Seismic hazard analysis using Deterministic Seismic Hazard Analysis (DSHA) Method: A case study in Southern Malang District, East Java

Indonesia is a country that has a high earthquake threat. The United States Geological Survey (USGS) recorded that Indonesia experienced 5,439 earthquakes with a magnitude of Mw 5. Geologically, Indonesia is located about 200 km from the subduction zone of the Indo-Australian and Eurasian plates. Malang Regency faces the subduction of the Indo-Australian and Eurasian plates. Earthquakes that occur in plate subduction areas will propagate in various directions, including to Malang Regency. Thus, Malang Regency, especially in the southern part, has a high earthquake threat. This study aims to calculate the earthquake threat using a deterministic approach. Earthquake threats can be calculated using the Deterministic Seismic Hazard Analysis (DSHA) method. This study obtained the Mc. Guirre and Donovan attenuation function. These two attenuation functions produce different Peak Ground Acceleration values. The calculations with the Mc. Guirre attenuation function has a range of PGA values from 38,76461 gals to 69,78215 gals. Meanwhile, the Donovan attenuation function calculation produces a range of PGA values from 41.97550 gals to 76.21376 gals. Based on the value of ground acceleration during an earthquake, Bantur, Gedangan, and Sumbermanjing sub-districts are the sub-districts with the highest earthquake threat level in Southern Malang Regency. Tirtoyudo Village, Tirtoyudo District, Malang District has a lower PGA value when compared to Gedangan District. However, the amount of damage to buildings in this subdistrict was much higher. Since the building structure in Tirtoyudo Village cannot withstand ground shaking due to the earthquake, there is damage to the structure of the building.


Introduction
Indonesia is located at the confluence of the world's large and several small plates (Syafitri, Bahtiar, & Didik, 2020).This condition causes Indonesia to have a high threat of earthquake disasters (Imani, Arman, & Sari, 2021;Noor, 2018;Rahman & Wijayanto, 2021).A large number of earthquake disasters can prove this from 2000-2021, which reached 5,439 events.The United States Geological Survey (USGS) records that Indonesia experiences hundreds of earthquakes with a magnitude of ≥ Mw 5 every year.Ihsan, & Palloan, 2019).This tectonic activity then makes the Java Island region an active tectonic area.This is what makes Java Island have a high level of seismicity (Widagdo & Permana, 2021).Thus, the East Java Region is also experiencing active plate movement.This tectonic movement triggers an earthquake in the subduction area.
Malang District is an area that has a high threat of earthquakes (Geological Agency, 2021).Geologically, it is located about 200 km from the plate subduction zone.Earthquakes that occur in the plate subduction area will propagate in various directions.The energy of this earthquake propagation then reached the East Java Region.This is what triggers the high threat of earthquakes in Malang District.
The area of Malang District is composed of rocks of tertiary age, some of which have been weathered.In addition, Malang District also has unconsolidated quaternary deposits.These geological conditions can strengthen the effects of earthquake shocks (Geological Agency, 2021).Based on the types of rocks that compose it, the Malang area can be considered earthquake-prone because amplification can occur when seismic waves pass through the medium, resulting in stronger shocks on the surface.
The Regional Agency for Disaster Management (BPBD) of Malang Regency reported two earthquakes occurring in 2021, April 10 and April 11, with the strength of the earthquake reaching a magnitude of 6.1 and 5.5, respectively.The higher the earthquake incident, the bigger the disaster threat (Susilanto, Ngadmanto, Sunardi, & Rohadi, 2019;Yuliana & Tejakusuma, 2016).In the two times earthquake incidents, the total damage to buildings reached 10,811.The data is the total number of 13 sub-districts in the research location.Details of the amount of damage to buildings can be seen in Table 1.Earthquake hazards can be calculated with a deterministic approach.It is done by using Deterministic Seismic Hazard Analysis (DSHA) method.It is a calculation method of Seismic Hazard Analysis by determining the ground motion parameter with the maximum earthquake magnitude and the closest distance from the source of the earthquake from the observation point (Mase, 2020;Tripathi & Zafar, 2016;Verma & Zafar, 2017).This method can be used to calculate the worst-case scenario of an earthquake hazard.The data obtained from USGS does not yet have the same type of magnitude.Therefore, before processing the data, it needs to do magnitude uniformity.This research used moment magnitude (Mw) as the uniformity basis of magnitude type.The moment magnitude (Mw) was used since the earthquake scale measurement of this magnitude type can represent the amount of energy released from the fault plane (Gunawan, Ratri, & Gunawan, 2020;Latifa, Meiliyadi, & Bahtiar, 2022; National Agency for Disaster Management Australia-Indonesia Facility for Disaster Reduction, 2010;Pasau, Bobanto, & Pandara, 2018;Somantri, 2021).Further, the magnitude conversion was done by using the attached formula in Table 3.This research limits on the earthquake magnitude is used.The earthquake used in this study is an earthquake which has Mw ≥ 5 magnitude.This limit is done to avoid underestimated calculation results.The earthquake of Mw ≥ 5 magnitude was chosen as a limitation based on the consideration that earthquakes with a strength of Mw ≥ 5 can have a significant impact on vibrations on the earth's surface and the resilience of building infrastructure (Pangaribuan, Rasimeng, Karyanto, & Rudianto, 2019).In addition, this research used earthquake data for the period 1958 to 2021.

