The study was led by scientists at the University of New Mexico with co-authors at the 91̽��, Rice University and Cornell University. All are part of an ambitious effort to use remote sensing to better understand the hidden passageways beneath one of the country’s most dangerous active volcanoes.

The , published Nov. 1 in Nature Communications, analyzes compressional waves traveling through the crust and reflecting off the mantle below the volcano. Results show that on one side the mantle is largely , a rare, moisture-absorbing, dark-green mineral that can look like a snake’s skin. But the mantle below the eastern half of the mountain is mostly , a common mineral that allows water — thought to play a key role in volcanic eruptions — to percolate up and into the overlying crust.

The paper presents the latest results from a July 2014 experiment that conducted a . The National Science Foundation project, nicknamed for imaging Magma Under St Helens, set off seismic waves to see how they travel under the mountain and generate a map of the volcano’s plumbing. The new paper is part of a larger collaboration that also involves researchers from the U.S. Geological Survey, Oregon State University and the Swiss Federal Institute of Technology in Zurich.

During that experiment, researchers from the University of New Mexico placed 900 autonomous seismographs within 15 kilometers (9.3 miles) of the crater, increasing the density of instruments right around the volcano. All sensors were deployed along the road and trail system at Mount St. Helens with an average spacing of 250 meters.

The iMUSH experiment set off 23 active-source explosions, with energy similar to small 2.0 magnitude earthquakes, over a two-week period. The resulting dataset provides high-resolution seismic imaging of deep crustal structure beneath this active arc volcano.

“We show that Mount St. Helens sits atop a sharp lateral boundary in Moho reflectivity,” said corresponding author , a postdoctoral researcher at the University of New Mexico. The lack of reflections to the west can be explained if there is a relatively cold wedge of serpentinite to the west of the mountain, because the compressional wave speed of serpentine is not very different than that of the overriding crust.

In this context, cold is less than about 700 degrees C, or 1,300 degrees F. Below that temperature the serpentinite binds the water into a crystal structure. Above that temperature, however, serpentinite is not stable, and water can percolate up through the hot mantle unimpeded and into the overlying crust.

“The melt that supplies Mount St. Helens is probably formed to the east, in the mantle wedge below Mount Adams, and then moves west through the magmatic system somehow,” Hansen concluded.

91̽��co-authors are and , both professors in the Department of Earth & Space Sciences. Other co-authors are Brandon Schmandt at the University of New Mexico, Alan Levander at Rice University, Eric Kiser at the University of Arizona and Geoff Abers at Cornell University.

“This is a nice result because it shows a very sharp boundary between where you have reflectivity and where you don’t, and that boundary between strong and weak reflectivity is pretty much directly beneath Mount St. Helens,” Creager said. “The density of data lets us see that this boundary between where there is reflectivity and where there isn’t is very sharp. Presumably what it’s telling us is the temperature of the mantle.”

Water is locked in various minerals inside the subducting oceanic plate. As the slab goes down, the temperature and pressure increase and the water is squeezed out of the crystals. The water then migrates up into the mantle of the overriding continental plate, where it reacts with olivine to become serpentinite to the west of Mount St. Helens, or olivine to the east. The temperature is key, Creager said, because that indicates where water in the descending ocean plate could be mixing with the rock to lower the melting temperature and form volcano-creating magma.

“An important question is: where is the water, and where isn’t it?” Creager said.

“This adds to a variety of other experiments that suggest that where it’s cold, this water is basically getting soaked into the mantle and turning olivine into serpentine and not going anywhere, so it can’t get up into the crust to form volcanoes,” he said. “When you get up into where there is olivine, the temperature is hotter, serpentine isn’t stable, so the water can play its role in the volcanic process.”

An iMUSH published in the spring that was led by Rice University suggested that most of the eruptive products came from one or more chambers at depths of 3 to 12 kilometers (about 2 to 7 miles), and that those chambers may be connected.

