Hazard analysis of earthquake in Pleret, Bantul Regency, Yogyakarta Special Region based on microtremor data

Pleret, one of the sub-districts in Bantul Regency, Yogyakarta, is prone to earthquakes. This area is traversed by an active fault system known as the Opak fault, with a maximum magnitude of 6.6 Mw. The local geological conditions can be determined based on microtremor data from 43 measurement points. The purpose of this research is to understand the dynamic characteristics of the local geological conditions, including the dominant frequency (f0), shear wave velocity (Vs30), peak ground acceleration (PGA), and to map the earthquake impacts on the population and structures. Microtremor measurements were conducted for approximately 20 minutes, revealing a dominant frequency range of 0.61-10.85 Hz. The HVSR curve inversion revealed the Vs30 range 213 to 826 m/s distribution. Peak Ground Acceleration (PGA) values varied from 0.16 to 0.26 g. Using InaSAFE software, the estimated affected population reached 49,052 people, with over 23,176 structures affected. The results of this mapping are expected to be utilized as a step towards enhancing disaster preparedness.


Introduction
The Pleret sub-district is part of the Bantul Regency, located south of Yogyakarta.During the 2006 Yogyakarta earthquake, Pleret suffered significant losses in casualties and building damage [1].This area is traversed by fault systems on the eastern part of the sub-district, notably marked by the presence of the Baturagung fault scarp, making Pleret susceptible to earthquakes.The Opak fault exhibits a strikeslip fault mechanism with a slip rate of 0.75 mm/year and has a maximum magnitude of 6.6 Mw [2].According to the Yogyakarta geological map, the research area consists of four formations: the Young Volcanic Deposits of Merapi Volcano (Qyu) in the western part, the Semilir Formation (Tmse) and a small portion of the Nglanggran Formation (Tmn) in the eastern part, and a small portion of the Alluvial Formation (Qa) in the central study area [3].
Earthquakes are natural occurrences that happen unexpectedly.Predicting when, where, or how powerful they will be is impossible.They have a significant impact and frequently cause substantial losses due to infrastructural damage and/or fatalities [4].The lack of information about earthquakeprone areas can increase the potential for building damage and loss of life.Building damage from earthquakes is not solely caused by the structural integrity of the buildings but is also influenced by local conditions [5].Given these factors, efforts to mitigate the impact of losses resulting from earthquake The purpose of this research is to understand the dynamic characteristics of the local geological conditions, including dominant frequency (f0), shear wave velocity (Vs30), and peak ground acceleration (PGA) based on the maximum magnitude of the Opak fault.Also, mapping the earthquake impacts on the population and structures is conducted.The results of this mapping are expected to be utilized as a step towards enhancing disaster preparedness.

Methods
This study utilized microtremor data from a total of 43 measurement points distributed across Pleret, Bantul Regency.Measurements were carried out using TDL-303s seismometers.The recorded microtremor data encompassed three components: horizontal component (E-W), horizontal component (N-S), and vertical component (Z).The geological map of Bantul Regency was also employed to assist in interpreting the outcomes of the microtremor analysis.

Horizontal to vertical spectral ratio (HVSR)
The microtremor data was recorded in (*.seed) format and subsequently analyzed using the Horizontal to Vertical Spectrum Ratio (HVSR) method.The sampling frequency was 100 Hz, with a frequency range of 0.5-15 Hz and a window length of 25 seconds.The H/V curve resulting from this processing describes each measurement point's dominant frequency and amplification factor.To evaluate the produced H/V curves, reliability criteria provided by SESAME were used [6].The HVSR process involves comparing amplitude spectra's vertical and horizontal components, expressed as [7].
Where   is Fourier spectra of the North-South component,   is Fourier spectra of the East-West component,   is Fourier spectrum of the vertical component.

