28 March 2025 Myanmar Earthquake: Strong Ground Motions, Macroseismic Intensity, Liquefaction, and ALOS‐2 ScanSAR Remote Sensing Damage Observations

Focus Section: Myanmar-Tingri Earthquakes 28 March 2025 Myanmar Earthquake: Strong Ground Motions, Macroseismic Intensity, Liquefaction, and ALOS-2 ScanSAR Remote Sensing Damage Observations Alejandro Cruz1 , 2, Mohammad Ghasemi3, Shaghayegh Karimzadeh*1 , Sadra Karimzadeh3 , 4, Aysegul Askan5, Masashi Matsuoka4, and Paulo B. Lourenço1 Abstract Cite this article as Cruz, A., M. Ghasemi, S. Karimzadeh, S. Karimzadeh, A. Askan, M. Matsuoka, and P. B. Lourenço (2025). 28 March 2025 Myanmar Earthquake: Strong Ground Motions, Macroseismic Intensity, Liquefaction, and ALOS-2 ScanSAR Remote Sensing Damage Observations, Seismol. Res. Lett. XX,1–18, doi: 10.1785/ 0220250258. On 28 March 2025, a devastating M 7.7 earthquake struck Myanmar along the right-lateral strike-slip Sagaing fault, followed by an M 6.7 aftershock twelve minutes later. This study presents a comprehensive assessment of the earthquake’s effects through recorded strong ground motions, macroseismic intensity observations, secondary hazards including liquefaction, and remote sensing-based damage detection. The supershear rupture spanned ∼460 km, with pronounced fault directivity effects resulting in high peak ground acceleration values of up to 0.57grecorded for the horizontal components and macroseismic intensities reaching modified Mercalli intensity = X in nearby cities. Analysis of acceleration time series and Fourier amplitude spectra at nearby recording stations revealed significant near-field effects and spectral exceedances over Myanmar National building code design levels, particularly in the short-period range, affecting mostly low-rise structures. Secondary hazard evaluation indicates widespread liquefaction, especially near the Irrawaddy River, supported by correlations between cumulative absolute velocity and liquefaction probability maps. Remote sensing analysis using multitemporal Advanced Land Observation Satellite-2 Synthetic Aperture Radar (SAR) coherence data identified severe building and road damage in urban centers. A differential coherence method, coupled with urban masking and OpenStreetMap road data, enabled spatial quantification of damage, revealing over 39,000 km of roads and 306 km2of buildings with moderate-to-heavy damage. The results provide critical insight into the earthquake’simpact and demonstrate the joint evaluation of ground-motion analysis and SAR-based damage mapping for rapid seismic response. Introduction Myanmar is located in a seismically active region with earthquakes primarily resulting from two main causes, as shown in Figure 1: (1) the subduction of the northward-moving India plate beneath the Burma plate, transitioning to a collision zone in the north, along the Andaman megathrust at a rate of 2–3.5 cm/yr; and (2) the northward motion of the Burma plate, driven by seafloor spreading in the Andaman Sea, occurring at a rate of 2.5–3.0 cm/yr (Thein et al., 2009). An earthquake of M7.7 occurred in Myanmar on 28 March 2025, followed by a second event of M6.7 twelve minutes later. The earthquakes occurred on the Sagaing fault, which lies between the Burma microplate and the Sunda plates as a main strike-slip fault system (e.g., Gögen et al., 2024). The same fault caused destructive 1. Department of Civil Engineering, University of Minho, ISISE, ARISE, Guimarães, Portugal, https://orcid.org/0000-0002-5631-0548 (AC); https://orcid.org/00000003-3753-1676 (ShK); https://orcid.org/0000-0001-8459-0199 (PBL); 2. School of Civil Engineering and Geomatics, Universidad del Valle, G-7, Cali, Colombia; 3. Remote Sensing Laboratory, Department of Remote Sensing and GIS, University of Tabriz, Tabriz, Iran, https://orcid.org/0000-0002-6992-2893 (MG); https:// orcid.org/0000-0002-5645-0188 (SaK); 4. Department of Architecture and Building Engineering, Institute of Science Tokyo, Tokyo, Japan, https://orcid.org/00000003-3061-5754 (MM); 5. Civil Engineering and Earthquake Studies Departments, Middle East Technical University, Ankara, Türkiye, https://orcid.org/0000-00034827-9058 (AA) *Corresponding author: [email protected] Copyright © 2025. The Authors. This is an open access article distributed under the terms of the CC-BY license, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Volume XX •Number XX •–2025 •www.srl-online.org Seismological Research Letters 1 Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/doi/10.1785/0220250258/7367882/srl-2025258.1.pdf by guest on 28 November 2025 events in the past, such as the 1930 M7.3, 1946 M7.7, and 1956 M7.0 earthquakes. The 2025 M7.7 event caused a supershear rupture of 460 km (Shahzada et al., 2025). With Advanced Land Observation Satellite (ALOS)-2 Synthetic Aperture Radar (SAR) data, extensive damage to the built environment during the 2025 Myanmar event in the cities of Mandalay, Sagaing, and Naypyidaw was observed due to near-field effects coupled with vulnerable building structures in the region. The ALOS-2 footprint covers a significant portion of central Myanmar (red area in Fig. 1), spanning from ∼94° to 99° E longitude and 21° to 25° N latitude. The macroseismic