Tokyo, Japan National Graduate Institute for Policy Studies Tsukuba, Japan Building Research Institute

FOCAL MECHANISM DETERMINATION OF LOCAL EARTHQUAKES IN MALAY PENINSULA

A Master Paper Submitted in Partial Fulfillment of the Requirement For the Degree of Master in Disaster Management

By Siti Norbaizura MAT SAID (MEE10505 Disaster Management Policy Program2010-2011 Seismology

September 2011

Abstract Since November 30th, 2007, small local earthquakes have been observed in the Malay Peninsula near the Bukit Tinggi area. The total number of these events in the Malay Peninsula is 30 until the end of 2010, including the newly recorded small earthquakes in Jerantut, Manjung and Kuala Pilah, which occurred in March, April and November 2009, respectively. Although hypocenters and magnitudes are determined for these events using data from the Malaysian National Seismic network, focal mechanisms have not been determined. This paper provides results of determination of their focal mechanisms. We selected three crust structure models, and compared the observed travel time differences between P and S waves to those computed for these models. Although they agree relatively well for all of these models, the model iasp91 explained the observations better. We analyzed four events that occurred in the Bukit Tinggi area for focal mechanism determination. We used three-component broadband waveform data recorded at stations of IPM and KOM of the Malaysian National Seismic network. We determined their focal mechanisms using polarity data of the first motions of P and S waves and their amplitude ratios. We used iasp91 for take off angle calculations. We obtained relatively well-constrained solutions for all four events. The focal mechanisms of the largest 3.5mb event, which occurred at 02:13 UTC, November 30, 2007 is a mostly strike slip with some dip slip mechanism, while that of three events are strike slip fault mechanisms. The maximum compressional (P) axes of the 02:13 UTC, November 30, 2007 are in the NNW-SSE direction, while that of three events are in the NW-SE direction. The minimum compressional (T) axes of the strike slip events are in the NE-SW direction. Since there is no surface trace of ruptures, or any other type of surface movement related to this activity, this result is important to improve our understanding of these seismic activities. Keywords: focal mechanism, polarity, compressional The author works for the Malaysian Meteorological Department

Table of Contents ABSTRACT TABLE OF CONTENTS 1 . I N T R O D UC T I O N . 1 1.1 Background..1 1.2 Tectonic Setting and Seismic Pattern in the Malay Peninsula from GPS Point of View4 1.3 Importance of the Study...4 2. THEORY AND METHODOLOGY..5 2.1 Focal Mechanism Solution using Polarities and Amplitude Ratios..5 2.2 Seismic Velocity Model for Malay Peninsula6 3 . D ATA. . . 11 4. RESULTS AND DISCUSSION .1115 4.1 Bukit Tinggi, November 30 th , 2007 (02:13UTC) 3.5mb..15 4.2 Bukit Tinggi, November 30 th , 2007 (12:42UTC) 3.2mb..18 4.3 Bukit Tinggi, December 12 th , 2007 (10:01UTC) 3.2mb..20 4.4 Bukit Tinggi, January 10 t h , 2008 (15:38UTC) 3. 0mb..22 4.5 Comparison of the Orientation of P and T axes, and the Observed Polarities..24 4.6 Relationship between Focal Mechanism Obtained and the Existing Faults..26 5. CONCLUSION...27 ACKNOWLEDGMENT..28 REFERENCES.........28

1. INTRODUCTION 1.1 Background Since instrumental recordings initiated more than 30 years ago by the Malaysian Meteorological Department (MMD), the Malay Peninsula is considered as an inactive earthquake zone except for induced earthquake activities around Kenyir Dam, Terengganu recorded between 1984 and 1985, with magnitudes ranging from 2.4 to 4.6 on Richter scale. As these Kenyir Dam impoundment earthquakes were not due to tectonic movements, it is believed that these activities will not recur should the current dam configurations prevail. However, in November 2007 until January 2008, there were 13 tremors recorded in the Bukit Tinggi area, with magnitude up to 3.5. According to the Malaysian Academy of Sciences (ASM) and the Malaysian Meteorological Department (MMD) in their Final Report of Seismic and Tsunami and Risks Study in Malaysia released in 2009, these earthquakes may have occurred along the Bukit Tinggi Fault Zone, which is believed to be a strike slip fault, and to be related to stress release as a result of the Sumatran earthquakes.

Figure 1. Locations of local weak earthquakes in the Malay Peninsula (2007 2010) and the seismic stations currently operational via the Malaysian National Seismic Network. Yellow circle, blue triangle, and purple triangle are events, broadband stations, and short period stations, respectively
1

The Malay Peninsula seems to have its fair share of local earthquakes, and that is considered as weak when these Bukit Tinggi earthquakes continued occurring. As a result of these, the total number of events in the Malay Peninsula increased to 30 until the end of 2010, including the newly recorded small earthquakes in Jerantut, Manjung and Kuala Pilah, which have occurred in March, April and November 2009, respectively. The distributions of these earthquakes and the locations of the seismic stations currently operational in the Malay Peninsula (Table 2) are shown in Figure 1. Table 1 and Figure 2 both show the characteristics of these events.

