Portal de Periódicos da CAPES

Miocene intra‐arc bending at an arc–arc collision zone, central Japan: Reply

1999; Wiley; Volume: 8; Issue: 1 Linguagem: Inglês

10.1046/j.1440-1738.1999.t01-1-00227.x

1440-1738

Masaki Takahashi, Kazuo Saito,

earthquake and tectonic studies

We appreciate the comment by J. R. Ali and S. J. Moss on our paper (Takahashi & Saito 1997), but we disagree with their criticism. The most important discrepancy between the Ali and Moss model and our reconstruction is in the location of the Subducted Izu–Bonin Ridge (SIBR) during the middle Miocene. We presented a middle Miocene reconstruction in which this then-active island arc was located in almost the same place as the present Izu–Bonin arc, but Ali and Moss insisted that the Izu–Bonin Ridge, and therefore the SIBR, was 300–400 km to the west. Their model is based on a reconstruction of the Philippine Sea Plate motion in which the Eurasia–Philippine Sea Plate Euler poles of 25–5 Ma was located at 15°N 160°E, and the amount of rotation was 34° clockwise (Hall et al. 1995a, 1995b). This model of Miocene Philippine Sea Plate motion was based on limited geophysical and geological data and has nonnegligible uncertainty. We prefer our interpretation that the active volcanic arc has been located near the present Izu–Bonin arc since the middle middle Miocene. Figure 1 shows some late Neogene sedimentary sequences around the South Fossa Magna. The Miura and Boso Peninsulas are located east of the Izu Peninsula, northern tip of the active volcanic Izu–Bonin arc. The Kakegawa area has been located on the forearc basin of the Honshu mainland. The Tomioka, Iwadono and Karasuyama areas have been situated between the volcanic arc of the Honshu and the outer arc (non-volcanic basement high). Thick marine sediment has been deposited in these basins since the middle or late Miocene. Some Late Cenozoic marine sequences around the South Fossa Magna, central Japan. The most remarkable feature of the Late Neogene sedimentary sequences in these basins is the frequent intercalated basaltic scoria beds that occur only in the Miura and Boso sequences. In contrast, the tuff layers within the marine sequence in other basins are all felsic. We studied the detailed geology and stratigraphy in the Tomioka (Oishi & Takahashi 1990), Iwadono and Karasuyama areas (Takahashi & Oda 1997), and recognized many tuff beds, which are all felsic. These felsic volcaniclastics were supplied by the volcanic arc of Honshu, in almost the same location as the present volcanic front, where felsic volcanism dominated during the Mio-Pliocene. We found more than 200 sets of tuff layers from these sequences, but only one fine-grained thin scoria bed (Og-13) was recognized in the Karasuyama area. In the Kakegawa area, more than 4000-m marine sediments have accumulated since the latest middle Miocene. Only nine key tuff layers have been found, which are all felsic and thin. It is quite reasonable that the small number of the tuff layers is attributed to their location (more than 100 km south of the volcanic arc of Honshu), and the westerlies transported volcanic ashes eastward. In contrast, thick marine sedimentary sequences in both the Miura and Boso Peninsulas contain numerous scoria beds (Fig. 2). Most of these scoria beds are a few to several tens of centimeters thick, and are composed of granule-size basaltic fragments. Based on careful observation of the morphology of the lapilli, scanning electron microscope (SEM) studies of the glass shards, and the presence of accretionary lapilli, Soh et al. (1989, 1991) concluded that these lapilli beds were the products of mafic phreatomagmatic, Surtseyan, eruptions. The presence of the pebble-size volcanic bomb within the hemipelagic sediments in the Miura Peninsula indicates that the source volcano was near the basin (see Fig. 7(a) of Soh et al. 1991). Stratigraphic positions of the scoria bed on the detailed stratigraphic column of the Kinone and lower Amatsu Formations in the Boso Peninsula. Photograph shows the outcrop of the scoria beds (dark color) within the hemipelagic siltstone (light color) of the lowest Amatsu Formation. FAD, first appearance datum; LAD, last appearance datum; RID, rapid increase datum; RDD, rapid decrease datum; RD, rapid decrease horizon; FO, first occurrence horizon; LO, last occurrence horizon. (1) Cande and Kent (1995); (2) Berggren et al. (1995); (3) Barron and Gladenkov (1995); (4) Takahashi and Danhara (1997); (5) Honda (1981); (6) Watanabe and Takahashi (1997); (7) Takahashi unpubl. data. Similarly, we can identify numerous basaltic fall deposits in the Boso sequence (Nakajima et al. 1981). The bulk composition of the lapilli are characterized by low-alkali tholeiite series basalt (Saito et al. 1992), which is similar to those in the Miura Peninsula. These petrochemical characteristics are similar to those of the modern Mihara-yama Volcano at Izu-Oshima Island, located at the volcanic front of the Izu–Bonin arc. The grain size of the scoria beds as well as sedimentary structures imply that the source vent was located not very far from the present Izu Peninsula. The absence of the scoria beds in the Kakegawa sequence indicates that the basaltic volcanoes might be located east of the Kakegawa area. The stratigraphic position of the scoria beds should constrain the timing of the volcanic