02이창욱1(13~22)(Fig.2~4)ok.qxp_원격37-1_2021 2021. 2. 24. 오전 11:01 페이지 13

Korean Journal of Remote Sensing, Vol.37, No.1, 2021, pp.13~22 ISSN 1225-6161 ( Print ) https://doi.org/10.7780/kjrs.2021.37.1.2 ISSN 2287-9307 (Online)

Article

Damage Proxy Map (DPM) of the 2016 Gyeongju and 2017 Pohang Earthquakes Using Sentinel-1 Imagery

† 1) · 2) Arip Syaripudin Nur Chang-Wook Lee

Abstrac t: The M L 5.8 earthquake shocked Gyeongju, Korea, at 11:32:55 UTC on September 12, 2016. One year later, on the afternoon of November 15, 2017, the M L 5.4 earthquake occurred in Pohang, South Korea. The earthquakes injured many residents, damaged buildings, and affected the economy of Gyeongju and Pohang. The damage proxy maps (DPMs) were generated from Sentinel-1 synthetic aperture radar (SAR) imagery by comparing pre- and co-events interferometric coherences to identify anomalous changes that indicate damaged by the earthquakes. DPMs manage to detect coherence loss in residential and commercial areas in both Gyeongju and Pohang earthquakes. We found that our results show a good correlation with the Korea Meteorological Administration (KMA) report with Modified Mercalli Intensity (MMI) scale values of more than VII (seven). The color scale of Sentinel-1 DPMs indicates an increasingly significant change in the area covered by the pixel, delineating collapsed walls and roofs from the official report. The resulting maps can be used to assess the distribution of seismic damage after the Gyeongju and Pohang earthquakes and can also be used as inventory data of damaged buildings to map seismic vulnerability using machine learning in Gyeongju or Pohang.

Key Word s: Gyeongju earthquake, Pohang earthquake, Sentinel-1, Damage proxy map

1. Introduction damage in and around the city (Kang et al ., 2019). The Pohang earthquake was just 40 km from the

An M L 5.8 earthquake shocked Gyeongju at 11:32:55 Gyeongju earthquake. The depth of the mainshock was UTC on September 12, 2016. The epicenter was 8.7 approximately 7.0 km, and several aftershocks were km from the south of the city and 15 km beneath the reported around the area (Kim et al ., 2018). Both surface (Kim et al ., 2016). One year later on November earthquakes were the largest to occur in the Korean 15, 2017, an M L 5.4 earthquake occurred in Pohang, Peninsula since earthquake monitoring was initiated by South Korea at 05:29:31 UTC, causing widespread the Korea Meteorological Administration (KMA) in

Received January 28, 2021; Revised February 6, 2021; Accepted February 9, 2021; Published online February 15, 2021 1) Combined MS/PhD Student, Department of Smart Regional Innovation, Kangwon National University 2) Professor, Division of Science Education, Kangwon National University

† Corresponding Author: Chang-Wook Lee ([email protected]) This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

