Consideration of grain and extent in landscape studies of terrestrial vertebrate ecology

The choice of scale (both grain and extent) of a research project can directly affect the results of the project itself, and the comparability of those results to similar research. This paper reviews 149 self-described landscape studies published between 1987 and 2001. From each paper, data for several variables were gathered that we proposed may influence the choice of scale. These variables included the taxa the research considered, the characteristics of the landscape of interest, the species response variables measured, the type of methodology employed by the study, the habitat type(s) surveyed, and the technique(s) used to create habitat maps. Our results demonstrated that only 61% of the papers identified both the grain and extent at which the research was conducted. Taxonomic group, landscape characteristics, and study type all appeared to influence both the choice of grain and extent for the study. Wide variation (standard deviation) in grain and extent among studies suggests that researchers either used some factor not considered here to select scale, or made decisions of the scale on pragmatic grounds, rather than on the appropriateness of the scale to the variables. Finally, 24 of the 149 studies reviewed researched the same combination of categories of the six variables that we considered, yet only for one combination was the grain and extent used within the same order of magnitude. This review highlights the need for more structured choice of scale in landscape studies, and in particular the need for more standardized size of grain and extent for greater comparability among studies.