PARTITIONING THE VARIATION AMONG SPATIAL, TEMPORAL AND ENVIRONMENTAL COMPONENTS IN A MULTIVARIATE DATA SET

We propose a method of partitioning the variation in a multivariate se t of data according to (i) environmental variables, (ii) variables des cribing the spatial structure in the data and (iii) temporal variables . This method is an extension of an existing method for partialling ou t the spatial component of environmental variation, using canonical an alysis. Our proposed method extends this approach by including tempora l variables in the analysis. Thus, the partitioning of variation for a data matrix of species' abundances or biomass can include, by our met hodology, the following components: (1) pure environmental, (2) pure s patial, (3) pure temporal, (4) pure spatial component of environmental , (5) pure temporal component of environmental, (6) pure combined spat ial/temporal component, (7) combined spatial/temporal component of env ironmental and (8) unexplained. In addition, permutation testing accom panying the analyses allows tests of significance for the relationship between the different components and the species data. We illustrate the method with a set of survey data of penaeid species (prawns) obtai ned on the far northern Great Barrier Reef, Australia. This extension is a useful tool for multivariate analysis of ecological data from sur veys, where space, time and environment commonly overlap and are impor tant influences on observed variation.