– 13 – 02이창욱1(13~22)(Fig.2~4)ok.qxp_원격37-1_2021 2021. 2. 24. 오전 11:01 페이지 14

Korean Journal of Remote Sensing , Vol .37, No .1, 2021

1978 and were among the most destructive events to Gyeongju earthquake did not cause surface deformation occur in Korea during the past century (Grigoli et al ., around the epicenter (Park et al ., 2018). On the other 2018; Woo et al ., 2019). hand, the Pohang earthquake resulted in surface Gyeongju earthquake resulted in 5368 damaged deformation in the eastern of the epicenter about properties with approximately $9.5 M (USD), 111 ±5 cm. This was the first time that surface deformation victims, 23 injured people (Han et al ., 2020). While could be detected by synthetic aperture radar (SAR) Pohang earthquake injured 135 residents, displaced imagery in Korea. Considering the uniqueness of more than 1,700 people into emergency housing, and both earthquakes, we choose Gyeongju and Pohang caused more than $75 M (USD) indirect damage to earthquakes as our study case. over 57,000 structures and over $300 M (USD) of total Damage proxy map (DPM) method using SAR economic impact, as estimated by the Bank of Korea satellites data has been shown to be useful for damage (Lee, 2019). More than 100 heritage buildings and mapping following an earthquake and other natural monuments sustained damage from the earthquakes disaster events, including the 2015 M W 7.8 Gorkha, (Sun, 2019). In Gyeongju, there is the Gyeongju Nepal earthquake using COSMO-SkyMed and ALOS- Historic Area that registered as a UNESCO world 2 satellites (Yun et al ., 2015), the 2019 typhoon cultural Heritage in November 2000, an area that Hagibis, Japan using Sentinel-1 satellites (Tay et al ., embodies the time-honored history and culture of 2020), the 2019 M W 7.1 Ridgecrest Earthquake in Gyeongju, the ancient capital of the Silla Kingdom California using Sentinel-1 satellites (Hough et al ., (57 BC-AD 935). Some damages were found in this 2020), and the 2014 eruption of Kelud volcano area, such as Dabotap Pagoda (dislocated banister), (Indonesia) using COSMO SkyMED satellites (Biass Cheomseongdae Observatory (shifted and tilted), and et al ., 2021). DPM method is part of an ongoing Gyeongju Gyochon Traditional Village (cracked walls) collaborative effort between the Jet Propulsion (Doo, 2016). This was the most damaging earthquake Laboratory (JPL) and the California Institute of to strike the Korean Peninsula for centuries. Technology, called the Advanced Rapid Imaging and Several studies have been carried out toward Analysis (ARIA) project. Gyeongju and Pohang earthquakes, such as groundwater This study aims to produce DPMs derived from system responses assessment to the Gyeongju earthquake European Space Agency (ESA) Sentinel-1 synthetic (Kim et al ., 2019), investigating related factor associated aperture radar (SAR) imagery to explore the distribution with the Gyeongju earthquake (Jin et al ., 2020), of local damage after the Gyeongju and Pohang detecting liquefaction phenomena using Sentinel-2 and earthquakes. DPMs were generated from subtracting Landsat-8 after Pohang earthquake (Baik et al ., 2019), interferometric coherence of pre- and co-events to investigating whether the Pohang earthquake was identify anomalous changes in the ground-surface induced events (Kim et al ., 2018), and modeling properties (Hough et al ., 2020). The resulted maps are static slip using interferometric synthetic aperture expected to be useful as a basis for the risk assessment, radar (InSAR) (Song and Lee, 2019). A government- prioritizing evacuation, and mitigation efforts. commissioned research team was formed and after a year-long study of the Pohang earthquake, the team concluded in 2019 that the earthquake was triggered 2. Study Area by enhanced geothermal stimulation (EGS) project (Lee, 2019). As the largest quake ever recorded, the The study areas were Gyeongju and Pohang,

– 14 – 02이창욱1(13~22)(Fig.2~4)ok.qxp_원격37-1_2021 2021. 2. 24. 오전 11:01 페이지 15

Damage Proxy Map (DPM) of the 2016 Gyeongju and 2017 Pohang Earthquakes using Sentinel-1 Imagery

observed to have occurred with magnitude 3.0 or more until May 31, 2018. A magnitude 4.3 occurred near the epicenter, 2 hours after the mainshock, and a magnitude 4.6 earthquake occurred 4 km to the southwest at around 20:03:04 UTC on February 11, 2018 (Kim et al ., 2018). Several faults are distributed within the study areas, including Dongrae, Moryang, Miryang, Ulsan, Wangsan, and Yangsan (Ellsworth et al ., 2019). Therefore, the probability of earthquake occurrence in Gyeongju and Pohang is considered relatively high.

3. Method

DPMs generated from the comparison of pre- and co-event SAR images can help identify damage caused Fig. 1. Gyeongju and Pohang, South Korea. The red and by earthquakes using remote sensing imagery (Yun et purple rectangles are the coverage of Sentinel-1 during descending observation. The blue and al ., 2015). The method relies on the reduction in the green lines indicate the study areas. coherence of the radar echoes between satellite-based SAR images taken before and after the earthquake Gyeonsangbuk-do, South Korea (Fig. 1). Gyeongju and to identify anomalous changes in ground surface Pohang are side-by-side cities and are located on the properties. Coherence measures the change in radar south-eastern edge of the Korean Peninsula. Gyeongju backscatter from the ground, a proxy for the ground- has a population of 254,853, an area of 1,324.82 km 2, surface properties changes. A low coherence implies a and consists of 23 administrative districts (Gyeongju large change to the ground surface that reflected the City Hall, 2021). Pohang has a population of 519,216, SAR radiation (Zebker and Villasenor, 1992). Changes an area of 1,127 km 2, and consists of 29 administrative can be caused by damage to the ground itself or damage districts (Pohang City Hall, 2021). to structures. Gyeongju earthquake was accompanied by 600 For the Gyeongju earthquake, we acquired two aftershocks, including M L 5.1 foreshock that occurred Copernicus C-band Sentinel-1 single look complex near the mainshock at 10:44:32 UTC and the largest (SLC) SAR images (Table 1), taken before the event aftershock (M L 4.5) occurred at 11:33:58 UTC on (August 11 and 23, 2016) and one after the event September 19, 2017 (Jin et al ., 2020). A series of (September 16, 2016). Pohang is located in between aftershocks of the Pohang earthquake was also two frames (470 and 475) path 61 of the Sentinel-1