Data Analysis Technique
The data analysis was done by using a deterministic approach.It was conducted by designing a certain earthquake scenario on the bedrock.It was formed by taking into account the earthquake that occurred at a certain location in a certain earthquake magnitude (Naik & Choudhury, 2015;Salsabil, Hilyah, Purwanto, & Fajar, 2018).Further, the deterministic approach is an applicable approach to be conducted to predict the ground motion which was caused by an earthquake (Patel, Solanki, & Thaker, 2022;Resta, Apriliyani, Nasri, & Dewi, 2021).Although it is applicable, the deterministic approach has flaws where it only focuses on the distance and magnitude of the earthquake, while the influence of the uncertainty of the earthquake is not considered (National Agency for Disaster Management Australia-Indonesia Facility for Disaster Reduction, 2010;Salsabil et al., 2018).
The DSHA analysis includes decisions and subjective opinions from relevant experts in determining the potential for an earthquake.Consequently, the experts have different perspectives from one another.This difference caused them difficulty agreeing on the earthquake potential (Sunardi & Sulastri, 2016;Utomo & Wahyudi, 2017).The powerless to make an agreement makes each expert have different formulas and constants in their calculations.Additionally, the DSHA calculation in this study used attenuation functions: 1) Mc.Guirre, and 2) Donovan.

Peak Ground Acceleration (PGA)
The acceleration is a parameter used to state the change in the object's speed from rest to a certain speed regardless of direction (Pakuani & Kurniawan, 2021;Suwandi, Sari, & Waslaluddin, 2017).The Peak Ground Acceleration (PGA) is the largest ground vibration acceleration value that has ever occurred in an area due to the propagation of earthquake waves (Hason, Hanoon, & Abdulhameed, 2021;Salahshoor, Lyubushin, Shabani, & Kazemian, 2018).The maximum PGA that was done at a certain point due to the earthquake can be calculated by seeing the magnitude and hypocenter distance and the predominant period of soil where it was located (Palupi, Raharjo, & Alfiani, 2020).
Moreover, the PGA outcome could indicate the earthquake threat in a certain place (Hsu, Wu, Liang, Kuo, & Lin, 2020;Shiuly, 2018).The PGA value is directly proportional to the threat of earthquakes in the region.In other words, the greater the PGA value that has occurred in a place, the greater the threat of an earthquake that may occur (Dari & Pujiastuti, 2021;Pakpahan, Tambunan, Mannesa, & Tambunan, 2021).It happened because the higher the PGA score, the greater the acceleration of ground vibrations in that region (Aditama & Fatimah, 2020).Thus, with a high PGA value, the primary effects of earthquakes will also be higher (Halimatusadiyah, Suryanita, & Yusa, 2021).The primary effect of earthquake is the effect of direct earthquake (Setyorini, 2020).This effect includes: 1) building structures damage, 2) fallen tree, 3) landslide, and 4) others.This damage was caused by the earthquake vibrations generated (Sungkowo, 2016).
In addition, the PGA values can be expressed in terms of g (gravitational acceleration = g) or cm/second square (1 g = 9,81 cm/s 2 ).Moreover, PGA could also be stated as gals where 1 gals equals 0.01 cm/second square (1 G = 981 gals).The PGA determination using equations is not always correct and has different methods for each equation depending on the parameters used.The PGA value at a certain point is also influenced by the geological structure of the area from the research point (Aditama & Fatimah, 2020;Murdiantoro, Sismanto, & Marjiyono, 2018).In addition to the geological structure, the PGA value is also influenced by the hypocenter distance.The closer an area is to the source of the earthquake, the greater the acceleration of the ground surface in that area (Douglas, 2002).The research flowchart is presented in Figure 1.

