ICES Journal of Marine Science · 2016 · 96 citations · 38 references
Life SciencesEcologyMarine SciencesMarine BiodiversityMarine EcologyMarine Invertebrate BiologyBenthic BiodiversityBenthic EcologyMacroecologyMacroinvertebrate EcologyBenthic Assemblage StructureBenthic Community StructureBenthic Invertebrate EcologyBenthic Invertebrate ZoologyBenthic Species DiversityMacroalgal EcologyMacrofaunal CommunitiesMacrofaunal Ecology
Abstract This study provides an inventory of the recent benthic macrofaunal communities in the entire Baltic Sea. The analyses of soft-bottom benthic invertebrate community data based on over 7000 locations in the Baltic Sea suggested the existence of 10 major communities based on species abundances and 17 communities based on species biomasses, respectively. The low-saline northern Baltic, characterized by silty sediments, is dominated by Monoporeia affinis, Marenzelleria spp., and Macoma balthica. Hydrobiidae, Pygospio elegans, and Cerastoderma glaucum dominate the community in sandy habitats off the Estonian west coast and in the southeastern and southern Baltic Sea. Deep parts of the Gulf of Finland and central Baltic Sea often experience hypoxia, and when oxygen levels in these regions recover, Bylgides sarsi was the first species to colonize. The southwestern Baltic Sea, with high salinity, has higher macrofaunal diversity compared with the northern parts. To spatially interpolate the distribution of the major communities, we used the Random Forest method. Substrate data, bathymetric maps, and modelled hydrographical fields were used as predictors. Model predictions were in good agreement with observations, quantified by Cohen's κ of 0.90 for the abundance and 0.89 in the wet weight-based model. Misclassifications were mainly associated with uncommon classes in regions with high spatial variability. Our analysis provides a detailed baseline map of the distribution of benthic communities in the Baltic Sea to be used both in science and management.
38
Leo Breiman · Machine Learning · 2001 · 126.2K citations · Full text
Classification and Regression by randomForest
Andy Liaw, Matthew C. Wiener · 2007 · 18.4K citations
A new method for non‐parametric multivariate analysis of variance
Marti J. Anderson · Austral Ecology · 2001 · 13.3K citations