The outbreak began so slowly that no one in Dallas perceived it at first.
In June 2012, a trickle of people began showing up in emergency rooms broiling with fever, complaining that their necks were stiff and that bright lights hurt their eyes. The numbers were initially small; but by the middle of July, there were more than 50 victims each week, slumping in doctors' offices or carried into hospitals comatose or paralysed from inflammation in their brains.
In early August, after nine people died, Dallas county declared a state of emergency: It was caught in an epidemic of what turned out to be West Nile virus, the worst ever experienced by a US city. By the end of the year, 1,162 people had tested positive for the mosquito-borne virus; 216 had become sick enough to be warded; and 19 were dead.
West Nile was not new to the United States. It had been a minor summer threat since August 1999, when it made 17 people sick in New York City. That was the virus' first entry into the US, and it expanded through it thereafter. It landed in Dallas in 2002, sickening 202 people and killing 13. When it moved on towards the West Coast, epidemiologists in the city thought West Nile would no longer be a threat. And events seemed to prove them right: Each year, there were just a handful of cases. In 2011, the year before the epidemic, there was only one.
"We all thought these things come as a flash in the pan: one big outbreak and then you don't see them again," Dr Robert Haley said.
CLEAR THE GARBAGE
IT JUST SNOWBALLS
Climate change is a threat multiplier. If there's one overarching theme that connects almost every way that climate change impacts us, it's that climate change takes a risk that already exists and enhances it. It's not inventing something new. It's taking something that we've already dealt with before, but giving it that extra oomph that makes it a bigger problem.''
DR KATHARINE HAYHOE, a director of Texas Tech's Climate Science Centre.
Dr Haley is director of epidemiology at the University of Texas Southwestern Medical Centre in Dallas and a former disease detective at the Centres for Disease Control and Prevention (CDC). After the last cases were recorded in the final days of 2012, he and a team of researchers studied the episode. Right away, they could see the geography of the illness: Victims were clustered in affluent ZIP codes where many owners had walked away from over-mortgaged mansions. Dr Haley and his team knew there would be abandoned swimming pools and potted plants there - perfect places for mosquitoes to breed unbothered. But the financial crisis was four years old in 2012. The homes had been neglected for years without triggering an epidemic; no matter how many mosquitoes bred in the summer, the deep cold that blankets central Texas a dozen days every winter would knock the bugs down again.
Except, Dr Haley remembered, it had not been very cold that year. There was only one night of hard frost; everyone had talked about it, grateful they did not have to dash out to wrap plants and turn off hose taps while thermometers plunged. The investigators downloaded federal weather data for each year since West Nile first arrived in Texas and plotted the metrics against the case counts. The epidemic year was an outlier on every measure, with the warmest winter, the warmest spring and the heaviest early rainfall in 10 years. It had been a freak weather event, and mosquitoes benefited from it. The insects survived the winter, so there were more of them to start with. They woke sooner, spilled out earlier from their winter hiding places and bit people in greater numbers than in any other year, transmitting so much virus that it made many people gravely ill.
That Dallas' unusual weather favoured the growth of mosquitoes might seem like random bad luck. But Dr Haley does not think of it as an accident. He considers it a warning. Climate change is turning abnormal weather into a common occurrence: Last year was the warmest year on record, the third in a row, and there were more heatwaves, freezes and storms in the US that caused US$1 billion (S$1.4 billion) or more in damage just in 2016 than in the years 1980 to 1984 combined. Anything that improves conditions for mosquitoes tips the scales for the diseases they carry: the West Nile virus that flattened Dallas, the dengue that returned to Florida in 2009 after 63 years, and the latest arrival, Zika, which gained a toehold in the US last year and is expected to surge this year. "These aberrant years are becoming more common," Dr Haley said. "Climate change is... altering the environment in ways that increase the potential for these diseases."
TILTING THE BALANCE IN FAVOUR OF MOZZIES
When the health effects of climate change are discussed, the planet- scale impacts get the attention: rising temperatures, which can cause death from overheating; earlier springs, which pump more pollen towards the allergic; run-off from violent storms, which washes faecal bacteria out of sewer pipes; changing airflows that trap ozone near the ground, stressing the systems of people living with heart disease.
