A European modeling study projects a steep rise in its dengue risk indicator under several future climate scenarios. In the CESM2 projection, the model’s annual attack-rate proxy was at least 150 times the 2024 baseline in SSP1-2.6 and SSP2-4.5, reaching 3.2% of the population, and at least 300 times the baseline in SSP5-8.5, reaching 6%.
Those percentages are not observed infection rates. They are outputs from a comparative scenario model, based on cumulative individuals reaching the model’s recovered category in each regional patch; the study presents them as an indicator rather than a point forecast.
A wide range of possible futures
The study also reported a population-at-risk index above 10 million in the optimistic and intermediate scenarios and above 40 million in the worst-case scenario. These are scenario-dependent model indices, not observed case counts.
The analysis divided the European Union, Switzerland, Liechtenstein and Norway into 1,200 NUTS3 patches. These were modeling units, not human participants. The framework combined climate-dependent vector abundance, transmission, human and vector mobility, population projections and a migration model, with mobility and vector-abundance parameters calibrated against European experimental and surveillance evidence.
Comparisons used the ACCESS-CM2, CESM2 and MRI-ESM2-0 climate projections alongside SSP1-2.6, SSP2-4.5 and SSP5-8.5, with 2024 as the present-day climate baseline. The model also varied the level of imported cases.
The map changes with the climate model
By 2030, the highest modeled indicators clustered around international airports in southern Europe, including Italy, Spain and Portugal. Lower and intermediate scenarios mainly intensified existing high-risk areas; SSP5-8.5 expanded the modeled pattern into Cyprus, Bulgaria, Romania, Hungary, Croatia, Slovenia, southern and western France and northern Spain.
The projected geography was not fixed. ACCESS-CM2 placed larger increases in western Europe, while CESM2 and MRI-ESM2-0 placed larger increases in eastern Europe. In the MRI-ESM2-0 runs, the lowest-emission scenario was stable, the intermediate scenario rose slightly with a tenfold increase from 2070, and the highest-emission scenario reached almost 4% of the European population in 2100.
Imported cases matter less as local conditions change
Imported cases remained important in the model. Lower and higher importation scenarios respectively reduced and increased the modeled risk indicator by at least 25%, although the population-at-risk index and high-risk patch count became less sensitive to importation over the century.
Using a threshold of a modeled attack rate above 20%, high-risk patches rose from almost zero in 2024 to at most 150 in lower and intermediate scenarios and more than 250 in SSP5-8.5. The authors interpret the combination of declining importation sensitivity and expanding patch counts as a gradual shift toward risk driven more by local environmental suitability.
Migration was tested as an extreme scenario
A separate population exercise tested climate-related redistribution. It was deliberately extreme—an upper-bound stress test rather than a demographic forecast—but the alternative populations systematically increased the modeled risk indicator. The difference was a few percentage points in lower and intermediate scenarios and tens of percentage points in the highest-emission scenario.
A scenario map, not a local outbreak forecast
Several assumptions limit how literally the numbers can be read. The mean-field approximation treats infected individuals as a continuous quantity, allowing the model to trigger an outbreak below one imported infected individual; the authors say this can affect local incidence, epidemic size and apparent timing.
The document is an arXiv version 1 preprint dated 20 August 2026. Its results are framed as comparative risk maps rather than precise forecasts of where outbreaks will occur.
Paper data and sources
Original title: Climate change and human mobility will shape dengue emergence risk in Europe
Authors: Charley Presigny, Paolo Baglioni, Pietro Rotondo et al.
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-20
DOI: Not available
Original paper · Full text