Identifying possible tsunami deposits on the Shizuoka Plain,Japan and their correlation with earthquake activity over the past 4000 years

Abstract

Three washover sand beds, ranging from 15 to 34 cm in thickness, were recorded in a lagoonal mud sequence on the Shizuoka coast of central Japan, which faces the Suruga Trough. The sand beds were composed of well-sorted and well-rounded beach sand derived by a marine inundation. The basal erosion surface, mud clasts, and presence of both inverse and normal grading suggest that the sand beds formed as a result of high-energy deposition. The sand beds are multilayered, with fine alternations of sand sheets and mud drapes, which are consistent with deposition from a long-period wave train. Radiocarbon dates obtained from the three washover deposits are around ad 1000, 3565–3486 cal. yr BP, and 4000 cal. yr BP. Vertical changes in the diatom assemblages suggest a gradually increasing inflow of seawater up to the second sand bed, followed by a rapid change to freshwater conditions just above the bed. The gradual increase of seawater inflow begins again just above the second sand bed. We conclude that this series of sequential environmental changes indicates interseismic subsidence and coseismic uplift. The study area experienced around 1 m of uplift during the ad 1854 Ansei Tokai earthquake (M_w 8.4). In contrast, the area has subsided gradually (c. 6 mm/yr) during the present interseismic period. The youngest sand bed may be correlated with the ad 1096 Eicho earthquake, which caused severe damage along the Shizuoka coast.

Keywords

coseismic uplift earthquake Holocene Shizuoka Plain Suruga and Nankai troughs tsunami deposits

Introduction

This paper aims to determine whether mega-earthquakes and their associated tsunamis, such as the 2011 Tohoku-oki event (M_w 9), have occurred in the past (and so may occur again in the future) along the Nankai and Suruga troughs, close to central and west Japan (Figure 1a and b). Since the 2011 Tohoku-oki earthquake and tsunami, developing a reliable estimate of the maximum earthquake and tsunami risk has become an important element of Japanese disaster control planning, and the Cabinet Office of the Japanese Government has developed a set of calculations of the likely tsunami height, inundation area, and damage caused by the maximum possible earthquake along the Nankai and Suruga troughs (http://www.bousai.go.jp/jishin/nankai/taisaku_wg/pdf/20130528_honbun.pdf).

Figure 1.

Location maps. (a) Map of Japan showing plate boundaries and the location of the Shizuoka area. 1: epicenter of the 2011 Tohoku-oki earthquake (M_w 9.0), 2: epicenter of the ad 1944 Tonankai earthquake (M_w 8.2), 3: epicenter of the ad 1854 Ansei Tokai earthquake (M_w 8.4), 4: epicenter of the ad 1707 Hoei earthquake (M_w 8.4). (b) Map of the Shizuoka area showing the distribution of tsunami wave heights (m) caused by the Ansei Tokai earthquake (after Watanabe, 1998), and indicating the location of the Oya lowland to the east of the Abe River. (c) Map of the study area in the Oya lowland, showing core sites and geomorphology (after Matsuda, 2006).

Japanese historical documents covering a period of more than 1300 years provide a record of tsunami genesis related to great earthquakes along the Nankai and Suruga troughs that stretches back to ad 684 Hakuho earthquake (e.g. Ando, 1975; Ishibashi, 1981; Usami, 2003). However, even this data set is not long enough to conclusively determine the likely maximum size of an earthquake, and its associated tsunami, that can be expected in this region. Paleotsunami research based on coastal geology has revealed traces of mega-tsunamis from the period prior to that covered by these historical documents, for example, at Hokkaido, north Japan (Hirakawa et al., 2000; Nanayama et al., 2003); at Sendai, northeast Japan (Minoura et al., 2001; Minoura and Nakaya, 1991; Sawai et al., 2012); and in Chile (Cisternas et al., 2005). According to this research, the earthquakes that cause mega-tsunamis are estimated to have much longer recurrence intervals than that of ‘lower intensity’ subduction zone earthquakes (c. M_w 8).

To estimate the size (inundation distance) of a paleotsunami, we must trace the tsunami deposits through the coastal lowlands from the former shoreline to its inundation limit (e.g. Nanayama et al., 2007). A serious problem facing paleotsunami researchers is artificial disturbance of the tsunami deposits, especially by agricultural activity. The coastal lowlands along the Nankai and Suruga troughs have been highly cultivated and urbanized for more than 2000 years. The Oya lowland on the southeast Shizuoka Plain is one of the few areas suitable for the field study of paleotsunami events in these regions (Figure 1). The lowland is a former lagoon and has not been heavily cultivated.

Kitamura et al. (2011) recorded the Holocene sequence preserved in the Oya lowland based on sediment cores recovered from two sites (Sites 1 and 2; Figures 1c, 2 and 3). These cores are 8 m long and consist mainly of clay that was deposited in a lagoon–delta system over the past 7300 years. Sand beds within the cores suggest the occurrence of high-energy currents, and these beds have a sharp base and are normally graded. Kitamura et al. (2011) classified these beds into Type A and Type B according to their lithological features. Type B beds are common throughout the lagoonal sequence and are characterized by a high mud content (>20%) and poorly rounded, very fine-grained sands. The Type A beds are rare (only two were recorded in the lagoonal sequence) and are characterized by a low mud content (5–15%) and well-rounded medium-grained sands.

Figure 2.

Locations of sample cores. (a) A 1887 1:25,000 scale topographic map (Kunosan district) showing the former course of the Oya River and (b) a 1:25,000 scale topographic map of the Shizuoka Tobu district (published by the Geospatial Information Authority of Japan).

Figure 3.

Stratigraphic columns reconstructed from the cores at the seven sites. See Table 1 for radiocarbon dates.

The grain shape (angular) of the Type B sands is similar to that of sands deposited in the river mouth bar of the Abe River (Figure 1b), implying that this deposit was derived from fluvial flooding. In contrast, the grain shape (subangular) of the Type A beds is similar to the foreshore and backshore sands of the Oya coast, suggesting that this deposit was most probably derived from marine flooding. Consequently, Kitamura et al. (2011) suggested a tsunami origin for the two Type A sand beds preserved within the lagoonal sequence; however, they did not obtain any definitive evidence to support this interpretation.

This paper considers the origins of the Type A and B sand beds based on five additional cores (Figures 1c, 2 and 3), and uses an interdisciplinary approach to interpret the sedimentological, geomorphological, and petrological data. Our results will help to assess the likelihood of further mega-tsunamis on the Shizuoka Plain and surrounding areas in the future.

Study area

Tectonic setting

The Shizuoka Plain is located on the Pacific coast of central Japan and faces the Suruga Trough, which marks the boundary of convergence between the Eurasian Plate and Philippine Sea Plate, together with the Nankai Trough (Figure 1a and b). This subduction zone is believed to be responsible for the region’s main earthquakes and tsunamis. Most earthquakes are initiated along one or other of two well-defined segments on the eastern and western sections of the subduction zone. Earthquakes rupturing the eastern segment are known as Tokai or Tonankai earthquakes, while those rupturing the western segment are known as Nankai earthquakes (e.g. Ando, 1975; Ishibashi, 1981). Analyses of historical documents and archaeological evidence (e.g. traces of liquefaction in ruins) suggest that great earthquakes (c. M_w 8) in this region have occurred with a recurrence interval of 90–200 years (e.g. Sangawa, 2001; Watanabe, 1998). The latest great earthquakes along the eastern Nankai Trough near the Shizuoka Plain occurred in 1854 (Ansei Tokai earthquake, M_w 8.4; Ando, 1975) and 1944 (Tonankai earthquake, M_w 8.2; Tanioka and Satake, 2001).

