The Story Behind the Hantavirus Outbreak at Yosemite
The outbreak, which killed three, didn’t only attack the lungs of its victims. It messed with the psyche of the rest of us, too.
Heading out the door? Read this article on the new Outside+ app available now on iOS devices for members! Download the app.
On December 10, Yosemite National Park began demolishing 91 tent cabins in Curry Village, a rustic encampment of 408 canvas-sided cabins jammed into a pine-and-cedar glade near the sloping shoulders of Half Dome. It was here that an outbreak of hantavirus began last summer, infecting at least 10 people and killing three.
But on Sunday, June 10, 2012, the campground seemed idyllic. That weekend held all the promise of early summer. The Curry Village swimming pool was open. The smell of hot dogs and nachos curled out of the snack bar. The sun bounced off the face of Glacier Point. Kids in “Go Climb a Rock” T-shirts shouted and chased each other on bikes.
Sometime that day, a 49-year-old woman from the Los Angeles area arrived at Curry Village’s front desk, a plain wood-floor office that’s often cacophonous with the sound of staffers checking guests in and out. A clerk handed her a key to one of the 91 “signature tent cabins” that opened three years ago—the “new 900s” as they were collectively known. Unlike the older cabins, which are sided with single-ply vinyl-coated canvas, the signature cabins boasted double-wall plywood construction and propane heaters, making them warmer and quieter than the older units.
Off she went, this Southern California lady, to enjoy her Yosemite vacation. We’ll call her Visitor One.
About the same time, another guest checked into Curry Village. He was a 36-year-old man from Alameda County, California, which encompasses Berkeley, Oakland, and the East Bay region. He was given the key to a cabin close to Visitor One’s. He dropped off his things and went about his business. We’ll call him Visitor Two.
We don’t know exactly how Visitors One and Two spent their four days in the park. Medical confidentiality laws forbid public-health officials from releasing their names, and they and their families have chosen to keep their stories private. Maybe they hiked to the top of Half Dome or enjoyed the giant sequoias of the Mariposa Grove. By the following Wednesday, June 13, both visitors had checked out of their Curry Village tent cabins and left the park.
Around Yosemite the summer unfolded quietly. The search-and-rescue team went out on minor events: an ankle fracture on the Panorama Trail, a fallen hiker on the Half Dome cable route. Rangers kept a wary eye on the Cascade Fire, a lightning-sparked wilderness blaze that smoldered through a red fir forest.
Then, in late June, Visitor One fell ill. She might have felt like she had the flu: chills, muscle aches, fever, headache, dizziness, fatigue. The flu goes away after a few days. This didn’t. We do know that, back home, she went to see her doctor. When presented with Visitor One’s symptoms, most physicians would have dismissed it as the flu or, at worse, low-level pneumonia. Her doctor didn’t. They talked about what she might have picked up and where. She mentioned her Yosemite trip. The doctor took the unusual step of calling Charles Mosher, a public-health officer for Mariposa County, which encompasses Yosemite, and asking if there were any known hantavirus cases in the area. “Based on her history and symptoms, [hantavirus] was a definite possibility,” Mosher recalled, so he and Visitor One’s doctor agreed that starting treatment for the virus while awaiting lab confirmation was the prudent way to go.
That was, given the circumstance, about the worst thing Visitor One could hear.
Hantavirus is a strange little zoonotic beast. Zoonosis is the movement of a pathogen from an animal to a human. David Quammen describes it in his recent book, Spillover, as “a word of the future, destined for heavy use in the 21st century.” Some of the most disturbing viruses to emerge in the past 30 years have been zoonotic: Ebola, Hendra virus, HIV, SARS, West Nile. Even as we wipe out more species, humans are coming into closer contact with other animals and inviting their viruses into our world, into our blood.
Hantavirus hit the American radar in 1993. That spring, a healthy young Navajo man and his fiancée died suddenly in the Four Corners area, where the borders of Arizona, New Mexico, Colorado, and Utah meet. Both had suffered acute respiratory failure, which means that they couldn’t breathe. An Indian Health Service doctor, going over death records, discovered five similarly healthy young people who’d all died under similar circumstances. Over the next few weeks, more patients turned up with the same symptoms. Because many victims were Navajo or came from the reservation region, the media initially dubbed it the Navajo flu.