The full iMUSH experiment includes four different lines of analysis. The UW-led team deployed 70 three-directional seismometers to detect the tiny earthquakes that happen about twice a day, either due to movement of tectonic plates or motion of magma within the volcano, over a two-year period. 91̽��researchers retrieved the instruments in August and are now analyzing their data. They hope to get a higher-resolution image of the deeper sections.

Overall, the goal is to create a more complete picture of the magma system beneath Mount St. Helens both to better understand and predict volcanic activity, and also to gauge the severity of the event when an eruption is imminent.

###

For more information, contact Hansen at stevenhansen@unm.edu and Creager at 206-685-2803 or kcc@uw.edu.

Adapted from a University of New Mexico .

NSF grants: 1445937, 1520875 and 1459047

The ambitious project, a joint undertaking by Earth scientists at Rice University, the 91̽��, the University of Texas at El Paso and other institutions, requires placing more than 3,500 active seismological sensors and 23 seismic charges around the volcano during the next few days.

“Mount St. Helens and other volcanoes in the Cascade Range threaten urban centers from Vancouver to Portland, and we’d like to better understand their inner workings in order to better predict when they may erupt and how severe those eruptions are likely to be,” said , a Rice professor of Earth science and lead scientist for the experiment.

Levander said the instruments will measure seismic waves generated by the detonation of charges in 23 boreholes that are each about 80 feet deep. Most of the detonations are scheduled in the evening on July 22-23, but Levander said area residents are unlikely to hear or feel them because of the depth of the boreholes and because the detonations will produce vibrations that approximate a magnitude 2 earthquake, which typically cannot be felt.

“Our shots will provide enough seismic energy to develop a clear picture of the mountain’s inner workings, but in most cases not enough to be felt and certainly no more than low-level seismic activity that occurs in the area on a weekly basis,” Levander said.

Levander said the detonations will take place in areas where the ground has already been disturbed, such as clear-cuts, quarries, gravel pits and garbage dumps.

Dozens of mostly student volunteers are expected to arrive Friday for two days of training about how to set up the 3,500 active seismic sensors that will gather the bulk of the experimental data.

“These sensors are basically a computer in a can with a small battery,” Levander said. “They’re about the size of a water bottle, but because of their limited power supply, we only have about two days to deploy the whole lot.”

He said the volunteers will also pick up all the active sensors — and the data they’ve collected — within a couple of days of the explosions.

A few more detonations will occur on the evening of July 30. A rearrangement of part of the sensor network will accompany those tests.

The coming tests follow two years of detailed planning and are part of a four-year project called , which could bring improvements in volcanic monitoring and advance warning systems at Mount St. Helens and other volcanoes.

The work area for the tests extends from Mount Rainier on the north to the Columbia River on the south, and from Interstate 5 on the west to Mount Adams on the east. An advance team of researchers has been in the area for weeks installing 70 . Levander said these instruments, which take several hours to set up and will be left in place for two years, are more sophisticated and sensitive than the active sensors.

“The active-source monitoring will provide very high-resolution images at a relatively shallow depth, while the passive experiment data will be at a lower resolution but will be at a much greater depth,” said , a 91̽�� professor of Earth and space sciences who is leading the passive monitoring.

Having both sets of monitors recording data from the active-source detonations will help scientists have a much clearer idea of how the deeper, harder-to-see structure compares with better-defined shallow structure, he said.

A third technique, magnetotelluric monitoring, which produces data based on fluctuations in Earth’s electromagnetic field, will also be used to image the subterranean structures.

Mount St. Helens was chosen for the study because it has been the most active volcano in the Cascade Range, erupting twice in the last 35 years, and is readily accessible for the researchers and their equipment, Levander said.

The National Science Foundation funded the research. Levander is principal investigator on the active source component of the project. He said the team hopes to publish its findings in 2015. Other institutions involved include Oregon State University, the Lamont-Doherty Earth Observatory, Eidgenössische Technische Hochschule of Zurich and the United States Geological Survey.

###

This story was adapted from a from Rice University and the 91̽��, prepared by Jade Boyd of Rice.