[S-VC34] 火山・火成活動および長期予測

[SVC34-06] Post-emplacement cooling of andesitic-dacitic lava flows and implications for ⁴⁰Ar/³⁹Ar dating at arc volcanoes

*Chris Conway¹、Andrew Calvert²、Graham Leonard³、Seiko Yamasaki¹、Osamu Ishizuka¹、Yumiko Harigane¹ (1.Geological Survey of Japan, AIST、2.U.S. Geological Survey、3.GNS Science)

キーワード：Ar/Ar dating, stratovolcanoes, andesite, lava flow

⁴⁰Ar/³⁹Ar ages for lava flows contribute greatly to organizing the geological maps, eruptive histories, and petrogenetic trends that underpin hazard assessments at active volcanoes. These age data are collected by analyzing the groundmass material of rock samples; therefore, it is important to understand how the cooling and crystallization of melt affects the distribution of K and Ar in lavas during their effusion and emplacement. We have undertaken a petrological and geochemical study of late Pleistocene to Holocene andesitic-dacitic lavas from Ruapehu volcano in Aotearoa New Zealand, for which ⁴⁰Ar/³⁹Ar ages were previously determined. Studied samples from the exposed interiors of lava flows yielded ages with relative 2σ precisions of 2–32% (e.g., 42.8 ± 1.0 ka; 8.8 ± 2.8 ka). Groundmass microlite phases in these lavas are plagioclase, orthopyroxene, and magnetite. New microanalytical data show that pre-existing melt was quenched to form rhyolitic glass at lava margins; however, melt fractionation during slow cooling in lava interiors formed subhedral sanidine (~11 wt.% K₂O) and patchy tridymite. The contribution of sanidine to the K/Ca budget of samples peaked through the middle stages of step-heating experiments, which resulted in consistent gas release spectra and calculation of high-precision plateau ages. Crystallization of groundmass sanidine in the analyzed samples was facilitated by slow cooling of felsic melt in the interior of relatively thick lava flows that formed due to their impoundment by glaciers on the flanks of Ruapehu volcano. In contrast to the apparent advantages of analyzing non-glassy rocks, samples with glassy groundmass material may be vulnerable to the issues of alteration, excess argon, and nuclear recoil, which lead to erroneous age calculations. For these reasons, collecting samples from the interior parts of lavas that have been exposed by erosion is recommended over taking glassy samples from lava flow margins.