Gondwana Research (Gondwana Newsletter Section) V. 8, No. 4, pp. 585-588. © 2005 International Association for Gondwana Research, Japan.

585

GNL

CORRESPONDENCE

Aftershock Propagation Characteristics During the First Three Hours Following the 26 December 2004 Sumatra-Andaman Earthquake Sujit Dasgupta, Basab Mukhopadhyay and A. Acharyya Geological Survey of India, Kolkata - 700 016, India, E-mail: [email protected] (Manuscript received May 1, 2005; accepted June 30, 2005)

Abstract Within three hours of the mainshock rupture of the 26 December 2004 Sumatra-Andaman earthquake, 45 aftershocks occurred that are distributed all along the mega-thrust fault plane and also along the West Andaman fault. Seven of these aftershocks struck sequentially and unilaterally from the mainshock in the south towards north within 2h 9m 50.76s indicating an overall rate of aftershock propagation to the tune of 167 meters/sec. Seismic moment calculated from fault parameters gives a value of 1.2 × 1030 dyne cm. Three separate fault segments are identified from distribution of aftershocks with propagation rates 330, 250 and 85 meters/sec in the southern, central and northern segments. These 7 unilaterally propagating shocks along the mega-thrust are probably not aftershocks of the mainshock rather these are sequentially triggered shocks each rupturing a small segment of the fault. Location of the mainshock and several aftershocks are guided by several lithospheric hinge faults identified previously. Key words: Sumatra-Andaman earthquake, aftershock propagation, triggered shock, fault segment, rupture velocity.

Introduction A natural disaster strikes when the previous incidence is almost forgotten and strangely it hits where it is least expected. Notwithstanding a few soft claims in the media on the forecast of this major earthquake that created havoc via the tsunami all along the Indian Ocean rim countries, the event could not have been predicted within a reasonable space, time and size window with the present knowledge of earthquake physics, statistics and tectonics. In spite of a host of organizations in India involved in the study of seismology, with around a hundred odd seismograph stations both old and new, our shyness in prompt and online data dissemination continues and this indecision is perhaps one of the several reasons why information on the tsunami wave that struck the Great and Car Nicobar by 7 am (IST) did not percolate the Indian mainland where it took two more hours to travel and was destined for devastation. Basic seismological data on which our understanding of this mega-event is derived even after six-months, is largely provided by the USGS website. From the study of NEIC earthquake catalogue both in the pre- and post-26th Gondwana Research, V. 8, No. 4, 2005

December 2004 scenario we had demonstrated (see http//: www.gsi.gov.in/suma_eq.htm) spatio-temporal variation of seismicity pattern; between 1st January and 26th November 2004 there are records of 260 events from the region, while in the period since 27th November till the great earthquake of 26th December there was a clear seismic quiescence of one month. We had further shown that all aftershocks that struck on 26th December 2004 (193 as listed on 08.02.2005) form three distinct linear clusters along the subduction mega-thrust and two more clusters along the West Andaman fault. We have revisited the NEIC catalogue (as on 08.06.2005) to find that 283 aftershocks are recorded on 26th December itself and in this note we analyze the spatio-temporal behavior of the aftershocks that occurred within the first three hours of the mainshock.

Analysis A chronological listing (Table 1) of 45 aftershocks that struck within three hours of the mainshock indicate that 7 such events (in bold Roman, Table 1) occur within and

586 Table 1. Chronological listing of aftershocks up to 3 hours following the 26 December 2004 Sumatra-Andaman Earthquake SL_NO