Table 1. List of local weak earthquakes in the Malay Peninsula, 2007-2010 (MMD, 2010)

Year 2007

Month Nov Dec

No. EQ 3 6 4 3 1 1 1 6 4 1 3.3 3.4 N 3.3 3.5 N 3.3 3.4 N 3.3 3.4 N 3.4 N 3.9 N 4.2 3.3 3.4 N 2.7 2.8 N 3.2 N

Epicenter (Lat/Long/Location) 101-102 E 101-102 E 101-102 E 101-102 E 101.8 E 102.5 E 100.7 101-102 E 102 E 102 E Bukit Tinggi Bukit Tinggi Bukit Tinggi Bukit Tinggi Bukit Tinggi Jerantut Manjung Bukit Tinggi Kuala Pilah Bukit Tinggi

Depth (km) <7 <5 <3 Shallow Shallow 50 23 < 10 < 15 5

Magnitude (mb) 3.2 3.5 2.6 3.5 2.4 3.4 2.5 3.3 2.6 3.2 2.8 1 - 2.8 3.0 3.5 2.6 3.2 ml 2.2

2008

Jan Mar May

2009

Mar Apr Oct Nov Dec

2010

No felt earthquakes in the Malay Peninsula

Magnitude (mb)

90

180

270

360

450

540

630

720

810

Day

Figure 2. Distributions of local weak earthquakes in the Malay Peninsula since its first occurrence (November 30th, 2007)

Table 2. List of seismic stations in the Malay Peninsula currently operational via the Malaysian National Seismic Network
STATION CODE/ LOCATION KUM IPM FRM KTM KGM JRM KOM Kulim, Kedah Ipoh, Perak FRIM Kepong, Selangor Kuala Terengganu, Terengganu Kluang, Johor Jerantut, Pahang Kota Tinggi, Johor LAT ( N) 5.2902 4.4795 3.2333 5.3283 2.0157 3.8867 1.7922 LONG ( E) 100.6492 101.0255 101.6333 103.1356 103.3190 102.4767 103.8467 ELEVATION (m) 074 247 097 033 103 055 049 23/4/2009 24/3/2006 3/11/2003 COMMISSIONING DATE

1.2 Tectonic Setting and Seismic Pattern in the Malay Peninsula from GPS point of View According to the Malaysian Minerals and Geosciences Department (MMGD), the Malay Peninsula is divided into two tectonic blocks, amalgamated in Bentong suture and has several inactive ancient faults, namely Lebir fault, Baubak fault, Bukit Tinggi fault, Mersing fault and Kuala Lumpur fault by which this region is teared into several parts. The Global Positioning System (GPS) measurement (1994-1996) under the Geodynamics of South and Southeast Asia (GEODEYSSEA) project shows that the Sundaland tectonic block, which covers most of the countries in the Southeast Asia is relatively stable, with a moving rate of 12 3 mm/yr to the east-southeast direction towards the Eurasia (Michel et al. 2001) and was rotating clockwise. However, the Sumatera mega-thrust earthquake in 2004 and the following Nias earthquake in 2005, had caused significant co-seismic and post-seismic displacement into Sundaland which impacted up to 16 cm deformation to the northern part of the Malay Peninsula and has changed its motion to the west direction. Post seismic motions (2006-2008) of the GPS station at northern and southern part of the peninsula, as shown in the following figure do not show similarity in both magnitude and direction, and as such, it does not move as one entity. It means that the northern part of the peninsula moved towards the west-southwest whilst the southern part moved with the opposite direction, and this anti-clockwise motion reflects the reverse of pre-seismic motion of the peninsula. Omar (2009) discussed that the stations motion indicates the Malay Peninsula may have seismic activities that come from inside. From November 2007 until January 2008, there were 13 tremors recorded in Bukit Tinggi area but unfortunately, the motions of the stations are not so clear that the movement of the faults in that area could not be detected. Thus, it is necessary to combine researches on geodesy, geology, and geophysics to improve an understanding of the seismic activities in the Malay Peninsula. 1.3 Importance of the Study Although hypocenters and magnitudes for these local weak earthquakes are determined by using data from the Malaysian National Seismic Network, solutions of the focal mechanisms have not been determined. The purpose of this paper is to address such issue by illustrating the focal mechanism of these local earthquakes. Proper detailed pictures of seismic activities in the Malay Peninsula are needed to improve the understanding of its structure and deformation which can later be used for disaster mitigation and planning of high rise building. This is a worthwhile effort because of the location of the Bukit Tinggi fault, reasonably close to the very capital of Malaysia, Kuala Lumpur city center.

