A Temperature-Driven Population Dynamics Model Based on Biweekly Insect Trap Counts for Stored-Product Pest Management

Insect trap records are widely used for monitoring stored-product pests but are less frequently applied in predictive decision-support systems. The present study aimed to evaluate whether biweekly trap data could support the development of a temperature-driven population dynamics model for stored-product pest management. A facility-level time series of aggregated biweekly insect counts was analyzed using a discrete negative binomial regression model, with mean temperature during the preceding 14 days and the number of insecticidal spraying applications within each monitoring interval as predictors. The fitted model showed that recent temperature was positively associated with expected insect counts, whereas spraying applications were negatively associated with expected counts. Specifically, each 1°C increase in 14-day mean temperature was associated with a 17.5% increase in expected biweekly insect counts, while each additional spraying application was associated with a 39.6% reduction. Scenario analysis was used to compare alternative temperature-triggered spray policies. A standard threshold policy produced outcomes very similar to those of the observed historical schedule, suggesting that the original intervention timing was already broadly aligned with a biologically plausible temperature-based rule. In contrast, a preventive intensified policy substantially reduced predicted insect counts but required a markedly higher number of spraying applications. Overall, the results indicate that routine biweekly trap data can support a practical, facility-level population dynamics model and can be used to quantify trade-offs between expected pest suppression and intervention effort. The proposed framework provides a proof of concept for transforming routine monitoring records into an operational tool for pest-management decision support.