YEAR

MO

DA

HR_MN_SEC

LAT

LONG DEPTH mb

0 1

2004 2004

12 12

26 26

00 58 53.45 01 17 10.33

3.30 95.98 4.94 94.27

30 30

8.9 5.5

2

2004

12

26

01 21 20.66

6.34 93.36

30

6.1

3

2004

12

26

01 22 25.59

7.42 93.99

30

6.0

4 5

2004 2004

12 12

26 26

01 25 48.76 01 30 15.74

5.50 94.21 8.83 93.71

30 30

6.1 5.5

6 7 8 9

2004 2004 2004 2004

12 12 12 12

26 26 26 26

01 33 22.38 01 40 7.13 01 48 52.07 01 52 43

7.76 5.84 5.43 10.38

93.71 93.15 94.46 92.12

25 30 51 12

5.5 5.3 5.7 5.2

10 11 12

2004 2004 2004

12 12 12

26 26 26

01 59 13.99 02 04 0.03 02 15 23.57

8.39 6.85 6.17

92.45 94.67 93.47

30 30 30

5.3 6.0 5.6

13

2004

12

26

02 15 49.5

12.26 92.28

20

5.3

14

2004

12

26

02 15 59.78 12.32 92.50

26

5.7

15 16 17 18 19 20 21 22 23 24

2004 2004 2004 2004 2004 2004 2004 2004 2004 2004

12 12 12 12 12 12 12 12 12 12

26 26 26 26 26 26 26 26 26 26

02 22 1.84 02 30 28.94 02 34 52.15 02 36 10.09 02 38 9.35 02 40 59.85 02 43 5.26 02 45 17.65 02 46 20.74 02 52 1.83

8.87 6.72 3.99 12.18 8.49 7.48 9.22 8.46 4.24 12.50

92.47 93.08 94.14 92.94 92.35 92.43 94.00 92.61 93.61 92.60

15 15 30 38 33 30 30 30 30 30

5.7 5.1 5.7 5.8 5.6 5.4 4.9 5.2 5.7 5.8

25 26 27 28 29 30

2004 2004 2004 2004 2004 2004

12 12 12 12 12 12

26 26 26 26 26 26

02 53 13.04 02 56 40.37 02 59 14.39 03 23 8.08 03 61 3.05 03 84 4.21

0.06 8.61 3.18 8.61 8.19 13.74

97.04 92.29 94.38 92.33 92.46 93.01

30 30 30 30 27 30

5.4 4.9 5.7 5.5 5.1 5.9

31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

2004 2004 2004 2004 2004 2004 2004 2004 2004 2004 2004 2004 2004 2004 2004

12 12 12 12 12 12 12 12 12 12 12 12 12 12 12

26 26 26 26 26 26 26 26 26 26 26 26 26 26 26

03 93 4.08 03 14 13.84 03 17 52.38 03 19 13.05 03 22 57.48 03 24 54.94 03 26 45.79 03 30 1.38 03 40 15.64 03 44 8.34 03 46 42.04 03 50 22.18 03 51 12.36 03 54 44.77 04 04 2.83

4.05 7.44 7.21 3.55 5.82 4.47 4.91 4.64 5.53 13.47 6.72 5.51 5.05 6.48 4.76

93.53 94.26 92.92 94.29 95.09 94.07 96.40 94.00 94.33 92.74 93.33 94.25 94.77 92.89 93.79

30 30 30 30 20 26 30 25 30 22 46 48 30 30 16

5.4 5.4 5.6 5.5 5.4 5.8 5.3 5.2 5.6 5.2 5.0 5.30 5.7 5.1 5.2

MAGNITUDE Ms Mw Mo 8.9

9.0

Remarks

3.95E+29 Main shock 1st aftershock along the plate interface 2nd aftershock along the plate interface 1st aftershock along the west Andaman Fault 2nd aftershock along the West Andaman Fault

3rd aftershock along the plate interface in the direction of rupture propagation

4th aftershock along the plate interface in the direction of rupture propagation 5th aftershock along the plate interface in the direction of rupture propagation

6th aftershock along the plate interface in the direction of rupture propagation

7th aftershock along the plate interface in the direction of rupture propagation

Gondwana Research, V. 8, No. 4, 2005

cmyk 587

define the entire mega-thrust rupture from the mainshock in the south to aftershock (No. 30) in the north. Usually aftershocks strike via residual stress to break small asperities left by the main rupture, but in the present case it appears

