Overwintering mortality in Halyomorpha halys (Hemiptera: Pentatomidae): insights from laboratory and field studies

Overwintering success in insects is shaped by both physiological state and environmental conditions. The invasive brown marmorated stink bug, Halyomorpha halys (Stål) (Hemiptera: Pentatomidae) survives temperate winters through a facultative adult diapause. We quantified overwintering mortality over 4 consecutive winters (2020/2021 to 2023/2024) in South Tyrol, Italy, using controlled laboratory conditions (9 °C, 8:16 h L:D) and a sheltered semi-field setup. Mortality in field-collected cohorts ranged from 48.0% to 65.0% in laboratory trials and 35.2% to 58.9% outdoors, with substantial variation among cohorts overwintered under identical conditions. We compared overwintering survival of reproductively active (fecund) versus diapause-induced (non-fecund) adults produced in the laboratory. Reproductively active individuals experienced near-complete mortality (99.3%), whereas diapause-induced adults showed markedly lower mortality (20.6%). Interannual differences suggested environmental effects, but temperature effects were not consistently associated with observed mortality patterns. Winters with lower minimum temperatures (down to -12.38 °C) and greater cumulative cold exposure were associated with elevated mortality (up to 58.9%). However, substantial mortality also occurred during milder winters, indicating context-dependent temperature effects. Overall, cohort identity was a dominant source of heterogeneity, likely reflecting variation in reproductive state, energetic reserves, and pathogen pressure, as well as other abiotic factors. These results indicate that overwintering mortality in H. halys cannot be captured by a single universal parameter. Improved forecasts of spring population density will require late-season sampling from confirmed overwintering sites and, where feasible, physiological screening to estimate diapause incidence and derive realistic overwintering mortality. Integrating diapause dynamics and associated changes in cold tolerance into mechanistic models may further improve forecasting accuracy.