In Gutenberg’s era, earthquakes were understood primarily as phenomena involving the release and propagation of seismic wave energy. Since the 1960s, seismic wave radiation has been explained by fault slip, and various characteristics of earthquakes have been successfully described by slip processes governed by friction laws on fault surfaces. As a result, the view that “earthquakes are fault slip” became widely accepted, and earthquake size is now usually represented not by radiated seismic energy but by seismic moment, a measure of fault slip. However, while earthquakes certainly involve fault slip, it is incorrect to assume that all fault slip represents an earthquake.

The discovery and increasing understanding of slow earthquakes in recent decades have made it necessary to reconsider a fundamental question: What exactly is an earthquake? What distinguishes slow earthquakes from regular earthquakes? Under what conditions does this distinction arise? Do regular earthquakes begin in a universal manner? Addressing such questions leads to a view of earthquakes as a coupled process of rock fracture and wave radiation that cascades through hierarchical heterogeneities spanning a wide range of spatial and temporal scales. I aim to discuss approaches for understanding—and ultimately forecasting—such complex, multiscale phenomena.

How to cite: Ide, S.: Rethinking Earthquakes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4764, https://doi.org/10.5194/egusphere-egu26-4764, 2026.

Earthquake source processes are described based on friction laws on faults, bulk constitutive behavior of the surrounding medium, and the stress state of an intricate fault network, including off-fault damage. The combination of these components forms a dynamic earthquake rupture model that accounts for the seismological, geodetic, and geological observations, as well as enables estimation of the earthquake energy budget. While such models help clarify key factors characterizing earthquake source processes, trade-offs among these factors often prevent identifying the dominant physical mechanism. For example, enhanced near-field high-frequency radiation can be attributed to fault roughness, structural heterogeneity, or coseismic off-fault damage (Okubo et al., 2019), each of which can produce similar observational signatures.

To mitigate the modeling uncertainties, information available from laboratory experiments can be utilized, such as fault geometry, bulk elastic properties, stress state, and frictional conditions. Here, I demonstrate this concept through a laboratory study aimed at controlling the size and location of earthquake source patches on a laboratory fault (Okubo et al., in revision). This approach provides a predetermined source configuration that can be incorporated into a dynamic rupture model. Circular gouge patches as earthquake sources were placed on a 4-meter-long laboratory fault in a large-scale biaxial apparatus, generating microearthquakes during the evolution of preslip or afterslip on the entire fault in stick-slip experiments. Acoustic emission waveforms carefully corrected for instrumental response and attenuation suggest that these earthquake clusters exhibit non-self-similar scaling. Using the controlled source geometry and the observed source parameters, we developed a dynamic rupture model that is quantitatively consistent with laboratory observations. Although this model is not a unique solution for explaining the observed non-self-similar scaling, given the limited information available to fully resolve the rupture process even under laboratory conditions, it complements previously proposed models for non-self-similar earthquakes and provides a useful basis for interpreting natural earthquakes.

Close integration of experiments and modeling is key to updating previous findings by addressing limitations in existing modeling frameworks. Large-scale experiments allow for spatially dense measurement arrays relative to the characteristic length scales of dynamic ruptures. Models quantitatively constrained by these high-quality measurements help clarify the details in source processes and play an important role in determining which observables, and at what resolution, are required to effectively monitor faulting activity.

How to cite: Okubo, K.: Advancing Knowledge of Earthquake Source Processes Through Dynamic Rupture Modeling with Natural and Laboratory Observables, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14630, https://doi.org/10.5194/egusphere-egu26-14630, 2026.

North-central Chile is a highly seismically active region. While the last megathrust earthquake occurred in 1730, the area has also experienced large events in recent decades, such as the 2015 Illapel earthquake (Mw 8.3), as well as numerous seismic sequences and persistent swarms. Although these phenomena are widespread along the Chilean subduction margin, their dynamics and potential connection to major earthquakes remain poorly understood. 

Within this framework, the ABYSS project has deployed Distributed Acoustic Sensing (DAS) interrogators along offshore telecommunication fiber-optic cables, complemented by temporary and permanent onshore seismic stations. This configuration offers a unique opportunity to monitor and investigate the offshore microseismicity in a region characterized by sparse permanent instrumentation and the absence of previous offshore sensors.

In this study, we develop a workflow to precisely relocate the seismicity recorded by the ABYSS network. We combine the probabilistic, non-linear hypocentral inversion using NonLinLoc with double-difference relocation using HypoDD, incorporating a 3D P- and S-wave velocity model and differential times derived from waveform cross-correlation on both DAS and onshore stations. Through this integrated approach, we identify and analyze clusters of seismicity associated with swarm activity and short-term seismic sequences. In particular, we apply the workflow to episodes such as the Tongoy swarm initiated on 30 December 2024, whose largest event reached Ml 5.3, and the offshore Ovalle sequence that occurred between October and November 2025.

Our goal is to precisely characterize these sequences by improving constraints on the geometry and spatio-temporal evolution, gaining insights into the processes driving this activity, and shedding light on how present-day swarm dynamics may relate to the occurrence of larger earthquakes along the Chilean subduction margin.

How to cite: Peralta, T., Flores, M. C., Rivet, D., Potin, B., Baillet, M., and Ruiz, S.: High-resolution relocation of seismic swarms using offshore DAS and onshore seismic data in north-central Chile, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-887, https://doi.org/10.5194/egusphere-egu26-887, 2026.

Numerical simulations of earthquake cycles provide essential insights into fault mechanics and the physical interpretation of frictional parameters. Here, we utilize a spring-block system governed by rate-and-state friction to systematically compare earthquake cycle behaviors under quasi-dynamic and fully dynamic conditions. Our simulations demonstrate that for both approaches, the static stress drop, dynamic stress drop, and peak stress scale linearly with the logarithm of the loading rate [ln(Vpl/V0)]; however, the scaling coefficients are distinct and are modulated by both frictional parameters and the system stiffness. Specifically, we observe stress overshoot during the coseismic phase in dynamic models, contrasting with the undershoot observed in quasi-dynamic simulations. Additionally, parameter sweeps reveal that stress drops decrease as the stiffness ratio kc/k increases. This study highlights the importance of the inertial term effect in interpreting earthquake cycle behaviors.

How to cite: Chai, L. and Hu, F.: Scaling of Stress Drop with Rate-and-State Frictional Parameters in Spring-Block Models, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6568, https://doi.org/10.5194/egusphere-egu26-6568, 2026.

Collision and relentless underthrusting of India beneath Eurasia resulted in large-scale deformation of the Indian lithosphere. Anisotropic parameters serve as a good proxies to decipher deformation in such complex orogenic collision zones. In this study, we present anisotropy characteristics of the crust beneath Sikkim Himalaya using harmonic decomposition of P-to-S converted phases identified in P-wave receiver functions (P-RFs). Analysis of azimuthal variation of these phases enabled parameterizing the crustal anisotropic properties, with depth. Initially, 11,087 high quality P-RFs were computed using waveforms of teleseismic earthquakes having magnitude  ≥ 5.5 and signal to noise ratio  ≥ 2.5 within an epicentral distance range of 30° - 100°, recorded at a network of 38 seismic stations deployed in Sikkim Himalaya and the adjoining foreland basin. Analysis of the first three harmonic degrees (i.e. k= 0, 1 and 2) reveals that the upper crustal anisotropy is oriented WSW-ENE to E-W, coinciding well with the trends of crustal microcracks and fractures. The mid to lower crustal anisotropy aligns predominantly with the dipping decollement layer along which the Indian plate is underthrusting Tibet. An orthogonal reorientation is observed within the extent of the Dhubri-Chungthang Fault Zone authenticating its role in segmenting the orogen. The lower crustal anisotropy is highly perturbed signifying a highly heterogeneous nature of the Moho.  Existence of multiple layers of anisotropy possessing distinct geometries varying with depth could be an indication of a highly complex deformational regime resulting from active crustal shortening.

How to cite: Kumar, G., Singh, A., Singh, C., Saikia, D., and Kumar, M. R.: Crustal Seismic anisotropy in Sikkim Himalaya: Implications for deformation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10972, https://doi.org/10.5194/egusphere-egu26-10972, 2026.

Regular monitoring of small to moderate sources of continuous earthquake events in the complex tectonics of Himalayan region helps in clearly defining the ongoing seismotectonic process. The study of moment tensor inversion to decipher the fault planes responsible for current seismic activity in the Kishtwar region of Northwest part of Himalaya has been undertaken by establishing a six-station network in 2022 and among them 15 events of shallow origin with magnitude ranging from M_L ~ 3.0 to 4.0 occurred in the local region of seismic network are used for the moment tensor inversion. A few number of studies didn’t able to clearly demarcate the actual scenario of seismotectonics in the northwest part of Himalaya due to its difficult terrain and complex geology. This area has been studied for fault plane solution by a software package ISOLA based on MATLAB programming environment. The source inversion is performed via iterative deconvolution method and synthetic seismogram is generated through green’s function computation via discrete wavenumber method using the regional crustal velocity model. However, the inversion is performed at several trial source position and at various frequency bands based on the epicenter distance and the magnitude of earthquake to find the best solution resulting from the maximum correlation between the recorded and synthetically generated waveforms. A 2D space-time grid search is performed for determining the optimal time and positon of earthquake generation. Perhaps calculating source parameters such as moment magnitude, centroid depth and fault parameters equally with describing uncertainty quantities such as variance reduction factor and condition number will deliver the reliability and stability to the solution. A strong follow-up uncertainty quantification can justify the best estimated fault plane solution. Quality of earthquake event can be calculated through their DC and CLVD percentage and maximum & minimum compression stress direction. Focal mechanism solution of these events following thrust with strike-slip focal mechanism and represents the compressional regime in north-northeastern direction. The centroid depth obtained by moment tensor inversion of all events falls within the depth zone of Main Himalayan Thrust (MHT) suggesting seismicity is concentrated along the major detachment in the region.

How to cite: Tiwari, S. and Gupta, S. C.: Moment tensor analysis and uncertainty quantification of local earthquake events: tectonic implication in the northwestern Himalayan region, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13656, https://doi.org/10.5194/egusphere-egu26-13656, 2026.

With the popularization of dense seismic array observations, tomographic imaging of subsurface velocity structures using surface wave dispersion data extracted via subarray partitioning has emerged as a new trend. The primary advantages of subarray-based dispersion data extraction lie in its reduced susceptibility to the inhomogeneous distribution of noise sources, which yields more stable and reliable dispersion measurements. Additionally, this approach enhances the energy of higher-order modes, thereby providing tighter constraints on subsurface velocity structures. Compared with the higher-order modes of Rayleigh waves, both the fundamental and higher-order modes of Love waves exhibit simpler dispersion characteristics with fewer mode crossings and overlaps, making them more favorable for joint inversion to constrain subsurface SH-wave velocity structures.

Traditional subarray surface wave imaging methods (e.g., SSWI) typically perform 1D velocity structure inversion at individual locations first, followed by stitching all 1D models to generate pseudo-2D or 3D velocity models. Despite its simplicity and computational efficiency, this direct stitching strategy is highly vulnerable to uneven station distributions, and the resultant velocity models may suffer from artificial velocity jumps. To address these limitations, Luo & Yao (2025) proposed a direct subarray surface wave imaging method (SSWDI), which eliminates the stitching step inherent in traditional methods and incorporates spatial smoothness constraints on velocity structures, thus enabling more robust inversion of subarray-derived dispersion data for subsurface imaging. However, the SSWDI method originally focused exclusively on the fundamental mode of Rayleigh waves. In this study, we further extend the SSWDI framework to accommodate both fundamental and higher-order modes of Love waves, and validate the improved method using both numerical synthetic data and field observational data.

Reference

Luo, S., and H. Yao (2025), Direct Tomography of S-wave Structure Using Subarray Surface Wave Dispersion Data: Methodology and Validation, Geophysics, 1–60, doi:10.1190/geo-2024-0515.

How to cite: Luo, S.: 3D SH-wave Velocity Tomography via Direct Inversion of Multimode Love Wave Dispersion Curves from Seismic Subarrays, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16937, https://doi.org/10.5194/egusphere-egu26-16937, 2026.

Seismic hazard assessment is crucial for the design of critical facilities, whose damage could lead to severe consequences. The design of such facilities typically requires the definition of seismic actions associated with recurrence periods on the order of 5,000-10,000 years. Earthquakes with such low frequencies are well documented in highly deforming regions, where paleoseismic records commonly encompass several seismic cycles of active faults. In contrast, in slow-deforming regions or areas of low seismicity, the scarcity of seismic data hinders the definition of seismogenic zones. In this context, geological studies of active seismogenic faults are essential, as they allow the characterisation of seismic behaviour over time spans far exceeding those covered by instrumental or historical records. These data can contribute to constraining fault’s seismic cycles and estimating earthquake magnitude–frequency distributions at the fault scale.

Despite their importance, the incorporation of faults into seismic hazard models remains challenging, particularly in low strain regions such as the western margin of the Valencia Trough. This region of the NE of Iberia (from the Vallès-Penedès Graben to the Valencia Depression) corresponds to a passive margin characterised by a basin-and-range structure, bounded by multiple NNE–SSW-oriented normal faults formed during the Neogene rifting episode. Those faults are usually associated with mountain fronts, although our recent studies have found some new faults crosscutting Pleistocene alluvial fans. These newly discovered faults are being studied by means of geomorphology, geophysics, paleoseismology and geochronology in order to estimate their seismic parameters. Several challenges arise when analysing these faults, including fault identification, incomplete geological records, and the need for complex dating techniques.

Moreover, in regions characterised by fault systems, fault interactions may play a significant role. In regions such as the studied area, these interactions may result in long quiescent periods followed by phases of increased activity or even cascading events. Under such conditions, distinguishing between quiescent and active phases is especially difficult, as recurrence intervals are expected to span several thousands of years in both cases.

In this work, we explore existing methodologies for the computation of seismic hazard incorporating geological data from faults and fault systems in slow-deforming regions, using the western margin of the Valencia Trough as a case study. To this end, a detailed geometric characterization of the fault system is carried out to establish the geometric relationships among faults. Recent morphotectonic analyses and newly acquired geological data are then used to constrain the seismic parameters of the studied faults and to estimate their earthquake frequency distributions. Finally, several alternative seismic source models are proposed, forming the basis for the construction of a logic tree for subsequent seismic hazard calculations. These
models, although in progress, provide a framework for improving seismic hazard assessments in slow-deforming regions, contributing to safer design of critical infrastructure.

How to cite: Ollé-López, M., García-Mayordomo, J., Scotti, O., and Masana, E.: A seismogenic modelling approach for rift-basin fault systems in slow-deforming regions: application to the western margin of the Valencia Trough, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20741, https://doi.org/10.5194/egusphere-egu26-20741, 2026.

The 2015 Mw 7.8 Gorkha earthquake was followed by numerous aftershocks that provided important information on active faulting in central Nepal. Accurate moment tensor estimations are essential for determining the source parameters of these seismic events. In this study, we determine double-couple and full moment tensor solutions for selected aftershocks of the 2015 Nepal earthquake sequence using a regional 1D velocity model.

The waveform data recorded by the temporary broadband network (NAMASTE) are used to analyse 51 aftershocks with M > 3.5. A library of Green’s functions is computed using the frequency–wavenumber method based on a 1D velocity model of the Nepal region. Synthetic waveforms derived from the Green’s functions are used to invert the waveform data for moment tensor estimation. Both body waves and surface waves are used in the inversion, and they contribute separately to the moment tensor solutions. The analysis focuses on regional waveforms in relatively higher frequency ranges.

Both double-couple–constrained and full moment tensor inversions are performed, and the resulting source parameters are examined in terms of waveform fit, centroid depth, and fault-plane orientation. This work presents a set of moment tensor solutions for the 2015 Nepal aftershocks using a 1D regional velocity model and provides a reference for future studies using more complex velocity structures.

How to cite: Lahon, P., Silwal, V., and Mahanta, R.: Double-Couple and Full Moment Tensor Solutions of the 2015 Nepal Aftershocks, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21099, https://doi.org/10.5194/egusphere-egu26-21099, 2026.

The alternative proxy parameters for seismic site amplification beyond the conventional time-averaged shear wave velocity of the upper 30 m (V_S,30) are investigated in this study with a focus on quantities that can be derived or constrained from surface wave-based measurements such as Spectral Analysis of Surface Waves (SASW) and Continuous Surface Wave System (CSWS) testing. Surface wave methods provide dispersion curves that are inverted to obtain near-surface shear wave velocity profiles, which are then used to construct synthetic one-dimensional layered models for ground response analysis. For each profile, two different candidate site parameters are evaluated, including V_S,30 and the impedance ratio between the surface layer and the underlying half-space. These parameters are chosen to reflect what can realistically be inferred from SASW/CSWS-derived velocity profiles, particularly the shallow stiffness and impedance contrasts that strongly influence amplification. Correlation analyses are carried out to quantify how well each parameter explains the variability in amplification across the synthetic suite. The results are used to assess whether the impedance ratio provides stronger or more consistent correlation with amplification than V_S,30, thereby offering guidance on how surface wave–based site characterization can be better integrated into proxy-based amplification and site classification schemes in seismic design practice.

How to cite: Singh, V. and Baidya, D. K.: Evaluating SASW/CSWS-Derived Proxies for Seismic Site Amplification, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21425, https://doi.org/10.5194/egusphere-egu26-21425, 2026.

The Rif’s belt is characterized by low to moderate seismic activity resulting from the continental collision between the African and Eurasian plates. This seismic activity, which involves devastation and human losses, requires an in-depth study of its origins and mechanisms. This study aims to identify the geological structures responsible for seismic activity in the eastern Rif by adopting an integrated methodological approach. The methodology relies on the use of a Geographic Information System (GIS) to process and analyze multiple geological, seismological, and geophysical datasets. Various filters were applied to magnetic and gravimetric data (vertical derivatives) to characterize the subsurface. The analysis of earthquake focal mechanisms helped identify active faults. The results show that the seismicity, with a NW-SE orientation, is localized within a fragile depression south of the city of Selouane. The final geological model highlights a system of faults and strike-slips oriented NE-SW and NW-SE. A significant spatial correlation is observed between epicenters and Messinian-aged NW-SE strike-slips, suggesting their reactivation. The analysis indicates that a system of dextral strike-slips is likely the source of this seismic activity. The proposed geodynamic model represents a major advancement in understanding local seismic activities and serves as an essential reference for future studies. These results significantly contribute to the assessment and management of seismic risks, thereby enhancing the safety and resilience of populations in this high-risk area.

KEYWORDS: Geodynamic model; Seismotectonic; Focal mechanism; Magnetic; Gravimetric; ·
Eastern Rif. 

How to cite: Iken, H., Nouayti, A., Nouayti, N., and Khattach, D.: An integrated geodynamic analysis of seismic sources in the Eastern Rif: Insights from geological, seismological, gravimetric, and aeromagnetic data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-22669, https://doi.org/10.5194/egusphere-egu26-22669, 2026.

Current demand for critical metals including Cu is outstripping current supply and will further escalate in the future. A significant source of Cu comes from porphyry deposits, which contribute to >60% of global Cu ore production. Many of these porphyry Cu deposits are found along convergent margins such as the Andes and the Sunda-Banda arc in Indonesia and these same arcs also host highly active volcanoes. Understanding the magmatic and geodynamic factors that contribute towards priming magmas for Cu fertility as opposed to volcanic eruptions can aid in identification of prospective targets for exploration.

Here we present analyses of apatite populations from known porphyry Cu deposits and active volcanoes along the Sunda-Banda arc in Indonesia. To gain a complete overview of the mineral associations and their information, we incorporate textural information to analyze both apatite inclusions and their mineral hosts, such as pyroxenes and amphiboles, as well as groundmass apatite. These mineral compositions will serve as input for thermodynamic models to constrain the volatile chemistry and budget, as well as the volatile saturation depths. The information gathered will be combined to test our working hypotheses that the magmas with high Cu fertility store at distinct depths, have geochemical signatures that suggest deep fractionation of garnet and amphibole, and are associated with anomalous geodynamic features such as slab tears.

