Ordination as a Tool for Analyzing Complex Data Sets

Abstract

Ordination, or indirect gradient analysis, has been widely used in plant ecology as a tool for examining relationships between environment and vegetation. Since the classic work of Bray & Curtis (1957) which popularized such applications, the mathematical models underlying ordination methodology have become increasingly sophisticated. Principal Components Analysis provided increased conceptual rigor and objectivity. The realization that the assumptions of linearity implicit in PCA were not consistent with representation of nonlinear species responses along environmental gradients (Beals 1973, Jeglum, Wehrhahn & Swan 1971, Whittaker & Gauch 1973) led to introduction of a series of nonlinear methods including Reciprocal Averaging, RA, (Hill 1973, 1974), Gaussian Ordination, GO, (Gauch, Chase & Whittaker 1974, Ihm & Groenewoud 1975) and Nonmetric Multidimensional Scaling, NMDS, (Fasham 1977, Prentice 1977). Subsequent comparative studies (e. g., Fasham 1977, Gauch, Whittaker & Singer 1980) have suggested these nonlinear methods to consistently and accurately recover the structure of simulated data.

The author gratefully acknowledges the continuing collaboration of Dr. Norman L. Christensen of Duke University. This research was supported by National Science Foundation grants DEB-7708743 and DEB-7804043 to R.K.P. and DEB-7707532 and DEB-7804041 to N.L.C.