Geomorphological changes along the Shizuoka coast during the 1854 Tokai earthquake have been reconstructed based on a local government survey conducted 38 years after the earthquake. According to the report, an estimated 1–2 m of uplift occurred around the Shizuoka Plain (Hatori, 1976; Ishibashi, 1984). However, the absence of any emerged coastal landforms suggests the accumulation of uplift through the Holocene in this area. This discrepancy between the historical and geomorphic data is probably related to interseismic subsidence. Tidal gauge data for the past 55 years show that the Shizuoka Plain is subsiding at a rate of around 6 mm/yr (Geospatial Information Authority of Japan, 2011) during the current interseismic interval.

Geomorphological setting

The Shizuoka Plain forms a Gilbert-type delta at the mouth of the Abe River (Figure 1b), which is mainly composed of Holocene sands and gravels (Tsuchi and Takahashi, 1972). The Oya lowland, to the southeast of the Shizuoka Plain, is bounded by the Udo Hills to the east and by the Abe River fan to the west (Figure 1b and c). Beach ridges and dunes up to 10 m high protect the lowland from Suruga Bay (Figures 1c and 2). Previous research revealed that the main part of the beach ridges formed during the Holocene transgression and held an embayment or lagoon, the forerunner of the lowland (Tsuchi and Takahashi, 1972). The landward margin of the Oya lowland broadly corresponds to the shoreline at 6–7 ka. At present, a large area of the Oya lowland has an artificially raised ground level (by c. 2 m), and has a height of 7–8 m above mean sea level (a.m.s.l.). According to Kitamura et al. (2011), the near-surface geology (up to c. 8 m deep) consists of alternating beds of clay and sand, mainly deposited in a lagoonal environment. This sequence also contains the Kikai-Akahoya (K-Ah) tephra layer, erupted at around 7300 cal. yr. BP (Kitagawa et al., 1995), in its lower section (Figure 3).

The coastline in the study area is wave dominated and microtidal, with a maximum tidal range of 160 cm on spring tides (gauged at Shimizu, which is 10 km northeast of the study area). The foreshore deposits are mainly well-sorted coarse sand with some pebbles. The backshore deposits consist mainly of well-sorted coarse sand. These coastal deposits do not contain molluscan shells or microfossils (Kitamura et al., 2011), due to the high-energy regime and unstable substrates that prevent benthic taxa from inhabiting these environments. The shoreface slope is relatively steep, reaching water depths of 20 m within 3 km of the coast (Shiba et al., 1990). According to Nemoto et al. (1988), sea floor deposits are mainly composed of gravel (<5 m deep) and well-sorted sand (5–30 m deep).

Historical background of the study area

Historical evidence of tsunamis in Shizuoka Prefecture shows that the area has been affected by six major tsunami events between the ad 1096 Eicho earthquake and the 1944 Tonankai earthquake. The estimated wave height around the coast of Suruga Bay was up to 6 m for the 1854 Ansei Tokai (Figure 1b) and 1707 Hoei earthquakes, but less than 2 m for the 1944 Tonankai earthquake (Watanabe, 1998); the 1944 earthquake did not rupture the plate boundary along the Suruga Trough. However, there are no historical documents suggesting the occurrence of tsunamis on the Oya lowland (Hatori, 1977).

In addition to the tsunamis originating from the Nankai and Suruga troughs, the Shizuoka area has been affected by tsunamis that have traveled considerable distances (teletsunamis). Teletsunamis that originated near Chile in 1960 and 2010 had wave heights of 1.3 and 0.5 m, respectively (Japan Meteorological Agency, 2012; Watanabe, 1998). According to Satake et al. (1996), the estimated height of the tsunami caused by the ad 1700 earthquake in Cascadia, off the Pacific coast of North America, was 2 m along the Shizuoka coast. The tsunami height associated with the 2011 Tohoku-oki earthquake was 1.0 m in this area (Japan Meteorological Agency, 2011).

Severe storm surges have also been recorded in the study area, and according to Arakawa et al. (1961), the three storm events of ad 1611, ad 1662, and ad 1953 are the largest to have occurred here. However, these events did not cause severe damage in the Oya lowland. We expect that the Oya lowland has not experienced any marine flooding related to storm events over the past 400 years.

Methods

A total of seven conventionally drilled cores were recovered for this study (Figure 1c). All cores were continuous, and were 8 m long with a diameter of 6 cm. One extra core was drilled at Site 1 to compensate for the lack of stratigraphic data from this site in the study of Kitamura et al. (2011). Coring sites were arranged along the longitudinal axis of the Oya lowland with a horizontal interval of 50–300 m (Figure 1c). The elevation of the coring sites was determined using an automatic level. The lithology, sedimentary structures, and macrofossil content of the cores were described, and a preliminary facies identification made. Further interpretation of the depositional environments was carried out by integrating the lithological and paleoecological (microfossil) data. The cores were divided into 1-cm-thick subsamples (c. 4.5 g), and the following analytical techniques were applied to the sediments.

Criteria for identifying tsunami deposits

Deposits from river flooding and storm surges (washover) may have similar sedimentary characteristics to tsunami deposits. We first separated the marine inundation deposits from the river flood deposits using lithological features such as grain roundness and sedimentary structures.

The structure of sediment deposited by tsunamis, which have much longer wavelengths than storm waves, may also provide criteria to differentiate the tsunami deposits from storm deposits (e.g. Fujiwara and Kamataki, 2007; Naruse et al., 2010; Switzer and Jones, 2008). Tsunami deposits often consist of multilayered sand beds composed of alternating graded sand sheets and mud drapes (e.g. Fujiwara, 2008; Fujiwara et al., 2012; Naruse et al., 2010). The multilayered structure records the repeated upflow and return flow (graded sand sheets), intercalated with slack-water periods during the transition between the two flows (mud drapes). We aimed to characterize the multilayered sand beds using grain-size distribution data.

Additionally, we aimed to discriminate between tsunami deposits and storm deposits using a fossil proxy. One useful indicator in this regard is the changing sedimentary environment caused by the coastal uplift and subsidence related to the seismic cycle. As mentioned above, gradual subsidence during the interseismic period, and rapid coseismic uplift, are typical of the study area. If such an abrupt environmental change coincides with marine inundation deposits, the deposits are most likely to be tsunami deposits associated with a great subduction zone earthquake (e.g. Atwater, 1987; Nelson et al., 1996b, 2008).