Table 1. Parameters of Sentinel-1 SAR data used for damage mapping

Coverage Pre-event scene date Orbit Direction Co-events scene date Frame Path 11 August 2016 Gyeongju Descending 16 September 2016 470 61 23 August 2016 23 October 2017 Pohang Descending 16 November 2017 470 & 475 61 4 November 2017

– 15 – 02이창욱1(13~22)(Fig.2~4)ok.qxp_원격37-1_2021 2021. 2. 24. 오전 11:01 페이지 16

Korean Journal of Remote Sensing , Vol .37, No .1, 2021

imagery (Fig. 1); therefore, we obtained six Sentinel-1 co-event interferometric pair ( γco-event > γpre-event ), so SLC images for the Pohang earthquake, with four that negative COD is obtained. A positive COD (or scenes before the event on October 23, 2017, and coherence loss) indicates surface changes between November 4, 2017, and two scenes after the event on the co-event scenes spanning the events, such as November 16, 2017. Furthermore, we need to merge major damage to a building significantly increase the the scenes before processing. The images were co- interferometric phase variance causing decrease in registered, the August 23, 2016 and November 4, 2017 coherence. Hence, the loss of coherence is most scenes as reference. This co-registration process has effective for detecting damage in built-up areas been done at sub-pixel accuracy to match one another caused by earthquake. However, COD is generally less scene perfectly. The co-registered images were cropped effective and less reliable in vegetation areas where to generate images that only contain the study area. coherence change may be random. Therefore, this We used the complex pixel value, c of the pre- study focused on DPM in urban, built-up areas as they processed SLCs for the change detection analysis, can be detected easily in SAR imagery. The greater loss where damage is inferred from loss of coherence or in coherence generally correlates with greater severity decorrelation between SAR images (Yun et al ., 2015). of the change, such as fully collapse building causing We computed the pre- and co-event interferometric more significant coherence loss than partial collapse coherences, γ (Equation 1 ), from pair of SLCs before (Tay et al ., 2020). the event and another pair spanning across the event, The threshold for significant coherence loss can be respectively (Tay et al ., 2020). chosen by comparing observed coherence changes with

* | < c1 c2 > | reported damage and areas in which it is known that no γ = * * , 0 ≤ γ ≤ 1 (1) < c1 c1 > < c1 c2 > damage occurred. Yun et al . (2015) compared DPM

where c1 and c2 are complex pixel values of two co- with National Geospatial-Intelligence Agency (NGA) registered SAR images and * denotes the complex analysis and the United Nations Operational Satellite conjugate. The resulting coherence is ranging from 0 Application Programme (UNOSAT) damage assessmen t (incoherent) to 1 (coherent). The coherence is equal to map. Tay et al . (2020) used high-resolution aerial 1 if the observation is identical in the two images imagery from the Geospatial Information Authority because of the stable object like buildings in the scene. of Japan (GSI). Here, coherence loss thresholds for The pre-event coherence represented change unrelated DPMs were chosen by considering the damaged area to the event and was assumed to be the background from the Korea Meteorological Administration (KMA) value. Then, we obtained a coherence difference report of the Gyeongju and Pohang earthquake (Korea

