When Mount St. Helens in southwestern Washington state began to demonstrate signs of volcanic unrest in the fall of 2004, scientists watched and waited to see if an eruption as devastating as the one that occurred there in 1980 would result. During that event, after two months of increased seismic activity, a magnitude 5.1 earthquake occurred, spawning an explosive eruption. Within 10 minutes, debris from an avalanche spread more than 15 miles west of the volcano. The eruption also felled trees and killed most wildlife in a 200-square-mile area, ejected debris and magma in a column several miles high, and distributed ash as far as 900 miles away.

Ultimately, Mount St. Helens did erupt again in 2004. Fortunately, that eruption was an effusive eruption, meaning it was dominated by the outpouring of lava onto the ground, instead of an explosive eruption, characterized by the violent fragmentation of molten rock. Although scientists cannot yet predict whether a particular eruption from the same volcano will be explosive or effusive, studying the geologic fluid mechanics of the volcano's magma (the term for molten rock located beneath the Earth's surface) increases our understanding of the problem. In the University's Department of Geology and Geophysics, newly arrived assistant professor Martin Saar is studying the physics of fluid flow in such a geologic context. Saar, the George and Orpha Gibson Chair in Hydrogeology and Geofluids, uses field work, laboratory experiments, and computer simulations to examine geologic fluid mechanics in a wide range of models. The research he performs has implications for geothermal energy exploration, volcanic and seismic hazard assessment, water management, construction of reservoirs, and remediation of environmental contamination.

Two kinds of volcanoes and two kinds of lava

In one aspect of his current research, Saar studies the characteristics of magma and lava (the term for molten rock when it erupts onto the Earth's surface). These two types of molten rock typically contain crystals and bubbles of suspended gasses. Saar's studies of magma rheology, the characteristics of deformation when stresses are applied to a material, inform scientists about the dynamics of volcanic eruptions--if they will be explosive or effusive--and about lava flow velocity and distance, known as emplacement. For example, the permeability of the liquid phase of magma for the gas bubbles suspended within it affects the ability of the magma to release its gasses. "If magma can lose bubbles while ascending within the volcano, gas can escape, decreasing the magma pressure similarly to a shaken soda bottle that has been opened slowly. This [degassing phenomenon] may be one of the most important factors for determining if an eruption will be explosive or effusive," stated Saar. When the bubble and crystal content of magma change, its fluid characteristics (such as viscosity and permeability) change. As a result, its likely association with an explosive eruption or its lava's emplacement characteristics will also change.

Saar is also interested in the transition between two well-characterized types of lava, commonly known by their Hawaiian names, pahoehoe (which has a smooth or ropy surface) and a'a (which has a rough, blocky surface). Understanding lava transition events and the emplacement characteristics of these two types of lava is important in regions like Hawaii, where learning more about the distance that lava can be expected to travel is critical for protecting life and property. Saar has collected (now solidified) lava samples along a pahoehoe to a'a transition from the 1974 Kilauea volcanic eruption in Hawaii and studied the relationship between their fluid mechanical properties and emplacement type. Related to this work, Saar and his students have developed a computer simulation to reconstruct, long after the partially molten rock has solidified, the fluid dynamics that did occur within magmas and lavas.

Overcoming obstacles

Much of Saar's work focuses on large-scale, complex systems that are inaccessible to standard measurements. "When dealing with the actual Earth itself rather than with simplified lab experiments, it is difficult to reduce the number of variables possibly influencing any given process of interest," noted Saar. "As a result, we don't know all the parameters when working on such large scales of time and space, but we can try to determine the parameters that appear to be most critical." In situations where all the variables cannot be controlled, referred to as underconstrained models, geologists routinely make multiple direct and indirect observations to improve, or further constrain, their results.

For Saar, the aim of using such multiple measures is to simplify a conceptual model of a real-world process to the point where an analytic mathematical solution can be found. However, sometimes conceptual models cannot be reduced to find analytic solutions. For these situations, Saar is developing and using large-scale computer models through a multi-processor Linux computer cluster. The goal of his computer analyses is to minimize the error between what the model predicts and what the data show. Having more data (from multiple measures) allows a model to be further constrained, providing a better-fitting model. "It is impossible to find one unique model for such underconstrained systems, but we can determine a best estimate," said Saar.

Just as such large-scale models present challenges, studying the rheology of magma and lava poses its own unique set of obstacles. Magma and lava are difficult to study both in the natural setting and in the laboratory due to the high melting temperature of rocks (magma exists between 650 and 1200 degrees Celsius). To simulate lava in the laboratory, Saar typically uses analog fluids. Starting with corn syrup, he mixes other materials in, such as bubbles and particles (poppyseeds). In doing so, he creates a multi-component system that exhibits complex rheological behavior that may be similar to lava and magma. To test some of the findings derived from experiments with analog fluids and computer simulations, Saar currently has a proposal pending to study rheology and permeability measurements on actual (re)molten rock samples.

Laying the groundwork

As part of his new post, Saar is forming a new hydrogeology and geofluids research group within the Department of Geology and Geophysics that is funded in part from an endowment from George and Orpha Gibson. "It's important to conduct fundamental research at the University," noted Saar. "It's also important to put numbers on things to look at phenomena in and on the Earth, and possibly other planets, in a quantitative way." The applied value of Saar's research, however, should not be overlooked. "We hope that what comes out of our labs will be picked up by people eventually and applied in the fields of groundwater resource management, geothermal energy exploration, or volcanic hazard assessment."

WRITTEN BY LINDA RAAB