Summary 1 The accidental introduction of the Asian strain of gypsy moth (AGM) Lymantria dispar (L.) to New Zealand poses a major threat to New Zealand's forestry industry. To aid eradication and control decisions in the event of its establishment, a model was developed for the effect of nuclear polyhedrosis virus (NPV) as biological control for AGM in New Zealand. 2 The model simulates within- and between-season gypsy moth population dynamics, including temperature-dependent development, density dependence through predation and resource limitation, and interactions with NPV. 3 Following its introduction to New Zealand, AGM is predicted to increase more quickly and have more severe outbreaks than the European strain. In the absence of predators, the model predicts initial outbreaks then damped oscillations to an equilibrium. 4 In the model, a single application of NPV (2500 GPIB ha−1) at the time of maximum larval density gave up to 80% suppression of peak larval densities in the following year. The same level of suppression was achieved in the absence of predators. 5 In the long term, the model predicted that spraying when an outbreak was just beginning gave best results (a 50–70% reduction of the following two outbreaks). Simulation of threshold spraying resulted in NPV application, on average, every 7 years and suppression of outbreak densities by 40–70%. Following a single application, NPV was maintained in the population as a classical biological control agent, giving approximately a 20% reduction in outbreak densities. 6 Eradication of AGM using NPV was possible if larval densities were very low.