Deforestation Tied to Changes in Disease Dynamics
Deforestation Tied to Changes in Disease Dynamics

Deforestation Tied to Changes in Disease Dynamics

Numerous studies link habitat destruction to malaria and other vector-borne diseases, but the relationship isn’t always clear.

Jan 29, 2019
Katarina Zimmer

ABOVE: © ISTOCK.COM, VAARA

When a malaria outbreak erupted in Malaysian Borneo in 2002, researchers were surprised to find that the culprit wasn’t Plasmodium malariae, the main mosquito-borne parasite known to infect humans in the area. Instead, the parasite’s DNA turned out to stem from P. knowlesi, colloquially known as “monkey malaria,” which is specialized to infect and proliferate in forest-dwelling macaques. A few accidental cases had been recorded in people over the years, but such an outbreak was unusual. And it didn’t stop there: P. knowlesi has since become the most common cause of malaria in Malaysia, and human infections are steadily rising throughout Southeast Asia.

It’s one of several instances of vector-borne pathogens that have popped up in humans in areas that are undergoing widespread deforestation. The forests of Borneo are being felled at a rapid rate, foremost to make way for palm oil plantations. Researchers have long suspected that the process may play a role in human outbreaks of vector-borne diseases, including malaria, dengue, and others, and evidence is increasingly mounting in support of that idea.

“The more we’re disturbing this natural habitat, the more we’re shaking the pot,” says Amy Vittor, an epidemiologist at the University of Florida’s Emerging Pathogens Institute. With a range of diseases, “the links are becoming clear that disturbance leads to downstream emergence events” in humans, she says.

A string of recent reports has bolstered the link between forest removal and P. knowlesi outbreaks. In a study published this month (January 16) in the Proceedings of the Royal Society B, a group of researchers investigated whether the association holds true for a small patch of rainforest in Malaysian Borneo, where locals fell trees for small-scale agriculture. The team used satellite data to document the change in forest cover over four years, and had a machine learning model analyze the data to see if forest loss could explain the pattern of P. knowlesi infections.

The proportion of cleared land was a strong predictor of P. knowlesi occurrence within a kilometer of the clearance, the researchers found. “The evidence we have from this and other studies strongly supports the association between deforestation and P. knowlesi in this setting,” writes coauthor Kimberly Fornace, an epidemiologist at the London School of Hygiene and Tropical Medicine, in an email to The Scientist.

Catherine Moyes, a spatial ecologist at the University of Oxford’s Big Data Institute, cautions against making claims about causal links between deforestation and P. knowlesi, noting that the algorithm the team used isn’t capable of proving cause and effect. Nevertheless, there is evidence for an association. Through a similar approach in 2016, her model predicted that some of the mosquito vectors that carry P. knowlesi are likely to occur in areas with forest disturbance. A separate statistical analysis by Fornace and her colleagues also found that P. knowlesi cases in humans were positively associated with forest loss over time.

The mechanisms that drive this relationship are unclear. Deforestation may simply be bringing humans into contact with forest-dwelling macaques and infected mosquitoes, as people move into cleared areas and macaques closer to human settlements. In addition, the clearing of trees may shift the locations of breeding sites for mosquitoes that thrive at forest edges. These mechanisms are thought to play a role in a range of vector-borne diseases. And as tropical forests are being felled worldwide at an alarming rate, scientists are becoming increasingly concerned that future pandemics may arise from humans’ destruction of forests.

Malaria and forests: An itchy relationship

The link between vector-borne disease and deforestation has been most thoroughly investigated in human-specialized forms of malaria. Some findings have been striking: One study estimated that a 4.3 percent increase in deforestation over a three-year timeframe was associated with a nearly 50 percent increase of malaria incidence in a small patch of Brazilian Amazon. In one large-scale analysis in 2017, researchers found a positive association between the rates of deforestation and malaria prevalence in 67 countries.

At the same time, other studies have seen no relationship between the disease and cutting down trees, and some even a negative correlation between deforestation and malaria incidence. An extensive 2016 review by Vittor and colleagues found contradictory evidence, depending on a number of factors, such as the type and scale of deforestation and human demographic changes in forested regions.

As forests disappear entirely, and the tree-fringed mosquito-breeding clearings along with it, malaria may likewise decline.

Despite the equivocal results, researchers are still confident there is a link between malaria and deforestation, albeit complex. A major reason for the discrepancies between studies’ results is the difference in habitat preferences across mosquito species, says infectious disease biologist Gabriel Zorello Laporta at Brazil’s Federal University of ABC in São Paulo. Some “deep forest” species thrive in untouched forested habitats, depending on shade to breed as larvae. The malaria forms they carry tend to decrease with deforestation, says Laporta, who recently wrote commentary in The Lancet on the topic. Other malaria-transmitting mosquitoes require sunny areas to breed, and would increase in abundance if large forest areas are cleared, he says. 

The main vector of malaria in the Amazon, Anopheles darlingi, appears to thrive in habitats along forest edges. “You’ll find them on the frontier where forest has been relatively recently cleared,” explains Bentley University entomologist Anthony Kiszewski. The insect breeds in warm, shady pools of water that often build up along roads without proper drainage, such as those that cut through rainforests, he says, or puddles behind logs and debris. A study published last year in Scientific Reports found that places with the highest incidence of malaria in the Brazilian Amazon were cleared forest patches between 0.1 and 5 square kilometers in size. These patches contain the shaded, watery, forest-edge habitat that creates for A. darlingi “a good environment for mosquito proliferation,” explains lead author Leonardo Suveges Moriera Chaves, an epidemiologist at the University of São Paulo. 

