We provide new data and insights into a 6 February 2018 $Mw$ 6.4 earthquake that shook the city of Hualien in eastern Taiwan at the leading edge of a modern arc–continent collision. Fatalities and damages were concentrated near the Milun fault and extended south to the northern Longitudinal Valley fault. Although the Hualien area has one of the highest rates of seismicity in Taiwan, the geologic structures responsible for active deformation were not well understood before this event. We analyzed Interferometric Synthetic Aperture Radar (InSAR) and Global Positioning System (GPS) data and produced a 3D displacement model with InSAR and azimuth offset of radar images to document surface deformation induced by this earthquake. The 3D displacement model was inverted to estimate slip on the Milun fault. We find that models assuming a single fault are inconsistent with observations of coseismic deformation and regional strain patterns, providing evidence for linked slip on a little‐studied offshore thrust belt. Based on data presented here and elsewhere, we propose a model for transpressive deformation in a zone of oblique convergence and left‐lateral wrench tectonics to explain this and a prior 1951 M 7.3 earthquake.

Nearby faults interact through stress changes induced by fault slip and viscoelastic flow. The process is, however, often elusive and can be geometry-dependent and time-variant. Here, we combine geodetic and field observations to characterize the interaction of two head-to-head, conjugate faults in eastern Taiwan during the 2022 Chihshang earthquake sequence. We map the coseismic slip on the Central Range fault and dynamically-triggered shallow slip on the Longitudinal Valley fault, which has been creeping interseismically. Overlapping of seismic and aseismic slip suggests that the Longitudinal Valley fault is capable of hosting a variety of distinct slip behaviors. Moreover, substantial slip on the Central Range fault suppresses Coulomb stress on the Longitudinal Valley fault, and vice versa, resulting in seismic bursts in an out-of-phase pattern on the two faults as seen in the hundred-year historical records. Such fault interaction implies the need for time-dependent seismic hazard reassessment for the complex fault system.

Landslides have caused extensive infrastructure damage and caused human fatalities for centuries. Intense precipitation and large earthquakes are considered to be two major landslide triggers, particularly in the case of catastrophic landslides. The most widely accepted mechanistic explanation for landslides is the effective-stress dependent shear strength reduction due to increases in pore water pressure. The Chashan landslide site, selected for the present study, has been intensively studied from geological, geophysical, geodetic, geotechnical, hydrological, and seismological perspectives. Our seismic monitoring of daily relative velocity changes (dv/v) indicated that landslide material decreases coincided with the first half of the rainy period and increased during the latter half of the rainy period. The geodetic surveys before and after the rainy period identified vertical subsidence without horizontal movement. The results from the multidisciplinary investigation enabled us to draw a conceptual model of the landslide recovery process induced by water loading. Where all sliding materials were stable (safety factor > 1.0), unconsolidated landslide colluvium and impermeable sliding surfaces trapped the seepage water to form a water tank, provided that compact forces were acting on the materials below the sliding boundary. The vertical force of compaction facilitates an increase in the cohesion and strength of landslide materials, thereby increasing the landslide materials’ stability. We demonstrated that the recovery process periodically occurs only under the combined conditions of prolonged and intense precipitation and the related stability conditions.

Understanding the causes of variability in earthquake rupture characteristics, particularly the distribution of shallow slip and the extent of off-fault deformation, is crucial for refining seismic hazard models and understanding fundamental rupture mechanics. The 28 March 2025 magnitude 7.7 earthquake on Myanmar’s Sagaing fault, the largest instrumentally recorded event in the region, is an extreme example of rupture behavior on mature, geometrically simple strike-slip systems. Here, we use Interferometric Synthetic Aperture Radar (InSAR) and optical satellite imagery to characterize the coseismic surface deformation field and infer the slip distribution along its ~500-km rupture. Our results reveal a remarkably sharp surface rupture with effectively zero shallow slip deficit along its entire length; peak slip consistently occurred at or near the surface, contrasting sharply with recent large earthquakes on less mature faults with significant off-fault deformation and shallow slip deficit. Furthermore, while the rupture smoothly propagated through seismic gaps, we show that it also re-ruptured segments previously broken in 1929, 1930 and 1956, but with reduced slip in these zones – consistent with a slip-predictable model. These observations highlight that rupture processes on mature plate boundary faults can be highly efficient, with implications for seismic hazard assessment on similar strike-slip fault systems worldwide.

Publishing timely warnings of areas potentially affected by landslides and subsequent events remains challenging in hazard assessment. Recently, seismic monitoring has emerged as a promising technique for landslide detection and location determination. However, the effectiveness of this approach is limited by location errors. To overcome this problem, we propose a hybrid method (a three-step process) through seismic analysis and consideration of prior geomorphological characteristics of areas and aspects in a landslide inventory to constrain the potential landslide locations. The first analysis entails a grid-based single-force inversion of 0.02–0.05 Hz seismic signals by spatial grid spacings of 0.2° and 0.05° to retrieve the single-force direction, the estimated landslide area, and a wide search range for a given source. The second analysis involves two location methods for high-frequency seismic signals (>1 Hz) with a finer spatial grid spacing of 0.01°, gradually reducing the landslide search region. By integrating seismology-determined information with a landslide inventory, potential landslide locations and warning areas could be identified effectively. Then we apply the proposed the three-step analysis method to the 2021 Silabaku landslide in southern Taiwan. The results showed that the seismic techniques we used progressively condensed the initial search range from 1831 to 770 km2, converging to a small area of 99 km2. The estimated landslide area was 0.80±0.04 km2, and the force direction was 154.09° ± 6.76°. Subsequently, according to the earlier information, we identified three possible landslide source locations from the 2020 landslide inventory. These findings have implications for warning for two road sections, one of which was subsequently damaged by debris flow after the Silabaku landslide. This study highlights the effectiveness of the hybrid three-step algorithm in constraining areas potentially affected by landslides, facilitating timely assessments for warnings in southern Taiwan.

This study addresses one of the most commonly-asked questions in synthetic aperture radar (SAR)-based landslide detection: How the choice of datatypes affects the detection performance. In two examples, the 2018 Hokkaido landslides in Japan and the 2017 Putanpunas landslide in Taiwan, we utilize the Growing Split-Based Approach to obtain Bayesian probability maps for such a performance evaluation. Our result shows that the high-resolution, full-polarimetric data offers superior detection capability for landslides in forest areas, followed by single-polarimetric datasets of high spatial resolutions at various radar wavelengths. The medium-resolution single-polarimetric data have comparable performance if the landslide occupies a large area and occurs on bare surfaces, but the detection capability decays significantly for small landslides in forest areas. Our result also indicates that large local incidence angles may not necessarily hinder landslide detection, while areas of small local incidence angles may coincide with layover zones, making the data unusable for detection. The best area under curve value among all datatypes is 0.77, suggesting that the performance of SAR-based landslide detection is limited. The limitation may result from radar wave’s sensitivity to multiple physical factors, including changes in land cover types, local topography, surface roughness and soil moistures.