Is it crazy to drill holes into live volcanoes? It is certainly a challenge, and costly, but scientific drilling of active volcanoes has been a steadily rewarding quest. We can learn about some important practical problems; for instance, what makes volcanoes explosive and how cities near volcanoes can be managed during eruptions.
Ordinary geologic fieldwork tells us a lot about volcanoes where they have been eroded, exposing their inner workings. But the picture is static. Learning more requires finding volcanoes that have had well-described eruptions and drilling into their fresh interiors. Rock samples and chemical tests give us a much better picture of the telltale details that increase useful, even practical understanding of eruptions.
Drilling the Right Kind of Volcano
Volcanoes are not as safe as most places, but with care drillers can cope with their dangers.
It is easy to imagine the worst: might molten lava blow out a drill hole, maybe even trigger an eruption? Well-grounded physics says not. Even the largest realistic hole would rapidly plug up with solidified lava, and the size of the hole would be less than a pinprick in relation to the mountain itself. In 2005 commercial geothermal drillers struck molten lava in Hawaii, and the magma only came up about 8 meters into a 20-centimeter hole before the drilling fluid froze it.
The greater danger is simply that of doing work on a live volcano. We have good information about the early signs of the eruption cycle; the key is to pick a volcano that is between eruptions. We also know that after an eruption ends, the lava vents solidify in short order. So the lava itself is a danger that can be avoided. A candidate volcano for drilling also needs to be well-behaved in other ways, without threatening landslides or venting large amounts of gas and steam.
Early Volcano Drilling Projects
The first serious attempts at volcanic drilling took place in Hawaii starting in 1960, just a year after spectacular eruptions at Kilauea Iki left behind a lake of lava. Core samples of freshly solidified lava taught us a great deal about how magma evolves as it rises and crystallizes. (The data from that project is used today to teach students in this online lesson.) The work was informed by decades of experience in drilling for geothermal power, and the scientists had a good idea of the conditions they would meet.
Later drilling at Inyo Domes, in eastern California, allowed us to study volcanic plumbing. A prime objective of the 1984 project was to learn how the lavas and volcanic gases behaved during the dome's last eruption 600 years ago. The drillers, as expected, found that the rock was almost entirely cold. The conduit that fed the eruption was a channel about 30 meters thick. This work was conducted not just for curiosity's sake but to aid in geothermal energy projects.
Breakthrough at Unzen
The volcano Unzen, in western Japan near the city of Shimabara, erupted over a four-year period starting in 1991. During that time it killed 44 people (including some volcanologists), displaced tens of thousands of residents, and destroyed some $2 billion in property. Destructive as it was, Unzen's eruption was not highly explosive. Volcano scientists seized the opportunity to drill into Unzen, and less than nine years after the lava stopped flowing a rig was pushing a drill string toward the volcano's throat.
The plan involved boring a curved hole into the volcano's flank, starting vertically but turning almost horizontal within a few hundred meters. As with routine geothermal holes, the drilling mud was manipulated to cool the surrounding rock and pressurize it so as to avoid any explosive hazard. It was thought that temperatures would be cool until very near the eruption conduit, which might be as hot as 600°C.
Drilling was slowed by highly porous materials in Unzen's flanks, but between February 2003 and July 2004 the hole reached and penetrated the 1991-1995 eruption conduit after nearly 2000 meters of drilling. Core samples were taken and instruments placed in the hole to monitor conditions. Nowhere was the mountain hotter than 200°C, a testimony to the hydrothermal circulation—the cooling role of groundwater. The conduit region, some 300 meters across, was found to consist mostly of volcanic breccia and several dikes of solid dacite lava between 3 and 30 meters in thickness.
Results were presented informally that year and continue to be published as the researchers combine their results into a scientific model of the volcano's evolution, chemistry and dynamics. Drilling was absolutely essential to tie surface observations of the eruption to the structure of the volcano.
The hunt for geothermal energy has prompted lots of volcano drilling, and when Puna Geothermal Venture broke into Hawaiian magma in 2005 it gave science a way to conduct the "ongoing interrogation of an active, docile magma," as volcanologist Bruce Marsh put it. There scientists have a chance of watching how basalt evolves into other types of lava. (See this BBC story for more detail.)
The International Continental Scientific Drilling Program (ICDP) is forging ahead after the success of the Unzen project; the next target in its volcano drilling projects is Mutnovsky, a volcano in eastern Russia that already supplies geothermal power to the city of Petropavlovsk. Here the 10-year plan is to start with already-planned geothermal drilling, then start a hole in an even hotter part of the mountain before attempting to penetrate Mutnovsky's active conduit.
ICDP plans are also proceeding at the Campi Flegrei caldera, a collapsed volcanic center near Naples, Italy within sight of Vesuvius. And on the Indian Ocean island of Réunion, geothermal drilling into the highly active volcano Piton de la Fournaise will have a scientific component.
Volcano drilling is coming of age.