Grain-size and roundness analysis

Beach deposits are continuously winnowed by wave and coastal currents, and generally consist of relatively well-sorted and well-rounded grains when compared with river sediments. Dune sand is mainly derived from the beach, and so has similar features to beach sands. Consequently, the sorting and roundness of sand grains could be used as an indicator to discriminate between riverbed and washover deposits in the Oya lowland (Kitamura et al., 2011).

The grain-size of both the core sediments and recent shoreface/beach and dune sediments from around the study area was analyzed (Figure 1c). For the core samples, we focused on the Type A sand beds reported by Kitamura et al. (2011) and the coarsest portions of the Type B sand beds. Analysis was conducted using the wet-sieving method at intervals of 1.0 phi between −3 phi (8 mm) and 4 phi (63 µm).

Grain roundness was measured in 23 samples (one sample from each sand bed) from the cores, and two samples of recent beach deposits. For each sand bed, we analyzed grain roundness in the sample with the coarsest mean grain-size. Each sample was cemented and arranged on a slide for examination under the microscope. In this paper, we focus on the mudstone fragments, because they are the major components of the samples. Roundness was calculated according to the scheme of Krumbein (1941) using more than 100 fine-sand-sized grains.

Diatom analysis

Diatom assemblages are reliable indicators of modern and ancient coastal environments, and are widely used as proxy indicators of depositional conditions, such as salinity and water depth, to support the lithological data (e.g. Kosugi, 1988). Changes in diatom assemblages in coastal sequences have been used to reconstruct the relative sea-level changes associated with crustal movements during the earthquake cycle (e.g. Atwater and Hemphill-Haley, 1997; Dawson et al., 1996; Mamo et al., 2009; Nelson et al., 1996a, 2008; Sawai et al., 2004, 2012; Shennan and Hamilton, 2006).

We analyzed a total of 32 subsamples from the clay beds at Sites 1 and 2 that were selected following a consideration of the lithofacies preserved in the cores. Preparation for microscopy followed Kosugi (1993), and diatom valves were analyzed under an oil immersion microscope. Identification of diatom species and floras was based on Kosugi (1988) and Ando (1990).

Radiocarbon dating

Radiocarbon dates (Table 1) were obtained from organic material using the accelerator mass spectrometry by Beta Analytic Inc., and they provided a chronological framework for this study. Ages were calibrated using the OxCal4.1 program (Bronk Ramsey, 2009) and the IntCal09 data set (Reimer et al., 2009).

Table 1.

¹⁴C dates from the study area obtained using the accelerator mass spectrometry technique (dating by Beta Analytic Inc.).

Sample no.	Site	Altitude (m a.m.s.l.)	Materials	δ¹³C (‰)	Conventional ¹⁴C age (yr BP)	Calibrated age (1σ) (cal. yr BP)	Calibrated age (1σ) (cal. yr BP)	Calibrated age (2σ) (cal. yr BP)	Lab number PLD
1	1	3.94	Plant materials	−25.14 ± 0.31	4410 ± 25	3093 bc (58.4%) 3011 bc	5043–4961 (58.4%)	5214–5197 (3.0%)	18,433
						2978 bc (7.4%) 2962 bc	4928–4912 (7.4%)	5051–4872 (92.4%)
						2949 bc (2.5%) 2944 bc	4899–4894 (2.5%)	5051–4872 (92.4%)
2	1	3.76	Plant materials	−29.67 ± 0.18	3280 ± 20	1608 bc (34.9%) 1571 bc	3558–3521 (34.9%)	3565–3451 (95.4%)	19,640
						1561 bc (11.2%) 1547 bc	3511–3497 (11.2%)
						1541 bc (22.1%) 1517 bc	3491–3467 (22.1%)
3	1	3.56	Plant materials	−28.93 ± 0.15	3890 ± 25	2457 bc (68.2%) 2345 bc	4407–4295 (68.2%)	4415–4247 (95.4%)	19,639
4	1	3.48	Plant materials	−29.09 ± 0.17	3815 ± 25	2290 bc (68.2%) 2205 bc	4240–4155 (68.2%)	4348–4335 (1.3%)	19,638
4	1	3.48	Plant materials	−29.09 ± 0.17	3815 ± 25	2290 bc (68.2%) 2205 bc	4240–4155 (68.2%)	4296–4145 (89.0%) 4123–4095 (5.1%)	19,638
5	1	0.66	Charcoal fragments	−29.96 ± 0.19	6435 ± 25	5468 bc (4.5%) 5463 bc	7418–7413 (4.5%)	7426–7310 (95.4%)	18,434
5	1	0.66	Charcoal fragments	−29.96 ± 0.19	6435 ± 25	5448 bc (63.7%) 5378 bc	7398–7328 (63.7%)	7426–7310 (95.4%)	18,434
6	1	0.29	Plant materials	−27.35 ± 0.16	7010 ± 25	5976 bc (28.1%) 5949 bc	7926–7899 (28.1%)	7934–7890 (33.0%), 7881–7791 (62.4%)	18,435
6	1	0.29	Plant materials	−27.35 ± 0.16	7010 ± 25	5919 bc (40.1%) 5879 bc	7869–7829 (40.1%)	7934–7890 (33.0%), 7881–7791 (62.4%)	18,435
7	1	0.21	Plant materials	−27.66 ± 0.16	6970 ± 25	5894 bc (58.9%) 5835 bc	7844–7785 (58.9%)	7924–7902 (5.5%), 7866–7721 (89.9%)	18,436
7	1	0.21	Plant materials	−27.66 ± 0.16	6970 ± 25	5826 bc (9.3%) 5811 bc	7776–7761 (9.3%)	7924–7902 (5.5%), 7866–7721 (89.9%)	18,436
8	1	−0.84	Plant materials	−28.59 ± 0.13	7015 ± 25	5979 bc (30.8%) 5946 bc	7929–7896 (30.8%)	7934–7793 (95.4%)	18,437
8	1	−0.84	Plant materials	−28.59 ± 0.13	7015 ± 25	5923 bc (37.4%) 5881 bc	7873–7831 (37.4%)	7934–7793 (95.4%)	18,437
9	2	5.26	Plant materials	−29.11 ± 0.16	5210 ± 20	4040 bc (38.0%) 4016 bc	5990–5966 (38.0%)	5995–5921 (95.4%)	18,438
9	2	5.26	Plant materials	−29.11 ± 0.16	5210 ± 20	4000 bc (30.2%) 3980 bc	5950–5930 (30.2%)	5995–5921 (95.4%)	18,438
10	2	4.76	Plant materials	−27.74 ± 0.18	2545 ± 20	791 bc (52.9%) 761 bc	2741–2711 (52.9%)	2746–2699 (61.9%)	18,439
						682 bc (15.3%) 671 bc	2632–2621 (15.3%)	2638–2616 (20.8%)
								2591–2542 (11.5%)
								2528–2518 (1.2%)
11	2	3.36	Plant materials	−27.23 ± 0.15	3650 ± 20	2116 bc (11.4%) 2099 bc	4066–4049 (11.4%)	4081–4036 (20.9%)	18,440
11	2	3.36	Plant materials	−27.23 ± 0.15	3650 ± 20	2039 bc (56.8%) 1976 bc	3989–3926 (56.8%)	4001–3898 (74.5%)	18,440
12	2	3.06	Plant materials	−26.86 ± 0.15	3670 ± 20	2128 bc (37.2%) 2089 bc	4078–4039 (37.2%)	4086–3926 (95.4%)	18,441
						2046 bc (23.3%) 2021 bc	3996–3971 (23.3%)
						1993 bc (7.7%) 1982 bc	3943–3932 (7.7%)
13	2	0.66	Plant materials	−29.76 ± 0.16	6865 ± 25	5767 bc (68.2%) 5717 bc	7717–7667 (68.2%)	7783–7780 (0.3%), 7760–7652 (91.8%)	18,442
13	2	0.66	Plant materials	−29.76 ± 0.16	6865 ± 25	5767 bc (68.2%) 5717 bc	7717–7667 (68.2%)	7645–7624 (3.3%)	18,442
14	3	3.13	Root	−28.28 ± 0.23	3350 ± 20	1682 bc (68.2%) 1618 bc	3632–3568 (68.2%)	3679–3669 (1.9%)	20,495
14	3	3.13	Root	−28.28 ± 0.23	3350 ± 20	1682 bc (68.2%) 1618 bc	3632–3568 (68.2%)	3641–3555 (82.2%), 3534–3486 (11.4%)	20,495
15	4	3.56	Wood	−26.31 ± 0.16	1095 ± 20	ad 899 (25.8%) ad 919	1051–1031 (25.8%)	1056–960 (95.4%)	19642
15	4	3.56	Wood	−26.31 ± 0.16	1095 ± 20	ad 948 (42.4%) ad 985	1002–965 (42.4%)	1056–960 (95.4%)	19642
16	4	3.48	Peach seed	−24.93 ± 0.17	955 ± 20	ad 1028 (23.9%) ad 1046ad 1092 (35.2%) ad 1121ad 1140 (9.1%) ad 1148	922–904 (23.9%), 858–829 (35.2%)	929–893 (31.1%), 874–795 (64.3%)	19,641
16	4	3.48	Peach seed	−24.93 ± 0.17	955 ± 20		810–802 (9.1%)	929–893 (31.1%), 874–795 (64.3%)	19,641
17	6	3.56	Root	−28.45 ± 0.24	2465 ± 20	750 bc (28.8%) 687 bc	2700–2637 (28.8%)	2706–2633 (31.8%)	20,496
						666 bc (11.2%) 642 bc	2616–2592 (11.2%)	2620–2436 (56.8%)
						592 bc (28.3%) 518 bc	2542–2468 (28.3%)	2413–2366 (6.8%)