2. THEORY AND METHODOLOGY 2.1 Focal Mechanism Solution using Polarities and Amplitude Ratios When surface faulting due to an earthquake is observed, it is easy to identify the seismic fault. Due to the fact that most of earthquakes have fault ruptures that do not reach the surface, it is not directly visible to the eyes. Seismological approaches such as focal mechanism is important for studying a seismic fault. Focal mechanism provides information related to earthquake faulting, plate motion and tectonic stress which causes an earthquake. Its solution can be represented by a focal sphere diagram, known as beachball, which has two nodal planes, of which one should be the seismic fault responsible for an earthquake. For those events with surface ruptures to be compared, the validity of the fault plane solution is confirmed by their consistency. The focal mechanism of an earthquake can be obtained using polarity (compressional or dilatational) of the initial motion of the observed P-waves radiating from the hypocenter. The polarity of the P-waves varies systematically with the azimuth and distance from the hypocenter, and as for the systematic P-polarity pattern observed, it is suggested that most earthquakes can be explained by a double couple force system. An excellent station coverage with well determined earth structure and well recorded impulsive P-wave first arrivals will provide a reliable estimation of fault-plane solution. On the contrary, polarities of the first arrivals are often ambiguous if the station density is low or if the velocity structure is known only approximately. Obtaining a focal mechanism solution will be even more difficult in case of sparse local seismic network as the P-motion data may not be enough to determine reliable solutions and the uncertainties of the solutions are increased with decreasing number of stations. It is important to determine accurate focal mechanism solutions for earthquakes with small magnitudes or those recorded by a small number of seismometers, perhaps with an unfavorable geographic distribution. A method using the initial motion polarity of S waves, in addition to P waves, and their amplitude ratios is described. By combining all first motion data into one data set, we can improve the data coverage to resolve a reliable fault plane solutions. We used the FOCMEC software package (Snoke 1984) from the Incorporated Research Institutions for Seismology (IRIS) which can be downloaded from http://www.iris.edu/software/downloads/processing/. This software package uses polarities of P and S waves (P, SV, SH) and amplitude ratios (SV/P, SH/P, SV/SH) as the input set for determination and display of earthquake focal mechanisms. Its main program, Focmec which is coded in Fortran 77 performs an efficient, systematic search of the focal sphere and reports acceptable solutions based on selection criteria for the number of polarity errors and errors in amplitude ratios. Assumptions on number of errors are needed in order to find satisfying solutions especially for those imperfect data whose range of possible solutions can be constrained to within 10% of the P and S amplitudes on vertical-component seismograms. For three-component seismograph data, the SH to P amplitude ratio provides what could be considered almost independent data and polarities and ratios

involving SH are likely to be more reliable than those that involving SV. The Focmec program creates two output files, a complete summary of information about all acceptable solutions, and a summary file that can be used as an input to the plotting program, Focplt. Directions of polarities in the FOCMEC package are considered in the position of observer facing the station with his/her back to the epicenter. Accordingly, the first motion of P arrival is either up or down, while the SH is either left or right, and upward or backward for SV. Using the Seismic Analysis Code (SAC), horizontal components for three-component seismographs are being rotated into radial and transverse component to separate SV and SH. For crustal events recorded at local or regional distances, the first arrivals may be the refracted phases Pn and Sn and if the Poissons ratio is constant in the crust and uppermost mantle, Pn and Sn will leave from the same point on the focal sphere and have identical travel paths. When choosing polarities and amplitudes for local mechanism determinations we must consider the fact that an SP arrival that comes in just before direct S affects the observed waveforms, particularly for SV incidence and also a low-velocity zone at the surface can complicate the waveforms (Booth and Crampin 1985). 2.2 Seismic Velocity Model for Malay Peninsula The focal mechanism maybe dependent on the choice of a velocity model since the take off angle, at which the ray leaves the source, for local earthquake studies needs an appropriate assumed velocity model. This study uses a crustal model determined by the CRUST2.0, specified on a 2x2 degree grid which takes advantage of a recent compilation of global sediment thickness on a 1x1 degree (Bassin 2000). It is available at http://igppweb.ucsd.edu/~gabi/crust2.html . This crustal model is composed of 360 key 1d-profiles and each profile has a 7-layer 1D-model (ice, water, soft sediments, hard sediments, upper crust, middle crust and lower crust). The crustal profiles for the Malay Peninsula available from a CRUST2.0 model, is shown down to the depth of 50 km (Figure 3). The crust of profile J1 consists of four uniform layers (soft sediments, upper crust, middle crust and lower crust) with Moho discontinuity at 31 km depth, while the crust of profile C2 consists of five uniform layers (water, soft sediments, upper crust, middle crust and lower crust) with Moho discontinuity at 30 km depth. Figure 4 shows a comparison for velocity structure between profiles of a CRUST2.0 model and reference model iasp91 (Kennet 1991), down to the depth of 50 km. The crust of iasp91 consists of two uniform layers with discontinuities at 20 and 35 km depth. The Moho depth differs by about 4 km from those of CRUST2.0 models, while it is close to that around 35 km, beneath IPM station (Din 2011). We checked the effect of the variety of velocity models available for Malay Peninsula on the travel-time difference of P and S-wave velocity (S-P). The calculation was performed using the Flexible Seismic Travel-Time and Raypath Utilities software package, or the TauP Toolkit (Crotwell 1999). The algorithms employed within the TauP package are based on the method of Buland and Chapman (1983). We compared the S-P travel-time from observation of nine events shown in Table 3

(using data from IPM station) with theoretical (calculated) S-P travel-time using three velocity models, J1, C2 and iasp91 (Figure 5). For determination of focal mechanism we avoided using any data with large residuals between the observed and the calculated ones. We also compared the P-phase arrivals picked up by the MMD with the theoretical P-phase arrivals computed by using three velocity models of J1 and C2 and iasp91 for events shown in Table 4. The result for the comparison, on the P-phase arrival, between the observed and the theoretical is shown by Figure 6. The root mean square (RMS) for both comparisons of S-P travel-time and P-phase arrival were calculated using Eq.(1) and the results are shown in Table 5.

RMS =

n i=1

i i n

(1)

where i is the i-th observed value, i is the i-th calculated value and n is the total number of values.

Figure 3. Comparison of P-wave velocity (), S-wave velocity (), and density () between crustal profiles of J1 and C2, from CRUST2.0

Figure 4. Comparison of P-wave velocity () and S-wave velocity () among three velocity models, J1 and C2 from CRUST2.0 and standard earths model iasp91

Table 3. Events considered in the comparison of the observed and the theoretical (calculated) S-P using data from IPM station
TIME (UTC) EPICENTER LAT 2009 2009 2008 2008 2007 2007 2007 2007 2007 Mac Apr Jan Mar Nov Nov Dec Dec Dec 27 29 10 15 30 30 4 9 12 1:46:25 13:53:55 15:38:00 0:50:00 2:13:00 12:42:00 10:12:00 12:55:00 10:01:00 3.9N 4.2N LONG 102.5E 100.7E 50 22 3.0 33 2.3 6.7 10 4.9 10 3.2 mb 2.8mb 3.0 mb 3.3 mb 3.5 mb 3.2 mb 3.0 mb 3.5 mb 3.2 mb DEPTH MAGNITUDE (KM) S-P TRAVEL-TIME COMPARISON OBSERVED 25.6 6.36 21.08 20.91 20.35 19.78 21.37 20.89 20.37 J1 19.9 6.79 19.24 17.45 19.33 18.92 18.68 19.08 18.68 C2 19.75 6.84 19.26 17.44 19.33 18.93 18.69 19.09 18.69 iasp91 19.79 6.5 19.41 17.53 19.21 19.25 19.01 19.39 19.01

YEAR

MONTH

DAY

3.39N 101.73E 3.33N 101.71E 3.36N 101.80E 3.31N 101.84E 3.36N 101.81E 3.33N 101.82E 3.47N 101.79E

Table 4. Events considered in the comparison of the picked up (by the MMD) and the theoretical (calculated) P-phase arrival using data from IPM station
ORIGIN TIME (UTC) YEAR MONTH DAY h 2009 2009 2009 2009 2009 2009 2009 3 4 10 10 10 10 11 27 29 7 7 7 7 29 01 13 21 21 22 22 16 m 46 53 21 26 09 20 15 s 25.1 54.8 26.0 06.0 47.1 59.7 05.2 LAT 3.9 4.1 3.3 3.4 3.3 3.3 2.7 EPICENTER DEPTH MAGNITUDE (KM) LONG Mb Ml 102.5 100.7 101.8 101.9 101.8 101.8 102.1 50 22 2 1 2 2 3 3.2 2.8 1.7 3.2 3.3 2.5 1.7 1.0 0.3 2.9 P-PHASE (MMD Picking) h 01 13 21 21 22 22 16 m 46 54 21 26 10 21 15 s 51.3 04.0 52.5 32.7 11.4 08.4 40.9 P 26.20 9.20 26.47 26.66 24.30 8.70 35.74 THEORETICAL P-PHASE J1 24.75 8.55 24.26 25.95 24.26 24.26 33.43 C2 25.54 9.27 24.94 27.12 24.94 24.94 34.10 iasp91 25.57 8.96 26.12 27.20 26.12 26.12 35.35

Table 5. The root mean square error (between observed and calculated) of S-P travel-time and P-phase arrival using three velocity model, J1 and C2 and iasp91 RMS J1 S-P travel-time P-phase arrival all excluding large residuals (>3s) 2.65 1.44 1.49 Velocity Model C2 2.68 1.44 1.01 iasp91 2.57 1.25 0.85

Although the differences noted in the RMS among three velocity models, J1, C2 and iasp91 are not so large, the lowest values of RMS are consistently given by the model of iasp91. As for S-P travel-time, the RMS values are much improved when we excluded two events having large residuals (more than 3 seconds of difference between the observed and the theoretical). These results suggest that the iasp91 model predicts the behavior of the seismic velocity in the Malay Peninsula better than the model from CRUST2.0. This is also consistent with results obtained by Din (2011) in which crustal thickness beneath IPM station is similar to the thickness of iasp91 (35km). Thus we used a velocity model from iasp91 for the calculation of a take off angle and for epicentral distances within the Moho headwave, the calculation of take off angle will be robust since the crustal waves travelling along the Moho discontinuity have the same incident angle.

Figure 5. Comparison of the observed and the theoretical (calculated) S-P using data from IPM station
10

Figure 6. Comparison of the observed and the theoretical (calculated) P-phase arrival using data from IPM station

3. DATA We used polarity and amplitude ratio from three-component broadband data from two stations, IPM and KOM for which the locations of stations and the distributions of the events are shown in Figure 7. Data were requested from the BREQ_FAST Request Manual (the batch access to the Incorporated Research Institutions for Seismology, Data Management Center, IRIS DMC archive) by electronically mailing a specially formatted file to breq_fast@iris.washington.edu. The DMC returned the seed data which can be downloaded via ftp://ftp.iris.washington.edu/pub/userdata/ and we used the Seismic Analysis Code (SAC), available at http://www.iris.edu/forms/sac_request.htm in analyzing the data. The vertical and horizontal (E-W) components from IPM station for the November 30th 2007 Bukit Tinggi event are shown in Figure 8, in which we performed a square of velocity time series and integration on the squared time series for both vertical and horizontal components, as to observe the onset of the P and S. We then compared the travel-time of the P and S-waves with the theoretical travel-time and rotated the horizontal component of the three-component seismogram as shown in Figure 9 to separate the SV and SH. We determined the polarity of the first (initial) motion of P, SV and SH and the amplitude for P and SH and made correction on the dissipation of the seismic velocity

11

based on the Eq.(2). exp [(/2)] where is the angular frequency, t is the phase time arrival and Q is the attenuation factor. (2)