He and his colleagues, he remembers, found a chalkboard and listed “every known possible disease, toxin, chemical, or occupational exposure.”
According to Mark Smolinski, an epidemiologist who studies emerging pandemics for the Skoll Global Threats Fund, a San Francisco non-profit founded by former eBay executive Jeff Skoll, symptoms included “a simple cold, a sudden fever—then trouble breathing and an immediate downhill course.” Back in 1993, Smolinski worked on the Four Corners mystery as a young public-health official. He and his colleagues, he remembers, found a chalkboard and listed “every known possible disease, toxin, chemical, or occupational exposure.”
Finally, virologists at the Centers for Disease Control and Prevention (CDC), working with tissue samples from the victims, discovered a genetic link to hantaviruses, a family of European viruses previously unknown in North America. The European virus caused kidney failure, but the Four Corners victims died of lung failure. What emerged in the Southwest that spring was a new strain of hantavirus endemic to North America. Scientists called the strain Sin Nombre, Spanish for nameless. Thirty percent of deer mice trapped in the Four Corners area were found to be carrying the Sin Nombre strain.
Mice infected with hantavirus shed the living virus in their urine, feces, and saliva. It takes just the right combination of timing, drying, and aerosolization for humans to become infected. A mouse pees or poops, the excrement dries onto dust particles, then those particles get swept up into the air. Human infection most commonly happens in confined spaces like houses, cabins, or storage areas, but it’s got to happen within 48 hours of the mouse shedding the virus—researchers have, so far, seen hantavirus survive for only two days outside a host.
Once a human inhales the virus, it can take up to five weeks to incubate. Then the victim comes down with flu-like symptoms that can linger for days. The symptoms might actually lessen for a bit. Then, all of a sudden, things take a turn for the worse. Much worse.
“It felt like the flu for about a week, and then one night I got horribly sick,” says Ethan Lindsey, a 34-year-old public-radio producer, who contracted hantavirus during a visit to Montana from his home in Bend, Oregon, in 2009. A month later, he ran a Fourth of July 5K. Felt great. Four days after that, he came down with what felt like the flu. “I was drained, bone weary, coughing,” he says. He went to the doctor, who couldn’t figure it out. It wasn’t swine flu, maybe just a mild case of pneumonia. Lindsey went home and rested, but continued to decline. “I felt horribly sick,” he says. “About midnight, my fiancée drove me to the hospital,” the St. Charles Medical Center in Bend. “I almost fainted on the way there.”
The turn from flu to respiratory failure happens so fast that one hantavirus survivor group on Facebook advises people who suspect they’ve been infected to designate a friend to monitor them around the clock. If things go south, a victim may be too sick to reach the hospital alone.
If the disease reaches that point—and it’s unclear what percentage of those infected with hantavirus actually do—it’s bad. Breathing begins to fail. Some victims describe the feeling of a band tightening around their chest or of being smothered with a pillow. “They put me on oxygen because it was becoming hard to breathe,” says Lindsey. “I remember the doctor telling me, ‘Listen, your oxygen saturation is at 60 percent and dropping fast. Once it hits 40 percent, major organs are going to fail and you’re going to die. We think we need to put you into a medically induced coma.’”
They put Lindsey under for nine days. Doctors hooked him up to a high-frequency oscillating ventilator because his lungs were filling with fluid. At one point his fever ran so high that nurses put his unconscious body in an ice bath.
Lindsey’s doctors still didn’t know it was hantavirus. “It’s so rare, why would they test me for that, right?” he says. The hantavirus test takes four to seven days. It’s extremely nuanced and difficult to interpret. There are a lot of false positives.
Even if they’d known, there wasn’t much more to be done. There is no known cure. The best that doctors can do is hook up a patient to a ventilator and let the body fight the virus on its own. The fortunate, like Lindsey, recover. But many are not fortunate.
“We had a 75 percent mortality rate in those Four Corners cases,” says Smolinski. “We wanted to know: Did this just pop up, or did something change? And were there some ecological pressures that caused it to evolve to the point where we started to recognize it?”
There were two answers. A study of autopsies in the Southwest indicated that there had been fatal cases of hantavirus pulmonary syndrome, or HPS (as the disease associated with the Sin Nombre strain came to be called), dating back to the 1950s. At that time, doctors had attributed the deaths to respiratory distress, cause unknown. In fact, CDC investigators found that traditional Navajo medicine recognized a similar disease—and associated its occurrence with mice.
