Geophysical Journal InternationalGeophys. J. Int. (2016) 204, 918–931 doi: 10.1093/gji/ggv466

GJI Seismology

Imaging architecture of the Jakarta Basin, Indonesiawith transdimensional inversion of seismic noise

E. Saygin,1 P.R. Cummins,1 A. Cipta,1 R. Hawkins,1 R. Pandhu,2 J. Murjaya,2

Masturyono,2 M. Irsyam,3 S. Widiyantoro4 and B.L.N. Kennett11Research School of Earth Sciences, The Australian National University, Canberra ACT 2601, Australia. E-mail: erdinc.saygin@anu.edu.au2Badan Meteorologi, Klimatologi, dan Geofisika, Jakarta, Indonesia3Department of Civil Engineering, Bandung Institute of Technology, Bandung, Indonesia4Global Geophysics Research Group, Faculty of Mining and Petroleum Engineering, Bandung Institute of Technology, Bandung, 40132, Indonesia

Accepted 2015 October 23. Received 2015 October 5; in original form 2015 July 10

S U M M A R YIn order to characterize the subsurface structure of the Jakarta Basin, Indonesia, a denseportable seismic broad-band network was operated by The Australian National University(ANU) and the Indonesian Agency for Meteorology, Climatology and Geophysics (BMKG)between October 2013 and February 2014. Overall 96 locations were sampled through suc-cessive deployments of 52 seismic broad-band sensors at different parts of the city. Oceanicand anthropogenic noises were recorded as well as regional and teleseismic earthquakes. Weapply regularized deconvolution to the recorded ambient noise of the vertical components ofavailable station pairs, and over 3000 Green’s functions were retrieved in total. Waveformsfrom interstation deconvolutions show clear arrivals of Rayleigh fundamental and higher ordermodes. The traveltimes that were extracted from group velocity filtering of fundamental modeRayleigh wave arrivals, are used in a 2-stage Transdimensional Bayesian method to map shearwave structure of subsurface. The images of S wave speed show very low velocities and a thickbasin covering most of the city with depths up to 1.5 km. These low seismic velocities andthe thick basin beneath the city potentially cause seismic amplification during a subductionmegathrust or other large earthquake close to the city of Jakarta.

Key words: Interferometry; Surface waves and free oscillations; Site effects; Seismic to-mography.

I N T RO D U C T I O N

Jakarta, Indonesia, is one of the world’s most densely populatedcities, with a population exceeding 10 million living in an area ofabout 600 km2 (greater Jakarta has a population of 28 million).Jakarta lies on a sedimentary basin that experiences one of themost rapid rates of subsidence in the world, over 26 cm yr−1,mainly due to groundwater extraction (Abidin et al. 2011; Nget al. 2012). Although there are no known active faults withinthe city itself (Harsolumakso 2001), its proximity to the JavaTrench 200 km to the south, where the Australian Plate subductsbeneath Java, suggests that the seismic hazard is high. In 1985,Mexico City sustained considerable casualties and damage due tothe Michoacan earthquake (Mw = 8.1) that occurred in the Mexi-can subduction zone over 300 km away. The shallow sedimentarylayers and basin architecture of the Mexico City Valley had a crit-ical influence on amplification and increased duration of seismicwaves that contributed to the heavy loss of lives and infrastruc-ture (Anderson et al. 1986; Kawase & Aki 1989; Furumura &Kennett 1998).

In urban basins such as Kanto, Los Angeles, Osaka, San Fran-cisco, and Seattle, a variety of data sets and methods are used toconstrain 3-D seismic velocity structure, in order to predict thelevel of strong ground motion and seismic hazard (Kagawa et al.2004; Brocher 2005; Koketsu et al. 2009; Delorey & Vidale 2011;Lee et al. 2014). These methods typically include borehole, gravity,seismic tomography, seismic refraction, and microtremor studies.In contrast, apart from sparse microtremor observations (Ridwanet al. 2014, 2015) the Jakarta basin has been the subject of veryfew geophysical or borehole measurements—and to our knowledgefew, if any, boreholes in the Jakarta basin have reached bedrock.

In this study, we use data from a temporary seismograph deploy-ment in the Jakarta basin in an attempt to characterize the shallowseismic velocity structure of the basin and surrounding area. InOctober 2013, 52 broad-band seismometers were installed acrossJakarta at public and private high-schools, and recording was con-ducted continuously until February 2014. During this time, half thestations were maintained at their original locations and half wereredeployed in three phases (each with one month duration) to coverthe northeastern, northwestern, and southern parts of the city with

918 C© The Authors 2015. Published by Oxford University Press on behalf of The Royal Astronomical Society.

at The A

ustralian National U

niversity on February 21, 2016http://gji.oxfordjournals.org/

Dow

nloaded from

Imaging architecture of Jakarta Basin, Indonesia 919

Figure 1. Map of the seismic network operated between October 2013 and February 2014 overlaid on the geological units of Jakarta (Turkandi et al. 1992).The approximate location of the experiment is shown with a red square on the world map (right) with plate boundaries and direction vectors. The administrativeboundary of Jakarta is shown with grey lines.

a minimum interstation spacing of 1–5 km. Overall, continuousseismic measurements were conducted at 96 different locations. InFig. 1, the distribution of the seismic stations is superposed on amap of the surface geology in the Jakarta area.

