Search

Deep-sea tsunami detection relies on Deep-ocean Assessment and Reporting of Tsunamis (DART), GNSS buoys, and cabled Ocean-Bottom Pressure (OBP) gauges, which are very expensive and difficult to maintain, and often suffer from vandalism or negligent damage. Here, we exploit the potential of establishing a less expensive and more robust island-based geodetic network for tsunami detecting, source reconstruction and warning. The network locates at the coastline of islands and uses a new technique: GNSS Interferometric Reflectometry (GNSS-IR). GNSS-IR retrieves sea levels from combination of the direct and reflected signals from the sea surface sent by satellites. To test the feasibility and efficiency of such a new geodetic network, we use the South China Sea region as an example, and compare its performance in reconciling the variable slip distribution on the Manila megathrust with the previously designed deep-sea monitoring system, i.e., DARTs and planned cable-based OBP gauges. We find that the newly designed GNSS-IR network could work equally well as the cabled OBP network in detecting tsunamis if the stations are built in strategically chosen locations. Combining GNSS-IR with a Kalman filter approach, we demonstrate that carefully situated coastal GNSS stations at global remote deep-ocean islands could function similarly to conventional tide gauges but with advantages of simultaneously measuring relative sea-level and land-height changes, meanwhile suffering lower risk from damaging sea-level events and potential vandalism. • Coastal GNSS-IR network built in strategically chosen locations could work effectively in detecting and reconstructing tsunami source. • Kalman filter approach can process the sparse sea-level retrievals measured by GNSS-IR to equally sampled data as conventional tide gauge. • Carefully situated coastal GNSS stations at remote deep-ocean islands could complement current tsunami warning system. [ABSTRACT FROM AUTHOR]

The South China Sea region is potentially threatened by tsunami hazards originated from multiple sources: the Manila subduction zone in the east, the Littoral Fault Zone (LFZ) in the north, numerous submarine landslides on the continental slopes and the volcanic islands in the Luzon Strait. Infrequent but potentially devastating tsunami hazard poses a great threat to the populous coastal region, fishery, oil and gas exploitation in the deep sea, etc. Here we review the recent progress in tsunami hazard assessment in the South China Sea region, focusing on two primary sources: submarine earthquakes and landslides. We sort and review the literature by the two commonly used approaches: deterministic and probabilistic tsunami hazard assessment for both source types. By simulating tsunamis generated by typical earthquakes originated from the Manila Trench, the LFZ and landslides in the continental slopes, we investigate their tsunamigenic mechanism and key tsunami characteristics in the South China Sea region. We point out the research gaps and highlight the key issues to be addressed in the future. [ABSTRACT FROM AUTHOR]

The South China Sea (SCS) coastlines are particularly vulnerable to tsunamis due to rapid urbanization and dense infrastructures. However, whether the SCS has experienced ocean-wide tsunamis is still under debate. Some geological records from the offshore islands inside the SCS imply a regional tsunami event may have occurred approximately 1000-yr-ago. We address these questions for the first time using forward numerical simulations in the SCS: what source could be responsible for such deposits, how the deposits were formed and at what conditions do the deposits could be preserved? Here, we assume the validity of the tsunami deposit hypothesis and use the forward model COMCOT-SED to investigate if these tsunami deposits can be linked to potential seismogenic tsunami sources inside the SCS, including megathrust earthquakes (Mw8.8–9.2) from the Manila subduction zone (MSZ) and the 1918 Mw7.5 Nan'ao earthquake from the Littoral Fault Zone (LFZ) in the northern continental shelf of the SCS. We also utilize a reported tsunami deposit from Qing'ao Embayment, Nan'ao Island, China, to exemplify the deposit formation process in a typical barrier-low coastal plain system. Quantitative analysis suggests large earthquakes from the MSZ may account for the reported geological deposits. The full-rupture earthquake scenario could best explain the depositions if such extreme case is geophysically and geologically plausible. If not, earthquakes (∼ Mw 9.0) covering the northern (16–23.5°N) or southern portion (14–19°N) of the MSZ could also be suitable candidates. Such great earthquakes can generate large tsunami waves (>7 m) that inundate the Qing'ao Embayment, transporting sediments and leaving tsunami deposits in low-lying areas. Conversely, tsunamis triggered by nearshore earthquakes in the LFZ cannot cause deposition in Qing'ao Embayment. The barrier-low coastal plain system is well-suited for preserving tsunami deposits, which remain unaltered by subsequent tsunami waves. Our findings offer a quantitative reference for paleo-tsunami research in the SCS and enhance global understanding of the tsunami deposition process in barrier-low coastal plain systems. • The full-rupture earthquake (Mw 9.0) from the MSZ best explains the reported plausible tsunami deposits in the South China Sea. • Tsunami waves >7 m generated by MSZ earthquakes could inundate the Qing'ao Embayment anddeposit sediments in low-lying areas. • The barrier-low coastal plain system in Qing'ao Embayment forms favorable conditions to preserve overwash tsunami deposits. [ABSTRACT FROM AUTHOR]

