Pyroclastic flows, lahars, and mixed avalanches generated during the 2006 eruption of Augustine Volcano: Chapter 10 in The 2006 eruption of Augustine Volcano, Alaska

Abstract

Each of the three phases of the 2006 eruption at Augustine Volcano had a distinctive eruptive style and flowage deposits. From January 11 to 28, the explosive phase comprised short vulcanian eruptions that punctuated dome growth and produced volcanowide pyroclastic flows and more energetic hot currents whose mobility was influenced by efficient mixing with and vaporization of snow. Initially, hot flows moved across winter snowpack, eroding it to generate snow, water, and pyroclastic slurries that formed mixed avalanches and lahars, first eastward, then northward, and finally southward, but subsequent flows produced no lahars or mixed avalanches. During a large explosive event on January 27, disruption of a lava dome terminated the explosive phase and emplaced the largest pyroclastic flow of the 2006 eruption northward toward Rocky Point. From January 28 to February 10, activity during the continuous phase comprised rapid dome growth and frequent dome-collapse pyroclastic flows and a lava flow restricted to the north sector of the volcano. Then, after three weeks of inactivity, during the effusive phase of March 3 to 16, the volcano continued to extrude the lava flow, whose steep sides collapsed infrequently to produce block-and-ash flows. The three eruptive phases were each unique not only in terms of eruptive style, but also in terms of the types and morphologies of deposits that were produced, and, in particular, of their lithologic components. Thus, during the explosive phase, low-silica andesite scoria predominated, and intermediate- and high-silica andesite were subordinate. During the continuous phase, the eruption shifted predominantly to high-silica andesite and, during the effusive phase, shifted again to dense low-silica andesite. Each rock type is present in the deposits of each eruptive phase and each flow type, and lithologic proportions are unique and consistent within the deposits that correspond to each eruptive phase. The chief factors that influenced pyroclastic currents and the characteristics of their deposits were genesis, grain size, and flow surface. Column collapse from short-lived vulcanian blasts, dome collapses, and collapses of viscous lavas on steep slopes caused the pyroclastic currents documented in this study. Column-collapse flows during the explosive phase spread widely and probably were affected by vaporization of ingested snow where they overran snowpack. Such pyroclastic currents can erode substrates formed of snow or ice through a combination of mechanical and thermal processes at the bed, thus enhancing the spread of these flows across snowpack and generating mixed avalanches and lahars. Grain-size characteristics of these initial pyroclastic currents and overburden pressures at their bases favored thermal scour of snow and coeval fluidization. These flows scoured substrate snow and generated secondary slurry flows, whereas subsequent flows did not. Some secondary flows were wetter and more laharic than others. Where secondary flows were quite watery, recognizable mixed-avalanche deposits were small or insignificant, and lahars were predominant. Where such flows contained substantial amounts of snow, mixed-avalanche deposits blanketed medial reaches of valleys and formed extensive marginal terraces and axial islands in distal reaches. Flows that contained significant amounts of snow formed cogenetic mixed avalanches that slid across surfaces protected by snowpack, whereas water-rich axial lahars scoured channels. Correlations of planimetric area ( A ) versus volume ( V ) for pyroclastic deposits with similar origins and characteristics exhibit linear trends, such that A=cV 2/3 , where c is a constant for similar groups of flows. This relationship was tested and calibrated for dome-collapse, column-collapse, and surgelike flows using area-volume data from this study and examples from Montserrat, Merapi, and Mount St. Helens. The ratio A/V 2/3 = c gives a dimensionless measure of mobility calibrated for each of these three types of flow. Surgelike flows are highly mobile, with c ≈520; column-collapse flows have c ≈150; and dome-collapse flows have c ≈35, about that of simple rock avalanches. Such calibrated mobility factors have a potential use in volcano-hazard assessments.