Extreme weather, ocean acidification, mass extinction, melting ice caps, rising sea levels – pick your favourite apocalypse and it is likely you will find a link to climate change. But the spread of pestilence? Could climate change be responsible for that, too?

It was thought that one of the few benefits of global warming might be fewer deaths from respiratory illnesses, since winter cold and poor ventilation tend to go along with a sharp rise in cases of flu. Scientists now are not so sure climate change will bring that bit of positive news for human health.

Robert Wallace, a phylogeographer at the Unversity of Minnesota, said he was sceptical initially that climate change would lead to more pandemic influenza, but recent studies from around the world have led him to change his mind. Wallace studies the way viruses, such as influenza or HIV, evolve and disperse among humans, and in particular how this happens in response to what humans themselves do.

In a talk at the University of Washington, Wallace pointed to work by virologist and avian flu expert Robert Webster who found that H5N1 samples are becoming increasingly viable at warmer temperatures. This strain of influenza typically breaks out in winter, but Webster and colleague Diane Hulse-Post showed that current strains survive in faeces at higher temperatures than previously thought. The World Health Organization suggested this may explain how the virus has resurfaced in summer months in Asia.

“If the results do hold, the implications are fundamental to the virus’s persistence in equatorial estuaries in which recombinants (viruses formed by recombining genetic material) appear to emerge across migrating bird species,” Wallace said.

Wallace cited the work of Cambridge University’s Derrick Smith, a specialist in infectious disease informatics who Science has described as “the unofficial cartographer of influenza”. Smith found that normal, seasonal influenzas are usually “seeded” in a few countries in Southeast Asia in the off-season, before emerging to spread to the rest of the world.

Seasonal influenzas in the North are arriving earlier in the colder-weather flu season and “in a nastier way,” said Wallace. He pointed to a 2013 study of climate and U.S. flu epidemics from 1997-98 onward by Arizona State University mathematician Sherry Towers who found a “significant association” between warm winters and severe epidemics that follow. After a warm winter the next year they saw early onset influenza A (H1N1) and B (H3N2) – the two main types of flu that cause seasonal epidemics.

“In the event of continued global warming, warmer than average winters are expected to occur more frequently,” said Towers. “Our work suggests that mild influenza seasons during unusually warm winters are a harbinger of the likelihood of an unusually severe season to come.”

A third factor, Wallace said is the impact of climate change on birds. As warming temperatures push wild birds to winter further north, they come in contact with new populations of poultry, “where avian influenzas epidemiologically incubate.”

Changing La Niña conditions – the move northward of warm currents – may contribute to the generation of new pandemic strains, he said, citing a 2011 paper by Columbia University climate scientist Jeffrey Shaman and his colleague Marc Lipsitch, an epidemiologist with Harvard’s Center for Communicable Disease Dynamics. They found that the four most recent human influenza pandemics (1918, 1957, 1968, and 2009) were preceded by La Niña conditions in the equatorial Pacific.

Increased intensity and frequency of La Niña conditions as a result of global warming, they hypothesized, bring different flu subtypes together in particular regions as birds and animals not normally found together are mixed, encouraging the simultaneous multiple infection of hosts and a jumbling up of influenza genetic material into new combinations.

The fundamental problem facing researchers investigating any link between climate change and infectious disease is the sheer complexity of separating out underlying causal relationships.

The seasonal nature of flu outbreaks – the winter months in the north and monsoon season in the tropics, suggests a climate link. But climatologist Christopher Fuhrmann of the U.S. National Oceanic and Atmospheric Administration is more cautious: “The seemingly simple question of why epidemics in temperate regions occur in the wintertime continues to remain elusive.”

Fuhrmann is concerned that the bulk of research has been conducted by public health and medical communities through epidemiological studies or under controlled lab conditions, with few contributions from other physical and social science fields. He called for more research on the relationship between climate and virulence, transmission, host susceptibility and respiratory-particle aerosol dynamics that go beyond simple correlations with meteorological variables.

A survey by University of Georgia ecologist Sonia Altizer and colleagues, covering climate effects on infectious disease more broadly beyond just human influenza, concluded that scientists still know relatively little about how climate change will drive host-pathogen evolution. They also argue that due to the regional variability of the impact of climate change, predicting and detecting local impacts might be more important than predicting global changes.

Researchers also do not have enough data to understand how vectors and disease agents respond to temperature.

“You’d think we would know how the life span of each important mosquito vector responded to temperature – but we don’t, and this sort of basic research is difficult to find funding for,” said Kevin Lafferty of the U.S. Geological Survey. “And don’t get me started about precipitation.”

“I think it is encouraging that more scientists are focusing on understanding the climate-disease relationship instead of just sounding the alarm.”

According to the WHO, statistical modelling has been applied to climate change impacts on malaria, dengue fever and, in the U.S. to encephalitis. While some models show a net increase in malaria over the next five decades, others show not so much. In addition, the global health body describes how predictions assume unchanged levels of human intervention.

If there is any certainty in this field, it is that humans do intervene in the face of heightened threat from infectious disease. Which means that socioeconomic factors and the scale of human response can confound our ability to detect the relationship between climate change and human pathogens. A future increase in per capita GDP, for example, might increase disease control efforts and this could offset increases in climate suitability for an infectious disease, notes Lafferty.

“Many scientists have moved away from the simplistic notion that climate will increase infectious diseases, and more toward predicting how changes in climate might lead to changes in infectious disease risk,” said Lafferty.

“There is now a general appreciation that climate suitability for infectious diseases won’t so much increase as shift toward the poles. If you live in the northern hemisphere, I suggest you look south. The infectious diseases they are dealing with there could be in your future.”