A protocol for classifying ecologically relevant marine zones, a statistical approach

• Published in: Estuarine, coastal and shelf science Vol. 83; no. 2; pp. 175 - 185
• Main Authors: Verfaillie, Els, Degraer, Steven, Schelfaut, Kristien, Willems, Wouter, Van Lancker, Vera
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 20.06.2009; Elsevier
• Subjects: abiotic variables; Animal and plant ecology; Animal, plant and microbial ecology; Belgian part of the North Sea; Biological and medical sciences; Brackish; Brackish water ecosystems; cluster analysis; Fundamental and applied biological sciences. Psychology; macrobenthic species; Marine; marine landscape approach; principal components analysis; Sea water ecosystems; Synecology; zonation; North Sea; ANE, North Sea; Belgium; marine landscape approach; abiotic variables; macrobenthic species; zonation; Belgian part of the North Sea; cluster analysis; principal components analysis; Cluster analysis; Marine environment; Brackish water environment; Benthic organism; Landscape; Principal component analysis
• ISSN: 0272-7714; 1096-0015
• DOI: 10.1016/j.ecss.2009.03.003

Mapping ecologically relevant zones in the marine environment has become increasingly important. Biological data are however often scarce and alternatives are being sought in optimal classifications of abiotic variables. The concept of ‘marine landscapes’ is based on a hierarchical classification of geological, hydrographic and other physical data. This approach is however subject to many assumptions and subjective decisions. An objective protocol for zonation is being proposed here where abiotic variables are subjected to a statistical approach, using principal components analysis (PCA) and a cluster analysis. The optimal number of clusters (or zones) is being defined using the Calinski–Harabasz criterion. The methodology has been applied on datasets of the Belgian part of the North Sea (BPNS), a shallow sandy shelf environment with a sandbank–swale topography. The BPNS was classified into 8 zones that represent well the natural variability of the seafloor. The internal cluster consistency was validated with a split-run procedure, with more than 99% correspondence between the validation and the original dataset. The ecological relevance of 6 out of the 8 zones was demonstrated, using indicator species analysis. The proposed protocol, as exemplified for the BPNS, can easily be applied to other areas and provides a strong knowledge basis for environmental protection and management of the marine environment. A SWOT-analysis, showing the strengths, weaknesses, opportunities and threats of the protocol was performed.