Fig. 1. Tectonic map of Sumatra-Andaman region (after Curray, 2005) with 45 aftershocks (solid circle: focal depth ≤40 km; solid triangle >40 km) that occurred within three hours of the mainshock (star) of 26 December 2004. The red line is the 40 km contour on top of the subducting Indian lithosphere Benioff Zone (after Dasgupta et al., 2003). Red solid circles are the seven aftershocks that occurred sequentially from south to north along the mega-thrust. The two green circles are sequential aftershocks along the West Andaman fault. The green region between the trench and 40 km contour is the total fault rupture area. I, II and III are fault segments along the plate interface (see Table 2) and IA denotes fault segment along WAF (see Table 3). Red star: volcano; N–Narcondam, B–Barren; ASR–Andaman spreading ridge; MPF–Mae Ping Fault; TPF–Three Pagodas Fault; SSF–Shan Scrap Fault; WAF–West Andaman Fault; RF–Ranong Fault; KMF–Khlong Marui Fault. Gondwana Research, V. 8, No. 4, 2005

different, with the mainshock triggering aftershocks 1 followed by 2, 9, 13, 14, 24 and 30 each triggering the next along the fault azimuth. These seven aftershocks occurred in temporal succession along the unilateral direction of rupture propagation and are shown as red solid circles (Fig. 1) on a tectonic map (Curray, 2005) superimposed with 40 km contour (red line in Fig. 1) on top of the Benioff Zone from Dasgupta et al. (2003). It took 2h 9m 50.76s to break up to the northern-most point of the mega-thrust since the mainshock, traversing a total of about 1300 km fault length with an average rate of about 167 m/sec. The actual fault length measured along the trench (Fig. 1) is however 1440 km. The downdip width of fault rupture varies between 90 and 173 km and the total rupture area is around 2.0 × 105 sq km. With an average slip of 15 metre [and rigidity (µ) as 4 × 1011 dyne/cm2], seismic moment (Mo) calculates to the order of 1.2 × 1030 dyne cm, a value very close to that given by Stein and Okal (2005). The mainshock along with aftershocks 1 and 2 define fault segment I that is 444 km-long and trends N40W. Average aftershock propagation rate is around 330 m/sec (Table 2), 6 to 7.5 times less than the modeled unilateral rupture velocity of 2.0 to 2.5 km/sec in the mainshock rupture segment (Yagi, 2005; Chen Ji, 2005; Yamanaki, 2005). The afterslip (or triggered slip) rate in the southern sector of fault segment I, between the mainshock and aftershock 1 (265 km), is around 242 m/sec while it is faster in the northern part (183 km) of 730 m/sec from shocks 1 to 2. Eight more aftershocks occurred in segment I close to and south of aftershock 1, and two more locate near aftershock 2 that broke smaller asperities intermittently during the first three hours. Segment II of the fault trends N17W and is around 469 km-long, defined by aftershock numbers 2 and 9 that locate on the southern and northern tip of the fault, respectively. It may be noted that except shock 7 that occurred near the junction of fault segments I and II, the other five aftershocks in between (2 and 9) locate along the West Andaman Fault. The aftershock propagation rate in fault segment II is of the order of 250 m/sec (Table 2), slower than that of segment I. Twelve aftershocks are recorded from this segment between Great and Car Nicobar that kept rocking the islands intermittently. Fault segment III is 386 km-long, trends N15E between aftershocks 9 and 30, and propagate unilaterally through shocks 13, 14 and 24. Average aftershock propagation velocity has further slowed down in this segment, which is of the order of 85m/sec (Table 2). Rate in the southern sector of fault segment III, between shocks 9 and 14, is about 158 m/sec and in the northern sector, between 24 and 30 is 147 m/sec, while it was very slow of 11 m/sec to break the 300 m patch between shocks 13 and 24.

cmyk

588 Table 2. Aftershock propagation rate along the megathrust. Aftershock No (see figure 1 and table 1)

Time after mainshock

Fault segment

Length (km) of the fault segment

Time (sec) for propagation of aftershock

Aftershock propagation rate (m/sec)

1 3

18 m 16.88 s(1096.88 s) 22 m 27.21s(1347.21s)

I (between mainshock and aftershock 2)

