On determining the statistical significance of discontinuities within ordered ecological data

Boundaries between adjacent ecosystem units may be important in determining ecosystem structure and function across heterogeneous landscapes. Such boundaries are potentially important sites for early detection of global climate change effects. Yet traditional data analysis methods focus primarily on homogeneous units rather than on the boundaries between them; thus, new methods are being developed for detecting, characterizing and classifying boundaries, eg split moving-window boundary analysis (SMW). SMW is a simple yet sensitive method for locating discontinuities that may exist within multivariate, serial data (ordered in one dimension) at various scales relative to the length of the data series. However, SMW is subjective and relative, and therefore locates apparent discontinuities even within random, serial data. The authors present two nonparametric methods for determining the statistical significance of discontinuities detected by SMW, a Monte Carlo method for determining the statistical significance of scale-dependent discontinuities (significant relative to only one scale), and a nonparametric, scale-independent method (also dependent upon scale size, but to a much lesser degree) that is more appropriate for locating statistically significant discontinuities that separate different, relative homogeneous groups of varying size along a series. They illustrate their application to locating boundaries between vegetation samples collected at systematic intervals across a desert landscape in southern New Mexico. -from Authors

Duke Scholars

Author James F. Reynolds Environmental Sciences and Policy