11:45 AM - 12:00 PM
[SSS11-10] Verifying variability and duplicability of near-surface fault deformation: re-excavation of pre-earthquake paleoseismic trenches, Futagawa fault, Central Kyushu, Japan
Keywords:Active fault, Paleoseismic trench survey, Dextral strike-slip fault, Futagawa fault, 2016 Kumamoto earthquake, Characteristic earthquake model
Long-term assessments of active faults are built based on their active history revealed by paleoseismic surveys. Thus, information about if a fault identified in a paleoseismic survey will repeat the same deformation is important in the long-term assessments. A coseismically displaced pre-earthquake paleoseismic trench could bring this valuable information.
There are only two examples that coseismically displaced pre-earthquake paleoseismic trenches were re-excavated and their pre- and post-earthquake walls were compared. The first survey (Crone, 1985) was conducted on the Lost River fault (a normal fault, Idaho, United States), which displaced a 7-yr old trench by ~2 m during the 1983 Borah Peak earthquake (Mw 6.9). In the re-excavation survey, it was comfirmed that the location and style of the 1983 slip was similar to those associated with the prior earthquake. The second example (Morris et al., 2023) is a survey on the Kekerengu fault (a dextral strike-slip fault, the South Island of New Zealand), which displaced a 10-month-old paleoseismic trench by ~9 m during the 2016 Kaikōura earthquake (Mw 7.8). The coseismically displaced trench was re-excavated, and it was revealed that at least one of slip locations during 2016 was different from that of the previous event. In addition, it was reported that despite the deformational expression of the strike-slip fault on the pre-earthquake wall was transtension, that on post-earthquake wall was transpression. Thus, different results were obtained in the two studies about the difference in faulting between pre- and post-earthquake trench walls. Thus, further studies are needed to verify the character of the near-surface faulting, such as conditions under which an active fault repeats similar deformation. Based on the background above, we re-excavated pre-earthquake paleoseismic trenches in the Tanaka site (Kumamoto Prefecture, 1996; Yoshioka et al., 2007), displaced by the Futagawa fault during the 2016 Kumamoto earthquake (Mw 7.0), to reveal coseismic behaviors of faults reported on the pre-earthquake trench walls and compare the deformational styles on pre- and post-earthquake trench walls.
At the Tanaka site, there are three pre-earthquake paleoseismic trenches (Fig. 1A), and they were dextrally displaced by 60 cm during the 2016 earthquake (Shirahama et al., 2016). In this study, a part of the paleoseismic trench by Kumamoto Prefecture (1996) (hereafter KP trench) and the Tanaka A trench by Yoshioka et al. (2007) were re-excavated (Fig. 1B, C). The east wall of the KP trench and the additionally excavated part of the west wall of the Tanaka A trench were successfully re-exposed.
Kumamoto Prefecture (1996) reported a fault cutting up to just below the cultivated soil on the east wall of the KP trench. After the re-excavation, it was confirmed that the trench wall was dextrally displaced by 40–50 cm along the fault previously reported. Despite the displacement, the appearance of and that of the stratigraphic deformation on pre- and post-earthquake walls are similar. Thus, it is indicated that the slip locus and the near-surface deformation style on the wall during the 2016 event was similar to those of the previous event. This is consistent with the result on the Lost River fault (Crone, 1985).
Yoshioka et al. (2007) reported that the additionally excavated part of the west wall of the Tanaka A trench exposed a fault cutting up to just below a silt layer with sand and cobble, which showed the age of 2,180±40 BP. This fault was recognized on the re-exposed trench wall. The lateral slip of 40–50 cm during the 2016 earthquake on the trench wall, however, is located 10–50 cm south of the previously identified fault. On the mainly displaced part, transtensional deformation style, in which sediment are depressed between marginal faults, was exposed, although the pre-earthquake wall displayed the transpressional deformation (Yoshioka et al., 2007). Thus, the locus and the deformational mode of faulting during the 2016 event are probably different from those during the prior event on the west wall of the Tanaka A trench. This survey result consists with that on the Kekerengu fault and supports the idea that active faults should be defined as a zone (Morris et al., 2023).
