Under the Volcano

Deaths rattle one of the riskiest disciplines in science

Long after the wounds have scarred over and the fractures have set, when the pain has faded and the nightmares dwindled, the mind echoes with a string of incessant

questions.

“Why did so many die?”

“Could the accident have been prevented?”

And the most haunting of all: “Who is to blame?”

Those are the doubts plaguing an increasing number of volcano scientists in the wake of several disasters that have shaken the small community to its foundations. Since 1975, 29 volcanologists have lost their lives in the line of duty, making this pursuit arguably the most dangerous field of scientific research today.

The deaths -- one as recent as January -- have cast a shadow over the field of volcanology, bringing negative publicity even as the Bush administration considers potentially devastating cuts to the budget for this type of research.

And the unwanted attention is about to explode with the publication next month of two books: Surviving Galeras (Houghton Mifflin) by Stanley Williams, a volcanologist at Arizona State University, and his coauthor, Fen Montaigne, and No Apparent Danger (HarperCollins), by a journalist named Victoria Bruce. Both books are being heavily promoted and cover primarily the same territory: the 1993 eruption that claimed the lives of six scientists and three tourists at a Colombian volcano called Galeras. The similarity ends there, however, for the books take opposite sides on whether Mr. Williams could have averted the disaster had he heeded warning signs.

In their push to sell a viewpoint, though, neither book captures the truth of the Galeras disaster in all of its maddening complexity. Nor do they reach out more broadly to chart where volcanology stands today after so many blows.

Volcano science is currently struggling through an awkward adolescence. A relatively young area of research involving some 400 scientists, it has tried to learn from recent mistakes and is indeed making headway with improved tools and safer practices. The discipline has the laudable goal of protecting vast populations threatened by active volcanoes, many of which reside in the world’s poorer nations. But the idealism and recklessness of youth can sometimes forge a fatal combination. Volcanology has not yet matured enough to avoid repeating some of its earlier errors -- a fact all too tragically illustrated last summer when an eruption in Indonesia killed two scientists and injured several others who had overlooked basic safety recommendations.

Such accidents hurt not only individual investigators but also the the reputation of the entire field, say some researchers. Others voice concern about the attention brought on by the mishaps and the two new books. “I’d hate to see the whole reason for our science and the kind of stuff that happened be sensationalized,” says one of the researchers injured in Indonesia last summer, Michael S. Ramsey, an assistant professor of geology and planetary science at the University of Pittsburgh. “If people start getting the idea that we are just a bunch of rogue scientists who get our kicks out of being close to a volcano, then it will impact on our ability to get funds and our ability to work with international scientists.”

With Mr. Ramsey and the more than two dozen other volcano scientists contacted by The Chronicle, a discussion of risk in their profession inevitably returns to the events at Galeras eight years ago. That accident simmers in the collective memory of the community, touching nearly everyone, because so many of their colleagues perished.

And the errors made at the mountain raised people’s consciousness, sparking a push for improved safety standards that carries on even now with mixed results. To understand the mistakes at Galeras and how they can be corrected, it is important to dissect what happened in Colombia in the days leading up to the disaster.

The fact that Galeras was capable of killing should not have surprised anybody involved in the accident, for they had come to visit the volcano precisely because it was so dangerous. Overlooking the 350,000 residents of a city called Pasto, Galeras had shuddered to life in 1988, only three short years after another Colombian volcano had claimed 23,000 lives. In the wake of that tragedy, volcanologists were particularly concerned about the growing numbers of earthquakes and other signs of unrest at Galeras.

“Between ’89 and ’91, the volcano was going crazy, all the indicators of activity were going crazy,” says John Stix, an associate professor at McGill University who first visited Galeras in 1989.

With relatively little known about the threat posed by the volcano, Mr. Stix teamed up with two other Galeras researchers to organize an international workshop for January 1993 aimed at attracting more people to study Galeras. The other two conveners were Arizona State’s Mr. Williams and Marta Lucía Calvache, who headed the Colombian geological survey’s volcano observatory in Pasto.

The workshop drew top researchers from around the world, who participated in lectures along with trips to the mountain to lay the foundation for future work.

At first, about 50 people signed up to visit the crater of Galeras, but the leaders knew that it would be unsafe to bring such a large group there. The route into the crater descends a precipitous slope where a person must hang onto a rope and can easily kick rocks onto those below, Mr. Stix explains. Moreover, volcanologists prefer to spend as little time as possible in craters, and big groups travel extremely slowly, he says. “Dealing with that kind of number in a crater is just insane.”

So the leaders limited the field trip to the dozen or so scientists who were going to conduct actual research in or near the crater.