G E O L O G Y A N D H I S T O R I C A LE A RT H Q UA K E S

Jakarta is located on the northern coast of the island of Java, In-donesia, which together with Sumatra to the west and Bali, Lombok,and Sumba to the east, comprises the southern edge of the SundaBlock. South of Java, about 200 km from Jakarta, the AustralianPlate subducts underneath the Sunda Block at the Java Trench, witha convergence rate of 67 mm yr−1 (Simons et al. 2007). Arc volcan-ism is prevalent in Java, where a number of active volcanoes exist tothe south of the city (Hamilton 1979). The Java Trench megathrust,the subducting slab beneath Java, and at least two possibly activecrustal faults within 100 km of Jakarta—the Cimandiri Fault to thesouth and Lembang Fault to the southeast—are all potential sourcesof earthquakes that might affect Jakarta.

Jakarta has not been subject to strong earthquake ground motionsince the explosive growth in population in the late 20th century—the population of greater Jakarta grew from about 0.8 million in1948 to over 28 million today (see, e.g. Cybriwsky & Ford 2001). Ithas, however, experienced large earthquakes historically. The mostsignificant of these occurred on 1699 January 5, when Dutch Batavia(the former name of Jakarta) experienced ‘an earthquake so heavyand strong that nothing comparable had ever been known to haveoccurred here, the movement having lasted with severe shakes andshocks for about three quarters of an hour’ (Coolhaas 1976, trans-lation by Reid (2012)). Another powerful earthquake, with shakinglasting more than 5 min was felt in 1757 and in 1834 an earth-quake similar to that in 1699 occurred (Wichmann 1918). The onlylarge, damaging earthquake near Jakarta to have been instrumen-

tally recorded occurred on 1903 February 27, with an Mw of 7.3(Engdahl & Villasenor 2002) and a hypocentre in the subductingslab somewhere between northwest Java and southeast Sumatra(Kanamori & Abe 1979).

Jakarta is located on a sedimentary basin with little topographicrelief, with elevation gradually increasing from a few metres atthe coast to 50 m at a distance 20 km from the coast. The surfacegeology is comprised of Holocene sand dunes and alluvium from thecoast to about 6 km southward to the city centre, whereas farthersouth it is dominated by Pleistocene alluvial fan deposits with apatch of Tertiary volcanic deposits in the west (Fig. 1, see alsoVan Bemmelen 1949; Turkandi et al. 1992). Alluvial fan depositsthicken to the north, so that the central part of Jakarta is composedof very thick alluvial fan sediments (>300 m) (Patmosukismo &Yahya 1974; Fachri et al. 2003). A recent study by Ridwan et al.(2014, 2015) mapped 1-D basin structure using 11 microtremormeasurements, and the results suggest depth to engineering bedrock(VS > 750 m s−1) of 600 m to in the north and around 360 m in thecentral part of the basin.

DATA A N D M E T H O D S

Data

We deployed Trillium Compact 20 s, three-component broad-bandsensors at every site. Sensors were installed in schools, generally ona concrete slab floor. Data were recorded by dataloggers designedand built by the Australian National University, which are poweredby batteries and have a 24-bit dynamic range. Recording was con-ducted at a 250 Hz sampling rate and instrument drift correctionswere applied by the digitizer at every hour based on GPS time. Asdescribed above, half of the 52 stations were deployed throughoutJakarta at 3–5 km spacing and left in place for the duration ofthe 3-month experiment (red circles in Fig. 1). The remaining

at The A

ustralian National U

niversity on February 21, 2016http://gji.oxfordjournals.org/

Dow

nloaded from

920 E. Saygin et al.

Figure 2. Comparison of Green’s functions for two station pairs retrieved with cross-correlation (black traces) and regularized deconvolution (red traces). Themap on the right shows the locations of these station pairs. Waveforms are filtered between 0.1 and 2 Hz and normalized to unity.

instruments were deployed so as to achieve denser spacing firstin northeastern Jakarta and then re-deployed after each monthto northwestern and southern Jakarta (green, blue, and black tri-angles, respectively, in Fig. 1). The data were downsampled to50 Hz and a zero-phase low pass filter was applied prior todown-sampling. We did not carry out instrument response re-moval, since all of the instruments are identical and deconvolu-tion (see below) cancels out the effect of the identical instrumentresponses.

Retrieval of Green’s Functions

After resampling the data, we follow Saygin & Kennett (2010,2012) in the processing of noise recordings. Instead of applyingcross-correlation on the noise records, we applied a regularizeddeconvolution to enable the use of a broader frequency range. Thedeconvolution is conducted on data of 300 s time interval with a60 s overlap, and we used 0.01 for the value of the regularizationparameter. After the construction of deconvolved traces, stacking isapplied on all of the available traces to create the final waveform. InFig. 2, Green’s functions for two different station pairs are shownfrom cross-correlation and regularized deconvolution. Waveformsfrom the regularized approach carry more energetic arrivals withhigher signal to noise ratio in the same frequency band.

In Fig. 3, Green’s functions are plotted for a reference stationlocated at the uppermost northeast of the array, where we termthe waves propagating away from the reference station as causal,and towards the station as acausal. There is clear propagation ofdispersed Rayleigh waves at acausal times, and a faster branchof energy exists at causal times. We identify the faster branch ofarrivals as higher order Rayleigh wave modes arriving before thedominant fundamental mode with a much lower signal to noise ratio.We confirm the nature these arrivals by doing simple modelling ofthe group velocities as in Figures A1 and A2. We identified andpicked fundamental mode Rayleigh wave Green’s functions on thedispersed wave trains visible on the vertical components. We verifythese picks by comparing them with the arrival time of the Lovewave Green’s function that appears on the same receiver pair’stransverse components.