a.m.s.l.: above mean sea level.

Results

The lagoon–delta system identified in the sample cores was classified into three sedimentary facies based on a combination of the lithology, sedimentary structures, grain-size, and fossil content of the deposits (Figures 3 –5): Facies 1 (shoreface/beach), Facies 2 (dune), and Facies 3 (lagoon/floodplain). Facies 3 is composed of silt and clay beds, and is intercalated with the sand and gravel beds.

Figure 4.

Fossil diatom assemblages and mud content at Site 1.

Figure 5.

Fossil diatom assemblages and mud content at Site 2.

Depositional facies

Facies 1 (shoreface/beach deposits)

This facies was found at Site 7 (−3.0 to +5.1 m a.m.s.l.), which was located on the beach ridge (Figure 1c). It is characterized by well-sorted medium sand beds with some gravels (<4 cm in diameter) in its lower part, and alternating beds of well-sorted sand and rounded pebble-size gravels in its upper part (Figure 3). The sand and gravel beds are 0.4–1.2 m thick, and generally show normal or inverse grading. The size and morphology of the gravels is similar to the gravels on the present-day beach in the study area. Facies 1 shows a general coarsening-upward trend and is overlain by Facies 2 (dune deposits). Mean values of grain roundness for the shoreface deposits (0.38–0.44) are slightly higher than those of the recent beach deposits (0.34–0.35; Figure 6). This is similar to studies elsewhere (e.g. Costa et al., 2012).

Figure 6.

Grain shape histograms for fine-sand-sized mudstone fragments from sandy sediments. Total sample size (n) and mean value ( $\bar{χ}$ ) are indicated in each case.

The well-sorted coarse sediments and stratified beds with no muddy matrix suggest that this facies was deposited under the influence of strong waves or currents, or both. The absence of marine fossils also supports this interpretation. The upward-coarsening trend within this facies indicates that the deposit represents a regressive phase. The stratigraphic position and vertical facies succession suggest that this facies represents shoreface and overlying gravelly beach deposits.

Facies 2 (dune deposits)

Facies 2, which was present only in the upper part of the Site 7 sequence (5.1–6.2 m a.m.s.l.; Figure 1c), consists of well-sorted medium sand beds showing a fining-upward trend (Figure 3). Mean values of grain roundness range between 0.45 and 0.46, and are similar to those of the shoreface deposits (Figure 6).

The sedimentary features and nature of the sampling sites, which is located at the beach ridge, clearly show that these sand beds are eolian dune deposits covering the beach deposits. The fining-upward trend in the sand beds suggests either a regressive process (i.e. an increase in the transport distance by wind) or storm deposits. The pebbles in this facies are interpreted as a washover deposit associated with storm.

Facies 3 (lagoon/floodplain deposits)

Facies 3 was found in all cores except Site 7 (Figures 3 –5 and 7). Foraminifera or volcanic clasts, except the K-Ah tephra layer, were not found in any of the sediment samples. Based on its lithological characteristics and occurrence, Facies 3 was subdivided into Subfacies 3.1, which mainly consists of clay beds, and Subfacies 3.2, which consists of gravel beds and occurs at Sites 4 and 5.

Figure 7.

Mud content at Sites 3–6.

Subfacies 3.1 is a sequence of alternating silty clay and clay beds with intercalations of sand beds up to 34 cm thick. The silty clay beds are generally massive and dark-bluish black in color, while the clay beds show fine (millimeter-scale) parallel-laminations and mark the lower to middle part of the facies (Figure 3). Diatom assemblages in Subfacies 3.1 show a clear contrast between the massive silty clay beds and the laminated clay beds (Figures 4 and 5). The massive silty clay beds are marked by the mixed occurrence of the marine species Caloneis westii and Diplonesis smithii, and freshwater taxa such as Pinnularia spp. and Synedra ulna. The laminated clay beds are dominated by the marine and brackish benthic species D. smithii (e.g. Sawai et al., 2008), and especially by the high-salinity species Nitzschia cocconeiformis (e.g. Kashima, 2008) in its lower part. The laminated clay beds contain limited occurrences of freshwater species.

Subfacies 3.2 consists of alternating pebble-size conglomerate beds with a clast-supported framework and is enclosed within Subfacies 3.1 (Figure 3). It ranges from 2.35 to 2.73 m in thickness. Each conglomerate bed is up to 1.4 m thick and shows normal grading. Subfacies 3.2 shows a fining-upward sequence as a whole. The distribution of grain roundness in the fine-grained sand of this subfacies is leptokurtic, with a mean grain roundness of 0.42 (Figure 6). At Site 4, fragments of pottery were incorporated within the gravel beds, one of which was dated to the 6th century based on its archaeological design.

