Untangling Multiple Factors in Spatial Distributions: Lilies, Gophers, and Rocks

Abstract

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. Ecological Society of America is collaborating with JSTOR to digitize, preserve and extend access to Ecology. This paper raises some general points about the interpretation of ecological data 40andfocuses on the ways in which standard statistical correlations fail to serve ecological ends. Although it may not be obvious why "ecological data" differ formance of organisms will in fact frequently pose two problems for correlation Emphasizing analysis. First, the habitats and landscapes from which data are gleaned tend to new ideas ~ be spatially patchy, so that nearby data points cannot be considered as independent new Iu~~a~ samples. New techniques are available to handle this problem. The second problem to is more basic and more recalcitrant. The theory of limiting factors leads us to expect that ecological factors may often impose upper ceilings on response vanistimulate research ables, without strongly influencing the distribution of the response variable below the ceiling. Bivariate scattergrams convey ecological information not only when in ecology they show linear or curvilinear relationships, but also when they fall into diffuse clouds, as long as the clouds have informative edges. Although ecologists care about such clouds, correlation analyses do not, and many ecologically informative data sets may be written off as "insignificant" because the wrong question is being asked. Asking the right question-is this cloud informative?-requires a new approach to data, new statistics, and new terminology. Thomson et al. bring some new statistical approaches to the attention of ecologists. Another goal of their paper is simply to encourage the objective examination of data for ecological meaning before, after, and beyond the application of "standard" analyses. AbstractDespite broad consensus on the power of experiments, correlational studies are still important in ecology, and may become more so as spatial studies proliferate. Conventional correlation analysis, however, (1) fundamentally conflicts with the basic ecological concept of limiting factors, and (2) ignores spatial structure in data, which can produce spuriously high correlations. Especially for field data, bivariate scattergrams often show "factor-ceiling" distributions wherein data points are widely scattered beneath an upper limit, due to the action of other factors. Although most ecological information in such a graph resides in the upper limit, standard correlation/regression does not characterize such limits. If other factors have been measured, path analysis may be useful, but otherwise, direct description of ecological ceilings is desirable. Objective methods for doing so are barely known to ecologists; we review rec...

Integrating heterogeneity across spatial scales: interactions between Atrina zelandica and benthic macrofauna

“…In the regressions, at least some of the weaknesses of the relationships were due to factor ceiling responses, i.e. correlations between variables that appear random except for a lack of points above or below a certain line (Thomsen et al 1996). Factor ceilings occur in our data, at least at the aggregate group level.…”

Spatial and temporal variability across life's hierarchies in the terrestrial Antarctic

“…Such an approach has rarely been applied to modelling the abundance and distribution of Antarctic species (most investigations focus on the assemblage level). Where this has been done, spatial structure has typically not been considered despite wide acceptance of its importance in ecology (Legendre 1993;Thomson et al 1996;Liebhold & Gurevitch 2002). Notable exceptions are the studies by Usher & Booth (1984, 1986, which appear to have been limited only by the availability of appropriate analytical techniques.…”

Spatial surface-pattern analyses and boundary detection techniques applied in forest ecology

“…En general, la forma del correlograma indica el patrón espacial descrito (Sokal y Oden, 1978a, 1978bLegendre y Fortin, 1989;Thomson et al, 1996). Existen así formas típicas como un correlograma cuyos valores disminuyen al aumentar la distancia en el caso de la I de Moran, y que corresponde a un gradiente o correlogramas con subidas y bajadas marcadas, que corresponden a patrones cíclicos.…”