M E T H O D S

Imaging with noise

Seismic noise is generally generated by a combination of ocean-solid earth coupling and human activity, and can propagate over

Figure 3. Retrieved Rayleigh wave Green’s functions between a permanentstation in the edge of network (northeast) and others. Waveforms werefiltered between 0.15 and 0.5 Hz and normalized to unity. Red and greenlines show group velocities of 0.5 and 0.25 km s−1, respectively. Inset mapshows the location of the source station (red star) and receiver stations (greytriangles).

large distances. The simultaneous recording of noise in a seismicnetwork can be utilized by, for example, using the cross-correlationoperation to extract the Green’s function between any two stationsin the network. This Green’s function will carry information aboutthe subsurface seismic velocity structure between the two stations.Most commonly, the dispersed surface wave components of thisGreen’s function is used to probe the underlying medium.

Imaging with seismic noise is an emerging technique and of-ten offers much more control on the survey design over traditional

at The A

ustralian National U

niversity on February 21, 2016http://gji.oxfordjournals.org/

Dow

nloaded from

Imaging architecture of Jakarta Basin, Indonesia 921

Figure 4. Geometry of straight ray path distributions at selected frequencies. The bar plot shows the distribution of number of ray path for the wholemeasurement band.

earthquake-based imaging techniques. Review papers by Weaver(2005) and Snieder & Larose (2013) outline the development of thetechnique since its inception with example studies. Seismic NoiseTomography was first applied by Shapiro & Campillo (2004) andShapiro et al. (2005), Sabra et al. (2005) to Southern California.Subsequent studies applied the noise tomography technique at avariety of scales, from upper mantle to upper crustal problems as-sociated with 100 to 10 s of km lateral scales, respectively (Gao& Shen 2014; Yang et al. 2007; Mordret et al. 2013; Saygin et al.2013).

We use over 3000 retrieved Rayleigh waves and measure theirgroup velocities at frequencies between 0.2 and 2 Hz by applyingnarrowband Gaussian filters. In Fig. 4, the distribution of the totalnumber of ray paths at each frequency and geometry of ray paths atselected frequencies are given. The coverage of the network offersa dense and balanced ray path distribution at the chosen bands ofthe measurements.

The measured traveltimes are used in a two-step tomographicscheme to image the Earth at different frequencies. In the for-ward part, we use the Fast Marching Method (FMM; Rawlinson &Sambridge 2004a,b) to trace the wave fronts for a given velocitymodel. FMM is especially useful to model seismic wave propaga-tion in 2-D where strong velocity anomalies can have a profoundeffect on ray geometry.

Transdimensional Bayesian seismic tomography

We use a Bayesian framework in the inversion of interstation travel-times at different frequencies to map the Rayleigh wave group ve-locity perturbation field across Jakarta. In inversion methods which

use Bayesian strategies, a probability density is specified a priorito avoid consideration of unrealistic Earth models. Bayes’ theoremessentially combines this prior probability density with a likelihoodfunction that expresses the probability of observing the data for agiven model, to obtain a posterior probability density distributionfor the model given the observed data (Gouveia & Scales 1998).In order to determine an appropriate prior probability distribution,in general information from previous work and knowledge of theexpected seismic velocity ranges are used (Scales & Tenorio 2001).In this work, we use uniform priors with wide ranges for the pa-rameters to be searched with a group velocity range between 0.1and 1.0 km s−1, and a data noise parameter assigned to each cellbetween 0.1 and 10 s.

One of the key advantages of adopting the Bayesian frameworkis that it provides a statistically rigorous appraisal of model uncer-tainty, which can be very important in interpretation of the results.It does this by sampling the posterior probability distribution, sothat instead of finding one best model, we obtain a distribution ofmodels from which statistics such as model mean and confidenceintervals can be computed.

In this paper, we use the reversible jump Markov chain MonteCarlo (rj-McMC) technique to simultaneously explore model andparameter space in what is termed transdimensional Bayesian inver-sion. This approach was introduced to the geophysics community byMalinverno (2002), and has since been used for inverting receiverfunctions and surface wave group and phase velocities (see, e.g.Bodin & Sambridge 2009; Agostinetti & Malinverno 2010; Bodinet al. 2012a,b).

We use the rj-McMC implementation of Bodin et al. (2012a), inwhich all the Rayleigh wave group velocities observed for a givenperiod are modelled with a parameterization defined by a number

at The A

ustralian National U

niversity on February 21, 2016http://gji.oxfordjournals.org/

Dow

nloaded from

922 E. Saygin et al.

Figure 5. Evolution of Earth models for measurements at 2 Hz represented with Voronoi cells for (a) 20 000 (10 000 burn-in, 10 000 after burn-in) iterations,(b) 500 000 (250 000 burn-in, 250 000 after burn-in) iterations with different number of chains.

of constant-velocity Voronoi cells distributed irregularly over thedeployment area. Each model is defined by the number of Voronoicells, their positions, and the values of Rayleigh wave group veloc-ity assigned to the cells. In the sampling of the Bayesian posteriordistribution, not only cell group velocities but also the number andpositions of cells are varied in different steps of each Markov chain.This allows not only the group velocity but also the level of detail inthe model to be determined by the data, thereby avoiding regulariza-tion and any arbitrary selection of regularization parameter values.The reversible jump technique automatically adjusts the underlyingparametrization of the model to produce solutions with an appro-priate level of complexity to fit the data to statistically meaningfullevels.

