Comparative Analysis of Selected Sampling Methods for Adult Bemisia tabaci (Homoptera: Aleyrodidae) in Cotton

Abstract

We evaluated the reliability and efficiency of various sticky traps and 2 direct-count sampling methods for estimating the abundance of Bemisia tabaci (Gennadius) adults in cotton, Gossypium hirsutum L., during a 3-yr period. Comparing sticky traps of various sizes, orientations, and placements, horizontally oriented traps (7.6 by 7.6 cm) with the upper surface exposed consistently captured the most adults per square centimeter over the season when placed at ground level at field edges followed by those placed within fields at ground level. Before insecticide applications, captures on all trap types and counts using a leaf-turn method and an oiled black pan method were highly correlated with the abundance of immatures in the field the same week, and even better correlated with immatures counted 1 wk later. After insecticide application, correlations were lower and more variable. Regression equations relating adult abundance to immature populations varied significantly between sites and over years for all adult sampling methods. Thus, robust predictive relationships could not be formulated. The 2 direct-count methods were highly correlated with one another, but the leaf-turn method was much less variable between individual samplers than the black pan method when estimating populations in the same field. Accounting for sample sizes needed for an acceptable level of precision and the per unit cost (time) of each sampling method, the black pan method was 3.5 times more costly than the leaf-turn method, and sticky traps were from 3.6 to 19.7 times more costly in estimating populations in the same fields. On average it took ≈6 min to estimate adult populations With a precision (SE/mean) of 0.25 using the leaf-turn method. Based on between-sampler variability, and cost considerations, the leaf-turn method was the most reliable and efficient technique for estimating adult abundance of those examined.