The Baiyun slide complex contains geological evidence for some of the largest landslide ever discovered in the continental slopes of the South China Sea. High‐resolution seismic data suggest that a variety of landslides with varied scales have occurred repeatedly in this area. The largest landslide reconstructed from bathymetric and seismic data has an estimated spatial coverage of ~5,500 km2 and a conservative volume of ~1,035 km3. Here, using geomorphological and geotechnical data, we construct a series of probable landslide scenarios and assess their tsunamigenic capacity. By treating the slides as deformable mudflows, we simulate the dynamics of landslide movements. The simulated landslide motions match the geophysical observations interpreted in previous studies. Particularly, we are able to reproduce the spatial distribution of observed runout, including the distance, shape, and deposit thickness, for the most credible slide scenario. We investigate tsunami impacts generated by different slide scenarios and highlight the importance of initial water depth, sliding direction, and nearshore bathymetry. The worst‐case scenario is capable of producing basin‐wide tsunami, with maximum wave amplitudes reaching ~5 m near Hong Kong and Macau, 1–3 m in western Philippines, and at least 1 m along central Vietnam, southeast Hainan, and southern Taiwan. The most noticeable phenomenon we observed is that the southern Chinese coast is the hardest‐hit region in all the simulated scenarios regardless of the diverse slide features. We conclude that the persistence of high tsunami impact is caused by the unique bathymetric feature of the wide continental shelf in front of southern China. Plain Language Summary: The tsunami hazard assessment in the South China Sea (SCS) region has been mostly focused on the earthquakes along the Manila Trench in the past, leaving other tsunami sources barely studied. Increasing seafloor exploration has revealed a large number of submarine landslides that are widely distributed in the continental slopes of the SCS. Although less frequent compared with earthquake‐generated tsunamis, the high vulnerability of SCS coastlines demands that we investigate the potential tsunami impacts generated by these submarine landslides. In this study, we assess the tsunamigenic potential of Baiyun landslide, which is characterized by one of the most well‐studied landslides and largest landslides ever discovered in the continental shelf of the SCS. In contrast with previous studies, which use simplified landslide source parameters, we construct six probable slide events with the key slide features inferred from the most updated geophysical data. By treating the landslide movement as deformable mudflows, we find consistency between simulated and observed runout characteristics. The tsunami simulation results suggest that Baiyun landslides have the capacity to generate basin‐wide tsunamis, like its earthquake counterpart from the Manila Trench. The most severely affected coastline is the southern China, which hosts a very high density of coastal cities and major infrastructures. Key Points: Treating Baiyun landslide as translational mudflow allows reproduction of the spatial distribution of observed runoutBaiyun landslide is capable of generating basin‐wide tsunami which would severely affect the coastlines of northern South China SeaSouthern Chinese coast is the hardest‐hit region in all scenarios due to the unique bathymetric feature of the wide continental shelf in front of it [ABSTRACT FROM AUTHOR]