As the Futagawa fault and the Kekerengu fault (Morris et al., 2023) show the variability in near-surface slip style between earthquakes, it is inferred that at least in the case of strike-slip faults, their near-surface deformation could change in the next faulting. The possible reason for this is that the position of en echelon fissures could change with each earthquake, depending on some factors such as difference in the property of the sediment.
There are only two examples that coseismically displaced pre-earthquake paleoseismic trenches were re-excavated and their pre- and post-earthquake walls were compared. The first survey (Crone, 1985) was conducted on the Lost River fault (a normal fault, Idaho, United States), which displaced a 7-yr old trench by ~2 m during the 1983 Borah Peak earthquake (Mw 6.9). In the re-excavation survey, it was comfirmed that the location and style of the 1983 slip was similar to those associated with the prior earthquake. The second example (Morris et al., 2023) is a survey on the Kekerengu fault (a dextral strike-slip fault, the South Island of New Zealand), which displaced a 10-month-old paleoseismic trench by ~9 m during the 2016 Kaikōura earthquake (Mw 7.8). The coseismically displaced trench was re-excavated, and it was revealed that at least one of slip locations during 2016 was different from that of the previous event. In addition, it was reported that despite the deformational expression of the strike-slip fault on the pre-earthquake wall was transtension, that on post-earthquake wall was transpression. Thus, different results were obtained in the two studies about the difference in faulting between pre- and post-earthquake trench walls. Thus, further studies are needed to verify the character of the near-surface faulting, such as conditions under which an active fault repeats similar deformation. Based on the background above, we re-excavated pre-earthquake paleoseismic trenches in the Tanaka site (Kumamoto Prefecture, 1996; Yoshioka et al., 2007), displaced by the Futagawa fault during the 2016 Kumamoto earthquake (Mw 7.0), to reveal coseismic behaviors of faults reported on the pre-earthquake trench walls and compare the deformational styles on pre- and post-earthquake trench walls.
At the Tanaka site, there are three pre-earthquake paleoseismic trenches (Fig. 1A), and they were dextrally displaced by 60 cm during the 2016 earthquake (Shirahama et al., 2016). In this study, a part of the paleoseismic trench by Kumamoto Prefecture (1996) (hereafter KP trench) and the Tanaka A trench by Yoshioka et al. (2007) were re-excavated (Fig. 1B, C). The east wall of the KP trench and the additionally excavated part of the west wall of the Tanaka A trench were successfully re-exposed.
Kumamoto Prefecture (1996) reported a fault cutting up to just below the cultivated soil on the east wall of the KP trench. After the re-excavation, it was confirmed that the trench wall was dextrally displaced by 40–50 cm along the fault previously reported. Despite the displacement, the appearance of and that of the stratigraphic deformation on pre- and post-earthquake walls are similar. Thus, it is indicated that the slip locus and the near-surface deformation style on the wall during the 2016 event was similar to those of the previous event. This is consistent with the result on the Lost River fault (Crone, 1985).
Yoshioka et al. (2007) reported that the additionally excavated part of the west wall of the Tanaka A trench exposed a fault cutting up to just below a silt layer with sand and cobble, which showed the age of 2,180±40 BP. This fault was recognized on the re-exposed trench wall. The lateral slip of 40–50 cm during the 2016 earthquake on the trench wall, however, is located 10–50 cm south of the previously identified fault. On the mainly displaced part, transtensional deformation style, in which sediment are depressed between marginal faults, was exposed, although the pre-earthquake wall displayed the transpressional deformation (Yoshioka et al., 2007). Thus, the locus and the deformational mode of faulting during the 2016 event are probably different from those during the prior event on the west wall of the Tanaka A trench. This survey result consists with that on the Kekerengu fault and supports the idea that active faults should be defined as a zone (Morris et al., 2023).
As the Futagawa fault and the Kekerengu fault (Morris et al., 2023) show the variability in near-surface slip style between earthquakes, it is inferred that at least in the case of strike-slip faults, their near-surface deformation could change in the next faulting. The possible reason for this is that the position of en echelon fissures could change with each earthquake, depending on some factors such as difference in the property of the sediment.