Forecasting Gypsy Moth (Lepidoptera: Lymantriidae) Defoliation with a Geographical Information System

Abstract

Maps of defoliation caused by the gypsy moth, Lymantria dispar (L.), as well as point estimates of egg mass density and counts of male moths captured in pheromone-baited traps collected from the Shenandoah National Park and George Washington National Forest from 1989- 1992 were assembled in a geographic information system. A maximum likelihood estimation procedure was used to fit 15 logistic regression models that predicted the probability of noticeable defoliation in 1-ha grid cells from various combinations of egg mass densities, counts of males in pheromone traps, presence of defoliation in the previous year, and distance to the expanding gypsy moth population front. Models that incorporated egg mass density estimates and distance to the infested front provided the most reliable predictions of defoliation probability. The performance of these models was comparable with decision errors encountered using various egg mass density thresholds alone. The errors associated with application of egg mass density thresholds and the various models were high and highlighted the need for improved methods for predicting defoliation.