A model for Obsidian Flow emplacement
As a magma body of rhyolite composition approaches the surface volatiles exsolve, separate, from the magma and rise to the plumes top. The decrease in the confining pressure leads to frothing of the volatile rich top section of the plume as ascension of the stratified mass occurs. This process of vesiculation leads to some number of explosive eruptions depositing tephra. After these eruptions the remaining volatile rich magma reaches the surface as highly inflated pumiceous lava. The transition from the explosive eruption to the pumiceous lava occurs once gas pressure in vesicles can no longer exceed strength of magma.
Later in the eruption following the pumiceous lava flow the event of the emplacement of obsidian occurs, this process happens with the bubble free obsidian flowing out over the earlier emplaced coarsely vesicular pumice (highly inflated pumiceous lava). The Obsidian is slightly less viscous but more dense than the coarse pumice. There remains a gradual transition from the coarse pumice to obisidian due to volatile decrease with the progression from pumice phase to obsidian phase. Essentially while obsidian is erupted vesicular material continues to be extruded.
As the obsidian rides over the coarse pumice, volatiles continue to evolve from the cooling upper flow surface of the obsidian flow. This process of the volatiles exsolving from the obsidian flow forms a finely vesicular layer that keeps the flow interior of the obsidian insulated and warm; this contact is also gradational, just as the coarse pumice to the obsidian was.
Within the volcanic vent that the obsidian flowed from the remaining magma from within cools and crystallizes and then erupts as a crystal rich rhyolite flow capping the vent mouth as a rhyolite dome.
The Big Obsidian Flow of Newberry Volcano, Central Oregon is very similar in the process of emplacement. Firstly the eruption began with explosive eruptions depositing pumice and lava blacks of 1 meter (3 feet) in diameter, after these eruptions pyroclastic flow occurred essentially taking the place of the pumiceous lava flow, however there remain deposits of course pumice within the region of Big Obsidian Flow which many indicate a short event of pumiceous flow. The final stage was the emplacement of Big Obsidian Flow that moved very slowly and according to Sherrod et al (1997) this lava flow probably moved a few meters to tens of meters a day. From observations in the field of the Big Obsidian Flow the presence of a thin visculated layer coats some of the surface of the obsidian flow. The final stage of the emplacement of a rhyolite dome is not present the vent is actually plugged by the obsidian flow indicating that the rhyolite stage was not present in the eruption or possibly the rhyolite magma was already to cool and harden to come to the surface of the flow.
This is an image of the Big Obsidian Flow of Newberry Volcano,a USGS Photograph taken in October 1987 by Willie Scott. The structure of the deposit indicates that the obsidian flowed over being erupted explosively.
Fink (1983) provided most of the information regarding the flow emplacement model of obsidian as well as the figures displayed.
Obsidian occurs as a flow, not as an explosive eruption in contrast to a vesiculated rhoylite pumice or dacite. This difference is due to the difference in composition, specifically gas content. Gases within highly viscous magmas can produce eruptive events due to the inability for the volatiles to escape easily from the magma so as they rupture they release an enormous amount of pressure producing an eruption such as the Mount. St. Helens eruption on May 18th 1980. For these explosive types of eruptions not occur in the emplacement of obsidian the gas content of the obsidian must be low. The average water content of obsidian is (0.3 wt %) where as crystalline rhyolite is <2.0 wt % water (Bakken Barbara., 1977). The low amount of water in comparison with rhyolite pumice indicates that the flow of obsidian must take place at the end stage of the explosive eruption phase of rhyolite magma (Bakken Barbara., 1977) end stage indicating after the vent of the volcano has released a large amount of gas through the explosive eruption stage. When it comes down to it the temperature ot the obsidian magma is the property controling the viscosity over any of the other properties, higher the temperature the lower the viscosity.
The viscosity of obsidian must be lower than rhyolite so it can flow; difference in eruption temperature is the greatest control over this difference. Initially the magma erupts at a temperature around 900 degrees C however this first eruption is rich in volatiles producing pumice, so this stage still has a greater viscosity than obsidian. As volatile rich pumice is released and the obsidian melt from the rhyolite is built up at a temperature around 900-700 degrees C the obsidian melt can be erupted at a low viscosity of a magnitude of 10^8 Pa s, which is an order of 3 magnitudes less than that of the original rhyolite pumice eruption, this therefore indicates that the eruption of the obsidian has a similar temperature of eruption as the initial pumice only difference is in the magma's gas content. After obsidian has been erupted what to follow is typically a rhyolite dome rich in crystals due to slow cooling of the magma; this event makes it clear that the reason for lack of crystals in the obsidian are due to being erupted at a temperature greater than the final rhyolite magma. This information comes to the conclusion that the flow of obsidian is dependent on the time the obsidian is erupted during the eruption, the temperature of the obsidian and the gas content.
Fink, J.H., (1983), Structure and emplacement of a Rhyolite Obsidian Flow: Little Glass Mountain, Medicine Lake Highland, Northern California. Geological Society of America Bulletin. 94 (362-380)
Sherrod., Mastin., Scott., Schilling. (1997). Newberry caldera, Oregon Big Obsidian Flow - Latest Eruptive Event Excerpt from: Sherrod, Mastin, Scott, and Schilling, 1997, Volcano Hazards at Newberry Volcano, Oregon. U.S. Geological Survey Open-File Report, 97(513)