intensity values in the nearby urban centers reached modified Mercalli intensity (MMI) values of X. The directivity effects were clearly observed, including larger recorded peaks in the forward direction as well as a skyscraper collapsing in Bangkok, Thailand. This collapse may also be attributed to the soft soil conditions and potential basin effects (Qodri, 2018). Given the scale of the event and the diversity of its impacts, a comprehensive assessment combining seismological data, macroseismic observations, secondary hazard mapping, and satellite-based damage detection is critical. This multidisciplinary study aims to document and analyze the ground-motion characteristics, evaluate the extent and nature of infrastructure damage, and investigate the geospatial patterns of secondary hazards, particularly liquefaction, using remote sensing techniques. The article presents ground-motion data and key insights from strong-motion records and macroseismic intensity. It also investigates secondary hazards, particularly liquefaction susceptibility and its spatial patterns, using geospatial indicators. Next, remote sensing analysis is conducted to map damage to roads and buildings using ALOS-2 SAR data. The study concludes by summarizing the main findings and their relevance for improving earthquake response and hazard mitigation strategies in the region. Observations on Recorded Strong Ground Motions and Macroseismic Intensity Values The mainshock was recorded at six strong ground motion stations within a 370 km rupture distance (Rrup) with rapidly varying peak ground-motion parameters (Fig. 2). Selected stations that recorded the 28 March 2025 M7.7 Myanmar earthquake, including observed (Fig. 2a) peak ground acceleration (PGA), (Fig. 2b) peak ground velocity (PGV), (Fig. 2c) pseudospectral acceleration (PSA) (T= 0.2 s), and (Fig. 2d) PSA (T= 1.0 s). (The surface projection of the fault is also presented). From the spatial distribution of the recorded data, it is observed that PGA, PGV, and PSA with 5% damping ratio values are lower north of the epicenter and much larger in the south. This observation is attributed to the fault directivity effects on the causative fault, which is the right-lateral strike-slip Sagaing fault segment. Despite the lack of data in the near field, it is observed that the maximum horizontal PGA reaches 0.57g at the closest station to the rupture with a corresponding shortperiod PSA value of almost 1.0g. In this study, we present detailed information on the recorded strong ground motions at the six nearest stations with reliable data and assess the recorded motions (Table 1). The information includes station coordinates, rupture distance (Rrup), epicentral distance (Repi), elevation, PGA, PGV, Housner intensity (HI), significant duration (D5–95) within 5% and 95% of Arias intensity (IA), and cumulative absolute velocity (CAV). The raw data are obtained from the Incorporated Research Institutions for Seismology website (Incorporated Research Institutions for Seismology [IRIS], 2025), baseline corrected, and filtered with a fourth-order Butterworth filter between 0.1 and 30 Hz. There is nopreciseinformationonthesiteclassesorVS30 values of the stations; only general information about VS30 is discussed later in the secondary hazard section; however, the elevations of the Figure 1. Tectonic map of Myanmar and nearby regions, highlighting major fault systems, plate boundaries, the epicenter of the 28 March 2025 M7.7 earthquake (yellow star), seismic stations (blue triangles), together with the Advanced Land Observation Satellite (ALOS)-2 PALSAR-2 data footprint used in this study (red box). Base map taken from Thein et al. (2009). The color version of this figure is available only in the electronic edition. 2Seismological Research Letters www.srl-online.org •Volume XX •Number XX •–2025 Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/doi/10.1785/0220250258/7367882/srl-2025258.1.pdf by guest on 28 November 2025 stations on the tectonic map allow us to establish a framework. From Table 1, it is observed that the stations within close distances from the rupture located along the forward direction on low elevations experienced larger ground motions, whereas the farther stations and those on the mountain edge recorded much smaller motions. We focus on the three closest stations to evaluate the acceleration time series and Fourier amplitude spectra (FAS) in Figures 3–5. On the same plots, we also compare the recorded response spectra (PSA) to the corresponding design spectra of Myanmar with 5% damping for a return period of 475 yr in the Myanmar National Building Code (MNBC) (2020). Among the recording stations, the NPW station is located at an Rrup of 2.75 km. The recorded PGA values in the horizontal directions are 511:5cm=s2(east–west) and 603:45 cm=s2 (north–south) with corresponding PGV values of 96.97 and 93.94 cm/s, respectively. The vertical component is almost Figure 2. Selected stations that recorded the 28 March 2025 M7.7 Myanmar earthquake, including observed (a) peak ground acceleration (PGA), (b) peak ground velocity (PGV), (c) pseudospectral acceleration (PSA) (T= 0.2 s), and (d) PSA (T= 1.0 s). (The surface projection of the fault