09:00 〜 09:15
[MGI32-06] 古伊豆-小笠原-マリアナ弧の背弧におけるメルト包有物の化学組成の時間変化（40-48 Ma）
キーワード：古伊豆-小笠原-マリアナ弧、メルト包有物、奄美三角海盆
Purpose
International Ocean Discovery Program Expedition 351 drilled Site U1438 (27°23’N, 134°19’E, water depth of 4700 m) located in the Amami Sankaku Basin on the rear-arc side of the proto-Izu-Bonin-Mariana (IBM) arc, and recovered volcaniclastic sedimentary cores (Arculus et al., 2015; Fig. 1). Previously we discussed the temporal evolution of the proto-IBM arc at Site U1438 based on melt inclusions from Unit III (30-40 Ma) (Brandl et al., 2017; Hamada et al., 2020). We performed statistical analysis of geochemical data of Unit III melt inclusions (n=237) and identified three major magma types: (1) enriched medium-K magmas, which form a tholeiitic trend (30-38 Ma); (2) enriched medium-K magmas, which form a calc-alkaline trend (30-39 Ma); (3) depleted low-K magmas, which form a calc-alkaline trend (35-40 Ma). However, we could not specify the timing of emergence of each magma type (38 Ma, 39 Ma and 40 Ma) because geochemical data of melt inclusions around 40 Ma were sparse.
Here we report results of new major and trace element analysis of melt inclusions from the bottom of Unit III and older age Unit IV (40-48 Ma; Fig. 1). The purpose of this study is to discuss temporal changes in the proto-IBM arc volcanism soon after subduction initiation at 52 Ma (Ishizuka et al., 2018).
Samples and methods
We analyzed major element composition of 22 melt inclusions collected from the bottom of Unit III and Unit IV (40-48 Ma) at Site U1438 using EPMA at the University of Leeds. Trace element composition of selected 19 larger melt inclusions were analyzed by LA-ICP-MS at JAMSTEC. Host minerals are either clinopyroxene (68≦Mg#≦86) or plagioclase.
Results
The oldest melt inclusions collected from Unit IV (48 Ma) are of the high-K series (Fig. 2). Melt inclusions collected from the bottom of Unit III (42 Ma) are of the medium-K series (Fig. 2). Trace element patterns of the new high-K and medium-K melt inclusions overlap (Figs. 3a and 3b). They are also silimar to previously reported medium-K melt incusions (30-40 Ma). Low-K melt inclusions (Fig. 2), which have flatter patterns of heavy rare earth elements (Fig. 3c), were only found in volcaniclastics in the age range of 35-40 Ma (Hamada et al., 2020). They have not discovered in older and younger age intervals.
Discussion
Considering that trace element patterns of high-K melt inclusions (48 Ma) and medium-K melt inclusions (30-42 Ma) overlap, we argue that their primary melts were derived by partial melting of an identical mantle source but differ in slab input and overall degree of partial melting. The degree of partial melting in the mantle wedge would increase over time through addition of slab-derived fluids. Slight differences in oxygen fugacity could result in formation of tholeiitic and calc-alkaline trends of the medium-K magmas (Hamada et al., 2020). Trace element patterns of low-K melt inclusions (35-40 Ma), especially the flatter patterns of heavy rare earth elements, suggests that the low-K primary melts were generated by larger degrees of partial melting of an ultra-depleted mantle source via further addition of slab-derived fluids. Such ultra-depleted mantle source can be the residue of extraction of the proto-arc basalts at the rear-arc side (Hamada et al., 2020). Occurrence of low-K magmas (40 Ma) would be delayed until dehydration of the subducting Pacific slab increased and reached a steady state. The volcanism of low-K magmas would disappear at 35 Ma probably due to mantle convection which replaced the shallow ultra-depleted mantle source with enriched mantle source (Hamada et al., 2020).
The described rear-arc volcanic evolution differ from that reported from the one reported from the volcanic front of the proto-IBM arc, where arc volcanism was initiated by eruption of arc tholeiitic lavas at 48 Ma, followed by eruption of boninites (e.g., Ishizuka et al., 2006; Umino et al., 2015). Moreover, boninitic melt inclusions (MgO>8 wt.%) were seldom found in Site U1438 sedimentary cores. We infer that temporal changes in arc volcanism were different between frontal-arc and rear-arc sides.
References
Arculus et al. (2015) Proceedings of the International Ocean Discovery Program 351. doi:10.14379/iodp.proc.351.103.2015.
Brandl et al. (2017) Earth and Planetary Science Letters 461, 73-84.
Gill (1981) Orogenic andesites and plate tectonics. Berlin: Springer, 390 pp.
Ishizuka et al. (2006) Earth and Planetary Science Letters 250, 385-401.
Ishizuka et al. (2018) Earth and Planetary Science Letters 481, 80-90.