(COD) by subtracting γpre-event from γco-event . Therefore, Meteorological Administration, 2018). the process could generate COD ranging from -1 to 1. A negative COD (or coherence gain) usually indicates surface changes occurring between the pre-event 4. Results and discussion scenes and is associated with changes not related to the event. Coherence gain could happen in agriculture area 1) Damage Proxy Map (DPM) when the fields full of crops. Then harvested during the The DPMs were generated from the Sentinel-1 period time of the pre-event interferometric pair, the dataset and geocoded to the Shuttle Radar Topography area has low coherence. After harvesting, leaving an Mission (SRTM) digital elevation model (DEM) 1 empty field, the area has a greater coherence in the arcsecond. Fig. 2 reveals the map view of Sentinel-1

– 16 – 02이창욱1(13~22)(Fig.2~4)ok.qxp_원격37-1_2021 2021. 2. 24. 오전 11:01 페이지 17

Damage Proxy Map (DPM) of the 2016 Gyeongju and 2017 Pohang Earthquakes using Sentinel-1 Imagery

DPMs overlaid with images from the 4-band PlanetScope According to the land cover map derived from the true-color scene taken on November 2, 2016, for the Korea Institute of Geoscience and Mineral Resources Gyeongju earthquake and November 16, 2017, for the (KIGAM), Fig. 2(a) reveals DPM after the Gyeongju Pohang earthquake. The assessment technique is most earthquake over residential areas, commercial areas, sensitive to the destruction of the built environment. agriculture, and vegetation areas. Fig. 2(b) and Fig. 2(c) Pixels are set to be relatively transparent that show the widespread COD in Gwangmyeong-dong corresponding to areas where decorrelation did not and Seonggeon-dong, respectively. These areas consist significantly change during the time spanning the of residential and commercial areas. Fig. 2(d) shows earthquake suggesting little to no destruction. Increased DPM that indicates COD over residential, commercial, opacity of the radar image pixels reflects increasing industrial, agriculture, and vegetation areas affected ground and building change or potential damage. Color by the Pohang earthquake. Some of the spatially large range from yellow to red indicates an increasingly and strong coherence loss in Pohang correspond to significant coherence change in the area covered by the commercial and residential areas located in Heunghae- pixel. Each pixel in the DPMs was registered to the eup (Fig. 2(e)), while Fig. 2(f) located in Duho-dong SRTM DEM and had a corresponding dimension of district, consists of residential areas. about 30 m. The DPMs show distribution of coherence loss areas 2) Comparison of DPM with official report after the earthquakes throughout Gyeongju and Pohang. We found a good correlation between the DPMs and

Fig. 2. Damage proxy maps (DPMs) overlaid over PlanetScope images for (a) the 2016 Gyeongju in (b) Gwangmyeong- dong and (c) Seonggeon-dong and for (d) the 2017 Pohang earthquake in (e) Heunghae-eup and (f) Duho-dong. Yellow to red pixels indicates increasingly more significant potential damage.

– 17 – 02이창욱1(13~22)(Fig.2~4)ok.qxp_원격37-1_2021 2021. 2. 24. 오전 11:01 페이지 18

Korean Journal of Remote Sensing , Vol .37, No .1, 2021

the released map from the KMA report about Gyeongju these areas suffered MMI VII to VIII. KMA defines and Pohang earthquakes from our qualitative validation. MMI VII to VIII as major structural parts such as The map was derived from field surveys and damage pillars, walls, and roofs, even in well-designed and built surveys data from local governments. The map shows buildings. Fig. 3 also shows some areas that suffered the distribution of Gyeongju and Pohang earthquakes MMI V to VI; however, the DPMs did not show magnitude using the Modified Mercalli Intensity any COD in these areas. The scales defined that the (MMI) scale (Korea Meteorological Administration, damages are inside buildings, such as minor crack 2018). Fig. 3(a) show DPM of the Pohang earthquakes walls and damage caused by dropping objects or tiles; indicated by red dots and Fig 3(b) show DPM of the therefore, SAR cannot detect a significant coherence Gyeongju earthquake indicated by blue dots. The change. DPMs overlaid on the released damage map from KMA reported that damages from the earthquakes KMA. The results were similar to the KMA report that are mainly distributed from MMI V to VIII. The

Fig. 3. DPMs of the (a) Pohang and (b) Gyeongju earthquakes indicated by blue and red dots, respectively, overlaid on the released damage map from KMA of the Gyeongju and Pohang earthquakes using the MMI scale. Modified from (Korea Meteorological Administration, 2018).

