LAUSR

BACKGROUND Developing a model that adequately explains pest population dynamics based on weather-related parameters is fundamentally important for proper pest management. Autocorrelation with past occurrences should be considered when modeling the relationship between the time series of pest occurrence data and meteorological factors; however, few attempts have been made to model these factors simultaneously. In this study, we constructed an autoregressive integrated moving average with exogenous variables (ARIMAX) model to represent the occurrence of the common cutworm, Spodoptera litura (F.) (Lepidoptera: Noctuidae), a major moth pest species in Asia, using the trap catch data of S. litura recorded approximately every 5 days. The multiple meteorological measurements taken over several past periods before S. litura occurrence were included as explanatory variables to evaluate their lag effects on future occurrences. RESULT It was suggested that temperature had the most important effect on S.litura occurrences among other meteorological factors (i.e., humidity, wind speed, and precipitation). Especially, higher temperatures during the larval/egg stage seemed to presage a higher moth abundance. When the model was fitted using independent data that were not used for calibrating the model, the model was able to capture trends in increases in the scale of occurrence, particularly after July, when the occurrence rapidly increased. CONCLUSION Past temperature condition, particularly during the larval and egg states, is suggested to be highly important for predicting future S. litura occurrences. The ARIMAX model proposed here will allow preventive measures to be taken, effectively safeguarding food resources against pest outbreaks. This article is protected by copyright. All rights reserved.