Scientists try to improve volcano forecasts . Lack of records and uniqueness of each volcano can make job difficult

When Colombia's Nevado del Ruiz volcano blasted itself into world headlines last week, it drove home to the world at large a point that scientists have known for some time: They may be good at explaining how volcanoes work, but they still have a lot to learn when it comes to pinpointing when a volcano may erupt. That fact poses a perpetual challenge for civil defense planners, volcanologists say. When researchers are able to predict eruptions at volcanoes such as Nevado del Ruiz with pinpoint accuracy, planners' jobs will be much easier.

``It's the difference between saying there's probably going to be a storm sometime in the next week and, `Quick, there's a hurricane coming down the street, get into your cellar,' '' says University of Chicago volcanologist Alfred T. Anderson. ``We are trying to become more certain about our predictions. That is important for civil protection efforts to be effective.''

Nevado del Ruiz is one volcano in a chain that stretches from northern Chile to Colombia, where a section of the Pacific Ocean floor plunges beneath the South American continental plate. It is also a segment of a so-called ``Ring of Fire'' that encircles the Pacific, where various plate boundaries are the sites of large numbers of earthquakes and volcanic activity.

Mt. St. Helens's eruption in 1980, a major eruption of Mexico's El Chichon volcano in 1982, the Mexican earthquakes in September, as well as last week's eruption in Colombia were all part of the same general process, where the Pacific floor slips under the American continents. In the case of volcanoes, the pressure that builds as one plate slides under another forces molten rock, or magma, from deep beneath the earth's crust up through volcanoes.

In a sense, Nevado del Ruiz's Nov. 13 eruption was no surprise to researchers who had been studying the volcano for the past year. Like Washington's Mt. St. Helens five years ago, Ruiz had been giving seismic hints since late 1984 that she was building toward a major eruption.

Yet scientists were hard pressed to tell exactly when that major eruption would be. Late last December, instruments began to pick up subterranean earthquakes from the volcano's core -- a sure hint of future eruptions. In March, a number of small explosions near the volcano's 17,655-foot summit further signaled impending violent activity. And on Sept. 11, Ruiz erupted with enough force to melt snow and ice on the mountain top and trigger a 20-mile-long mud slide.

As speculation grew as to when the actual eruption might take place, geologists from Ecuador, Costa Rica, and the United States formed a consortium to aid Colombia in establishing an emergency response plan to deal with an eruption.

Dr. Darrell Herd, deputy chief of the US Geological Survey's (USGS) Office of Earthquakes, said last Thursday that Colombian scientists were attempting ``to become prepared for the disaster'' but that the eruption came ``too soon, before all the plans had been implemented.''

It appears that skepticism among some local officials about the scientists' pronouncements thwarted some of the best-laid plans. Reports are still sketchy, but some local officials may have ignored warnings from government officials even after the eruption had taken place. According to one account, a local radio announcer relayed the warning of explosions on the mountain top, then advised residents not to worry because it was probably a false alarm.

Researchers are hobbled in their attempts to understand and predict the behavior of individual volcanoes because each one acts in its own special way. A volcano's activity is determined by a number of factors, particularly the chemical makeup of its magma, or lava, which powers volcanic explosions.

For example, Hawaii's volcanoes, which spew rather runny lava, tend to erupt in spits and spurts. Lava in other volcanoes, such as Mt. St. Helens and Nevado del Ruiz, is much thicker. Because thicker lava traps gases and allows immense pressues to build, it tends to explode violently from the confines of the mountain.

Hence, volcanologists often depend on reams of historical data from a given volcano to help them predict how that volcano might behave in the future. When the data is available the system works well. In Hawaii, the USGS runs the Hawaiian Volcano Observatory on the top of Mt. Kilauea, the world's most active volcano. The observatory has been in continuous operation since 1912. Scientists there are able to predict eruptions with near total accuracy.

In the case of Nevado del Ruiz, however, scientists didn't have the luxury of a long historical record to guide them.

The volcano's last major eruption occurred on March 12, 1595, although it experienced smaller eruptions in 1828 and 1829.

``It is extraordinarily difficult to make predictions of a volcano's behavior when it's been dormant for nearly 400 years,'' says Dr. Richard P. Hoblitt, of the USGS's volcanic hazards prediction project.

Nevado del Ruiz is also a more difficult type of volcano to study than Kilauea because it is a more dangerous volcano to study. Where Hawaiian eruptions frequently involve lava dribbling down volcano sides, eruptions in volcanoes such as Ruiz frequently involve part of the mountainside being blasted away.

Scientists analyzing the aftermath of Ruiz do have the precedent of Mt. St. Helens, however, and the physical similarity between the two is said to be strong. As with Mt. St. Helens, Ruiz took time to build up to its major eruption, gave plenty of warning that one was on the way, and wreaked most of its early havoc through mud slides.

Those similarities, and others, are guiding scientists who are trying to determine what Ruiz will do next.

``With each eruption, the picture becomes a little clearer,'' says Dr. Anderson. For example, scientists have accurately predicted 16 of the past 18 minor eruptions at Mt. St. Helens.

In addition, scientists are able to use data gleaned from instruments deployed in the aftermath of the St. Helens eruption.

After that occurrence, researchers embarked on aggressive projects to fit St. Helens and other volcanoes around the world with computer-operated instruments that constantly monitor subterranean vibrations, groundswell caused by piling magma, and gas emissions. They have also embarked on global projects to collect historical volcanic data.

Scientists have been playing close attention to 13 of Mt. St. Helens's sister volcanoes in the Cascades. Several of the mountains, including Lassen Peak, Mt. Baker, Mt. Ranier, and Mt. Hood -- have erupted in the past 200 years.

Steam and melt water has been seeping from Mt. Hood since 1975. That, scientists say, is a strong indication of volcanic activity deep inside the mountain. All are capped with ice and snow, and an eruption would be accompanied by the sorts of mud slides and floods that characterized the St. Helens and Nevado del Ruiz eruptions.