– 18 – 02이창욱1(13~22)(Fig.2~4)ok.qxp_원격37-1_2021 2021. 2. 24. 오전 11:01 페이지 19

Damage Proxy Map (DPM) of the 2016 Gyeongju and 2017 Pohang Earthquakes using Sentinel-1 Imagery

damage from the Gyeongju earthquake impacted an area of 1.09 km 2 and the Pohang earthquake yielded urbanization/dry areas and agriculture areas. Of the an area of 1,32 km 2. KMA reported a total of 4.316 total 4,316 earthquake damage, 68% occurred in earthquake damage, 69 damaged buildings by the residential areas, followed by 7% in commercial areas, Gyeongju earthquake, and 20,074 earthquake damage 7% in public facilities, 5% in rice fields, and 3% in with 504 damaged buildings by the Pohang earthquake fields. The Pohang earthquake damage occurred in (MMI VII and VIII). The Pohang earthquake caused urbanization/dry areas, agriculture areas, forest areas, more damage than the Gyeongju earthquake, one of and bare land. Among them, the damage to residential which was due to the depth of the epicenter. The areas was the largest. Of the total 20,074 earthquake shallower the epicenter of an earthquake, the more damage in Pohang, 66% occurred in residential areas, damage it causes. The epicenter depth of the Pohang followed by 12% in other areas, 9% in coniferous earthquake was 7 km, while the epicenter depth of the forests, 8% in commercial areas, and 1% in rice fields. Gyeongju earthquake was 15 km. Also, the surface In the case of MMI VIII, both earthquakes mainly deformation caused by the Pohang earthquake affected damage residential areas. However, 25% of the the occurrence of damage in Pohang. Gyeongju earthquake damaged area was rice filed (Korea Meteorological Administration, 2018). 3) Seismic characteristic analysis Here, the total damaged areas from DPMs were Comparing the seismic waveform and spectrum of calculated by multiplying the total number of pixels by both earthquakes, high-frequency energy was relatively the pixel cell size, which Gyeongju earthquake yielded high in the Gyeongju earthquake, and low-frequency

Fig. 4. DPM in (a) Gwangmyeong-dong corresponding to collapsed (b) roof and (c) walls after Gyeongju earthquake and (d) DPM corresponding to damaged (e) walls after Pohang earthquake in Handong Global University.

– 19 – 02이창욱1(13~22)(Fig.2~4)ok.qxp_원격37-1_2021 2021. 2. 24. 오전 11:01 페이지 20

Korean Journal of Remote Sensing , Vol .37, No .1, 2021

energy was large in the Pohang earthquake (Korea damaged buildings to map a seismic vulnerability using Meteorological Administration, 2018). Generally, low- machine learning in Gyeongju or Pohang. frequency energy impacts high-rise buildings, and high- frequency energy can damage low-rise buildings. Therefore, the Gyeongju earthquake caused damage to 5. Conclusion roofs and fences, and cracked walls on the first and second floor was common. The Pohang earthquake, The Damage Proxy Maps were produced from which has low-frequency energy, damage high-rise ESA Sentinel-1 SAR data for the 2016 Gyeongju buildings such as school and apartments. earthquake and the 2017 Pohang earthquake. DPMs Fig. 4 shows significant decorrelation associated were generated by comparing pre- and co-events with damaged buildings after the Gyeongju and Pohang interferometric coherences to identify anomalous earthquake. Fig. 4(a) shows DPM in Gwangmyeong- changes that indicate damaged by the earthquakes. dong corresponding to collapsed (Fig. 4(b)) walls DPMs manage to reveal coherence loss in residential and (Fig. 4(c)) roof of a house after the Gyeongju and commercial areas in both Gyeongju and Pohang earthquake (Ministry of Public Safety and Security, earthquakes. We found that our results show a good 2017). Fig. 4(d) shows DPM over Handong Global correlation with the KMA report with MMI values of University associated with (Fig. 4(e)) damaged walls more than VII. The color scale of Sentinel-1 DPMs by the Pohang earthquake (Lee, 2017). Based on the indicates an increasingly significant change in the area MMI scale by KMA, these were categorized as MMI covered by the pixel, delineating collapsed walls and VII or VIII. roofs from the official report. With the frequent six-day With frequent revisits for six days and freely revisit and freely accessible of the Copernicus Sentinel- accessible of the Copernicus Sentinel-1 satellites, the 1 satellites, the Sentinel-1 dataset could provide a fast Sentinel-1 dataset can provide fast and reliable damage and reliable damaged map after a disaster occurred maps after a disaster using the DPM method to assist using the DPM method to assist mitigation strategies mitigation strategies to reduce casualties. However, and reduce the number of victims. The results can also given the SAR pixel size, the DPM from Sentinel-1 be used as inventory data of damaged buildings to map does not, as expected, have sufficient resolution to seismic vulnerability using machine learning in identify all limited damage to individual structures and Gyeongju or Pohang. may not reveal, for example, some MMI VIII in a high density of buildings such as in Heunghae-eup district could not be detected. As these earthquakes occurred Acknowledgments in Korea, the Korean government could use the high- resolution KOMPSAT-5 SAR satellite (2.2 m) operated This research was supported by Basic Science by Korea Aerospace Research Institute (KARI) to Research Program through the National Research make damaged area maps utilizing the DPM method. Foundation of Korea (NRF) funded by the Ministry Nevertheless, our results managed to reveal damaged of Education (No.2019R1A6A1A03033167) and also buildings after the Gyeongju and Pohang earthquakes supported by a grant from the National Research in residential, commercial, and industrial areas. The Foundation of Korea (No. 2019R1A2C1085686). resulting map can be used to assess the distribution of seismic damage and can be used as inventory data of