Laporta points out that this pattern may not hold up over time: As forests disappear entirely, and the tree-fringed mosquito-breeding clearings along with it, “what you’re going to see is a malaria decrease.” 

Chris Drakeley, a coauthor on the macaque study in Malaysia, agrees: Numbers of P. knowlesi infections spike in the immediate aftermath of deforestation, but fall several years later as the areas are converted into large-scale oil palm plantations, he says. 

See “Predicting Future Zoonotic Disease Outbreaks

In Africa, the main transmitters of malaria are A. gambiae mosquitoes, which lay their eggs only in sunlit pools, and have been shown to prefer farmland habitat over shady forests and swamps. Interestingly, a recent statistical analysis by the Washington, DC-based Center for Global Development, a nonprofit focusing on international development, found no correlation between recent forest loss and malaria rates across 17 countries in sub-Saharan Africa. This was surprising given that previous studies had suggested a positive relationship between the two, explains coauthor Jonah Busch, now at the San Francisco–based Earth Innovation Institute, which researches sustainable development in tropical nations. It’s possible people could have some resistance to malaria already, he speculates, or there could be sufficient prevention strategies in some areas, or fewer people are moving into deforested areas. 

Kiszewski points to a different reason: Many places in this region were deforested decades ago, giving mosquitoes ample opportunity to establish themselves in cleared areas. “If malaria is already present in a community with mosquito habitat, removing another patch of forest is not going to increase the intensity of transmission much on a population basis,” he writes in an email to The Scientist

Zoonotic disease from the forest

While some researchers are teasing out the nuances in the relationship between malaria and deforestation, others are investigating the transmission of some rare zoonotic diseases that humans have come into contact with.  

In 2010, a forested region in eastern Panama known as Darien saw a spike in cases of a disease caused by South American eastern equine encephalitis virus, or Madariaga virus. Like its cousin, Venezuelan equine encephalitis virus, it’s known to infect horses on occasion. The 2010 outbreak was the first ever documented in humans, and is thought to have affected around 100 people, especially children. In many, the virus caused severe neurological problems and is suspected to have resulted one death.

How humans contracted the disease is a mystery, explains Vittor, who is studying the disease. It’s not even clear what the natural host of the virus is. In 2012, she and her colleagues sampled the blood of nearly 600 different mammal species in the area in search of antibodies to Madariaga virus, which would signify an infection. The team found two rodent species that had high levels of antibodies in their blood, and may be possible culprits: the forest-dwelling long-whiskered rice rat and the short-tailed cane mouse, which lives in pasture areas. Both inhabit the young forests that grow back after virgin forest has been cut down. Vittor thinks a mosquito is responsible for transmitting the disease, but it’s not clear which species. 

Vittor is currently exploring the possibility that the outbreak is linked to widespread clearing and disturbance of local forests in the region. “The fragmentation brings people and animal species [together] that normally would not be interacting,” she says, as humans and forest-dwelling creatures come into contact in cleared areas.

The more we’re disturbing this natural habitat, the more we’re shaking the pot.

—Amy Vittor, University of Florida

Human outbreaks of other zoonotic diseases have been linked to fragmented forests, such as Ebola, which is believed to stem from fruit bats. This is not only because disrupted forest areas bring infected animals and humans together, but also because disturbance to forests can alter the communities of forest-dwelling animals in a way that favors disease transmission. For instance, populations of bat species that feed on insects tend to decline in fragmented forests, whereas those of fruit-eating species tend to increase, possibly because fruit-bearing plants are quick to grow back after a forest is cut down. It’s the latter variety that are thought to be the amplifying hosts for Ebola, although recent evidence suggests that insectivorous bats may have been the culprit in the West African outbreak several years ago. Changes in community structure “might be an explanation of why in these altered, anthropogenically disturbed habitats you might find that there’s higher risk of having emerging pathogens,” Vittor suggests.

For some zoonotic diseases, having a diverse community of animals can help buffer the spread of disease transmission to humans—something that has been shown for Lyme disease, notes Kiszewski. In the northeastern US, the bacterium that causes Lyme disease—Borrelia burgdorferi—is transmitted via ticks, but can infect many mammals in forest communities. Of these, only the white-footed mouse can truly sustain infections over long periods of time and pass them onto ticks. In diverse communities, many bacteria end up in dead-end hosts, such as squirrels, raccoons, or deer. By “absorbing a lot of the infection,” he says, diverse communities can limit the number of ticks that carry the disease and can transmit it to humans—a so-called “dilution effect.” Forest loss tends to decrease biodiversity, and thereby increase the risk of humans contracting the disease, he says. However, it’s unclear whether the phenomenon may be widespread among zoonotic diseases, he says. 

There are many examples of vector-borne diseases affecting human populations nowadays that have their origins in forests—such as Zika, chikungunya, and dengue. “All of these were probably initially contained in forest settings in a very limited zoonotic cycle,” Vittor says. Understanding how deforestation may lead to zoonotic disease outbreaks in humans will aid in developing better strategies to prevent them, and possibly ways to predict future disease outbreaks, she adds. Some efforts towards this goal are underway: For instance, EcoHealth Alliance, a New York–based nonprofit that studies infectious diseases, has several projects to track the spread of wildlife-borne diseases in places where humans have intruded into untouched areas. 

Although there is still much research needed on the mechanisms by which deforestation can result in the spread of vector-borne and zoonotic infections to humans, Vittor says, the idea is rapidly gaining traction: “We humans are in our environment and we influence it, and it influences us.”