Today, researchers believe that hantavirus coevolved with North American rodents. Deer mice are the virus’ main host, though variations appear in white-footed mice, cotton rats, and rice rats as well. The rodents are considered reservoirs, which means that they harbor the virus without being affected by it.
To Smolinski’s second question, there were indeed ecological pressures. An unusually wet winter led to a bumper crop of nuts on the region’s piñon pines that spring. Flush with food, mice populations exploded. The virus hadn’t mutated. There were simply more mice about, which meant more chances for people to come in contact with the virus.
The 1993 outbreak ultimately claimed 27 lives—a mortality rate of 56 percent. That’s shockingly high. The Ebola virus, top of the scary charts, has a mortality rate of 30 to 90 percent. But there was one piece of good news: hantavirus didn’t transmit person-to-person. It moved from mouse to human, but humans seemed to be dead-end vessels. Infected patients weren’t contagious.
The hanta scare died down after 1993. But the virus didn’t go away. About 30 people come down with hantavirus in the U.S. every year. The mortality rate remains high: 37 percent of patients diagnosed with hantavirus die. It shows up all over the place, but predominantly in the West. New Mexico: 91 cases since 1993. Colorado: 78. Montana: 35. California: 48.
In most cases, the carriers of the virus turned out to be deer mice. Which is a challenge. You can’t wipe out the disease by killing all the deer mice. (The deer mouse, Peromyscus maniculatus, prefers rural fields, forests, and outbuildings; it has nothing to do with deer, despite its name. The mice you trap in your basement are more likely house mice, Mus musculus, which aren’t known carriers.) For North American carnivores, deer mice are like nature’s wheat. They’re the food staple for birds, cats, reptiles, and canines. They grow like a crop during spring and summer, and with a certain percentage of them grows the hanta.
Visitor one was lucky. Her body fought off the virus, and she recovered. Tests eventually confirmed that it was, as Charles Mosher suspected, hantavirus.
At that point—mid-July—there was no cause for alarm at the park. Aside from the original 1993 outbreak, medical experts believed hantavirus almost never occurred in clusters. Though she visited Yosemite, Visitor One could have caught it anywhere on her visit to the eastern Sierra.
“When we saw that first case, we assumed it was isolated—like all were, to that point,” says Danielle Buttke, a veterinary epidemiologist who works for the National Park Service. Wildlife diseases are Buttke’s specialty. She deals with tick-borne relapsing fever at Grand Canyon National Park, rabid beavers at Delaware Water Gap National Recreation Area, and West Nile virus in mosquitoes at Fire Island National Seashore.
Buttke got a call about Visitor One from the California Department of Public Health in July. A CDPH official told her a case of hantavirus had been reported in a California resident who’d visited Yosemite in June. Health officials couldn’t be certain that she’d contracted it in Yosemite.
Then Visitor Two came down with the flu. He wasn’t so lucky. The 36-year-old Alameda County man checked himself into the hospital on July 30. On July 31, he was dead.
A quick check of lodging records revealed that Visitor One and Visitor Two had one commonality: both stayed in the signature tent cabins, the new 900s.
Alarm bells rang. Buttke met a state public-health team, including Mariposa County public-health officer Charles Mosher, in Curry Village. Their weapon of choice: pencils.
“As we go through those dwellings, we’re trying to identify any gaps a quarter-inch or larger,” Buttke says. That’s all a mouse needs to squeeze into a cabin. “If you can stick a pencil through a gap, it needs to be addressed.”
During their inspections, the team discovered something interesting. “When you pulled back that canvas, you could see there were mice living in the void space between the walls,” Buttke says. “They were nesting in the insulation.”
Suddenly, the picture became clearer. One of the outbreak’s big mysteries—why did the hanta victims come only from the new 900s and not the older tent cabins?—found a possible solution.
Hantavirus isn’t a long-lived bacterium like anthrax, which can remain viable in dormant spores for decades. “Typically it survives only a couple days,” said Charles Chiu, an infectious-disease physician at the University of California at San Francisco. Chiu, one of the nation’s leading virus researchers, heads UCSF’s Viral Diagnostics and Discovery Center. He spent part of his summer analyzing the Yosemite hantavirus strain. (It turned out to be very similar but not identical to the Sin Nombre strain.)