– 20 – 02이창욱1(13~22)(Fig.2~4)ok.qxp_원격37-1_2021 2021. 2. 24. 오전 11:01 페이지 21

Damage Proxy Map (DPM) of the 2016 Gyeongju and 2017 Pohang Earthquakes using Sentinel-1 Imagery

References Intensity, and Remote-Sensing Data, Bulletin of the Seismological Society of America , Baik, H., Y.-S. Son, and K.-E. Kim, 2019. Detection 110(4): 1506-1516. of Liquefaction Phenomena from the 2017 Jin, K., J. Lee, K.-S. Lee, J.B. Kyung, and Y.-S. Kim, Pohang (Korea) Earthquake Using Remote 2020. Earthquake Damage and Related Factors Sensing Data, Remote Sensing , 11(18): 2184- Associated with the 2016 ML = 5.8 Gyeongju 2197. Earthquake, Southeast Korea, Geosciences Biass, S., S. Jenkins, D. Lallemant, T.N. Lim, G. Journal , 24(2) :141-157. Williams, and S.-H. Yun, 2021. Forecasting Kang, S., B. Kim, S. Bae, H. Lee, and M. Kim, 2019. and Planning for Volcanic Hazards, Risks, and Earthquake-Induced Ground Deformations in Disasters , Elsevier, Amsterdam, NLD. the Low-Seismicity Region: A Case of the Doo, R., 2016. Artifacts Damaged in Record-Breaking 2017 M5.4 Pohang, South Korea, Earthquake, Earthquake in Gyeongju, https://www.the Earthquake Spectra , 35(3): 1235-1260. jakartapost.com/life/2016/09/19/artifacts- Kim, H.-S., C.-G. Sun, and H.-I. Cho, 2018. Geospatial damaged-in-record-breaking-earthquake-in- Assessment of the Post-Earthquake Hazard gyeongju.html , Accessed on Jan. 18, 2021. of the 2017 Pohang Earthquake Considering Ellsworth, W.L., D. Giardini, J. Townend, S. Ge, and Seismic Site Effects, ISPRS International Journal T. Shimamoto, 2019. Triggering of the Pohang, of Geo-Information, 7(9): 375-390. Korea, Earthquake (Mw 5.5) by Enhanced Kim, J., J. Lee, M. Petitta, H. Kim, D. Kaown, I.-W. Geothermal System Stimulation, Seismological Park, S. Lee, and K.-K. Lee, 2019. Groundwater Research Letters , 90(5): 1844-1858. System Responses to the 2016 ML 5.8 Gyeongju Grigoli, F., S. Cesca, A.P. Rinaldi, A. Manconi, J.A. Earthquake, South Korea, Journal of Hydrology , López-Comino, J.F. Clinton, R. Westaway, C. 576:150-163. Cauzzi, T. Dahm, and S. Wiemer, 2018. The Kim, K.-H., J.-H. Ree, Y.H. Kim, S. Kim, S.Y. Kang, November 2017 M w 5.5 Pohang Earthquake: and W. Seo, 2018. Assessing Whether the 2017 A Possible Case of Induced Seismicity in South Mw 5.4 Pohang Earthquake in South Korea Korea, Science , 360(6392): 1003-1006. Was an Induced Event, Science , 360(6392): Gyeongju City Hall, 2021. Population Status, www. 1007-1119. gyeongju.go.kr , Accessed on Jan. 26, 2021 (in Kim, Y.H., J. Rhie, T.-S. Kang, K.-H. Kim, M. Kim, Korean). and S.-J. Lee, 2016. The 12 September 2016 Han, J., J. Kim, S. Park, S. Son, and M. Ryu, 2020. Gyeongju Earthquakes: 1. Observation and Seismic Vulnerability Assessment and Mapping Remaining Questions, Geosciences Journal , of Gyeongju, South Korea Using Frequency 20(6): 747-752. Ratio, Decision Tree, and Random Forest, Korea Meteorological Administration, 2018. Pohang Sustainability , 12(18): 7787-7809. Eartquake Analysis Report , Korea Meteorological Hough, S.E., S.-H. Yun, J. Jung, E. Thompson, G.A. Administration, Seoul, Korea (in Korean). Parker, and O. Stephenson, 2020. Near-Field Lee, H.-I., 2017. Follow the Manual They Created... Ground Motions and Shaking from the 2019 Handong University Student Overcomes Mw 7.1 Ridgecrest, California, Mainshock: Earthquake, https://www.chosun.com/site/data/ Insights from Instrumental, Macroseismic html_dir/2017/11/21/2017112100251.html ,

