Submitted to: Environmental Entomology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: April 5, 2005
Publication Date: October 3, 2005
Citation: Prasifka, J.R., Hellmich II, R.L., Dively, G.P., Lewis, L.C. 2005. Assessing the effects of pest management on non-target arthropods: the influence of plot size and isolation. Environmental Entomology. 34:1181-1192. Interpretive Summary: Using insecticides to control crop pests sometimes has harmful unintended effects on insects and other organisms that are not pests, called "non-target" groups. Some of these non-target groups include species that benefit crop production or are valued by humans for other reasons. The new technology of using genetically-modified (transgenic) plants that produce insecticidal substances in their tissues is also considered a possible threat to some non-target species, particularly beneficial insects. Tests of whether an insecticide or genetically-modified plant harms non-target insects are usually conducted in small plantings (plots) of a crop, which may be physically isolated from other plants. This research questions if the use of plots that are much smaller than typical fields, or are physically isolated, influences whether a test will detect a negative effect on non-target species. This information is important to designing accurate, science-based evaluations of the safety of insecticides and genetically modified plants that produce insecticidal substances. Results showed that in very small plots (less than 20 feet wide), the number of some non-target insects is influenced by the surrounding area. This makes such small plantings of genetically-modified varieties less likely to detect differences (positive or negative) compared to conventional management with insecticides. This effect of size is likely less critical for tests with insecticides. Results also showed that isolated plots can either attract or repel different types of insects. Based on the results, it is suggested that researchers generally should avoid using very small plots (less than 20 feet wide) because the surrounding environment strongly influences results. Also, when planning research it is important to consider whether one of the non-target groups being studied is likely to be affected by isolated plots. This information is useful for all stakeholders interested in potential non-target effects of transgenic crops.
Technical Abstract: Evaluations of field research on the non-target effects of pest management, particularly the production of transgenic crops with insecticidal properties, suggest the methods used are sometimes unlikely to detect real differences among treatments. Among potential problems, abundance estimates may be scale-dependent for many arthropods, which move among experimental plots and between experimental plots and the surrounding environment. Insecticide-disturbed plots of field corn in a range of sizes in 2003 (0.03 - 0.53 ha) and 2004 (0.01 - 0.13 ha) were used to test for effects of scale on non-target arthropod abundance. Possible effects of artificially isolating plots by removal of vegetation around plot borders were also investigated in 2003. Community and taxon-based analyses showed abundance of foliar (above-ground) arthropods depended on plot size and isolation. While abundance of foliar arthropods was generally greater in smaller plots, isolation treatments suggested some taxa may have been either repelled or attracted to isolated plots. Levels of some epigeal (ground-dwelling) taxa were also size or isolation-dependent, but community-based analysis did not indicate a strong collective response to treatments. Recommendation of a practical but rigorous minimum plot size for non-target studies may not be appropriate because responses to plot size varied among taxa. However, because arthropod movement into and out of plots can reduce differences between treatments, results suggest the use of small plots (width < 9 m) for non-target studies on transgenic crops generally should be avoided. Similarly, the taxon-specific effects of isolating plots should be considered when planning studies or interpreting results.