For a human to be infected by such a short-lived virus, you need an ongoing infestation.
In Curry Village, deer mice might have occasionally scurried across the floor of the older soft-sided tents. But in the hard-sided 900s they lived in the walls, continually shedding virus.
Then there’s the food. If you walk around Curry Village, you can’t swing a cat without hitting “Mother Curry’s Camp Etiquette,” signs listing the camp rules. Yosemite has a well-known bear problem, and food is strictly prohibited inside the canvas cabins, so hungry bears won’t rip through the walls. Guests store food in outdoor bear-proof boxes instead.
One of the most disconcerting aspects of Yosemite’s hanta outbreak, in retrospect, was the period between the reopening of the sanitized 900s in early August and the public acknowledgment of the outbreak on August 16.
But the new 900s had different rules. “Hard-sided cabins may store food inside cabins,” proclaimed the Etiquette. “Keep food out of sight, and doors and windows latched.”
How to attract mice: 1) Offer cozy nests; 2) Provide food.
“We found evidence of food in the cabins, evidence of mice eating food dropped on the floor,” Buttke says.
Mouse populations are extremely dynamic. Conception to birth: 21 days. They can reproduce six weeks after birth. Given a safe nesting ground and a bountiful food supply, a mouse population can explode. Here’s where things really get interesting. When a hanta-infected male starts fighting for territory, he bites and scratches his competitors, which spreads the virus mouse to mouse. In field and forest, predators such as foxes, coyotes, and owls rein in a mouse’s urge for territorial expansion. Leave the nest to fight, and you risk getting eaten. In predator-free Curry Village, not so much. So while 14 percent of California’s mouse population may carry hanta, the infection rate within a double-walled cabin could be much higher.
In early August, the park shut down the 91 signature tent cabins. Cleaning crews gave every cabin a sanitizing deep clean. Maintenance workers sealed gaps and tightened up the vinyl-coated canvas draped over the double-plywood walls. They hammered fresh one-by-fours around the base of each of the new 900s. No way any mouse was squeezing through that. When CDPH officials inspected and cleared the fortified cabins, the park reopened the 900 series to visitors.
But with that long incubation period, people kept getting sick.
On August 12, Visitor Three died. He was a 45-year-old man from Pennsylvania, and we don’t know much about him except for the fact that he stayed in a signature tent cabin prior to the end of July.
“With the third case, we realized this was extremely unusual,” Buttke said. “Something was definitely not right.”
On August 16, Yosemite and CDPH officials publicly acknowledged the hantavirus cases for the first time. The news exploded. Set a killer virus in a globally famous setting like Yosemite and you’ve got a story that, well, goes viral.
That wasn’t all bad. Curry Village suffered a rash of cancellations and Yosemite’s visitation numbers dipped, but park officials wanted to alert the 1,700 guests who’d overnighted in the 900s that summer to be aware of the situation. If they felt themselves coming down with the flu, the Park Service advised them to get tested for hantavirus. A handful of early-summer guests turned up positive.
One of the most disconcerting aspects of Yosemite’s hanta outbreak, in retrospect, was the period between the reopening of the sanitized 900s in early August and the public acknowledgment of the outbreak on August 16. During that window, visitors were checked into signature tent cabins without being informed about the outbreak. Al and Pamela Oligino from Laguna Beach, California, met their son’s family at Yosemite on August 13. They stayed three nights in cabins 966 and 954. “One day, three official-looking guys came by and started inspecting the empty cabin next to us,” Oligino says. “We found out a few days later through the press that there’d been this hantavirus outbreak, which was very alarming. They knew weeks before, yet they didn’t tell us we were at risk.” At one point, Pamela Oligino says, a Curry Village staff member loaned her a broom to sweep out their cabin. Sweeping can send the virus into the air and potentially into the lungs. “Obviously, one employee missed the training session,” she said.
On August 12, Visitor Three died. He was a 45-year-old man from Pennsylvania, and we don’t know much about him except for the fact that he stayed in a signature tent cabin prior to the end of July.
Tom Medema, Yosemite’s chief of interpretation, says park officials relied on state public-health inspectors for guidance on closings, openings, and public alerts. “We’re not disease experts in the way they are,” he says. The belief was that the extra precautions taken to rid the cabins of mice, along with redoubled cleaning efforts, had reduced the risk of infection to near-zero. And, in fact, nobody who stayed in the new 900s after the end of July contracted the virus.