HVSR curve inversion
The H/V curves obtained from the processing are then subjected to inversion using OpenHVSR, to obtain shear wave velocity values.The first step in the inversion process is preparing input data, consisting of the H/V curves in format (*.txt) and initial model parameter data, including primary wave velocity (  ), secondary wave velocity (  ), layer density (), and layer thickness (ℎ).The inversion process uses the Monte Carlo propagation method, which minimizes the misfit function [8].The misfit calculation is expressed as follows: Where   () is the simulated curves and   () is the experimental curves.The calculation of shear wave velocity down to a depth of 30 meters from the ground surface (Vs30) is done using the following equation [9].
Where N represents the number of layers up to a depth of 30 meters, ℎ stands for the thickness of the layer (m), Vsi denotes the shear wave velocity (m).The value of Vs30 is classified based on soil type according to the criteria set by the National Standardization Agency (SNI) 1726:2019 can be seen in Table 1 below [10].

Peak ground acceleration (PGA)
One of the approaches that can be used to determine the Peak Ground Acceleration (PGA) is the Deterministic Seismic Hazard Analysis (DSHA) method.PGA represents the maximum ground shaking acceleration at a location due to an earthquake during a specific period.The magnitude of PGA indicates higher earthquake risk.The value of PGA is influenced by several factors, including the earthquake magnitude, distance to the location, and earthquake depth [11].In the deterministic method, earthquake scenarios are pre-determined.These scenarios encompass information about earthquake events that are likely to occur at a specific location with a specific magnitude.One of the advantages of DSHA is its ease of use in estimating earthquake motions for worst-case scenarios [12].In this research, the OpenQuake software is used to calculate the PGA values using the DSHA method.The earthquake source is based on data from the worst-case earthquake along the Opak fault with a maximum magnitude of 6.6 Mw, resulting in a shallow crustal earthquake model.PGA values can be used to estimate the intensity of earthquake shaking at the surface.The Modified Mercalli Intensity (MMI) scale is commonly used for assessing earthquake shaking intensity.The relationship between PGA and MMI can be obtained based on the equation [13].
The earthquake hazard level can be classified based on the PGA value, with reference to other research Widigdo (2006) presented in Table 2 [14].

Estimate the number of populations and structures affected
The InaSAFE software is used to assess the population's exposure and structures to earthquake effects.
The key features of InaSAFE include impact analysis, predicting minimum needs, and generating maps and analysis tables.Combining hazard and exposure data in InaSAFE results in scenario-based estimations of the losses incurred by potential disaster hazards.The input required for conducting an analysis using InaSAFE consists of three layers: 1) The earthquake hazard layer comprises earthquake shaking intensity data (MMI) with pixel resolution 10 x 10 meters.2) The exposure layer comprises population data from the Central Statistics Agency of 2022 and building data from OpenStreetMap that offers free geospatial data and can be customized to meet specific requirements.3) Aggregation layer, representing administrative boundaries for the research area in vector format.
To estimate the number of evacuees, InaSAFE assumes that the population exposed to shaking greater than MMI scale V will be evacuated.The regulation of BNPB Head, number 7 year 2008 on Guidelines for Providing Assistance for Basic Needs Fulfillment, serves as a reference for InaSAFE in calculating the estimated minimum needs of evacuees, as outlined in Table 3.

Dominant frequency result
Based on the processing results, it is revealed that the Pleret area exhibits a distribution of dominant frequencies ranging from 0.61 to 10.85 Hz, as depicted in Figure 1.The overall distribution of dominant frequencies in the research area tends to be low, as indicated by the green color legend on the map.The lowest dominant frequency value was recorded in Wonokromo Village at station point S2 with a value of 0.61 Hz.Areas with high dominant frequency values are indicated by colors ranging from yellow to red, scattered to the northeast of Wonolelo Village, dominated by the Semilir Formation, and to the north of Segoroyoso Village, which the Alluvial Formation characterizes.

Figure 1. Dominant frequency map
A relationship exists between dominant frequency values and sediment thickness; as the dominant frequency value increases, the sediment thickness is thinner [15].Amplification is the phenomenon of wave enlargement resulting from significant structural differences between soil layers.A region with a small dominant frequency and high amplification has the potential to experience greater levels of earthquake damage than a region with a sizeable dominant frequency and low amplification [16].

