Adult female density was measured at two paired locations within each patch, at the edge and at 2 m from the edge. In 2000, each sample consisted of 8 unidirectional sweeps with a sweep net (parallel to the edge and approximately 0.75 m in length). In 2001, planthopper densities were measured with a D-vac insect vacuum (Rincon-Vitova, Ventura, California, USA) with a 0.08-m2 sampling head. For each sample, the head was placed at 10 points for 3 s each, and approximately 0.5 m apart along the edge or interior. Planthopper numbers per sample (in both 2000 and 2001) were converted to numbers per cordgrass stem by estimating mean stem densities (per m2) in each replicate patch (separate estimates for the edge and interior). This was accomplished by measuring the average number of stems within a 25 × 25 cm sampling frame placed at three random locations at the edge and interior of the patch.
We offer several lines of evidence to suggest that samples taken at 2 m inward from the patch edge are representative of the patch interior in general, and that potential edge effects likely do not extend this far into the patch interior. First, a pilot study indicated that planthopper densities between 1 m and 3 m into the patch interior were lower than the densities at the patch edge (K. J. Haynes, unpublished data). Second, significant differences in plant quality were found to also exist between the patch edge and only 2 m into the interior (see Results). Finally, planthoppers are relatively immobile when moving within a patch (Cronin 2003), suggesting that the intermixing of individuals between the edge and 2 m into the interior is likely to be low.
Cronin, J. T. 2003. Movement and spatial population structure of a prairie planthopper. Ecology 84:1179–1188.