is also presented). The color version of this figure is available only in the electronic edition. Volume XX •Number XX •–2025 •www.srl-online.org Seismological Research Letters 3 Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/doi/10.1785/0220250258/7367882/srl-2025258.1.pdf by guest on 28 November 2025 twice the horizontal component, attributed to the near-field source effects such as those observed in the 2023 Kahramanmaraşsequence (Altindal and Askan, 2024). At the NPW station, D5–95 of the horizontal components is around 13 s at this station, HI is 288 cm and 300 cm in the east–west and north–south directions, respectively. Such high levels of HI usually correlate closely with structural damage (e.g., Askan et al.,2022). At station NPW, there are strong near-field effects as observed in the form of velocity pulses, also consistent with the enhanced low-frequency content in the FAS. The final observation is on the comparison of the acceleration response spectra (PSA) against the design spectra defined in the MNBC (2020) with 5% damping for a return period of 475 yr. As the station is located at a low elevation, we use the design spectra for sites C and D for the sake of comparison. It is noted that, according to the national building code MNBC (2020), the short-period site coefficient (Fa) is the same for site classes C and D, which explains the similarity of spectral values in the plateau. The difference between classes C and D appears in the long-period site coefficient (Fv), which controls velocity-dominated site effects. It is observed that the recorded response spectra exceed the design spectra for both type C and D soils, up to periods of 0.3 s. This observation, coupled with the structural vulnerability, could have caused damage to the low-rise residential buildings. Within T=0.3–1.3 s range, the recorded spectra are below the design spectra. For the longer periods, again, the recorded horizontal spectra exceed the corresponding design amplitudes. Station TGI is located in the fault-normal direction, east of the Sagaing fault at an Repi of 192.29 km. From the map, this station is at an elevation of 1457.5 m. As a result of the hardrock conditions and increased distances, both PGA and PGV values are much lower than those recorded at the station NPW. The recorded PGA values are 81:13 cm=s2in the east–west and 73:95 cm=s2in the north–south directions. The vertical PGA is almost half of that of the horizontal ones. The PGV values are much lower, around 3 cm/s, for all components. D5–95 for this station is around 50 s, whereas HI reaches 25 cm. The recorded response spectrum for this station is compared against the design spectra for rock and stiff soil conditions, sites B and C. It is observed that the recorded values are below both design spectra at all period ranges. TABLE 1 Information on the Recorded Stations Station Latitude (°) Longitude (°) Rrup (km) Repi (km) Elevation (m) Channel PGA (cm=s2) PGV (cm/s) HI (cm) D5–95 (s) IA (cm/s) CAV (cm/s) NPW 19.779° N 96.138° E 2.75 265.89 158 HNE 511.47 96.97 288.2 13.4 2.78 1625.12 HNN 603.45 93.94 300.1 13.2 3.13 1621.48 HNZ 1111.9 65.56 228.7 9.9 3.3 1351.84 TGI 20.768° N 97.034° E Not available 192.29 1457.5 HHE 81.13 3.35 15.5 51.8 0.26 857.17 HHN 73.95 2.63 14.6 49.0 0.2 754.29 HHZ 41.75 3.01 12.6 42.7 0.05 340.57 NGU 21.206° N 94.917° E 113.95 147.15 70.4 HNE 45 6.2 31.1 65.7 0.096 588.60 HNN 58.1 8.87 35.1 72.2 0.09 557.04 HNZ 17.65 4.5 12.1 100.4 0.02 359.60 YGN 16.865° N 96.153° E 157.14 607.07 20 HNE 24.3 11.7 24.75 118.5 0.04 437.95 HNN 14.7 7.8 18.08 118.1 0.04 499.16 HNZ 19.5 3.7 15.24 120.8 0.01 256.51 CHTO 18.814° N 98.944° E 272.22 506.00 420 HN1 9.52 5.9 8.66 33.85 0.004 104.18 HN2 8.21 3.97 9.26 34.97 0.004 114.77 HNZ 12.9 4.44 10.83 33.41 0.005 118.65 KTN 21.286° N 99.590° E 370.59 416.70 832 HNE 10.63 2.54 12.18 87.31 0.015 267.78 HNN 13.22 4.48 16.82 90.76 0.024 332.03 HNZ 7.31 1.87 9.97 86.69 0.006 165.93 Channel, components of recorded ground motion: HH, high frequency and high sensitivity; HN, high gain, high-sampling rate; E, east–west direction; N, north–south direction; Z, vertical direction; 1, east–west direction; 2, north–south direction. CAV, cumulative absolute velocity; D5–95, significant duration (time between 5% and 95% of Arias intensity); HI, Housner intensity; IA, Arias intensity; PGA, peak ground acceleration; PGV, peak ground velocity; Repi, epicentral distance; and Rrup, closest distance to rupture. 4Seismological Research Letters www.srl-online.org •Volume XX •Number XX •–2025 Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/doi/10.1785/0220250258/7367882/srl-2025258.1.pdf by guest on 28 November 2025 Finally, the station NGU is located at a lower elevation of 70.4 m and an Rrup distance of 113.95 km in the fault-normal direction, west of the Sagaing fault. The PGA values are recorded as 45:00 and 58:10 cm=s2for the east–west and north–south components, respectively. For the Z component, Figure 3. Station NPW: (a) acceleration time series; (b) Fourier amplitude spectrum (FAS) and observed pseudoacceleration response spectra (PSA) compared with the code spectra. The color version of this figure is available only in the electronic edition. Volume XX •Number XX •–2025 •www.srl-online.org Seismological Research Letters 5 Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/doi/10.1785/0220250258/7367882/srl-2025258.1.pdf by guest on 28 November 2025 this value is much less, 17:65 cm=s2. The PGV values are 6.20 and 8.87 cm/s for the corresponding horizontal directions. Although in the Z direction, it is almost half. The significant duration is longer at this station, reaching 65.70 and 72.20 s, mostly due to surface waves at this low elevation and longer Figure 4. Station TGI (a) acceleration time series; (b) FAS and observed pseudoacceleration response spectra (PSA) compared with the code spectra. The color version of this figure is available only in the electronic edition. 