Volcano's Deadly Warning homepage

After the eruption of Nevado del Ruiz killed more than 23,000 people in Colombia in 1985, the U.S. Office of Foreign Disaster Assistance asked the U.S. Geological Survey to design a program to help foreign governments cope with volcano crises. The result was the Volcano Disaster Assistance Program, or VDAP. Based at the Cascades Volcano Observatory in Vancouver, Washington, this crack unit is standing by at all times, ready to lend a hand at the next volcano showing signs of blowing up. NOVA asked VDAP's chief, volcanologist Dan Miller, what it's like to be on the team.

NOVA: Earlier this year you spent five weeks in Ecuador. What were you doing?

Miller: In early December 2001 we started getting phone calls and e-mails from our colleagues down at the Geophysical Institute in Quito about earthquakes that were occurring under Cotopaxi volcano. Cotopaxi is a huge volcano, nearly 20,000 feet high. It's covered with massive glaciers, and it has a long and ugly history of producing eruptions with devastating lahars or mudflows that go down the valleys leading away from the volcano.

By Christmastime it became clear that an intrusion was taking place. That is, molten rock or magma was moving up from some deep location under Cotopaxi into the edifice itself. As a consequence, there was an official request for a VDAP team to help our colleagues there install some additional seismic monitoring equipment and put in mudflow warning equipment around Cotopaxi.

Early in January five of us left for Ecuador. We spent most of our time either in the office upgrading equipment to receive the new signals or in the field around Cotopaxi installing acoustic flow monitors.

NOVA: What are those?

Miller: They are telemetered stations that we put near the valley floors of volcanoes that are covered with snow and ice. They are detectors that can pick up vibrations that lahars or mudflows produce as they move down a valley. When lahars are detected, warnings can be given to people living downstream so that they can seek high ground. We installed a network of nearly a dozen of these in the drainages that head north and also west and then south from Cotopaxi. It's actually one of the most extensive mudflow monitoring networks in the world.

NOVA: And how is Cotopaxi at the moment?

Miller: For the time being, installing the equipment on the volcano seems to have helped it go back to sleep. It's a fitful rest, though, and there are days when additional swarms of earthquakes occur. Cotopaxi endangers more than 100,000 people, who live on valley floors that have been repeatedly swept by massive lahars from even modest-sized eruptions of the volcano.

NOVA: Are there any other volcanoes that are currently in pre-eruption stage?

Miller: In Ecuador, there is another volcano that has been erupting for over two and a half years now. It's called Tungurahua, and it woke up in September of 1999. VDAP has sent several teams of scientists to help instrument that volcano and to work with colleagues there. That eruption has been producing spectacular nighttime displays of incandescent magma fragments or lava blocks that are shot up out of the crater and land on the flanks of the volcano. It has produced volcanic ash columns that have risen from a few thousand meters to as much as 10,000 meters above the summit crater. The ash falls have been carried off toward the west from the volcano and have seriously damaged crops and some villages west of the volcano.

Success at Pinatubo

NOVA: What do you consider VDAP's greatest triumph?

Miller: Without a doubt the 1991 eruption of Mt. Pinatubo in the Philippines, which was the third-largest eruption of the 20th century. When unrest began, VDAP's little team of eight people, augmented by other scientists from the U.S. Geological Survey and using funding and support from the U.S. Office of Foreign Disaster Assistance, were invited to join scientists from the Philippine Institute of Volcanology and Seismology. We worked together to install VDAP monitoring equipment to build a volcano observatory on very short notice. Data from the instruments were telemetered back to a location on Clark Air Base, which is very close to Pinatubo.

“About 70,000 people lived in villages that were totally destroyed a few days after they left.”

The scientific team did an incredible job of interpreting the data coming in and of understanding what was about to happen. They completed a preliminary but extremely accurate volcanic hazards assessment and a hazards zonation map, which indicated that Clark Air Base and many tens of thousands of people living around Mt. Pinatubo were directly at risk from an eruption, which looked like it was about to begin. They communicated that information to U.S. military officials, the Philippine government, and mayors of nearby towns.

In the end, Philippine officials evacuated about a million people from around the volcano. About 70,000 of those people lived in villages that were totally destroyed a few days after they left. So literally tens of thousands of lives were saved, not to mention hundreds of millions of dollars worth of military equipment, which was moved out of harm's way.

NOVA: When an eruption occurs, do you custom design a team and the equipment you'll send?

Miller: Absolutely. When unrest begins at a volcano, we always wait for an official invitation, which usually comes through the State Department or the U.S. Agency for International Development. We think about the volcano and what kind of eruptions have occurred there in the past. Then we select a team of scientists with the kind of expertise that is requested and required, and we take the kinds of equipment with us that will help out.

Usually we train local scientists to operate and maintain the equipment, which we invariably give to the host country. It's a nice feeling to help out another country with something that has the chance of saving lives, and then to offer them the equipment, no strings attached. We'll say, "We're going to head home in a couple of weeks, and we're going to leave this stuff here. We'll help you with spare parts and come down when you need help." They'll often look at us as if to say, "Well, what do you want from us?" And we can honestly reply, "Nothing. This is your equipment now, and we hope that it serves you well."

Tools by the trunkload

NOVA: What equipment do you bring along?

