This is the authors' abstract. We don't add key points for this paper.
Accurate timing of insect emergence—especially for invasive pests—is critical for surveillance and timely interventions in smart agricultural systems. Yet, many species lack experimental developmental data required for traditional phenology models. Here, we present an empirical, binary logistic regression model that predicts the first adult emergence of the invasive beetle Anoplophora horsfieldii using citizen‐science occurrence data (iNaturalist) and degree‐day accumulation. We developed a dual‐binning strategy in constructing binary data that balances early emergence sensitivity using 10 DD bins with accurate saturation estimation using 20 DD bins. The estimated binary-based logistic model of slope = 0.0146 and intercept = -14.471 yielded a 50% emergence probability at 990.0 DD and demonstrated a robust fit. The model was validated using independent field observations from Jeju Island (2024 and 2025), which fell within the early transition zone (21.8 % emergence probability). In validation using independent 2025 citizen‐science data, ROC analysis yielded an AUC (Area Under the Curve) of 0.871, indicating the model's good discriminatory ability. Our results highlight that meaningful emergence predictions can be achieved from fragmented, presence–absence data, offering a data-driven tool for early detection and biofix estimation of non-native pests. By aligning with core smart agriculture principles—predictive analytics, early warning, and scalable data integration—this framework supports operational decision‐making and enables real‐time phenology forecasting in data‐limited agricultural and managed landscape contexts.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Heo et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: