MAGMA MIXING AT KARYMSKY: PETROLOGIC CONSTRAINTS AND MODEL

1
MAGMA MIXING AT KARYMSKY PETROLOGIC CONSTRAINTS
AND MODEL
Pavel Izbekov1, John Eichelberger1 and Boris
Ivanov2
1 Alaska Volcano Observatory, Geophysical
Institute, UAF, Fairbanks 2 Institute of Volcanic
Geology and Geochemistry, Petropavlovsk-Kamchatsky
, Russia
2
Introduction

• Mixing of compositionally distinct magmas is well
  documented by the presence of enclaves and zoning
  of phenocrysts in igneous rocks
• The question is how fast can the compositionally
  distinct magmas mix?
• We present results of a detailed petrologic study
  of andesite and basalt that erupted
  simultaneously at Karymsky volcanic center
  beginning in January 1996.
• Continuous eruption of Karymsky offered an unique
  opportunity to determine compositional variations
  of its magma caused by basaltic recharge. These
  variations indicate that mixing can be
  surprisingly fast and thorough.

3
Location
4
Geological background

• Karymsky volcano and Academy Nauk caldera belong
  to a chain of volcanoes, calderas, and maars, the
  location of which is controlled by a local
  north-trending fault.
• Karymsky is a 5300-yr-old andesitic
  stratovolcano located in the center of a
  7900-yr-old caldera. During the past 500 yr, the
  volcano has been in a state of frequent, but
  intermittent eruptive activity.
• Academy Nauk caldera is centered 9 km south of
  Karymsky on the same fault system. Since its
  caldera-forming event (ca. 40,000 yr B.P.), the
  volcanic activity within the caldera was confined
  to phreatomagmatic eruptions of basalt, which
  have occurred at least twice since 5000 yr B.P.

5
1996 eruption of Karymsky and Academy Nauk
The most recent episode of volcanic activity at
Karymsky started on January 2, 1996, after 13 yr
of dormancy. It began with simultaneous eruption
of andesite from the central vent of Karymsky
volcano and basalt from a new vent, which formed
in the northern part of Academy Nauk caldera.

• The magmas erupted simultaneously.
• The erupted magmas had strongly contrasting bulk
  compositions. Academy Nauk vent produced basalt
  (52.2 wt SiO2), while Karymsky summit vent
  erupted andesite (62.4 wt SiO2).
• The eruptive vents are located along the same
  active fault.
• Significant ground deformation occurred between
  eruptive vents. Extension between eruptive vents
  occurred gradually, rather than catastrophically.

The eruption of basalt coincident with the start
of the most recent cycle of activity at Karymsky
is suggestive of eruptive triggering.
The eruption of basalt coincident with the start
of the most recent cycle of activity at Karymsky
is suggestive of eruptive triggering.
6
Questions to answer

• Was there a mixing of andesite and basalt?
• If yes, how can we explain the homogeneity of
  andesite? Was the mixing that fast?

7
Samples and analytical techniques

• Electron microprobe (major elements)
• Cameca SX-50 at University of Alaska Fairbanks
• 15 kV acceleration voltage
• 10 nA beam current
• 5 micron beam (10 microns for glass)

• LA-ICP-MS (Ba and Sr)
• Micromass Platform ICP- HEX-MS at Michigan State
  University
• Cetac LSX 200 laser ablation system equipped with
  UV laser
• 30 micron beam.

Samples of pyroclastics of Karymsky vs. time of
their eruption. Length of hori-zontal bars
corresponds to approximate time of lava flows
effusion.
8
Petrography
Modal Abundances, vol.
9
Karymsky Glass composition vs. time
Compositions of volcanic ash glass plotted
against the date of eruption. Melt of Karymsky
andesite, which erupted in February 1996, was the
most mafic. Error bars correspond to 2s in
electron microprobe analyses of ash samples.
10
Karymsky and Academy Nauk plagioclases
BSE images of Academy Nauk (a) and Karymsky (b)
plagioclases and their corresponding
compositional profiles. Note contrasting textures
and compositions.
11
Karymsky Calcic cores in plagioclases

• Approximately 25 vol. of plagioclase phenocrysts
  in Karymsky andesite contain calcic cores.
• Composition and texture of cores match those of
  plagioclases in Academy Nauk basalts.
• Widths of sodic rims are consistent with
  introduction of calcic cores at the onset of
  eruption (2.5 109 mm/s plagioclase growth rate)

12
Karymsky Xenocrysts of olivine
Photomicrograph (a) and a simplified sketch (b)
of Karymsky andesite from the lava flow, which
effusively erupted during April-August 1996. Note
olivine xenocryst attached to the calcic core of
plagioclase both likely introduced to andesite
by basaltic replenishment in January, 1996.
Composition of olivines is shown in figure C.
13
Summary

• By late February 1996 Karymsky erupted texturally
  homogeneous andesites, no mafic enclaves were
  found.
• Within two months of the onset of eruption, the
  composition of melt of Karymsky andesite, as
  recorded by glass in tephra, shifted toward a
  more mafic composition and then gradually
  returned to its original state and remained
  constant for the following 4 years.
• Andesite contains xenocrysts of basaltic origin,
  i.e calcic plagioclase and olivine, at least part
  of which was most likely introduced by basaltic
  replenishment in January 1996, at the onset of
  the eruptive cycle.

14
Viscosity constraints on mixing
The plausibility of mixing can be roughly tested
using estimates of viscosities for Academy Nauk
basalt and Karymsky andesite.

• Isobaric crystallization of basalt was modeled
  using COMAGMAT algorithm (Ariskin)
• Viscosity was calculated using Shaw (1972) model
• Volume of injected basalt was assumed negligible
  comparative to the volume of the stored andesite
  and did not affect the temperature of the
  produced hybrid significantly.

Variations of viscosity vs. temperature for
Academy Nauk basalt (2 wt. H2O) and Karymsky
andesite (1.5 wt. H2O) at 200 MPa total pressure.
Variations of viscosity vs. temperature for
Academy Nauk basalt (2 wt. H2O) and Karymsky
andesite (1.5 wt. H2O) at 200 MPa total pressure.
15
Comparative model
16
Conclusion

• Homogeneity of andesite erupted by late February
  1996 suggests to us that thorough mixing of
  injected basalt and andesite occurred in a period
  of time as short as two months.
• Perhaps the rapid and effective blending was
  facilitated by an only modest contrast in
  viscosities and temperatures between the magmas
  and by a vigorous fluid dynamic regime in
  Karymsky reservoir.
• Karymsky is a well-mixed end member case that
  reflects the short recurrence interval of
  recharges to the system. In contrast, Trident
  volcano in Alaska may represent an intermediate
  case, where both clotting and direct mixing
  occurs, and Kizimen volcano in Kamchatka a
  poorly-mixed end member case, where clotting
  along is dominant.