On 4 January 1907, an earthquake occurred off the west coast of Sumatra, Indonesia, with an instrumental surface-wave magnitude (MS) in the range of 7.5–8.0 at periods of ~ 40 s. The tsunami it generated was destructive on the islands of Nias and Simeulue, where it killed hundreds and gave rise to the legend of the S'mong. This tsunami was also observed in other parts of the Indian Ocean basin. Relative to its instrumented magnitude, the size of the tsunami was anomalous, qualifying the event as a "tsunami earthquake". However, unusually for a tsunami earthquake, the shaking on Nias was severe (7 EMS). We revisit the 1907 earthquake with a multidisciplinary approach by extracting evidence describing shaking effects or the tsunami from written documents and by acquiring new seismograms. Combining these, we discriminate two large earthquakes within an hour of each other with clear differences in seismological character. The first we interpret to be a tsunami earthquake with characteristic low levels of shaking, an estimated average seismic moment (M0) of 2.5 × 1028 dyn cm (MW ≈ 8.2) in the frequency band 6–8 mHz, and an epicentral location close to the front of the Sunda Megathrust. The seismograms we analyzed also document a regular growth of moment with period, approaching MW ≈ 8.4 at the longest resolvable period (~ 170 s). For the second earthquake that caused damage on Nias, we estimate MS ≈ 7 based on seismograms and phase data. We also identify two MS ≈ 6 aftershocks within 24 h of the mainshock. Additionally, we present a dataset of 88 locations within the Indian Ocean basin where the tsunami was observed. Using a subset of these, we forward modeled the tsunami to propose a seismic rupture model extending along the Sunda Megathrust for about 220 km (~ 94.7°E to ~ 97°E) with a maximum modeled slip of ~ 21 m. Our new rupture model provides an acceptable fit to our new dataset of tsunami runup and inundation values from 88 local and far-field locations in the Indian Ocean basin. We also urge caution against an over-reliance on the S'mong legend for tsunami evacuation as its premise, that a tsunami will only follow an earthquake with very severe ground motions, is rendered ineffective for tsunami earthquakes. [ABSTRACT FROM AUTHOR]

On 6 February 2023, a local tsunami was recorded in the southeastern Mediterranean Sea following the Mw 7.7 Turkey–Syria inland strike‐slip earthquake. Due to the lack of underwater observation, the tsunami generation mechanism remains mysterious. To understand the source mechanisms, we analyzed the tsunami waveforms of four nearby tide gauges and located possible sources using a backward tsunami ray tracing approach. We then conducted forward numerical modelings for a range of possible source parameters. We show that there were probably two tsunami sources, inside and outside Iskenderun Bay, which may be related to thick coastal sediments. A source inside the Bay with a characteristic length of 7 km produced dominant periods of 10–30 min with negative initial motion, possibly generated by a landslide. Another source of 6 km length outside the Bay produced dominant periods of 2–10 min with positive initial motions, possibly related with liquefaction. Plain Language Summary: Inland earthquakes, especially inland strike‐slip earthquakes rarely produce tsunamis since tsunami waves are normally generated by vertical displacements accompanied with fault ruptures. On 6 February 2023, the Turkey–Syria strike‐slip inland earthquake of Mw 7.7 mysteriously generated a small‐scale tsunami in the southeastern Mediterranean Sea. Due to the complexity of the earthquake source and lack of underwater observation, how the tsunami waves were generated remains mysterious. To understand the tsunami origins, we analyzed the tsunami waveforms in different aspects and adopted numerical modelings to explore the possible sources. We find that there existed two tsunami sources probably generated by landslides inside the Iskenderun Bay and related with liquefactions outside the Bay. We therefore highlight the potential threat caused by the disaster chain of coastal strike‐slip earthquakes. Key Points: We find distinctly different tsunami wave properties inside and outside the Iskenderun BayWe obtain possible tsunami sources that can well explain the observed tsunami waveformsUnderestimated and unpredictable tsunami origins due to the disaster chain caused by coastal strike‐slip earthquakes call for more attention [ABSTRACT FROM AUTHOR]