Facies 3 is interpreted as a lagoon/floodplain system comprising muddy lagoonal and back-marsh deposits (Subfacies 3.1) as well as gravelly river-channel deposits (Subfacies 3.2). It forms the upper part of the Holocene sedimentary sequence. Our cores were too widely spaced to reconstruct the complete geometry of Subfacies 3.2, or the detailed lithological variations of the floodplain sediments. It is often difficult to distinguish between crevasse-splay, channel, and residual channel deposits; therefore, a distinction is made between muddy lagoon/back-marsh deposits and gravelly channel-fill deposits.

The lower-middle part of Subfacies 3.1 is marked by laminated clay beds with marine/brackish diatom assemblages. The laminated clay is surrounded by massive silty clay with a mixed assemblage of marine and freshwater diatoms. This turnover of diatom assemblages suggests a temporal change in seawater influx into the lagoon/floodplain system. Subfacies 3.2 consists of coarse and graded sediments that were deposited under the influence of strong currents, and its restricted distribution across the lagoon/floodplain system suggests that it is a channel-fill deposit.

The K-Ah tephra layer (c. 7300 cal. yr BP) was identified beneath the laminated clay beds in all cores, except at Site 7 (Figure 3). A total of 17 radiocarbon dates were obtained from Subfacies 3.1 (Figure 3; Table 1). Five samples below the K-Ah tephra layer (Samples 5–8 and 13) show conformable ages with the tephra, indicating that the lower part of Subfacies 3.1 formed during the period 8000–7300 cal. yr BP. Two radiocarbon dates (Samples 11 and 12) suggest that the laminated clay was deposited before 4000 cal. yr BP. Most of the samples from the upper part of Subfacies 3.1 (2, 4, 10 and 14–17) suggest ages younger than about 3500 cal. yr BP. Samples 1 and 3 from Site 1 and Sample 9 from Site 2 yield older ages (5214–4872, 4415–4247, and 5995–5921 cal. yr BP, respectively). These older ages were rejected because they represent reworked older deposits associated with either erosion or bioturbation.

An increased seawater influx into the study area, marked by the base of the laminated clay beds just above the K-Ah tephra layer, corresponds to the peak of the Holocene transgression around Japan (Nakada et al., 1991). From this, we can infer that Subfacies 3.1 represents a single transgression–regression cycle, and we interpret the sequence of depositional environments as estuarine during the transgressive stage (lower massive clay beds), a salty lagoon at the peak of the transgression (laminated clay beds), and lagoonal to back-marsh during the high sea-level stand and regressive stage (upper massive clay beds).

Identification of marine incursion sand beds

Roundness and grain-size data for the sand beds sampled from the cores and the modern deposits are shown in Figures 6 and 8, respectively. The sand beds can be separated into two groups based on mean grain-size and mud content. One group contains the shoreface and dune deposits of Site 7, as well as the present-day beach sand (Figure 8), and this group corresponds to Type A of Kitamura et al. (2011). The other group plots in the region of finer mean grain-size and higher mud content (Figure 8), and corresponds to the Type B sand beds of Kitamura et al. (2011). The sand bed at 3.20–3.35 m a.m.s.l. (core depth: 4.56–4.41 m) at Site 2, which was identified as possible Tsunami Deposit 2 by Kitamura et al. (2011), shows intermediate characteristics between Types A (well rounded) and B (finer mean grain-size and higher mud content). Two Type A sand beds were identified in the cores. On the basis of stratigraphic correlation, we labeled them as Sand Beds T2 and T1 in ascending order (Figure 3). Sand Bed T1 is widely preserved along a 0.8-km transect in the study area, but Sand Bed T2 was found only at Site 2. An additional Type A sand bed (Sand Bed T0), recorded at Site 4, is 30 cm thick and composed of medium sand (Figure 8).

Figure 8.

Mean grain-size plotted against mud content for Sand Beds T0, T1, and T2; other sand beds in the studied cores; and present-day beach deposits. Sample 17 is not plotted in the figure because the mean grain-size was greater than −1 phi.

The grain-size distribution and roundness of the Type A sand beds are similar to those of the ancient shoreface and recent beach deposits (Figures 6 and 8), indicating that they are probably washover deposits. The lower contact with the lagoonal deposits is mainly sharp and erosive, and the sand beds are characterized by cross-laminations in some cores (Kitamura et al., 2011). These sedimentary features suggest deposition from a high-energy flow. The upper contact with the lagoonal deposits is also relatively sharp.

The presence of multiple graded structures in the Type A sand beds is clearly shown in the results of the grain-size analysis. For example, Sand Bed T1 at Site 1 was classified into a lower and an upper sand layer, separated by a mud drape (Figures 9 and 10). The upper sand layer is generally thinner and finer than the lower layer. Sand Bed T1 at Site 6 was also separated into lower and upper sublayers by a mud drape. Inverse-normal grading is clearly observed in the lower layer of Sand Bed T1 at Site 6 (Figure 9). Based on the presence of the mud drapes (suggested by the high mud content of samples from the sand bed), we distinguished inverse and normally graded sublayers (Figure 9), and the number of distinguishable sublayers was three for Sand Bed T2 at Site 2, five for Sand Bed T1 at Site 2, and five for Sand Bed T1 at Site 3 (Figure 9).

Figure 9.

Vertical profiles of grain-size and mud content in Sand Beds T0, T1, and T2.

Figure 10.

Sedimentary structures in Sand Bed T1 at Site 1.

Radiocarbon dating

The two radiocarbon dates from just below and above Sand Bed T2 (Samples 11 and 12, respectively) indicate that the sand bed was deposited around 4000 cal. yr BP. However, as the youngest ¹⁴C age of the wood samples from Sand Bed T1 is 3565–3451 cal. yr BP (Sample 2), the actual depositional age of the bed appears to be younger than 4000 cal. yr BP. On the other hand, the oldest age obtained from roots recovered from the bed is 3679–3486 cal. yr BP (Sample 14), providing a maximum depositional age. Thus, the depositional age of Sand Bed T1 is estimated to be 3565–3486 cal. yr BP.

Two fragments of plant material were also recovered and dated from Sand Bed T0 at Site 4. Their ages are in the ranges 1056–960 cal. yr BP (Sample 15) and 929–795 cal. yr BP (Sample 16). Consequently, the depositional age of Sand Bed T0 is constrained to the period around 900–1000 cal. yr BP, if we assume that the pottery fragment, which was dated to the 6th century, was either reworked from underlying deposits or transported from an older archaeological site.

Diatom analysis

The preservation of diatoms is generally poor in sandy sediments, although silt and clay usually yield sufficient diatoms valves for paleoenvironmental reconstruction. Marine and brackish diatoms such as D. smithii and C. westii dominate the diatom assemblages in the silt and clay beds beneath Sand Bed T1. Small numbers of freshwater species were also found in the sediments beneath Sand Bed T1.

At Site 1, marine and brackish diatom species show an increase-upward trend toward Sand Bed T1, but decrease temporarily just above Sand Bed T1, before again increasing upward in the clay beds (Figure 4). The freshwater diatom S. ulna shows the opposite trend to the marine and brackish species; that is, it shows temporary increases just above the Sand Bed T1 before decreasing upward again through the silt beds.

