USGS
Hawaiian Volcano 
Observatory

Kilauea

Mauna Loa

Earthquakes

Other Volcanoes

Volcanic Hazards

March 11, 1999

A weekly feature provided by scientists at the Hawaiian Volcano Observatory.


Magmatic path from Kilauea's summit to vent

Kilauea's summit magma chamber is connected to the rift-zone vents like a water tank linked by hose to an irrigation system. The hose comprises the dikes that lie at 3-4 km depth along the trace of the rift zones. This relation gives us a good idea of what's going to happen at the end of the hose by watching for changes at the water tank. As the faucet is turned to supply more water, the hose trembles. We monitor the status of Kilauea's summit reservoir and of Pu`u `O`o by watching seismicity and ground deformation, in other words, by watching for the trembling.

How quickly do downrift sites respond to changes at the summit for a long-established magmatic system like the one that leads to Pu`u `O`o? As little as two hours are required for a magmatic pulse to be transmitted the 19 km from summit to Pu`u `O`o. What travels from summit to rift vent during that time, however, isn't a unit of magma, but, instead, a pressure pulse. The process is analogous to suddenly pressurizing an already-full garden hose. The water that squirts from the business end of the hose was already there, waiting to be forcefully ejected. And the water suddenly added to the upstream supply won't spew out for some time (seconds in a water hose, weeks or perhaps months in the volcano's hose.)

Surprisingly, the increased volume of magma that surges from the vent following each large summit tilt change is hotter than the magma that issues during the steady eruption preceding the surge. The slight difference, between 2 and 6?C, is determined by analyzing the magnesium content of the glass, which decreases when temperature drops and magnesium-bearing crystals of olivine are precipitated from it.

One might ask, why should there be any temperature difference at the eruption site? Why doesn't the pressure pulse just force out more of the same stuff that was being erupted before the surge began. Two explanations come to mind, both of them requiring the storage of magma beneath or uprift of Pu`u `O`o. Magma cools slightly as it resides in the chamber under Pu`u `O`o. During steady-state eruptions, this process is relatively undisturbed; incoming magma deep in the chamber forces slightly cooled magma out from the chamber's upper reaches. During a surge, the pressure pulse disrupts the system and forces some of the deeper, hotter magma to erupt abruptly through the cooler, capping magma.

In the other explanation, the pressure pulse opens an alternative subterranean pathway. The magma forced into this pathway completely bypasses the magma chamber enroute to the surface.

Regardless of which near-surface route it takes, the magma that erupts then enters the shallow tube system to flow as lava to the coast. Unlike the pressurized dikes and fissures, the tube typically is only about one-third full. The pressure pulse is transformed to a volumetric surge of lava flowing in the open tube. If sufficiently voluminous, this surge will overfill the tube and gush out from skylights as surface lava flows. The lava that stays in the tube will get from Pu`u `O`o to the coast within three hours.

Eruption Update

Lava continued to erupt from Pu`u `O`o and flow through a network of tubes from the vent to the sea. No surface flows from breakouts of the tube system were observed on the coastal flats, and lava is entering the ocean near Kamokuna. The public is reminded that the ocean entry areas are extremely hazardous, with explosions accompanying frequent collapses of the new land. A large collapse occurred on Monday morning, March 8. At least 8.2 hectares (20 acres) of the shoreline bench slipped beneath the waves.

No felt earthquakes were reported during the week ending on March 11.

HomeVolcano WatchProductsPhoto GalleryPress Releases
How Hawaiian Volcanoes Work

The URL of this page is http://hvo.wr.usgs.gov/volcanowatch/archive/1999/99_03_11.html
Contact: hvowebmaster@usgs.gov
Updated: 18 Mar 1999