The unpredictable weather patterns stimulated by climate change affect infectious diseases as well as chronic ones. Warmer weather encourages food-borne organisms like salmonella to multiply more rapidly, and warmer seas foster the growth of bacteria like Vibrio that make oysters unsafe to eat. Spikes in heat and humidity have less visible effects, too, changing the numbers and distribution of the insect intermediaries that carry diseases to people.
When former vice-president Al Gore spoke at a meeting on climate and health in Atlanta in February, he chose to start his talk not with a starving polar bear or a glacier falling into the sea, but with images of mosquitoes and ticks. "Climate change is tilting the balance, disrupting natural ecosystems and giving more of an advantage to microbes," Mr Gore said, standing in front of a giant image of Aedes aegypti, the mosquito species that transmits yellow fever and dengue, and now the Zika virus too. "Changing climate conditions change the areas in which these diseases can take root and become endemic."
Right now, yellow fever is causing an epidemic in South America, and dengue has been increasing in Central America. But in the US, the most alarming disease linked to mosquitoes is Zika, which can cause devastating birth defects.
Zika has been a persistent concern since January last year, when a Houston man became the mainland US' first case, arriving back from a trip to El Salvador with the fever, rash and red eyes of full-blown infection. Now more than 5,200 US residents have come down with the virus, at least one in every state except Alaska. The vast majority were infected by being bitten outside the country, and a small number by having sex with someone who was infected that way. But more than 220 people caught Zika from local mosquitoes carrying the virus.
Aedes aegypti are found in more than half the states, from California to Florida, and in cities as far-flung as San Francisco, Kansas City and New Haven; entomologists have found that they regularly survive through the winter in sheltered spots in Washington DC. Unlike the salt-marsh mosquitoes that whine through beach towns at twilight or the night-biting Culex that carry West Nile between birds and humans, aegypti prefer proximity to people; we are their favourite meal. To get to us, they fly into houses and conceal themselves in closets and under beds and furniture. They have evolved to breed in the tiny pools of water we carelessly create around us: in an abandoned tyre, the saucer under a houseplant, even an upturned bottle cap.
Like West Nile, Zika can cause high fevers and paralysis - but unlike West Nile, it can also trigger catastrophic birth defects in the children of women infected while they are pregnant. It appears to destroy brain tissue while a foetus is growing, causing the skull to collapse. It also seems to cause brain damage, and eye, ear and joint abnormalities later on - though what will happen to babies as they grow is uncertain, because all the children born to Zika-infected mothers are still toddlers. The CDC has identified that 1,311 women who were pregnant in the US in the past year were possibly infected with Zika; 56 of their children were born with Zika-related birth defects. In seven cases, the pregnancies ended early, and the foetuses were shown to have been affected. The CDC recently announced that nearly 10 per cent of women infected while they were pregnant had a child with a birth defect - 15 per cent, if they were infected in their first trimester.
The combination of an ugly virus and a stealthy predator is unnerving - especially because in the year since that first Houston case, it has become clear that the US is more vulnerable to Zika than thought. Like generals basing their strategy on the last war they fought, public health experts have set up their defences based on what worked for previous threats. The traps that health departments bought to catch Culex mosquitoes are not attractive to Aedes. The spraying with pesticide by trucks and airplanes that knocks down nuisance mosquitoes cannot reach ones that have sneaked into buildings. The best defence against Aedes mosquitoes turns out to be not big municipal gestures but small individual actions: destroying their habitat by emptying the pools of water where they reproduce, and keeping them from eating by repairing window screens and wearing bug repellent.
Those strategies require that landlords and the municipal authorities pay attention to housing repairs and garbage pick-up, and ask families who probably have other priorities to stay alert to conditions they did not cause.
Since the start of the Zika epidemic, a few disease experts have been warning that the parts of the US where Aedes mosquitoes flourish - the gulf coast, especially its largest cities - are also those that possess the worst poverty and municipal neglect, and that are particularly vulnerable to an outbreak.
