Earthquake clusters analysis in Central and East Java 1 region , Indonesia , based on hypocenter determination and 2 relocation with waveform cross-correlation 3 4

The Central and East Java region, which is part of Sunda Arc, has relatively high seismicity rates due to the convergence between two major tectonic plates in Indonesia region, the Indo-Australian plate that subducts under the Eurasian plate. Many devastating earthquakes in the study area occurred as results of these plates interaction, such as the 1994 Banyuwangi earthquake (Mw 7.6) and the 2006 Yogyakarta earthquake (Mw 6.3). This study aims to determine the precise earthquake location and analyze the pattern of seismicity distribution around Central and East Java, Indonesia. We manually re-picked P and S-wave arrival time recorded by Agency for Meteorology, Climatology and Geophysics (BMKG) of Indonesia network for the times period of January 2009-September 2017. We then determined the earthquake location by a non-linear method. To improve the accuracy of earthquakes location, we relocated 1,127 out of 1,529 events using a double-difference algorithm with waveform cross-correlation data. Overall, the seismicity around Central and East Java regions are dominantly distributed in the south of the island, e.g. Kebumen, Yogyakarta, Pacitan, Malang, and Banyuwangi cluster. These clusters are probably related to the subduction activity. Meanwhile, the shallow depth earthquakes that are clustered in mainland indicate the activity of inland faults in the region, e.g. Opak Fault, Kendeng Thrust, and Rembang-Madura-Kangean-Sakala (RMKS) fault zone. Several other active inland faults have not shown significant seismicity over the times period, i.e., Pasuruan Fault, Lasem Fault, Muria Fault, Semarang Thrust, Probolinggo Fault, etc.


Introduction
This manuscript is a non-peer reviewed preprint submitted to EarthArXiv. It has been under reviewed for publication to Geoscience Letters on 13 Sept 2020 with submission ID GOSL-D-19-00015R2. Newer versions may be moderately different with slight variations in content.
This manuscript is a non-peer reviewed preprint submitted to EarthArXiv. It has been under reviewed for publication to Geoscience Letters on 13 Sept 2020 with submission ID GOSL-D-19-00015R2. Newer versions may be moderately different with slight variations in content. 5 determine the Vp/Vs ratio of the observed data (Fig 2b). To determine the hypocenter 97 location, we applied a non-linear method using the NLLoc program (Lomax et al. 2000) 98 and the global 1D seismic velocity model of AK135 (Kennett et al. 1995). The 99 algorithm used in this program is the oct-tree importance sampling to produce an 100 estimation of the posterior density function (PDF) for the hypocenter location in 3D 101 (Lomax and Curtis 2001). The same method was also previously implemented to

147
Hypocenter determination result consists of 1,529 events located using 11,192 phases 148 for each P and S-wave (Fig 3). The observed arrival times were plotted in the Wadati 149 diagram to independently check the linear relationship between phases data (Fig 2b).

157
We also estimated the uncertainty of observed data by using the waveform cross- 172 173 Several earthquakes that may be generated by the same source mechanism will produce 174 high waveform similarity at a common station. Therefore, the waveform cross-175 correlation process ensures the consistency of P and S-waves phase identification. We 176 computed the cross-correlation functions for P and S waves using a time window of 0.2 177 sec before and 2 sec after onset of P-arrival time and 1.4 sec before and 5 sec after S-178 arrival time onset. We used Butterworth filter between 1-6 Hz and coefficient 179 correlation criteria that are greater than 0.7. Figure 6 shows an example of the cross-180 correlation result at RTBI and PWJI station. The output of the waveform cross-181 correlation process that saved as inputs for HypoDD is lag time and coefficient 182 correlation.

184
We applied both catalog and cross-correlation differential time data into HypoDD to 185 improve the quality of event clustering and minimalize the eliminated events to relocate.

186
The weighting of the distance between paired events for catalog data (  10 the Java subduction system. The steeper dipping angle of the slab is likely to cause the 221 earthquake occurrence rate to be higher towards the east. Reported GCMT focal 222 mechanism shows dominantly thrusting mechanism, even though some of them also 223 have normal faulting mechanism (Fig 12).   In the northern part of East Java, there are clustered shallow seismicity around Rembang 267 and Madura (Fig 11). They probably associated with the same mechanism of Rembang-

312
All authors read and approved the final manuscript.

314
This manuscript is a non-peer reviewed preprint submitted to EarthArXiv. It has been under reviewed for publication to Geoscience Letters on 13 Sept 2020 with submission ID GOSL-D-19-00015R2. Newer versions may be moderately different with slight variations in content.
14 made by using The Generic Mapping Tools developed by Paul Wessel and Walter H.F.

319
This manuscript is a non-peer reviewed preprint submitted to EarthArXiv. It has been under reviewed for publication to Geoscience Letters on 13 Sept 2020 with submission ID GOSL-D-19-00015R2. Newer versions may be moderately different with slight variations in content.
This manuscript is a non-peer reviewed preprint submitted to EarthArXiv. It has been under reviewed for publication to Geoscience Letters on 13 Sept 2020 with submission ID GOSL-D-19-00015R2. Newer versions may be moderately different with slight variations in content.
This manuscript is a non-peer reviewed preprint submitted to EarthArXiv. It has been under reviewed for publication to Geoscience Letters on 13 Sept 2020 with submission ID GOSL-D-19-00015R2. Newer versions may be moderately different with slight variations in content.
This manuscript is a non-peer reviewed preprint submitted to EarthArXiv. It has been under reviewed for publication to Geoscience Letters on 13 Sept 2020 with submission ID GOSL-D-19-00015R2. Newer versions may be moderately different with slight variations in content.   This manuscript is a non-peer reviewed preprint submitted to EarthArXiv. It has been under reviewed for publication to Geoscience Letters on 13 Sept 2020 with submission ID GOSL-D-19-00015R2. Newer versions may be moderately different with slight variations in content. This manuscript is a non-peer reviewed preprint submitted to EarthArXiv. It has been under reviewed for publication to Geoscience Letters on 13 Sept 2020 with submission ID GOSL-D-19-00015R2. Newer versions may be moderately different with slight variations in content. This manuscript is a non-peer reviewed preprint submitted to EarthArXiv. It has been under reviewed for publication to Geoscience Letters on 13 Sept 2020 with submission ID GOSL-D-19-00015R2. Newer versions may be moderately different with slight variations in content. This manuscript is a non-peer reviewed preprint submitted to EarthArXiv. It has been under reviewed for publication to Geoscience Letters on 13 Sept 2020 with submission ID GOSL-D-19-00015R2. Newer versions may be moderately different with slight variations in content.