This method has been applied to seismic noise tomography ina number of previous studies: Young et al. (2013) applied it usingphase velocity measurements of interstation Green’s functions inTasmania, and Zulfakriza et al. (2014) used it with group velocitymeasurements in central Java.

As was done in these previous studies, we accelerate convergenceby running a large number (128) of Markov chains in parallel. Ateach step of each Markov Chain, a velocity model is proposedand then either rejected or accepted to become a member in anensemble of acceptable solutions. The Markov Chains are run formany iterations producing new models, and eventually convergeto a sampling of the posterior distribution that is independent ofthe initial model. Once this ‘burn-in’ stage has been achieved, westart to extract models to be used in the ensemble averages. Atthe end, this ensemble can be used to calculate mean Rayleighwave group velocity models for selected periods through takinga simple arithmetic average over the ensemble values for Rayleighwave group velocity at that point, and their associated uncertainties.In addition to the velocity models, data noise is also regarded as aparameter to be determined in the inversion.

In Fig. 5, the evolution of ensemble mean Earth models is shownwith varying number of chains and burn-in and post burn in steps.With an increasing number of chains, the resulting ensemble aver-ages are much smoother. When the number of chains is low, the

averaging process over a relatively short number of runs cannotproduce smooth velocity models, so that the sharp edges of Voronoicells are visible. A degree of convergence is observed even for alower numbers of steps with higher number of chains. In Fig. 6,the distribution of Voronoi cells for a randomly selected chain isshown for every simulation shown in Fig. 5. It can be seen that theshorter runs have not converged, which is remedied by increasingthe number of iterations.

Ray paths were iteratively updated with FMM five times, usingthe mean velocity models from previous runs to accommodate theeffects of velocity changes in wave propagation, and so include raybending. In Fig. 7, results from iterative updates of ray paths areshown for measurements at 1 Hz. In each panel, the mean velocitymodel from the previous run was used to calculate the ray pathgeometry, except in the first panel where straight ray paths wereused. Iterative updating of ray paths is important especially in aregion such as Jakarta where there is a strong variation of Rayleighwave group velocities in the measurement band. The effect of thebending of ray paths is reflected in the images, as is particularlynotable between Fig. 7(a) and the other parts of Fig. 7. While the raypaths undergo considerable change after the first update, the changein ray path geometry is much smaller in subsequent updates.

The computations were carried out in a parallelized environment,utilizing a total of 128 compute cores. Each core ran a separatechain with 250 000 burn-in steps and 250 000 post burn-in stepsat each period. The models generated during the burn-in phaseare rejected and not used the ensemble averages. To avoid anydependency during the ensemble averaging, each chain was thinnedby taking every 50th model.

R E S U LT S : G RO U P V E L O C I T Y I M A G E S

In Fig. 8, we show the results of the Rayleigh wave group velocitymaps from 0.2 Hz to 2 Hz. The corresponding uncertainty of theeach image is given in Fig. 9.

at The A

ustralian National U

niversity on February 21, 2016http://gji.oxfordjournals.org/

Dow

nloaded from

Imaging architecture of Jakarta Basin, Indonesia 923

Figure 6. Randomly selected chains for the same runs with Fig. 5. Curves are coloured with red (first half) for depicting burn-in and blue (second half) postburn-in states.

Because of the low velocity of basin-fill sediments, surface wavesin the period range 0.2 to 2 Hz are most sensitive to structure atdepths <1 km. In the western and northwestern parts of Jakarta typ-ical velocities of the Rayleigh wave velocities are 200–400 m s−1.Towards the east the velocity anomalies exhibit more spatial vari-ability, with a number of small high velocity bodies surrounded bylow velocity material. South of latitude 6.3◦S, a large block existswith high group velocity across all of the measured period rangewith velocities up to 0.8 km s−1. At around 0.6 Hz, the ray path cov-erage reaches a maximum, where the low velocity anomalies showsimilar patterns to those at 2 Hz. This indicates that the extent ofthe basin does not change over a wide range of depths. The resultsindicate very low velocities across the region comparable to or evenlower than the Kanto Basin (Koketsu & Kikuchi 2000; Denolle et al.2014) or Taipei Basin (Huang et al. 2010).

T R A N S D I M E N S I O NA L 1 - D S H E A RWAV E I N V E R S I O N S

A common way of inverting the surface wave velocity maps fromseismic noise tomography is to conduct pixel by pixel inversionof dispersion curves extracted from these maps. This 2-stage ap-proach has been applied successfully at a number of domains toimage depth-shear wave velocity structure. Brenguier et al. (2007)imaged the magma chamber of Piton de la Fournaise volcano fromthe inversion of dispersion curves extracted from surface velocitytomographic sections. Saygin & Kennett (2012) imaged large scalestructures across Australia from the point inversion of dispersioncurves. Mordret et al. (2014) applied depth inversion to phase ve-locity maps constructed from a very dense data set of Valhall Life ofField Seismic network, and recently Pilia et al. (2015) constrainedthe shear wave velocity in the southeast of Australia from the Trans-dimensional inversion of group velocity measurements of Green’sfunctions retrieved from noise.