Figure 7. Locations of broadband stations (IPM and KOM) and events used for determination of focal mechanism. Inverted triangles are stations and circles are events of Bukit Tinggi

(a)

(b)

Figure 8. Time series of vertical channel, (a) and horizontal channel, (b) of IPM station for the Bukit Tinggi event, occurred at 02:13UTC, November 30, 2007, with magnitude of 3.5 mb and depth of 2.3 km. The upper trace is the velocity seismogram, the middle trace is squared velocity and the lower trace is an integration of the middle time series
12

EW

NS

Radial

Transverse

Figure 9. Velocity of three-component seismogram data of IPM station for the Bukit Tinggi event, occurred at 02:13UTC, November 30, 2007, with magnitude of 3.5 mb and depth of 2.3 km. The horizontal component of the first three time series (EW and NS) were rotated into radial and transverse components in order to separate the SV and the SH

The polarities in the FOCMEC package are designated by the directions with respect to an observer facing the station with his/her back to the epicenter. Accordingly, the SAC convention has positive radial as forward and positive transverse as to right with respect to an observer facing the station. The following figures show the component for the P and the S arrivals at IPM station for the 02:13UTC, November 30, 2007 Bukit Tinggi event. The basis for the polarity choices for IPM station Focmec input file is D (down on vertical) for P arrival (Figure 10), and as for S arrival, F (up on radial and vertical) for SV, and L (down on transverse) for SH (Figure 11).

13

Coordinate Rotation

East-West North-South Radial Transverse Vertical Figure 10. Components for P arrivals at IPM station for the 02:13UTC, November 30, 2007 Bukit Tinggi event. The back azimuth is 145.18, so the direct arrivals come from the southeast

Vertical

Radial

Transverse

Figure 11. Components for S arrivals at IPM station for the 02:13UTC, November 30, 2007 Bukit Tinggi event. Positive for the radial component is towards the northwest and positive for the transverse component is towards the northeast

14

4. RESULTS AND DISCUSSION We analyzed four events occurred near the Bukit Tinggi area which having small RMS (Table 6) and determined their focal mechanisms. The Focmec program output all acceptable solutions (possible mechanisms) based on input parameters (station identifiers, azimuths, take off angles at the source, polarities and amplitude ratios from among P, SV, and SH arrivals). As for selection criteria, no error is allowed. The maximum number of solutions is set to 100,000. We allowed 0.1 for the maximum log10 of ratio, and selected 0.01 and 0.02 for lower bound (cutoff value) of P and S radiation factor, respectively. The ratio was indeterminate when both calculated values were less than the chosen cutoff values. Table 6. List of the Bukit Tinggi events analyzed for focal mechanism determination YEAR MONTH DAY TIME (UTC) 2007 2007 2007 2008 Nov Nov Dec Jan 30 30 12 10 02:13 12:42 10:01 15:38 EPICENTER LAT 3.36N 3.31N 3.47N 3.39N LONG 101.80E 101.84E 101.79E 101.73E DEPTH MAGNITUDE (KM) 2.3 6.7 10 3.0 3.5 mb 3.2 mb 3.2 mb 3.0 mb

4.1 Bukit Tinggi, November 30, 2007 (02:13 UTC) 3.5 mb We obtained a relatively well-constrained solution for the largest 3.5mb event, which occurred at 02:13 UTC, November 30, 2007. Table 7 shows the input parameters determination of the focal mechanism and Figure 12 shows the focal sphere projections of the lower hemisphere for 58 possible solutions returned by the Focmec program. The focal mechanism of this event is a mostly strike slip with some dip slip and the maximum compressional (P) axes are in the NNW-SSE direction. Figure 13 shows the distributions of the dip, strike, and rake angles for two fault planes in which, for the first fault plane, they fall in the range of 45 to 75, 10 to 60, and 20 to 50 respectively, while for the second fault plane they fall in the range of 50 to 70, 270 to 310, and 130 to 170, respectively.

15

Table 7. Input parameters for the November 30, 2007, 02:13 UTC Bukit Tinggi event. D = Dilatational, F = Away from event, L = Left and H = Log10 (SH/P) Station IPM IPM IPM IPM KOM KOM KOM KOM Azimuth 325.23 325.23 325.23 325.23 127.25 127.25 127.25 127.25 Take-off angle 46.17 46.17 46.17 46.17 46.17 46.17 46.17 46.17 D F L H D F L H 0.77248 0.85281 Key Log10(S/P)

Figure 12. Lower-hemisphere projections of the focal sphere for the November 30, 2007, 02:13 UTC Bukit Tinggi event . P is maximum compressional axis and T is minimum compressional axis

16

FAULT PLANE 1

FAULT PLANE 2

(a)

(b)

(c)

Figure 13. Histogram for the distributions of focal mechanism parameters of the November 30, 2007, 02:13 UTC Bukit Tinggi event for two fault planes, (a) dip angles, (b) strike angles, and (c) rake angles

17

4.2 Bukit Tinggi, November 30, 2007 (12:42 UTC) 3.2 mb The solutions of the November 30, 2007, 12:42 UTC Bukit Tinggi event are well-constrained in which we obtained a strike slip faulting, with the P axes distributions in the NW-SE direction and the T axes distributions in the NE-SW direction. Table 8 and Figure 14 show the input parameters to infer the focal mechanism and the focal sphere projections of the lower hemisphere for 10 possible solutions returned by the Focmec program, respectively. The dip and rake angles of the first plane fall within the range of 72 to 85 and of -40 to 0, respectively while for the second plane, the dip and rake angles fall within the range of 58 to 82 and -180 to -160. The strike angles, falling in two ranges, between 170 and 190, and between 340 and 360 for the first fault plane, and between 80 and 100, and between 270 and 280 for the second fault plane, as shown in Figure 15.