eruptions. Using the geochronological and microbiostratigraphic data, we estimate that the oldest scoria bed (Am-1 in the Amatsu Formation) is ~ 13 million years old (Fig. 2). The scoria beds are intercalated from the lowest Amatsu Formation to the uppermost Anno Formation (~ 3 Ma) continuously. This fact implies that the basaltic activity was initiated at 13 Ma, or active volcanoes had migrated to near the Boso Peninsula at 13 Ma, and its activities continued until 3 Ma. This geological evidence is therefore an important constraint for the location of the Izu-Bonin arc since the middle middle Miocene. The presence of the scoria beds within the Amatsu, Kiyosumi, and Anno Formations contradicts Ali and Moss’s model. If their 12.5-Ma reconstruction was correct, it would need other volcanic activities between the Kakegawa area and Miura Peninsula, immediately offshore from the South Fossa Magna. Their model would also indicate that the Pacific Plate was subducted under the Philippine Sea Plate at a location so far west of the present Izu–Bonin arc so that central Japan was facing the westward-migrating Pacific Plate. The stratigraphic range of the scoria beds indicates that basaltic eruptions had continued since 13 Ma until recently near the Boso forearc basin, probably almost the same locality as the northern tip of the present Izu–Bonin arc (e.g. present Izu Peninsula or Izu-Oshima island). As the SIBR came close to the South Fossa Magna at 7.5 Ma as mentioned by Ali and Moss, expected basaltic activities should have been supplying both the Miura and Boso sedimentary basins with scoria between 13 Ma and at least 7.5 Ma. This more than 5-million-year-long basaltic activity on the moving Pacific Plate requires two guesses as shown in Fig. 3(a,b). In the model shown in Fig. 3(b), a hot spot is the source of the 13–7.5-Ma basaltic products. Figure 3(a) shows another model, in which a long volcanic chain parallel to the Pacific Plate motion (almost westward: Engebretson et al. 1985) supplied the basaltic products to the Miura and Boso forearc basins. Both models are rather far-fetched. Expected basaltic volcanic activities deduced from the scoria fall deposits within the marine sequence in the Miura and Boso Peninsulas. A and B indicate two far-fetched explanations for the volcanic activity between the Kakegawa area and the Miura Peninsula, based on the assumption that the subducted Izu–Bonin Ridge (SIBR) and Philippine Sea–Pacific Plate boundary were located significantly west of the present location of the Izu-Bonin arc during 13–7.5 Ma, as claimed by Ali and Moss. In addition, eastward migration of the SIBR volcanic arc along the convergent plate margin during the Mio-Pliocene would be expected to leave some signature in the geological record of the accretionary prism of the Shimanto Belt and overlying Miocene forearc basin sediments such as progressive structural indentation of the margin, but no such geological evidence has been documented. In fact, structural and petrochemical studies of the late Oligocene–early Miocene rocks of the Shimanto accretionary prism at Cape Muroto, Shikoku island, indicate that the Shikoku Basin fossil spreading ridge on the Philippine Sea Plate has had little east–west motion since 15 Ma (Hibbard & Karig 1990a, 1990b). Their conclusion is concordant with our model. Consequently, considering (i) the frequent intercalation of the scoria beds within the hemipelagic sediments of the Mio-Pliocene Miura and Boso forearc basins; (ii) the absence of the scoria beds within the late Miocene–Pliocene sequence in the Kakegawa Area; (iii) petrochemical and SEM studies, as well as sedimentological observations, on these volcaniclastics; (iv) the stratigraphic range of the basaltic fall deposits within the Boso forearc sediments; and (v) other geological, petrochemical and geophysical investigations along the Southwest Japan margin, we are compelled to conclude that the basaltic volcanism continued from 13 Ma till recently immediately offshore the Miura and Boso Peninsulas. These lines of evidence have led to the interpretation that the basaltic volcanism was associated with the proto-Izu–Bonin arc which subducted or accreted at the South Fossa Magna since the middle middle Miocene. The tectonic significance of the volcaniclastics within the Miura Group was first discussed by Matsuda (1962). Many geological and geophysical investigations have been carried out to find the missing volcanic belt around the South Fossa Magna (Matsuda 1980; Ogawa et al. 1985, 1992; Ito 1986; Soh & Taniguchi 1988; Arima et al. 1991; Taniguchi et al. 1991a, 1991b). The exact timing of the collision of the Kushigatayama, Misaka and Tanzawa Blocks is still controversial (Matsuda 1989). Recently, Amano and coworkers have done detailed investigation in the South Fossa Magna (Amano et al. 1997; Martin & Amano 1997), which will make the collision history clearer. We are grateful to Drs Shuichi Tokuhashi, Terumasa Nakajima, and Mahito Watanabe of the Geological Survey of Japan for discussions on the geology and stratigraphy of the Boso Peninsula. We also thank Dr Wonn Soh of the Kyushu University, Prof. Yujiro Ogawa of the University of Tsukuba, and Prof. Tokihiko Matsuda of the Seinan-Gakuin University for comments and ideas on the origin of the basaltic volcaniclastics within the Miura and Boso sedimentary sequences.