We extracted over 1500 dispersion curves from the group velocitytomograms with an equal spacing of 0.6 km in each direction, for

frequencies between 0.2 and 2 Hz. The curves were inverted witha Transdimensional Bayesian scheme to map the variation of theshear wave depth structure across Jakarta. In Bodin et al. (2012b),a detailed description of the method is given. This approach fol-lows the same methodology used in the inversion of traveltimes forcreating the group velocity images of this study. The model is pa-rameterized with a variable number of layers defined with Voronoicells, and thicknesses, all of which are unknowns in the inversions.For the prior that define the search space, we used 0.3–2.7 km s−1

for layer velocities, and the number of the layers are allowed to varybetween 2 and 30 for characterizing the first 5 km. Each dispersioncurve inversion is run for 40 000 steps for the burn-in and another100 000 steps for the post burn-in. Each resulting velocity modelposterior distribution is from an ensemble of 72 chains run in par-allel, which is sufficient to explore the parameter space. In general,the misfit between observed and modelled dispersion curves is low.An example of an inverted dispersion curve is given in Fig. 10. Theinverted dispersion curves and shear wave velocity-depth modelsare derived as averages over the respective posterior distributions.

In Figs 11 and 13, shear wave velocity cross-sections are shownwhich illustrate the structural variation of the basin across Jakarta.The associated uncertainty is presented in Fig. 12. The very lowvelocity cover (VS < 1.2 km s−1) extends to depths over 500 min northern Jakarta. Another successive unit follows this very lowvelocity unit extending to depths of 1.5 km. The spatial extent of thebasin cover is mainly around northern Jakarta (6.1◦S–6.25◦S), andthe presence of low-velocity material diminishes with increasingdepth. Most of the basin fill with velocities less than 1.5 km s−1

is in northwest Jakara. At 2 km, the signature of the low velocitybasin has largely disappeared. However, around these depth surfacewaves are sensitive to a large range of depths.

The inverted shear wave velocities show a high level of similar-ity to the sedimentary rocks from other domains such as the SanFrancisco Bay area. A considerable amount of work has been donefor this region to produce a compressional and shear wave velocitymodel varying with depth. Brocher (2005) compiled measurementsfrom number of different methods to create empirical relationships

at The A

ustralian National U

niversity on February 21, 2016http://gji.oxfordjournals.org/

Dow

nloaded from

924 E. Saygin et al.

Figure 7. Change of earth models with iteratively updated ray paths at 1 Hz. (a) Straight ray paths. (b) Ray paths traced in velocity model (a). (c) Ray pathstraced in model from (b). (d) Ray paths updated from panel (c). (e) Updated from panel (d) and (f) updated from panel (e).

at The A

ustralian National U

niversity on February 21, 2016http://gji.oxfordjournals.org/

Dow

nloaded from

Imaging architecture of Jakarta Basin, Indonesia 925

Figure 8. Rayleigh wave group velocity tomograms from Transdimensional Bayesian Seismic Noise Tomography between 2 and 0.2 Hz after five updates ofray path geometry: (a) 2 Hz, (b) 1.8 Hz, (c) 1.4 Hz, (d) 1.0 Hz, (e) 0.6 Hz and (f) 0.2 Hz.

at The A

ustralian National U

niversity on February 21, 2016http://gji.oxfordjournals.org/

Dow

nloaded from

926 E. Saygin et al.

Figure 9. Uncertainties of Rayleigh wave tomograms from Transdimensional Bayesian Seismic Noise Tomography between 2 and 0.2 Hz after five updates ofray path geometry: (a) 2 Hz, (b) 1.8 Hz, (c) 1.4 Hz, (d) 1.0 Hz, (e) 0.6 Hz and (f) 0.2 Hz.

at The A

ustralian National U

niversity on February 21, 2016http://gji.oxfordjournals.org/

Dow

nloaded from

Imaging architecture of Jakarta Basin, Indonesia 927

Figure 10. Example of an inverted dispersion curve from northern part ofJakarta with Transdimensional Inversion. The left panel shows observed andinverted curves and right panel shows the corresponding shear wave velocityvariation with depth.

between depth, and seismic velocities. As an example for the GreatValley Sequence of sedimentary rocks, typical shear wave velocitiesare about 1.27 km s−1 at 1 km depth and around 1.8 km s−1 at 2 kmdepth, which are very similar to results that we obtained from ourshear wave inversions.

D I S C U S S I O N A N D C O N C LU S I O N

For the first time, we developed a high-resolution shear wave ve-locity model of the Jakarta Basin from a 2-stage TransdimensionalBayesian inversion of Rayleigh wave Green’s functions retrievedfrom seismic noise. We conducted Rayleigh wave tomography byusing Green’s functions retrieved from seismic noise to estimate 2-D Rayleigh wave group velocity maps for different periods. Then,the pixel inversion of dispersion curves across the study area mapsthe 3-D shear wave velocity variations in the basin. The low-velocitybasin covers most of the northern part of Jakarta at shallow depths,where the basin boundaries shrink with the increasing depths. Theinfluence of the basin mostly diminishes at around 1–1.5 km, where

the deepest part is below central Jakarta extending to depths of1.5 km. Even below the basin, shear wave velocities are still rela-tively low with an average of 1.8 km s−1 at 2.0 km. In the measure-ment band, surface waves are mostly sensitive to the first 2.5 km.Towards the south, the general trend is toward increasing velocities,however the resolution at the tip of the region is lower compared tothe central part of Jakarta due to the coverage of ray paths.

We argue that in the case of a large earthquake, the city of Jakartawill be exposed to enhanced seismic hazard due to amplificationand prolonged duration of seismic wave motion. Along with theamplification of seismic waves due to the impedance contrast atthe basin-basement contact, the geometry of the basin can increasethe duration of strong ground motion by focussing and trappingof seismic waves (Furumura & Kennett 1998). In future work, wehope to consider the influence of the basin on the character ofearthquake-generated ground motion.

