How climate change is expanding mosquito habitat and spreading diseases to new regions — and how AI can help predict their next move
Malaria has been called “the undisputed queen of parasitic diseases.” It’s a protozoa that causes high fever and chills and that, if left untreated, can lead to severe complications and death — all transmitted via nothing more than a mosquito bite. In the mid-20th century many countries successfully eradicated this mosquito-transmitted disease through insecticide spraying and environmental engineering. The Centers for Disease Control was founded in 1946 with a mission to eradicate malaria on American soil; the US has been essentially malaria-free since 1951.
But perhaps not for much longer. This past summer, 8 locally-acquired cases of malaria were reported in Florida and Texas, the first such cases in 20 years. This could be just the tip of the iceberg — and it’s not just Americans who need to worry. As reported in a recent BlueDot webinar, projections show that 100 million more cases of malaria are expected worldwide in the next 15 years. The worst-case scenario models show that, within 50 years, 90 percent of the world’s population will be at risk.
The situation is similar, though perhaps not as drastic, for all mosquito-borne diseases: be it Zika, chikungunya, dengue or others, case counts will rise in the years ahead, and outbreaks will appear in unexpected places. Mosquitoes may move slowly, but the diseases they carry are projected to spread rapidly. But it’s possible to stay ahead of the curve if you understand the reasons behind the change, and know what to look for.
Mosquito-borne diseases: up, over and across
BlueDot’s surveillance team, which has been tracking mosquito-borne disease outbreaks for years, has lately been chronicling some noticeable and surprising pattern changes. Those changes fall into three broad types of phenomena, each one as startling as the next.
1. Historically large outbreaks in endemic countries.
Countries that are no strangers to mosquito-borne diseases are experiencing outbreaks larger than they’ve ever seen before. Dengue fever is perhaps the best example: outbreaks in some countries have increased by multiple orders of magnitude. Those countries aren’t concentrated in a specific region of the globe, either; we’re seeing this phenomenon in all corners of the globe.
| Country | Cases 2023 | Increase vs 2022 |
|---|---|---|
| Bangladesh | 309,087 | 451% |
| Cambodia | 26,912 | 184% |
| Malaysia | 100,936 | 97% |
| Mexico | 45,982 | 355% |
| Peru | 35,375 | 914% |
| Thailand | 136,655 | 223% |
In the case of Bangladesh, the country experienced an explosive increase in dengue cases in 2023. It was observed that the country’s increased rainfall and expanded monsoon season promoted an ideal environment for mosquito populations to thrive — and for dengue to spread.
2. Outbreaks in non-endemic countries.
The second phenomenon is outbreaks in countries that have no prior history with a particular mosquito-spread disease, and whose populations are considered “immunologically naïve” — meaning they have never been exposed to a specific antigen.
Uruguay, for example, experienced a substantial outbreak of chikungunya in 2023. The outbreak came on the heels of a larger outbreak in nearby Paraguay, with 19 of Uruguay’s cases due to travel between the two countries. But at least 51 of Uruguay’s cases were due to local, mosquito-borne transmission. Now that the disease has taken hold, Uruguayans are likely to be dealing with similar outbreaks in the years ahead.
3. Geographic migration.
There are more than 3,500 species of mosquito on the planet; they are native to every continent except Antarctica. But only a select few species carry the pathogens that are a danger to humans — and they are carrying those pathogens farther and farther afield.
Consider what happened in Ethiopia in 2022: after a 20-year decline in malaria cases, Ethiopia experienced a spike of more than 100,000 cases in a three-month period. The culprit: Anopheles stephensi, a mosquito that is native to India and Iran, but that has been steadily moving eastward across Africa and has been identified as far away as Nigeria and Ghana.
Riding the wave of climate change
Though each of these phenomena represents a very different situation, they all share a common thread: mosquitoes and the diseases they carry are thriving amid the planet’s changing climate.