Our ongoing work advances current understanding on magma storage and transfer along and across fertile magmatic arcs, aiming to better understand magmatic pre-conditioning for porphyry copper deposit formation to complement exploration efforts to find copper deposits in the geological records.

How to cite: Utami, S. B., Ubide, T., Rosenbaum, G., Li, W., Handini, E., Wood, S., Handley, H., and Goode, L.: Insights into the copper accumulation potential of magmas along the Sunda-Banda arc, Indonesia from apatite and its mineral hosts, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-938, https://doi.org/10.5194/egusphere-egu26-938, 2026.

Abstract:

Strike-slip basement faults and their related segments are crucial for oil and gas exploration. These faults are considered favorable channels for hydrocarbon migration. The multistage activities of these faults influence the development of hydrocarbon-bearing structures. They can also produce fracture systems that enhance reservoir properties and boost oil and gas production. Understanding how strike-slip fault segments and their associated structures affect hydrocarbon accumulation is essential for geological research and exploration planning. This study aims to characterize the geometry and structural evolution of the strike-slip basement fault with Pan-African or Arabian trends, investigate the relationship between fault segments, and assess their impact on the distribution of hydrocarbon traps. This research focuses on the structural and tectono-sedimentary analyses of the Kazerun fault system based on processing and interpretation of the surface data (e.g., satellite images and aeromagnetic data) and the subsurface data (e.g., 2D and 3D seismic and well data) in the Zagros orogenic belt, SW Iran. The relationship between the segmented strike-slip fault zone and hydrocarbon reservoirs is analyzed through map view patterns and profile features. Results reveal that the Arabian-trending Kazerun fault system comprises segmented dextral strike-slip faults and is considered a transform and wrench fault. These faults display various planar configurations, including linear, en-echelon, horsetail splays, and irregular geometries in the map view. Based on the seismic data interpretation, three structural styles develop along the Kazerun strike-slip fault zone, including vertical or oblique, pull-apart (negative flower structure), and push-up (positive flower structure) segments. Releasing and restraining bends and oversteps formed at the tail end of the Kazerun strike-slip fault segments. In the study area, salt diapirism occurred along the pull-apart segment and the releasing bend. Hydrocarbon traps are developed in the push-up segment and the restraining bend. Fractures are less prominent in the vertical segments but more developed in push-up and pull-apart segments, which act as pathways for fluid migration and improving reservoir quality. The push-up segment and restraining bend exhibit a higher degree of branching fractures, making them the most significant for reservoir development. This research shows that strike-slip fault segmentation (in the form of fault overlapping or stepping) and their lateral linkage control the reservoir distribution and connectivity. Recognizing the growth and lateral connections of strike-slip fault segments is crucial for structural analysis and predicting fault-controlled reservoirs. These findings offer valuable insights into the structural characteristics of strike-slip fault zones and can enhance oil and gas exploration in the Zagros fold-and-thrust belt and other similar regions.

Keywords:

Strike-slip basement fault, Segmentation pattern, Oil/Gas fields, Zagros orogenic belt, SW Iran

How to cite: Soleimany, B., Tajmir Riahi, Z., Payrovian, G. R., and Sepahvand, S.: Impact of segmentation pattern of the Pan-African trending strike-slip basement fault on the spatial distribution of hydrocarbon traps in SW Iran, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1377, https://doi.org/10.5194/egusphere-egu26-1377, 2026.

Skarn-type Cu–Fe–Au mineralization in the Middle–Lower Yangtze River Metallogenic Belt (MLYRB) is closely associated with Early Cretaceous intermediate to felsic magmatism; however, the links between magmatic evolution and ore-forming efficiency remain poorly constrained. In the Tonglushan ore field, one of the largest Cu–Fe–Au skarn systems in eastern China, multiple intrusive phases are spatially distributed, providing an ideal opportunity to investigate how magmatic processes control metallogenic potential. Here we present new geochronological and geochemical constraints on quartz monzodiorite porphyry, quartz monzodiorite, quartz diorite, and their mafic microgranular enclaves (MMEs) from different sectors of the Tonglushan ore field.

Zircon U–Pb ages indicate synchronous emplacement of all intrusive phases and MMEs at ca. 142–140 Ma. Whole-rock geochemistry and Sr–Nd–Hf isotopes indicate that these intrusive rocks belong to a high-K calc-alkaline to weakly adakitic series and were derived from an enriched lithospheric mantle source modified by slab-derived components, followed by extensive fractional crystallization. The MMEs record efficient mixing between mafic and felsic magmas, highlighting the role of mafic recharge in supplying heat and metal components to the evolving system. Estimates of magmatic water contents and oxygen fugacity from zircon compositions reveal systematic variations among different intrusions. The Jiguanzui and Tonglushan quartz monzodiorite porphyries are characterized by high water contents and elevated oxidation states, consistent with intense Cu–Au and Cu–Fe–Au mineralization, whereas the weakly mineralized Zhengjiawan quartz diorite exhibits lower values. These observations suggest that, beyond structural controls, the metallogenic fertility of intrusions in the Tonglushan ore field was primarily governed by fractional crystallization, mafic magma input, and the development of highly hydrous and oxidized magmatic systems.

Our study demonstrates that integrated whole-rock and zircon geochemical indicators provide effective tools for evaluating the ore-forming potential of skarn-type Cu–Fe–Au mineralization related intrusions in the MLYRB.

How to cite: Zhang, M. and Tan, J.: Magmatic controls on skarn-type Cu–Fe–Au mineralization in the Tonglushan ore field, Middle–Lower Yangtze River Metallogenic Belt, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2162, https://doi.org/10.5194/egusphere-egu26-2162, 2026.

Low-temperature thermochronology provides key constraints on the post-mineralization exhumation and preservation of orogenic gold deposits. In this study, we investigate the exhumation histories of the Anjiayingzi and Jinchanggouliang gold deposits, located respectively in the Kalaqin metamorphic core complex (MCC) and the Nuluerhu magmatic dome within the Chifeng–Chaoyang metallogenic belt on the northern margin of the North China Craton.

Both deposits formed in the Early Cretaceous (~130 Ma), but at significantly different depths (5.6–7.1 km for Anjiayingzi and 1.0–2.6 km for Jinchanggouliang), and are currently exposed at the surface, implying differential post-mineralization exhumation. Zircon and apatite (U–Th)/He and fission-track analyses were conducted on ore-hosting rocks to reconstruct cooling and exhumation histories. Combined age–elevation relationships and thermal history modeling reveal that the Anjiayingzi deposit experienced multi-stage, rapid exhumation totaling ~6.75 km since mineralization, with the most intense exhumation occurring between 130 and 80 Ma. In contrast, the Jinchanggouliang deposit underwent slower and more limited exhumation, with a total exhumation of ~2.50 km over the same period.

The contrasting exhumation histories coincide with an Early Cretaceous regional extensional regime affecting the northern margin of the North China Craton. We suggest that tectonic setting plays a first-order role in controlling post-mineralization exhumation. Deposits hosted within MCCs are characterized by rapid extensional denudation related to detachment faulting, whereas deposits hosted in magmatic domes are mainly exhumed through regional uplift and surface erosion. These results emphasize the importance of structural architecture in governing the exhumation, preservation, and exposure of gold deposits in extensional orogenic systems.

How to cite: Li, A. and Fu, L.: Post-mineralization exhumation of gold deposits on the northern margin of the North China Craton: constraints from low-temperature thermochronology, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2243, https://doi.org/10.5194/egusphere-egu26-2243, 2026.

The Southeast Anatolian Suture Belt hosts the oceanic and continental remnants of the southern Neotethyan realm. During the Late Cretaceous, the southern Neotethyan domain experienced an Andean-type magmatism on its northern continental margin (the Tauride-Anatolide Platform), characterized by the Baskil Magmatics. The plutonic part of this unit is intruded by numerous dikes, which are the primary focus of this study. The U-Pb zircon dating of the dikes and their granodioritic host rocks indicates that their emplacement occurred within a narrow interval, between 81-79 Ma. The dikes vary chemically from basalt to dacite, while the host rocks range from andesitic to dacitic. On the normal mid-ocean ridge (N-MORB)-normalized plots, all samples exhibit negative Nb anomalies. Trace element systematics reveals that this dike system is chemically heterogeneous, consisting of five distinct chemical types. The elemental and isotope ratios indicate varying contributions from depleted and enriched components. All chemical types, with relative Nb depletion, suggest incorporation of slab-derived and/or crustal additions. This interpretation is further supported by the EM-2-like Pb isotopic ratios. Based on the variability in elemental and isotopic composition, this intrusive system appears highly heterogeneous, likely due to the combined effects of mantle source, crustal contamination, and fractional crystallization. The bulk geochemical characteristics of the studied dikes and their host rocks suggest that these intrusives formed at a continental arc. Considering the available paleontological and geochronological age data, it appears that the intraoceanic subduction and continental arc magmatism in the Southern Neotethys occurred simultaneously; the former created the Yüksekova arc-basin system, whereas the latter formed the Baskil Arc.

Note: This study was supported by project Fübap-MF.15.12.

How to cite: Ural, M., Sayit, K., Koralay, E., and Göncüoglu, M. C.: Geochemical and Geochronological approaches of Baskil Dikes (Elazığ, Eastern Turkey): Discrimination between the Late Cretaceous Continental and Oceanic Arc-related Magmatism in the Southern Neotethys, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2274, https://doi.org/10.5194/egusphere-egu26-2274, 2026.

The Kaladgi Basin, an E–W trending intracratonic basin in the northern part of the Dharwar Craton, preserves favourable structural and stratigraphic conditions for sandstone-hosted and unconformity-related U–REE mineralization. In the study area, the Neoproterozoic Cave Temple Arenite (CTA) of the Badami Group unconformably overlies deformed Mesoproterozoic rocks of the Bagalkot Group. The crystalline basement of the Kaladgi Supergroup comprises Meso- to Neoarchaean Peninsular Gneiss and the Chitradurga Greenstone Belt. This association of cratonic basement, schist belt, and basin-margin fault and fold systems provides an excellent structural framework for hydrothermal fluid circulation and mineralization.

Detailed thematic mapping at 1:25,000 scale in the Ramdurg–Suriban sector reveals that NNW–SSE–oriented Dharwarian stress generated a series of anticlines and synclines involving the Saundatti Quartzite, Malaprabha Phyllite, and Yaragatti Argillite, as constrained by conjugate fracture analysis and S–C fabric development. An E–W trending tectonic fault defines the contact between the Peninsular Gneissic Complex and Saundatti Quartzite, with comparable faulted contacts also developed within the Bagalkot Group. Intense faulting resulted in silicification, chalcedonic brecciation, and pervasive hydrothermal alteration along these contact zones. Transverse normal faults with associated brecciation accommodate strain related to the main E–W structure and indicate episodic reactivation of the basin architecture.

Fusion ICP–MS analysis of 20 bedrock samples collected proximal to these fault zones shows U₂₃₈ concentrations exceeding twice the threshold values of National Geochemical Mapping (NGCM) stream sediment sample. Uranium enrichment is spatially associated with Malaprabha Phyllite, first-cycle CTA, and silicified banded hematite quartzite veins of the Hiriyur Formation. Chondrite-normalized (La/Yb)_n versus (Eu/Eu*)_n systematics indicates a dominantly low-temperature basinal brine hydrothermal system characterized by low (La/Yb)_n <25 and negative Eu anomalies. Redox-sensitive (Ce/Ce*)_n versus (Eu/Eu*)_n plots further indicate reducing fluid conditions. In contrast, quartz–chlorite veins developed within sheared Malaprabha Phyllite and younger dolerite record comparatively higher-temperature fluids, marked by Eu²⁺ mobilization ((Eu/Eu*)_n > 0.8) and negative Ce anomalies. These results suggest that reactivated, structure-controlled tectonites acted as effective fluid pathways, with the TTG-dominated Peninsular Gneissic Complex serving as a likely uranium source and contributing to localized U–REE mineralization along the basin margin.

How to cite: Mahanandia, A., Lal, M. M., Singh, T. G., Nandhagopal, N., and Singh, S.: Structure-controlled Uranium + REE mineralization in low temperature basinal brine hydrothermal system at the contact of Kaladgi Basin and Peninsular Gneissic Complex, South India, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2923, https://doi.org/10.5194/egusphere-egu26-2923, 2026.

Gravity observations are widely used in volcanic monitoring to infer subsurface mass redistributions, commonly interpreted in terms of magma intrusion. However, gravity changes may also arise from thermo-poro-elastic (TPE) processes associated with temperature and pore-pressure variations in fluid-saturated reservoirs. Neglecting these effects can lead to ambiguous or misleading interpretations of gravity signals during volcanic unrest.

The recent development of TPE inclusion models allows us to describe the mechanical fields induced by fluid-saturated rock volumes undergoing pore-pressure and temperature variations. These sources can coexist with magmatic sources within volcanic systems and are typically located at shallower depths than the deep magmatic reservoir, which acts as the primary engine by releasing hot fluids. These exsolved fluids rise from depth and either accumulate in, or migrate through, overlying brittle rock volumes, which respond to thermal and pore-pressure perturbations and therefore act as secondary sources of deformation and gravity change. In this work, we consider a disk-shaped TPE inclusion, a geometry that has been successfully applied in previous studies to represent deformation fields that are predominantly radial and associated with axisymmetric sources.

The results show that gravity variations induced by a TPE inclusion depend strongly on the fluid phase. Both liquid water and gaseous fluids can produce the same significant ground uplift, but lead to different gravity residuals: negative for liquid water and minor but positive for gaseous fluids. In contrast, condensation or vaporization of a thin layer near the surface can generate large gravity changes without notable deformation. As a result, heating and pressurization of a TPE inclusion can mask or weaken the gravitational signature of magma ascent, complicating the interpretation of gravity data and highlighting the need to account for hydrothermal effects when estimating magma volumes during unrest.

Gravity data collected over the past decades at the Campi Flegrei caldera (Italy) provide an ideal test site for applying our model and offer intriguing insights into both past and current unrest phases, although our results are applicable to any volcanic system with an active hydrothermal system. These findings highlight the importance of incorporating TPE effects into gravity data interpretation and integrated volcano monitoring strategies. Accounting for them improves our ability to distinguish between magmatic and hydrothermal contributions, leading to more robust assessments of subsurface dynamics and volcanic hazards.

How to cite: Nespoli, M., Bonafede, M., and Belardinelli, M. E.: Thermo-Poro-Elastic effects as hidden drivers of gravity signals in volcanic systems, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3001, https://doi.org/10.5194/egusphere-egu26-3001, 2026.

Abstract: Unlike ionosphere, troposphere is nondispersive and delays cannot be determined from observations of signals at different radio frequencies. In GNSS data processing, station height, receiver clock error and tropospheric delay are highly correlated to each other, especially in kinematic situations. Although zenith hydrostatic delay can be provided with sufficient accuracy, zenith wet delay, which is more spatially and temporally varying than hydrostatic component, has to be carefully processed. Usually, temporal dependence of tropospheric delays at zenith is modeled as a random walk process with a solely given process noise rate σ_rw in GNSS processing. The usually used σ_rw is a constant throughout whole process session and is in range of 3~10 mm per sqrt hour. This setting is generally appropriate for desirable GNSS positioning estimation in normal conditions. However, modeling zenith tropospheric delay by using a constant σ_rw in whole session will be unsatisfactory in cases of special weather conditions, e.g., the shower case. The σ_rw is a measure of magnitude of typical variation of zenith path delay or its residual after calibration in a given time. Values of σ_rw that are too large could weaken strength for geodetic estimation, while values that are too small may introduce systematic errors, since a strong constraint for tropospheric unknowns is imposed to stabilize the system. The random walk model for wet delay must be constrained approximately to "correct" value to obtain optimum parameters estimates. Assuming temporal change of tropospheric delay at an arbitrary station can be described by random walk model, the process noise levels were calculated by some scholars. They employed water vapor radiometric, surface meteorological measurements and numerical weather model data set for optimum selection of σ_rw. In general, although a lot of efforts have made to optimize post-processing and/or real-time GNSS tropospheric delay estimation, stochastic modeling of zenith wet delay remains insufficiently investigated, especially for kinematic applications. Since temporal variation of zenith wet delay depends on water vapor content in atmosphere, it seems to be reasonable that constraints should be geographically and/or time dependent. In this work, we first investigate sensitivity of both station coordinates and zenith wet delay estimators on different σ_rw values, and then try to propose to take benefit from post-processed static or kinematic estimated tropospheric delay to obtain the optimum σ_rw. The general objective is that if zenith tropospheric delays are of different variation characteristic, e.g., relatively stable or rapid changing, then a varying σ_rw, e.g., small or large value, could be employed, which should be more theoretically feasible compared with a invariant σ_rw. The initial results show that the new method can efficiently obtain epoch-wise σ_rw values at different stations. Compared to results from conventional constant σ_rw value, time-varying noise rate can improve precision of PPP solutions. We note that this first results represent performance view at several selected stations, more works should be done to draw global or even long-term conclusions.

This work is supported by National Natural Science Foundation of China (42304010), Youth Foundation of Changzhou Institute of Technology (YN21046).

How to cite: Wang, M., Lu, B., and Zhong, Q.: The initial results about optimum the random walk process noise rate for GNSS tropospheric delay estimation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4327, https://doi.org/10.5194/egusphere-egu26-4327, 2026.

The Yanshan-Liaoning metallogenic belt (YLMB), the second-largest molybdenum deposit cluster in China, hosts over twenty porphyry molybdenum deposits, including the large-scale Caosiyao, Sadaigoumen, and Dasuji deposits, as well as the newly discovered medium- to large-scale Qiandongdamiao, Zhujiawa, and Taipingcun deposits. Geochronological data indicate that the duration of molybdenum mineralization spanned ca. 100 Myrs, from the Triassic to Early Cretaceous (240–140 Ma). However, the reasons for such a prolonged or multi-period metallogenic event, and the magmatic and geodynamic processes controlling the spatial–temporal distribution of these deposits, remain poorly understood.

Here we summarize the geological, chronological and geochemical data from selected molybdenum deposit to reconstruct the temporal–spatial distribution and tectonic setting of ore- metallogenic history in the YLMB. The formation of molybdenum deposit in the YLMB can be divided into three periods of 240–220 Ma, 185–180 Ma and 160–140 Ma. The ore-forming intrusions among these three periods illustrate an overall characteristic that metaluminous to peraluminous, high-K calc-alkalic to shoshonite series acidic rocks, and the source of intrusions is the Archaean–Paleoproterozoic lower crust. Through in-depth analysis of Sr-Nd-Hf isotopic data, we find that the magma source that during the 185-180 Ma stage is relatively younger, mainly reflecting the partial melting of Paleoproterozoic crust, whereas the magma source that during the 240–220 Ma and 160–140 Ma stages likely are contained both from the Paleoproterozoic and Neoarchean crust. Further calculations using trace element content ratios reveal a shallower magma source along the magma evolution during the 240–220 Ma period, which supported by the gradual decrease trend in crustal thickness. In contrast, the calculation of crustal thickness during the 185–180 Ma and 160–140 Ma stages show an increase trend, suggested an thicken process in the depth of the magma source.

Spatially, the porphyry molybdenum deposits formed during these three periods exhibit distinct geographic distributions. Deposits formed at 240–220 Ma are mainly located in the northern part of the YLMB, including the Chengde-Zhangbei-Fengning district. Those formed at 185–180 Ma are primarily located in the Liaoxi district, eastern part of the YLMB while deposits formed at 160–140 Ma are located in the southern part of the YLMB, particularly in the Xinghe-Zhangjiakou-Xinglong district. We propose that the variations of the spatial–temporal distribution and geochemical characteristics of the molybdenum deposit formed during different periods in the YLMB are controlled by variations of their geodynamic settings. The porphyry molybdenum deposits formed in 240–220 Ma are under the post-collision or post-orogenic extension environment between the North China Plate and the Siberian Plate in the Middle Triassic. Deposits formed in 185–180 Ma are under the extension environment in the early stage of the Yanshanian movement, and porphyry molybdenum deposits formed in 160–140 Ma are in the strong extrusion environment in the main stage of the Yanshanian movement.