Eventually, though the mice found their way back in. The inspectors that Al Oligino noticed continued to check the cabins for rodent signs. In late August, they found them. Check and mate. On August 28, Yosemite officials permanently closed all 91 signature tent cabins. On December 10, park maintenance crews began removing the entire double-wall 900 series. They’ll be replaced by tent cabins built in the traditional single-wall style. “Our hope is that they’ll be online before next summer,” said Medema.
The irony is that the new 900s were opened in 2009 to keep visitors out of harm’s way. In October 2008, a section of rock cleaved away from Glacier Point’s 3,200-foot granite wall. A load equal to about 500 dump trucks thundered down and crushed an unoccupied cabin. In response, park officials condemned more than 200 cabins in Glacier Point’s drop zone. The new 900s went up, away from the falling rock. Nobody, it was thought, should die because they were assigned an unlucky cabin.
Looking back on it, the Yosemite hantavirus outbreak appears to be a case study in modern virus hunting and the psychology of disease. What gave the story legs was its combination of the well-known and the unknown: iconic brand name meets mysterious peril.
The effects of the story were something to see. I visited Yosemite in mid-September, and I can tell you I had to tamp down my own anxiety to do so. I considered bringing my kids along to see El Capitan but ultimately decided against it. No sense possibly exposing them. On the way to Yosemite, I stopped at Home Depot and bought a respirator mask. Just in case.
When I arrived at Curry Village on a sunny autumn day, I counted only a handful of lodgers. At the bike rental stand, I tested a fat-tire cruiser and asked the counter guy if business had slowed. He glanced at a barn full of unrented two-wheelers. “Not by choice, unfortunately,” he said. “Gotta love the media.”
I wandered around for a while, maskless. The place was quiet. A breeze moved through the trees. I poked my head inside one of the soft-sided 500s. In the 900 area, I noticed that park employees were living in tent cabins next to the notorious signature cabins, separated only by a wooden fence. Which threw me. Why would they be safe when cabins 20 feet away were too risky to enter? I couldn’t decide if I was a fool for poking my nose into a potential virus chamber or an overreacting hypochondriac.
What gave the story legs was its combination of the well-known and the unknown: iconic brand name meets mysterious peril.
I wasn’t the only one with qualms. Susan Smartt is the CEO of NatureBridge, which runs outdoor-education programs in the West. Every year more than 30,000 schoolkids and teachers enjoy NatureBridge programs at Yosemite, Washington’s Olympic National Park, California’s Channel Islands, and the Santa Monica Mountains.
The hantavirus outbreak forced Smartt to think hard about risk. NatureBridge students often stay in the signature tent cabins during the school year. Shortly after the outbreak was announced, NatureBridge suspended its Yosemite program. “The health and well-being of our students have to come first,” Smartt says. “Our programs come second—or we won’t have programs.”
Smartt was in a tough spot, though. NatureBridge exists to connect kids with the great outdoors, especially urban kids with little experience in nature. They come from the inner cities of Oakland and San Francisco and from as far away as Texas and Japan. “Some have never seen snow,” she says. Those kids—and their parents—may already feel trepidation about heading into the woods. Add a killer virus into the mix and you’ve got a tough sell.
Smartt kept in close contact with Yosemite officials. When they felt that the outbreak had been contained, Smartt started thinking about bringing the kids back—but not to Curry Village. She worked with Delaware North, the park’s lodging concessionaire, to find space in the Cedar Lodge and the Wawona Hotel. “I’m on the phone this week firming up our reservations,” she said in early October.
The good news: NatureBridge resumed operations on October 14. The bad news: 62 school groups had their Yosemite visits canceled. “There were some schools where the school board said: We can’t do it this year. It just feels too risky,” Smartt says.
That’s a shame. For some kids, this might have been their only chance to experience the outdoors. And what an experience! Yosemite. It’s like being introduced to church in the Sistine Chapel. But I couldn’t blame those school boards. After all, I kept my own kids home.
I left Curry Village and wandered upvalley toward El Capitan. The strange psychology of disease worked my brain. Yosemite’s three hantavirus deaths freaked out a lot of people. Yet during that same summer, more than 5,000 cases of West Nile virus broke out across the United States. It was the nation’s worst West Nile outbreak ever; by late November, 247 people would die. The mayor of Dallas authorized aerial spraying of insecticide. But most people didn’t cancel vacation plans.
