ordinationpackage

Ordination package

(a bit dated now, here just for the record)

The ordination package used for the analysis of multiple characteristics of ecological systems over time is a simple ordination of the data (Principal components analysis modified to handle missing data) with a colour coded graphical output. It has been found useful for analysing ecosystem change on the Eastern Scotian Shelf (see figures to the right).

The code itself is a series of simple R functions which can be found here https://github.com/jae0/aegis/R/pca*:

NOTE: the following has been subsumed by pca.analyse.data.r, pca.basic.r in https://github.com/jae0/aegis/

The following is an example of how to install and use the package.

data( essdata ) # load a sample dataset

rownames( essdata ) = essdata$yr # the rownames must be set

essdata$yr = NULL # remove year as we do not want to use it as an ordination variable

essdata.scaled = scale( essdata, center=TRUE, scale=TRUE )

pcastats = sorted.ordination.plot( essdata.scaled )

# this function also creates PDF figures and data outputs

in the working directory which is: getwd()

str( pcastats ) # look at contents of the variable, etc

Data is returned to pcastats after which you can continue to do more refined analyses.

NOTE: This source code simple enough and is provided as is with no strings attached. But you will gain good karma if you share your modifications with the rest of the world. Please send me a note describing how you use it or if you make any improvements to it.