Researchers at the University of Málaga announced that outbreaks of the Ebola virus are tied to large-scale climate patterns rather than occurring at random intervals, pointing toward a new method for early epidemiological surveillance.
Tracking Teleconnections from the Caribbean to Central Africa
By analyzing information on 49 Ebola outbreaks recorded across Africa between 1976 and 2025 and comparing them against 26 climate oscillation indices, the researchers found that shifts occurring thousands of kilometers away from Central Africa share a statistical relationship with the disease's emergence.
The team used data up to 2018 to build their predictive models, reserving the years 2019 through 2025 to test whether the models could successfully anticipate new outbreaks. Their findings highlight four key climate oscillation indices tied to this distribution: the Caribbean Sea surface temperature index (CAR), the Indian Ocean Dipole (DMI), the quasi-biennial oscillation of stratospheric winds (QBO), and the Pacific Decadal Oscillation (PDO), with the latter serving as the most prominent macroclimatic factor in the model.
As detailed in Stochastic Environmental Research and Risk Assessment, these remote climatic indicators do not directly cause Ebola. Instead, researchers explain that large oscillations modify regional weather patterns, such as rainfall across the African continent, which in turn alters ecological conditions that favor the appearance of outbreaks. Jesús Olivero, a professor specializing in Biogeography and Macroecology at the University of Málaga and another of the study's signatories, clarifies that this does not mean these remote climate indicators directly cause Ebola, but rather that they can act as an anticipatory clue
regarding environmental conditions closer to the biological phenomenon.
Climate Patterns Provide Early Warning for Central Africa
The findings arrive as Central Africa continues to grapple with an epidemic.
While the newly identified climate patterns offer a potential early warning system, researchers emphasize that the predictive capability remains limited. The useful window for these large-scale climate oscillations sits between six and twelve months prior to a potential event. Consequently, further study is required before this approach can function as an autonomous predictive system, though it already provides a viable framework to flag vulnerable periods and regions for targeted public health preparation.
