10:45 AM - 12:15 PM
[SCG45-P06] The role of backthrust in the development of the Coulomb wedge from sandbox experiments
Keywords:accretionary wedge, sandbox, digital image correlation
1. Introduction
Some shallow slow earthquakes observed at the plate boundary are thought to occur within the subduction wedge. Therefore, it is important to understand the deformation processes that occurs inside the tip of the wedge to establish the mechanism of shallow slow earthquakes. In this study, we constructed a Coulomb wedge in analogue (sand box) experiments using dry sand (Toyoura silica sand) and analyzed images of the developed wedge using digital image correlation (DIC). In particular, we traced shear bands such as the backthrust (BT), a forward-dipping fault at the rear of the wedge, and the frontal thrust (FT), a backward-dipping fault at the tip of the wedge. At the same time, load measurement was carried out using a load cell to explore how much the BT contributes to intra-wedge deformation.
2. Method
The experimental method was as follows. An adhesive sheet was placed inside an acrylic box with a load cell and an actuator. A 16-mm-thick layer of Toyoura sand covered the bottom of the box, and a camera was placed next to the sand layer. The actuator was pulled over a distance of 250 mm at a velocity of 0.4 mm/s to create a wedge. Up to 500 images were obtained at an interval of 1 image every 0.4 mm of sheet movement over a displacement of 50–250 mm. The obtained images were imported into a PC and analyzed using DIC. The linear region with a principal strain of >1.0% and located close to the fixed wall was identified as the BT, and the distance from the intersection of the BT and FT to the fixed wall was measured. Results were compared with load data to investigate the correlation between the location of the BT and the load.
3. Results
As a result of the experiment, it was clarified that the timing and place of formation of FT and BT changed depending on filling rate of sand layer. The high-filling-rate layer (filling rate of about 67%) has fewer shear bands, the displacement period of one shear band is longer, and the shear bands. It was concluded that these characteristics are caused by the degree of strain weakening during the formation of shear bands. In addition, it was found from the experiment that the displacement of the BT greatly contributed to the deformation within the wedge. The displacement of the FT is large, but it is accumulated when the FT is formed at the wedge tip. The FT does not displace much when the next FT newly occurs and is positioned inside the wedge. In the low-filling-rate layer experiment, it was clarified that the deformation inside the wedge was not localized but occurred over a wide area of the wedge.
4. Discussion
Based on the scaling law, the deformation discussed in the DIC analysis in this study corresponds to the deformation process every 190–420 years within the several km at the tip of the natural accretionary complex. there is a spatial and temporal gap between analogue experiments and observational data. Thus, improving the resolution of each experiment, analysis, and observation and accumulating data are future tasks to improve the comparison between the analogue experiments and the prototypes.
Some shallow slow earthquakes observed at the plate boundary are thought to occur within the subduction wedge. Therefore, it is important to understand the deformation processes that occurs inside the tip of the wedge to establish the mechanism of shallow slow earthquakes. In this study, we constructed a Coulomb wedge in analogue (sand box) experiments using dry sand (Toyoura silica sand) and analyzed images of the developed wedge using digital image correlation (DIC). In particular, we traced shear bands such as the backthrust (BT), a forward-dipping fault at the rear of the wedge, and the frontal thrust (FT), a backward-dipping fault at the tip of the wedge. At the same time, load measurement was carried out using a load cell to explore how much the BT contributes to intra-wedge deformation.
2. Method
The experimental method was as follows. An adhesive sheet was placed inside an acrylic box with a load cell and an actuator. A 16-mm-thick layer of Toyoura sand covered the bottom of the box, and a camera was placed next to the sand layer. The actuator was pulled over a distance of 250 mm at a velocity of 0.4 mm/s to create a wedge. Up to 500 images were obtained at an interval of 1 image every 0.4 mm of sheet movement over a displacement of 50–250 mm. The obtained images were imported into a PC and analyzed using DIC. The linear region with a principal strain of >1.0% and located close to the fixed wall was identified as the BT, and the distance from the intersection of the BT and FT to the fixed wall was measured. Results were compared with load data to investigate the correlation between the location of the BT and the load.
3. Results
As a result of the experiment, it was clarified that the timing and place of formation of FT and BT changed depending on filling rate of sand layer. The high-filling-rate layer (filling rate of about 67%) has fewer shear bands, the displacement period of one shear band is longer, and the shear bands. It was concluded that these characteristics are caused by the degree of strain weakening during the formation of shear bands. In addition, it was found from the experiment that the displacement of the BT greatly contributed to the deformation within the wedge. The displacement of the FT is large, but it is accumulated when the FT is formed at the wedge tip. The FT does not displace much when the next FT newly occurs and is positioned inside the wedge. In the low-filling-rate layer experiment, it was clarified that the deformation inside the wedge was not localized but occurred over a wide area of the wedge.
4. Discussion
Based on the scaling law, the deformation discussed in the DIC analysis in this study corresponds to the deformation process every 190–420 years within the several km at the tip of the natural accretionary complex. there is a spatial and temporal gap between analogue experiments and observational data. Thus, improving the resolution of each experiment, analysis, and observation and accumulating data are future tasks to improve the comparison between the analogue experiments and the prototypes.