© JSRM All rights reserved.

Volume 15, Number 1, June 2019, pp.1-4

[Summary]

Stress Buildup and Drop in Inland Shallow Crust Caused by the 2011

Tohoku-oki Earthquake Events

Kiyotoshi SAKAGUCHI*, Tatsuya YOKOYAMA**, Weiren LIN*** & Noriaki WATANABE*

* Graduate School of Environmental Studies, Tohoku University, Aobaku, Sendai 980-8579 Japan

** OYO Corporation, Minamiku, Saitama 336-0015 Japan

*** Graduate School of Engineering, Kyoto University, Saikyoku, Kyoto 615-8540 Japan

Received 15 05 2019; accepted 10 06 2019

ABSTRACT

This article is the summary of the published paper (Sakaguchi et al. 2017). In this study, to examine the change in in-situ stress

between before and after the 2011 Tohoku-oki earthquake, we performed stress measurements after the earthquake in the Kamaishi

mine. The in-situ stress measurement period was from 1991 to 2016. The results showed that the magnitudes of the

three-dimensional principal stresses and the vertical stress drastically increased during the mainshock and, at one year after the

earthquake, were more than double those before the earthquake. The principal stress magnitudes then decreased with time, and

returned almost to the pre-earthquake levels at about three years after the earthquake. The increasing and decreasing trends in stress

in the Kamaishi mine can be interpreted in terms of the effects of coseismic rupture behavior of the Tohoku-oki earthquake

mainshock and the occurrence of aftershocks in the Sanriku-oki low-seismicity region (SLSR), where the Kamaishi mine is located.

The drastic increase in stress suggests that the SLSR may act as a barrier to further rupture propagation. In addition, the consistency

between the change in measured stress and the change in seismicity in the Kamaishi regions suggests that the results of stress

measurements, even those at a much shallower depth than the earthquake source fault, can be useful for understanding rupture

propagation behavior.

Keywords: Stress change, Rock stress measurement, the 2011 Tohoku-oki earthquake, CCBO technique

1. INTRODUCTION

The Tohoku-oki earthquake that occurred on March 11,

2011 was in the largest class of earthquakes. The Tohoku

region experienced crustal movement of up to 5.3 m in the

horizontal direction and up to 1.2 m in the vertical direction

(subsidence) within a short period. This crustal disturbance

with a large displacement rate is likely to have had a major

impact on the crustal stress field at a relatively shallow depth.

Therefore, it is important that we clarify the change in in-situ

stress between before and after the 2011 Tohoku-oki

earthquake.

2. OVERVIEW OF THE IN-SITU STRESS

MEASUREMENT

The measurement location is the Kamaishi mine in

northeast Japan. The Kamaishi mine is located about 170 km

northwest of the earthquake epicenter. In this study, we

selected a measurement station that is about 5 km from the

mine opening of the 550 mL site at the Kamaishi mine. The

overburden at this measurement station is about 290 m. This

measurement station is not affected by galleries or the goaf

cavern in this mine. In addition, several stress measurements

have been performed at the 550 mL site before the

Tohoku-oki earthquake (Sakaguchi et al., 1995a; Sakaguchi et

al., 1995b; Sugawara and Obara, 1999; JNC, 1999). In-situ

stress measurements were performed using the Compact

Conical-ended Borehole Overcoring (CCBO) technique

(Sakaguchi et al., 1994). The CCBO technique is the one of

the suggested methods of the ISRM (Sugawara and Obara,

1999).

Figure 1 shows a plan view of the drift of the 550 mL site

at the Kamaishi mine. This figure also shows the

measurement stations used in this study after the mainshock

and those in other studies before the mainshock.

Figure 2 shows a plan view of the measurement station in

this study. In-situ stress measurements were performed four

times after the mainshock. The first measurement (SKO-1)

was performed from February 27 to March 1, 2012, the

second measurement (SKO-2) was performed from

December 17 to December 19, 2012, the third measurement

(SKO-3) was performed from March 10 to March 12, 2014,

and the fourth measurement (SKO-4) was performed from

March 14 to March 18, 2016, which represent, approximately,

one, two, three and five years after the mainshock,

respectively.

2 K. SAKAGUCHI et al. / International Journal of the JSRM vol.15 (2019) pp.1-4

Figure 1. A plan view of the drift the 550 mL site at the

Kamaishi mine. The measurement station used in this study

and those in other studies with the CCBO technique are

indicated by stars.

Figure 2. A plan view of the measurement station. The

measurement station is located where two galleries (width of

about 5.5 m and height of about 7 m) are adjacent. In-situ

stress measurements were performed in four boreholes

(SKO-1, SKO-2, SKO-3 and SKO-4) that are denoted by

thick black solid lines and red broken line. Strains were

recorded at every 3 mm of overcoring advance using

data-logging equipment and a PC.

3. RESULTS OF IN-SITU STRESS MEASUREMENT

Figure 3 shows the orientation of the principal stresses

(1, 2, 3) by a lower hemisphere stereographic projection.

The maximum principal stress 1 is oriented in a North-South

direction both before and after the mainshock. The

intermediate principal stress 2 was in an East-West

horizontal direction, and the minimum principal stress 3 was

in a vertical direction. This is consistent with reverse fault

stress and was maintained for up to five years after the

mainshock.

Figure 4 shows (a) the magnitude of the principal stresses

and (b) the ratio of the vertical stress v to the overburden

pressure pv as estimated from the rock density and the depth

from the ground surface just above the measurement station.

The magnitudes of the principal stresses at one year after

were two to three times greater than those before the

mainshock. However, at two years after the mainshock, the

magnitudes of the intermediate principal stress and the

minimum principal stress were almost the same as those

before the mainshock, while the magnitude of the maximum

principal stress was still large compared to that before the

mainshock. The magnitudes of the principal stresses at three

years and five years after the mainshock were almost the

same as those before the mainshock.

Figure 3. The orientation of the principal stresses (1, 2, 3)

by a lower hemisphere stereographic projection. 1 (circle) is

the maximum principal stress, 2 (square) is the intermediate

principal stress and 3 (star) is the minimum principal stress.

(a) The result that were obtained before the earthquake (K-1 -

K-5). (b) The result from at one year after to five years after

the 2011 mainshock.

The magnitude of the vertical stress σv was almost the same

as the overburden pressure except at 1 year after the

earthquake: the magnitude of σv at 1 year post-earthquake

was approximately 2.4 times greater than the overburden

pressure (Fig. 4(b)). Theoretically, the static vertical stress

has to be in a state of mechanical equilibrium with the

overburden pressure. In the study area, the topography is

steep: three tall mountains with altitudes of approximate

1,300 m are located within 3 km of the stress measurement

station. These three mountains are approximately 300–400 m

higher than the ground surface immediately above the

measurement station. As the height difference is larger than

the overburden ~290 m above the measurement station, a

higher vertical stress magnitude of more than two times of the

overburden pressure can be considered to be possible for a

short period of time (1 or 2 years).

(a)

(b)

K. SAKAGUCHI et al. / International Journal of the JSRM vol.1 (2019) pp.1-4 3

Figure 4. Annual trends in the stress state. The results after

the Tohoku-oki earthquake are mean values. (a) The three

principal stress magnitudes (σ1, σ2, σ3). These mean values of

principal stresses were calculated by averaging the principal

stress at every measuring point. (b) The ratio of the vertical

stress v to the overburden pressure pv calculated as

(gravitational acceleration) × (depth) × (average density

determined from rock samples to be 2.7 ton/m3). The error of

overburden pressure pv was evaluated assuming that the

measured value of overburden (depth) has an error ±10 m.

4. DISCUSSION

Figure 5 shows the the total slip distribution of larger than

5 m of the 2011 Tohoku-oki earthquake (Yagi and Fukuhata,

2011; Ye et al., 2012). Figure 6 shows the magnitude-time

plot for the off Kamaishi region (Ariyoshi et al., 2014). In

Figure 5, the pink rectangular frame shows the Sanriku-oki

low-seismicity region (SLSR) where located near the

northern termination of the Tohoku-oki mainshock rupture by

Ye et al. (2012). A rectangular region with small slipping is

distributed off Kamaishi (Figure 5). The Kamaishi region is

believed to correspond to the outside edge of the slipping

region of the 2011 Tohoku-oki earthquake, and this

rectangular shape area may have made the slipping stop.

Earthquakes had occurred off Kamaishi every approximately

5.5 years before the 2011 Tohoku-oki earthquake (Figure 6).

The magnitude of these earthquakes ranged from M = 4.7 to

M = 5.1. However, the interval between these earthquakes

decreased and the magnitude of the earthquakes off Kamaishi

increased after the Tohoku-oki mainshock. Furthermore, the

interval between earthquakes off Kamaishi gradually

increased at one year after the Tohoku-oki mainshock. At

about the same time, the magnitudes of these earthquakes

returned to the same level as those before the Tohoku-oki

mainshock.

Figure 5. The positional relationship between the Kamaishi

mine and the epicentre of the Tohoku-oki earthquake (yellow

star with a red outline). The epicentre of repeaters off

Kamaishi is marked by a red star and the location of the

Kamaishi mine is indicated by a black star (modified from

Ariyoshi et al., 2014). The total slip distribution of slip

greater than 5 m for the 2011 Tohoku-oki earthquake is also

plotted, from Yagi and Fukuhata (2011) and the epicentres of

earthquakes (M > 5) in the Sanriku-oki low-seismicity region

(SLSR), from Ye et al. (2012).

Figure 6. Magnitude-time plot for the off Kamaishi region

with close-up in the sky-blue-colored time window (insite),

lower figure, from Ariyoshi et al.. The vertical red line shows

the occurrence time of the 2011 Tohoku-oki earthquake.

These results suggest the following scenario regarding the

changes in the crustal stress in the Kamaishi mine after the

(a)

(b)

4 K. SAKAGUCHI et al. / International Journal of the JSRM vol.15 (2019) pp.1-4

Tohoku-oki mainshock. Slipping behavior off Kamaishi

increased the magnitude of the crustal stress at the Kamaishi

mine at one year after the Tohoku-oki mainshock. As a result,

the number of earthquakes off Kamaishi increased. These

earthquakes led to a decrease in the magnitude of the crustal

stress at the Kamaishi mine beyond one year after the

Tohoku-oki mainshock. The magnitude of the crustal stress in

the Kamaishi region then decreased beyond two years after

the Tohoku-oki mainshock, and the earthquakes off Kamaishi

decreased.

Iinuma et al. (2016) showed that the cumulative

postseismic slip (for the period from 23 April 2011 to 10

December 2011) of the 2011 Tohoku-oki earthquake in

offshore Kamaishi was larger than that of surrounding regions.

Moreover, Toda et al. (2011), Hiratsuka and Sato (2011) and

Sato et al. (2012) showed that the Coulomb stress change

(ΔCFF) around the Kamaishi mine exhibited a positive trend

after the Tohoku-oki earthquake, and Bletery et al. (2014)

demonstrated that the stress drop altered to a negative trend

after the mainshock. Additionally, Ishibe et al. (2017) showed

that the temporal changes in median ΔCFF from 2000 to the

middle of 2015. The median values of ΔCFF rapidly

increased just after the Tohoku-oki mainshock, after which

the median ΔCFF gradually decreased to background levels

approximately 3 years after the mainshock. Uchida et al.

(2013) showed that the stress drop due to earthquakes off

Kamaishi (20 Mar. 2011–23 Sep. 2011) was 2.4 MPa to 10.4

MPa. These observations also support our interpretation of

the stress change pattern examined in this study.

In addition, the consistency between the change in

measured stress and the change in seismicity in the Kamaishi

regions suggests that the results of stress measurements, even

those at a much shallower depth than the earthquake source

fault, can be useful for understanding rupture-propagation

behavior.

5. CONCLUSION

The direction of the maximum principal stress did not

change much between before and after the Tohoku-oki

mainshock. However, the directions of the intermediate

principal stress and the minimum principal stress after the

mainshock were different than those before the mainshock;

i.e., the intermediate principal stress was in the East-West

direction horizontally, while the minimum principal stress

was in the vertical direction.

The magnitudes of the principal stresses at one year after

were two to three times greater than those before the

mainshock. However, at two years after the mainshock, the

magnitudes of the intermediate principal stress and the

minimum principal stress were almost the same as those

before the mainshock. The magnitudes of the principal

stresses at three years and five years after the mainshock were

almost the same as those before the mainshock. The

magnitude of the vertical stress at one year after the

mainshock was about 2.4 times greater than the weight of the

overburden. Moreover, the magnitude of the vertical stress at

five years was almost the same as the weight of the

overburden.

