Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
8845450 | Ecological Indicators | 2018 | 13 Pages |
Abstract
We used canonical correspondence analyses, including partial correspondence analysis and principal response curves to analyze change in composition and metrics and the relative contribution of environmental variables for each biological data set. We carried out analyses between sites and along a temporal gradient, complemented by linkage trees and logistic regressions. Conclusions are more complex than the formulated hypotheses. Although both invertebrate and fish communities revealed a similar ability to assess impairment, invertebrates responded more strongly to local disturbances, whereas the fish composition or traits/guilds were more sensitive to larger scale variation. Results for temporal variation analyses suggested that invertebrate traits more accurately detected change in environmental factors but variation in the taxonomic composition of fish species also followed a temporal gradient. Differences on community response reinforce the importance of a combined approach, using types of community data, to detect environmental impacts.
Keywords
Related Topics
Life Sciences
Agricultural and Biological Sciences
Ecology, Evolution, Behavior and Systematics
Authors
Edna Cabecinha, Samantha Hughes, Rui Cortes,