Our findings demonstrate the multi-period metallogenic history of the YLMB, highlighting the critical role of magma source, storage depth, and geodynamic setting in controlling the formation of porphyry molybdenum deposits.

How to cite: Jiang, C., Liu, Q., Cao, L., Li, A., and Fu, L.: Magmatic and geodynamic processes control on the formation of porphyry molybdenum deposits: Insights from the Yanshan-Liaoning metallogenic belt, northern margin of North China Craton, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4396, https://doi.org/10.5194/egusphere-egu26-4396, 2026.

Microcontinents are isolated fragments of continental crust surrounded by oceanic lithosphere. They commonly occur in modern ocean and are also recognized in orogenic system. They can be accreted onto continental margins through collision and subduction during ocean-continent subduction process, and lead to migration of subduction zone toward the oceanic side. However, it is not well understood whether and how this process can be recorded by metamorphism. In this study, a high grade metamorphic-magmatic terrane is recognized along the previously defined Qilian block. The Datong-Mengyuan terrane (DMT) is separated from the low-medium grade metamorphic basement of the Qilian block (QLB) by dextral strike-slip ductile shear zone and ophiolite mélange. The petrology and texturally-controlled U-Pb multi-mineral geochronology reveal that the mafic and felsic granulites from the DMT record two significance events of metamorphism. The earlier event experienced a pressure and temperature conditions of 11.4–13.7 kbar and 735–805°C at ca. 500 Ma, and later stage records a pressure and temperature conditions of 5.5–9.6 kbar and 790–840°C at ca. 460 Ma. We suggest that the earlier Cambrian high pressure granulite facies metamorphism is resulted from collision and thickening related to the accretion of the DMT to the Qilian block, and the later low-medium pressure granulite facies overprinting formed by decompression heating, which happened in continental arc setting and is associated with shift of subduction zone toward the ocean. These findings provide a critical example of metamorphic record on the microcontinent accretion and convergent plate boundary dynamics.

How to cite: Mao, X. and Zhang, J.: Metamorphism records microcontinent accretion and subduction relocation: an example from early Paleozoic Qilian Orogenic Belt, NW China, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4872, https://doi.org/10.5194/egusphere-egu26-4872, 2026.

The distribution of rare gases within the Earth’s interior has caught the attention of scientists for the past few years. The inertness and volatility of noble gases make them excellent tracers for understanding the chemical evolution of Earth’s mantle and atmosphere. Previous studies indicate that noble gases can be found associated with clathrates, form their own oxides, or, in some cases, noble gases such as helium and xenon can even bond with Fe under extreme pressure (p) - temperature (T) conditions like those in Earth’s core. However, the ability of lower mantle mineral phases to house rare gases remains poorly understood, leaving important gaps in knowledge. Helium and argon are noble gases of interest in this study. The isotopes ⁴He and ⁴⁰Ar are produced from the radioactive decay of ²³⁸U and ⁴⁰K within the Earth’s interior, while ³He and ³⁶Ar are regarded as primordial, introduced during the accretion of Earth. Dong et al. (2022) revealed that noble gases can become reactive under mantle pressure conditions. Still, their ability to be incorporated into mantle minerals via adsorption needs to be thoroughly studied, as there are many limitations in the experiments conducted to measure the solubility of noble gases in minerals under mantle p-T conditions. In this study, we investigated the adsorption behavior of helium and argon on the (001) plane of periclase (MgO) by employing first-principles density functional theory (DFT) calculations.

Adsorption energies were estimated across pressures ranging from 0 to 125 GPa, representative of conditions throughout Earth’s interior, i.e., approximately up to the Core Mantle Boundary (CMB). At ambient pressure, both helium and argon showed negative adsorption energies, indicating stable adsorption relative to isolated species (MgO, Ar, He). These energies became increasingly negative with pressure, becoming notably negative beyond 75 GPa which corresponds to lower mantle pressures. This may be due to the accelerated reactivity of noble gases at extreme pressure conditions, as reported in previous studies. Additionally, under all pressure conditions argon exhibited stronger adsorption than helium, indicating enhanced argon retention in lower mantle conditions. However, further investigations into the mechanical and dynamical stability of these adsorbed structures are required to completely understand the mechanisms governing noble gas occurrence in the Earth’s lower mantle.

How to cite: Karangara, A. and Kumar Das, P.: Adsorption of Helium and Argon on the (001) Surface of Periclase: A First-Principles Study, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4988, https://doi.org/10.5194/egusphere-egu26-4988, 2026.

Soil CO₂ emission is a key proxy for investigating fluid migration processes associated with volcanic and tectonic activity. In particular, the analysis of the spatial distribution of geochemical anomalies represents an effective tool for identifying active structures and zones of ongoing deformation. Numerous studies have shown that faults and fracture systems play a fundamental role in controlling the localization and evolution of surface geochemical anomalies.

Vulcano Island (Aeolian Archipelago, Italy) is characterized by intense hydrothermal activity and persistent soil CO₂ emissions, providing a natural laboratory to investigate the relationships between fluid circulation and active tectonic structures. In this study, we present an integrated analysis of soil CO₂ fluxes based on results obtained from periodic surveys and continuous soil CO₂ flux records acquired at key sites across the island.

Periodic measurements are performed on fixed spatial grids, allowing the production of soil CO₂ flux maps and the identification of areas characterized by elevated degassing rates. At selected sites, the carbon isotopic composition of gases is analyzed to constrain gas sources.

These spatial datasets provide insights into the structural control exerted by the main tectonic lineaments on gas release at the surface. Continuous CO₂ flux monitoring enables the investigation of temporal variations and transient degassing signals potentially related to seismic and tectonic processes. In particular, the recent volcanic crisis at Vulcano Island, started on 2021, characterized by a marked increase in soil CO₂ flux, allowed a more detailed identification of preferential CO₂ emission pathways, highlighting zones of enhanced permeability associated with fault and fracture systems.

This work is carried out within the framework of the CAVEAT project (Central-southern Aeolian islands: Volcanism and tEArIng in the Tyrrhenian subduction system), which aims to provide a comprehensive understanding of the current geodynamics of the southern Tyrrhenian region.

How to cite: De Gregorio, S., Camarda, M., Capasso, G., Di Martino, R. M. R., Pisciotta, A., and Prano, V.: Soil CO₂ Emissions as Indicators of Fluid Pathways in Volcanic–Tectonic Environments: Insights from Vulcano Island, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5147, https://doi.org/10.5194/egusphere-egu26-5147, 2026.

Mg-bearing minerals, including brucite [Mg(OH)₂], lizardite [Mg₃(Si₂O₅)(OH)₄] and iowaite [Mg₆Fe³⁺₂(OH)₁₆Cl₂·4H₂O] are variably reactive with carbon dioxide (CO₂) at Earth’s surface conditions and can be used to mineralize and sequester this greenhouse gas. Here, we assess the impact of temperature (5, 20 and 40 °C) on the rate of CO₂ mineralization of these minerals. At each temperature, mineral powders (~100 mg ) were placed in a 7.5-litre flow-through reactor that was supplied with humidified laboratory air (0.042% CO₂; 100% RH) at ~200 mL/min. Subsamples (n = 54) of each mineral were collected over 3 months and analyzed (XRD, TIC, BET) to ascertain the amount and rate of carbonation as a function of time, temperature, and mineral feedstock.

Preliminary X-ray diffraction (XRD) results show the formation of dypingite [Mg₅(CO₃)₄(OH)₂·5H₂O] and a decrease in the abundance of brucite over time. The 003 peak of iowaite shifted to smaller d-spacings, indicating replacement of chloride by carbonate ions and a transition to a more pyroaurite-rich [Mg₆Fe³⁺₂(CO₃)(OH)₁₆·4H₂O] composition. Total Inorganic Carbon (TIC) measurements were used to determine the amount and rate of carbonation as a function of time, temperature, and mineralogy.

The results of this study will help us estimate the carbonation kinetics of these minerals in ultramafic ores and mine tailings under different temperature conditions relevant to large-scale deployment of CO₂ mineralization at mines across the globe.

How to cite: Sulemana, S. Z., Wilson, S., Moyo, A., Akhtar, S., Power, I. M., and Sleep, S.: Temperature-dependence of CO2 drawdown into Mg-bearing minerals., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6294, https://doi.org/10.5194/egusphere-egu26-6294, 2026.

The main focus of the study is to calibrate Sentinel-1 InSAR Line-of-Sight (LOS) velocities along a ~700 km North-South transect extending from the Black Sea coast (Kastamonu-Samsun) to the Mediterranean (Mersin-Gaziantep). This transect encompasses diverse tectonic regimes, including the North Anatolian Fault Zone, the Central Anatolian Block, and the junction of the East Anatolian Fault Zone. This complex structure of the transect requires detailed analysis of the GNSS-InSAR calibration procedure including validation. 

Across the study region, processed LiCSAR products are integrated with 3D velocities derived from the continuous local CORS network (21 stations) and an extensive campaign-based GNSS network (200 stations). For calibration, GNSS velocities are first projected into the satellite LOS geometry using LOS vectors derived from coherent InSAR pixels within a 1-km radius. The velocity bias (ΔVlos) is calculated at continuous GNSS locations. This correction surface is propagated using various conventional and Machine Learning techniques independently, including Kriging, Weighted Least Squares (WLS) based Quadratic Surface fitting, Thin Plate Spline (TPS) and Radial Basis Functions (Gaussian, Multiquadric, and Inverse Multiquadric). To address specific error sources, the contributions of topography-correlated atmospheric delays and local spatial trends are also analyzed by Geographically Weighted Regression (GWR) and Random Forest regression. Cross-validation is applied to assess the quality of each model individually where spatial random sampling and plate boundaries are also considered. This study presents preliminary results for obtaining a validated basis for generating up-to-date velocity fields over Türkiye.

How to cite: Elvanlı, M. and Durmaz, M.: Comparative Analysis of Machine Learning and Geostatistical Approaches for GNSS-InSAR Integration: A Case Study in Anatolia, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7178, https://doi.org/10.5194/egusphere-egu26-7178, 2026.

Geomagnetic storms (GS) significantly perturb the near-Earth environment, leading to enhanced thermosphere density, increased non-conservative forces, and degraded satellite orbit determination, particularly for Doppler-based techniques such as DORIS. In this study, we investigate and improve DORIS orbit determination performance during GS conditions by developing storm-adapted processing strategies. Storm days were classified using geomagnetic indices and categorized into moderate to severe storm levels (G3-G5).

Four distinct processing strategies were implemented and evaluated: a standard operational solution and three experimental storm-adapted solutions, designed through systematic modifications of drag constraints and observation-elimination criteria. These strategies were tested through targeted daily and weekly experiments conducted across multiple DORIS-equipped satellites, with a particular emphasis on periods of intense storms.

The storm-adapted strategies demonstrate clear performance improvements relative to the standard solution during geomagnetic storms. The modified strategies reduce orbit residual RMS in all orbital components, improve Length-of-Day (LOD) variance by approximately 40-80%, and decrease LOD mean biases by nearly 60%. Additionally, Earth Rotation Parameters (ERP) exhibit notable improvements, with reductions of approximately 22–25% in both bias and variability for the polar motion components (X/Y pole). Among the tested configurations, the combined strategy, particularly when applied with zero-rotation constraints, consistently delivers the best performance during intense storm conditions (Kp ≥ 8+). These results demonstrate that storm-adapted DORIS processing strategies significantly enhance orbit and geophysical parameter estimation during disturbed space-weather conditions.

How to cite: Kumar, V., Stepanek, P., Filler, V., Balasubramanian, N., and Dikshit, O.: Impact of Storm-Adapted DORIS Processing on Orbit Quality and Earth Rotation Parameters During Geomagnetic Storms , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7764, https://doi.org/10.5194/egusphere-egu26-7764, 2026.

Abstract: The study of fluid inclusions and sulfur isotope characteristics of barite deposits is crucial for tracing the source of ore-forming materials and predicting prospecting targets. Current research indicate that the Sangmuchang barite deposit in northern Guizhou is primarily hosted within the joint fractures of dolomites in the Sinian Dengying Formation and Cambrian Qingxudong Formation. The contact between ore bodies and surrounding rocks is distinct, with the orebodies occurring as veins and lenticular. The ore textures are mainly veinlets, stockworks, massive, and banded, while the ore structures consist of inequigranular tabular-columnar blastic, fine-crystalline, and arenaceous texture.  Fluid inclusion studies reveal that  the inclusions are single-phase aqueous inclusions. Microthermometric measurements of 33 inclusions show that their homogenization temperatures range from 81°C to 182°C, with an average of 132°C; Salinity values vary from 9.61 wt.% NaCl eqv to 20.63 wt.% NaCl eqv, with an average of 17.53 wt.% NaCl eqv. Ten sulfur isotope analyses from the deposit show that the δ³⁴SV-CDT values range from 40.89‰ to 46.95‰, with a mean of +44.51‰.The characteristics of fluid inclusion salinity, temperature and sulfur isotopes suggest that the ore-forming fluids of this barite deposit are characterized by moderate-low temperature and moderate-high salinity. These ore-forming fluids were mainly derived from basin brines, with contributions from meteoric water. The significant enrichment of heavy sulfur isotopes and homogeneous sulfur isotope composition reveal that the sulfur source of ore-forming materials in this barite deposit is a relatively singular source for the sulfur in the ore-forming materials, which is similar to the δ³⁴S characteristics of Sinian marine evaporites, suggesting a close genetic relationship between the sulfur source and evaporites. Therefore, the Sangmuchang barite deposit is interpreted as a moderate-low temperature hydrothermal deposit.  It was formed by the migration of moderate -low temperature hydrothermal fluids in the sedimentary basin, which leached ore-forming materials from underlying and surrounding barium-rich evaporite sequences, followed by precipitation within structural fracture zones under the mixing of meteoric water. The structural fracture zones and areas indicative of fluid migration pathways along the basin margin are important targets for exploration prediction. Keywords: ore-forming fluid; fluid inclusion; sulfur isotope; barite; northern Guizhou

How to cite: Chen, Y., Wang, J., and Liu, Z.: Study on the Source of Ore-Forming Materials of the Sangmuchang Barite Deposit in Northern Guizhou, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8505, https://doi.org/10.5194/egusphere-egu26-8505, 2026.

Uranium is a strategically important metal with applications in nuclear energy, medicine, radiometric dating, food processing, industrial radiography, material sciences, and catalysis. This study presents a detailed microtextural and geochemical investigation of uranium mineralization from the Bodal uranium mine, Mohla-Manpur-Chowki, Central India. Uranium occurs as both crystalline and colloidal precipitates, with coffinite [U(SiO₄)_1-x(OH)_4x] and gummite representing the dominant uranium-bearing phases. The mineralization is spatially and genetically associated with altered metabasalts. Petrographic and geochemical evidence indicates that late-stage hydrothermal alteration played a crucial role in uranium remobilization and ore enrichment. Sulphide minerals, including cobaltite (CoAsS), galena (PbS), arsenopyrite (FeAsS), and chalcopyrite (CuFeS₂), are intimately associated with uranium phases and likely acted as effective reductants and sorption substrates, facilitating uranium precipitation under reducing conditions. The ore assemblage is accompanied by abundant accessory minerals such as zircon, allanite, and apatite. Substitution of U⁴⁺ for Zr⁴⁺ in zircon locally records uranium-rich hydrothermal fluids and contributes to zirconium enrichment. Collectively, these observations suggest that hydrothermal fluid–rock interaction and redox-controlled precipitation were the dominant processes responsible for uranium enrichment at the Bodal mine.

Keywords: Uranium mineralization; Hydrothermal alteration; Redox-controlled precipitation; Bodal mine; Central India

How to cite: Ganveer, S., Mallick, S. P., and Pruseth, K. L.: Hydrothermal remobilization and redox trapping of uranium in metabasalts of the Bodal mine, Central India, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11094, https://doi.org/10.5194/egusphere-egu26-11094, 2026.

The Kirazlı porphyry and high-sulfidation (HS) epithermal system is situated in the central Biga Peninsula of northwestern Türkiye, a region characterized by the protracted closure of the Tethyan oceanic branches and the subsequent collision of Gondwana-derived continental fragments with the Sakarya Zone. This geodynamic framework facilitated the development of diverse tectono-magmatic environments, leading to the formation of porphyry and associated hydrothermal mineralization during the Cenozoic. Based on established geochronological data, magmatism in the Biga Peninsula occurred in five discrete chronostratigraphic episodes: Paleocene to Early Eocene (65–49 Ma), Middle–Late Eocene (49–35 Ma), Late Eocene to Early Oligocene (35–23 Ma), Late Oligocene to Middle Miocene (~23–14 Ma), and Late Miocene to Pliocene (14–5 Ma). Mineralization within the Kirazlı district is temporally constrained to two primary intervals—Late Eocene to Early Oligocene and Oligocene to Early Miocene corresponding to specific magmatic pulses and structurally mediated by major regional shear zones.

Integration of the ages of fault-hosting lithologies, structural styles, fault geometries, and paleostress reconstructions indicates three distinct tectonic phases consistent with the regional Cenozoic evolution: (1) NW–SE extension (Phase-1), (2) NNE–SSW extension (Phase-2), and (3) NE–SW extension (Phase-3). Detailed field observations, petrographic analysis, and microstructural investigations of oriented samples demonstrate that the porphyry and HS-epithermal stages were governed by these shifting stress regimes. B- and D-veins associated with the porphyry stage exhibit preferred orientations along an ENE–WSW strike, consistent with the NW–SE extensional regime of Phase-1. In contrast, late-stage quartz veins within the HS-epithermal overprint formed under a NNE–SSW extensional stress field, aligning with the Phase-2 tectonic pulse.

Analysis of fault planes for both Phase-2 and Phase-3 indicates that ENE–WSW and NE–SW strike directions are common to both phases. Phase-3 displays kinematic and geometric features characteristic of the modern transtensional NE–SW and strike-slip regime currently active in the Biga Peninsula. Correlation of these structural data with magmatism–mineralization age constraints indicates that the porphyry and HS-epithermal components of the Kirazlı system were emplaced during distinct tectonic periods. This evolution reflects the transition from a post-collisional setting to the current extensional and strike-slip dominated regime of western Anatolia.

How to cite: Çam, M., Kuşcu, İ., and Kaymakcı, N.: Tectono-Magmatic Evolution and Structural Controls on the Kirazlı Porphyry-High Sulfidation Epithermal System, Biga Peninsula, NW Türkiye, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11195, https://doi.org/10.5194/egusphere-egu26-11195, 2026.

El Chichón is an active volcano in Chiapas, Mexico, that features a hydrothermal system characterized by thermal springs, fumaroles and an acid crater lake. Many studies have focused on tracking the geochemical evolution of its fluids since its last eruption in 1982 and some have specifically aimed to evaluate the geothermal potential.  This work assesses the evolution of the geothermal potential through time using published geochemical data (1983-2025). We use geochemical diagrams, temperatures estimated with geothermometers and water-rock interaction analysis to identify the main system changes that influence the geothermal potential estimations. Given that El Chichón has been considered  a geothermal prospect since the 1980s, we discuss the possible uses of this resource in terms of its recent active seismicity, the risk scenarios and the local socio-cultural context. 

How to cite: Salas Ferman, J. L., Jácome Paz, M. P., Campion, R., Armienta, M. A., and Inguaggiato, S.:  Temporal evaluation of El Chichon´s geothermal potential in the period of 1983-2025. , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15407, https://doi.org/10.5194/egusphere-egu26-15407, 2026.