The irony of the conference’s timing was that Galeras seemed to be going into hibernation in the months leading up to the January meeting. Just a half-year earlier, the volcano had been all fire and fury. In July 1992, enough gases accumulated inside the mountain that they erupted through a giant plug of rock -- called a lava dome -- that covered much of the volcano’s crater. Galeras blew its cork, much like a shaken champagne bottle.

Then it went nearly silent. The number of earthquakes dwindled, and the volcano exhaled markedly less gas than previously. Mr. Stix says, “I remember thinking in late ’92 and [early] ’93, ‘What’s going on here? Is the volcano on a path of returning to quiescence or not?’ That was one of the main questions on everybody’s mind.”

As visitors arrived for the conference, the seismometers at the Pasto observatory were picking up clues that could answer Mr. Stix’s question. The mountain was trembling with small, unusual earthquakes that left a screw-shaped trace on the recording paper of the seismometers. These tremors were so-called long-period quakes, ones that resemble a low rumble more than a shrill whistle. And they vibrated the ground with a tone much purer than most other earthquakes.

At the time, though, nobody in Colombia knew how to translate those messages from the volcano. “It was not clear to us that this signal was a signal of an impending eruption, although we were worried about that,” says Fernando Gil Cruz, a seismologist at the Volcanological and Seismological Observatory in Manizales, Colombia. Only later, after the disaster, would Colombian seismologists recognize the importance of the tremors that they dubbed"tornillos,” the Spanish word for screws.

There was one person who might have spotted the danger at the time, but he was sitting thousands of miles away in his office at the U.S. Geological Survey in Menlo Park, Calif. A seismologist who specializes in studying long-period earthquakes, Bernard Chouet had traveled to Galeras in November 1991 and had collaborated with Mr. Gil Cruz. After watching the volcano in action, he hypothesized that deep long-period tremors could signal that the volcano was becoming plugged, building up pressure that could lead to an eruption.

Mr. Chouet says that he wrote a report about this idea and gave it to the Colombian scientists. Then the pattern of activity before the July 1992 eruption matched his predictions, he says. “The model was being followed. It was being confirmed.”

He contends that when the tornillos started appearing again two weeks before the 1993 conference, their presence should have raised a red flag for the Colombians. “It should have been obvious,” he says. “I’m sure it was. At the very least, I would imagine that they would have been worried.”

At the time of the conference, however, Mr. Chouet was not there to examine the records or voice his hypothesis. Scientists with the U.S. Geological Survey were barred from going because the U.S. government determined that political instability in Colombia made travel there too dangerous.

Other volcanologists, however, disagree with Mr. Chouet’s contention that the seismic warning signs were clear, especially given that understanding of long-period earthquakes was less advanced eight years ago. Stephen McNutt, a seismologist at the University of Alaska at Fairbanks and the Alaskan Volcano Observatory, attended the 1993 conference and looked at the seismic data after the eruption. “Not being familiar with Galeras, and showing up there for the first time, and based on my experience elsewhere, it was not obvious,” says Mr. McNutt, who before 1993 had studied long-period earthquakes at other volcanoes.

Barry Voight, a professor of geology at Pennsylvania State University at University Park, agrees that the indicators were not easy to read. “I have plenty of admiration for [Mr. Chouet’s] model,” says Mr. Voight, “but you have to realize that small numbers of long-period events are not at all rare on active volcanoes, and there is not a one-to-one correspondence between a few such events and the imminent occurrence of an explosion. ... So my feeling is that -- before hindsight entered the picture -- there was plenty of room for alternative interpretations regarding the hazard implications of the signals.”

Mr. Gil Cruz says he was indeed concerned about the long-period quakes but not overly so because they were happening at a rate of just one or two a day, far less than the dozens of such quakes that preceded the blast in July 1992. Moreover, no new lava dome had formed in the crater, so the situation seemed quite distinct from the one that preceded the blast in July 1992, he says.

In her book, Ms. Bruce says that Mr. Gil Cruz and Fernando Muñoz warned Mr. Williams and others about the tornillos the night before the trip and that there was an argument about what these signs meant. But Mr. Gil Cruz says, “It’s not true. I think Victoria Bruce is mixing up some stories about me and Stanley Williams. Years before, I discussed other topics with Stanley and we discussed [them] strongly. But not in the [1993] workshop and not about tornillos.” (Ms. Bruce responds by saying “He’s told different reporters different things.”)