Table 8. Input parameters for the November 30, 2007, 12:42 UTC Bukit Tinggi event, D = Dilatational, F = Away from event, R = Right, L = Left and H = Log10 (SH/P) Station Azimuth IPM IPM IPM IPM KOM KOM KOM 325.05 325.05 325.05 325.05 126.9 126.9 126.9 Take-off angle 46.17 46.17 46.17 46.17 46.17 46.17 46.17 D F R H D L H 0.5800 0.8367 Key Log10(S/P)

Figure 14. Lower-hemisphere projections of the focal sphere for the November 30, 2007, 12:42 UTC Bukit Tinggi event . P is maximum compressional axis and T is minimum compressional axis 18

FAULT PLANE 1

FAULT PLANE 2

(a)

(b)

(c)

Figure 15. Histogram for the distributions of focal mechanism parameters of the November 30, 2007, 12:42 UTC Bukit Tinggi event for two fault planes, (a) dip angles, (b) strike angles, and (c) rake angles

19

4.3 Bukit Tinggi, December 12, 2007 (10:01 UTC) 3.2 mb As for the Bukit Tinggi event occurred at 10:01 UTC, December 12, 2007, the results are close with those of Bukit Tinggi, 12:42 UTC, November 30, 2007 event in that the focal mechanism of this event is strike slip. P axes distributions are in the NW-SE direction and the T axes distributions are in the NE-SW direction. Table 9 shows the input parameters to infer the focal mechanism and Figure 16 shows the focal sphere projections of the lower hemisphere for 6 possible solutions returned by the Focmec program. From the dip, strike, and rake angles distributions for two fault planes as shown in Figure 17, we obtained a range of 70 to 82 for dip angles, and -40 to -10 for rake angles while strike angles fall in two ranges, of 160 to 180 and of 330 to 360 for the first fault plane and 55 to 80 for dip angles, -180 to -150 for rake angles, while strike angles fall in two ranges, of 70 to 90 and 260 to 280 for the second fault plane. Table 9. Input parameters for the December 12, 2007, 10:01 UTC Bukit Tinggi event. D = Dilatational, F = Away from event, R = Right, and H = Log10 (SH/P) Station Azimuth Take-off angle IPM IPM IPM IPM KOM KOM KOM 322.8 322.8 322.8 322.8 129.0 129.0 129.0 46.2 46.2 46.2 46.2 46.2 46.2 46.2 D F R H D R H 0.4270SH 0.7619 SH Key Log10 (S/P)

Figure 16. Lower-hemisphere projections of the focal sphere for the December 12, 2007, 10:01 UTC Bukit Tinggi event . P is maximum compressional axis and T is minimum compressional axis
20

FAULT PLANE 1

FAULT PLANE 2

(a)

(b)

(c)

Figure 17. Histogram for the distributions of focal mechanism parameters of the December 12, 2007, 10:01 UTC Bukit Tinggi event for two fault planes, (a) dip angles, (b) strike angles, and (c) rake angles

21

4.4 Bukit Tinggi, January 10, 2008 (15:38 UTC) 3.0mb The focal mechanism of the January 10, 2008, 15:38 UTC Bukit Tinggi event shows similar results with the previous two Bukit Tinggi events (the 12:42 UTC, November 30, 2007, and the 10:01 UTC, December 12, 2007 Bukit Tinggi event) in which the obtained focal mechanism is of strike slip faulting with P and T axes distributed in the NW-SE and NE-SW, respectively. Table 10 and Figure 18 show the input parameters for the focal mechanism determination and the focal sphere projections of the lower-hemisphere for 2 possible solutions returned by the Focmec program, respectively. From Figure 19 we obtained a range of 80 to 85, 0 to 10 and 170 to 180, and -20 to -10 for the distributions of dip, strike, and rake angles, respectively, for the first fault plane. As for the second fault plane, the dip and the rake fall in the range of 75 to 80 and -180 to -170, respectively, while the strike fall in two ranges, 90 to 100 and 260 to 270.

Table 10. Input parameters for the January 10, 2008, 15:38 UTC Bukit Tinggi event. D = Dilatational, R = Right, L= Left, and H = Log10 (SH/P) Station Azimuth
IPM IPM IPM KOM KOM KOM

Take-off angle 46.17 46.17 46.17 46.17 46.17 46.17

Key D R H D L H

Log10(S/P)

322.02 322.02 322.02 126.84 126.84 126.84

0.8073

1.0125

Figure 18. Lower-hemisphere projections of the focal sphere for the January 10, 2008, 15:38 UTC Bukit Tinggi event. P is maximum compressional axis and T is minimum compressional axis
22

FAULT PLANE 1

FAULT PLANE 2

(a)

(b)

(c)

Figure 19. Histogram for the distributions of focal mechanism parameters of the January 10, 2008, 15:38 UTC Bukit Tinggi event for two fault planes, (a) dip angles, (b) strike angles, and (c) rake angles

23

4.5 Comparison of the Orientation of P and T axes, and the Observed Polarities Figure 20 shows the distributions of the P and T axes of these four Bukit Tinggi events. The P axes of the largest 3.5mb Bukit Tinggi event are in the NNW-SSE direction while the three strike slip events are consistently in the NW-SE direction. The T axes for the three strike slip events are also consistent, for which in the NE-SW direction. We also compared the polarity of the first motion of P-phase for five Bukit Tinggi events, of which three are events described earlier while two others are events occurred on December 9, 2007 and March 15, 2008. The consistent dilatational first motion of P waves for all events recorded at IPM station are shown by the seismograms in Figure 21. The similar waveforms imply that the mechanisms of the events are very similar and that the observed polarity at a given station should be the same for each event in cluster (Hardebeck 2002).

