[S-VC54_1PO1] Volcanic and igneous activities, and these long-term forecasting

[SVC54-P03] A preliminary estimation of water content of the mantle beneath Changbaishan Volcano, northeast China

*Takeshi KURITANI¹, Satoshi OKUMURA², Tetsuya YOKOYAMA³, Yoshinori ITO², Michihiko NAKAMURA², Haiquan WEI⁴ (1.Osaka City University, 2.Tohoku University, 3.Tokyo Institute of Technology, 4.China Earthquake Administration)

Keywords:mantle, water content, China

In northeast China, Cenozoic intraplate volcanic products are widely distributed. Geophysical studies have suggested that the underlying mantle transition zone is remarkably hydrous (Kelbert et al., 2009) and contains remnants of the subducted Pacific slab (Fukao et al., 1992); therefore, the Pacific slab stagnation and its relation to observed magmatism has received growing attention (e.g., Ohtani and Zhao, 2009; Richard and Iwamori, 2010). Beneath the Changbaishan volcanic field, a prominent low-velocity anomaly with a plume-like shape has been imaged in the upper mantle by P-wave tomography, which is suggestive of an upwelling of a mantle plume from the mantle transition zone (e.g., Zhao et al., 2009). In this study, to characterize the nature of the transition zone-derived mantle plume, the water content of the source mantle is estimated for basaltic products from the Changbaishan volcano. Basaltic scoria samples were collected from a cinder cone, located about 20 km to the northeast of Tianchi volcano. One scoria sample was used for preliminary analysis of glass inclusions in some plagioclase phenocrysts. Basaltic lavas, which occur with abundant mantle xenoliths, were also collected from the outcrop near the cinder cone to know the primitive magma composition at the volcano. The MgO contents of the scoria and the lava are 5.1 wt.% and 9.1 wt.%, respectively. Major element compositions of quenched glass inclusions in the scoria sample were analyzed using EPMA, and the water contents were estimated by the difference of the analytical total of the major element analysis from 100 wt.%. Through calibration using an in-house standard glass sample of known water content, the water contents of the glass inclusions were obtained to be 0.15-3.4 wt.%. The FT-IR analysis was also performed for one glass inclusion of the estimated water content of 0.15 wt.% by EPMA, which yields the total water content of 0.2 wt.%. Given that 3.4 wt.% represents the original water content of the melt without leakage, the H₂O/K₂O ratio of the melt of 0.90 is obtained. If we assume that the H₂O/K₂O ratio of the melt was not affected significantly by magmatic processes and the ratio is essentially constant in basaltic magmas at Changbaishan volcano, the water content of the primitive magma (2.4 wt.% in K₂O) is estimated to be 2.2 wt.%. The source mantle for the Changbaishan basalts may contain ~0.5% sediment component (Kuritani et al., 2011), and the Ce content of the source mantle is estimated to be ~1.1 ppm using the Ce content of the sediment component of 57.3 ppm (Plank and Langmuir, 1998) and that of the depleted mantle of 0.77 ppm (Salters and Stracke, 2004). If we assume that Ce and H₂O behave similarly during mantle melting (e.g., Michael, 1995), the compositions of the primitive basalt lava (Ce: 70 ppm, H₂O: 2.2 wt.%) yield the water content of the source mantle of ~350 ppm. This estimated water content is significantly higher than that of the normal depleted mantle (~120 ppm; Salters and Stracke, 2004), suggesting that the transition zone-derived mantle plume is hydrous compared with the surrounding ambient upper mantle. In this preliminary study, we have analyzed only seven glass inclusions in a single sample, and therefore, the water content of ~350 ppm may represent the minimum estimate. It is necessary to increase the number of data by EPMA and FT-IR analyses to more reliably estimate the source water content for the Changbaishan basalts.