With a warming climate and changing precipitation, malaria will be on the move in Africa. And a new study suggests that when factoring in hydrology, the area on the continent favorable for the mosquito-borne disease will shrink, and its seasonality will shift. The result highlights how considering processes such as flashing rivers and seeping water can help scientists tease out the disease’s complicated future.
Researchers know that malaria rates are affected by changing temperatures and precipitation. Both the Anopheles mosquitoes that ferry the disease between people and their Plasmodium parasites, which are responsible for the infection, are temperature sensitive. The mosquitoes also depend on standing water to complete their life cycle.
Historically, health researchers and epidemiologists have used rainfall as a proxy for surface water and thus malaria transmission. By comparing case locations with precipitation levels, they’ve deduced that malaria spreads where there’s around 60 to 80 millimeters of rainfall per month, said Mark Smith, a hydrologist at the University of Leeds in England who led the new work. Researchers have used that threshold to determine where the disease may spike.
“It was quite a shock to see that was how it was done,” Smith said. He saw firsthand that rainfall patterns didn’t necessarily explain where malaria-carrying mosquitoes bred. Around a decade ago, he was studying landscape processes in Tanzania in a place that met the rainfall threshold for malaria transmission. But no one was paying attention to the areas where a drying river left puddles or the groundwater that seeped into divots on the ground, he said.
“There’s kind of a disconnect between the actual mechanistic processes that create larval habitat and what we can predict from precipitation.”
Climate simulations use different greenhouse gas emission scenarios to predict future temperatures and precipitation. That made it relatively simple to forecast future malaria hot spots on the basis of different greenhouse gas scenarios. But “there’s kind of a disconnect between the actual mechanistic processes that create larval habitat and what we can predict from precipitation,” said Sadie Ryan, a medical geographer at the University of Florida in Gainesville who wasn’t part of the work.
To get a better sense of the climatic suitability for malaria, Smith and his colleagues combined seven global hydrological models that simulate processes such as water runoff, infiltration into the ground, flow through rivers, and evaporation and include land use and the depth of the water table. Pairing these models with four climate simulations representing emission scenarios from low to high, the team could figure out where surface water occurs and where temperatures would be suitable for malaria throughout Africa.
But because each hydrological model emphasizes different factors in the water cycle to produce its picture of water on the ground, the researchers needed to figure out how to best combine them.
They looked back to 1900, before malaria control measures such as insecticides and bed nets cut transmission rates. The team ran their simulations from 1875 to 1900, then weighted the contribution of each hydrological model until they found the combination that best mimicked malaria distribution in 1900.
Previous malaria forecasts have considered hydrology at a local scale, but this is the first study to cover the entire continent.
Filling in Hydrological Details
The team used their combined model to forecast malaria transmission on the African continent after 2025. They compared environmental suitability for the disease from 2026 to 2050, 2050 to 2075, and 2075 to 2100 with that from 1985 to 2005.
Across time, large areas may become better for malaria transmission, whereas other large areas may become worse. Overall, the team found a decrease in the area in Africa suitable for malaria spread starting in 2025, they reported in the journal Science. But population is slated to rise in many of the areas forecast to host the disease, Smith noted, so determining how transmission rates may change in the future is difficult.
“Where you do see a change, we’re seeing much more pronounced changes.”
On a broad scale, the result looks similar to those of previous precipitation-only simulations. But “where you do see a change, we’re seeing much more pronounced changes,” Smith said. For instance, in some areas, the new model forecasts that the climate will be suitable for transmitting malaria for fewer months of the year—in some cases by more than 4 months—than when rainfall alone is considered.
The new model gives a much more complex and nuanced picture of how the disease may spread in the years to come. With hydrology incorporated, it picks out areas near large rivers such as the Niger and the Zambezi for year-round transmission. But the model’s outputs are still too coarse to be helpful in making decisions about where to deploy malaria interventions, Smith cautioned. And it doesn’t capture how the spread of the disease is influenced by human behaviors, such as irrigating plants and leaving out containers that fill with standing water.
As computing power continues to increase in the decade to come, the models may aid public health interventions, for instance, by creating a data set that could feed into a warning system that could issue alerts about the potential for a bad malaria season, Smith said.
The work is a “great step in the right direction,” said Arne Bomblies, a hydrologist at the University of Vermont in Burlington who wasn’t involved with the study. The current study didn’t look at differences among mosquito species, such as temperature tolerances and breeding environment. Researchers could next look at how hydrological processes overlap with where certain species dominate, he said. With this focus on geography and the water cycle, this approach could help researchers understand the factors involved with malaria transmission. Hydrology is “an integral link between climate and malaria,” he said.
—Carolyn Wilke (@CarolynMWilke), Science Writer