Egg mass sampling plans for gypsy moth management programs

The goal of this research was to develop gypsy moth egg mass sampling plans that reflect the influence of habitat, changes in egg mass distribution, and provide populations densities or density categories for making control decisions.

Sequential egg mass sampling plans for gypsy moth management programs in urban and suburban habitats were developed from 0.01 ha fixed-radius plot samples collected in Loudoun, Fairfax, and Arlington Counties, Virginia. The sampling plans were develop from Wald's sequential probability ratio test and is based on a negative binomial distribution. Operating characteristic and average sample number curves were used to determine the acceptability of the sampling plans. Three sampling plans were developed for the action thresholds of 618, 1,236, and 2,471 egg masses/ha.

The use of binomial sampling for low density (<618 egg masses/ha) gypsy moth populations in continuously forested habitats was examined. Fixed- and variable-radius plot egg mass samples were collected in 28 study areas in Virginia, Maryland, and Massachusetts. The relationship between egg mass density and the proportion of trees with zero egg masses was developed. Binomial sampling resulted in a higher relative variability and lower relative efficiency than the fixed- and variable-radius plot sampling method. Binomial sampling was determined not to be an effective sampling method for gypsy moth populations below 618 egg masses/ha.

Fixed- and variable-radius plot egg mass samples were taken when leaves were present (summer) and absent (winter) in 136 sample sites in Virginia. A significant difference between summer and winter counts was determined. The relationship between summer and winter counts was quantified using nonparametric Statistics. Winter egg mass counts were found to be 14 to 34 percent higher than summer egg mass counts . The probability of a summer egg mass count exceeding an action threshold was constructed by fitting a logistic curve to empirical data for the action thresholds of 618 and 1,236 egg masses/ha. Egg mass counts need to compensate for differences between summer and winter counts. Alternatively, the probability of a summer egg mass count exceeding an action threshold should be utilized.