6Seismological Research Letters www.srl-online.org •Volume XX •Number XX •–2025 Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/doi/10.1785/0220250258/7367882/srl-2025258.1.pdf by guest on 28 November 2025 wavetrains, whereas it is even higher, 100.40 s, in the Z direction. In terms of HI, the value is around 35 cm for both horizontal directions, and in the Z direction, it reduces to 20 cm. When the response spectrum is compared with the design spectrum for site classes C and D, it is observed that the Figure 5. Station NGU: (a) acceleration time series; (b) FAS and observed pseudoacceleration response spectra (PSA) compared with the code spectra. The color version of this figure is available only in the electronic edition. Volume XX •Number XX •–2025 •www.srl-online.org Seismological Research Letters 7 Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/doi/10.1785/0220250258/7367882/srl-2025258.1.pdf by guest on 28 November 2025 recorded motions are much lower than the design levels. This observation is consistent with the lack of severe damage in these areas. For the three stations, CAV values in the three components are higher than 0.16 g.s (157 cm/s), which has been identified as a conservative threshold distinguishing damaging from nondamaging ground motions for well-designed and wellconstructed buildings, as defined by the MMI scale (EPRI, Palo Alto and U.S. Department of Energy, 2006). Figure 6a shows the spatial distribution of MMI as reported by the U.S. Geological Survey (USGS) (2025), as well as the MMI values at the selected stations in this study. It is observed that along the fault plane toward the south of the epicenter, the MMI values reach X+, consistent with the very heavy damage in that direction. The spatial distribution of the reported MMI values also confirms the strong directivity effects observed during the 28 March 2025 Myanmar event. Next, in Figure 6b, we compare the MMI values reported in the DYFI system (USGS, 2025) against the MMI values at the stations obtained with conversions from corresponding recorded PGA and PGV values. For conversion, we use the relationships by Bilal and Askan (2014), which were developed based on PGA and PGV using data from masonry and reinforced-concrete structures in Türkiye affected by major earthquakes. As demonstrated in the studies by Erberik (2008a,b) and later applied and validated by Karimzadeh et al. (2018) for Erzincan, these models were successfully verified against observed damage in regions dominated by masonry and RC building typologies. To better represent the structural typologies, both PGA and PGV values were employed, capturing the behavior of rigid masonry and more flexible reinforced-concrete structures (Erberik, 2008a,b;Karimzadeh et al., 2018). The computed MMI values from these conversions show good agreement with reported observations, with only minor deviations noted at the closest NPW station. Given that Myanmar exhibits similar construction typologies, we consider the adopted equations to be reliable within an expected range of uncertainties. Furthermore, comparison with reported data demonstrates consistent agreement, as illustrated in Figure 6b. It is observed that the computed MMI values from PGA and PGV closely match the reported data, except for minor differences in MMI values at the closest NPW station. Secondary Hazards Secondary effects, such as tsunamis, landslides, liquefaction, and fire, account for ∼25%–40% of total economic losses and fatalities associated with earthquakes worldwide (Gómez et al., 2022). Liquefaction was a dominant secondary hazard in the studied earthquake, with substantial infrastructure damage consistently attributed to its effects in prior studies. In multiple sources, severe infrastructure damage is attributed to widespread liquefaction, especially in areas south of Mandalay, such as Inwa Town, near the Irrawaddy River (e.g., National Remote Sensing Centre, 2025), where there are local conditions, such as a shallow groundwater table, that contributed to soil instability (Wang et al., 2025). Although field investigations confirmed the presence of liquefaction, this section evaluates the potential spatial extent of the phenomenon using U.S. Geological Survey (USGS, 2025) data and correlates it with various geospatial proxies. The estimated probability of liquefaction taken from USGS (2025) is based on the work of Allstadt et al. (2022), which is an adaptation of the model of Zhu et al. (2017). The model results carry