Traditional methods for determining geopotential and height require successive transfers of leveling and gravity measurements, which are prone to error accumulation, face challenges in transoceanic applications, and are generally time-consuming, labor-intensive, and inefficient. Based on the principles of general relativity, an alternative approach using high-precision time-frequency signals to determine geopotential can overcome these limitations. In this study, simulation experiments were conducted to determine geopotential differences using BDS and Galileo five-frequency undifferenced carrier phase time-frequency transfer technology. The simulations employed clocks with different performance characteristics, utilizing precise clock offsets and multi-frequency observation data from both systems. The results show that the frequency stability achieved by BDS and Galileo five-frequency undifferenced carrier phase time-frequency transfer can reach approximately 3×10⁻¹⁷. The root mean square of the determined geopotential differences corresponds to centimeter-level equivalent height accuracy, and the convergence accuracy of the geopotential difference by the final epoch can reach better than 3.0 m²·s⁻². Given the rapid development of GNSS multi-frequency signals and ongoing improvements in the precision of products such as code and phase biases, geopotential determination based on Galileo and BDS multi-frequency signals is expected to have broader application prospects in the future. This study was supported by the National Natural Science Foundation of China project (No. 42304095), the Key Project of Natural Science Research in Universities of Anhui Province (No. 2023AH051634), the Chuzhou University Research Initiation Fund Project (No. 2023qd07).

How to cite: Xu, W. and Song, J.: Geopotential Difference Determination via BDS and Galileo Multi-Frequency Time-Frequency Signals, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15725, https://doi.org/10.5194/egusphere-egu26-15725, 2026.

How well mixed is Earth's mantle? Are there primordial reservoirs? What fraction of the mantle feeds surface volcanism? We attempt to address these questions using large-scale Lagrangian particle tracking in time-reversed and forward convection models. We track particles backward in time using a Back-and-Forth Nudging (BFN) method applied to time-reversed thermal convection, initialized with a present-day seismic–geodynamic–mineral physics model (Glisovic & Forte, 2016, 2025). We likewise carried out long-term (multi-hundred-million-year) forward-in-time mantle convection simulations initialized with present-day mantle structure inferred from tomography. In all cases, we employ mantle viscosity structure that has been independently constrained and verified against a wide suite of present-day geodynamic observables that include free-air gravity anomalies, dynamic surface topography, horizontal divergence of plate velocities, excess core-mantle boundary ellipticity, and glacial isostatic adjustment data. A voxel-based analysis quantifies sampling density, residence time, and flux throughout the mantle.

We use different particle starting conditions, each designed to address a specific aspect of mantle mixing. To identify long-lived isolated regions, we track uniformly distributed particles both forward and backward in time, calculating residence times to locate candidate reservoirs. To estimate the sampling of lower mantle material in the upper mantle, we initialize particles in the D" layer and track them forward to determine what fraction reaches the upper mantle. To address plume dynamics and sampling, we place cylindrical arrays of particles beneath present-day hotspots and track them backward, using the statistical evolution of their standard deviation to quantify mixing along transport pathways, with transit time, and voxel analysis. To measure upper-to-lower mantle exchange, we initialize particles uniformly in the upper mantle. By combining these approaches, we systematically identify regions of low flux and high residence time, candidates for reservoirs. We further take a statistical approach based on voxel density sampling to quantify mixing across the volume of the mantle.

How to cite: Johnston, G., Anderson, M., Forte, A., and Glišović, P.: How Well Is the Mantle Sampled? A Global Voxel-Based Analysis of Residence Time and Flux from Forward- and Reverse-Time Mantle Convection, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15990, https://doi.org/10.5194/egusphere-egu26-15990, 2026.

Accurate geoid models are essential for converting GNSS-derived heights into physically meaningful elevations and for ensuring consistency in modern height reference systems. This study presents a unified geodetic framework for refining gravimetric geoids using GNSS/leveling residuals through physically interpretable fitting models. Five correction representations are evaluated, ranging from local Cartesian planar surfaces to geodetically consistent spherical formulations of increasing degree. The analysis demonstrates that low-order models effectively remove regional bias and tilt but show limited predictive stability. To enhance robustness, iteratively reweighted least squares is applied to mitigate the influence of outliers while preserving deterministic structure. Higher-order geodetic models are stabilized using ridge regularization, with the regularization strength selected objectively through leave-one-out cross-validation. This strategy ensures numerical conditioning while directly optimizing predictive performance. Results show that the full degree-2 geodetic model offers the best balance among accuracy, stability, and physical interpretability. It reduces long-wavelength distortions while maintaining consistent in-sample and cross-validated performance. The proposed approach supports reliable GNSS-based height determination in modern vertical datum realization and height modernization efforts.

How to cite: Abdalla, A. and Dwira, C.: Geodetic degree-based Models for Robust Regional Geoid Refinement, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16148, https://doi.org/10.5194/egusphere-egu26-16148, 2026.

Sulfur released by magmatic activity strongly impacts the climate and is essential for ore mineralisation. Many porphyry systems contain up to billions of tons of sulfur, far exceeding the sulfur capacity of silicate melt and therefore requiring an additional, efficient S‑transfer mechanism.

We present a unique mafic rock (SiO₂ = 53–59 wt.%, MgO = 5.3–7.3 wt.%) containing ~15–20 vol.% anhydrite, ~30–40 vol.% biotite and ~40–50 vol.% plagioclase from the largest porphyry–epithermal system in China. Magmatic anhydrite, indicated by textural relations and LREE‑rich compositions, yields bulk‑rock S contents of ~2–3 wt.%, far above experimental S solubilities.

Plagioclase shows sharp core–rim decreases from An₅₀–₇₀ to An₂₅–₄₅, recording strong CaO depletion caused by sulfate saturation. Extensive sulfate saturation also suppressed amphibole/orthopyroxene and removed a large proportion of LREEs from the melt, producing flat REE patterns in co-crystallised apatite. Biotite exhibits pronounced Ba depletion from core to rim. Because Ba partitions strongly into sulfate melt, not into anhydrite, this Ba zoning is best explained by the formation of a sulfate melt, rather than by crystallisation of anhydrite from a silicate melt.

Nd isotopic compositions (ԑNd(t) ≈ -1.0) indicate that the magma was derived from partial melting of the mantle wedge. We suggest that ascent of this oxidised, sulfur‑rich mafic magma led to decompression-driven oxidation of S²⁻ to S⁶⁺, sulfate saturation, and exsolution of an immiscible sulfate melt. This discrete sulfate‑melt migrated upward and provided an efficient pathway for long‑distance transfer of large amounts of sulfur to porphyry systems. This sulfate‑melt exsolution process is a previously unrecognised mechanism that relaxes the constraint imposed by the sulfur capacity of silicate melt, and LREE‑depleted apatite associated with abundant magmatic sulfate phases may serve as an indicator of sulfate‑melt exsolution and a proxy for porphyry mineralisation potential in the upper crust.

How to cite: Huang, W., Humphreys, M., and Liang, H.: Magmatic sulfate‑melt exsolution as a mechanism for excess sulfur in porphyry systems, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-22040, https://doi.org/10.5194/egusphere-egu26-22040, 2026.

The Local Magnitude (M_L), was the earliest proposed magnitude scale, allows for rapid determination based on observed amplitudes and a zero magnitude reference amplitudes (A₀) derived from local events. However, the amplitudes are susceptible to external factors, the physically robust Moment Magnitude (M_w) was proposed. The previous studies showed a 1:1 relationship between M_L and M_w for M_L < 6.5 in Southern California; however, M_L in Taiwan tend to overestimate when compare to M_w due to different regional attenuation characteristics. Although the previous study has recalibrated the logA₀ attenuation model for shallow earthquakes in Taiwan, deep events exhibit an even more significant overestimation, averaging overestimate 0.528. Therefore, this study aims to discuss deep earthquakes in the Taiwan and establish a new logA₀ attenuation model for deep events. Since models relying solely on hypocentral distance (R) result in depth-dependent residuals, a depth term (D) was incorporated to account for the physical characteristic of deep seismic waves often propagating through high-Q plates. The derived regression model is:

logA₀ = 0.097 - 1.587logR - 0.0014R + 0.417logD ± 0.273

The results demonstrate that logA₀ attenuation varies distinctly with distance at different depths, aligning with Richter mentioned that different depth events require distinct calibration. Furthermore, the logA₀ value at a hypocentral distance 100 km differs from that of shallow event models, indicating the difference in attenuation properties. The recalibrated M_L demonstrates no depth dependency and a consistent 1:1 relationship with M_w, with a standard deviation ±0.160. This study proposed model provides a rapid and precise M_L calculation. This new model enhances the reliability of real-time hazard assessment and reduces magnitude conversion errors in catalog combination.

How to cite: Lin, X.-Y. and Wu, Y.-M.: Discussion of Local Magnitude (ML) Scale for Deep Earthquakes in Taiwan, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3299, https://doi.org/10.5194/egusphere-egu26-3299, 2026.

Constraining seismic velocity discontinuities at depths of ~410 km (D410) and ~660 km (D660) is critical for understanding mantle dynamics. Receiver function (RF) studies have revealed the widespread presence of a low-velocity layer (LVL) immediately above D410, supporting the global water‐filter model. Numerical experiments further predict the occurrence of double LVLs above D410 under specific conditions. However, the relationship between LVL occurrence and underlying mantle processes remains poorly understood. Conventional RF analyses rely on careful selection of high-quality traces from multiple teleseismic events at individual stations, followed by slant stacking within a time window around the P-to-S converted phase (P410S) from D410. This approach reduces data coverage and reproducibility and depends on reference Earth velocity models to align RFs. Because true mantle velocity structure and discontinuity depths deviate from these models, slant stacking is often ineffective, particularly for weak and complex converted phases associated with LVLs above D410. In this study, we apply an automated, data-driven machine learning approach to delineate LVLs above D410 using RFs from global seismic stations. Our method aligns the P410S phase without invoking theoretical travel times, instead relying on patterns inherent in the data. We demonstrate that the resulting alignment is physically meaningful and directly reflects velocitystructure near D410. This automated framework significantly improves efficiency, objectivity, and reproducibility in RF analysis. Our results reveal three global patterns of seismic velocity structure near D410: (1) a thick LVL associated with cold mantle regions and subducting slabs; (2) a double LVL, with variable inter-layer spacing, linked to hot mantle and fast upwelling; and (3) a thin LVL correlated with slower upwelling. These observations indicate that LVLs above D410 are a global feature consistent with the water‐filter model, while their detailed characteristics reflect variations in mantle upwelling style beneath seismic stations. 

How to cite: Koireng, T. R. and Bharadwaj, P.: Spatial variations of the low-velocity layer above the 410-km discontinuity and their relationship to mantle dynamics, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7533, https://doi.org/10.5194/egusphere-egu26-7533, 2026.

The Anatolian Plate has a complex tectonic structure. This is mainly caused by the northward movement of the Arabian Plate and subduction along the Hellenic Arc. Many large earthquakes in Türkiye occur along the right-lateral North Anatolian Fault and the left-lateral East Anatolian Fault. Western Anatolia is dominated by north-south extension. This extension formed the Aegean Extensional Province, which includes E-W trending grabens and NW-SE oriented active faults. One of the key structures in this area is the Simav Fault Zone (SFZ). In this study, we focused on the 28 September 2025 Kütahya-Simav (M_w 5.5) earthquake located on the SFZ. Earthquake catalogs provide basic source information. However, they are often not enough to fully describe fault geometry and fault segmentation. For this reason, a detailed source analysis is needed. In this study, we present a moment tensor solution for the 2025 Kütahya-Simav earthquake. The analysis is carried out using the ISOLA and regional waveform inversion. Broadband seismic data from the AFAD and KOERI networks are used. Both point-source and possible multiple-source inversions are performed. These inversions are used to estimate the focal mechanism, seismic moment, and centroid depth. The quality of the solutions is evaluated using Variance Reduction (VR) and Condition Number (CN) values. The results are compared with previous studies and historical data from the Simav Fault Zone. This allows us to obtain a more reliable solution than standard catalog results. The findings explain the present stress pattern and fault behavior in the Kütahya-Simav region.

How to cite: Altuntaş, N. and Taymaz, T.: Source Characteristics of the 28 September 2025 Mw 5.5 Kutahya-Simav Earthquake Sequence in Western Turkiye, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7814, https://doi.org/10.5194/egusphere-egu26-7814, 2026.

The Western Churchill Province (WCP) of northern Canada is a complex assemblage of Archean to Proterozoic crust, yet its lithospheric architecture remains incompletely constrained. We present a comprehensive receiver function study of crustal structure across the WCP using approximately 5,000 high-quality P-wave receiver functions recorded at 39 broadband seismic stations between 2000 and 2025. Data quality was ensured using DeepRFQC, a machine-learning–based automated quality control framework.

Crustal thickness and bulk composition were estimated through H–κ stacking, while harmonic decomposition and differential evolution inversion of receiver functions using RAYSUM were applied to investigate crustal anisotropy, dipping interfaces, and seismic velocity structure. The mean Moho depth across the WCP is 40 ± 5 km, with pronounced lateral variability. The deepest Moho is observed in the northernmost and southernmost regions, whereas the central WCP exhibits comparatively thinner crust.

Bulk VP/VS ratios are relatively uniform (1.76–1.79), consistent with predominantly felsic crust, although elevated values near northern Hudson Bay suggest localized mafic intrusions. Harmonic decomposition reveals coherent azimuthal patterns indicative of crustal anisotropy and dipping structures, with mid-crustal discontinuities identified at approximately 9 and 30 km depth.

The orientations of harmonic components correlate with regional magnetic anomalies and independent SKS shear-wave splitting measurements, implying strong coupling between crustal structure and lithospheric mantle fabric. Comparisons with Bouguer gravity anomalies indicate that gravity variations are primarily controlled by subsurface density variations, as reflected in VP/VS ratios, rather than by Moho topography. These results provide new constraints on the crustal architecture and lithospheric coherence of the Western Churchill Province.

How to cite: Sabermahani, S., Frederiksen, A., and Drayson, D.: Receiver Function Constraints in the Western ChurchillProvince, Northern Canada, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8116, https://doi.org/10.5194/egusphere-egu26-8116, 2026.

Currently, when an earthquake with a magnitude (ML) of 3.5 or greater occurs ob the Korean Peninsula, the Korea Meteorological Administration (KMA) performs a post-event precision analysis to address parameters not captured during the real-time detection phase. To ensure consistent and operationally application, we developed the Post-Earthquake Precision Analysis System, which standardizes the entire workflow from data input to staged outputs.The system enhances location accuracy through precise hypocenter relocation, including relative relocation, and characterizes rupture geometry and kinematics using P-wave first-motion polarity and waveform inversion for focal mechanism solutions. Additionally, spectrum-based source parameter analysis is integrated to quantify the physical characteristics of the seismic source. To improve the completeness of the aftershock catalog, the system employs template matching to detect micro-seismic events that often go unnoticed.The Post-Earthquake Precision Analysis System was applied to major seismic events in the Korean Peninsula, yielding high-precision relocated hypocenters, focal mechanisms, and source parameters. We also generated enhanced aftershock lists and systematically organized the spatio-temporal evolution of each sequence in a standardized format. This study presents an operationally-ready analytical workflow and a core output framework for the systematic post-earthquake precision analysis of domestic seismic activity.

How to cite: Byun, A.-H., Jo, E., Choi, M., Choi, Y., Min, K., and Park, S.-C.: Establishment of a precision post-earthquake analysis system and its application to major earthquake on the Korean Peninsula, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8789, https://doi.org/10.5194/egusphere-egu26-8789, 2026.

A network of fourteen broadband seismic stations has been established in the foreland basin of the Bhutan Himalaya to investigate the seismically active Dhubri-Chungthang Fault Zone (DCFZ). To assess station reliability and data integrity, a detailed analysis of ambient noise was carried out. Power spectral density (PSD) estimates were generated for each site, and the recorded noise from January 2024 to December 2024 was compared with globally recognised reference models. The findings indicate that the observed noise levels remain consistently within the global limits. The study also reveals that the instrumental tilt exerts a significant influence on the horizontal channels of broadband sensors. Mapping noise across multiple period bands shows the impact of different noise sources, including human activity, surface and body waves, and atmospheric pressure variations. Temporal fluctuations in noise amplitudes reveal seasonal changes in the short-period, microseism, and long-period bands. Overall, the results characterise the spatial and temporal variability of ambient noise in the area while confirming the stability and robustness of the seismic installations.

How to cite: Bala, S., Singh, C., and Singh, A.: Seismic noise environment in and around Dhubri-Chungthang Fault Zone, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9652, https://doi.org/10.5194/egusphere-egu26-9652, 2026.

The Lower Taylor Valley, owing to the McMurdo Dry Valleys, are one of the coldest and driest desert ecosystems on the planet and represent a key natural laboratory for investigating permafrost, glacier dynamics, and subsurface geochemical and geophysical processes. In recent years, a systematic and dense field measurements of soil gas concentrations and fluxes revealed an anomalous greenhouse gas concentration not randomly distributed, but forming an elongated E–W oriented zones that follow the main principal axes of the valley. To investigate the structure of the permafrost and shallower crustal layers,  as well as the preferential path controlling the multigas emissions, the geochemical surveys have been complemented, for the first time in this area, by passive seismic observations. To do that, a passive seismic experiment was carried out between December 24, 2024, and January 23, 2025, using an array of 15 three-component nodal sensors deployed in the central portion of the Lower Taylor Valley. The array geometry was designed in a spiral configuration to enhance azimuthal coverage, reaching a maximum aperture of 1.5 km. Despite the severe Antarctic environmental conditions, we successfully recorded about 25 days of continuous seismic data, constructing thus a dataset of high-quality recordings. Preliminary analyses revealed coherent seismic arrivals consistently recorded across the array. The signals are characterized by regular, high-frequency repeating peaks in the seismograms and were interpreted as icequakes originating from the Commonwealth Glacier at the northern boundary of the valley. Additionally, we also identified a variation in signal amplitude and frequency content among stations, suggesting a possible difference in local environmental conditions or subsurface properties affecting the seismic signal.

How to cite: Baccheschi, P., Hailemikael, S., Sciarra, A., Wilson, G. S., Florindo, F., Beagley, J., Mazzoli, C., Davidson, L., Berquist, C., and Ruggiero, L.: Seismological and Geochemical Monitoring of Greenhouse Gas Emissions in the Lower Taylor Valley (McMurdo Dry Valleys, Antarctica), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10323, https://doi.org/10.5194/egusphere-egu26-10323, 2026.

The 22 January 2024 Mw 7.0 Wushi earthquake (WSEQ) struck the Wushi Basin along the southern margin of the Tian Shan, Xinjiang, China. As the largest earthquake in the region since the 1992 Mw 7.3 Suusamyr earthquake (Kyrgyzstan), the WSEQ is among the few large earthquakes captured by high-quality observational datasets—including the earthquake early warning network operated by China Earthquake Administration before and a dense temporary seismic array deployed immediately after the mainshock in the Tian Shan region. This provides a rare opportunity to investigate fault network activation and rupture behavior in the tectonically active southern Tian Shan.

We picked and associated Pg/Sg phases using a neural network and the REAL method, respectively, followed by relocating all events via NonLinLoc and the double-difference relocation method. Focal mechanisms were also determined for aftershocks with M ≥4.

Our results are as follows: 1) The seismogenic fault of the WSEQ strikes northwest, correponding to a moderately-dipping, oblique-reverse main-fault rupture; 2) The mainshock triggered reactivation of a series of small-scale faults, and aftershock distribution is closely linked to these active faults in the study area; 3) Aftershocks are concentrated on subfaults rather than the main fault; and a shallowly buried subfault produced distinct surface rupture whereas the main fault is completely blind. This fault network, misaligned with the prevailing background stress field, likely forms through a reactivation of inherited weak planes.

These results illustrate that structural inheritance strongly controls fault geometric architecture and underscores the complexity of seismic activity and rupture behavior within the active fault network.

How to cite: Yin, X., Li, T., Cui, H., Chen, J., Chen, J., and Guo, B.: Interlacing ruptures of the 2024 Wushi Mw 7.0 earthquake (Chinese Tian Shan) revealed by dense seismic array observations, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10685, https://doi.org/10.5194/egusphere-egu26-10685, 2026.