Miller: We have three complete volcano observatories sitting on the shelf ready to go right now—two for international use and one for domestic use. They include three primary tools that we use to forecast volcanic eruptions. First, a telemetered volcanic seismic monitoring network to detect the earthquakes that often precede eruptions. Second, tiltmeters and electronic distance measuring equipment to monitor bulging or deformation that results from magma pushing up against the solid rock of the volcano. Finally, devices to measure sulfur dioxide and carbon dioxide. These two diagnostic gases associated with magma are fairly easy to detect. When the flux of these gases at the surface increases with time, we become concerned about magma rising close to the surface and about the increased likelihood of an eruption. (For more details on techniques and equipment, see Can We Predict Eruptions?.)

NOVA: How much equipment do you bring?

Miller: Lots. In September 1994, for example, we received an urgent request from the Rabaul Volcano Observatory in Papua New Guinea. An eruption was under way, and it had already destroyed most of their monitoring network, including their tiltmeters and seismometers. Within a week, we had sent a three-person team with 38 trunks of equipment.

The equipment is modular, and so each of these trunks weighed less than 70 pounds, which is the maximum amount that you can ship as excess baggage. But each had a solar panel, batteries, and either a tiltmeter or a seismometer as well as all of the cabling and radios. All 38 trunks were flown on commercial airliners to Port Moresby, Papua New Guinea, where they were put on a military C-130 and flown into a little airstrip near Rabaul. From there they were carried by truck to the Rabaul Volcano Observatory, then flown out by helicopter in modules to the field sites and installed.

NOVA: The VDAP team was right on the volcano during the eruption?

Miller: Yes. We use great caution once an eruption appears to be imminent or, worse yet, has begun. But you need to get the seismic instruments within a few miles of the volcano, and tiltmeters need to be installed on the volcano's flanks. Sometimes that's closer than you want to be. At Pinatubo, when it became clear that a large eruption was imminent, the team finally decided not to go back to the mountain to fix or replace damaged equipment. It was simply too dangerous to go close, even with a helicopter.

NOVA: How soon after the team left did it erupt?

Miller: Four days. Actually, the big eruption occurred then. There were dangerous eruptions occurring everyday, even when that decision was made. So we try to do things in as safe a way as possible. But it's always difficult to anticipate what a volcano will do. Each is different; each has a unique plumbing system. (For more on the successful evacuation around Rabaul, see Planning for Disaster.)

The future of forecasting

NOVA: Eruption prediction is an inexact science. How soon might it approach an exact one?

Miller: It's very frustrating that, even with equipment installed and the most experienced team members that we can assemble, it's extremely difficult to accurately forecast exactly what the volcano is going to do, when it's going to do it, and how big an eruption there will be. Part of the frustration is that scientists don't make decisions about land use, or how to respond to the unrest, or whether or not to evacuate. That's the reponsibility of civil defense and elected officials. But these are life-and-death decisions, and they have huge political and economic consequences.

“I and my colleagues were always pretty close to the helicopter, ready to start it up and leave at a moment's notice.”

If there's a failed eruption, or a so-called "false alarm," everybody's angry, money is lost, and both scientists and public officials lose credibility. By the same token, if scientists don't understand what's about to happen, or public officials don't believe what the scientists think is about to happen, and people are not evacuated, and an eruption occurs and people are killed, then everyone is even angrier. We do the very best we can to provide good, accurate information to public officials. But we're never in a position where we can say we're confident that an eruption will occur within "x" number of days and be of a certain size and destroy a certain area.

When will forecasting get better? It's improving year by year. Every time we work on a volcano crisis, we learn more about how to interpret the subtle and sometimes very sophisticated signals that volcanoes give as magma moves around. There are a whole suite of different kinds of earthquakes, for instance: volcano tectonic earthquakes, long-period earthquakes, volcanic tremor. It's a very, very complicated business. However, compared to earthquake predictions, we're extremely lucky; no one has any ability to forecast earthquakes.

NOVA: I've read that you don't think of yourselves as "cowboys," despite the risks you're taking.

Miller: We don't consider it to be very dangerous. We try to be sensible; we have families, and we want to do this over the long term. Since we work for the U.S. government, our policy has always been that we will never ask any of our scientists to do anything that they're uncomfortable with. We discuss as a team what our objectives are, what kind of equipment we'd like to get installed, and what kind of observations we'd like to make, and then scientists decide whether they're willing to do it or not and on what terms.

NOVA: Ever had any close calls yourself?

Miller: I'm a pretty cautious person. I worked at Mt. St. Helens for years before the 1980 eruption, for seven or eight months during that year, and off and on ever since. But when the volcano was erupting, and when it was restless between eruptions, I was pretty darn careful. I spent very, very little time up in the crater, because it was quite a dangerous place because of rock falls off the crater rim and explosions on the dome. When I worked all over the blast zone in the summer of 1980, I and my colleagues were always pretty close to the helicopter, ready to start it up and leave at a moment's notice.

Volcanoes Talking Bernard Chouet on a seismic signal that foretells eruptions. Emergency Response Team How a crack unit of volcanologists tackles volcanic unrest abroad. Anatomy of a Volcano Distinguish magma from tephra, lava from lahar, etc. Seismic Signals Discover the hidden volcanic signatures that volcanologists seek.

Interview conducted and edited by Peter Tyson, editor in chief of NOVA online. Note: This interview has been updated from an earlier version that appeared on NOVA's "Deadly Shadow of Vesuvius" Web site.