She says no one there had “the clout or the experience to stand up to Williams and make the call: Galeras is dangerous. Do not go into the volcano.” But Mr. Gil Cruz, Ms. Calvache, and their colleague Mr. Muñoz all told The Chronicle that they decided that it was safe enough to enter the crater. They would not have let their friends go in otherwise, says Mr. Muñoz, who is deputy manager for the geoenvironmental hazards division of the Colombian geological survey in Bogota. “Nobody knew what was going on,” he says.

Each person recalls a different version of the pre-trip discussion. Mr. Gil Cruz says that he talked about the need to work quickly with the other leader of the field trip, Arlés Zapata of the Pasto observatory, and does not remember whether Mr. Williams was even present at the meeting. Ms. Calvache and Mr. Stix say that they did not gather to discuss the volcano’s activity the night before the trip, but they did meet two or three days earlier.

Whatever was said at those meetings, the information about the tornillos should have been conveyed to the scientists who were going on the field trip, and it wasn’t, Ms. Bruce contends in an interview.

The next morning, those scientists and another group traveled over the rutted twisty road up to the top of the volcano. They knew the volcano was dangerous -- it had erupted six months earlier -- but they had no specific apprehension about the mountain that day. On a video tape captured by a Colombian television crew at the summit, a volcanologist named Patty Mothes from Ecuador can be seen telling a reporter that scientists could not tell whether the eruption would happen next week or in five years.

There is something distinctly eerie about watching Ms. Mothes make that statement as some of her friends shouldered their backpacks and prepared to descend toward Galeras’s crater -- and their deaths.

The group had already been pared down to investigators with a clear reason for going into the danger zone. Some planned to take samples of the gases spewing out of fissures, others would monitor the force of gravity on the crater lip, looking for signs that molten rock might be moving beneath the mountain.

To reach their quarry at the mouth of the volcano, the scientists would have to clamber over a route that Dante would have appreciated. The top of the mountain looked something like an amphitheater, a little over a mile across with a wide opening to the west -- the legacy of a past eruption. Within the center of this half-bowl stood a 450-foot-high cone and inside that cone lay the crater of Galeras. At the time the researchers started, however, they could see none of that landscape. A thick blanket of clouds obscured the view, making it seem as if they were stepping off into a void.

Mr. Williams recalls being particularly concerned about the weather and the possibility that researchers might get stuck on the mountain. “I told people to make sure you have lots of food, take a flashlight, take water, because we might get stuck on the volcano. Take rain pants because it’s going to get cold,” he says.

From the end of the road along the upper edge of the amphitheater, the group descended by rope down the face of a sloping cliff. They crossed a low area called “the moat” and then started to climb the outer wall of the cone.

In the thin air at 14,000 feet, the scientists made slow progress, recalls Andrew W. Macfarlane, an associate professor of earth sciences at Florida International University. Panting from the effort, he reached the top of the cone sometime during mid-morning, while fog still obscured the bottom of the crater.

This was Mr. Macfarlane’s first trip into an active volcano, and he was dressed like most of his colleagues, prepared for the cold and rain but lacking any sort of protective gear. Academic researchers, at that time, wore such equipment relatively rarely. Only two researchers on the trip that day had on hard hats, and they were associated with Los Alamos National Laboratory.

Andy Adams of Los Alamos remembers people making comments about his preparations. “Here I am, all dressed up with a hard hat, got a gas mask, safety boots, insulated coveralls, and they’re kind of tittering at me.”

Before the trip, the Colombian scientists from the observatory had emphasized being quick around the crater, but the issue of speed was not stressed by the people leading the trip, Mr. Williams and Mr. Arlés Zapata.

Mr. Williams did suggest that Mr. Adams -- who moved more slowly than others and had finished his work -- should start back. But the leaders did not hurry the other researchers in the three plus hours they spent around the crater.

In his defense, Mr. Williams says he felt that he didn’t need to lecture people. “Among us, we had lots and lots of knowledge. So people respected one another’s opinion. I didn’t order people about how to behave or what to wear. ... And I don’t work for the agencies that have written rules.”

Yet it was one of those very rules, imposed by Los Alamos, that saved Mr. Adams’s life. He had walked back down the side of the cone and started his way up the wall of the amphitheater when a roar split the air at 1:41 p.m. Sizzling-hot rocks the size of softballs rained out of the sky and pelted him on the head. “I’m a firm believer that I might not be here talking to you if I had not been wearing that hard hat,” he says.

His colleagues standing much closer to the crater were not as lucky. Mr. Macfarlane remembers hearing a sound “reminiscent of close thunder or a sonic boom, and for a split second it was not clear to everyone what was going on,” he said in a letter written shortly after the blast. He turned around and saw a black cloud shooting up out of the crater. Rocks as big as television sets fell from the sky, and everybody in his group started running down the side of the cone. “The violence of the impacts was incredible, and when the falling blocks hit boulders on the ground, they shattered and sprayed hot, sharp shrapnel. When the blocks split open, they were glowing hot inside, and the fragments would just lie there and hiss.”