substantial uncertainty because they depend on globally generalized predictor variables and ShakeMapderived ground-motion intensities (Allstadt et al.,2022), which may not fully capture local ground conditions or site-specific effects. According to the data provided by the USGS (2025), the liquefaction probability due to this earthquake is estimated to be extensive in both severity and spatial extent. In this work, the analysis explores the relationships between geospatial proxies and the physical factors that influence liquefaction susceptibility in the Myanmar region (Fig. 7). For instance, VS30 serves as a proxy for soil density, as noted by Zhu et al. (2015), and it is a critical parameter in liquefaction analysis, as soil stiffness significantly influences liquefaction susceptibility. Higher VS30 values indicate stiffer soils, which are generally less prone to liquefaction compared to loose, softer soils (Yilmaz et al., 2021;Cetin et al., 2022). Figure 7a shows the VS30 distribution in the Myanmar region, based on USGS (2025) data. As shown in Figure 7, liquefaction probability tends to be higher in areas classified as site classes E and D according to the National Earthquake Hazards Reduction Program (Building Seismic Safety Council [BSSC], 2020). The probability of liquefaction also appears elevated along the trace of the Sagaing fault. However, despite the widespread presence of softer soils in the central region, the estimated liquefaction probability is relatively low in that area, particularly on the western vertex of the Sagaing fault. In this area, the expected liquefaction probability is lower than 0.5%. This apparent anomaly may coincide with the spatial extent of geological unit TM, corresponding to the upper Pegu Group. This unit presents a north–south facies gradient that reflects a transition from nonmarine environments in the north to marine environments in the south. As a result, the deposits are characterized by thin-bedded fine to medium-grained sandstones interbedded with thick sequences of clays and siltstones, which exhibit various sedimentary structures (Ridd and Racey, 2015). The thick clay and silty layers within this geological unit are a key factor contributing to the low liquefaction susceptibility in this area. One notable observation is in Yangon City, which shows a high estimated liquefaction probability, aligning with previous findings that indicate greater vulnerability in the southern part of the city compared to the northern part (Htet et al., 2018). These observations are consistent with the USGS 8Seismological Research Letters www.srl-online.org •Volume XX •Number XX •–2025 Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/doi/10.1785/0220250258/7367882/srl-2025258.1.pdf by guest on 28 November 2025 liquefaction map (USGS, 2025) and documented field reports (e.g., National Remote Sensing Centre, 2025). Another relevant proxy is PGA, which is directly related to earthquake-induced loading (Zhu et al., 2015). However, the relationship between PGA and liquefaction occurrence (Fig. 7b) is not straightforward. Considering the recorded PGA values, Figure 6. (a) Spatial distribution of modified Mercalli intensity (MMI) according to U.S. Geological Survey (USGS) (USGS, 2025) along with the causative fault and the recording stations on the location map of Myanmar; (b) MMI values according to the USGS —Did You Feel It? (DYFI) (USGS, 2025) and conversion equations of Bilal and Askan (2014). The color version of this figure is available only in the electronic edition. Volume XX •Number XX •–2025 •www.srl-online.org Seismological Research Letters 9 Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/doi/10.1785/0220250258/7367882/srl-2025258.1.pdf by guest on 28 November 2025 to red pixel classifications, indicating widespread and severe infrastructure damage. The linear patterns of damage visible in Figure 10a along road segments, coupled with the concentrated clusters of damaged buildings in Figure 10b, collectively confirm the devastating impact of the earthquake on the interconnected urban fabric. Table 2provides a comprehensive quantitative summary of the classified damage levels for both road networks and building structures within the study area. The analysis revealed that a substantial 3,960 km of roads sustained heavy damage, with an additional 35,019 km experiencing possible moderate damage. Conversely, 74,918 km of roads were categorized as having no observable damage. For buildings, the extent of damage is equally significant. A total of 100:36 km2of built-up areas were classified as possibly heavily damaged, and 205:12 km2as possibly moderately damaged. In contrast, 503:41 km2of buildings exhibited no discernible damage. The spatial distribution of the damaged areas from analyses of SAR data is observed to be consistent with the distribution of ground motions. These quantitative metrics, derived from the differential coherence analysis, corroborate the visual interpretations from the potential damage maps, providing approximate numerical values for the extent of earthquake-induced infrastructure devastation. The cumulative extent of severe and moderate damage highlights the critical need for large-scale recovery