Variation of downdip limit of the seismogenic zone near the Japanese islands: implications for the serpentinization mechanism of the forearc mantle wedge
Variation of downdip limit of the seismogenic zone near the Japanese islands: implications for the serpentinization mechanism of the forearc mantle wedge
复制标题
DOI：
10.1016/j.epsl.2004.12.027
复制
发表时间：
2005-03
影响因子：
5.3
通讯作者：
T. Seno
中科院分区：
文献类型：
--
作者：
T. Seno
俯冲带（DLT）板间逆冲型地震的下倾界限，要么取决于板块界面地壳物质在350-450 °C的脆-韧性转变，要么取决于上板块弧前莫霍面。后者是由于俯冲板块上方弧前地幔楔的可能蛇纹石化作用，使冲断带以稳定的方式滑动。我研究这一标准在日本附近的俯冲带，太平洋和菲律宾海板块俯冲。我发现，在大多数N。旧太平洋板块俯冲的本州岛，关东和南岛。九州岛是菲律宾海板块伊豆-小笠原海脊和九州-帕劳海脊俯冲的地方，DLT位于60-70 km的深度，比上板块的莫霍面深。这表明该判据不成立，地幔楔不可能在这些地区被蛇纹化。与此相反，在北日本岩手近海地区，在本州岛和小笠原，旧的太平洋板块俯冲，DLT位于20-30公里的深度周围的莫霍面的上板块，这表明地幔楔是蛇纹在这些地区。在日本西南部，年轻的四国盆地正在俯冲，DLT的深度约为300公里，这与莫霍面相吻合，但也与板块界面处的350 °C温度相吻合。虽然地震层析成像表明地幔楔是蛇纹化的，但DLT并不能帮助检验这一标准。假设地幔楔的水化作用是由脱水俯冲地壳释放的水通过水力压裂作用进行的，我认为地幔楔的应力状态将决定地幔楔的蛇纹石化程度。在北东向逆断层型应力状态下，本州、关东和S.在九州，水力压裂是不受欢迎的，与反复板间地震有关的逆冲带的裂缝为水向上倾方向流动提供了管道，导致地幔楔没有蛇纹化。地幔楔没有蛇纹石化反过来又引起板间地震。与此相反，在岩手和小笠原地区的正断层型应力体制和日本西南部的走滑断层型应力体制中，水力压裂是有利的，导致地幔楔的蛇纹化。逆冲带的摩擦性质受稳定滑动控制，导致莫霍面以下没有板间地震。因此，地幔楔的应力状态决定了它是否发生了蛇纹化，进而决定了DLT的深度。
It has been proposed that the downdip limit of interplate thrust-type earthquakes in subduction zones (DLT) is determined either by the brittle–ductile transition around ∼350–450 °C of the crustal material at the plate interface or the Moho of the forearc of the upper plate. The latter is due to possible serpentinization of the forearc mantle wedge above the subducting plate, which makes the thrust zone slide in a stable manner. I examine this criterion in subduction zones near Japan, where the Pacific and Philippine Sea plates are subducting. I show that in most of N. Honshu where the old Pacific plate is subducting, and in Kanto and S. Kyushu where the Izu-Bonin and Kyushu-Palau ridges of the Philippine Sea plate are subducting, the DLT is located at a depth of 60–70 km, which is deeper than the Moho of the upper plate. This shows that the criterion does not hold and the mantle wedge is not likely to be serpentinized in these regions. In contrast, in the region off Iwate in N. Honshu and in Bonin, where the old Pacific plate is subducting, the DLT is located at a depth of 20–30 km around the Moho of the upper plate, suggesting that the mantle wedge is serpentinized in these regions. In SW. Japan where the young Shikoku Basin is subducting, the DLT is around ∼30 km in depth, which coincides with the Moho, but also with the 350 °C temperature at the plate interface. Although seismic tomography indicates that the mantle wedge is serpentinized, the DLT here does not help to examine the criterion. Assuming that hydration of the mantle wedge is conducted through hydro-fracturing by water released from the dehydrating subducted crust, I propose that the stress state of the wedge would determine the extent of serpentinization of the mantle wedge. In the reverse-fault-type stress regime of N. Honshu, Kanto, and S. Kyushu, hydro-fracturing is not favored, and fractures in the thrust zone associated with repeated interplate earthquakes provide conduits for water to flow in the updip direction, resulting in no serpentinization of the mantle wedge. No serpentinization of the mantle wedge in turn causes interplate earthquakes. In contrast, in the normal-fault-type stress regime of the region off Iwate and Bonin, and the strike-slip fault-type stress regime of SW. Japan, hydro-fracturing is favored, resulting in serpentinization of the mantle wedge. The frictional property at the thrust zone is governed by stable sliding, resulting in no interplate earthquakes deeper than the Moho. Thus, the stress regime of the mantle wedge would determine bifurcation whether it is serpentinized or not, and then the depth of DLT.