Murdoch University Research Repository

Welcome to the Murdoch University Research Repository

The Murdoch University Research Repository is an open access digital collection of research
created by Murdoch University staff, researchers and postgraduate students.

Learn more

A user-friendly quantitative approach to classifying nearshore marine habitats along a heterogeneous coast

Valesini, F.J., Clarke, K.R., Eliot, I. and Potter, I.C. (2003) A user-friendly quantitative approach to classifying nearshore marine habitats along a heterogeneous coast. Estuarine, Coastal and Shelf Science, 57 (1-2). pp. 163-177.

Link to Published Version:
*Subscription may be required


A scheme, which can be readily used by fisheries and environmental managers and ecologists, has been developed for quantitatively classifying the different habitats found in nearshore marine waters along the heterogeneous lower west coast of Australia. Initially, 25 beach sites, representing a wide range of nearshore environments, were separated into six a priori habitat types on the basis of characteristics that could readily be observed and were likely to influence the extent to which a particular (fish) species occupies a particular habitat. Focus was thus placed on such features as the degree of exposure to wave activity and whether or not seagrass and/or reefs were present in the nearshore vicinity. Subsequently, quantitative data for 27 environmental variables, considered likely to characterise the six habitat types, were obtained for each of the 25 sites from readily accessible sources. When the latter data were subjected to multidimensional scaling (MDS) ordination, the points for the sites representing only three of those six habitat types formed discrete groups. The bvstep routine in the Primer v5.0 statistical package (Clarke & Gorley, Primer v5.0: User Manual/Tutorial, Primer-E Ltd, Plymouth, 2001) was thus used to select a subset of the 27 environmental variables that would provide a better resolution of the six a priori habitat types. This process involved matching the distance matrix constructed from the quantitative environmental data with a matrix constructed from scored data that reflected the criteria for the initial a priori classification scheme. A subset of seven environmental variables gave the best correlation between the two matrices (ρ=0.823), and thus provided the optimal set of quantitative data for discriminating between the six a priori habitat types. These variables comprised both the direct and north-westerly fetches, the minimum distance from the shoreline to the 2 m depth contour, the distance from the shoreline to the first offshore reef chain along a south-westerly transect, and the relative contributions of bare sand, subtidal reef and seagrass. Data for these characteristics at any nearshore site along the coastline can readily be recorded by managers and ecologists and subjected to the ‘nearest-replicate’ classification procedure developed in this study to ascertain the habitat type to which that site should be assigned. Current work is using MDS ordination, in conjunction with associated statistical tests and the bvstep routine, to elucidate the extent to which the compositions of assemblages of fish, benthic macroinvertebrates, meiofauna and zooplankton in nearshore waters along the lower west coast of Australia are related to habitat type(s).

Item Type: Journal Article
Murdoch Affiliation(s): Centre for Fish and Fisheries Research
School of Biological Sciences and Biotechnology
Publisher: Academic Press
Item Control Page Item Control Page