Hamada et al. (2020) Journal of Petrology 61: egaa022.
Umino et al. (2015) Geology 43, 151-154.
International Ocean Discovery Program Expedition 351 drilled Site U1438 (27°23’N, 134°19’E, water depth of 4700 m) located in the Amami Sankaku Basin on the rear-arc side of the proto-Izu-Bonin-Mariana (IBM) arc, and recovered volcaniclastic sedimentary cores (Arculus et al., 2015; Fig. 1). Previously we discussed the temporal evolution of the proto-IBM arc at Site U1438 based on melt inclusions from Unit III (30-40 Ma) (Brandl et al., 2017; Hamada et al., 2020). We performed statistical analysis of geochemical data of Unit III melt inclusions (n=237) and identified three major magma types: (1) enriched medium-K magmas, which form a tholeiitic trend (30-38 Ma); (2) enriched medium-K magmas, which form a calc-alkaline trend (30-39 Ma); (3) depleted low-K magmas, which form a calc-alkaline trend (35-40 Ma). However, we could not specify the timing of emergence of each magma type (38 Ma, 39 Ma and 40 Ma) because geochemical data of melt inclusions around 40 Ma were sparse.
Here we report results of new major and trace element analysis of melt inclusions from the bottom of Unit III and older age Unit IV (40-48 Ma; Fig. 1). The purpose of this study is to discuss temporal changes in the proto-IBM arc volcanism soon after subduction initiation at 52 Ma (Ishizuka et al., 2018).
Samples and methods
We analyzed major element composition of 22 melt inclusions collected from the bottom of Unit III and Unit IV (40-48 Ma) at Site U1438 using EPMA at the University of Leeds. Trace element composition of selected 19 larger melt inclusions were analyzed by LA-ICP-MS at JAMSTEC. Host minerals are either clinopyroxene (68≦Mg#≦86) or plagioclase.
Results
The oldest melt inclusions collected from Unit IV (48 Ma) are of the high-K series (Fig. 2). Melt inclusions collected from the bottom of Unit III (42 Ma) are of the medium-K series (Fig. 2). Trace element patterns of the new high-K and medium-K melt inclusions overlap (Figs. 3a and 3b). They are also silimar to previously reported medium-K melt incusions (30-40 Ma). Low-K melt inclusions (Fig. 2), which have flatter patterns of heavy rare earth elements (Fig. 3c), were only found in volcaniclastics in the age range of 35-40 Ma (Hamada et al., 2020). They have not discovered in older and younger age intervals.
Discussion
Considering that trace element patterns of high-K melt inclusions (48 Ma) and medium-K melt inclusions (30-42 Ma) overlap, we argue that their primary melts were derived by partial melting of an identical mantle source but differ in slab input and overall degree of partial melting. The degree of partial melting in the mantle wedge would increase over time through addition of slab-derived fluids. Slight differences in oxygen fugacity could result in formation of tholeiitic and calc-alkaline trends of the medium-K magmas (Hamada et al., 2020). Trace element patterns of low-K melt inclusions (35-40 Ma), especially the flatter patterns of heavy rare earth elements, suggests that the low-K primary melts were generated by larger degrees of partial melting of an ultra-depleted mantle source via further addition of slab-derived fluids. Such ultra-depleted mantle source can be the residue of extraction of the proto-arc basalts at the rear-arc side (Hamada et al., 2020). Occurrence of low-K magmas (40 Ma) would be delayed until dehydration of the subducting Pacific slab increased and reached a steady state. The volcanism of low-K magmas would disappear at 35 Ma probably due to mantle convection which replaced the shallow ultra-depleted mantle source with enriched mantle source (Hamada et al., 2020).
The described rear-arc volcanic evolution differ from that reported from the one reported from the volcanic front of the proto-IBM arc, where arc volcanism was initiated by eruption of arc tholeiitic lavas at 48 Ma, followed by eruption of boninites (e.g., Ishizuka et al., 2006; Umino et al., 2015). Moreover, boninitic melt inclusions (MgO>8 wt.%) were seldom found in Site U1438 sedimentary cores. We infer that temporal changes in arc volcanism were different between frontal-arc and rear-arc sides.
References
Arculus et al. (2015) Proceedings of the International Ocean Discovery Program 351. doi:10.14379/iodp.proc.351.103.2015.
Brandl et al. (2017) Earth and Planetary Science Letters 461, 73-84.
Gill (1981) Orogenic andesites and plate tectonics. Berlin: Springer, 390 pp.
Ishizuka et al. (2006) Earth and Planetary Science Letters 250, 385-401.
Ishizuka et al. (2018) Earth and Planetary Science Letters 481, 80-90.
Hamada et al. (2020) Journal of Petrology 61: egaa022.
Umino et al. (2015) Geology 43, 151-154.