Yield loss to arthropods in vegetationally diverse agroecosystems

Five methods were evaluated for measuring the effects of arthropod pests on crop yields in vegetationally diverse agroecosystems, and the conditions for vegetational diversity to improve yields were examined. Two methods measured absolute yield benefits in which polyculture yields were higher than yields in monocultures. A review of the literature indicated that absolute yield benefits occurred only rarely and only when the arthropod pests caused severe yield losses in monocultures. One method was judged of limited use. The remaining two methods measured marginal yield benefits, in which yield loss in polycultures is less than that in monocultures. Statistical tests for factorial experiments that estimate this marginal benefit are proposed. A review of 41 cases suggested that this benefit occurred only if polycultures had lower pest populations than monocultures, and even then, it occurred intermittently. Severe competition from the other plants in the polyculture might limit the ability of the crop to compensate for pest injury and crop tolerance, or resistance to pest injury might otherwise limit yield losses in polycultures. In addition, the data suggested that pest injury is less likely to exceed economic injury levels in polycultures than in monocultures.