The active tectonic condition in Sumatra Island has generated several significant earthquakes followed by massive losses. Therefore, the seismic hazard scheme must be carefully conducted. Here, we analyzed the hypocenters distribution by relative relocation for stochastic analysis with machine learning to determine the seismic hazard condition. The well-relocated hypocenters figure several clusters ranging from 20 – 40 km to the megathrust zone to 0 – 20 km to the active faults. Total earthquakes were reduced from 12111 to 6598 after decluttering, which provided the values of magnitude completeness of 4.3, a value of ~8.4, and b-value of 0.98, indicating a highly stressful condition. The peak ground acceleration in Sumatra Island for 10% and 2% in 50 years range with the highest values of 0.6 – 0.8 gal along active faults while the lowest value with 0.0 – 0.3 gal in the megathrust zone. An example of observation from the crustal earthquake in Langsa city suggests an important validation between numerical model and observation. A similar condition between the model and observation will provide a good concept that can be used to construct a mitigation plan.

Microseismic observation is a mandatory tool for analysing and imaging the progress of source earthquake mechanism. The observation of low-quality signal is usually found because of false detection, transient signals by natural noise, or related to seismometer condition and some human activities. Therefore, we try to figure the microseismic phenomenon after major earthquake in the Jogjakarta fault that was generated by an active fault namely Opak Fault. We used the seismic data recording from the seismic project that was installed in 2006 cover the Jogjakarta region. We used one week data to see the potential detection and highlight the possible of positive or negative false detection. With total 16 stations, we got ~500 events in only one week recording. We used a various threshold with 110 to scan all the dataset and we obtained ~450 events with 50-100 events with possible of false positives. Meanwhile, the threshold of 112 has < 50 events that could be suitable with the waveforms. The example results of 120 thresholds figure a strong event that is located by highly resolution of stack coherent from some stations with precise of P and S phase fitting. This study present the first automatic earthquake locations that can provide more detail of seismic structure information in the Jogjakarta region.

Tsunami warning is one of many important reports to save lives and reduce the damage for local peoples. A moment magnitude of P-wave (Mwp) and the rupture time duration (Tdur) can be used as the quickly parameters to diseminate the tsunami warning. In this paper, we analyze the seismic waveform from global network to get Mwp and Tdur of South-West Coast of Sumatera earthquake. Mwp was calculated using automatic and manual phase picking of P phase. The results of this study show a well-analyzed relationship between P wave from automatic and manual picking, Mwp and time duration, respectively. The result also give an encouraging studies for the early warning system that will be set up in the future in the region.

After Palu Earthquake that occurred on September 2018, another moderate earthquake on April 2019 with 6.8 Moment Magnitude (Mw) at a depth of 17 km occured in Banggai Islands. The earthquake occurred on the shallow depth that generated by an active fault with horizontal mechanism on the Banggai’s tectonic system. To explain the tectonic system in the Banggai Islands, we present a well-calculated 1-D velocity model by solving the coupled hypocentre-velocity inversion for 385 local earthquakes that recorded by BMKG regional network. The earthquakes was selected based on azimuthal gap, minimum number of stations and root mean square of travel‐time residuals. Technically, the fit solution are simultaneously inverted for total 81 inital models and will give an unique final model. The model is constructed by using Velest program that analyse the fit velocity model from body wave traveltimes (P and S wave), together with station corrections. The final 1-D velocity model will be very useful to conduct another high precision relative relocation and make a focal mechanism inversion.