Geological Area of the Southern Part of Malang
The geological area of the southern part of Malang is dominated by Quaternary, Neogene, and Paleogene rocks.There are seventeen types of geology located in the southern part of Malang, consisting of Qal, Qas, Qlks, Qlv, Qpj, Qpkb, Qpvb, Qvj, Qvk, Qvs, Qvtm, Tmn, Tmw, Tmwl, Tomi, Tomm, Tomt.The geological difference led to the differences in dispersed rock materials.Different rock materials will give different responses and impacts when transmitted by earthquake waves (Giang, Pham, & Nguyen, 2022;Pirhadi, Tang, & Yang, 2019).This difference can occur due to differences in the level of rock cohesiveness, density, and rock strength (Karimzadeh, Feizizadeh, & Matsuoka, 2017;Maringue et al., 2022).Paleogene-aged rocks have material that is loose, not solid, and not compact (Fasulo & Ridgway, 2021 Therefore, when exposed to energy from earthquake waves, the material will be easily moved. The southern part of Malang district has an active minor fault.The existence of a minor fault can trigger a high peak ground acceleration in the surrounding area when an earthquake wave passes through the fault.Based on the geological map of the area, this location has four active minor faults that are scattered in the eastern part.Geologically, the existence of the active minor fault located in the southeast of Malang causes this area to have a higher threat of earthquake disasters compared to areas that do not have minor faults.Thus, the western part has a lower threat of earthquake disasters when compared to the eastern part.A geology map of southern Malang can be found in Figure 2.  4. The minimum value is obtained from calculating locations that are far from the epicenter, while the maximum value is obtained from calculations at locations close to the epicenter (Gandomi, Soltanpour, Zolfaghari, & Gandomi, 2016;Hason et al., 2021).In the deterministic approach, the results obtained were influenced by the distance of the earthquake hypocenter from the research location and the magnitude of the earthquake (Caccavale, Sacchi, Spiga, & Porfido, 2019;Concha et al., 2020;Pouryari, Mahboobi Ardakani, & Hassani, 2022).Therefore, the farthest location from the earthquake will have a small PGA value, while the closest location will have a great PGA value (Kowsari, Ghazi, Kijko, Javadi, & Shabani, 2021;Mehta & Thaker, 2022;Stewart, Luco, Hooper, & Crouse, 2020).In other words, the closer a location is to the earthquake hypocenter, the greater the PGA value (Douglas, 2002;Mendoza, Ramos, & Dimalanta, 2022).
According to the calculations of Donovan and Mc.Guire formula, it was revealed that the results of both formulas have a difference.The maximum and minimum values from the calculation of the Donovan attenuation function have a greater value when compared to the results calculated using the Mc.Guirre attenuation function (Sari & Rahmatullah, 2021;Sung, Abrahamson, & Huang, 2021).In its application in Southern Malang District, Donovan's attenuation function is closer to the actual value.It is shown in the calculation results at coordinates -8.347S 112.59E.The calculation of the Donovan attenuation function at these coordinates produces a value of 71.26312 gals.This value is close to the actual value, where the PGA value at that coordinate is 75.1738gals.At the same time, the value for the calculation uses the Mc.Guirre attenuation function on the coordinates -8.347S 112.59E is 65.31240 gals.The real PGA value can be obtained by taking the field measurement using the accelerometer.

Earthquake Hazard Map
The map of earthquake threat is designed according to the PGA value.The more the PGA value, the greater the acceleration of the ground vibrations caused by an earthquake (Tauhidur & Chhangte, 2021).The greater the ground acceleration, the greater the shock caused by the earthquake (Gogoi, Baruah, & Sharma, 2023;Wibowo & Sembri, 2016).Thus, the threat of earthquakes in the area is also getting bigger.The distribution and variation of PGA values will affect the level of earthquake threat in an area.

Earthquake Hazard Map with M.Guirre Attenuation Function Approach
The PGA value with Mc.Guirre attenuation function is then visualized in the form of earthquake hazard map.Additionally, the distribution of PGA values with the Mc.Guirre attenuation function greatly affects the earthquake hazard (Ulfiana, Rumy, Pratama, & Ariy, 2018;Zera, Fauziah, Nafian, & Ramadhani, 2022).The earthquake hazard map with the Mc.Guirre attenuation function approach can be seen in Figure 3.