The rheological behaviors of submarine landslide materials significantly affect the landslide's motion and the associated tsunami hazard. In this study, the effects of rheological behaviors on the landslide dynamics and tsunami hazard are numerically investigated. The Herschel-Bulkley rheological theory and nonlinear shallow water equation model are adopted. Taking one of the large landslides in the South China Sea (SCS), the Baiyun Slide, as an example, we investigate the impacts of remolding rates and initial/residual yield strengths on the speed and runout of landslide and the resulting tsunami wave. Further comparison of the Newtonian viscous model and the rheological model reveals the importance of the materials' properties on landslide dynamics and associated tsunami hazard. The resonance characteristics of landslide-generated tsunamis are studied through fast Fourier transform (FFT) analysis. A parameter set of (τy,0, τy,∞, Γ) = (12 kPa, 3 kPa, 0.05) is found to be optimal for the Baiyun Slide. The slide speed and runout distance are mainly affected by the residual yield strength and remolding rate. The computed arrival time is not sensitive to the rheological parameters, but a smaller residual yield strength and a larger remolding rate result in a larger tsunami amplitude. The standing wave and dominant tsunami wave periods with high energy range from 15 and 26 min. This study reveals that the rheological parameters of landslide have considerable effects on submarine landslide motion and landslide tsunami amplitude, which is helpful for the landslide-generated tsunami assessment. [ABSTRACT FROM AUTHOR]

Due to an unfortunate oversight we did not publish the hypothetical slip model [ABSTRACT FROM AUTHOR]

The 10th South China Sea Tsunami Workshop (SCSTW-10) held in Singapore during October 10-11, 2018 marks the 10 year anniversary of the workshop, which was initiated by Professor Philip L. F. Liu in 2007. In the last 10 years, the topics of SCSTW have expanded from the tsunami hydrodynamics to very diverse disciplines, to include big data analysis, and development of process-based models, in the process, enhancing our understanding of tsunamis. These papers cover topics related to hydrodynamics of tsunami waves, tsunami hazard assessment, new technology for tsunami warning and mitigation. [Extracted from the article]

In this study, we use Pleiades satellite stereo imagery to map shallow‐water bathymetry specifically for the purpose of assessing tsunami hazard in coastal regions. In order to calculate the refraction correction factor, we independently measure the refractive indices of seawater mixed with a variety of phytoplankton, which was previously unquantified. We find that the refractive index of seawater increases with cell density of phytoplankton, ranging from 1.3410 to 1.3425. Through rigorous error analysis, we demonstrate that the change in refraction due to heterogeneous distribution of phytoplankton is negligible, suggesting that the technique is applicable to the global ocean. Using the Penghu Islands as a case example, we show that photogrammetric bathymetry is able to reveal tsunami impacts more accurately and reasonably than the commonly used General Bathymetric Chart of the Oceans data, highlighting the importance of high‐resolution and high‐accuracy bathymetry for tsunami hazard assessment. Plain Language Summary: Tsunami hazard assessment relies heavily on coastal bathymetry (underwater topography) as it ultimately determines tsunami‐wave propagation and inundation. However, high‐resolution and high‐accuracy coastal bathymetric data are scarce worldwide because they are rather difficult and expensive to obtain from traditional methods such as light detection and ranging (LiDAR). Here, we use 0.5‐m resolution Pleiades satellite stereo imagery as an alternative to map shallow‐water bathymetry specifically for the purpose of assessing tsunami hazard in coastal regions. A key step in this method is refraction correction (when the ray of light passes the water/air interface, it bends), in which the refractive index of the media is vital. We measure for the first time the refractive indices of seawater mixed with a variety of phytoplankton. We find that although the refractive indices of seawater change with heterogeneous distribution of phytoplankton, errors arising from such changes are negligible. This demonstrates that the method is generic and can be applied to the global ocean, without any requirements for in situ data. Using the newly derived photogrammetric bathymetry, we simulate a tsunami scenario to show the improvement in our ability of assessing tsunami hazard. Key Points: Satellite photogrammetry provides an effective means of deriving high‐resolution coastal bathymetryError analysis shows that changes in refractive index of seawater due to heterogeneous distribution of phytoplankton are negligibleHigh‐resolution bathymetry captures near‐shore tsunami impacts more accurately than the commonly used General Bathymetric Chart of the Oceans data [ABSTRACT FROM AUTHOR]

Abstract The potential tsunami hazard for atoll islands inside the South China Sea (SCS) has not been investigated previously. In this study, we assess the tsunami hazard for the Xisha Archipelago using a deterministic approach. The tsunami source considered in this study is hypothesized earthquakes in the Manila subduction zone. We employ COMCOT (COrnell Multi-grid Coupled Tsunami model) for tsunami simulations. Different with previous deterministic assessments, our study investigates the uncertainties caused by bathymetric resolution and rupture complexities (here refers to rupture speed and direction variations related to fault geometry complexity). Comparison between results of different resolution suggests that coarser grids may over/underestimate tsunami hazards around mid-ocean atoll islands and this depends on the surrounding bathymetry. But results of different datasets show that GEBCO may lead to a lower result than the multibeam data. We also investigate the uncertainties caused by different rupture speed and direction, which are highly affected by subducting seamounts/ridge. Different rupture direction will shift the hardest hit locations to areas, which are possibly underestimated in cases of instantaneous ruptures. Also, the distribution of slip and location of the rupture zone can affect the local tsunami wave height significantly. We suggest that near-island high-resolution bathymetric data and dynamic rupture process should be considered in future tsunami hazard assessment in locations with similar geological structures. Highlights • Tsunami hazard evaluation inside the SCS is taken for the first time, mainly focusing on the small scattered reef islands. • Tsunami hazard on atoll islands may be over/underestimated when using low-resolution bathymetry data. • Rupture complexity related to seamounts/ridge subduction along the Manila trench should be taken into consideration. [ABSTRACT FROM AUTHOR]

On 30 October 2020, an Mw 6.9 normal faulting earthquake occurred off the northern coasts of Samos Island, Greece. The earthquake with moderate magnitude generated the largest tsunami in the eastern Mediterranean since the 1956 Amorgos event. Here, we investigate the earthquake source characteristics and examine the factors contributing to the exacerbated tsunami using an integrated approach, including the source inversion from InSAR and GPS data, tsunami simulation, and spectral analysis of tsunami waveforms. The results show that a set of conjugate faults, determined in this study, can explain the geodetic surface deformation equally well. Our tsunami modeling results indicate that the shallow north‐dipping fault was most likely responsible for the mainshock. With the key features of our preferred source models, we infer that the tsunami was intensified by several conditions: (a) the spatial distribution of major coseismic slips coincided with the deepest portion of the Samos Basin which served as an energy reservoir for the tsunami generation; (b) the tsunami was generated in a semi‐closed bay which trapped most of the tsunami energy; (c) the east‐west orientation of the mainshock fault and the location of the concentrated coseismic slips directed the majority of the tsunami energy toward Sığacık Bay, which formed the extremely high tsunami of 3.8 m; (d) the primary tsunami wave period, close to natural oscillation frequency of the harbor, resulted in the observed significant oscillation. We conclude that the tsunami of the 2020 Samos earthquake was exacerbated by both the source characteristics and tsunami resonance property. Plain Language Summary: Tsunamis are generally generated by megathrust earthquakes in subduction zones, like the 2004 Mw 9.1 Sumatra‐Andaman earthquake. However, a normal‐faulting earthquake on 30 October 2020, in the Aegean Sea generated the largest tsunami in the eastern Mediterranean since 1956. We investigate how the Mw 6.9 normal earthquake could produce tsunami waves of up to 3 m in height and lead to such severe tsunami damage on the coasts of Izmir, Turkey and Samos Island, Greece. We explore the earthquake source characteristics and the factors exacerbating the effects of the tsunami. We evaluate six representative fault models derived from this study and others by comparing their capability of reproducing the observational data, from which we infer the most likely source features. We also analyze the period components of tsunami waveforms to examine how local bathymetry affects the tsunami behavior. We find that the shallow depth of the earthquake with concentrated slips in the deepest water area of the Kuşadası bay is the key factor in producing adequate initial tsunami energy. The tsunami period coincides with the characteristic oscillation periods of the local bays, causing larger tsunami amplitude and longer duration. Key Points: We find the key features of the tsunami source of the 2020 Samos Earthquake which can best explain a variety of observationsWe identify the favorable conditions which contribute to the exacerbated tsunami hazard in the Kuşadası BayTsunamigenic potential of active faults located in confined sea areas may pose unique tsunami hazard and deserves more attention [ABSTRACT FROM AUTHOR]

