Rotten to the core – hydrothermal activity and volcano sector collapse

The 1980 eruption of Mt. St. Helens provided insight into a type of volcanic event that was previously unfamiliar to geologists.  Intrusion of magma into the northern flank of the volcano in the months  prior to the eruption led to bulging and consequent instability of the side of the mountain, and also to hydrothermal alteration and consequent weakening of the adjacent rocks.  A moderate earthquake (M5.1) caused the collapse of the entire side of Mt. St. Helens in what has been described as the largest landslide in history.  This debris avalanche was immediately followed by the enormous lateral blast, and then by a major plinian eruption than lasted for several hours.

Hummocks of debris avalanche material at Mt. St. Helens (author photo)

This type of event is now known as sector collapse, and examination of the deposits around other volcanoes has shown that it is quite common in volcano evolution.  The Mt. St. Helens debris avalanche was indeed huge, but it appears to have been small in comparison with some other similar events.  A prehistoric sector collapse at California’s Mt. Shasta is estimated to have been 20 times larger than the one at Mt. St. Helens.  Sea-floor deposits around the island of Hawaii show evidence of sector collapses many times larger still.  Sector collapse is potentially catastrophic.  For example 15,000 were killed by sector collapse, and an ensuing tsunami, on Japan’s Mt. Unzen in 1792.

Several different processes can contribute to or trigger sector collapse, including deformation related to magma intrusion (eg. the bulge at Mt. St. Helens), hydrothermal alteration to clay minerals which weakens the rock, percolation of water into fractured rocks, earthquakes and eruptions themselves.  Another less well understood process is the effect of heat from intruding magma.

As described by Reid (2004) intrusion of magma into the water-saturated rock of a volcano not only promotes hydrothermal alteration, but heating of the water within the rock porosity can also increase the pore-fluid pressures to the point where stability of the mountain is compromised.   Reid shows that these fluid-pressure effects can propagate through the rock much more quickly than the thermal effects.   Rock at some distance (100s of m) from the intrusion does not need to be heated significantly to experience an increase in pore-fluid pressure.

Modeled change in pore-fluid pressures in a typical composite volcano following intrusion of a 0.2 cubic km intrusion (red)
Modeled effective stresses within a composite volcano as a consequence of thermal pressurization of pore fluids.  Areas with higher stresses (orange and red) are potentially unstable.  Vectors show groundwater hydraulic gradients.

Of a number of historic sector collapse events examined by Reid, all show some evidence of hydrothermal activity prior to collapse, and although some were also characterized by alteration of the rock, and/or a triggering earthquakes, some show no indications of either of these.  The important point is that while edifice deformation (bulging) and hydrothermal alteration of rocks can be observed as warning signs of a possible sector collapse, stress changes related to pore-fluid pressures cannot be easily detected.  Furthermore, because of the rapid propagation of fluid-pressure effects, warning times may be short.  In other words, sector collapse could occur almost without warning, and in populated areas the results could be devastating.


Reid, M, 2004, Massive collapse of volcano edifices triggered by hydrothermal pressurization, Geology, V. 32, p. 373-376. (May 2004)

Steven Earle, 2004. Malaspina University-College, Geology Department, Return to Earth Science News