²³⁸U-²³⁰Th-²²⁶Ra systematics in lavas of Miyakejima volcano erupted in the last 500 years: implications for the evolution of shallow plumbing system by magma replenishment

Abstract

In order to understand the timescale of the evolution of shallow plumbing system beneath active volcanoes at subduction zones, we investigated U-series disequilibria of lavas erupted in the last 500 years (1469AD – 1983AD) at the Miyakejima volcano, Izu arc, Japan. Miyakejima lavas of this period have ²³⁸U-²³⁰Th-²²⁶Ra disequilibria enriched in ²³⁸U and ²²⁶Ra, which are due to the addition to the mantle wedge of slab-derived fluids (Yokoyama et al., 2003). These lavas show 1) a trend almost parallel to the equiline in a (²³⁰Th/²³²Th)-(²³⁸U/²³²Th) diagram, and 2) positive linear correlation in (²³⁸U/²³⁰Th)-1/Th and (²²⁶Ra/²³⁰Th) diagrams. Fractional crystallization cannot produce these trends. Instead, magma mixing between differentiated magma having lower (²³⁸U/²³⁰Th) and (²²⁶Ra/²³⁰Th)₀ ratios and higher Th content (andesitic end-member magma: AEM), with less-differentiated magma having higher (²³⁸U/²³⁰Th) and (²²⁶Ra/²³⁰Th)₀ ratios and lower Th content (basaltic end-member magma: BEM), is expected. However, it is not simple to determine the origin of individual end-components. The AEM is characterized by elevated (²³⁰Th/²³²Th) ratio (>1.40), while only old lavas erupted >7 kyr ago have such high values at Miyakejima. Though, this signature contradicts with the observation in the (²²⁶Ra/²³⁰Th)₀–1/Th diagram that the AEM should have ²²⁶Ra-²³⁰Th disequilibrium. The most plausible explanation is that, at least during the last 500 years, the AEM, probably originated from the magma of 7 kyr ago, has been evolved by intermittent replenishment of the BEM that had ²²⁶Ra-²³⁰Th disequilibrium higher than the AEM. This process can enhance the degree of ²²⁶Ra-²³⁰Th disequilibrium for the AEM, with keeping the (²³⁰Th/²³²Th) ratio higher than the BEM. Then, to explain the trend of lavas in the (²²⁶Ra/²³⁰Th)₀–1/Th diagram, (²²⁶Ra/²³⁰Th) ratio of the BEM would not be constant but decrease with the elapse of the time, suggesting that the magma chamber for the BEM is at closed system during the last 500 years. This result well matches with petrological observations that the temperature of the BEM decreases with the progress of fractional crystallization during this period.