Shear wave velocity (Vs30) result
The value of Vs30 represents the average shear wave velocity calculated down to a depth of 30 meters from the ground surface.The range of Vs30 values in Pleret ranges from 213 to 826 m/s, as depicted in

Peak ground acceleration (PGA) result
The distribution of PGA values in Pleret ranges from 0.16 to 0.26 g as seen in the Figure 3. Based on the map, the PGA values are low and scattered to the northeast of Pleret, represented by the green color legend covering Bawuran and Wonolelo Villages.Meanwhile, the highest PGA value is located to the east of Segoroyoso Village, specifically at observation station point 6.This is due to the calculations involving the distance to the earthquake source.Areas that are relatively close to the earthquake source, in this case, the Opak fault, will have higher PGA values compared to areas that are farther away from the fault.The greater the PGA value in a location, the higher the potential seismic hazard that could occur.

Estimate the number of population and structures affected
Based on the data processing using InaSAFE, an estimated 23,176 structures in Pleret are exposed to the earthquake effects, comprising 16 educational buildings, healthcare buildings, and 23,150 residential with MMI scale VI to VII can be seen in Figure 4. Scale VII is distributed in the Young Volcanic Deposits and Alluvial Deposits, administratively located in Wonokromo, Pleret, and Segoroyoso villages.Meanwhile, scale VI is distributed in the villages of Bawuran and Wonolelo, composed of the Semilir Formation.On scale VI, it may result in minor damage to non-structural parts of the building, such as hairline cracks on the walls and some roof tiles shifting and falling.Meanwhile, on Scale VII, it is likely to cause many building structures to experience light to moderate damage, including numerous cracks in building walls, broken glass, and falling roof tiles [18].
Population exposed to the research area, divided across five villages, can be seen in Table 4. Furthermore, Table 5. presents the estimated minimum weekly evacuation needs, including food, clean water, basic family supplies, hygiene packs, and toilets.

Earthquake hazard level
An earthquake hazard map can be created by calculating the PGA value in an area.Based on the classification in Table 2, it is found that Pleret is included into two earthquake hazard categories, which are classifications high II and high III.Results of calculating the PGA value obtained at the research location showed the highest PGA value 241 gal at microtremor station 6.While the lowest PGA value is 177 gal at station 3. Segoroyoso Village is prone to a higher earthquake hazard compared to other villages due to its location along the Opak fault.According to historical data from the 2006 earthquake, Segoroyoso Village suffered significant damage.The earthquake caused at least 2,155 building damaged and 96 fatalities [19].The closer an area is to the earthquake's source, the greater the risk of an earthquake occurring.

Figure 2 .
The distribution of low Vs30 values is indicated by the red color legend on the map, which covers the western part of Pleret, including Wonokromo and Pleret Villages, and the northern to eastern parts of Segoroyoso Village.The areas with low Vs30 values to the south of Pleret are dominated by the Young Volcanic Deposits of the Merapi Volcano, characterized by flatlands and valleys composed of sand and silt.Meanwhile, the low Vs30 value at measurement point 6 has geological characteristics related to Alluvial Deposits.High Vs30 values are found in the eastern part of the research area, encompassing Bawuran and Wonolelo Villages, predominantly characterized by the Semilir Formation.Based on the classification according to SNI 1726:2019, the research area is categorized into three soil types, explicitly belonging to soil class D (175<Vs30<250 m/s), class C (350<Vs30<750 m/s), and class B (750<Vs30<1500 m/s).Classes C and D represent moderately to stiff soils that dominate this research area due to a significant portion of Pleret being part of the Young Volcanic Deposits formation.Fundamentally, layers with low Vs also have low levels of stiffness.Locations with a low Vs indicate a denser sediment layer than those with a high Vs.The thickness of the sediment layer can also indicate the presence of bedrock; the denser the sediment layer, the greater the possibility that bedrock exists, and vice versa [17].

Figure 4 .
Figure 4. Structures Affected by Earthquake Hazard Map This can be achieved by mapping earthquake hazards based on local geological characteristics through microtremor measurements.
2 disasters are necessary.

Table 2 .
Eathquake hazard level based on PGA value

Table 3 .
Minimum assistance standards

Table 4 .
Estimating population affected

Table 5 .
Estimated minimum needs of evacuees