"That's an invitation to Zika, right there," said Dr Peter Hotez, founding dean of the National School of Tropical Medicine at Baylor College of Medicine, as he glared at a heap of garbage in Houston's Fifth Ward - a historic - and historically neglected - black community.
The garbage was resting in a grass-lined ditch where another city would have put a sidewalk, an old accommodation to Houston's drenching afternoon thunderstorms. The ditch was damp, and so was the heap. There was a sodden mattress, a turned-over box spring, a pile of torn drywall, old tyres and a dozen plastic bottles lying on their sides; there were scummy puddles caught in the tyres and an inch of old rainwater in the bottles.
The heap was stacked against a sign that said "No Dumping".
Scofflaws would offload trash when no one was watching, trusting that the neighbourhood's lack of political pull would keep them from being held accountable. It was clear Dr Hotez was infuriated by the garbage. In a colder city, it might have represented victimless cheating; in an area where the virus might land, it was a threat to mothers and children.
"We have no historical expertise in how to do Aedes aegypti control," Dr Hotez told me. "We've never done it, and now we're playing catch-up in the middle of an epidemic." He glared at the garbage again. "We have the perfect mix of factors here for Zika transmission to begin."
ZIKA: A HISTORY
Researchers in 1947 trapped and caged a feverish rhesus monkey outdoors in the Zika Forest in Uganda, part of a study looking for yellow fever. The researchers who were monitoring the monkeys injected a sample of the sick monkey's blood into mice. The mice fell sick. The team harvested a pathogen from their brains - not yellow fever, the virus they expected, but something no one had seen before. They named it Zika. Since 1995, that virus has been stored in Galveston, Texas, at the World Reference Centre for Emerging Viruses and Arboviruses, created by Dr Robert Shope and Dr Robert Tesh at Yale University at first. In the collection, among the largest in the world, Zika was seen as one of the least important samples. No agency had ever written a grant to study it.
Dr Shope died in 2004. Three years later, Zika caused its first known outbreak, infecting three- quarters of the population of Yap, a remote island in the Pacific, and leaving almost a quarter with fever, rashes, red eyes and joint pain. Yap lies 800kmfrom any population centre, and no one could demonstrate how the virus landed there - but in 2013, it leapt east across thousands more kilometres, sickening an estimated 19,000 people in French Polynesia. And then it jumped again, landing in Brazil and igniting a worldwide epidemic in 2015, focusing attention for the first time not just on the orphan virus but also on the Galveston lab that harboured it.
"Zika surprised everyone," Dr Tesh said. "Just as West Nile virus did. When West Nile came to the US in 1999, we knew it had occurred in Africa and the Middle East, but nobody thought of it as a serious problem. No one thought Zika would be of much significance either."
Dr Tesh, whom I met in a small office attached to his lab inside the University of Texas Medical Branch tower, spent much of his career hopscotching from the US to developing nations, examining viruses in the places where they emerged. "When Aedes aegypti was introduced to Panama in the 1980s, they started a big campaign of control," he told me. "They got all the school kids to go out and turn over containers. That worked for a year or two or three, and then people lose interest. The same thing happened (in the US), when West Nile appeared in 2002. Everyone in Houston was very scared. They put out messages: Don't go to the baseball game at night. Wear repellent. Wear long- sleeved shirts. Then, after a year or two or three, who doesn't go out to watch the fireworks? Or a concert, or a baseball game? People forget."
In the past century, according to the Environmental Protection Agency, the average temperature in Texas has risen as much as 17 deg C - a noticeable twitch in a system that was stable for millenniums. Rainfall has increased in the central and eastern parts of the state, where most of the population resides. On the coast, the sea has risen nearly 5cm per decade. Scientists have noticed the impact inland, too. "When I came to El Paso eight years ago, you had to look to find Aedes aegypti here," Dr Doug Watts, principal investigator at the University of Texas at El Paso's Mosquito Ecology and Surveillance Laboratory, told me. "Now you find them in just about every backyard."
