Constrained Quadratic Ordination
This chapter describes an extension of the RR-VGLM class, called quadratic reduced-rank vector generalized linear models (QRR-VGLMs), which allow for constrained quadratic ordination (CQO). QRR-VGLMs have a lot of potential applications in ecology where species–site data are collected. Topics include the basic ideas of ordination, the species packing model, latent variables as gradients, CQO estimation and interpretation, biplots, and calibration. Unconstrained QO (indirect gradient analysis) is also described. Some real-data examples are given.
KeywordsEnvironmental Variable Brown Trout Ordination Axis Niche Width Reference Species
- Gauch, Hugh G., J., G. B. Chase, and R. H. Whittaker 1974. Ordinations of vegetation samples by Gaussian species distributions. Ecology 55(6):1382–1390.Google Scholar
- Jongman, R. H. G., C. J. F. ter Braak, and O. F. R. van Tongeren (Eds.) 1995. Data Analysis in Community and Landscape Ecology. Cambridge: Cambridge University Press.Google Scholar
- Kooijman, S. A. L. M. 1977. Species abundance with optimum relations to environmental factors. Annals of Systems Research 6:123–138.Google Scholar
- ter Braak, C. J. F. 1995. Calibration. See Jongman et al. (1995), pp. 78–90.Google Scholar