Binomial Sequential Sampling Plans for Macadamia Felted Coccid (Hemiptera: Eriococcidae) Infesting Hawaii Macadamia Orchards

Abstract

The macadamia felted coccid, Eriococcus ironsidei (Williams) is an invasive pest of macadamia orchards that causes severe damage to the trees and reduces nut yield in Hawaii. Their feeding results in distortion and stunting of new growth, and with high population densities, dieback of entire branches may result. Ten macadamia orchards were sampled on the Hawaii island over a period of 18 mo to characterize the dispersion of the pest in the field, and to use those data as the basis for the development of a binomial sampling plan. Analyses based on Taylor's Power Law and Iwao's mean crowding index indicated an aggregated pattern of dispersion of the pest in the field. Curvilinear relationships between mean crawler density and the proportion of infested trees based on four tally thresholds (crawlers per unit area, T > 0, 5, 10, 20) were plotted. Results showed that a tally threshold T > 10 was the best predictor to model mean crawler density. Binomial sequential sampling plans based on three critical proportions of trees infested (0.40, 0.60, and 0.80) were developed. The adoption of a binomial sequential sampling plan for use in decision making can contribute to improve the management of E. ironsidei and optimize the number of samples needed to make decisions to control this pest.