In the northern region of the South China Sea, located at the intersection of the continental shelf and slope, there have been recorded four earthquakes with magnitudes M ≥ 6.0. This pattern of seismic activity suggests the presence of a previously unidentified active submarine fault. Notably, no active faults have been documented in this area prior to our investigation, with earlier studies predominantly focusing on the structural morphology and sedimentary basin evolution associated with the South China Sea's spreading phenomenon. Through the examination of seismic profiles and historical earthquake records, we have identified a newly discovered fault, termed the Slope Fault Zone (SFZ). The SFZ exhibits diverse structural components, including a listric normal fault in the western sector and a strike-slip fault zone in the eastern sector. Analysis of shallow seismic profiles and borehole data has revealed that the SFZ intersects with strata from the late Pleistocene epoch, confirming its classification as an active fault. The derived fault parameters indicate that the western segment of the SFZ has the potential to generate earthquakes of considerable magnitude, ranging from M 6.7 to 7.2. Furthermore, given the significant occurrence of submarine landslides along the southern boundary of the SFZ, there exists a risk that such seismic events could trigger slope failures and subsequent tsunamis. In summary, our research has unveiled the presence of an active fault capable of precipitating submarine earthquakes, landslides, and tsunamis. • An unreported marine fault in the northern South China Sea is discovered and named SFZ. • The Pearl River Mouth segment of the SFZ was active since 120,000 years ago. • The Pearl River Mouth segment of the SFZ can produce Mw 6.7–7.2 earthquake. [ABSTRACT FROM AUTHOR]

The distribution of slip during an earthquake and how it propagates among faults in the subduction system play a major role in seismic and tsunami hazards, yet they are poorly understood because offshore observations are often lacking. Here we derive the slip distribution and rupture evolution during the 2016 M w 7.9 Kaikōura (New Zealand) earthquake that reconcile the surface rupture, space geodetic measurements, seismological and tsunami waveform records. We use twelve fault segments, with eleven in the crust and one on the megathrust interface, to model the geodetic data and match the major features of the complex surface ruptures. Our modeling result indicates that a large portion of the moment is distributed on the subduction interface, making a significant contribution to the far field surface deformation and teleseismic body waves. The inclusion of local strong motion and teleseismic waveform data in the joint inversion reveals a unilateral rupture towards northeast with a relatively low averaged rupture speed of ∼1.5 km/s. The first 30 s of the rupture took place on the crustal faults with oblique slip motion and jumped between fault segments that have large differences in strike and dip. The peak moment release occurred at ∼65 s, corresponding to simultaneous rupture of both plate interface and the overlying splay faults with rake angle changes progressively from thrust to strike-slip. The slip on the Papatea fault produced more than 2 m of offshore uplift, making a major contribution to the tsunami at the Kaikōura station, while the northeastern end of the rupture can explain the main features at the Wellington station. Our inversions and simulations illuminate complex up-dip rupture behavior that should be taken into consideration in both seismic and tsunami hazard assessment. The extreme complex rupture behavior also brings new challenges to the earthquake dynamic simulations and understanding the physics of earthquakes. [ABSTRACT FROM AUTHOR]

We propose an optimal deployment scheme of the Seafloor Observation Network (SON) in the South China Sea (SCS), aiming at early warning of potential tsunami hazards based on the data assimilation approach. The SON is composed of offshore bottom pressure gauges (OBPGs). We design the initial location of OBPGs along the isobaths of 500, 1000, and 2000 m. The energy distribution of the tsunami wavefield based on Empirical Orthogonal Function (EOF) analysis is used as a reference. Then, synthetic tsunamis generated by stochastic slip sources along the Manila Trench are adopted to evaluate the performance of the SON. The OBPGs at a depth of 1000 m can make an accurate early warning based on tsunami data assimilation at a reasonable engineering cost. Finally, based on the quantitative evaluation, the optimal deployment scheme of the SON is suggested. Our results indicate that at least three stations are required to cover the coast along southern China to forecast the tsunami in the SCS successfully. We note that our method could also be applied to other regions to design the SON against potential tsunami hazards. • A tsunami early warning system is designed for the South China Sea. • Data assimilation approach is adopted for tsunami forecasting. • Offshore stations along the 1000-m isobath are most cost-effective. • At least three stations are required for constructing tsunami warning system. [ABSTRACT FROM AUTHOR]