Krakatau

Products of the 1883 eruption, Krakatau Volcano, Indonesia

Sampled in the post-assembly field trip, 2000 IAVCEI

We joined in this field trip led by Dr. Igan Stawidjaja.

The deposits of the 1883 eruption characteristically contain obsidian block and vesicular pumices. In some pyroclasts, vesicular and dense obsidian parts co-exist within a grain. We suggested that the obsidian is degassed and collapsed magma and not a quenched one. Here some results of water content and textural analysis are presented to show this.

Updated 1st on July 27, 2000

Large pumice

Large pumice as large as 30 cm in dameter.

<- Bread crust surface of the pumice Photo by the courtesy of Dr. Takarada (GSJ)

This floats in the sea water. Floating pumice with me (my head). Photo by the courtesy of Dr. Takarada (GSJ)

The pumice consists of three parts

(1) outer part: dense with laminar texture sub-parallel to surface morphology (black part). The very surface part (ca. 3-4mm) is sometimes very vesicular with high number density fine bubbles (white to gray).

(2) middle part: very vesicular with fine bubbles and pipy structure as long as 10 cm.

(3) inner part: very vesicular with spherical large bubbles connecting with each other (lowest part of this sample).

The thin section of the outer part of the pumice (obsidian part). Direction is similar to the sample photo above. The laminar texture and low microlite content are characteristic. Phenocrysts consist of Plagioclase, pyroxenes, and Fe-Ti oxide minerals. The initial temperature is estimated to be 961 ±14 ℃ by Spencer and Lindsley (1981). Results by pyroxene thermometer (Lindsley, 1983) are consistent (or a little higher T) to this value . No hydrous minerals are observed.

Initial Water content estimatedby petrological hygrometer

Initial water content estimated from Housh and Luhr (1991) is ca. 3 wt.% which is consistent to water content of melt inclusion measured by FTIR(Mandeville et al., 1996)

Water content measurement by Karl-Fisher method

Water content of these parts are measured by Karl-Fisher method.

It is higher for the vesicular part (ca. 1.0 wt.%), whereas it is systematically lower for the dense part (ca. 0.7 wt.%). Both indicate extensive exsolution, but not to equilibrium at 1 bar and 900-1000℃.

Interpretation

(1) Outer part seems to consist of degassed melt with collapsed bubbles. The vesiculr outermost part of ca. 3-4 mm thick would be the quenched part by the contact with sea water (which might have been much thicker, but thinned by erosion).

(2) Middle part seems to be pathway of gas by mechanism of filtration flow such that proposed by Alidivirov and Dingwell (2000).

(3) Inner part would be expanded later than cooling of the outer part. The expansion and exsolution seem to have occurred extensively to have overpressure. This resulted in pipe formation and degassing by filtration flow progressed. Due to the two properties (solid outermost part and expanding overpressured inner part), bubbles in outer part would have been pushed from both sides, and have collapsed to be obsidian.

These deformation and collapse of bubbles were possible probably because of high temperature and very low crystallinity which facilitate degassing of magma.

Implication

This structure strongly indicates strong temperature gradient within the large pumices. These pumices unlikely to have transferred heat to water and air efficently enough. Thus, the amount of these large pumice would change the column height by lowering the effective heat release to atmosphere. And this may be a cause that this eruption is characterized fundamentally by pyroclastic flows.

Reference

Alidivirov and DIngwell (2000) Bull. Volcanol., 100, 413-421.

Housh and Luhr (1991) Am. Mineral., 76,477-492.

Lindsley (1983) Am. Mineral., 68,477-493.

Mandeville et al. (1996) J.Volcanol. Geotherm. Res., 74, 243-274.

Spencer and Lindsley (1981) Am. Mineral., 66,1189-1201.

HOME

Large pumice	Large pumice as large as 30 cm in dameter. <- Bread crust surface of the pumice Photo by the courtesy of Dr. Takarada (GSJ)
	This floats in the sea water. Floating pumice with me (my head). Photo by the courtesy of Dr. Takarada (GSJ)
	The pumice consists of three parts (1) outer part: dense with laminar texture sub-parallel to surface morphology (black part). The very surface part (ca. 3-4mm) is sometimes very vesicular with high number density fine bubbles (white to gray). (2) middle part: very vesicular with fine bubbles and pipy structure as long as 10 cm. (3) inner part: very vesicular with spherical large bubbles connecting with each other (lowest part of this sample).
	The thin section of the outer part of the pumice (obsidian part). Direction is similar to the sample photo above. The laminar texture and low microlite content are characteristic. Phenocrysts consist of Plagioclase, pyroxenes, and Fe-Ti oxide minerals. The initial temperature is estimated to be 961 ±14 ℃ by Spencer and Lindsley (1981). Results by pyroxene thermometer (Lindsley, 1983) are consistent (or a little higher T) to this value . No hydrous minerals are observed.
Initial Water content estimatedby petrological hygrometer	Initial water content estimated from Housh and Luhr (1991) is ca. 3 wt.% which is consistent to water content of melt inclusion measured by FTIR(Mandeville et al., 1996)
Water content measurement by Karl-Fisher method	Water content of these parts are measured by Karl-Fisher method. It is higher for the vesicular part (ca. 1.0 wt.%), whereas it is systematically lower for the dense part (ca. 0.7 wt.%). Both indicate extensive exsolution, but not to equilibrium at 1 bar and 900-1000℃.
*Interpretation*	(1) Outer part seems to consist of degassed melt with collapsed bubbles. The vesiculr outermost part of ca. 3-4 mm thick would be the quenched part by the contact with sea water (which might have been much thicker, but thinned by erosion). (2) Middle part seems to be pathway of gas by mechanism of filtration flow such that proposed by Alidivirov and Dingwell (2000). (3) Inner part would be expanded later than cooling of the outer part. The expansion and exsolution seem to have occurred extensively to have overpressure. This resulted in pipe formation and degassing by filtration flow progressed. Due to the two properties (solid outermost part and expanding overpressured inner part), bubbles in outer part would have been pushed from both sides, and have collapsed to be obsidian. These deformation and collapse of bubbles were possible probably because of high temperature and very low crystallinity which facilitate degassing of magma.
*Implication*	This structure strongly indicates strong temperature gradient within the large pumices. These pumices unlikely to have transferred heat to water and air efficently enough. Thus, the amount of these large pumice would change the column height by lowering the effective heat release to atmosphere. And this may be a cause that this eruption is characterized fundamentally by pyroclastic flows.