作者: R. M. Warwick , K. R. Clarke
DOI: 10.1017/S0025315400037528
关键词:
摘要: Statistical methods for analysing changes in community structure fall under the three general headings of univariate, graphical/distributional and multivariate. These are applied to a variety benthic data (macrobenthos, meiobenthos, corals, demersal fish), from localities (intertidal/subtidal, temperate/tropical) over both spatial temporal scales. Four conclusions emerge this comparative study: (1) The similarity between sites or times based on their univariate properties is usually different clustering multivariate analyses. (2) Species dependent (multivariate) much more sensitive than species independent (univariate graphical/distributional) discriminating times. (3) In examples where one component fauna has been studied, may give results components, whereas tend same results. (4) By matching ordinations subsets environmental an ordination faunistic data, key variables responsible change be identified.