Preharvest sampling plan for larvae of the wheat stem sawfly, Cephus cinctus (Hymenoptera: Cephidae), in winter wheat fields

Abstract

AbstractThe wheat stem sawfly, Cephus cinctus Norton, is the most important insect pest of dryland wheat, reducing both harvest efficiency and head weight. When wheat lodged (flattened) by C. cinctus is straight-cut, harvest efficiency is affected in three ways: (1) lodged stems are difficult to collect with the combine, so mature wheat is lost; (2) the combine has to operate more slowly to collect as many of the lodged stems as possible; and (3) there is increased risk of damage to the combine due to lowering of the combine header to recover lodged stems. An alternative harvest method is to swath the wheat before lodging occurs and place it in windrows prior to combining. When swathing is used, harvest efficiency is not affected by the C. cinctus infestation level; however, this type of harvesting is more expensive and labor-intensive than straight- cutting. Thus, swathing should be used only when the crop is heavily infested by C. cinctus, while straight-cutting should be used when the level of C. cinctus infestation is lower. Adapting harvesting practices to the C. cinctus infestation level requires a sampling plan that can reliably and cost-effectively predict the infestation level in a given field. In this study, we (i) evaluate the relationship between sampling effort along field edges and precision of the estimated average infestation level in 6 wheat fields, and (ii) generate a Gaussian regression analysis of the infestation level up to 200 m into 18 wheat fields using two variables, distance from the edge and average infestation level at the edge, as explanatory variables. We conclude that little additional precision for estimates of infestation level is gained by collecting more than 10 samples along field edges. We also conclude that the average C. cinctus infestation level along field edges can be used successfully to predict infestation levels up to 200 m into winter wheat fields. The proposed sampling should be implemented about one month before harvest to provide growers with enough time to make logistical preparations for harvest.