In Brazil's Atlantic Forest, three of four mosquito meals are now human

By David Ogilvy, Chief Marketing Officer at Mosticare Global | Published 2026-05-06

A team of Brazilian biologists has just published the genetic contents of two dozen mosquito stomachs collected in fragments of the Atlantic Forest in Rio de Janeiro state. Eighteen of them were full of human blood. One mosquito had bitten a wild dog. One had bitten a frog. Six had fed on birds. The rest of the menu, in a forest that once teemed with mammals, was made up almost entirely of us.

The paper, published in Frontiers in Ecology and Evolution on 15 January 2026 by Dálete Cássia Vieira Alves, Sérgio Lisboa Machado, Jeronimo Alencar, and colleagues from the Oswaldo Cruz Institute and the Federal University of Rio de Janeiro, is small in scale but big in implication. The researchers trapped 1,714 mosquitoes across two protected sites — the Guapiaçu Ecological Reserve and Sítio Recanto Preservar — and from those captured 145 engorged females. After DNA sequencing of the cytochrome b gene from each blood sample, they could confidently identify the host species in 24 cases. Three of every four meals had come from a human being.

That is not how forest mosquitoes are supposed to feed.

The forest is empty, so the mosquitoes adapted

The Atlantic Forest is one of the most biologically rich biomes on Earth and one of the most diminished. It once covered roughly 1.3 million square kilometres along Brazil's east coast — an area larger than France, Germany, and Italy combined. Today, less than 30% remains, broken into thousands of small fragments and surrounded by farms, towns, and motorways. The forest's mammal populations — primates, agoutis, deer, pacas — have collapsed in step with its trees.

Mosquitoes, however, do not collapse. They adapt. Sérgio Lisboa Machado, one of the authors, put it plainly to Mongabay: "Once the vertebrate population decreases, moving for other habitats, mosquitoes ... go in search of new blood sources." The new blood source is the species that tends to live around the edges of fragmented forest: ours.

Nine different mosquito species in the study had fed on humans, including Aedes albopictus — the tiger mosquito already familiar to readers in southern France, Italy, and Spain — alongside Aedes scapularis, Coquillettidia fasciolata, Psorophora ferox, and several Anopheles species. The authors describe a "clear tendency" for the captured species to feed predominantly on people. In ecological terms, this is a shift from zoophilic to anthropophilic biting behaviour, and it is happening in the wild rather than in a laboratory.

Why this matters beyond a Brazilian forest

Mosquitoes are not just irritants. The species in question carry yellow fever, dengue, Zika, chikungunya, and Mayaro virus. When mosquitoes bite a wider range of wildlife, pathogens cycle through species that often act as buffers; many vertebrates clear or dilute the virus, and the link to human cases stays weak. When mosquitoes bite humans almost exclusively, that buffer disappears. Every infected mosquito becomes a more direct route from forest to person.

The pattern has been documented before in scattered studies — in Amazonian fragments, in West African urban fringes, and in Sri Lankan estate landscapes — but this is one of the cleanest demonstrations using modern genetic methods in the Atlantic Forest specifically. It joins a growing body of evidence that biodiversity loss is not only an aesthetic or ethical problem but an epidemiological one.

For European readers, the temptation is to file the finding under "Brazilian problem". That would be a mistake. The same Aedes albopictus that drank Brazilian human blood in this study is now established in 369 NUTS-3 regions across 26 European countries, according to the European Centre for Disease Prevention and Control. Europe's own forests are heavily fragmented; its peri-urban mosaics of housing, gardens, motorways, and patches of woodland look — from a mosquito's point of view — strikingly like the Atlantic Forest at the edges. The lesson is not that we will see Mayaro in Madrid next summer. The lesson is that the simpler the surrounding ecosystem, the more concentrated the mosquito's attention on us.

A small caveat the headlines tend to skip

It is worth being honest about the numbers. Only 24 of the 145 engorged females yielded a successful host identification — about 17% of the captured engorged females, or 38% of the samples that made it to DNA amplification. The authors are explicit about this limitation in their methods. A sample of 24 is a strong signal but not the last word. The follow-up question, which the authors flag for future work, is whether the same pattern holds across more sites, more seasons, and after controlling for trap location relative to human dwellings. A trap closer to a research station will, unsurprisingly, catch mosquitoes that have just fed on the researchers.

The shape of the result, though, is consistent with what other groups have found and with what mosquito ecology would predict. When the buffet of wild hosts shrinks, the menu narrows. We are usually still on it.

What to watch next

Three things, if you follow this strand of vector ecology. First, replication: are similar blood-meal ratios reported in studies underway in the Cerrado and Amazon, where deforestation is faster and patches are larger? Second, surveillance: do European trap networks — the UKHSA's 1,300-trap programme, France's ARS sites, Italy's Emilia-Romagna network — start sequencing blood meals as well as counting mosquitoes? Genetics is cheap now; the data exists, it just needs collecting. Third, policy: do environmental ministries begin to count mosquito-borne disease risk as part of the cost of forest fragmentation? It would be unusual. It would also be sensible.

Mosticare's view on this is straightforward. Vector ecology does not stop at the edge of a tropical reserve; it follows the same logic everywhere humans simplify a landscape. The most enduring protection a household can take against mosquitoes is the one that does not depend on which animals the local forest still contains: a physical barrier between people and biting insects, around the bed and in the garden. The science behind that conclusion has not changed. The forest, on the other hand, is changing fast.

Sources cited

1. Alves, D. C. V., Machado, S. L., Silva, J. dos S., de Almeida, N. M., Dias, R., Silva, S. O. F., & Alencar, J. (2026). Mosquito blood-feeding patterns in Atlantic Forest fragments of Rio de Janeiro. Frontiers in Ecology and Evolution, 15 January 2026. https://www.frontiersin.org/journals/ecology-and-evolution/articles/10.3389/fevo.2025.1721533/full
2. Bascomb, B. (2026, 16 January). Mosquitoes in Brazil's Atlantic Forest prefer human blood. Mongabay. https://news.mongabay.com/short-article/2026/01/mosquitoes-in-brazils-atlantic-forest-prefer-human-blood/
3. European Centre for Disease Prevention and Control. (2025, June). Aedes albopictus — current known distribution: June 2025. https://www.ecdc.europa.eu/en/publications-data/aedes-albopictus-current-known-distribution-june-2025