Figure 3. Peak Ground Acceleration Spread Map in Southern Part of Malang District
According to the earthquake hazard map in the southern part of Malang District, it was discovered that Bantur, Gedangan, and Sumbermanjing Subdistricts are the places with a high threat of earthquake disaster.These three subdistricts have a zone with ground acceleration with a variation 52.5-76.2gals.These three subdistricts are directly adjacent to the sea, so this area meets the subduction of the Indo-Australian and Eurasian plates.In addition, Donomulyo, Tirtoyudho, and Ampelgading Subdistricts are also directly adjacent to the sea.However, Bantur, Gedangan, and Sumbermanjing Subdistricts have more indented structures into the sea.It brings them closer to the earthquake source.The magnitude of the earthquake threat in these areas is strongly influenced by the distance from the earthquake source, where the three subdistricts have the closest distance to the subduction of the Indo-Australian and Eurasian plates.
Besides, the other subdistricts which are further away from the earthquake source have smaller variations in PGA values.Kalipare, Sumberpucung, Kepanjen, Gondanglegi, Turen, Dampit, Tirtoyudo, and Ampelgading Subdistricts have PGA value variations 45-55 gals.Based on the value of ground movement acceleration, the threat of earthquakes in this area is smaller when compared to the Bantur, Gedangan, and Sumbermanjing Districts.
Further, according to the earthquake hazard map with Mc.Guirre attenuation, Southern part of Malang District is dominated by the ground movement acceleration value of 55.1-57.5 gals reaching 28,589.06Ha or 18.09% of the total area.The second position is occupied by a PGA value of 57.6-60 gals with an area of 24,398.64Ha or 15.78% of the total area.In contrast, the area with the lowest PGA value is 47.6-50 gals which only has an area of 14,928.84Ha or 2.84% of the total area (The National Center for Earthquake Studies, 2017).Details of the area of PGA values with the Mc.Guirre attenuation function can be seen in Table 7.

Earthquake Hazard Map with Donovan Attenuation Function Approach
The approach to earthquake hazard map is different between Mc.Guirre and Donovan attenuation function since both have different earthquake potential constants, making different results.The earthquake hazard map is also different, with a distribution of PGA values.Besides, the Earthquake Hazard Map with the Donovan attenuation function approach can be seen in Figure 4. Bantur, Gedangan, and Sumbermanjing Subdistricts are still the three subdistricts with the highest earthquake threat level.However, it is different from the earthquake threat map with Mc.Guirre attenuation function, which has PGA variation range of 57.5-80 gals.The ground acceleration value in Donomulyo District is also different.In Donomulyo, it has value variations of 50.0-70.0gals.Based on this value, Donomulyo District has a higher earthquake hazard when compared to the earthquake hazard map using Mc.Guirre attenuation function.
Meanwhile, even though Kalipare, Pagak, Sumberpucung, Kepanjen, Gondanglegi, Turen, Dampit, Tirtoyudo, and Ampelgading Districts have lower earthquake threats, the ground acceleration value in this area is still higher when compared to the ground acceleration using the Mc.Guirre attenuation function.Accordingly, this area has a peak ground acceleration variation of 45.0-60.0gals.Otherwise, using the Donovan attenuation function, this area has a ground acceleration variation of 45.0-65.0gals.Thus, the earthquake hazard level using the Donovan attenuation function is higher when compared to the earthquake hazard level using the Mc.Guirre attenuation function.
The southern subdistricts of Malang were dominated by PGA value 55.1-57.5 gals, which the large reached 32.562,14 Ha or 18.09% of the total area.In the meantime, the PGA value 67.7-70.0gals be the value with the least area, namely 1490.37 Ha or 0.83% of the total area.Further, PGA value 70.1-80 gals has an area of 8,117.02Ha or 4.5% of the total area.Although the area with a PGA value of 67.6-80 gals only covers 5.32%, this area is an area with the highest threat of earthquakes in the Southern Malang Region.Details of the area of PGA values with the Donovan attenuation function can be seen in Table 8.Based on the calculation results, the earthquake hazard map using the Donovan attenuation function has better accuracy when compared to the Mc.Guirre attenuation function.This is based on the peak ground acceleration value with the Donovan attenuation function closer to the actual peak acceleration value (Ahnan & Agung, 2021;Anugrayanti, Arsyad, & Tiwow, 2021;Ariani, 2020;Suwandi et al., 2017).The peak ground acceleration value obtained using the accelerometer at coordinates -8.347S, 112.59E is 75.1738gals (Sinadinovski, Pekevski, Dojcinovski, & Cernih, 2018).This value is only 3.9106 gals greater than the results calculated using the Donovan attenuation function at these coordinates.While calculations using the attenuation function have a difference of 9.8614 gals.Thus the earthquake hazard distribution map using the Donovan attenuation function is more accurate when compared to the Mc.Guirre attenuation function when applied to the Southern part of Malang District.