Türkiye is one of the most seismically active regions due to the complex tectonic collisions relative to the Eurasian, African, and Arabian plates in the Eastern Mediterranean region. These tectonic motions have formed two major strike-slip structures: the right-lateral North Anatolian Fault Zone (NAFZ) and the left-lateral East Anatolian Fault Zone (EAFZ). The province of Sındırgı-Balıkesir is located in the southern Marmara region at central-western Türkiye. It is positioned near the southern branches of the NAFZ and the Aegean graben system, where extensional forces in Western Anatolia are observed. In this region, the active faults generally exhibit NE-SW and NW-SE orientations. On October 27, 2025, a magnitude, M_w 6.0 earthquake occurred in Sındırgı-Aktaş near Balıkesir. Understanding the source characteristics of this event is crucial for defining the local fault mechanisms and tectonics, improving seismic hazard and risk assessments, and performing more accurate ground motion analyses in this region. This study aims to resolve the source parameters and rupture characteristics of this earthquake. The seismological waveform data retrieved from the national networks. operated by Disaster and Emergency Management Authority (AFAD) and Kandilli Observatory and Earthquake Research Institute (KOERI). The earthquake source parameters were estimated using the ISOLA software package. To achieve high precision in estimating the seismic moment and focal mechanism, both point-source and potentially multiple-source inversions are performed. This comparison aims to better characterize the fault behaviour within the Sındırgı segment with respect to previously published PDE reports in the region.

How to cite: Küçük, D. and Taymaz, T.: Source Characteristics of the 27 October 2025 Mw 6.0 Sındırgı-Balıkesir Earthquake Sequence in Western Türkiye, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10736, https://doi.org/10.5194/egusphere-egu26-10736, 2026.

In its routine analysis the Federal Seismological Survey at BGR evaluates local seismic earthquakes in Germany as well as teleseismic events. For the analysis of teleseismic events we use the seismic stations in Germany and in some cases stations from surrounding countries. Because the stations are far from the events, we apply array methods (f-k analysis) to identify distinct seismic phases and locate events. We use waveforms of the densely spaced Gräfenberg array sites (GRF) with an interstation distance of 10 to 15 km or of the large aperture German Regional Seismic Network (GRSN) stations if coherent phase picking is possible.
The majority of detected and evaluated phases are first-arriving phases, such as P phases and, at more distant epicentre distances, Pdiff and PKP phases. However, in the case of stronger events, we can also detect later phases, such as PP, PS, SS, PcP, ScS, SKS and others. These secondary phases are identified well via slowness, azimuth and travel time.
Here, we show some striking examples with PKP phases as well as later phases.
Events with PKP phases are numerous and mostly located within the subduction zones at Fiji and Tonga. Due to the epicentral distances of around 145 degrees the German stations are close to the caustic of PKP branches. Therefore, PKPdf, PKPbc, and PKPab often show strong amplitudes. In some cases all three branches as well as their corresponding depth phases are visible and can be picked.
Some of the later phases, such as SKP, are rarely observed. We investigate to what extent the magnitude, the focal depth and also the focal mechanism are responsible for the amplitude of the phase and thus for their visibility at German stations. In order to determine the influence of the focal mechanism, we use the solutions of international agencies and calculate radiation coefficients for the respective phase in the direction of the stations under consideration. This approach may help to decide whether an observed depth phase is a pP or sP phase, for example, and thus enables a better depth determination.
Another special feature are very late phases, such as PKKP. They run on the long path around the Earth to the station and have an azimuth shifted by 180 degrees to the short path. These phases may be misinterpreted as independent events. Here, we show an example.

How to cite: Plenefisch, T., Donner, S., Gaebler, P., Hartmann, G., Roß, O., Stammler, K., and Steinberg, A.: Observation and analysis of core and secondary teleseismic phases at broadband stations in Germany using array methods, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10804, https://doi.org/10.5194/egusphere-egu26-10804, 2026.

The Hälsingland region in central Sweden, within the Proterozoic Fennoscandian Shield, lies near a transition from relatively shallow Moho depths (<45 km) in the west to deeper Moho depths (>50 km) in the east. The region lies close to a broad east-west transition in lower-crustal seismic velocities reported in regional models. Furthermore, it is a seismically active region, host the southernmost surface scarp of a glacially triggered fault identified in Sweden, and therefore, represents a region of elevated seismic hazard. The reason for the elevated seismicity in the region is unknown, but it has been proposed that it is hosted by the postulated Hudiksvall fault crossing the region with an NNE-SSW orientation.

In this study, we employ inversion of teleseismic receiver functions (RFs) and apparent S-wave velocity (Vs_app) for 45 available permanent and temporary seismological stations between approximately 60–63°N to map key crustal discontinuities in the region, including the boundary between upper and lower crust, the thickness of high-velocity lower crust (HVLC), and the Moho depth.

We compare the crustal model to Bouguer gravity, aeromagnetic data, and seismicity, allowing us to test candidate structures that may correlate with the occurrence of earthquakes, such as the reported NNE–SSW-trending geophysical anomaly that has been interpreted as a possible crustal boundary and the postulated NW-dipping, NNE-striking Hudiksvall fault previously inferred from 3D geophysical-geological modelling. On a regional scale, our model is consistent with previous crustal models, showing thickening of the crust to the east and south. However, the increased station density due to new data from temporary stations in the area reveals finer-scale details. Comparison with the Swedish National Seismic Network earthquake catalogue shows that clusters of seismicity in the study area tend to occur preferentially near depth changes in crustal discontinuities. Most prominently, the seismicity localizes along the greatest change in crustal thickness, as well as upper crustal thickness while the relationship with lower crustal thickness is more complex. Our preliminary analysis suggests that seismicity is focused along major changes in crustal architecture. Whether these changes in crustal structure are related to the postulated Hudiksvall Fault can currently not be determined.

How to cite: Ak, E., Schiffer, C., Lund, B., and Eggertsson, G.:  The Crustal Structure in the Hälsingland Region, Central Sweden, and Implications for Seismotectonics, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11035, https://doi.org/10.5194/egusphere-egu26-11035, 2026.

According to the China Earthquake Networks Center (CENC), an M_S6.8 earthquake occurred in Dingri County, Tibet Autonomous Region, China, on January 7, 2025, with a focal depth of 10 km. The epicenter was located in the southern part of the Tibetan Plateau. Due to the northward push of the Indian Plate, a series of north-south trending rifts have formed within the block containing the epicenter. The seismogenic fault is identified as the Dengmocuo Fault in the southern segment of the Shenzha-Dinggye Rift. Within one week after the mainshock, 55 earthquakes of M_S≥3.0 were recorded, including one aftershock of M_S≥5.0—an M_S 5.0 event on January 13th. Focal mechanism solutions from different institutions consistently indicate that the mainshock was an extensional rupture event with a nearly north-south striking plane, essentially consistent with the trend of the Shenzha-Dinggye Rift. Based on the CENC catalog, earthquakes of M_L≥3.0 within the aftershock zone are overall distributed along a north-south orientation. The epicentral distribution map shows that, bounded by latitudes 28.8°N and 28.6°N, the aftershock zone can be divided into three main areas: northern, central, and southern. The mainshock is located in the southern area. M_L≥3.0 aftershocks are primarily distributed in the northern and southern areas, with fewer and more scattered events in the central area.

We collected waveform data for earthquakes of M≥3.0 within the aftershock zone from January 7 to 14. After quality screening, we determined the focal mechanism solutions for moderate and small earthquakes based on P-wave first motions, obtaining solutions for a total of 30 events. The results show that the focal mechanisms in the northern and central areas are predominantly strike-slip, although the number of solutions from the central area is limited. The focal mechanisms in the southern area are relatively complex, mainly characterized by extension with a subordinate strike-slip component. Subsequently, we inverted the regional stress field. Given that the aftershocks are basically aligned north-south with a narrow east-west distribution, we calculated the stress field from south to north at intervals of 0.2 degrees using a radius of 20 km. The calculation results show that the orientation of the maximum principal compressive stress (σ₁) within the aftershock zone is essentially north-south, indicating that the overall rupture is dominated by east-west extension. Furthermore, the R-values from north to south are 0.8, 0.5, 0.1, and 0.2, respectively. This reveals a gradational stress pattern across the entire aftershock zone: "strong compression in the north → weak planar stress in the central area → weak compression in the south," with no abrupt changes. This suggests that the post-mainshock stress adjustment is continuous and controlled by the regional tectonic setting, with no significant stress discontinuity. A transition in the stress state from compression in the north to extension in the south is observed.

How to cite: Ma, Y. and Wang, Y.: Study on the Stress Characteristics of the Aftershock Sequence of the 2025 Dingri, Tibet MS6.8 Earthquake, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11123, https://doi.org/10.5194/egusphere-egu26-11123, 2026.

How to cite: van Zelst, I., Lythgoe, K., Gilligan, A., and Jenkins, J.:  Why is Eeyore talking about earthquakes? The fascinating seismology story behind Winnie-the-Pooh , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11569, https://doi.org/10.5194/egusphere-egu26-11569, 2026.

California, located along the transform boundary between the Pacific and North American plates, hosts a complex fault system, a long history of damaging earthquakes, and frequent small earthquakes. While earthquakes arise from the buildup and release of elastic stress, detailed knowledge of principal stress orientations, absolute stress magnitudes, and fault instability remains limited. Understanding stress distribution and its interaction with faults is key to assessing tectonic evolution and seismic hazards. Here we obtain 810,562 high-quality focal mechanisms in California combining machine-learning-based phase pickers and REFOC algorithm, representing 51.5% of all relocated earthquakes from 1981 to 2021. These mechanisms enable us to invert for 2D and 3D stress models of California’s crust, including principal stress orientations, faulting style, R-ratio, and fault instability (how close each fault’s orientation is to optimal failure) for 350 major faults. The results reveal a heterogeneous stress field: transpressional regimes dominate the northern and central San Andreas Fault (SAF) system, strike-slip regimes dominate the southern SAF system, and transtensional regimes prevail in the Walker Lane and Eastern California Shear Zone. Stress rotations over ~50 km are closely related to fault geometry, interactions, and strain partitioning. Most faults with low instability are either less optimally oriented to fail under the background stress field or located in areas with recent major ruptures. These findings underscore the tight coupling between stress and fault systems in California and the value of continuous stress monitoring and improved modeling for time-dependent seismic hazard assessments and understanding ongoing tectonic processes.

How to cite: Cheng, Y., Bürgmann, R., Taira, T., Liao, Z., and Allen, R.: Stress model of California: Fault–stress interactions across a complex plate boundary system from focal mechanisms of small earthquakes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11610, https://doi.org/10.5194/egusphere-egu26-11610, 2026.

Earthquake locations derived from seismic phase arrival times are highly dependent on the velocity model used to predict theoretical travel times. In operational monitoring, simplified 1D velocity models are commonly used to ensure rapid processing, but such approximations may introduce systematic biases in hypocentral estimations, particularly in regions characterized by strong lateral heterogeneities. This issue is especially relevant in mainland France, where complex crustal structure challenges standard localization strategies.

In this study, we focus on quantifying the impact of different velocity models on earthquake locations, with particular emphasis on the contribution of a national 3D velocity model derived from passive seismic imagery. To minimize the influence of network geometry and isolate the effect of velocity structure, we rely on a high-quality reference catalog extracted from more than 88 000 seismic events reported in the CEA catalog since 1963 (Mazet-Roux, 2025). Event selection is based on Ground Truth (GT) criteria, which assess epicentral location accuracy solely from network geometry (Bondár et al., 2001; Bondár et McLaughlin, 2009). The 22 000 events satisfying the GT5 95% criteria, associated with an epicentral accuracy better than 5 km with a 95% confidence level, define our reference event set.

We relocate this reference catalog using two advanced location algorithms capable of incorporating complex velocity structures. NonLinLoc performs a fully probabilistic search of the solution space and estimates full hypocentral uncertainty distributions (Lomax, 2008; Lomax et al., 2009), while iLoc applies a hybrid, iterative approach optimized for high-precision earthquake location (Bondár and McLaughlin, 2009; Bondár and Storchak, 2011). These methods are first compared using a common 1D velocity model (. The impact of a regional 3D velocity model, obtained through the homogenization of multiple regional models constrained by passive seismic tomography (Arroucau, 2020; Arroucau et al., 2021), is then assessed using the probabilistic NonLinLoc approach.

Regardless of the location algorithm or velocity model considered, more than 95% of the reference events exhibit epicentral differences smaller than 5 km, reflecting the robustness of the GT-based event selection. The differences mainly concern depth estimates and the associated uncertainties.

Indeed, using a common 1D velocity model, significant differences arise in depth determination and associated uncertainties. NonLinLoc systematically converges toward free-depth solutions with quantified uncertainties, whereas iLoc fixes the depth for more than 55% of the events when depth is poorly constrained.

The introduction of the 3D velocity model leads to systematic changes in hypocentral depths, with inversions yielding statistically deeper events on average and uncertainty ellipses becoming better constrained compared to 1D velocity model, despite a slight statistical increase in depth uncertainty for reference events. Differences in depth uncertainties seem to reveal regional variability and possible dependence on event depth. Comparisons with well-documented seismic sequences and previous studies are discussed to better interpret the observed differences between velocity models.

How to cite: Pantobe, L., Mazet-Roux, G., Bollinger, L., Vallage, A., Bertin, M., and Vergne, J.: Relocation of a reference seismic events catalog: influence of 1D and 3D velocity models and location methods, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11852, https://doi.org/10.5194/egusphere-egu26-11852, 2026.

We present new three-dimensional P- and S-wave velocity models of the northern Iberian Peninsula and Pyrenees using arrival times of local earthquakes and seismic ambient noise.

The arrival time dataset has been built in two steps. First, we have merged the existing seismic bulletins of permanent seismic networks in the region (IGN, ICGC, OMP). Second, we have compiled continuous waveforms of all temporary experiments in the region from 2010 to present and have been automatically picked using the deep-learning picker PhaseNet, substantially increasing data coverage in regions with sparse permanent instrumentation. Using this augmented arrival time dataset, we have inverted it simultaneously for P and S wave velocity structure and earthquake relocation. Since the region is too large for the flat-earth approximation used in the tomography code we have obtained multiple overlapping smaller models and calculated the final model by averaging the individual models.

To further enhance structural resolution, particularly in areas with limited earthquake ray coverage, we incorporated results from ambient noise tomography based on inter-station surface-wave dispersion measurements. These data provide complementary constraints on the shallow crust and improve lateral continuity of the velocity model, especially in aseismic regions and across major sedimentary basins.

The obtained P and S wave velocity models provide detailed images of the subsurface structure of the region, including parts that were poorly imaged before. The addition of arrival times picked at temporary stations has been of fundamental importance to illuminate the central part of the region, because of its low seismic activity and lack of permanent stations. Particularly well imaged are the sedimentary basins, including the southern Aquitaine basin, and the northern Ebro and Duero basins and their connection along the Rioja Trough.

How to cite: Villaseñor, A. and Chevrot, S.: Local earthquake and ambient noise tomography of the northern Iberian Peninsula and Pyrenees , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11950, https://doi.org/10.5194/egusphere-egu26-11950, 2026.

Repetitive rupture of the same asperity has been documented in various tectonic settings through the observation of highly correlated waveforms from co-located earthquakes. These events, known as repeaters, are of particular interest because they provide key constraints on small-scale fault processes and can be used to investigate slip rates on faults, aseismic deformation, earthquake nucleation... In fewer cases, highly similar earthquakes exhibiting polarity inversions at all recording stations have been identified. These so-called anti-repeaters are thought to rupture the same asperity with an inverted focal mechanism and are associated with specific driving processes. To our knowledge, no intermediate case involving rupture of the same asperity with a change in slip direction (rake) has been reported, despite the fact that such a scenario is theoretically plausible.

In this study, we search for such events within the aftershock sequence of the M7.1 Miyagi-Oki earthquake (26 May 2003), in the northeastern Japan subduction zone. This intermediate-depth intraslab sequence (70 km depth) exhibits a high seismicity rate and is well recorded by a dense seismic network. Using data from the 17 closest three-component broadband stations, we compute waveform coherence, cross-correlation and anti-correlation (flipped traces) for both P and S waves of close event pairs. We identify 40 pairs of highly similar earthquakes displaying polarity inversions at several (but not all) stations. After performing relative hypocenter relocation using correlation-derived time delays, we retain ten co-located pairs with high-quality waveform similarity and polarity inversion. By comparing measured amplitude ratios with synthetic radiation pattern ratios, we invert the rake change for each event pair.

At this stage, the physical mechanism responsible for this newly identified class of similar earthquakes, that we name ‘rake-changing repeaters’, is uncertain. A change in rake between successive ruptures of the same asperity likely reflects a highly localised modification of the stress field. It could be driven by transient pore-fluid pressure variations or stress perturbations induced with nearby moderate slip. The identification of rake-changing repeaters opens new perspectives for investigating the complexity of local faulting processes at depth, and complements existing insights from repeaters and anti-repeaters.

How to cite: Costes, L., Marsan, D., and Gardonio, B.: Rake-changing repeaters : a new class of co-localised similar earthquakes with both correlated and anti-correlated waveforms, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12485, https://doi.org/10.5194/egusphere-egu26-12485, 2026.

Full-waveform inversion (FWI) has matured over the last decade to produce high-resolution global wavespeed models down to minimum periods of 30 s. While current models agree on large scales, they exhibit structural differences at shorter wavelengths depending on the assimilated data and inversion strategy. REVEAL is a global-scale FWI model utilizing a multi-scale, whole-waveform inversion strategy. To mitigate cycle-skipping, data is incrementally bandpass-filtered and inverted from long to short periods. Currently, REVEAL inverts all phase windows in 1-hour seismograms; this conservatively preserves known body-wave structure while incorporating new surface wave structure.

However, the use of 1-hour seismograms limits the model's sensitivity to the Southern Hemisphere, where receiver density is significantly lower than in the Northern Hemisphere. To address this, we present preliminary work on a next-generation model that incorporates 3.5-hour seismograms. These longer time series capture both minor- and major-arc surface wave arrivals. Including major-arc waves (epicentral distances > 180°) significantly increases sensitivity to structure in the Southern Hemisphere. Starting from the current REVEAL model, we revise the long-period structure (60–200 s) using a multi-scale approach, beginning with 100–200 s and terminating at 60–200 s. We simulate full viscoelastic wavefields using the spectral-element solver Salvus on global fourth-order cubed-sphere meshes. Adjoint sources are calculated using a time-frequency phase misfit, and optimization is performed via mini-batch stochastic trust-region L-BFGS.

How to cite: Schiller, C. J., Keating, S., Afanasiev, M., Marty, P., and Fichtner, A.: Global full-waveform inversion using a surface wave orbital, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12709, https://doi.org/10.5194/egusphere-egu26-12709, 2026.

The global coda-correlation wavefield is a mathematical representation of the seismic wavefield generated by large earthquakes, computed by cross-correlating the long-lasting coda waves recorded by a worldwide network of seismometers. Unlike the common assumptions that the cross-correlation functions are Green’s functions between station pairs, earthquake coda-correlation features arise from cross-terms of in reverberating body waves of common slowness that share common slowness characteristics a subset of propagation legs (e.g., Hrvoje & Phạm, 2018; Phạm et al., 2018). This wavefield has offered new constrains on the deep Earth's interior (e.g., Hrvoje & Phạm, 2018; Ma & Tkalčić, 2024). However, quantitative interpretation of these features for tomographic imaging has been hindered by the lack of distributed sensitivity kernels relating observed correlation signals to Earth structure heterogeneities.

In this study, we develop a forward modeling framework for the earthquake coda-correlation wavefield bypassing conventional cross-correlation of late-coda waveforms. The forward-modeled correlograms reproduce features obtained through conventional stacking with a significantly improved signal-to-noise ratio. Building on this framework, we derive finite-frequency traveltime banana-doughnut sensitivity kernels using adjoint methods (e.g., Tromp et al., 2010; Fichtner, 2014), which quantify how traveltime measurements of correlation features depend on perturbations in P-wave velocity, S-wave velocity, and density.