Climate change is having adverse effects on weather patterns, which are creating conducive environments for mosquito larvae to thrive. It was observed that these affects may have contributed to Bangladesh’s 2023 historic dengue fever increase. Furthermore, the malaria-carrying Anopheles stephensi mosquito species is proving to be well-adapted to both dry climates and urban environments, with increased presence of man-made reservoirs that accompany urbanization. And Uruguay’s chikungunya outbreak shows how, as local climates become more hospitable to invasive mosquito species, human travel can unexpectedly give those species a free ride.
3 Top Takeaways on Mosquito-Borne Diseases
1. Mosquito habitat is expanding. Changing climate conditions are opening up new regions of the world for the mosquito species that carry diseases such as malaria, dengue, chikungunya and others.
2. The outbreaks are getting bigger, and showing up in unusual places. In countries that are no strangers to mosquito-borne diseases, outbreaks are hitting historical highs. And places that were once considered free from those dangers — and whose populations have never been exposed to their antigens — are now experiencing their own outbreaks.
3. Artificial intelligence can help predict what’s next. BlueDot is employing AI to predict which regions could soon host invasive vector species, and to identify emerging mosquito-borne disease outbreaks faster than ever before — crucial intelligence that can help mitigate the impact of disease.
The bad news is that the effort to combat climate change won’t necessarily help combat disease change. “Even if we were able to reverse climate change tomorrow, mosquito habitat has already changed and will continue to change,” says Dr. Hernan Acosta, Medical Scientific Liaison at BlueDot. “For the time being, humans and mosquitoes are stuck in a toxic marriage.”
But climate change is not a random event: by examining its patterns, humans can predict which regions of the world, previously inhospitable to certain mosquito species, are likely to provide them with a new and comfortable place to call home in the years to come.
BlueDot’s own modeling shows that some Canada-US border regions, for instance, may become environmentally suitable to Aedes aegypti and Aedes albopictus mosquito species in the next 10 years, potentially exposing new major population centres to the mosquito-borne disease chikungunya. “That doesn’t necessarily mean that the disease will appear there,” explains Dr. Acosta, “but it’s more important than ever for those places to be aware of the risks their populations might face.”
Tracking mosquito-borne disease with artificial intelligence
Given the astonishing projections for mosquito-borne diseases, and especially their likelihood to pop up in unexpected places, we need better and more constant detection of emerging threats and outbreaks. Recourse to official sources alone won’t be enough to stay abreast of the issue.
It’s BlueDot’s core mission to develop early-warning intelligence about outbreaks: to identify them at the earliest stages of their development and to forecast their potential spread, no matter where in the world they might first appear. “The best way to describe it is that we are always looking for needles in haystacks,” says Dr. Acosta. “And we are now using artificial intelligence as a kind of magnet, so that we can find the needles faster.”
As part of our ongoing disease surveillance, BlueDot has always scanned the world for articles and other digital sources about potential outbreaks — a process that has become exponentially faster through the use of Large Language Models (LLMs), the technology that powers many AI applications.
Thanks to AI, BlueDot is now able to scan and process the contents of thousands of sources each day, in multiple languages, to search for credible reports of emerging mosquito-borne disease outbreaks. And our LLM knows what to look for, and how to identify credible sources, because we have trained it under the supervision of human expertise. When it comes to disease surveillance, this is no small task: it requires oversight by epidemiologists, immunologists, virologists, environmental scientists and more.
But the result is worth the effort. A purpose-built, properly-trained LLM can do all the primary-source work of disease surveillance in a matter of seconds to a depth of detail no human can match. It can find reports of specific diseases and syndromes. It can isolate locations, assemble timelines of events and cross-reference sources. It can also help our experts project the course of a disease — including a mosquito-borne illness —and disseminate information quickly.
And when our system detects a potential outbreak of a vector-borne disease such as malaria, dengue or chikungunya in an unusual location, we bring our human expertise to bear upon the LLM’s findings — including our geographical modeling of mosquito habitat expansion — to confirm it and turn the information into actionable intelligence.
“What AI does at BlueDot is help us to quickly separate the signal from the noise,” says Dr. Acosta. “It augments the surveillance techniques we have built through years of expertise. It’s no substitute for human intelligence, but it provides us with an increase in capacity we would not otherwise have.”