Hanta outbreaks are consistently rare—about 30 isolated cases a year. Lyme disease, after rising to nearly 30,000 annual cases in 2009, has held steady at 22,000 to 24,000 cases over the past few years, the vast majority of them in the Northeast, mid-Atlantic, and Upper Great Lakes regions. West Nile, by contrast, has gone on a tear. Since its first appearance in New York 13 years ago (62 cases), it’s gone nationwide. The CDC reports that this year’s 5,245 cases—which it called an epidemic—were spread over 48 states. Only Alaska and Hawaii were immune.
I can’t punch that germ in the nose and make it run away. Because of that, diseases are especially freakish.
While we’re talking about relative risk, consider this: last summer the Merced River, which runs through the Yosemite Valley, claimed four drowning victims. All four drowned within hiking distance of Curry Village. Most people never heard about those fatalities.
Mark Schaller, a professor of psychology at the University of British Columbia, has studied the human psychological response to disease. I asked him about the very different reactions to hantavirus, West Nile, and the river deaths. “The emotions aroused by the perception of threat don’t always correlate with a rational assessment of the risk,” he said. “We’re often responding to the more ancient part of our brain, and historically, infectious diseases have been a huge threat.”
“Think of a charging bear or a roaring river,” he says. “Those are big physical threats that can kill us or our children. But I can see the bear, I can whack it with a stick, I can avoid the threat. Same with the river. Sure, I could drown, but I can see the damn river!”
The psychological power of disease, by contrast, lies partly in its invisibility. “These viruses could be lurking anywhere,” Schaller says. “Our usual sensory apparatuses—eyes, ears, nose—can’t detect them. I can’t punch that germ in the nose and make it run away. Because of that, diseases are especially freakish.”
I think it also has to do with the calculus of each disease. Mosquitoes pick up West Nile from birds and transfer it to humans. With an annual average of 3,500 cases nationally, it’s unusual but not rare. And 80 percent of people infected with West Nile have no symptoms at all. Twenty percent will develop flu-like symptoms. Fewer than one percent will suffer life-threatening encephalitis or meningitis—and those folks are more likely to be elderly, immune compromised, or very young. So, West Nile: wildly more prevalent than hantavirus, but exponentially more survivable.
Hanta, from its very first case, has mowed down the young and the hale.
Hanta, on the other hand, is extremely rare. In 2012 there was about one hantavirus case for every 175 West Nile cases in the U.S. But somehow that rarity doesn’t lessen the scariness. In fact, it does the opposite. The flu has killed multitudes more people than the Ebola virus. And yet.
The mortality rate doesn’t help matters. Thirty-seven percent is an eye-opener. The mind can dismiss West Nile victims as frail and elderly. Not gonna happen to me. Hanta, from its very first case, has mowed down the young and the hale.
Finally, there’s a sense that Hanta is not playing fair. It’s a virus waging asymmetrical warfare. By infecting victims in Yosemite’s tent cabins, it attacked in one of America’s most sacred secular spaces. Our parks are the home of wholesomeness; those tent cabins are chapels of health and vigor. A haunted house is one thing. A haunted church scares us shitless.
On my way to El Cap I stopped in at Camp Four, the legendary climber’s hangout. September is the traditional start of climbing season in the valley. Hantavirus didn’t seem to scare any of them away. “Camp 4 is FULL,” said a sign at the ranger’s station.
I spoke with three young guys who’d just come off the rock. Russell Facente, Justin Loyka, and J.D. Merritt planned to be in Yosemite for most of September. They talked about hanta and about taking care with tents and food. But the virus didn’t seem to phase them.
“I’m from the Southwest,” Facente told me. “So I’m pretty familiar with hantavirus. I understand it’s around. But you take precautions. Keep an eye out for mouse droppings.”
Familiarity. In this case it doesn’t breed contempt. It breeds calm. It breeds rationality.
I continued on to the foot of El Capitan and I touched its granite and was sorry my kids weren’t there. Next time they will be.
Contributing editor Bruce Barcott is the author of The Last Flight of the Scarlet Macaw: One Woman’s Fight to Save The World’s Most Beautiful Bird.