We compute sensitivity kernels for prominent correlation features (P*, ScS*, and others) at station separations ranging from 30° to 180°. The kernels reveal extensive sampling of Earth's mantle, outer core, and inner core, with spatial-sensitivity patterns fundamentally different from those of body waves in direct seismic wavefield. For antipodal station pairs, correlation features exhibit strong sensitivity throughout the deep interior, including regions poorly sampled by traditional seismic phases. Our results confirm that correlation features form through constructive interference of body-wave pairs with similar slowness, rather than representing Green's functions between station pairs.

This work establishes the theoretical foundation for coda-correlation tomography, enabling future three-dimensional imaging of Earth’s internal structure with unprecedented sampling of the deep interior. The sensitivity kernels provide a pathway to exploit the wealth of information contained in earthquake coda for high-resolution mantle and core tomography.

Reference
Ma, X. & Tkalčić, H. (2024) Low seismic velocity torus in the Earth's outer core: Evidence from the coda correlation wavefield, Sci. Adv., 10, 35, https://doi.org/10.1126/sciadv.adn55.

Fichtner, A. (2014). Source and processing effects on noise correlations. Geophys. J. Int., 197(3), 1527-1531. https://doi.org/10.1093/gji/ggu093 

Phạm, T-S., Tkalčić, H., Sambridge, M. & Kennett, B.L.N. (2018) The Earth's correlation wavefield: late coda correlation, Geophys. Res. Lett., 45, https://doi.org/10.1002/2018GL077244.

Tkalčić, H., & Phạm, T.-S. (2018). Shear properties of Earth's inner core constrained by a detection of J waves in global correlation wavefield. Science, 362, 329. https://doi.org/10.1126/science.aau7649

Tromp, J., Luo, Y., Hanasoge, S., & Peter, D. (2010). Noise cross-correlation sensitivity kernels. Geophys. J. Int.,183(2), 791-819. https://doi.org/10.1111/j.1365-246X.2010.04721.x 

How to cite: Wei, Z., Wang, S., Phạm, T.-S., and Tkalčić, H.: Sensitivity Kernels for Earthquake Coda-Correlation Wavefield Features and Applications to Global Seismology , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13130, https://doi.org/10.5194/egusphere-egu26-13130, 2026.

Two earthquakes with Mw 6.1 occurred in Sındırgı–Balıkesir, western Türkiye, within a few months (10 August 2025 and 27 October 2025). This short time separation provides a valuable opportunity to compare spatiotemporal changes in post-event seismicity and faulting parameters under broadly similar regional conditions, while also allowing potential differences in detection capability and catalogue completeness between the two periods to be handled explicitly.

In this study, a homogenised regional earthquake catalogue is analyzed within an explicit magnitude-of-completeness (Mc) framework to minimize bias arising from temporal variations in detectability when comparing the two post-event periods. Completeness is assessed in a time-dependent manner, and all subsequent analyses are conditioned on the estimated Mc to ensure a like-for-like comparison between the August and October sequences. The aftershock sequences are examined in both space and time, spatiotemporal clustering and migration patterns are evaluated from the mapped aftershock distributions, while temporal decay is characterized using the Omori–Utsu law by estimating and comparing sequence-specific decay parameters. In parallel, b-values are estimated from the Gutenberg–Richter frequency–magnitude distribution and their spatial variations are evaluated to identify potential heterogeneity in the post-event stress state and rupture environment across the source region.

To link seismicity patterns with faulting behaviour, focal-mechanism solutions are used to determine the dominant faulting styles, the principal P- and T-axis orientations, and the regional stress regime associated with each sequence. The joint interpretation of aftershock decay, b-value variability, and mechanism-derived faulting characteristics provides an integrated, catalogue-consistent comparison of post-event behaviour for the August and October 2025 earthquakes, and yields a consolidated description of how seismicity and faulting evolve in the Sındırgı region of western Türkiye following closely spaced moderate–large events.

How to cite: Tamtaş, B. D.: Seismicity characteristics of Sındırgı-Balıkesir in western Türkiye in the context of the 10 August 2025 (Mw=6.1) and 27 October 2025 (Mw=6.1) earthquakes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13613, https://doi.org/10.5194/egusphere-egu26-13613, 2026.

High-resolution imaging of crustal structure is critical for understanding deformation processes, lithospheric rheology, and seismic hazard in continental collision zones. Afghanistan and its surrounding regions lie at the junction of the Indian, Eurasian, and Arabian plates, hosting diverse tectonic environments that include major strike-slip fault systems, thick foreland and intracontinental basins, hot orogenic cores, and ongoing continental subduction beneath the Hindu Kush. Despite frequent damaging earthquakes, existing seismic velocity models are limited by sparse station coverage and the sensitivity of conventional methods to long wavelengths or near-vertical ray paths, resulting in poor resolution at seismogenic depths. Here, we address these limitations using earthquake inter-event interferometry, which exploits dense inter-source ray coverage to enhance lateral resolution in regions with uneven seismic networks and provides new constraints on crustal structure across Afghanistan and adjacent areas.

We analyzed a high-quality regional dataset of earthquakes recorded between 2006 and 2019 across Afghanistan, eastern Iran, Pakistan, Tajikistan, Turkmenistan, and Uzbekistan. Vertical-component seismograms were processed to retrieve empirical Green’s functions from both earthquake–station surface waves and inter-event correlations of Rayleigh-wave coda. Inter-event interferometry was applied under stationary-phase and minimum-separation conditions, and all paths were stacked using phase-weighted stacking to enhance signal coherence. Rayleigh-wave group velocities were measured with a multiple-filter technique and inverted for period-dependent two-dimensional group-velocity maps using fast-marching surface-wave tomography, with regularization optimized through L-curve analysis and resolution assessed by checkerboard tests. Local dispersion curves were then inverted for one-dimensional shear-wave velocity profiles and assembled into a quasi-three-dimensional Vs model extending to ~60 km depth.

The resulting shear-wave velocity model reveals pronounced lateral and vertical heterogeneity that closely tracks major tectonic provinces. Foreland and intracontinental basins appear as shallow low-velocity domains reflecting thick, weakly consolidated sediments, with contrasting depth evolution between rapidly strengthening foreland crust and basins that retain mid-crustal weakening near major fault systems. The Pamir and western Himalayan collision zone is characterized by a strong upper crust overlying laterally extensive middle- to lower-crustal low velocities, interpreted as thermally and fluid-weakened thickened crust associated with shortening, partial melting, and ductile flow. Farther east, the Hindu Kush exhibits narrow, localized low-velocity anomalies vertically linked to intense intermediate-depth seismicity, consistent with fluid-controlled weakening above the north-dipping Indian lithosphere. To the south, the Makran subduction zone forms the most vertically continuous low-velocity system, extending from the shallow accretionary prism to Moho-proximal depths and reflecting thick sediment accumulation, underplating, hydration, and high pore-fluid pressures above the subducting Arabian plate, bounded by a cold, mechanically strong backstop at the Sistan–Makran transition.

This study presents the first regional application of earthquake inter-event interferometry in Afghanistan, providing new constraints on crustal rheology, fault and suture architecture, and an improved seismic velocity framework for geodynamic analysis, earthquake location, and hazard assessment in a complex continental collision zone.

How to cite: Kazemnia Kakhki, M., Bokelmann, G., Shirzad, T., Sadidkhouy, A., and Esteve, C.: High-Resolution Crustal Imaging of Afghanistan from Earthquake Inter-Event Interferometry, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13717, https://doi.org/10.5194/egusphere-egu26-13717, 2026.

Jointly with the 33rd expedition of Bulgarian Antarctica institution, the National Center of Meteorology (United Arab Emirates) installed its first broadband seismic station in Livingstone Island in Antarctica. This deployment aims to detect earthquakes, ice sheet dynamics, and volcanic activities. In addition to the exciting seismic stations on the island, the seismic installation includes two advanced seismometers, Trillium 120 and Trillium 40 (Nanometric), to accurately detect local, regional and tele seismic events.

The seismic station will be added value to the existing networks on the island as well as covering the gap analysis for more accurate data collection. In addition, since the station located near to active volcanic eruption (Deception Island), It will provide crucial data to interpret the interactions between seismic, cryosphere changes, and volcanic activities.

The project highlights continuous and reliable seismic recording in harsh polar environments. During the Bulgarian Expedition 34 (2025–2026), the data were collected and analyzed, demonstrating that there were no gaps or interruptions throughout the entire year of 2025.

The station is expected to become a vital tool for future studies, potentially integrating high-resolution seismic monitoring with glaciological and meteorological data collection. This comprehensive approach will enhance our understanding of the dynamic relationships among tectonic, cryosphere, and volcanic processes.

How to cite: Alameri, B., Alebri, K., Hamidatou, M., Alkaabi, A., and Megahed, A.: Continuous Seismic Observation under Extreme Environmental Conditions , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14941, https://doi.org/10.5194/egusphere-egu26-14941, 2026.

Seismological signatures of aseismic slip include tectonic tremors, repeating earthquakes, and earthquake swarms. They are most commonly documented in subduction margins, where their depth distribution and temporal behavior reflect slab geometry, fluid release, and plate interface rheology. Their occurrence in continental collision belts, however, remains unclear. Taiwan provides a unique natural laboratory to address this gap. Here we investigate how earthquake swarms interact with repeating earthquakes (REs) and tectonic tremors, and how their coupling varies with depth along the west-dipping Central Range Fault.
Using a 24-year seismic catalog, we show that tremors (mostly >25 km), REs (15–25 km), and swarms (<20 km) align spatially along the same fault system but exhibit distinct recurrence behaviors and interaction thresholds. We find that temporal interaction among these phenomena is strongly depth-dependent and controlled by spatial overlap. Swarms and REs display the strongest coupling within shared fault segments, consistent with asperities being repeatedly loaded by surrounding aseismic creep and transient slip-rate accelerations, supported by the rapid diffusivity of catalogued swarms (several m²/s). In contrast, tremors occur in the lower crust under near-critical frictional conditions and generally interact with seismogenic activity only during periods of elevated aseismic slip, most notably following M6-class earthquakes. Tremors alone exhibit strong tidal modulation (~90% during increasing tidal levels), indicating markedly higher stress sensitivity at depth. Elevated Vp, Vs, and Vp/Vs ratios further delineate a high-stiffness, high–pore-fluid-pressure lower-crustal environment that enables tremor generation beneath the mountain root. Overall, these observations indicate that swarms, REs, and tremors represent depth-stratified manifestations of aseismic slip, with interaction style and stress sensitivity systematically varying with depth.

How to cite: Peng, W. and Chen, K. H.: Depth-dependent interaction of tectonic tremors, repeating earthquakes, and earthquake swarms in a collision zone of Taiwan, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15736, https://doi.org/10.5194/egusphere-egu26-15736, 2026.

The Miryang Fault is an inferred fault located in the southeastern Korean Peninsula and is a part of the Yangsan Fault System, which trends NNE-SSW. Although the Miryang Fault appears as a distinct topographic lineament, its activity and characteristics remain poorly constrained due to a scarcity of direct geological and seismological evidence. We investigated seismic activity along the Miryang Fault to delineate subsurface fault structures using earthquake detection, high-precision hypocenter relocation, and focal mechanism analysis. We employed energy-ratio-based automatic detection and determined relative hypocenter locations using the double-difference method. Focal mechanism solutions were derived using P-wave first-motion polarities. Epicenters are concentrated to the west of the surface trace of the Miryang Fault and generally align with its strike. The hypocenters exhibit a strongly linear spatial distribution, distinctively separated into northern and southern clusters. The seismicity tends to become more spatially scattered northward, particularly beyond the northern termination of the surface trace of the Miryang Fault. The southern cluster is characterized by a higher frequency of larger earthquakes and a lower Gutenberg–Richter b-value compared to the northern cluster. Focal depths range from 5 to 20 km, with the southern region showing a narrower range and a concentration at greater depths. Principal Component Analysis of the hypocentral distribution reveals that the fault geometries for both clusters trend NNE-SSW and dip nearly vertically. Both clusters extend approximately 30 km along the strike. Focal mechanisms indicate predominantly dextral strike-slip motion, with strike and dip consistent with the geometry inferred from the spatial distribution. These findings provide new insights into the seismotectonic characteristics of the Miryang Fault and underscore its potential role in the active tectonics of the southeastern Korean Peninsula.

How to cite: Heo, D., Han, J., Kang, T.-S., Kim, S., and Rhie, J.: Subsurface geometry of the Miryang Fault, southeastern Korean Peninsula, inferred from high-precision earthquake relocation and focal mechanism analysis, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16016, https://doi.org/10.5194/egusphere-egu26-16016, 2026.

Dynamic processes operating at continental plate margins, including associated igneous activity, lead to progressive reactivation and modification of the lithosphere and are recorded in the evolving structure of the continental crust. Crustal evolution is strongly controlled by crust–mantle interaction at the Moho, where changes in temperature, composition, and density can be induced by underplating or lithospheric destruction. Consequently, the Moho represents a critical interface at which continental growth, mechanical weakening, and lithospheric removal may occur. Seismic imaging of the uppermost mantle therefore provides direct constraints on the processes governing lithospheric evolution. In this study, we present three-dimensional P- and S-wave tomography of the uppermost mantle beneath the southern Korean Peninsula, derived from Moho-refracted waves. The tomographic models are used to estimate thermal and compositional variations by comparing observed seismic velocities with petrological predictions. A basalt–harzburgite mechanical mixture (MM) was adopted as the baseline model, while an equilibrium assemblage (EA) and an orthopyroxene (Opx)-enriched mantle were additionally considered to evaluate the effects of compositional variability on seismic velocities.

The results showed pronounced regional contrasts in seismic velocity, temperature, and composition. The Gyeonggi Massif and Gyeongsang Basin are characterized by relatively low Vp (≈7.65 km/s) and low Vp/Vs (≈1.73), consistent with a hot, harzburgite-rich mantle and a thin lithosphere. In contrast, the Yeongnam Massif exhibits high Vp (≈7.89 km/s) and high Vp/Vs (≈1.76), indicating a relatively cold mantle with a higher basaltic fraction (40%) within a thick lithosphere. These characteristics are consistent with independent constraints on lithospheric thickness. The hot and thinned harzburgite-dominated mantle is interpreted as the result of lithospheric modification by delamination or thermal erosion. In contrast, basaltic components preserved within the cold and thick lithosphere are interpreted as basaltic underplating associated with Mesozoic subduction, maintained by a stabilized lithospheric root. However, the combination of low Vp (≈7.65 km/s) and exceptionally low Vp/Vs (≈1.70) observed in the western Gyeonggi Massif cannot be explained by either the MM or EA models, whose minimum Vp/Vs values are 1.74 and 1.73, respectively. This discrepancy requires anomalous mantle conditions, such as strong Opx enrichment or pronounced seismic anisotropy. Despite their close spatial proximity, these contrasting mantle properties indicate that distinct geological processes operated beneath different regions of the Korean Peninsula during the Mesozoic–Cenozoic, and that the resulting thermal and compositional signatures are preserved in the uppermost mantle.

How to cite: Kim, M., Song, J.-H., and Kim, S.: Crust–Mantle Evolution beneath the Korean Peninsula Constrained by Temperature and Composition Estimated from Uppermost Mantle Tomography, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16313, https://doi.org/10.5194/egusphere-egu26-16313, 2026.

The mantle wedge seismicity of the Atacama segment of the Northern Chilean subduction, a region with complex slab geometry, has recently been mapped in a high-density seismicity catalog. In order to better constrain the underlying mechanisms responsible for its occurrence, we investigate the spatio-temporal evolution of the seismicity along the plate interface and within the overriding (upper) plate. We find that the b-value of the mantle wedge seismicity is consistently greater than 1, averaging around 1.5, and that for earthquakes of similar magnitude, there are few, if any, aftershocks in the mantle wedge compared to the interface seismicity. We also estimate the seismic wave velocities Vp and Vs according to the amount of serpentinization and compare to the results of recent high quality tomography images. We estimate that the region of active seismicity, located mainly between 450-550°C isotherms, corresponds to a partially serpentinized part of the mantle wedge while the corner of the cold nose, which corresponds to a fully serpentinized zone, is deprived from earthquakes.

How to cite: Gardonio, B., Münchmeyer, J., Brunet, F., Hernández-Soto, N., Auzende, A.-L., and Socquet, A.: The Chilean Mantle Wedge seismicity as a marker for serpentinization., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17464, https://doi.org/10.5194/egusphere-egu26-17464, 2026.

Surface wave tomography yields high resolution models for the seismic velocities
in the earth’s upper mantle. To extend the imaging into the mantle transition
zone and into the lower mantle below, additional constraints from teleseismic
body waves, which probe these deeper regions, are needed. Here, we combine a
large new dataset of teleseismic P and SV travel times obtained via an automatic
picking algorithm (Stampa et al., 2024), with waveform fits from Automated
Multimode Inversion (AMI; e.g. Dou et al. (2024)) in a joint inversion. The
sensitivity volumes for the teleseismic body waves used in the inversion are
approximated using ray tracing in a spherically symmetric background model.
For the P and SV arrival-time data sets, travel-time corrections are calcu-
lated to account for the differences induced by the ellipticity of the earth. The
frequency-dependent effects of anelasticity on the propagation of the teleseismic
waves are estimated and accounted for, using the measured central frequencies
of the waves’ arrivals. Travel time anomalies resulting from crustal structure
are estimated and accounted for using the ECM1 crustal model.
The preliminary results show a marked improvement in the resolution of
structure at depths below 400 km, in particular in regions of dense station
coverage, including Europe, North America, and the subduction zone regions
around the Pacific Ocean.

References

Dou, H., Xu, Y., Lebedev, S., de Melo, B. C., van der Hilst, R. D., Wang, B., &
Wang, W., 2024. The upper mantle beneath Asia from seismic tomography,
with inferences for the mechanisms of tectonics, seismicity, and magmatism,
Earth-Science Reviews, 255, 104841.

Stampa, J., Eckel, F., Keers, H., Lebedev, S., Meier, T., & AlpArray and
SWATH-D Working Groups, 2024. Automated measurement of teleseismic
P-, SH-and SV-wave arrival times using autoregressive prediction and the
instantaneous phase of multicomponent waveforms, Geophysical Journal In-
ternational , 239(2), 936–949

How to cite: Stampa, J., Lebedev, S., Meier, T., Keers, H., and Henriksen Krogh, J.: Joint inversion of a new global data set of teleseismic P and SV arrival times and the waveforms of surface and regional body waves, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18156, https://doi.org/10.5194/egusphere-egu26-18156, 2026.

Earthquake rupture forecasting is a critical component of seismic hazard analysis and requires identifying physically plausible rupture paths across complex fault networks. At its core, this task could be formulated as a large-scale combinatorial optimization problem, which involves selecting an optimal subset of fault segments from a set of candidates. Such problems pose significant challenges for traditional algorithms, as the number of admissible multi-fault ruptures grows combinatorially. Current operational workflows rely on locally applied plausibility filters. While computationally efficient, such greedy local heuristics risk excluding globally competitive rupture scenarios, particularly in regions with dense fault connectivity and competing rupture pathways. 

This work investigates whether quantum annealing hardware can serve as a sampling accelerator for exploring the ensemble of physically plausible ruptures beyond what is currently accessible to classical approaches. Designing a problem formulation that remains scientifically meaningful while respecting the constraints of current quantum hardware (size, noise, etc.) is nontrivial, and naïve encodings often collapse under these limitations. We encode rupture plausibility modeling as low energy solutions to a quadratic unconstrained binary optimization (QUBO) problem defined on a fault-network middle graph. Binary variables represent activated fault segments, while local interaction terms encode Coulomb stress transfer (as a proxy for pairwise rupture compatibility), continuity preferences, and branching behavior. Here, we adopt a classical-quantum hybrid workflow: the quantum annealer is used to sample globally competitive rupture candidates, while classical post-processing implements physical constraints to filter out non-physical ruptures. 

The workflow is demonstrated on a subset of the fault network used in the 2023 USGS National Seismic Hazard Model including over 200 fault segments, using both simulated thermal and quantum annealing on D-Wave hardware. Results show that our hardware-aware formulation and conditioning enable robust sampling on comparatively large fault-network instances. Views on quantum computing are polarized: some overstate its power, while others dismiss it as impractical. Our results help bridge this by establishing a concrete, testable pathway for integrating quantum annealing into rupture-modeling workflows on existing purpose-built quantum devices.