At Site 2, the vertical changes in diatom species beneath Sand Bed T1 are similar to those at Site 1. Marine and brackish diatoms such as D. smithii and C. westii show an upward increase in abundance toward Sand Bed T1, while the freshwater diatom S. ulna shows a gradual upward decrease (Figure 5). A small number of diatom valves were found in the clay beds just above Sand Bed T1. Freshwater diatoms such as Pinnularia spp. dominate over the marine and brackish species in the upper horizon at Site 2.

Discussion

Identification of tsunami deposits

Lithological features (especially grain roundness) suggest that the Type A Sand Beds T0, T1, and T2 are washover deposits. The sedimentary features of these sand beds, the erosional basal surface, and presence of mud clasts suggest the influence of erosive currents on the muddy substrate. The inverse-normal grading in the sand beds also suggests deposition from a high-energy flow (e.g. Le Roux and Vargas, 2005; Lowe, 1982; Moore et al., 2011; Sohn, 2000). However, such sedimentary features are commonly observed in both tsunami and storm surge deposits. Here, we discuss the origins of Sand Beds T0, T1, and T2 based on their sedimentary structures and diatom assemblages.

Causes of multilayered sand beds

At some sites, Sand Beds T0, T1, and T2 had a multilayered structure; that is, fine alternations of graded sand sheets and mud drapes. These sand beds further show a fining-upward trend, with thinner and finer-grained sand sheets occurring in the upper section of the sand beds. Each couplet of graded sand sheet and mud drape represents a single waning sediment flow. The mud drapes were deposited from suspension under low-energy conditions, and the alternation of graded sand sheets and mud drapes suggests the repeated occurrence of sediment flows, each of which was separated by a sufficient slack-water period to allow the deposition of mud drapes. Multilayered structure is common in washover deposits but are not characteristic of tsunami deposits. When multiple washover deposits are amalgamated, they show a fine alternation of graded sand sheets and mud drapes similar to Sand Beds T0, T1, and T2. However, the general thinning- and fining-upward sequences of sand sheets in Sand Beds T0, T1, and T2 cannot be explained by multiple storm events because multilayered sand beds caused by multiple storm events would consist of sand sheets with random thickness and grain-size profiles, which reflect the difference in intensity and sediment source of each storm event.

The combination of (1) multilayered sand beds, (2) mud drapes separating each sand sheet, and (3) a general thinning- and fining-upward sequence can be a useful tool to the identification of tsunami deposits (Fujiwara, 2008; Fujiwara et al., 2012; Naruse et al., 2010). Tsunami wave trains have long wavelengths, of over tens of minutes, and each tsunami wave generates a strong sediment flow mainly at the coast, but slack-water stages sufficient to allow the formation of mud drapes also occur between each tsunami run-up and backwash depending on coastal configuration. After the biggest wave, smaller waves arrive at the coast in the later stages of the tsunami, which leads to the development of the general thinning- and fining-upward sequence seen in tsunami deposits. One typical waveform of tsunamis includes biggest wave in its middle stage (e.g. Watanabe, 1998).

Abrupt environmental change suggesting coseismic uplift

As described above, the sedimentary features suggest that Sand Beds T0, T1, and T2 were formed when a tsunami inundated the lagoon. Associated with the deposition of Sand Bed T1, a rapid environmental change was recorded in the sediments of the lagoon.

Diatom assemblages suggest an abrupt shift from a marine-brackish environment to a freshwater environment after the deposition of Sand Bed T1 at Site 1 (Figure 4). At Site 2, marine-brackish diatoms disappear abruptly after the deposition of Sand Bed T1 (Figure 5). At Sites 1 and 2, the diatom assemblages suggest that the influence of seawater gradually increased in the lagoon before the deposition of Sand Bed T1. Although local differences were observed in the change of depositional environment, the abundance of marine-brackish diatoms increased again above the Sand Bed T1 (Figure 4).

According to Carver and McCalpin (1996), coseismic land-level changes are reflected by abrupt replacement of indicative assemblages, while gradual changes expected from longer term interseismic motions involve a transition in species composition over centimeters or decimeters of stratigraphic section. We therefore conclude that the abrupt change in sedimentary environment, indicated by the changes in diatom assemblages in this lagoon on the western coast of the Suruga Bay, were not caused by build-up of dune or a regional (eustatic) sea-level change or storms, but instead by crustal movements associated with the earthquake cycle of the subduction zone fault. Both build-up of dune and a regional sea-level rise caused a transition in species composition over centimeters or decimeters of stratigraphic section. Sea level around Japan was essentially stable during the depositional period of Sand Bed T1 (e.g. Ota et al., 1990). If the environmental changes had been caused by storm events, environmental conditions would have recovered to their prestorm state fairly quickly.

As stated above, the 1854 Ansei Tokai earthquake caused uplift of the western coast of Suruga Bay by about 1.0 m. After the deposition of Sand Bed T1, the depositional environment changed abruptly from marine-brackish to freshwater (Figure 4). This shows that the study area rose after the deposition of Sand Bed T1, so reducing the marine influence. We suggest that this abrupt change in sedimentary environment was caused by coseismic uplift.

Interseismic subsidence

Interseismic subsidence is indicated between the deposition of Sand Beds T2 and T1 (Figure 5), and also following the deposition of Sand Bed T1 (Figure 4), by the gradual changes in diatom assemblages. A gradual increase in marine-brackish diatoms, and a gradual decrease in freshwater diatoms, suggests the increasing influence of the sea. This may be related to subsidence of the coast around the study area. As noted above, the Shizuoka area has been subsiding at a rate of around 6 mm/yr during the interseismic period following the 1854 Ansei Tokai earthquake.

Correlation of tsunami deposits along the Shizuoka coast

We conclude that Sand Bed T1 is a tsunami deposit related to the paleo-Tokai/Tonankai earthquake. Sand Beds T0 and T2 have similar sedimentary features to Sand Bed T1, although they lack the diatom data indicating coseismic uplift. These two Type A sand beds may have been deposited by tsunamis. We estimated the distribution of these tsunami deposits along the Shizuoka coast by comparing the depositional ages with previously identified tsunami deposits and historical documents.

The depositional ages of the Sand Beds T0, T1, and T2 were estimated to be around ad 1000, 3565–3486 cal. yr BP, and 4000 cal. yr BP, respectively, from ¹⁴C dating (Figure 3). The age of Sand Bed T0 corresponds well to the ad 1096 Eicho Tokai earthquake, which coincided with a destructive tsunami along the Shizuoka coast. Sand Bed T1 may be correlated to the tsunami deposit reported by Fujiwara et al. (2013) from a drowned valley in western Shizuoka Prefecture about 80 km west of the present study area, which was deposited around 3400 cal. yr BP. At present, there is no historical or geological evidence to indicate the origin of Sand Bed T2. However, the lack of evidence of the vertical crustal movement implies that Sand Bed T2 is not a tsunami deposit related to the paleo-Tokai/Tonankai earthquake.

