Climate Intelligence for Dengue Preparedness

Dengue has spread dramatically over recent decades. The World Health Organization reports that around half of the world’s population is now at risk, with outbreaks increasingly affecting large cities. Rapid urbanization, global travel and a warming climate have all contributed.

For city and state health departments, dengue poses a distinctive preparedness challenge. Transmission is concentrated in time - often around and after the monsoon - and in space, in particular neighbourhoods. When a surge arrives, it can overwhelm hospitals, laboratories and blood services within weeks.

Why climate matters for dengue

Aedes mosquitoes, the main vectors of dengue, breed in small containers of clean standing water: buckets, discarded tyres, water storage tanks, construction sites, flower pots. Climate shapes transmission in several ways:

• Rainfall creates and replenishes breeding sites, typically with a lag of several weeks before cases rise.
• Temperature affects mosquito development, survival and the time it takes for the virus to develop inside the mosquito.
• Humidity influences adult mosquito survival.

These relationships are not simple. Very heavy rain can flush breeding sites; water storage during dry periods can sustain breeding. Effects differ between cities and even between neighbourhoods.

From climate data to climate intelligence

Climate data alone is not intelligence. To be useful for preparedness, it needs to be:

1. Harmonized with surveillance data at the same geography and time step.
2. Modelled in ways that capture realistic lags and local differences.
3. Communicated with honest uncertainty, so that decision-makers know how much weight to give a forecast.
4. Connected to the decisions it is meant to inform: ward-level vector control, community engagement and hospital readiness.

What better preparedness could look like

Imagine a municipal team that, several weeks before the expected seasonal rise, can see which wards are likely to experience the steepest increases. Source-reduction drives and larviciding could be concentrated there. Hospitals serving those wards could review bed capacity, testing supplies and platelet availability. Community engagement campaigns could be timed for maximum effect.

None of this requires certainty. It requires a forecast that is good enough, early enough, and accompanied by a clear explanation of what is driving it.

Pitfalls to avoid

• Over-reliance on a single driver. Climate is important, but immunity, serotype shifts, population movement and reporting changes also shape dengue trends.
• Ignoring under-reporting. Many dengue infections are mild or unreported. Surveillance trends reflect testing and care-seeking as well as transmission.
• Mistaking correlation for mechanism. Models should be validated out-of-sample and reviewed by entomologists and epidemiologists.

Epinova’s approach

Epinova’s initial dengue focus is Maharashtra, India - a large and diverse state where dense cities and monsoon seasonality shape transmission. We are developing short-term forecasting that integrates climate signals with surveillance data, designed to support ward-level preparedness rather than headline predictions.

As with all our work, forecasts will carry visible uncertainty, models will be validated before use, and decisions will remain with the public health professionals who know their cities best.