Refining Insect Pest Monitoring: How Crop and Pest Identity Influence the Choice of Sampling Unit and Technique.

This study aimed to determine the most suitable sampling unit (apical, middle, or basal leaf strata) and sampling technique (direct counting or tray beating) for sap-sucking pest groups (aphids, thrips, lace bugs, and whiteflies) across different crops in an agroforestry system (AFS). The study was conducted with kale, okra, tomato, and scarlet eggplant over three cropping seasons (one in 2023 and two in 2024). The main pest species were Lipaphis erysimi on kale and Aphis gossypii on okra (aphids); Corythaica passiflorae on scarlet eggplant and Gargaphia lunulata on okra (lace bugs); Bemisia tabaci on kale and tomato (whiteflies); and Frankliniella schultzei on tomato and Thrips spp. on okra (thrips). A generalized linear mixed model (GLMM) with negative binomial distribution was used to evaluate the effects of sampling technique, sampling unit, and crop on pest density. Optimal sampling combinations were defined based on higher mean densities and lower coefficients of variation (CV). Direct counting was the most effective technique for aphids, whiteflies, and lace bugs, whereas tray beating was more efficient for thrips. However, the optimal sampling unit varied strongly according to crop-pest interactions. These results indicate that, although sampling techniques can be partially standardized by pest group, the selection of the sampling unit must consider crop-specific interactions. This approach improves sampling efficiency and supports decision-making in Integrated Pest Management programs in diversified agroecosystems such as AFS.