Dengue is a mosquito-borne viral disease whose transmission is strongly influenced by climate variability. In the Caribbean, where small island developing states are increasingly exposed to climate extremes, the capacity to anticipate outbreaks with sufficient lead time is critical for effective public-health preparedness. In collaboration with the Caribbean Institute for Meteorology and Hydrology, the Caribbean Public Health Agency, and the Ministry of Health and Wellness in Barbados, the Global Health Resilience Group at the Barcelona Supercomputing Center (BSC) developed a climate-informed early warning system (EWS) to predict dengue outbreak risk up to three months in advance.
The EWS is informed by an impact-based modelling framework that integrates both long-lag and short-lag climate drivers to represent the compound and cascading effects of successive extreme events on dengue transmission. Analysis of historical dengue incidence and hydrometeorological data in Barbados identified a consistent climatic sequence preceding outbreaks, characterised by prolonged dry conditions approximately five months beforehand, elevated temperatures at three-month lead times, and exceptionally wet conditions in the month immediately prior to outbreak onset. Explicitly modelling the interaction between these drivers substantially improved predictive performance compared to conventional approaches, with the system correctly identifying around 80 % of recorded outbreaks during validation for the period 2012–2022.
The operational relevance of this approach was demonstrated ahead of the International Cricket Council Men’s T20 World Cup hosted in Barbados in 2024. The model indicated a high probability of dengue transmission months in advance, supporting targeted vector control and enhanced surveillance in areas of increased risk. Building on this application, the Barbados Ministry of Health and Wellness is adopting the model as part of its national dengue early warning strategy, representing an important step towards embedding climatedriven disease forecasting within routine public-health decision-making.
