Abstract
Dengue poses a rapidly increasing threat to global health, with Southeast Asia among the most affected regions. Climate-informed early warning systems can help mitigate outbreaks; however, prediction of large outbreaks with sufficient lead time remains a challenge. In this study, we quantified the role of climatic variation and serotype competition in shaping dengue risk in Singapore using over 20 years of weekly case data. We integrated these insights into a forecasting framework capable of predicting dengue outbreaks up to two months ahead and generated counterfactual projections to assess the impact of novel interventions, such as Wolbachia. While a climate-informed model improved predictive power by 54% compared to a seasonal baseline, including serotype information increased this to 60%, better explaining interannual variation in dengue incidence. By incorporating serotype competition as a proxy for population immunity, this work advances the field of climate-informed dengue prediction and demonstrates the value of long-term virus surveillance.