– 21 – 02이창욱1(13~22)(Fig.2~4)ok.qxp_원격37-1_2021 2021. 2. 24. 오전 11:01 페이지 22

Korean Journal of Remote Sensing , Vol .37, No .1, 2021

Accessed on Jan. 27, 2021 (in Korean). Tay, C.W.J., S.-H. Yun, S.T. Chin, A. Bhardwaj, J. Jung, Lee, K.-K., 2019. Summary Report of the Korean and E.M. Hill, 2020. Rapid Flood and Damage Government Commision on Relation between Mapping Using Synthetic Aperture Radar in the 2017 Pohang Earthquake and EGS Project , Response to Typhoon Hagibis, Japan, Scientific The Geological Society of Korea, Seoul, KOR. Data , 7(1): 100-108. Ministry of Public Safety and Security, 2017. Gyeongju Woo, J.U., M. Kim, D.H. Sheen, T.S. Kang, J. Rhie, F. 9.12 White Book , Ministry of Public Safety Grigoli, W.L. Ellsworth, and D. Giardini, 2019. and Security, Seoul, KOR. (in Korean). An In Depth Seismological Analysis Revealing Park, S.-C., H. Yang, D.K. Lee, E.H. Park, and W.-J. a Causal Link Between the 2017 M W 5.5 Lee, 2018. Did the 12 September 2016 Pohang Earthquake and EGS Project, Journal Gyeongju, South Korea Earthquake Cause of Geophysical Research: Solid Earth , 124(12): Surface Deformation?, Geosciences Journal , 13060-13078. 22(2): 337-346. Yun, S.-H., K. Hudnut, S. Owen, F. Webb, M. Simons, Pohang City Hall, 2021. Pohang City Hall, www. P. Sacco, E. Gurrola, G. Manipon, C. Liang, E. pohang.go.kr , Accessed on Jan. 26, 2021 (in Fielding, P. Milillo, H. Hua, and A. Coletta, Korean). 2015. Rapid Damage Mapping for the 2015 Sang-sun, S, 2019. Toward Better Risk Preparedness Mw 7.8 Gorkha Earthquake Using Synthetic for Cultural Heritage, http://www.koreanheritage. Aperture Radar Data from COSMO –SkyMed kr//inside/view.jsp?articleNo=15 , Accessed on and ALOS-2 Satellites, Seismological Research Jan 19, 2021. Letters , 86(6):1549-1556. Song, S.G. and H. Lee, 2019. Static Slip Model of the Zebker, H.A. and J. Villasenor, 1992. Decorrelation in 2017 Mw 5.4 Pohang, South Korea, Earthquake Interferometric Radar Echoes, IEEE Transactions Constrained by the InSAR Data, Seismological on Geoscience and Remote Sensing , 30(5): Research Letters , 90(1): 140-148. 950-959.

– 22 –