A rock clipped Mr. Macfarlane just above his eye with enough force to daze him and open a gash. He was eventually hauled out of the amphitheater on a stretcher, his teeth chattering from the cold and shock, his face a ghostly white.

The five scientists standing in the crater or on the crater lip died instantly in the eruption. The blast shredded their bodies so thoroughly that investigators could find only separate parts of a single person, hurled more than the length of a football field from where he had stood. The explosion also claimed a sixth scientist and 3 tourists who were farther from the crater at the time.

No kind of protective gear would have saved any of these people, says Peter J. Baxter, a consulting physician at the University of Cambridge who was at the conference and studies how volcanoes kill. In the Galeras blast, “the bodies that were found were quite badly mutilated, so a helmet would not have helped,” he adds.

Yet Dr. Baxter disagrees with Mr. Williams’s contention that helmets would not have made a difference to anybody on the volcano that day. He cites Mr. Williams as a case in point: When Galeras blew, a rock flying out of the crater smashed in the skull of the Arizona State scientist. Reeling from the impact and unable to protect himself, Mr. Williams got pummeled by rocks that crushed one leg and nearly severed the other.

A helmet, concludes Dr. Baxter, “would have protected him from the head injury he got.”

That blow has caused permanent difficulties for Mr. Williams, making his research a struggle, he says. The 48-year-old is not shy about mentioning his cognitive problems repeatedly in interviews, asserting that his injury has impeded his ability to speak and spell.

What’s more, he says, the damage and medical treatments have magnified his already abrasive personality: “I’ve become irritable, I’ve become impatient. I get angry too easily.”

Connected with that, Mr. Williams persisted for years to claim that he was the sole survivor of the group near the crater, and he launched scathing attacks against Mr. Macfarlane and others who were at the site and challenged that story.

Originally, Mr. Williams seems to have believed that he was closer to the blast than other survivors were -- a point disputed by others present that day. Eager for attention, he later simplified the tale until he became the only survivor.

Following the accident, several reporters described the accounts of other survivors, and it was clear that Mr. Williams’s version was wrong. Still, he clung to the story, and it continued to circulate in places such as The New York Times. Mr. Williams stuck with his version even after selling his book proposal, until his co-author, Mr. Montaigne, says he finally forced the issue.

In the end, the publishing houses lapped up Mr. Williams’s dramatic account as well as Ms. Bruce’s controversial case concerning his responsibility. Both book proposals received in the neighborhood of a half-million dollars each, acknowledged Mr. Williams and a representative of HarperCollins.

For most volcanologists, however, the importance of the Galeras disaster reaches far beyond the world of flashy proposals and princely advances. With six of their colleagues dead and several others wounded, the accident forced some scientists to take a hard look at how they practiced their profession. Mr. McNutt from Alaska contacted the International Association of Volcanology and Chemistry of the Earth’s Interior, and it initiated a study of mishaps that have occurred in the past several decades.

The committee ended up issuing safety guidelines in 1995 that can be viewed at its Web site (http://www.iavcei.org/). Besides recommending helmets with chin straps, the group called for scientists visiting volcanoes to develop comprehensive safety plans, approach dangerous areas only when absolutely essential, and spend the minimum time necessary in those places.

The guidelines may help counter an unhealthy attitude that some volcanologists have toward safety, contends Mr. Macfarlane. “There are people in that line of work who really have a cowboy mentality, almost a kind of locker-room mentality about what kind of risks they take.”

To a certain extent, the guidelines and the vivid examples of accidents in the early 1990’s have made a difference. “I think there’s been a whole sea change,” says Dr. Baxter. “Most volcanologists I speak to are now much more cautious than pre-Galeras.” Prior to that accident, he says, many researchers did not fully appreciate the risks involved in their work.

“I think people’s experiences are now building up as more and more people work on volcanoes and realize that it’s really quite foolish not to be fully equipped.”

Yet the lessons of Galeras and the official warnings have not stopped researchers from wandering into active volcanoes without adequate protection. At a workshop in Japan last year, some scientists wore helmets while visiting volcanoes and others did not, according to attendees.

The lapses took a more tragic turn last July in Indonesia. After a meeting of the volcanologists’ international association -- the same one that had earlier issued safety guidelines -- on the island of Bali, a group of Indonesian volcanologists invited some visiting researchers on a trip to climb Mount Semeru, the tallest peak on the nearby island of Java. A notoriously active volcano, Semeru spews a cloud of ash and steam skyward roughly every 40 minutes.

