Skip to main content

Application of fuzzy cluster analysis to Lake Simcoe crustacean zooplankton community structure

Buy Article:

$50.00 plus tax (Refund Policy)

Abstract:

Fuzzy clustering generates cluster membership weights that indicate how tightly each object is linked to its cluster relative to other clusters of a dendrogram. In a fuzzy clustering of the crustacean-zooplankton taxa of Lake Simcoe, a large (720 km2) hardwater lake in Ontario, Canada, we show how the membership weights can be used to rank all taxa for their contribution to the sampling unit (SU) classification, where the total number of SUs was 84 (7 years × 12 sampling sites). The validity of the results was confirmed by comparison with other more traditional methods of identifying variables important for object classifications and by permutation tests of matrix correlation before and after removal of low-ranked and highly ranked species. Fuzzy clustering of Lake Simcoe SUs also revealed (i) the likelihood of trends in zooplankton community composition over the 7-year period and (ii) differences in composition possibly related to sampling-station depth. In particular, the shallowest sampling station in southern Cook's Bay had a zooplankton community structure that differed significantly from other stations during all years of the study. As a preliminary screening or data exploration tool, fuzzy clustering is particularly useful for analysis of ecological data.

La classification floue assigne à chaque objet d'un dendrogramme un poids qui reflète le degré d'appartenance de l'objet à son groupe par rapport aux autres groupes. Lors d'une classification floue des taxons du zooplancton du lac Simcoe, un grand (720 km2) lac d'eau dure d'Ontario, Canada, il a été possible d'ordonner tous les taxons selon leur contribution à la classification de l'unité d'échantillonnage (SU), où le nombre d'unités était de 84 (7 années × 12 sites d'échantillonnage). Ces résultats ont été validés par comparaison avec des méthodes plus traditionnelles d'identification des variables importantes dans la classification d'objets et des tests de permutation des matrices de corrélation, avant et après le retrait des espèces de rangs inférieurs et supérieurs. La classification floue des SU du lac Simcoe a aussi révélé (i) la vraisemblance des tendances dans la composition des communautés zooplanctoniques au cours des 7 années et (ii) la possibilité que les différences de composition soient reliées à la profondeur à la station d'échantillonnage. En particulier, la station d'échantillonnage la moins profonde dans la partie sud de la Baie de Cook avait une communauté zooplanctonique dont la structure différait de celle de toutes les autres stations au cours de toutes les années. La classification floue s'avère être un outil particulièrement utile pour le tri préliminaire et l'exploration des données écologiques.[Traduit par la Rédaction]

Document Type: Research Article

Publication date: 2001-02-01

More about this publication?
  • Published continuously since 1901 (under various titles), this monthly journal is the primary publishing vehicle for the multidisciplinary field of aquatic sciences. It publishes perspectives (syntheses, critiques, and re-evaluations), discussions (comments and replies), articles, and rapid communications, relating to current research on cells, organisms, populations, ecosystems, or processes that affect aquatic systems. The journal seeks to amplify, modify, question, or redirect accumulated knowledge in the field of fisheries and aquatic science. Occasional supplements are dedicated to single topics or to proceedings of international symposia.
  • Information for Authors
  • Submit a Paper
  • Subscribe to this Title
  • Terms & Conditions
  • Sample Issue
  • Reprints & Permissions
  • Ingenta Connect is not responsible for the content or availability of external websites
  • Access Key
  • Free content
  • Partial Free content
  • New content
  • Open access content
  • Partial Open access content
  • Subscribed content
  • Partial Subscribed content
  • Free trial content
Cookie Policy
X
Cookie Policy
Ingenta Connect website makes use of cookies so as to keep track of data that you have filled in. I am Happy with this Find out more