Simulation of community patterns

Programs are described for the simulation of coenoplanes-mathematically defined, two-dimensional community patterns characterized by scattered and broadly overlapping Gaussian reponse surfaces for species populations. Field data have been studied to derive a set of assumptions regarding realistic properties of coenoplanes and ways in which some of these properties should be variable. Additional assumptions define properties of samples and sets of samples taken from coenoplanes. The programs thus provide a basis for tests of different ordination techniques and effects of characteristics of sample sets on these.