ABSTRACT

Sweep sampling is the quickest, least expensive, and most commonly used method of sampling rangeland grasshopper (Acrididae) communities to determine species composition. However, few studies have examined the accuracy of sweep sampling relative to actual community composition because the latter is difficult to determine. We made such comparisons by sampling grasshoppers within enclosures that prevented their escape during sampling. We first took a single sweep sample within the enclosure, then exhaustively sampled the enclosure to remove all remaining grasshoppers. Comparison of the grasshoppers in the initial sample to those remaining in the enclosure was used to test the hypothesis that an accurate estimate of community composition was provided by the initial sweep sample. We sampled during a 5-wk period to determine if accuracy changed seasonally and swept in both mowed and unmowed enclosures to relate the results to the physical structure of the vegetation. The initial sweep samples, which are equivalent to the single samples taken in many studies, provided good estimates of the relative abundance of most species. Mean differences between estimated and actual proportions of particular species were generally less than 0.03. Even though the deviations were usually relatively small, the relative abundances of some species were sometimes underestimated or overestimated in the samples. In particular, Phoetaliotes nebrascensis (Thomas) was sometimes overestimated to such an extent that it caused other species to be underestimated. Our results suggest that correction factors could be developed to adjust sweep sample estimates to account for consistent biases in sampling, but this correction would have to be done on a community-by-community and habitat-by-habitat basis.

community ecology, rangeland, sampling accuracy

¹Accepted for publication 15 June 1999.
²To whom correspondence should be addressed.
³Bee Biology and Systematics Laboratory, USDA-ARS, Utah State University, Logan, Utah 84322.