02:13 UTC November 30, 2007 3.5 mb

12:42 UTC November 30, 2007 3.2 mb

10:01 UTC December 12, 2007 3.2 mb

15:38 UTC January 10, 2008 3.0 mb

Figure 20. Focal sphere projections of lower-hemisphere (using Focplt program) for all acceptable solutions of four Bukit Tinggi events. P, T and B are the maximum compressional, minimum compressional and null axes, respectively
24

(a)

(b)

(c)

(d)

(e) Figure 21. The polarity of the first motion of P-waves observed at IPM station for the following Bukit Tinggi events, (a) the November 30, 2007, 02:13 UTC, (b) the November 30, 2007, 12:42 UTC, (c) the January 10, 2008, 15:38 UTC, (d) the December 9, 2007, 12:55 UTC, and (e) the March 15, 2008, 00:50 UTC

25

4.6 Relationship between Focal Mechanism Obtained and the Existing Faults We show the epicenter of the four events that we analysed, their focal mechanisms and the faults in the study area, namely the Bukit Tinggi fault and the Kuala Lumpur fault in Figure 22. Comparing the fault orientations and the strikes of the nodal planes, the relationship between these two faults and the focal mechanisms obtained is not clear. Shuib (2009) discussed that the relationship between earthquakes and geologically mapped faults in the Bukit Tinggi area cannot be proven due to no surface ruptures observed. The difference in the movement rate between the west and the east based on the GPS data observed in the Southeast Asia within a ten-year span (between 27 November 1994 and 25 December 2004 ), may cause some pushing that could release energy to trigger movement on any weak plane, such as in the Bukit Tinggi fault zone (Lat 2009). The stress regime for this area indicates that the force in play is compressional in the NE-SW direction (Simmons et al. 2007). This is contradict with the orientation of compressional axes obtained from the determination of focal mechanisms, in which in the NW-SE and NNW-SSE. One possible explanation to these Bukit Tinggi earthquakes is that they are due to the weak-zone-normal extension mechanisms (Hurukawa and Imoto 1992).

Figure 22. Map showing the locations of four Bukit Tinggi events (yellow circles) and the respective focal mechanisms obtained by using polarities and amplitude ratios of body wave arrivals from data recorded at stations of IPM and KOM. Also shown are the existing ancient and inactive faults, namely the Bukit Tinggi fault and the Kuala Lumpur fault

26

Fatt et al. (2011) conducted a relocation study of local earthquakes in the Malay Peninsula and showed that hypocenters of these events are better improved by using a method of modified joint hypocenter determination (MJHD) developed by Hurukawa and Imoto (1990), and HYPOCENTER program (Lienert and Havskov 1995). The MJHD program shows that epicenters are scattered along the Bukit Tinggi fault (striking NW-SE directions and dipping toward NE ) and Kuala Lumpur fault (striking NW-SE directions and dipping toward SW) with depth between 6 and 30 km. The HYPOCENTER program shows that the epicenters are scattered at the Bukit Tinggi fault with depth between 0 to 10 km with similar orientation of nodal plane to the MJHD program. Their study however, did not relocate the four Bukit Tinggi events that we analysed. Thus, relocation of these earthquakes is important to improve the accuracy of the results of this study.

5. CONCLUSION We determined focal mechanisms for four Bukit Tinggi events that occurred in 2007 and 2008 using the polarity data of the first motions of P and S waves, and their amplitude ratios from three-component broadband waveform data recorded at stations of IPM and KOM. Although constraint for dip angles is weaker for two Bukit Tinggi events on November 30, 2007 (12:42UTC) and on December 12, 2007 (10:01UTC), we obtained relatively well-constrained solutions for these four events; three events have strike slip faulting while one (and the largest) has a mostly strike slip with some dip slip. The T axes of the strike slip events are in the NE-SW direction while their maximum P axes are in the NW-SE direction. The P axes of the largest event are in the NNW-SSE direction. The consistency of the first motion polarity of P-waves observed for all Bukit Tinggi events together with relatively well-constrained P and T axes direction, suggest that these earthquakes might have similar mechanisms. This fits the characteristics of cluster earthquakes or earthquake swarms (Lat 2009). The validity of the relative motion of the strike slip mechanism, unfortunately, cannot be confirmed geologically since these Bukit Tinggi earthquakes share significant similarities with other intraplate earthquakes in that there are lack of surface ruptures (Shuib 2009). Due to an absence of this surface movement, the result of this study serves for an important basis in ascertaining the focal mechanisms of the earthquakes near Bukit Tinggi area, and will improve an understanding of seismic activities in the Malay Peninsula.