Conclusion

We identified three possible tsunami deposits, that is, Sand Beds T0 (around ad 1000), T1 (3565–3486 cal. yr BP), and T2 (around 4000 cal. yr BP), from the Oya lowland, which lies southeast of the Shizuoka Plain and faces the Suruga Trough. These possible tsunami deposits are composed of well-sorted beach sand and intercalated silt beds that accumulated in a lagoonal environment. We used grain roundness, grain-size, sedimentary structures, diatom assemblages, and ¹⁴C dating to determine the origin and age of the tsunamis, and the vertical crustal movement related to past Tokai earthquakes.

Sand Beds T0, T1, and T2 have erosional bases and contain mud clasts, and range in thickness from 15 to 34 cm. They exhibit inverse-normal grading, which suggests deposition from a high-energy sediment flow. These sand beds are characterized by a multilayered structure composed of fine alternations of sand sheets and mud drapes, which suggests deposition from a long-period wave train.

Based on changes in diatom assemblages, we suggest that Sand Bed T1 can be linked to historical Tokai earthquakes associated with an abrupt uplift of the Oya lowland. The diatom assemblages may further suggest gradual subsidence during the interseismic period. The youngest sand bed, T0, may be correlated with the ad 1096 Eicho earthquake.

As the preexisting geomorphological context remains unclear, it is difficult to estimate the size of the tsunami waves that deposited these sand beds. Further geomorphological and geological research is needed to reconstruct the specific characteristics of paleotsunamis in this region.

Footnotes

Acknowledgements

We thank Dr Takao Ubukata and Professor Toshiaki Masuda for their advice, Stallard Scientific Editing () for improving the English in the manuscript, and staff at Shizuoka City for assistance in the field.

Funding

This study was funded by Shizuoka University, Paleo Labo Co., Ltd, Fujiwara Natural History Foundation, and Strategic Funds for the Promotion of Science and Technology.

References

Ando

(1990) Environmental indicators based on freshwater diatom assemblages and its application to reconstruction of paleo-environments. Quarterly Journal of Geography 42: 73–78 (in Japanese with English Abstract).

Ando

(1975) Source mechanisms and tectonic significance of historical earthquakes along the Nankai trough, Japan. Tectonophysics 27: 119–140.

Arakawa

Ishida

Ito

(1961) Nihon takashio shiryo. Tokyo: Meteorological Research Institute, 272 pp. (in Japanese).

Atwater

(1987) Evidence for great Holocene earthquakes along the outer coast of Washington State. Science 236: 942–944.

Atwater

Hemphill-Haley

(1997) Recurrence Intervals for Great Earthquakes of the Past 3,500 Years at Northeastern Willapa Bay (Professional Paper 1576). Washington, DC: US Geological Survey, 108 pp.

Bronk Ramsey

(2009) Bayesian analysis of radiocarbon dates. Radiocarbon 51: 337–360.

Carver

McCalpin

(1996) Paleoseismology of compressional tectonic environments. In: McCalpin

(ed.) Paleoseismology. California: Academic Press, pp. 183–270.

Cisternas

Atwater

Torrjon

. (2005) Predecessors of the giant 1960 Chile earthquake. Nature 437: 404–407.

Costa

PJM

Andrade

Dawson

. (2012) Microtextural characteristics of quartz grains transported and deposited by tsunamis and storms. Sedimentary Geology 275–276: 55–69.

10.

Dawson

Smith

Ruffman

. (1996) The diatom biostratigraphy of tsunami sediments: Examples from recent and middle Holocene events. Physics and Chemistry of the Earth 21: 87–91.

11.

Fujiwara

(2008) Bedforms and sedimentary structures characterizing tsunami deposits. In: Shiki

Tsuji

Yamazaki

. (eds) Tsunamiites—Features and Implications. Amsterdam: Elsevier, pp. 51–62.

12.

Fujiwara

Kamataki

(2007) Identification of tsunami deposits considering the tsunami waveform: An example of subaqueous tsunami deposits in Holocene shallow bay on southern Boso Peninsula, Central Japan. Sedimentary Geology 200: 295–313.

13.

Fujiwara

Sato

Ono

. (2013) Researches on tsunami deposits using sediment cores: 3.4 ka tsunami deposit in the Rokken-gawa lowland near Lake Hamana, Pacific coast of central Japan. Journal of Geography 122: 308–322 (in Japanese with English Abstract).

14.

Fujiwara

Sawai

Shishikura

. (2012) Sedimentary features of the 2011 Tohoku earthquake tsunami deposit, on the central Kujukuri coast, east Japan. Quaternary Research (Daiyonki-kenkyu) 51: 117–126 (in Japanese with English Abstract).

15.

Geospatial Information Authority of Japan (2011) Outline of the inundation area by the tsunami from the 2011 off the Pacific coast of Tohoku earthquake. Available at: http://www.gsi.go.jp/kikaku/kikaku60003.html (in Japanese) (accessed 6 September 2012).

16.

Hatori

(1976) Documents of tsunami and crustal deformation in Tokai district associated with the Ansei earthquake of Dec. 23, 1854. Bulletin of the Earthquake Research Institute 51: 13–28 (in Japanese with English Abstract).

17.

Hatori

(1977) Field investigation of the Tokai tsunamis in 1707 and 1854 along the Shizuoka coast. Bulletin of the Earthquake Research Institute 52: 407–439 (in Japanese with English Abstract).

18.

Hirakawa

Nakamura

Echigo

(2000) Giant tsunami along the Pacific coast of the Tokachi region. Chikyu Monthly 31(Suppl.): 92–98 (in Japanese).

19.

Ishibashi

(1981) Specification of a soon-to-occur seismic faulting in the Tokai district, central Japan, based upon seismotectonics. In: Simpson

Richards

(eds) Earthquake Prediction – An International Review (Maurice Ewing Series 4). Washington, DC: AGU, pp. 297–332.

20.

Ishibashi

(1984) Coseismic vertical crustal movements in the Suruga Bay region. Quaternary Research (Daiyonki-kenkyu) 23: 105–110 (in Japanese with English Abstract).

21.

Japan Meteorological Agency (2011) Available at: http://www.jma.go.jp/jma/press/1002/28b/kaisetsu201002281650.pdf.

22.

Japan Meteorological Agency (2012) Available at: http://www.seisvol.kishou.go.jp/eq/2011_03_11_tohoku/tsunami_jp.pdf.

23.

Kashima

(2008) The quantitative reconstruction of salinity changes using diatom assemblages in inland saline lakes in the central part of Turkey during the Late Quaternary. Quaternary International 105: 13–19.

24.

Kitagawa

Fukusawa

Nakamura

(1995) AMS ¹⁴C dating of varved sediments from Lake Suigetsu, central Japan and atmospheric ¹⁴C change during the Late Pleistocene. Radiocarbon 37: 371–378.

25.

Kitamura

Fujiwara

Kobayashi

. (2011) Preliminary study on drill cores for evidence of run-up tsunami deposits from Holocene sediments in the southeast Shizuoka Plain, Shizuoka Prefecture. Geoscience Reports of Shizuoka University 38: 3–19 (in Japanese with English Abstract).

26.