and reconstruction efforts. Although direct validation of the damage estimates was not feasible due to the absence of detailed ground observations, the coherence-based method employed here has been widely applied and validated in previous earthquakes (Karimzadeh and Mastuoka, 2017;Adriano et al., 2018;Hasanlou et al., 2021;Karimzadeh et al., 2022), demonstrating its robustness for post-disaster damage mapping. Although the SAR-based coherence analysis provides valuable insights into the spatial distribution of damage, several sources of uncertainty should be considered. First, temporal baselines between SAR acquisitions can pose decorrelation unrelated to structural damage, particularly when acquisition intervals are long. Second, vegetation dynamics, especially in periurban zones, may reduce coherence and complicate damage discrimination. Third, radar imaging geometry effects (layover, shadowing, and foreshortening) may obscure heavily affected structures in dense urban or mountainous regions. Finally, the lack of extensive field data limited our ability to directly validate the coherence-derived results. Despite these limitations, the methodology has been extensively tested in prior seismic events and has consistently demonstrated reliable performance in identifying damaged areas. In this context, the observed agreement between the spatial distribution of strong ground motions and our coherence-based damage estimates provides confidence that the results presented in Table 2represent a robust first-order approximation of actual damage patterns. Conclusions Analysis of the recorded seismic data from the 28 March 2025 Myanmar earthquake revealed strong directivity effects and high ground-motion parameters, particularly at near-field stations located in the forward rupture direction and on softer soil conditions. The recorded spectra at the near-field station exceeded the Myanmar National Building Code values for short periods, which may explain the widespread damage to low-rise buildings. Liquefaction was one of the most prominent secondary hazards, especially in areas with soft soil conditions and shallow groundwater, such as near the Irrawaddy River. The relationship between CAV and estimated liquefaction probability appeared to be more consistent than PGA or PGV alone, indicating its potential as a better proxy for rapid postevent liquefaction assessment. Through advanced SAR-based coherence analysis, we were able to detect and map extensive infrastructure damage. The results indicated that over 39,000 km of roads and 306 km2 of built-up areas suffered moderate-to-heavy damage. The coherence-based damage mapping method proved particularly useful for highlighting changes in urban areas, with strong agreement between building and road damage zones, also consistent with the spatial distribution of ground motions. Overall, this study underscores the importance of integrating strong-motion data, secondary hazard modeling, and remote sensing in postearthquake damage assessments. The insights gained can inform future emergency response planning and seismic risk mitigation strategies for Myanmar and other tectonically active regions. Data and Resources All data used in this article came from published sources listed in the references. Declaration of Competing Interests The authors acknowledge that there are no conflicts of interest recorded. Acknowledgments This work has been partly funded by the STAND4HERITAGE project that has received funding from the European Research Council (ERC) TABLE 2 Quantitative Summary of Earthquake-Induced Possible Damage to Roads and Buildings Possible Damage Road (km) Building (km2) No damage 74,918 503.41 Moderate damage 35,019 205.12 Heavy damage 3,960 100.36 16 Seismological Research Letters www.srl-online.org •Volume XX •Number XX •–2025 Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/doi/10.1785/0220250258/7367882/srl-2025258.1.pdf by guest on 28 November 2025 under the European Union’s Horizon 2020 research and innovation program (Grant Agreement Number 833123), as an Advanced Grant. This work was partly financed by FCT/MCTES through national funds (PIDDAC) under the R&D Unit Institute for Sustainability and Innovation in Structural Engineering (ISISE), under reference UID/04029/Institute for Sustainability and Innovation in Structural Engineering (ISISE), and under the Associate Laboratory Advanced Production and Intelligent Systems ARISE under reference LA/P/ 0112/2020. This work is financed by national funds through FCT– Foundation for Science and Technology, under Grant Agreement 2022.12016.BD is attributed to the first author. The Advanced Land Observation Satellite (ALOS)-2 PALSAR-2 images are provided by the JAXA (Japan Aerospace Exploration Agency) working group. Sara Rengifo kindly provided technical assistance in preparing the base maps for three figures and contributed valuable geological insights. The authors also extend their sincere thanks to the two anonymous reviewers for their insightful comments and suggestions, which greatly improved the quality of this article. References Adriano,B.,S.Koshimura,S.Karimzadeh,M.Matsuoka,andM.Koch (2018). Damage mapping after the 2017 Puebla earthquake in Mexico using