Binomial sampling plan for classifying population density of Thrips palmi (Thysanoptera: Thripidae) in potato

Abstract

Binomial sequential classification sampling plans were developed for use in classifying populations of Thrips palmi Karny, below or above a mean intervention threshold density (mIT) for management decision making on the fall potato on Cheju Island, Korea. The proportion of potato leaves with at least T (tally threshold) thrips (PT) was related to mean thrips density (m) with an empirical model 1n(-1n(1-PT)) = y+δ1n(m). The PT-m relationship fit the data well for T values of 1-10, with T=3 having the best fit. Wald's sequential probability ratio test was used to formulate sequential sampling stop lines relative to mIT values of 5 and 10 thrips per potato leaf with a series of T. The sampling plans were evaluated using the operating characteristic and average sample number functions. Simulation analysis indicated that a binomial model with T = 3 was best, and less than 60 samples, on average, were needed to classify populations relative to either mIT value. The binomial sequential sampling plan was tested with sequential resampling simulation using 8 independent data sets for the validation. The binomial sequential classification sampling plans presented here should enhance the efficiency of pest management programs based on the prescriptive suppression of T. palmi on fall potato.