—---

This work is supported by the ICSC National Research Centre for High Performance Computing, Big Data and Quantum Computing (CN00000013, CUP D53C22001300005) within the European Union-NextGenerationEU program (National Recovery and Resilience Plan (PNRR) - Mission 4 Component 2 Investment 1.4.)

How to cite: Stallone, A., Dukalski, M., Diaferia, G., and Milner, K.: Using Quantum Annealing for Identifying Plausible Earthquake Rupture Paths in Fault Networks, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18185, https://doi.org/10.5194/egusphere-egu26-18185, 2026.

We succeeded to run Obspy—the popular seismic Python toolkit—entirely in the browser using WebAssembly and a novel, package distribution called
emscripten-forge. This breakthrough eliminates the need for local installations or complex backends for small to medium sized data.

Now, anyone can perform core seismic data analysis—including reading, processing, and visualizing seismological data—directly in their web browser,
on many devices, from desktops to mobile phones. For developers, Obspy's capacities can be embedded in ordinary static web sites that are highly
scalable, without any complex backend to maintain. Additionally, Large Language Models (LLMs) can command Obspy directly in the browser, allowing to build seismic applications driven by natural language.

The implications are profound. The lack of complex installations, backend maintenance, or specialized deployments, drastically simplifies using and
building seismic data analysis tools. Educators can incorporate hands-on seismology into curricula without technical overhead. Researchers can share
apps or notebooks, and explore data without fighting non-reproducible compute environments. Engineers in remote settings gain access to powerful
analytical tools on mobile devices. Users can interact with seismic data in the browser in natural language, lowering the barrier for non-experts and
professionals alike.

We will demonstrate this with JupyterLite, showcasing a fully functional, Jupyter environment with Obspy and scientific Python tools running completely in the browser, without any server and with LLM assistance. We will also highlight the limitations of in-browser execution, particularly regarding multi-threading, memory constraints and compute overhead. All of the demonstrated technology is open source, published under permissive licenses, and therefore available for anyone to use, modify, and build upon.

How to cite: Meschede, M., Corlay, S., and Beier, T.: Obspy + WebAssembly: Running AI-Assisted Seismic Analysis in the Browser, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18563, https://doi.org/10.5194/egusphere-egu26-18563, 2026.

The Core-Mantle Boundary (CMB) is a critical interface influencing
Earth’s thermal and chemical evolution. Among the various features
in the CMB region, the D” seismic reflector—located approximately
300 km above the CMB—is of particular interest. Mineral physics
links this reflector, which generates distinct P- and S-wave reflec-
tions and increases S-wave velocity by 2–3 % and P-wave velocity by
0.5–2 %, to a phase transition from bridgmanite to post-perovskite.
If these reflections stem from the phase transition, variations in re-
flector depth directly reflect lateral changes in temperature and rock
composition near the CMB. In this study, we modelled phase transi-
tions across a range of temperature–composition conditions to pro-
duce synthetic seismograms, demonstrating how these factors influ-
ence velocity jumps, reflector depth, and reflectivity. We also com-
piled data from over 65 previous studies measuring the D” reflector’s
depth, addressing the lack of a coherent global map since the identifi-
cation of post-perovskite in 2004. Our compilation shows that shal-
lower reflector depths are found in seismically fast (cold) regions,
while deeper reflectors are found in seismically slow (hot) regions,
consistent with theoretical predictions of post-perovskite occurrence.
This global map establishes a consistent framework for comparing
phase transition models with seismic observations.

How to cite: Pahlings, J., Houser, C., Hernlund, J., and Thomas, C.: Mapping the D” Seismic Reflector: Insights from Phase Transition Modelling and Global Seismic Observations, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18989, https://doi.org/10.5194/egusphere-egu26-18989, 2026.

Reliable, stable, and internally consistent moment magnitude (M_w) estimates are essential for seismic hazard assessment and earthquake source characterization. Full waveform inversion provides physically robust seismic moment estimates; however, its applicability is often constrained by data quality, network geometry, and modeling assumptions. Conversely, coda-based approaches are a more efficient alternative that are less sensitive to radiation pattern and path effects. This study presents a systematic comparison between coda-derived moment magnitudes (M_w-coda) estimated using the Qopen framework and those (M_w-ISOLA) obtained from moment tensor inversion conducted through the ISOLA software.

Analysis is based on the same dataset examined by Özkan et al. (2024), comprising 303 local earthquakes (2.5 ≤ M_L ≤ 5.7) recorded by 49 broadband stations in the Sea of Marmara region, NW Türkiye. Qopen is utilized to jointly invert S-wave and coda-wave envelopes by employing Radiative Transfer Theory. This process yields frequency-dependent attenuation parameters and coda-derived source spectra, from which the M_w-coda is estimated. For a subset of these events with sufficient waveform quality and azimuthal coverage, independent seismic moment estimates are obtained through time-domain waveform inversion using ISOLA.

Overall, we aim to evaluate the agreement and consistency between M_w-coda and M_w-ISOLA in terms of magnitude scaling, scatter, and potential systematic bias, with particular emphasis on magnitude range, source depth, and signal-to-noise conditions in a tectonically complex region characterized by strong lateral heterogeneity and frequency-dependent attenuation.

How to cite: Özkan, B., Eken, T., Gaebler, P., Yolsal-Çevikbilen, S., and Taymaz, T.: Comparisons of Source Parameter Estimates Based on Moment Tensor and Coda Analysis Methods in the Sea of Marmara, NW Türkiye, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19204, https://doi.org/10.5194/egusphere-egu26-19204, 2026.

The b-value of the Gutenberg-Richter law represents a fundamental parameter for characterizing earthquake size distributions and assessing crustal stress conditions.

Numerous studies have demonstrated a significant relationship between the b-value and the tectonic stress regime. While low b-values are often associated with stress accumulation and potential precursors to large earthquakes, their applicability as universal precursors remains debated.

This study investigates the spatial and temporal evolution of the b-value preceding large earthquakes (M≥5.5) in regions characterized by contrasting tectonic settings and fault kinematics. We examine geographically and tectonically different areas—such us Italy, China, New Zealand, and Myanmar—which encompass different stress regimes and plate boundary configurations. By systematically analyzing local seismic catalogs from these contrasting regions, we assess whether b-value variations constitute a universal feature of the seismic cycle, or are primarily modulated by region-specific crustal properties.

Each study area is selected according to the seismogenic structures responsible for the target earthquakes. The completeness magnitude (Mc), defined as the lowest magnitude threshold above which the catalog reliably records all or nearly all earthquakes in the region, is rigorously determined as a function of time according to well-established methodologies.

We estimate Mc using the most appropriate method for each region, including either the maximum curvature method  or the 90% goodness-of-fit criteria, to ensure robust results.

The b-value is subsequently estimated using the maximum likelihood approach over appropriate spatiotemporal windows preceding each mainshock.

We present preliminary results showing that systematic b-value drops are observed in most cases. Our findings support the hypothesis that such variations represent a robust indicator of progressive stress accumulation. This comparative approach suggests that b-value monitoring can provide valuable precursory signals for seismic hazard assessment across different tectonic contexts.

How to cite: Orlando, M., De Caro, M., and Montuori, C.: Spatial and temporal evolution of the b-value: A comparative analysis across different tectonic settings, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19351, https://doi.org/10.5194/egusphere-egu26-19351, 2026.

We present the analysis of the first known video of co-seismic slip on a natural fault. It was captured during the 2025 Mw 7.7 Mandalay earthquake (Myanmar) by a CCTV camera located a few meters away from the fault. By direct image analysis of the footage, we measure the slip and slip-rate functions from the natural coseismic rupture. The results show that the rupture propagated as a slip-pulse, with a local slip duration of 1.4 s and a maximum slip rate of 3.5 m/s ± 20%. We then fit two steady-state slip-pulse models to the measured slip-rate function, allowing us to estimate the mechanical properties of the fault: the slip-stress curve, the slip-weakening distance, the breakdown work and the energy release rate. This study shows the value of direct on-fault slip measurements for constraining those mechanical parameters, that are key inputs in dynamic rupture models.

How to cite: Latour, S., Lebihain, M., Bhat, H. S., Twardzik, C., Bletery, Q., Hudnut, K. W., and Passelègue, F.: Direct estimation of earthquake source properties from a single CCTVcamera (2025 Mw 7.7 Mandalay earthquake, Myanmar), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21737, https://doi.org/10.5194/egusphere-egu26-21737, 2026.

In late 2023, the Reykjanes Peninsula in Iceland experienced an intense seismic and volcanic episode. A major earthquake swarm began on 24 October 2023, driven by magmatic intrusion beneath the region, and its frequency and intensity escalated dramatically on 10 November. This activity culminated in the Sundhnúksgígar crater chain eruption on 18 December. During the November swarm, the high density of tremors caused significant challenges for the automatic earthquake location system, reducing its reliability. To address this issue, we applied the extended Integrated Particle Filter (IPFx) method to continuous seismic data recorded during the eruption period.

The IPFx method, originally developed for Japan’s Earthquake Early Warning (EEW) system, integrates single-station P-wave detection with a network-based particle filter approach to estimate event locations and magnitudes in real time. It processes continuous waveform data from multiple stations, enabling rapid and accurate earthquake detection even during intense seismic sequences. We analyzed three days of continuous data (9–11 November 2023) and compared IPFx-derived locations with the manually reviewed catalog of the Icelandic Meteorological Office (IMO).

Initial application of the IPFx method using its default configuration—Japanese velocity structure and no historical seismicity—resulted in large offshore location uncertainties due to limited azimuthal coverage near the eruption site. To improve accuracy, we incorporated the South Iceland Lowland (SIL) velocity model used in Iceland and regional historical seismicity into the particle filter’s sampling and likelihood functions. These modifications reduced average location errors by approximately 50%. Furthermore, the IPFx method successfully distinguished multiple closely spaced events during periods of high seismicity, demonstrating its potential for generating reliable automatic earthquake catalogs under challenging conditions.

Our findings highlight the adaptability of the IPFx method for real-time seismic monitoring in volcanic regions with sparse station coverage. By improving earthquake location accuracy during swarm activity, this approach can enhance early warning capabilities and contribute to hazard mitigation efforts in Iceland and similar tectonic settings.

How to cite: Yamada, M., Jónsdóttir, K., and Erlendsson, P.: Enhancing Earthquake Detection During the 2023 Reykjanes Swarm Using the Extended IPF Method, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-466, https://doi.org/10.5194/egusphere-egu26-466, 2026.

Earthquake localization using Distributed Acoustic Sensing (DAS) is challenging due to the single-component directional sensitivity of DAS systems. We propose a novel approach for localization that is based on dense DAS recordings and constraints from known structural heterogeneity of the subsurface. Our method employs full-waveform simulations to generate synthetic DAS wavefield images for a range of potential earthquake source locations. A deep convolutional neural network (CNN), based on a U-Net architecture, is trained on these images to map DAS-recorded wavefield patterns to earthquake source coordinates, without the need for identification and picking of P- and S-wave arrivals. We evaluate this wavefield-based localization technique using a challenging synthetic case study involving DAS recordings in a single vertical borehole, representative of monitoring configurations commonly deployed at geothermal platforms. We consider different velocity models of varying geological complexity. The results show that the CNN effectively learns location-specific wavefield signatures influenced by subsurface heterogeneity. Uncertainties can be reduced significantly by adding recordings from a second borehole. While the results are based on idealized 2D synthetic modeling, the method offers a promising approach for improving microseismic monitoring when detailed information on the heterogeneous velocity structure is available (such as that derived from seismic surveys).

How to cite: Komeazi, A., rümpker, G., and Limberger, F.: Exploring wavefield-based location imaging in heterogeneous media: a borehole DAS example, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5526, https://doi.org/10.5194/egusphere-egu26-5526, 2026.

Accurate knowledge of existing fiber-optic cable geometry is essential for applications of distributed acoustic sensing (DAS), however the true positions of buried or installed fibers are often uncertain due to slack, bends, or deviations from documented routes. We present two passive, seismology-based approaches for cable localization that exploit information contained in DAS recordings. Case A employs ambient noise cross-correlations with reference points to estimate relative travel times, whereas Case B uses the differential arrivals of plane waves from distant earthquakes with linearly independent slowness vectors. Both approaches can be formulated in a least-squares framework that allows for the joint estimation of propagation velocity and geometry, thereby reducing biases from noise and model assumptions. Synthetic experiments show that cable positions can be recovered with an accuracy better than 100 m, even when apparent velocities are uncertain or the medium exhibits heterogeneity. The two methods provide independent geometric constraints that complement other sources of information on cable routing.

How to cite: Rümpker, G., Limberger, F., and Komeazi, A.: Passive seismological approaches for localizing near-surface fiber-optic cables with DAS, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5599, https://doi.org/10.5194/egusphere-egu26-5599, 2026.

The seismic moment tensor (MT) delivers valuable information about the physical source process of a seismic event. It allows to distinguish between earthquakes, explosions and volcanic events, constraints the orientation of an earthquake rupture, and represents the most accurate estimate of the released seismic energy.

The computation of absolute MTs by waveform inversion is a data intensive task that is oftentimes feasible only for the largest events in a data set. Relative MTs rely on less subsurface information and may be computed for a large number of closely spaced weaker seismic events that are connected to an absolute MT through relative amplitude measurements. The relative MT method assumes that the Green’s functions between events is similar and that relative amplitudes are measured below the corner frequency of the largest event. Under these assumptions, the relative amplitude between seismograms can be attributed to the difference in moment tensor between events. Compared to absolute methods, path and site effects cancel out and do not need to be considered.

We here present relMT, an accessible, research-grade, open-source software package that facilitates computation of relative moment tensors for a large variety of data sets. The software takes as inputs seismic waveform, event locations, ray take-off angles, and a reference MT, as well as waveform headers and a configuration file. In synopsis, the similar waveforms are aligned to sub-sample accuracy under consideration of possible polarity reversals for P-waves and planar polarization of S-waves. Amplitude ratios between the aligned seismograms are measured on single seismic stations in a principal component framework. The relative amplitudes are combined mathematically with ray take-off angles, relative event distances and one absolute reference moment tensor in a linear system of equations. The solution of the equation system with algebraic methods yields all relative moment tensors at once. The uncertainty of the solutions is quantified using the bootstrap method. The software is under active development on GitHub (https://github.com/wasjabloch/relmt).

We illustrate the application of relMT using data sets of induced seismicity and tectonic aftershock seismicity. For the induced seismicity of the enhanced geothermal system in Helsinki, Finland, we are able to lower the magnitude threshold for which MTs can be computed from 0.5 to -0.5. For aftershocks in the Pamir highlands of Central Asia from 4.0 to 2.0.  For the data sets, this represents a 3- to 30-fold increase in the number of recovered MTs.

How to cite: Bloch, W., Oye, V., Drolet, D., Plourde, A., and Bostock, M.: relMT – Software to Determine Relative Seismic Moment Tensors Illustrated with Tectonic and Induced Seismicity, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7124, https://doi.org/10.5194/egusphere-egu26-7124, 2026.

Recent advances in seismology, driven by the deployment of dense seismic networks and the development of machine-learning-based earthquake detection, have enabled the generation of high-quality seismic catalogs with unprecedented spatial and temporal resolution. These dense microseismic datasets provide a robust foundation for detailed waveform-based analyses, that allow individual earthquakes to be reliably linked to fault segments and enable constraints on fault geometry and slip style at fine spatial scales.

We present an integrated seismological workflow that starts from automated earthquake detection using machine-learning techniques (PhaseNet) applied to continuous seismic recordings in the Val d’Agri region (Southern Italy). The resulting high-resolution microseismic catalog is then analyzed through waveform similarity-based clustering to identify events associated with the same seismogenic structures, followed by high-precision relative relocation to delineate fault segments, and Bayesian moment tensor inversion to robustly characterize faulting style.

This waveform-based workflow enables the association of earthquakes with individual seismogenic structures, allowing to resolve fault geometries and slip styles at fine spatial scales. Results indicate that seismicity predominantly clusters on steeply southwest-dipping normal faults, with focal mechanisms consistent with the regional extensional stress regime.

These analyses illustrate how machine-learning-driven seismic monitoring combined with waveform-based analysis can bridge the gap between large microseismic datasets and fault-scale imaging. Beyond increasing detection rates, this workflow provides new insights into the geometry and kinematics of active faults in regions affected by diffuse seismicity.

How to cite: Caredda, E., Cesca, S., and Morelli, A.: From Machine-Learning detection to fault imaging: high-resolution seismology in the Val d'Agri (Southern Italy), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7959, https://doi.org/10.5194/egusphere-egu26-7959, 2026.

Differential travel-time observations from waveform cross-correlation are among the most precise measurements in observational seismology and underpin high-resolution relative relocation. In large modern catalogs, however, HypoDD-style pair files (dt.cc) often contain outliers, cycle skipping, and internally inconsistent links that can bias downstream workflows, particularly cluster-based relocation pipelines. We present DDSync, a lightweight preprocessing method that treats the differential-time measurements for each station–phase as a weighted event graph and solves a graph synchronization problem to recover a self-consistent set of relative arrival-time proxies (one scalar per event and station–phase). The baseline estimator is a sparse weighted least-squares solution of a graph Laplacian system, which implicitly averages over redundant constraints in dense graphs and yields stable long-baseline differentials without enumerating paths.

DDSync adds two robustness layers. First, it computes an edgewise inconsistency diagnostic from the global fit (a loop-closure-style residual) and prunes grossly inconsistent links using a MAD-based threshold, with automatic re-identification of connected components. Second, it refines the solution using iteratively reweighted least squares, with a Huber loss to downweight remaining heavy tails while preserving connectivity. Beyond producing cleaned and synchronized dt.cc files, DDSync estimates uncertainty in the synchronized results by approximating per-event variance of the inferred potentials using a stochastic diagonal estimator of the inverse reduced Laplacian, and propagating these to conservative pairwise σ estimates and weights for downstream inversions.

We evaluate DDSync on the Ridgecrest synthetic benchmark of Yu et al. (2025), where differential times are perturbed with Laplacian-distributed errors and outliers added, and show near complete removal of gross outliers and strong tightening of residual distributions relative to ground truth, reducing error by about a factor of 5. The inferred uncertainty on the denoised observations capture station–phase and event-specific constraint quality and provide a practical, uncertainty-aware weighting scheme for relocation and related inverse problems. We also highlight a subglacial volcanic example with distinct event family types (icequake and VT events) where pruning preferentially removes inconsistent links that bridge otherwise separated event families, improving interpretability and robustness for analysis of dense catalogs.

How to cite: Heimisson, E. R., Winder, T., and Yu, Y.: DDSync: Denoising and outlier removal in differential travel-time observations using graph synchronization, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8169, https://doi.org/10.5194/egusphere-egu26-8169, 2026.

Local magnitude (M_L) is widely used for reporting the size of small earthquakes, but to achieve consistent physical scaling and cross-regional comparability, moment magnitude (M_W) and related source parameters such as corner frequency (f_c) are required. Routine M_W estimation for small events is often hampered by low signal-to-noise ratios and unstable spectral fitting in conventional frequency-domain approaches. Recent time-domain approaches have therefore estimated M_W from peak S-wave displacement amplitudes measured in multiple narrow-band filters, but commonly rely on frequency and distance-dependent empirical attenuation curves that are inherently region specific. We present a generalized time-domain method that estimates moment magnitudes from peak S-wave displacement amplitudes measured on vertical-component seismograms filtered into ten fixed narrow bands with center frequencies spanning 0.1-30 Hz.

The method applies one-way Butterworth bandpass filters with constant bandwidth (0.2 Hz) and three poles, selected to provide stable spectral equivalence across center frequencies. For each band, the peak S-wave displacement is converted to an equivalent displacement spectral amplitude using a constant factor calibrated against Fourier displacement spectra. Source and path effects are corrected using geometrical spreading and intrinsic anelastic attenuation. Multi-band amplitudes are interpreted with a Brune source model (Brune, 1970) using a nested grid search to retrieve the long-period level (Ω₀) and corner frequency (f_c). M_W is then calculated from the resulting seismic moment.

