence assemblage composition. Many studies have emphasized that bioassessments need to account for natural environmental gradients, but there is little consensus on how bioassessments should account for the impacts of non-native species. In particular, non-native trout species have been introduced into many high-quality streams that probably meet the expected reference conditions for bioassessment in a given region. The goal of this study was to test whether the presence of large, piscivorous, non-native Brown Trout (Salmo trutta) at reference sites altered interpretations of taxonomic completeness indices (TCIs) based on fish assemblages. We used fish data from 215 sites in wadeable streams in northwestern Pennsylvania to compare the performance of 3 TCIs that used different modeling approaches to account for Brown Trout impacts. One model accounted for the presence of non-native Brown Trout as a covariate (covariate model), another model censored reference sites with non-native Brown Trout from the reference pool (censored model), and a final model used all reference sites without accounting for the presence of Brown Trout (unaccounted model). TCIs based on observed-to-expected ratios were able to distinguish reference conditions from altered conditions in the covariate and censored models. In contrast, the unaccounted index could not distinguish reference from altered conditions and was, thus, unable to accurately assess biological integrity, probably because large Brown Trout reduce native species richness. Our results provide a framework for how bioassessment practitioners can use different approaches to account for non-native species impacts, especially when considering which criteria are most important for defining reference conditions. Accounting for the effects of non-native species with these approaches should improve the ability of bioassessments designed to summarize the interactive effects of all potential human stressors on stream assemblages.
Environmental Science / Studies
This work is licensed under a Attribution 4.0 International (CC BY 4.0) License.
Mark A. Kirk and Scott A. Wissinger, "Accounting for non-native Brown Trout in biological assessments: Implications for selecting reference conditions," Freshwater Science 38, no. 4 (December 2019): 790-801.
Published article
University of Chicago Press