An empirical model for predicting insects' diapause termination and phenology: An application to <i>Cydia pomonella</i>

Abstract Diapause is a vital survival strategy for insects, enabling them to conserve energy and endure adverse conditions. Understanding how diapause affects insect phenology and population dynamics is crucial for the effective management of insect pests. Predictive pest phenological models can be invaluable tools for providing essential information to support management strategies. This study presents a modelling framework to incorporate diapause into phenological models when biological information on variables regulating and functions describing diapause induction and termination are lacking or limited. In our framework, insect phenology is divided into a set of phases characterized by specific events (diapause induction and termination) and processes (development of diapausing and post‐diapausing biological stages). The phenology is simulated by a stage‐structured model based on the Kolmogorov equation, and the temperature‐dependent development rate functions are described by the Brière functional form. Our modelling framework was tested on a case study involving the prediction of the phenology of the codling moth, ( Cydia pomonella L. 1758). Model calibration and validation were performed using four time‐series adult trap catch data collected in the Emilia Romagna Region from 2021 to 2023. The calibration procedure allowed obtaining realistic parameters related to the temperature threshold triggering diapause termination and the development rate function of post‐diapausing larvae and pupae. Model validation proved successful in simulating both the initial emergence and the overall phenological patterns of adults across the three observed generations. The methodological framework proposed here aims to facilitate the introduction of diapause in phenological models improving also their predictive abilities. The model may serve as an accurate and knowledge‐based tool for planning and implementing pest monitoring and control actions based on the realistic predictions provided by the model on the phenological status of the pest.