Risk Assessment for Tomato Fruitworm in Processing Tomato Crop-Egg Location and Sequential Sampling

SIMPLE SUMMARY: The tomato fruitworm, Helicoverpa armigera, is a key pest of several crops. It can cause particularly extensive damage in crops of processing tomatoes. Risk assessment can be a tedious and costly task if sampling protocols require a large number of plants. Sequential sampling allows sampling of a reduced number of plants when population densities are much lower or much higher than the economic or control threshold. Additionally, for crop protection purposes, sampling for classification (to assess if population density is lower or higher than the economic threshold) is adequate and requires much less effort. We studied the preferred location of eggs found on plants and then described the spatial pattern of oviposition in processing tomatoes using Taylor’s power law. Eggs were found more frequently in the exposed canopy in the upper and middle-upper strata, directly below open flower clusters, with an aggregated spatial pattern. A sequential plan was developed for 20 and 80 plants, as minimum and maximum sample sizes, respectively. This reduces sampling efforts and costs when compared to the fixed number sampling plan, and provides acceptable precision in decision-making for this pest in a processing tomato crop. ABSTRACT: Helicoverpa armigera is one of the key pests affecting processing tomatoes and many other crops. A three-year study was conducted to describe the oviposition preferences of this species on determinate tomato plants (mainly the stratum, leaf, leaflet, and leaf side) and the spatial pattern of the eggs in the field, to form a sequential sampling plan. Eggs were found mainly in the exposed canopy, on leaves a (upper stratum) and b (upper-middle stratum) and significantly fewer eggs on leaf c (middle-lower stratum) below flower clusters. This vertical pattern in the plant was found in all phenological growth stages. The spatial pattern was found to be aggregated, with a trend towards a random pattern at lower densities. A sequential sampling plan was developed, based on Iwao’s method with the parameters of Taylor’s power law, with minimum and maximum sample size of 20 and 80 sample units (plants), respectively (two leaves/plant). For its validation, operating characteristic (OC) and average sample number (ASN) curves were calculated by means of simulation with independent data sets. The β-error was higher than desirable in the vicinity of the economic threshold, but this sampling plan is regarded as an improvement both in effort and precision, compared with the fixed sample plan, and further improvements are discussed.