444

1347

330

9

53 m 49.55s(3229.55s)

II (between aftershocks 2 and 9)

469

1882

250

13 14 24 30

1h 16 m 56.05 s (4616.05 s) 1h 17 m 06.33 s (4626.33 s) 1h 53 m 08.38 s (6788.38 s) 2h 09 m 50.76 s (7790.76 s)

III (between aftershocks 9 and 30)

386

4561

85

Table 3. Aftershock propagation rate along the West Andaman Fault. Aftershock No (see figure 1 and table 1) 1 2 5

Time after mainshock 18 m 16.88 s(1096.88 s) 23 m 32.14 s(1412.14 s) 31 m 22.29 s(1882.29 s)

Fault segment

Length (km) of the fault segment

IA (between aftershocks 1 and 5)

The main earthquake of 26th December has loaded the entire fault system in the region both in the subducting and overriding plates and transferred stress particularly to the West Andaman fault. Several large aftershocks locate along this fault system but rarely occur north of 10° latitude. Aftershock 1 from the mega-thrust is unilaterally followed northwards to shock 5 through aftershock 3 (green dots). This N-S segment (IA) is about 440 km long with an afterslip propagation rate of around 246 m/sec (Table 3).

Conclusion From the present analysis and our previous studies (Dasgupta et al., 2003; Mukhopadhyay et al., 2005) we make the following conclusions: (1) The fault rupture propagated unilaterally northwards from the mainshock epicentre in three distinct segments of 444, 469 and 386 km-length defined by seven aftershocks. (2) If distribution of aftershocks defines the fault rupture area then sequential and unilateral aftershock propagation rate should approximate the rate of rupture propagation. (3) This rate is much slower than the modeled rupture velocity of 2.0 to 2.5 km/sec and varies from 330 in segment I, 250 in segment II to 85 m/sec in segment III. (4) These seven unilaterally propagating shocks along the mega-thrust are probably not aftershocks (sensu stricto) of the mainshock, rather these are sequentially triggered shocks each rupturing a small segment of the fault. (5) The main earthquake on the mega-thrust has

440

Time (sec) for propagation of aftershock 1785

Aftershock propagation rate (m/sec) 246

loaded the West Andaman Fault wherein the aftershock propagation rate is around 246 m/sec. (6) Previous studies (Dasgupta et al., 2003) from the region indicate that the location of the mainshock and several aftershocks are guided by the presence of a number of lithospheric hinge faults (see Fig. 4 of Dasgupta et al., 2003). The epicentre of the earthquake of 26th December, 2004 locates close to the hinge of lithospheric fault f16. A cluster of aftershocks that locate SSW of Car Nicobar in fault segment II is traversed by the fault f12. The junction of fault segments II and III (shock 9) is marked by a similar hinge fault f11 in the subducting lithospheric slab. The very slow propagation rate (11 m/sec) across the 300 m patch between shocks 14 and 24 may be attributed to the barrier caused by the fault f10.

References Curray, J.R. (2005) Tectonics and history of the Andaman Sea region. J. Asian Earth Sci. (in press). Dasgupta, S., Mukhopadhyay, M., Bhattacharya, A. and Jana, T.K. (2003) The geometry of the Burmese-Andaman subducting lithosphere. J. Seismol., v. 7, pp. 155-174. Chen, J. (2005) http://www.gps.caltech.edu/~jichen/ Earthquake/2004/aceh/aceh.html Mukhopadhyay, B., Dasgupta, S. and Acharyya, A. (2005) Aftershock investigation of 26th December, 2004 earthquake. http://www.gsi.gov.in/suma_eq.htm Stein, S. and Okal, E.A. (2005) Speed and size of the Sumatra earthquake. Nature, v. 434, pp. 581-582. Yagi,Y. (2005), http://iisee.kenken.go.jp/staff/yagi/eq/ Sumatra2004/Sumatra2004.html Yamanaki, Y. (2005) http://www.eri.u-tokyo.ac.jp/sanchu/ Seismo_Note/2004/EIC161e.html

Gondwana Research, V. 8, No. 4, 2005