10:15 AM - 10:30 AM
[MIS13-06] Investigation of paleo-tsunami deposit in southern Sado Island, Japan, and associated tsunami source
Keywords:Japan Sea, Tsunami sedimentology, Paleo-earthquake
Findings on spatial and temporal distributions of paleo-tsunami deposits are quite limited in the Japan Sea coasts compared with the Pacific coasts Sadakane et al. (2019) conducted a preliminary survey in the Hamochi area, southern Sado Island, to establish the paleo-tsunami history in this region. They found an event sand layer that suggests a possible prehistoric tsunami about 3,760 years ago, although it lacks strong evidence for tsunami origin.
In this study, we conducted a tsunami deposit survey in the Hamochi area, sedimentological and geochemical analysis of samples, and numerical modeling of tsunami propagation and inundation, to clarify the distribution and sedimentological characteristics of the tsunami deposit and associated tsunami source. The study area consists of a coastal lowland with 1 km in width, 2 km in-depth, and 3 to 10 m in elevation, and a shore-parallel beach ridge with elevations ranging from 5 to 8 m. The survey was conducted in November 2020. Columnar geological samples were collected by simple impact boring at six locations 600-900 m away from the coast. During the field observation, an event layer was identified in the background sediments of the wetland. The event layers were found at the same horizons in the core depth and consisted of 2-8 cm thick fine-grained sand.
Changes in grain size and geochemistry of the samples were analyzed to examine tsunami origin of the event layer. Results showed that the event sand layer had a normal and reverse grading similar to known modern tsunami deposits. Geochemical signatures such as Ca, Sr indicated a seawater origin of the sand layer. These results suggest that the event layer is highly likely a tsunami deposit. Radiocarbon (14C) dating returned 3700-4000 cal yr BP as a formation age of the tsunami deposit. This age is close to the age of the Ev. 10 layer found from Kamo Lake, although the regional correlation is not so far possible.
Tsunami numerical modeling was performed using the TUNAMI-N2 code. Five fault models around Sado Island, including the Kakuta-Yahiko fault, which is known as a reverse fault extending from the western edge of the Echigo Plain to off Niigata, were examined as the tsunami sources. Results show three fault models can explain the distribution of the tsunami deposits in the Hamochi lowland. In addition, considering activity history and average slip rate of the faults, the F42 fault (Mw about 7.4; Japan Sea Earthquake Research Association, 2014), which has a total length of 55.8 km, a width of 17.7 km, and a slip of 4.65 m, located in the west of Sado Island, was selected as a most likely candidate. Despite these findings, more research is needed for regional distribution of the contemporary tsunami deposits in other areas to determine the tsunami source.
In this study, we conducted a tsunami deposit survey in the Hamochi area, sedimentological and geochemical analysis of samples, and numerical modeling of tsunami propagation and inundation, to clarify the distribution and sedimentological characteristics of the tsunami deposit and associated tsunami source. The study area consists of a coastal lowland with 1 km in width, 2 km in-depth, and 3 to 10 m in elevation, and a shore-parallel beach ridge with elevations ranging from 5 to 8 m. The survey was conducted in November 2020. Columnar geological samples were collected by simple impact boring at six locations 600-900 m away from the coast. During the field observation, an event layer was identified in the background sediments of the wetland. The event layers were found at the same horizons in the core depth and consisted of 2-8 cm thick fine-grained sand.
Changes in grain size and geochemistry of the samples were analyzed to examine tsunami origin of the event layer. Results showed that the event sand layer had a normal and reverse grading similar to known modern tsunami deposits. Geochemical signatures such as Ca, Sr indicated a seawater origin of the sand layer. These results suggest that the event layer is highly likely a tsunami deposit. Radiocarbon (14C) dating returned 3700-4000 cal yr BP as a formation age of the tsunami deposit. This age is close to the age of the Ev. 10 layer found from Kamo Lake, although the regional correlation is not so far possible.
Tsunami numerical modeling was performed using the TUNAMI-N2 code. Five fault models around Sado Island, including the Kakuta-Yahiko fault, which is known as a reverse fault extending from the western edge of the Echigo Plain to off Niigata, were examined as the tsunami sources. Results show three fault models can explain the distribution of the tsunami deposits in the Hamochi lowland. In addition, considering activity history and average slip rate of the faults, the F42 fault (Mw about 7.4; Japan Sea Earthquake Research Association, 2014), which has a total length of 55.8 km, a width of 17.7 km, and a slip of 4.65 m, located in the west of Sado Island, was selected as a most likely candidate. Despite these findings, more research is needed for regional distribution of the contemporary tsunami deposits in other areas to determine the tsunami source.