A C K N OW L E D G E M E N T S

We thank editor Gabi Laske, reviewers Marine Denolle, Martha Sav-age and one anonymous reviewer for their constructive criticismson the manuscript. We are grateful to the Australian Departmentof Foreign Affairs and Trade’s Australia-Indonesia Facility for Dis-aster Reduction for funding the experiment. Seismograms used inthis study were collected as part of the Jakarta Experiment (2013–2014) using instruments purchased from ARC Linkage Infrastruc-ture, Equipment & Facilities grant no LE120100061. We thank Gov-ernor of DKI Jakarta as well local government of Jakarta for theirpermission for the experiment and sites. This research was under-taken with the assistance of resources provided at the NCI NationalFacility systems at the Australian National University through theNational Computational Merit Allocation Scheme supported by the

Figure 11. Shear wave velocity slices across Jakarta from the inversion of group velocity dispersion curves.

at The A

ustralian National U

niversity on February 21, 2016http://gji.oxfordjournals.org/

Dow

nloaded from

928 E. Saygin et al.

Figure 12. The corresponding uncertainty of shear wave velocity slices given in Fig. 11.

Figure 13. Shear wave vertical profiles (1, 2 and 3) for selected three lines shown in the inset map.

at The A

ustralian National U

niversity on February 21, 2016http://gji.oxfordjournals.org/

Dow

nloaded from

Imaging architecture of Jakarta Basin, Indonesia 929

Australian Government. Some of calculations were performed onthe Terrawulf cluster, a computational facility supported throughAuScope and the Australian Geophysical Observing System(AGOS). This work was supported by resources provided by thePawsey Supercomputing Centre with funding from the AustralianGovernment and the Government of Western Australia. Pythonpackages Matplotlib (Hunter 2007) was used in drafting the fig-ures (except maps), Basemap was used in drafting maps and ObsPy(Beyreuther et al. 2010) was used for the file format conversions,and data management.

R E F E R E N C E S

Abidin, H.Z., Andreas, H., Gumilar, I., Fukuda, Y., Pohan, Y.E. & Deguchi,T., 2011. Land subsidence of Jakarta (Indonesia) and its relation withurban development, Nat. Hazards, 59(3), 1753–1771.

Agostinetti, N.P. & Malinverno, A., 2010. Receiver function inversion bytrans-dimensional Monte Carlo sampling, Geophys. J. Int., 181(2), 858–872.

Anderson, J., Bodin, P., Brune, J., Prince, J., Singh, S., Quaas, R. & Onate, M.,1986. Strong ground motion from the Michoacan, Mexico, earthquake,Science, 233(4768), 1043–1049.

Beyreuther, M., Barsch, R., Krischer, L., Megies, T., Behr, Y. & Wasser-mann, J., 2010. Obspy: A python toolbox for seismology, Seismol. Res.Lett., 81(3), 530–533.

Bodin, T. & Sambridge, M., 2009. Seismic tomography with the reversiblejump algorithm, Geophys. J. Int., 178(3), 1411–1436.

Bodin, T., Sambridge, M., Rawlinson, N. & Arroucau, P., 2012a. Transdi-mensional tomography with unknown data noise, Geophys. J. Int., 189(3),1536–1556.

Bodin, T., Sambridge, M., Tkalcic, H., Arroucau, P., Gallagher, K.& Rawlinson, N., 2012b. Transdimensional inversion of receiverfunctions and surface wave dispersion, J. geophys. Res., 117(B2),doi:10.1029/2011JB008560.

Brenguier, F., Shapiro, N.M., Campillo, M., Nercessian, A. & Ferrazzini,V., 2007. 3-D surface wave tomography of the Piton de la Fournaisevolcano using seismic noise correlations, Geophys. Res. Lett., 34(2),doi:10.1029/2006GL028586.

Brocher, T.M., 2005. Compressional and shear wave velocity versus depth inthe San Francisco Bay Area, California: rules for USGS bay area velocitymodel 05.0. 0, Tech. rep.

Coolhaas, W.P., 1976. Generale Missiven van Gouverneurs-Generaal enRaden aan Heren XVII der Verenigde Oostindische Compagnie, Deel VI:16981713, Nijhoff.

Cybriwsky, R. & Ford, L.R., 2001. City profile: Jakarta, Cities, 18(3), 199–210.

Delorey, A.A. & Vidale, J.E., 2011. Basin shear-wave velocities beneathSeattle, Washington, from noise-correlation Rayleigh waves, Bull. seism.Soc. Am., 101(5), 2162–2175.

Denolle, M.A., Miyake, H., Nakagawa, S., Hirata, N. & Beroza, G.C., 2014.Long-period seismic amplification in the Kanto basin from the ambientseismic field, Geophys. Res. Lett., 41(7), 2319–2325.

Engdahl, E.R. & Villasenor, A., 2002. 41 global seismicity: 1900–1999, Int.Geophys., 81, 665–690, XV–XVI.

Fachri, M., Djuhaeni, H.L. & Ramdhan, A., 2003. Stratigraphy and hy-drostratigraphy in Jakarta groundwater basin, Bull. Geol. Bandung. Inst.Technol., 34(3), 169–189, in Indonesian.

Furumura, T. & Kennett, B., 1998. On the nature of regional seismic phases–III. The influence of crustal heterogeneity on the wavefield for subductionearthquakes: the 1985 Michoacan and 1995 Copala, Guerrero, Mexicoearthquakes, Geophys. J. Int., 135(3), 1060–1084.