high-resolution Alos2 Palsar2 data, IGARSS 2018-2018 IEEE International Geoscience and Remote Sensing Symposium, IEEE, 870–873, doi: 10.1109/IGARSS.2018.8517933. Alaska Satellite Facility (2024). ALOS PALSAR—radiometric terrain correction, available at https://asf.alaska.edu/datasets/daac/alospalsar-radiometric-terrain-correction/ (last accessed June 2025). Allstadt, K. E., E. M. Thompson, R. W. Jibson, D. J. Wald, M. Hearne, E. J. Hunter, J. Fee, H. Schovanec, D. Slosky, and K. L. Haynie (2022). The US Geological Survey ground failure product: Near-real-time estimates of earthquake-triggered landslides and liquefaction, Earthq. Spectra 38, no.1,5–36, doi: 10.1177/87552930211032685. Altindal, A., and A. Askan (2024). Traditional seismic hazard analyses underestimate hazard levels when compared to observations from the 2023 Kahramanmaras earthquakes, Commun. Earth Environ. 5, no. 1, 14, doi: 10.1038/s43247-023-01148-y. Askan,A.,Z.Gülerce,Z.Roumelioti,D.Sotiriadis,N.S.Melis,A. Altindal,B.Akbaş,E.Sopaci,S.Karimzadeh,I.Kalogeras,et al. (2022). The Samos Island (Aegean Sea) M 7.0 earthquake: Analysis and engineering implications of strong motion data, Bull. Earthq. Eng. 20, no. 14, 7737–7762, doi: 10.1007/s10518-02101251-5. Bilal, M., and A. Askan (2014). Relationships between felt intensity and recorded ground-motion parameters for Turkey, Bull. Seismol. Soc. Am. 104, no. 1, 484–496, doi: 10.1785/0120130093. Bozzoni, F., R. Bonì, D. Conca, C. G. Lai, E. Zuccolo, and C. Meisina (2021). Megazonation of earthquake-induced soil liquefaction hazard in continental Europe, Bull. Earthq. Eng. 19, no. 10, 4059– 4082, doi: 10.1007/s10518-020-01008-6. Bray, J. D., and J. Macedo (2017). 6th Ishihara lecture: Simplified procedure for estimating liquefaction-induced building settlement, Soil Dynam. Earthq. Eng. 102, 215–231, doi: 10.1016/J.SOILDYN.2017.08.026. Building Seismic Safety Council (BSSC) (2020). NEHRP (National Earthquake Hazards Reduction Program) recommended seismic provisions for new buildings and other structures, FEMA P-2082–1, Washington, D.C. Cetin,K.O.,S.Altun,A.Askan,M.Akgün,A.Sezer,C.Kıncal,Ö.C. Özdağ,Y.İpek, B. Unutmaz, Z. Gülerce, et al. (2022). The site effects in Izmir Bay of October 30 2020, M 7.0 Samos Earthquake, Soil Dynam. Earthq. Eng. 152, 107,051, doi: 10.1016/j.soildyn.2021.107051. Dieter, J., and R. Bamler (1994). Phase statistics of interferograms with applications to synthetic aperture radar, Appl. Opt. 33, no. 20, 4361–4368. EPRI, Palo Alto and U.S Department of Energy (2006). Program on Technology Innovation: Use of Minimum CAV in Determining Effects of Small Magnitude Earthquakes on Seismic Hazard Analyses,available at https://www.nrc.gov/docs/ML0603/ML060320738.pdf (last accessed June 2025). Erberik, M. A. (2008a). Fragility-based assessment of typical mid-rise and low-rise RC buildings in Turkey, Eng. Struct. 30, no. 5, 1360– 1374, doi: 10.1016/j.engstruct.2007.07.016. Erberik, M. A. (2008b). Generation of fragility curves for Turkish masonry buildings considering in-plane failure modes, Earthq. Eng. Struct. Dynam. 37, no. 3, 387–405, doi: 10.1002/eqe.760. Ghasemi, M., S. Karimzadeh, and B. Feizizadeh (2021). Urban classification using preserved information of high dimensional textural features of Sentinel-1 images in Tabriz, Iran, Earth Sci. Inform. 14, no. 4, 1745–1762, doi: 10.1007/s12145-021-00617-2. Gögen, B., S. Karimzadeh, and P. B. Lourenço (2024). Probabilistic seismic hazard analysis for Bagan (Myanmar), Acta Geophys. 72, no. 4, 2193–2221, doi: 10.1007/s11600-023-01263-6. Gómez, J. C., M. Pittore, N. Brinckmann, J. Lizarazo-Marriaga, S. Medina, N. Tarque, and F. Cotton (2022). Scenario-based multi-risk assessment from existing single-hazard vulnerability models. An application to consecutive earthquakes and tsunamis in Lima, Peru, Nat. Hazards Earth Syst. Sci. 23, 2203–2228. Hasanlou, M., R. Shah-Hosseini, S. T. Seydi, S. Karimzadeh, and M. Matsuoka (2021). Earthquake damage region detection by multitemporal coherence map analysis of radar and multispectral imagery, Remote Sens. 13, no. 6, 1195, doi: 10.3390/rs13061195. Hoffmann, J. (2007). Mapping damage during the Bam (Iran) earthquake using interferometric coherence, Int. J. Remote Sens. 28, no. 6, 1199–1216, doi: 10.1080/01431160600928567. Htet, W. P., M. Thant, and D. Arini (2018). Probability of liquefaction hazard map for Yangon City, Myanmar, doi: 10.1063/1.5047312. Incorporated Research Institutions for Seismology (IRIS) (2025). IRIS Wilber-3: Find stations, available at https://ds.iris.edu/wilber3/ find_stations/11952284 (last accessed June 2025). Karimzadeh, S., and M. Mastuoka (2017). Building damage assessment using multisensor dual-polarized synthetic aperture radar data for the 2016 M 6.2 Amatrice earthquake, Italy, Remote Sens. 9, no. 4, 330, doi: 10.3390/rs9040330. Karimzadeh, S., and M. Matsuoka (2021). A preliminary damage assessment using dual path synthetic aperture radar analysis for the M 6.4 Petrinja earthquake (2020), Croatia, Remote Sens. 13, no. 12, 2267, doi: 10.3390/rs13122267. Karimzadeh, S., A. Askan, M. A. Erberik, and A. Yakut (2018). Seismic damage assessment based on regional synthetic ground motion dataset: A case study for Erzincan, Turkey, Nat. Hazards 92, no. 3, 1371–1397, doi: 10.1007/s11069-018-3255-6. Karimzadeh, S., M. Ghasemi, M. Matsuoka, K. Yagi, and A. C. Zulfikar (2022). A deep learning model for road damage detection after an earthquake based on synthetic aperture radar (SAR) and field Volume XX •Number XX •–2025 •www.srl-online.org Seismological Research Letters 17 Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/doi/10.1785/0220250258/7367882/srl-2025258.1.pdf by guest on 28 November 2025 datasets, IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 15, 5753– 5765, doi: 10.1109/JSTARS.2022.3189875. Kramer, S. L., and R. A. Mitchell (2006). Ground motion intensity measures for liquefaction hazard evaluation, Earthq. Spectra 22, no. 2, 413–438, doi: 10.1193/1.2194970. Lee, J.