Kosugi

(1988) Classification of living diatom assemblages as the indicator of environments, and its application to reconstruction of paleoenvironments. Quaternary Research (Daiyonki-kenkyu) 27: 1–20 (in Japanese with English Abstract).

27.

Kosugi

(1993) Diatom. In: Japan Association of Quaternary Research (ed.) A Handbook of Quaternary Research, vol. 2. Tokyo: University of Tokyo Press, pp. 245–252 (in Japanese).

28.

Krumbein

(1941) Measurement and geological significance of shape and roundness of sedimentary particles. Journal of Sedimentary Petrology 11: 64–72.

29.

Le Roux

Vargas

(2005) Hydraulic behavior of tsunami backflows: Insights from their modern and ancient deposits. Environmental Geology 49: 65–75.

30.

Lowe

(1982) Sediment gravity flows: II. Depositional models with special reference to the deposits of high-density turbidity currents. Journal of Sedimentary Petrology 52: 279–297.

31.

Mamo

Strotz

Dominey-Howes

(2009) Tsunami sediments and foraminiferal assemblages. Earth-Science Reviews 96: 263–278.

32.

Matsuda

(2006) Environmental changes of the Toro Ruins in Shizuoka Prefecture during 3100–1200 years ago from channel and floodplain deposits. Report of reexcavation of the Toro Ruins 1–27, The Board of Education Shizuoka City, Shizuoka (in Japanese).

33.

Minoura

Nakaya

(1991) Trances of tsunami preserved in inter-tidal lacustrine and marsh deposits: Some examples from northeast Japan. The Journal of Geology 99: 265–287.

34.

Minoura

Imamura

Sugawara

. (2001) The 869 Jogan tsunami deposit and recurrence interval of large-scale tsunami on the Pacific coast of northeast Japan. Journal of Natural Disaster Science 23: 83–88.

35.

Moore

Goff

McAdoo

. (2011) Sedimentary deposits from the 17 July 2006 western Java tsunami, Indonesia: Use of grain size analyses to assess tsunami flow depth, speed, and traction carpet characteristics. Pure and Applied Geophysics 168: 1951–1961.

36.

Nakada

Yonekura

Lambeck

(1991) Late Pleistocene and Holocene sea level changes in Japan: Implications for tectonic histories and mantle rheology. Palaeogeography, Palaeoclimatology, Palaeoecology 85: 107–122.

37.

Nanayama

Furukawa

Shigeno

. (2007) Nine unusually large tsunami deposits from the past 4000 years at Kiritappu marsh along the southern Kuril Trench. Sedimentary Geology 200: 275–294.

38.

Nanayama

Satake

Furukawa

. (2003) Unusually large earthquakes inferred from tsunami deposits along the Kuril Trench. Nature 424: 660–663.

39.

Naruse

Fujino

Suphawajruksakul

. (2010) Features and formation processes of multiple deposition layers from the 2004 Indian Ocean Tsunami at Ban Nam Kem, southern Thailand. Island Arc 19: 399–411.

40.

Nelson

Jennings

Kashima

(1996a) An earthquake history derived from stratigraphic and microfossil evidence of relative sea-level change at Coos Bay, southern coastal Oregon. Geological Society of America Bulletin 108: 141–154.

41.

Nelson

Sawai

Jennings

. (2008) Great earthquake paleogeodesy and tsunamis of the past 2000 years at Alsea Bay, central Oregon coast, USA. Quaternary Science Reviews 27: 747–768.

42.

Nelson

Shennan

Long

(1996b) Identifying coseismic subsidence in tidal-wetland stratigraphic sequences at the Cascadia subduction zone of western North America. Journal of Geophysical Research 101(B3): 6115–6135.

43.

Nemoto

Sato

Akai

. (1988) Submarine topography, bottom material, and geological structure of the continental shelf off Mt. Udo, western Suruga Bay. Journal of the School of Marine Science and Technology, Tokai University 27: 61–82 (in Japanese with English Abstract).

44.

Ota

Umitsu

Matsushima

(1990) Recent Japanese research on relative sea level changes in the Holocene and related problems – Review of studies between 1980 and 1988. Quaternary Research (Daiyonki-kenkyu) 29: 31–48 (in Japanese with English Abstract).

45.

Reimer

Baillie

MGL

Bard

. (2009) IntCal09 and Marine09 radiocarbon age calibration curves, 0–50,000 years cal BP. Radiocarbon 51: 1111–1150.

46.

Sangawa

(2001) Recent results of paleoseismological study based on earthquake traces excavated at archaeological sites. Annual Report on Active Fault and Paleoearthquake Researches No. 1, Geological Survey of Japan/AIST, Tsukuba, pp. 287–300 (in Japanese with English Abstract).

47.

Satake

Shimazaki

Tsuji

. (1996) Time and size of a giant earthquake in Cascadia inferred from Japanese tsunami records of January 1700. Nature 379: 246–249.

48.

Sawai

Fujiwara

Kamataki

. (2008) Marine incursions of the past 1500 years and evidence of tsunamis at Suijin-numa, a coastal lake facing the Japan Trench. The Holocene 18: 517–528.

49.

Sawai

Namegaya

Okamura

. (2012) Challenges of anticipating the 2011 Tohoku earthquake and tsunami using coastal geology. Geophysical Research Letters 39: L21309. DOI: 10.1029/2012GL053692.

50.

Sawai

Satake

Kamataki

. (2004) Transient uplift after a 17th-century earthquake along the Kuril subduction zone. Science 306: 1918–1920.

51.

Shennan

Hamilton

(2006) Coseismic and pre-seismic subsidence associated with great earthquakes in Alaska. Quaternary Science Reviews 25: 1–8.

52.

Shiba

Nemoto

Suruga Bay Collaborative Research Group . (1990) Geological structure of the Udo Hill and its adjacent continental shelf, western Suruga Bay. Journal of the Faculty of Marine Science and Technology, Tokai University 30: 47–65 (in Japanese with English Abstract).

53.

Sohn

(2000) Depositional processes of submarine debris flows in the Miocene fan deltas, Pohang Basin, SE Korea with special reference to flow transformation. Journal of Sedimentary Research 70: 491–503.

54.

Switzer

Jones

(2008) Large-scale washover sedimentation in a freshwater lagoon from the southeast Australian coast: Tsunami or exceptionally large storm. The Holocene 18: 787–803.

55.

Tanioka

Satake

(2001) Detailed coseismic slip distribution of the 1944 Tonankai earthquake estimated from tsunami waveforms. Geophysical Research Letters 28(6): 1075–1078.

56.

Tsuchi

Takahashi

(1972) Geologic history of alluvial plains on the coast of the Tokai Region, the Pacific side of central Japan. Memoirs of the Geological Society of Japan 7: 27–37 (in Japanese with English Abstract).

57.

Usami

(2003) Materials for Comprehensive List of Destructive Earthquakes in Japan. Tokyo: University of Tokyo Press, 416 pp. (in Japanese).

58.

Watanabe

(1998) Comprehensive List of Tsunami That Struck the Japanese Islands. 2nd Edition. Tokyo: University of Tokyo Press, 238 pp. (in Japanese).