PREDICTING LARVAL ABUNDANCE OF THE BERTHA ARMYWORM, MAMESTRA CONFIGURATA WLK., IN MANITOBA FROM CATCHES OF MALE MOTHS IN SEX ATTRACTANT TRAPS
AbstractTwo cone-orifice-type traps baited with (Z)-11-hexadecen-1-ol acetate (1.0 mg) and (Z)-9-tetradecen-1-ol acetate (0.05 mg) were placed in each of 5–13 canola (Brassica spp.) fields per year for 7 years at locations distributed throughout the area in Manitoba subject to outbreaks of the bertha armyworm, Mamestra configurata Wlk. (Lepidoptera: Noctuidae). The number of male moths per trap varied from 0 to 821 in the 90 fields trapped. Larval density, in the trap fields and in 185 nearby canola fields, varied from 0 to 105 per square metre. There was a direct relationship between the number of moths per trap and the density of late-instar larvae (L4–L6) for the trap fields and for nearby (within 5 km) fields but variability was loo high to allow accurate prediction of larval densities for individual fields. However, the number of moths per trap could be used to predict the proportion of fields in the vicinity of the trap field that would have larval densities above the economic threshold. No fields with above-threshold larval densities occurred with captures of <20 moths per trap. The proportion of fields with larval densities above the economic threshold increased from 0.10 with 20–30 moths per trap, lo 0.19 with 30–60 moths per trap, 0.64 with 70–100 moths per trap, and 0.91 with over 100 moths per trap. Contamination of catches by non-target species of noctuid moths had insignificant effects on the accuracy of these predictions because even if they were misidentified as M. configurata the number of moths per trap changed only marginally. Populations of M. configurara were higher and local outbreaks occurred more frequently in canola-growing areas adjacent to morainic hills along the Manitoba Escarpment, whereas populations were low in areas of the Manitoba Lowlands and of the Western Uplands. A proposed monitoring system for M. configurata in Manitoba is described.