A climate-informed dengue transmission model with Bayesian decision support

It is crucial to model the spread of dengue under different weather conditions so that we can understand how it changes with the seasons and plan targeted actions. Conventional deterministic models frequently overlook the time-varying impact of environmental factors, hence constraining their forecast accuracy and policy significance.

This research formulates a non-autonomous, climate-sensitive SEIR-SI dengue model with crucial parameters like transmission rate, vector mortality, and recruitment are articulated as explicit functions of temperature, humidity, and precipitation. The basic reproduction number R₀ is analytically determined using a next-generation operator formulated from the evolution operator of the linearized infection subsystem, hence providing the threshold condition that R₀<1 keeps the disease-free equilibrium stable, when R₀>1 means the persistence of the disease. The model is integrated into a Bayesian inference framework utilizing affine-invariant MCMC to predict intervention intensities and climate-response coefficients, while measuring posterior uncertainty.

The climate-informed model formulation improved the fit by approximately 20%-25% relative to constant-rate baselines, accurately capturing the seasonal peak timing and amplitude. The posterior medians indicated that the controls were moderate yet synergistic: larval source management 0.28 [0.13-0.41], nets 0.12 [0.08-0.15], and spraying 0.19 [0.17-0.21]. Scenario analyses have shown that the efficacy of fixed strategies is compromised by warmer, humid weeks. Specifically, the incidence was increased by nearly 19% when the number of infections increased by +2°C warming and +5% humidity, or by approximately 10% for each factor.

This approach provides a more flexible and useful representation of dengue dynamics by combining theoretical foundations with Bayesian uncertainty quantification. The findings highlight the need for portfolio-based, climate-responsive interventions-such as larval source management, personal protection, and targeted spraying during high-risk climatic windows.