Gao, H. & Shen, Y., 2014. Upper mantle structure of the cascades fromfull-wave ambient noise tomography: evidence for 3D mantle upwellingin the back-arc, Earth planet. Sci. Lett., 390, 222–233.

Gouveia, W.P. & Scales, J.A., 1998. Bayesian seismic waveform inversion:parameter estimation and uncertainty analysis, J. geophys. Res., 103(B2),2759–2779.

Hamilton, W.B., 1979. Tectonics of the Indonesian Region, no. 1078, USGovt. Print. Off.

Harsolumakso, A., 2001. Struktur geologi dan daerah genangan, BuletinGeologi, 33, 29–45.

Helmberger, D. & Wiggins, R.A., 1971. Upper mantle structure of midwest-ern united states, J. geophys. Res., 76(14), 3229–3245.

Huang, Y.-C., Yao, H., Huang, B.-S., van der Hilst, R.D., Wen, K.-L.,Huang, W.-G. & Chen, C.-H., 2010. Phase velocity variation at pe-riods of 0.5–3 seconds in the Taipei basin of Taiwan from cor-relation of ambient seismic noise, Bull. seism. Soc. Am., 100(5A),2250–2263.

Hunter, J.D., 2007. Matplotlib: A 2D graphics environment, Comput. Sci.Eng., 9(3), 90–95.

Kagawa, T., Zhao, B., Miyakoshi, K. & Irikura, K., 2004. Modeling of 3Dbasin structures for seismic wave simulations based on available informa-tion on the target area: case study of the Osaka basin, Japan, Bull. seism.Soc. Am., 94(4), 1353–1368.

Kanamori, H. & Abe, K., 1979. Reevaluation of the turn-of-the-centuryseismicity peak, J. geophys. Res., 84(B11), 6131–6139.

Kawase, H. & Aki, K., 1989. A study on the response of a soft basin forincident S, P, and Rayleigh waves with special reference to the longduration observed in Mexico city, Bull. seism. Soc. Am., 79(5), 1361–1382.

Kennett, B., 2009. Seismic Wave Propagation in Stratified media, ANU EPress.

Koketsu, K. & Kikuchi, M., 2000. Propagation of seismic ground motion inthe Kanto basin, Japan, Science, 288(5469), 1237–1239.

Koketsu, K., Miyake, H. & Tanaka, Y., 2009. A proposal for a standardprocedure of modeling 3-D velocity structures and its application to theTokyo metropolitan area, Japan, Tectonophysics, 472(1), 290–300.

Lee, E.-J., Chen, P., Jordan, T.H., Maechling, P.B., Denolle, M.A. & Beroza,G.C., 2014. Full-3-D tomography for crustal structure in southern Cali-fornia based on the scattering-integral and the adjoint-wavefield methods,J. geophys. Res., 119(8), 6421–6451.

Malinverno, A., 2002. Parsimonious Bayesian Markov chain Monte Carloinversion in a nonlinear geophysical problem, Geophys. J. Int., 151(3),675–688.

Mordret, A., Landes, M., Shapiro, N., Singh, S., Roux, P. & Barkved, O.,2013. Near-surface study at the Valhall oil field from ambient noise sur-face wave tomography, Geophys. J. Int., doi:10.1093/gji/ggt061.

Mordret, A., Landes, M., Shapiro, N., Singh, S. & Roux, P., 2014. Ambientnoise surface wave tomography to determine the shallow shear velocitystructure at Valhall: depth inversion with a neighbourhood algorithm,Geophys. J. Int., 198(3), 1514–1525.

Ng, A. H.-M., Ge, L., Li, X., Abidin, H.Z., Andreas, H. & Zhang, K., 2012.Mapping land subsidence in Jakarta, Indonesia using persistent scattererinterferometry (PSI) technique with ALOS PALSAR, Int. J. Appl. EarthObs. Geoinf., 18, 232–242.

Patmosukismo, S. & Yahya, I., 1974. The basement configurationof the north west Java area, in Proc. 3rd Ann. Convention, pp.129–152. http://archives.datapages.com/data/ipa/data/003/003001/129ipa0030129.htm.

Pilia, S., Rawlinson, N., Cayley, R.A., Bodin, T., Musgrave, R., Reading,A.M., Direen, N.G. & Young, M.K., 2015. Evidence of micro-continententrainment during crustal accretion, Nature Scientific Reports, 5(8218),doi:10.1038/srep08218.

Rawlinson, N. & Sambridge, M., 2004a. Multiple reflection and transmissionphases in complex layered media using a multistage fast marching method,Geophysics, 69(5), 1338–1350.

Rawlinson, N. & Sambridge, M., 2004b. Wave front evolution in stronglyheterogeneous layered media using the fast marching method, Geophys.J. Int., 156(3), 631–647.

Reid, A., 2012. Historical Evidence for Major Tsunamis in the Java Sub-duction Zone, Asia Research Institute (ARI), National University ofSingapore.

at The A

ustralian National U

niversity on February 21, 2016http://gji.oxfordjournals.org/

Dow

nloaded from

930 E. Saygin et al.

Ridwan, M., Afnimar, A., Widiyantoro, S., Irsyam, M. & Yamanaka, H.,2014. Estimation of S-wave velocity structures by using microtremorarray measurements for subsurface modeling in Jakarta, J. Math. Fundam.Sci., 46(3), 313–327.