-S., K. W. Hoppel, S. A. Mango, and A. R. Miller (1994). Intensity and phase statistics of multilook polarimetric and interferometric SAR imagery, IEEE Trans. Geosci. Remote Sens. 32, no. 5, 1017–1028, doi: 10.1109/36.312890. Liu, H., C. Song, Z. Li, Z. Liu, L. Ta, X. Zhang, B. Chen, B. Han, and J. Peng (2024). A new method for the identification of earthquakedamaged buildings using sentinel-1 multitemporal coherence optimized by homogeneous SAR pixels and histogram matching, IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 17, 7124–7143, doi: 10.1109/JSTARS.2024.3377218. Myanmar National Building Code (MNBC) (2020). Myanmar National Building Code-2020, Ministry of Construction, Nay Pyi Taw, Myanmar. National Remote Sensing Centre (2025). Myanmar earthquake 2025: Damage assessment report, available at https://www.nrsc.gov.in/ nrscnew/MyanmarEarthquake2025-main/index_earthquake.html (last accessed June 2025). Qodri, M. F. (2018). Geophysical site investigation and ground response analysis of Bangkok subsoil due to earthquake, Chulalongkorn University, Bangkok, Thailand, doi: 10.58837/CHULA.THE.2018.133. Ridd, M. F., and A. Racey (2015). Chapter 4 Onshore Petroleum Geology of Myanmar: Central Burma Depression, Vol. 45, Geological Society of London, 21–50, doi: 10.1144/M45.04. Rosen, P. A., S. Hensley, I. R. Joughin, F. K. Li, S. N. Madsen, E. Rodriguez, and R. M. Goldstein (2000). Synthetic aperture radar interferometry, Proc. IEEE 88, no. 3, 333–382, doi: 10.1109/ 5.838084. Santoro, M., U. Wegmüller, C. Lamarche, S. Bontemps, P. Defourny, and O. Arino (2015). Strengths and weaknesses of multi-year Envisat ASAR backscatter measurements to map permanent open water bodies at global scale, Remote Sens. Environ. 171, 185–201, doi: 10.1016/j.rse.2015.10.031. Seymour, M. S., and I. G. Cumming (1994). Maximum likelihood estimation for SAR interferometry, Proc. of IGARSS ’94-1994 IEEE International Geoscience and Remote Sensing Symposium, IEEE, 2272–2275, doi: 10.1109/IGARSS.1994.399711. Shahzada, K., U. A. Noor, and Z.-D. Xu (2025). In the wake of the March 28, 2025 Myanmar earthquake: A detailed examination, J. Dynam. Disasters 1, no. 2, 100,017, doi: 10.1016/j.jdd.2025.100017. Shimada, M., T. Itoh, T. Motooka, M. Watanabe, T. Shiraishi, R. Thapa, and R. Lucas (2014). New global forest/non-forest maps from ALOS PALSAR data (2007–2010), Remote Sens. Environ. 155,13–31, doi: 10.1016/j.rse.2014.04.014. Tamkuan, N., and M. Nagai (2017). Fusion of multi-temporal interferometric coherence and optical image data for the 2016 Kumamoto earthquake damage assessment, ISPRS Int. J. Geo-Inform. 6, no. 7, 188, doi: 10.3390/ijgi6070188. Thein, M., T. Myint, S. Thura Tun, and T. Lwin Swe (2009). Earthquake and tsunami hazard in Myanmar, J. Earthq. Tsunami 3, 43–57. Toutin, T. (2004). Review article: Geometric processing of remote sensing images: Models, algorithms and methods, Int. J. Remote Sens. 25, no. 10, 1893–1924, doi: 10.1080/0143116031000101611. Touzi, R., A. Lopes, J. Bruniquel, and P. W. Vachon (1999). Coherence estimation for SAR imagery, IEEE Trans. Geosci. Remote Sens. 37, no. 1, 135–149, doi: 10.1109/36.739146. U.S. Geological Survey (2025). M 7.7 - Myanmar region, U.S. Geological Survey, available at https://earthquake.usgs.gov/earthquakes/ eventpage/us7000pn9s (last accessed June 2025). Wang, T., Y. Zhou, J. Chen, X. Wang, H. Bi, and X. Wang (2025). Field survey of building damage at Mandalay during 2025 Myanmar Mw 7.7 earthquake, Earthq. Eng. Eng. Vib. doi: 10.1007/s11803-0252325-5. Yilmaz, C., V. Silva, and G. Weatherill (2021). Probabilistic framework for regional loss assessment due to earthquake-induced liquefaction including epistemic uncertainty, Soil Dynam. Earthq. Eng. 141, 106,493, doi: 10.1016/J.SOILDYN.2020.106493. Zhu, J., L. G. Baise, and E. M. Thompson (2017). An updated geospatial liquefaction model for global application, Bull. Seismol. Soc. Am. 107, no. 3, 1365–1385, doi: 10.1785/0120160198. Zhu, J., D. Daley, L. G. Baise, E. M. Thompson, D. J. Wald, and K. L. Knudsen (2015). A geospatial liquefaction model for rapid response and loss estimation, Earthq. Spectra 31, no. 3, 1813– 1837, doi: 10.1193/121912EQS353M. Manuscript received 9 August 2025 Published online 20 October 2025 18 Seismological Research Letters www.srl-online.org •Volume XX •Number XX •–2025 Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/doi/10.1785/0220250258/7367882/srl-2025258.1.pdf by guest on 28 November 2025