Level of Damage to Buildings Due to Earthquakes in Research Fields
The subdistricts in the Southern part of Malang Regency have almost the same damage but with different amounts for each sub-district.In general, building damage is caused by the inability of building structures to withstand ground vibrations caused by earthquakes (Erdogan & Yilmaz, 2019;Wang et al., 2016).Building damage was identified into three classes: 1) Minor Damage, 2) Moderate Damage, and 3) Major Damage.Details of the level of damage to buildings can be seen in Table 9.Based on this classification, it is possible to classify the level of damage to buildings in each sub-district at the research location.By conducting a survey, it can be seen that each subdistrict has three levels of damage, from minor damage to major damage.Documentation sampling was carried out at each level of damage to buildings in each district.Documentation of the level of damage to buildings in each of the three districts with the highest amount of damage can be seen in Table 10.

Application of Detailed Post-Earthquake Damage Scale Mapping in Tirtoyudo Village
Detailed scale mapping, also known as microscale mapping, is carried out to map the damage after the earthquake with a magnitude of 6.1 on April 10, 2021, and a magnitude of 5.5 on April 11, 2021.Based on data from BPBD Malang Regency, Tirtoyudo Village is one of the villages with the highest amount of building damage.According to this, Microscale Mapping was carried out to determine the distribution of the level of damage to buildings in Tirtoyudo Village.
Data acquisition is done using UAV, the data is then visualized into a map.Based on the data acquisition result map, it can be seen that the damage to buildings in Tirtoyudo Village is almost evenly distributed.Light damage to buildings is the level of damage with the largest number of houses.Then, it was followed by moderate and severe levels of damage.The building damage map can be seen in Figure 5.  Tirtoyudo District was far greater than that in Gedangan District, namely, 2707 damages, while Gedangan District only amounted to 232 damages.The damage was caused by the fact that in the village of Tirtoyudho, almost all were permanent buildings made of bricks.Buildings made of wood are rare.Based on the survey results, only two buildings were made of wood.Buildings made of wood have a good ability to respond to ground vibrations, so they are more resistant to earthquakes (Alih & Vafaei, 2019;Hariyanto, & Widyowijatnoko, 2022;Queirós, Calejo Rodrigues, & Pereira, 2016;Sugino, Tanishima, Tanaka, Kashiwa, & Hayashi, 2022).In addition, based on interviews conducted with stakeholders, this can happen because of the building structure.Communities build buildings without regard to their resistance to earthquakes.Thus, when an earthquake occurs, the building structure cannot withstand the ground's vibrations, resulting in damage.

Conclusion
Southern Malang District has varying Peak Ground Acceleration (PGA) values.These variations are different for each attenuation function.Mc.Guirre attenuation function has a variety of PGA values from 38.76461 gals to 69.78215 gals.The Donovan attenuation function varies from 41.97550 gals to 76.21376 gals.The calculated value with the Donovan attenuation function is closer to the actual ground acceleration value.The PGA value with the Donovan attenuation function only has a difference of 3.9106 gals with the accelerometer measurement results.Based on the PGA value, the Southern part of Malang Regency has various earthquake threats.The sub-districts with the highest threat of earthquakes are Bantur, Gedangan, and Sumbermanjing Subdistricts.They have the shape of an area that juts into the sea.Thus, these three districts are the districts closest to the subduction of the Indo-Australian and Eurasian Plates, where the plate's subduction is one of the main sources of earthquakes.Tirtoyudo District has a lower PGA value compared to Gedangan District.However, the Tirtoyudo District had a much larger amount of damage, namely 2707 damage.This is because the building structures in Tirtoyudo District cannot withstand the ground shaking caused by the earthquake, so the buildings are damaged.Other subdistricts also experienced similar damage, although not as much as the amount of damage in Tirtoyudo Subdistrict.

Figure
Figure 1.Research Flowchart

Figure
Figure 2. Geological Map

Figure 4 .
Figure 4. Peak Ground Acceleration Distribution Map in the Southern Part of Malang District

Figure 5 .
Figure 5. Map of Distribution of Levels of Damage to Buildings Due to the Earthquake Based on the PGA value, this area has a smaller ground acceleration value than Gedangan District, which has a PGA value of 74.28163 gals.Tirtoyudho District only has a ground acceleration of 58.31669 gals.Based on this value, Tirtoyudo District, especially Tirtoyudo Village, should have a smaller threat of earthquakes.However, the amount of damage in

Table 5 . Value of Peak Ground Acceleration (gals) in Each District
Hamdy, Gaber, Abdalzaher, & Elhadidy, 2022).Donovan attenuation function has the PGA value range 41.97550 gals -76.21376 gals.Meanwhile, the PGA value for Mc.Guire attenuation function ranges from 38.76461 gals -69.78215 gals.The minimum and maximum calculation values can be seen in Table 6.