Year-round warmth and wet climates make Colombia and Puerto Rico natural dengue hotspots.
A mosquito-borne viral disease that causes flu-like symptoms—high fever, severe headaches, muscle and joint pain, and rash—dengue is prevalent year-round in both countries, with seasonal peaks typically occurring between May and November.
It’s the most common mosquito-borne disease worldwide. According to the World Health Organization, dengue has surged in prevalence over the past two decades, from about 500,000 reported cases globally in 2000 to more than 14.6 million in 2024. The Americas carry the heaviest burden, accounting for more than 13 million of the cases reported in 2024. Because a dengue infection can be asymptomatic or mild, the true case count is likely much higher. Now, in an effort to get ahead of future outbreaks, a University of Wisconsin-Madison environmental engineer is leading a project to monitor and predict dengue in Colombia and Puerto Rico.
Paul Block, a UW-Madison professor of civil and environmental engineering, is collaborating with Jorge Osorio, a professor of pathobiological sciences. They’re working with the Dengue Branch of the Division of Vector-Borne Diseases at the Centers for Disease Control and Prevention, the One Health Center through a UW-Colombia partnership, and the Red Cross Red Crescent Climate Centre. NASA is funding the project.
Block specializes in subseasonal-to-seasonal models, which make big-picture predictions for a given season several months in advance. These models might predict total snowfall during a winter in Madison, Wisconsin, for example, rather than day-to-day weather events. They tend to be most useful for government agencies and organizations like utilities as they prepare for an upcoming season.
“We’re applying these same types of long-lead predictions—one or three or even six months ahead—to infectious disease challenges,” Block says. “The dengue burden is large in these locations. If we can predict the number of expected dengue cases in the upcoming season, it gives managers an opportunity to prepare.”
Block aims to produce seasonal forecasts that public health and community agencies could use to inform decisions on preventive measures or interventions. Preparations could include procuring and deploying larvicide and insecticide, distributing protective equipment such as netting, and acquiring vaccines. Forecasts could also provide more lead time for shorter-term interventions, like targeted vaccination drives during outbreaks.
One Health teams will collect mosquitoes in Colombia and Puerto Rico to test whether they are carrying dengue and, if so, which strain of the disease. The researchers will also leverage “fever clinic” records and broader public health records going back 15 years. Once they have data, researchers will evaluate the prediction model’s performance against historical dengue data.
They will also incorporate climatological data gathered from NASA satellites into their models. “The remotely sensed data can give us information about surface conditions, like how green it is and the soil moisture state,” Block says. “We’ll use that to collect land surface and hydrology parameters to see how well those correlate with dengue burden. Right now, we don’t know how different land surface and atmospheric conditions are associated with dengue, specifically, or if at all.”
Local stakeholder feedback will shape how the team designs its models and implements them in Colombia and Puerto Rico. That’s an essential step, Block says, because every public health decision carries costs.
“We have to tailor the model so that it’s applicable to the social context of the local setting,” he says. “What that looks like in Puerto Rico could be different from Colombia.”
There are also less tangible factors that can impact how a disease spreads through communities. For example, dengue is more frequently found in urban settings. Year-to-year weather changes influence how residents store water, which then affects how much standing water is available for mosquitoes to breed, especially in cities.
“The citizens of these areas obviously know much better than we do about what’s possible and what’s palatable in their communities,” Block says. “It’s not just a question of making the model, but how do we fit it within these constraints? We could ignore the social aspect and local conditions and just build the model, and it might be pretty good. But if it doesn’t fit with what an agency needs or could realistically use, there may be little to no value.”
Featured image caption: Dengue is the most prevalent mosquito-borne viral infection in the world. Civil and Environmental Engineering Professor Paul Block is leading a NASA-supported project to monitor dengue levels in Colombia and Puerto Rico. iStock photo.