27

ACKNOWLEDGEMENT I wish to thank the International Institute of Seismology and Earthquake Engineering of the Building Research Institute (IISEE, BRI) and the Japan International Cooperation Agency (JICA) for funding and making this study possible. My heartiest gratitude goes to my supervisor Dr. T. Hara, who for almost one year made sure that i received most comprehensive guides and supports. I would also like to thank Dr. Hurukawa and Dr. Yokoi for their constructive comments that improved the clarity of this manuscript. Thanks are also due to the Director, Geophysics and Tsunami Division of the Malaysian Meteorological Department, Mr. Rosaidi Che Abas for permitting me to use data of the Malaysian National Seismic Network and to colleagues in the Malaysian Meteorological Department especially Zamuna Zainal, Hasliza Halim and Azrin Irwan, for providing useful suggestions and assistances. My profound respect and heartfelt appreciation goes to all staff and students in the IISEE, BRI for their accomplishments and collective efforts towards realizing the one year course study, through hardship following the Great Tohoku Earthquake on March 11. May this study reaches the readers, with a level of quality adequate to its purpose.

REFERENCES Aki, K., and Richards, P.G., 2002, Quantitative Seismology Second Edition, University Sciences Books Bassin, C., Laske, G. and Masters, G., 2000, The current limits of the resolution for surface wave tomography in North America, EOS Trans Agu, 81, F897 Booth, D.C. and Crampin, S., 1985, Shear-wave polarizations on a curved wavefront at an isotropic free surface, J. Geophys. Res., astr. Soc., 83, 31-35 Buland, R. and Chapman C.H., 1983, The Computation of Seismic Travel Times, Bull. Seis. Soc. Am., 73, 1271-1302 Crotwell, H.P., Owens, T.J. and Ritsema, J., 1999, The TauP Toolkit : Flexible Seismic Travel-time and Ray-Path Utilities, Seis. Res. Lett, 70, 154-169 Hardebeck, J. L. and Shearer, P. M., 2002, A new method for determining first-motion focal mechanisms, Bull. Seis. Soc. Am., 92, 2264-2276 Hurukawa, N. and Imoto, M., 1990, Fine structure of an underground boundary between the Philippine Sea and the Pacific plates beneath the Kanto district, Japan, J. Seis. Soc. Japan, 43, 413-429 Hurukawa, N. and Imoto, M., 1992, Subducting oceanic crusts of the Philippine Sea and Pacicif plates and weak-zone-normal compression in the Kanto district, Japan, Geophys. J. Int., 109, 639-652 Kee, T.C., Omar, K., Simons, W.F.J., Abu Samad, 2005, GPS for regional crustal deformation measurement: case study of Aceh earthquake 2004 and Nias earthquake 2005, South East Asia Survey Congress

28

Kennet, B. L. N., and Engdahl, E. R., 1991, Traveltimes for global earthquake location and phase identification, Geophys. J. Int., 105, 429-465 Kisslinger, C., Bowman, J. R. and Koch, K., Determination of focal mechanism from SV/P amplitude ratios at small distances, Phys. Earth Planet. Inter., 30, 172-176 Lat, C.N., 2007, Bukit Tinggi earthquakes: November 2007-January 2008, Geological Society of Malaysia, 55, 81-86 Lienert, B., R. and Havskov, 1995, A computer program for locating earthquakes both locally and globally, Seis. Res. Lett., 66, 26-36 Michel, G. W., Yu, Y. Q., Zhu, S. Y., Reigber, C., Becker, M., Reinhart, E., Simons,W., Ambrosius, B., Vigny, C., Chanmot-Rooke, N., Pichon, X. L., Morgan, P. and Matheussen, S., 2001, Crustal motion and block behaviour in SE Asia from GPS measurements, Earth Planet. Sci. Lett., 187, 239 244 Shuib, M.K., 2009, The recent Bukit Tinggi earthquakes and their relationship to major geological structures, Geological Society of Malaysia, 55, 67-72 Simons, W. J. F., Socquet, A., Vigny, C., Ambrosius, B. A., Abu, S. H., Promthong, C., Subarya, C., Sarsito, D. A., Matheussen, S., Morgan, P. and Spakman, W., 2007, A decade of GPS in Southeast Asia : Resolving Sundaland motion and boundaries, J. Geophys. Res., 112, 06420 Snoke, J.A., Munsey, J.W., Teague, A.C., and Bollinger, G.A, 1984, A program for focal mechanism determination by combined use of polarity and SV-P amplitude ratio data, Earthquake Notes, 55, 3, 15 Web site: CRUST2.0; http://igppweb.ucsd.edu/~gabi/crust2.html Web site: IRIS DMC Software Processing; http://www.iris.edu/software/downloads/processing/ Website: Omar, K. and Jhonny, 2009, Crustal Deformation Study in Peninsular Malaysia using GPS; http://www.fksg.utm.my/research_group/gng/publications/2009/10.pdf Website: Fatt, C.M., Zainal, Z., Ramachandran, D., Mokhtar, Z.A., Wahab, A.A. and Abas, M.R.C., 2011, Study on hypocenter relocation of the local earthquakes in Malay Peninsula using the Modified Joint Hypocenter Determination and HYPOCENTER Programs; http://www.met.gov.my/images/pdf/research/researchpaper_201102.pdf Web site : Seismic Analysis Code (SAC); http://www.iris.edu/forms/sac_request.htm Din, Z. A., 2011, Determination of crustal structure beneath broadband station in Malaysia using receiver function analysis, Bulletin of the IISEE, 45, 19-24

29