The effects have rippled farther north too. Since 1980, the amount of time when conditions are ideal for mosquitoes - more warmth, more humidity - has risen by five days in 125 American cities, according to the news and research organisation Climate Central. In 10 cities, the mosquito season has grown by a month. In 21 cities - on parts of the Atlantic coast, in much of Florida, in Mobile, New Orleans, Beaumont outside Houston and south to Corpus Christi - mosquitoes are active at least 190 days per year.
"Climate change is certainly expanding the geographic range of mosquito species, and inevitably the diseases follow them," said Associate Professor Nikos Vasilakis at the University of Texas Medical Branch and a member of its Zika research effort. "But it also shortens what we call the extrinsic incubation period, the time it takes from when a mosquito takes a blood meal to when it becomes infectious. The standard is 14 days, but in warmer periods we can see it as short as nine or 10 days."
If the impact of climate change on mosquitoes and the diseases they carry were predictable, anticipating what comes next might be simple.
It is not. The perturbations that cause a moist early spring like what Dallas had in 2012, favouring mosquito breeding, can equally cause devastating floods - like the wall of water that swept through central Texas in May 2015 and killed 11 people - that will scour mosquito eggs from wherever they have been laid.
Warming temperatures that allow mosquitoes to move north into new territory may also make their current territory inhospitable. In 2012, researchers at Texas Tech University estimated that in Chicago, rising temperatures would expand the length of the season for the mosquito that carries dengue - but in Atlanta and Lubbock, Texas Tech's home turf, summers would become so hot and dry that the risk of transmission would shift to spring and autumn, when residents would not be on guard. The unpredictability will increase the challenge of preparing for diseases whose incidence will also increase.
"Climate change is a threat multiplier," Dr Katharine Hayhoe, one of those researchers and a director of Texas Tech's Climate Science Centre, told me. "If there's one overarching theme that connects almost every way that climate change impacts us, it's that climate change takes a risk that already exists and enhances it. It's not inventing something new. It's taking something that we've already dealt with before, but giving it that extra oomph that makes it a bigger problem."
One other factor complicates the calculation of how much of a threat mosquito-borne diseases pose, this summer and in years to come: the global movement of people. In its month-long lifetime, an Aedes mosquito is unlikely to fly more than 400m from where it hatched. It is possible that an insect carrying the virus could fly into an airplane in one country, ride for thousands of kilometres and fly out into another; every few years, mosquitoes taking a trip like that cause clusters of what is called "airport malaria".
But it is much more likely that Zika will arrive in the bloodstream of an infected person, as yellow fever did in enslaved Africans in the 17th century and dengue in Caribbean traders in the 19th. Mosquitoes carry viruses from one person to another, but it is the movement of those people - for pleasure or family ties or business, in flight from weather or strife - that transports the viruses over borders and into new homes.
There is no city in the parts of the US hospitable to Zika that embodies migration more than Houston does. Its reach to the Gulf of Mexico and halfway to Austin is borderless and porous, a jostling sprawl of people who arrive on direct flights from 30 other countries and through one of the busiest container ports in the US.
One-fourth of its residents were born somewhere else. Harris County, which surrounds it, has 4.5 million residents, who speak 145 languages and live in 34 municipalities across 4,600 sq km.
The possibility of locating the arrival of Zika in that complexity "is a needle in a haystack, times two", said Dr Umair Shah, executive director of Harris County's Public Health Department. Dr Shah is a physician from Cincinnati who studied philosophy as an undergraduate and speaks three languages.
The shock of a new disease is almost always followed by vilification of the people who are believed to have carried it - from the Eastern European Jews blamed for cholera in 1890s New York to the healthcare workers forcibly quarantined because of Ebola fears. Dr Shah, whose parents were immigrants from Pakistan, is conscious of how migration complicates preparing for diseases, and of the need to educate, without blaming.
"There's a narrative that people want to paint, that the reason this is happening to us is because an individual of Latin American origin comes into our country and gives us this disease," he said. "But it could be a businessman who goes to a Zika-affected country, gets bitten, doesn't know he has Zika because he doesn't develop symptoms, comes home, doesn't wear mosquito repellent while he's here because he doesn't know he is a risk, gets bitten again - and bam, it's in our population."
•Maryn McKenna is a senior fellow at the Schuster Institute for Investigative Journalism at Brandeis University.