Native Plants Journal - Article

Systematic experimental designs for mixed species plantings

Abstract:

Systematic designs save considerable space and require many fewer seedlings than conventional mixture designs. Systematic designs are based on large plots where species composition varies gradually. Conventional mixture designs may be impractical due to the large areas and tremendous numbers of seedlings they require. The systematic designs also provide splendid demonstration areas for scientists and land managers to observe the effects of a gradient of species composition. One basic design incorporates a large triangular plot; in concept this plot is identical to the well-known soil textural triangle. The intent of the designs is to produce a response surface over species composition, rather than test for significant differences between two specific species compositions. Another design superimposes a species composition gradient on a Nelder's design, which systematically varies planting density. It is possible to study mixtures in multiple strata, such as overstory trees and herbaceous understory. The systematic mixture designs are most effective when considering 2 to 4 species; when more species are considered, a screening study based on a diallel-like design is indicated. The designs may be applied to any stationary organism.

Issue & Pages:

Fall 2001 Pages: 90-96

Article Download:

2-2NPJ90-96.pdf (PDF document)

Authors:

• JCG Goelz

Keywords:

Competition experiment, diversity, interference, mixture experiments, multiple species, Nelder's design, response surface design