The group had no plans to conduct research on the volcano nor did they originally discuss approaching the active crater. But after an arduous climb to the summit over a 24-hour span, several of the scientists decided to head downhill to get a closer look at the erupting crater about 500 feet below.

Mr. Ramsey from the University of Pittsburgh walked part of the way toward the crater while his colleagues ventured closer. As the next eruption began, Mr. Ramsey started to take a photograph when he saw rocks flying out of the crater. He dropped the camera and ran a few steps away.

Then his mind flashed back to some conversations he had once had with his graduate co-adviser, Mr. Williams, who recounted the lessons of the Galeras disaster. People had survived best in that eruption when they protected their heads from falling rocks instead of trying to run.

Like his colleagues on Semeru, Mr. Ramsey had no helmet so he had to improvise. “I crunched down. Got my face in the dirt. ... So I had my backpack on my back and my camera bag over the back of my neck. A large fragment hit the camera bag and knocked it right out of my hand. I’m pretty convinced that if I hadn’t had my bag there, I would have -- at the least -- been severely injured.” He ended up with bruises, broken toes, and minor burns.

Those closer to the crater fared much worse. Two of the Indonesian scientists took blows to the head and died, while others suffered a range of injuries. Paul Kimberly of the Smithsonian Institution was knocked out and then bombarded with rocks that caused third-degree burns, broken bones, and a deep wound to the shoulder. Lee Siebert of the Smithsonian suffered lesser gashes and burns.

Eight months later, the Smithsonian scientists have largely recovered and have avoided describing the events on the mountain, they say, out of deference to the dead and to prevent harming relationships with their Indonesian colleagues. The two were willing, however, to highlight their own mistakes in hopes of educating others.

“One needs to make sure that the gain justifies the risk in a situation like that,” says Mr. Siebert. “It wasn’t necessary to go to the crater. That’s one of the lessons of this trip that we would hope that volcanologists would draw.”

Aside from lacking helmets or other protective gear, the scientists had not made any plans about what they would accomplish at the crater and what to do in case of any difficulties. “The decision was made somewhat spontaneously, and there was not adequate discussion,” Mr. Siebert says.

Part of the problem, says Mr. Kimberly, was that the scientists made their decision in the thin air at 12,000 feet, and language differences hampered communication within the group. Another problem, says Mr. Kimberly, was “the draw due to curiosity,” a powerful force for scientists.

The recent accidents force the question: Is it critical for volcano scientists to risk as much time as they do in and around craters?

“A resounding no,” answers Mr. Chouet, the USGS seismologist. He uses seismic sensors placed some distance from active craters and says such devices reveal the most about what is going on beneath a volcano. “It is not necessary for people to put themselves close. I think this is more a call to adventure than it is a need for science. ... It’s a little bit like a Formula One car racer. You go there and you get your adrenaline rush and you feel really alive.”

Most other scientists, though, discount that view. “There are some types of data that you’re only going to get from sampling in a [volcanic] vent,” says Thomas Casedevall, a regional director of the U.S. Geological Survey in Denver, whose research involves collecting gases from craters.

He denies that volcanologists are thrill-seekers and notes that even seismologists can get into trouble. Mr. Chouet himself almost died in 1975 when a 50-hour-long blizzard struck while he was working on the flank of Mount Baker in Washington. Although Mr. Casedevall grants that volcanologists need to be more careful, he says that the risk should be put into context. “From a personal perspective, do I feel safer driving the 2 1/2 miles to work or working in an active crater? I think I feel a little more safe working in an active crater because I feel a little more in control.”

To reduce the risks of their job, scientists are trying to develop tools to monitor volcanoes from a distance. But such methods can not yet replace the observing powers of a human, says Donald A. Swanson, scientist-in-charge at the Hawaiian Volcano Observatory. He warns against erring too much on the side of caution by placing limits on volcano studies. “You never know what you’re going to learn from the research that might enable one to minimize the hazards in the future.”

And those hazards hang over the heads of millions of people around the world who live in the shadow of dangerous volcanoes -- a figure that injects a sense of urgency and purpose into the work of scientists studying those mountains.

Mr. Ramsey, who still bears scars from the Semeru eruption, says that the life-saving potential of volcano research far outweighs the risks run by scientists. “If you were doing the observations to clear out the tens of thousands or millions of people in harm’s way, I think that’s worthwhile.”

ANATOMY OF A DISASTER

http://chronicle.com Section: Research & Publishing Page: A18