Macroinvertebrate and Fish Community Metrics: Confounding Effects and Consistency over Time
- PDF / 1,122,968 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 75 Downloads / 174 Views
GENERAL WETLAND SCIENCE
Macroinvertebrate and Fish Community Metrics: Confounding Effects and Consistency over Time Katya E. Kovalenko 1
&
Valerie J. Brady 1 & Jan J.H. Ciborowski 2 & George E. Host 1 & Lucinda B. Johnson 1
Received: 12 March 2019 / Accepted: 14 October 2019 # Society of Wetland Scientists 2020
Abstract Macroinvertebrate and fish metrics are an essential tool in freshwater biomonitoring. Yet, many well-known stream metrics do not perform well as indicators of anthropogenic stress in lakes and wetlands, and there is a need to better understand the role of potential confounding factors such as latitude or environmental variables acting at regional scales. We attempt to untangle the relationships among metrics and watershed stressors while accounting for potential confounding factors using hierarchical partitioning, and test metric consistency across time, focusing on the Laurentian Great Lakes wetland monitoring programs conducted 10 years apart. Our results show that many frequently used metrics have high temporal variability and are significantly affected by spatial factors, most notably richness-type metrics used in numerous regional monitoring studies. Only a few metrics (invasive fish species richness and relative abundance, mayfly family richness and the relative abundance of Ephemeroptera, Trichoptera, Sphaeriidae and Odonata) had slightly better and more consistent correlations with watershed land use. We suggest that explicit consideration of confounding factors is essential in the context of large-scale monitoring programs, and focus on the less general (such as habitat-specific) metrics may be a more promising approach. Keywords Freshwater coastal wetlands . Lentic biodiversity . Watershed land-use . Hierarchical partitioning . Bioassessment . Large-scale monitoring
Introduction Electronic supplementary material The online version of this article (https://doi.org/10.1007/s13157-019-01239-7) contains supplementary material, which is available to authorized users. * Katya E. Kovalenko [email protected] Valerie J. Brady [email protected] Jan J.H. Ciborowski [email protected] George E. Host [email protected] Lucinda B. Johnson [email protected] 1
Natural Resources Research Institute, University of Minnesota Duluth,5013 Miller Trunk Highway, Duluth, MN, USA
2
Department of Biological Sciences, University of Windsor, 401 Sunset Ave, Windsor, ON N9B 3P4, Canada
Coastal wetlands of the world’s largest lakes harbor a significant proportion of freshwater biodiversity, but are also epicenters of human activity (Vadeboncoeur et al. 2011), placing this biodiversity at risk from the intense anthropogenic stress. Metrics that describe attributes of macroinvertebrate and fish assemblages affected by this stress are embedded in many approaches used in freshwater biomonitoring (e.g., Karr 1981; Rosenberg et al. 2008). A large body of literature focuses on identifying appropriate metrics that can serve as biotic indicators of changes occurring in freshwater ecosystems (e.g., Stoddard et al. 2008;
Data Loading...