Ridwan, M., Widiyantoro, S. & Irsyam, M., 2015. Identification of engi-neering bedrock in Jakarta by using array observations of microtremors,Procedia Earth and Planetary Science, Vol. 12, pp. 77–83.

Sabra, K.G., Gerstoft, P., Roux, P., Kuperman, W. & Fehler, M.C., 2005.Extracting time-domain Green’s function estimates from ambient seismicnoise, Geophys. Res. Lett., 32(3), doi:10.1029/2004GL021862.

Saygin, E. & Kennett, B., 2012. Crustal structure of Australiafrom ambient seismic noise tomography, J. geophys. Res., 117(B1),doi:10.1029/2011JB008403.

Saygin, E. & Kennett, B.L., 2010. Ambient seismic noise tomography ofAustralian continent, Tectonophysics, 481(1), 116–125.

Saygin, E., McQueen, H., Hutton, L., Kennett, B. & Lister, G., 2013. Struc-ture of the Mt Isa region from seismic ambient noise tomography, Aust.J. Earth Sci., 60, 707–718.

Scales, J.A. & Tenorio, L., 2001. Prior information and uncertainty in inverseproblems, Geophysics, 66(2), 389–397.

Shapiro, N.M. & Campillo, M., 2004. Emergence of broadband Rayleighwaves from correlations of the ambient seismic noise, Geophys. Res. Lett.,31(7), doi:10.1029/2004GL019491.

Shapiro, N.M., Campillo, M., Stehly, L. & Ritzwoller, M.H., 2005. High-resolution surface-wave tomography from ambient seismic noise, Science,307(5715), 1615–1618.

Simons, W. et al., 2007. A decade of GPS in Southeast Asia: re-solving Sundaland motion and boundaries, J. geophys. Res., 112(B6),doi:10.1029/2005JB003868.

Snieder, R. & Larose, E., 2013. Extracting Earth’s elastic wave re-sponse from noise measurements, Annu. Rev. Earth Planet. Sci., 41(1),183–206.

Turkandi, T., Sidarto Agustiyanto, D. & Purbohadiwidjojo, M., 1992. Geo-logic Map of the Jakarta and Kepulauan Seribu Quadrangle Java.

Van Bemmelen, R., 1949. The Geology of Indonesia, Vol 1A, General Ge-ology of Indonesia and Adjacent Archipelagoes, Government PrintingOffice, The Hague, Martinus Nijhoff, Sole Agent.

Weaver, R.L., 2005. Information from seismic noise, Science, 307(5715),1568–1569.

Wichmann, A., 1918. Die erdbeben des Indischen Archipels bis zum Jahre1857, vol. 20, J. Muller.

Yang, Y., Ritzwoller, M.H., Levshin, A.L. & Shapiro, N.M., 2007. Ambientnoise Rayleigh wave tomography across Europe, Geophys. J. Int., 168(1),259–274.

Young, M.K., Rawlinson, N. & Bodin, T., 2013. Transdimensional inversionof ambient seismic noise for 3D shear velocity structure of the tasmaniancrust, Geophysics, 78(3), WB49–WB62.

Zulfakriza, Z., Saygin, E., Cummins, P., Widiyantoro, S., Nugraha, A., Luhr,B.-G. & Bodin, T., 2014. Upper crustal structure of central Java, Indonesia,from transdimensional seismic ambient noise tomography, Geophys. J.Int., 197, 630–635.

A P P E N D I X A : M O D E L L I N G G R E E N ’ SF U N C T I O N S

We show results of our modelling for two different velocity modelsto show the effective propagation of first-mode Rayleigh waves aswell as fundamental mode (see Fig. A1). This is also clearly visibleon the observed section (see Fig. A2) at causal and acausal times. Weuse the reflectivity method (Kennett 2009) to compute the Green’sfunction responses for two different Earth models. Vertical dipoleforce is used as source in the calculations.

A P P E N D I X B : R E G U L A R I Z E DD E C O N V O LU T I O N

We use regularized deconvolution as previously employed inSaygin & Kennett (2010, 2012). This approach is implementedin the frequency domain with a water-level type regularization op-erator (Helmberger & Wiggins 1971). If the velocity fields recordedat stations A and B are given with v(xA) and v(xB), respectively thenwe can denote the deconvolution with

�(ω) = v(xA, ω)v∗(xB, ω)

ρ(ω), (B1)

where ∗ is complex conjugation and ω is angular fre-quency. The denominator is regularized with a water-level typeparameter c,

ρ(ω) = max[ρ1, ρ2], (B2)

with

ρ1 = v(xB, ω)v∗(xB, ω), (B3)

ρ2 = c max[v(xB, ω)v∗(xB, ω)]. (B4)

The parameter c determines the level of spectral filling in the de-nominator, which prevents numerical instabilities in the divisionoperation. The resulting signal after transforming back to the timedomain is not modulated by the square of the ambient noise spec-trum and has a broader response.

Figure A1. Dispersion curves for fundamental and first mode of Rayleigh wave for a fast velocity model (left) and a slower model (right).

at The A

ustralian National U

niversity on February 21, 2016http://gji.oxfordjournals.org/

Dow

nloaded from

Imaging architecture of Jakarta Basin, Indonesia 931

Figure A2. Corresponding vertical–vertical component Green’s functions for velocity models given in the Fig. A1. Note the prominent early arrival of firstmode of Rayleigh wave computed from faster velocity model. Waveforms are filtered between 0.1 and 0.5 Hz and normalized to their maximum.

at The A

ustralian National U

niversity on February 21, 2016http://gji.oxfordjournals.org/

Dow

nloaded from