The increasing and decreasing trends in the crustal stress

in the Kamaishi mine are believed to be influenced by both

the slipping behavior off Kamaishi at the Tohoku-oki

earthquake and repeated earthquakes off Kamaishi.

REFERENCES

Ariyoshi, K., Uchida, N., Matsuzawa, T., Hino, R., Hasegawa, A.,

Hori, T. & Kaneda, Y., 2014. A trial estimation of frictional

properties, focusing on aperiodicity off Kamaishi just after the

2011 Tohoku earthquake, Geophys. Res. Lett., doi:

10.1002/2014GL061872.

Bletery, Q. et al., 2014. A detailed source model for the Mw 9.0

Tohoku-Oki earthquake reconciling geodesy, seismology, and

tsunami records, J. Geophys. Res., 119, pp. 7636–7653,

doi:10.1002/2014JB011261.

Hiratsuka, S. & Sato, T., 2011. Alteration of stress field brought about

by the occurrence of the 2011 off the Pacific coast of Tohoku

Earthquake (Mw 9.0), Earth Planets Space, 63, pp. 681–685,

doi:10.5047eps.2011.05.013.

Iinuma, T. et al., 2016. Seafloor observations indicate spatial

separation of coseismic and postseismic slips in the 2011 Tohoku

earthquake, Nature Communications, 7, 13506,

doi:10.1038/ncomms13506.

Ishibe, T., Ogata, Y., Tsuruoka, H. & Satake, K., 2017. Testing the

Coulomb stress triggering hypothesis for three recent megathrust

earthquakes, Geosci. Lett., 4, 5, doi:10.1186/s40562-017-0070-y.

Japan Nuclear Cycle Development Institute, 1999, JNC TN7410

99-001.

Sakaguchi, K., Takehara, T., Obara, Y., Nakayama, T. & Sugawara,

K., 1994. Rock stress measurement by means of the Compact

Overcoring method, Journal of MMIJ, 110, pp. 331-336.

Sakaguchi, K., Huang, X., Noguchi, Y. & Sugawara, K., 1995a.

Application of Conical-ended Borehole technique to

discontinuous rock and consideration, Journal of MMIJ, 111, pp.

283-288.

Sakaguchi, K., Takeda, H. & Matsuki, K., 1995b. In-Situ rock stress

measurement using improved Downward Compact

Conical-Ended Borehole Overcoring technique, Journal of MMIJ,

126, pp. 418-424.

Sato, T., Hiratsuka, S. & Mori, J., 2012. Coulomb stress change for

the normal-fault aftershocks triggered near the Japan Trench by

the 2011 Mw 9.0 Tohoku-Oki earthquake, Earth Planets Space,

64, pp. 1239–1243, doi:10.5047/eps.2012.04.003.

Sugawara, K. & Obara Y., 1999. Draft ISRM suggested method for in

situ stress measurement using the Compact Conical-ended

Borehole Overcoring (CCBO) technique, Int. J. Rock Mech. Min.

Sci., 36, pp. 307-322.

Toda, S., Lin, J. & Stein, R., 2011. Using the 2011 Mw 9.0 off the

Pacific coast of Tohoku Earthquake to test the Coulomb stress

triggering hypothesis and to calculate faults brought closer to

failure, Earth Planets Space, 63, pp. 725–730,

doi:10.5047/eps.2011.05.010.

Uchida, N., Shimamura, K., Matsuzawa, T. & Okada, 2013. T.

Postseismic response of repeating earthquakes around the 2011

Tohoku-oki earthquake: Moment increase due to the fast loading

rate, J. Geophy. Res., 120, pp. 259–274,

doi:10.1002/2013JB010933.

Yagi, Y. & Fukuhata, Y., 2011. Rupture process of the 2011

Tohoku-oki earthquake and absolute elastic strain release,

Geophys. Res. Lett., 38, L19307, doi:10.1029/2011GL048701.

Ye, L., Lay, T. & Kanamori, H., 2012. The Sanriku-Oki

low-seismicity region on the northern margin of the great 2011

Tohoku-Oki earthquake rupture, J. Geophys. Res., 117, B02305,

doi:10.1029/2011JB008847.