We validated the approach using 12,025 records from 490 earthquakes in and around the southern Korean Peninsula (2017–2022). The time-domain M_W agrees closely with reference M_W from displacement-spectral fitting with uncertainty assessment (R² = 0.97). To validate the generalization of our method, we applied it to a two-week subset of the 2019 Ridgecrest sequence (4–18 July 2019), comprising 115,309 seismograms from 5,073 earthquakes. The resulting magnitudes also showed strong agreement with Trugman (2020) (R² = 0.91) without a region-specific correction curve.

The proposed method is directly compatible with real-time workflows, and we are integrating it into an Earthworm-based pipeline to output source parameter estimates shortly after S-wave arrival. To support this implementation, we developed modules for real-time IIR filtering and moment-magnitude estimation by adapting and extending Earthworm modules. This provides an efficient and practical route to real-time M_W estimation in operational settings.

How to cite: Hong, Y., Doo, M., and Sheen, D.-H.: A generalized time-domain approach for routine moment magnitude estimation from S-wave peak displacement amplitudes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8875, https://doi.org/10.5194/egusphere-egu26-8875, 2026.

Deep learning methods have gained significant attention in seismic denoising due to their superior ability to extract weak signals from prestack data, which are often smoothed out by traditional techniques. While conventional convolutional neural networks can be trained in an end-to-end manner, they often fail to capture the underlying data distribution. Generative models are capable of reconstructing more realistic seismic signals by learning the distribution. The primary generative models include Variational Auto-Encoders (VAEs), Generative Adversarial Networks (GANs), and the more recently proposed Denoising Diffusion Probabilistic Models (DDPMs). VAEs offer stable training but tend to yield results of limited quality. GANs can produce high-quality outputs via an adversarial discriminator but suffer from unstable training. DDPM could provide a favorable balance between output quality and training stability. However, the supervised training paradigm relies on high-quality labeled data, which is often scarce in geophysical applications. This limitation frequently leads to constrained generalization ability. Consequently, there is significant signal leakage for strong reflection events when applied to unseen data from different work areas.

To address this, we propose a novel discriminator-constrained diffusion model. Our key innovation is the integration of a discriminator into the DDPM framework. This adversarial component provides a powerful constraint during the training process. The hybrid training objective combines the standard diffusion loss with the adversarial loss, guiding the model to preserve critical reflections while removing noise.

We validate our method through comprehensive experiments. On synthetic data containing various noise intensities. Our method has an improvement of 0.7 dB for noisy data with 1% Gaussian noise compared to standard DDPM. More importantly, in cross-field tests, the proposed method has an improvement of 2 dB. Visualizations of denoised sections and difference profiles confirm that our approach better preserves reflections.

In conclusion, the incorporation of adversarial training into the diffusion process offers a robust solution to the generalization challenge in deep learning-based seismic denoising. Our work demonstrates a promising pathway for applying advanced generative models to practical geophysical data with limited labels.

How to cite: Zhang, Y.: Discriminator-Augmented Denoising Diffusion Probabilistic Models for Seismic Data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10528, https://doi.org/10.5194/egusphere-egu26-10528, 2026.

We present new local magnitude models for Switzerland, derived using a top-quality input dataset of ground-motion intensity measures, obtained through an automated and consistent waveform processing workflow (https://doi.org/10.1785/0120250032). The input dataset includes Wood-Anderson displacement response amplitudes calculated from 150,000 waveforms generated by about 15,000 earthquakes in the region of interest. This dataset is substantially larger than those used in previous magnitude studies in Switzerland and contains a large number of recordings at short distances for smaller events (about 5,000 earthquakes with magnitudes lower than 1, and about 1500 records at hypocentral distances shorter than 5 km), which were sparse in earlier studies. The parametrization of the local magnitude model is based on a detailed investigation of attenuation characteristics of the Wood-Anderson response amplitudes, comprising: (i) linear and logarithmic distance terms to represent different physical mechanisms of seismic wave propagation; (ii) hinge distances to allow modelling the effects of Moho reflections and the associated changes in attenuation rate; (iii) regional adjustments (Swiss Alps vs Swiss northern Foreland) based on the length and location of the surface projection of the source-to-site ray paths. Model coefficients are determined through mixed-effects regressions, thus allowing the derivation of station-magnitude correction terms consistent with the reference rock-like ground type used for mapping seismic hazard in Switzerland. We assess and validate the model’s performance via uncertainty analyses, validation against recent data not used in model calibration, validation against recordings of major Swiss events, and comparisons with results obtained from alternative, fully independent modelling approaches (including non-parametric and 2D cell-based methods). The new model yields systematically lower magnitudes, with an average difference of 0.1-0.2 magnitude units for smaller events with magnitudes below 2.5 and for earthquakes located in the Swiss northern Foreland, compared to the currently authoritative catalogue magnitudes. Based on the new candidate magnitude model, we present and discuss an updated empirical scaling relationship between local and moment magnitudes.

How to cite: Altindal, A., Cauzzi, C., Bindi, D., Diehl, T., Deichmann, N., Clinton, J., and Wiemer, S.: Towards New Magnitude Models for Switzerland, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10958, https://doi.org/10.5194/egusphere-egu26-10958, 2026.

The transition from manual to deep-learning automated seismic phase picking has revolutionized seismology applications such as seismic catalog building and fault structures analysis. However, the reliability of these AI-driven catalogs is often hindered by a "black-box" approach to model selection and decision thresholds. While deep learning models like PhaseNet (Zhu and Beroza, 2019) offer unprecedented efficiency, their performance is sensitive to the data they were trained on and to the probability thresholds used to define a "phase pick". 

In this work, we present a comparative study focused on the Amatrice–Visso–Norcia 2016-2017 seismic sequence in Central Italy. We investigate the influence of the training models and the threshold variation on the phase picking detections.  

In particular, we compare the performance of the default PhaseNet model (STEAD) against i) a model trained on the AQ2009 dataset (specific for the Central Apennines, from Bagagli et al., 2023) and ii) a model obtained through Transfer Learning on STEAD fine-tuned with the INSTANCE model (Michelini et al., 2021) via the SeisBench platform (Woollam et al., 2022). We also analyze how varying the confidence threshold (from 0.1 to 0.9) affects the final catalog's completeness and precision.

Preliminary results show that regional training significantly outperforms default models in specific noise conditions and that the optimal threshold is influenced by station geometry and signal-to-noise ratios. By providing a statistical framework for automated threshold calibration, this study offers a roadmap for more objective and reproducible signal detection, applicable not only to seismology but to any domain dealing with continuous time-series classification.

How to cite: Fonzetti, R., Bailo, D., Valoroso, L., De Gori, P., and Chiarabba, C.: The Impact of Probability Thresholds and Model Training on Phase Picking Performance, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11512, https://doi.org/10.5194/egusphere-egu26-11512, 2026.

A high-resolution earthquake catalog and detailed quantification of earthquake source parameters are essential for constraining fault structure and earthquake interactions. As a candidate site for the next-generation gravitational wave detector (i.e., Einstein Telescope), the Lower-Rhine Embayment region requires a comprehensive assessment of the fault distribution inferred through seismicity and earthquake source properties. In this study, we build an enhanced earthquake catalog using both permanent seismic stations and new data from temporary deployments together with AI-based techniques, including signal enhancement with a decoder-autoencoder denoiser, seismic phase detection using PhaseNet, and event association with PyOcto. We further refine earthquake locations with the NLL-SSST-coherence algorithm and then apply an automatic quality-control filter using event association to remove false detections resulting from the misinterpretation of teleseismic signals as local ones due to event denoising. We detect 3900 events for the period 2019 to 2025, with 2101 of them being classified as earthquakes. The enhanced catalog shows increased hypocentral depths along the NW-trending Sandgewand fault, with a maximum depth of 20 km at the southern end of the fault system. 

We also present the first results of a source-parameter catalog for earthquakes that occurred between 2000 and 2025 based on the dataset of Hinzen et al. (2021). Focal mechanisms for selected earthquakes in the region are determined with SKHASH by combining S/P ratios and first-motion polarities obtained from PhaseNet+. We fit individual earthquake spectral parameters, including corner frequency and seismic moment, and calculate stress-drop values for M_L≥1.5 events based on a circular crack model. Preliminary results indicate a median stress-drop value of 2.5 MPa across the region, with slightly higher stress-drop values observed on the Sandgewand fault relative to the Rurrand fault. In addition, we use the Distributed Acoustic Sensing (DAS) recordings to compute focal mechanism and corner frequency estimates and compare the results with broadband seismic stations for two earthquakes captured by DAS observations in the Netherlands in 2025. The enhanced spatial sampling density of DAS data provides additional constraints on earthquake source parameters and enables fault movement estimation that is difficult with seismic station data alone.

How to cite: Chen, X., Carrasco, S., Roth, M. P., Bocchini, G. M., and Harrington, R. M.: An enhanced catalog and earthquake source parameter estimations in the Lower-Rhine Embayment region, western Central Europe, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11625, https://doi.org/10.5194/egusphere-egu26-11625, 2026.

Distributed Acoustic Sensing (DAS) enables continuous assessment of transport infrastructure conditions through passive surface wave analysis. However, operational deployment may be problematic as the dispersion curves from thousands of frequency-velocity dispersion images are generated during routine monitoring. Passive DAS recordings from train-induced vibrations exhibit severe fragmentation, strong higher-mode interference, and temporal variability driven by seasonal moisture changes that consistently challenge manual or semi-automated picking methods. 

We present a training-free hybrid algorithm that combines marker-controlled watershed segmentation with physics-informed trajectory optimization to extract fundamental mode of dispersion curves from complex operational DAS data. Applied to a 15-month monitoring campaign on a 350 m railway embankment in the UK, the methodology operates directly on dispersion images exported from standard MASW processing software. The algorithm proceeds through five stages: (1) binary masking with morphological noise suppression establishes candidate energy regions; (2) watershed transformation with internal markers separates touching fragments that should constitute independent segments; (3) bidirectional amplitude-maximum propagation with adaptive vertical search radii extracts local trajectory estimates within each isolated fragment; (4) velocity band filtering combined with forward monotonic chaining reassembles disconnected segments by enforcing kinematic consistency and rejecting physically-implausible connections; and (5) global sigmoid fitting with constrained horizontal extension produces smooth, inversion-ready dispersion curves validated against aliasing boundaries. 

Validation against manual picking demonstrates that the algorithm bridges spectral gaps exceeding 5 Hz, correctly isolates fundamental from higher modes even when energy amplitudes are comparable, and maintains trajectory continuity through severe fragmentation where conventional peak-following methods fail. Beyond immediate operational utility, automated extraction from real-world DAS railway data enables generating computationally labeled training datasets that preserve physical consistency and interpretability. We demonstrate how this computer vision approach produces high-quality dispersion curve labels across diverse geological settings and complexity levels. 

How to cite: Shrivastava, S., Trafford, A., Saqlain, M., and Donohue, S.: Automated Extraction of Rayleigh Wave Dispersion Curves from DAS Railway Monitoring, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13989, https://doi.org/10.5194/egusphere-egu26-13989, 2026.

The open-source, seismological software SeisComP is widely used for earthquake monitoring world-wide.  Its default automatic phase association and location module is scautoloc, which was originally developed to monitor large earthquakes with global or large-regional networks.  It detects and locates earthquakes iteratively, beginning with P picks from only the nearest stations and improving the earthquake location as additional picks arrive. This process may take more than 15 minutes and the availability of early intermediate results is therefore essential for time-critical applications like tsunami early warning.

Requirements for local earthquake monitoring are quite different.  The number of stations is usually smaller and the seismic wave travel times are typically below one minute. Small networks greatly benefit from the use of S picks and custom velocity models to optimize their locations, neither of which are currently supported by scautoloc.

In an effort to improve local and regional monitoring capabilities in SeisComP using only open-source software, we adapted the phase associator PyOcto [1, 2] to the SeisComP framework.  PyOcto was specifically designed for fast processing of large amounts of regional network picks. This is particularly important because of the vast improvements in the amount of picks obtained using machine learning techniques [3, 4].  Its efficient implementation and low computational overhead also make PyOcto a perfect phase associator for real-time earthquake monitoring.

The new SeisComP module scoctoloc [5] leverages the use of PyOcto to improve processing of local and regional network data in SeisComP. The ability to process both P and S picks and the convenient support for either homogeneous (0D) or custom layered (1D) velocity models overcome the main limitations of scautoloc. Small networks may choose to run scoctoloc as a drop-in replacement of scautoloc, though both modules may also be run in parallel.

The factors limiting the real-time processing speed are the network dimension (and hence seismic travel times) and pick latency. The latter depends on the latency of the data telemetry and on the processing delay imposed by the picking technique used.  In our real-time test setup, a virtual seismic network in Northern Chile, earthquake locations are usually produced within three minutes after an event using P and S picks produced by the scdlpicker module [4]. Where processing speed is more crucial than optimum pick accuracy, the standard SeisComP scautopick with S picking enabled is still the picker of choice.

The use case that PyOcto was developed for originally, the bulk processing of huge amounts of picks read from a database, is supported by scoctoloc as well.  In addition it is possible to run "pick playbacks" from a SeisComP database in order to simulate real-time operation. This mode is useful in order to fine-tune the configuration parameters relevant for real-time monitoring based on past events.

[1] Münchmeyer, J. (2024). PyOcto: A high-throughput seismic phase associator. Seismica. doi:10.26443/seismica.v3i1.1130.
[2] https://github.com/yetinam/pyocto
[3] Münchmeyer et al. (2022), https://doi.org/10.1029/2021JB023499
[4] https://github.com/SeisComP/scdlpicker
[5] https://github.com/jsaul/scoctoloc

How to cite: Saul, J., Münchmeyer, J., and Tilmann, F.: scoctoloc - A regional phase associator and locator module for SeisComP, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14439, https://doi.org/10.5194/egusphere-egu26-14439, 2026.

We present an updated and extended seismotectonic earthquake catalog of Switzerland and surrounding regions. This SECOS25 catalog serves as input for the seismic-source component of Switzerland’s next generation seismic hazard model to be developed by the Swiss Seismological Service (SED) over the coming years. The SECOS25 catalog includes 51 years of instrumental seismicity detected and located by the SED between 1975 and 2025. For the digital era of the SED catalog (phase picks and seismograms available in digital form) starting in 1984, hypocenters were consistently relocated in absolute terms using a recent 3D P and S-wave crustal velocity model. Starting from these improved hypocenters, double-difference relative relocations were performed at different scales (local clusters as well as regional scales), combining differential times from manual picks and waveform cross correlations. A merging procedure was developed that selects the preferred location method (bulletin location, absolute relocation, relative relocation) based on location-quality criteria for each hypocenter. The proposed procedure provides various hypocenter uncertainty measures and ensures the maximum possible hypocenter-location accuracy and precision for each event. Local magnitudes were revised using a new set of consistent amplitude measurements (doi.org/10.1785/0120250032) and a new magnitude model. For each earthquake, we provide moment magnitudes from native methods if available (either from moment tensors or spectral fitting) or revised scaling relationships. Finally, we link the hypocenters with catalogs of moment tensors (containing about 80 solutions) as well as first-motion focal mechanisms (containing about 600 solutions).

The SECOS25 catalog serves as a base for further down-stream seismotectonic components of the seismic-source model. This includes heat maps of seismic activity and moment release across Switzerland as well as maps of deformation regimes and stress orientations derived from the analysis and inversion of focal mechanisms. Finally, we apply an enhanced version of the HyFi method (doi.org/10.1029/2023JB026352) to the SECOS25 catalog to systematically identify previously unknown seismically active fault segments and their orientations. The derived information will contribute to a refined definition of seismic zonation as well as faulting styles and preferred rupture orientations required for hazard computations. The SECOS25 catalog and derived products therefore also contribute to an improved understanding of present-day seismotectonic processes in the Central Alpine region.

How to cite: Diehl, T., Heilig, J., Altindal, A., Truttmann, S., Cauzzi, C., Deichmann, N., Clinton, J., Herwegh, M., and Wiemer, S.: Towards an Improved Seismic Source Catalog for Switzerland, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14799, https://doi.org/10.5194/egusphere-egu26-14799, 2026.

The accumulation of decades of continuous seismic observations, combined with the emergence of new sensing technologies (e.g., distributed acoustic sensing-DAS- and nodes) and novel computing infrastructure, presents both outstanding challenges and opportunities for the observational geophysical community to tackle data processing at petabytes scale. Methodologies, open-source software practices, and cyberinfrastructure have advanced to a point where mining petabyte-scale archives can be done within a single day of cloud computation (Ni et al, 2025a,b). This contribution reviews research workflows centered around seismic event monitoring with large-scale seismometers and regional DAS networks. We evaluate strategies for both cloud infrastructure (Ni et al, 2025) and edge DAS units (Shi et al, 2025a,b) for a seismic event monitoring pipeline that leverages deep learning for rapid feature extraction, such as classification of seismic source type and picking of P- and S-wave arrivals. Leveraging and advocating for open-source software, we discuss computational considerations and strategies to improve the performance of pre-trained deep learning models through transfer-learning and model architecture adaptation. We illustrate these findings with the United States NSF-National Geophysics Facility archive operated by the EarthScope Consortium, as well as diverse experiments from the University of Washington FiberLab.

How to cite: Denolle, M., Ni, Y., Shi, Q., Rose, A., and Lipovsky, B.: PetaScale Data-Driven Seismology: Geohazard Discovery via Large Array Data Mining at the Edge, on Premise, and on the Cloud, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15608, https://doi.org/10.5194/egusphere-egu26-15608, 2026.

Probabilistic seismic hazard assessment (PSHA) is challenging in cratonic regions such as Sweden, where the characteristics of strong motion (detrimental for structures) are uncertain due to a lack of data. The current approach is therefore to use available data sets of small-to-moderate earthquakes to identify seismogenic areas and to adapt models from more seismically active regions. One issue encountered in this process is estimating the moment magnitude (Mw) of these earthquakes. In fact, evaluation of local magnitude (Ml) is preferred for magnitude <4 earthquakes due to the difficulty of deriving Mw using standard methods.

In this work, we propose to estimate the Mw of earthquakes in Sweden using a generalised inversion technique (GIT), and to investigate the uncertainty of this measurement in this seismic context. To do so, a ground-motion data set of Ml=0-4.1 earthquakes recorded by the Swedish National Seismological Network (SNSN) is compiled, and the non-parametric generalised inversion of Oth et al. (2011) is applied. This approach identifies the source spectrum of earthquakes, the apparent attenuation of the region and the site amplification of stations by performing a spectral decomposition of the Fourier amplitude spectrum (FAS). The parameters of the Brune (1970) source model and the frequency-dependent elastic and anelastic attenuation models are derived from these terms in post-inversion.

We present here the resulting source parameters (Mw and corner frequency fc) that are analysed and compared with the moment based Ml computed by the SNSN. The consistency of these estimates is then evaluated using higher-sample-rate recordings from a temporary network that was deployed for four years in Sweden’s most seismic active region. The 100 Hz sampling rate of the permanent stations limits the estimation of Mw and fc for earthquakes of magnitude <2, as their fc is expected to exceed the Nyquist frequency (~45 Hz). The 200 Hz sample rate set for the temporary network enables the source spectrum to be derived up to 90 Hz and therefore allows the veracity of the Brune (1970) model (which has been derived from a shorter frequency band) to be analysed.

Applying this inversion algorithm also provides new insights into anelastic attenuation in cratonic regions and into site amplification observed at hard rock/bedrock sites (which constitute the majority of SNSN installations). Knowing the apparent attenuation and site conditions is useful for future studies, especially for correcting the FAS at stations to measure Mw using other approaches or models, and for achieving a quasi real time estimate.

How to cite: Buscetti, M. and Lund, B.: Estimation of the moment magnitude and its uncertainty for small-to-moderate earthquakes in Sweden using a generalised inversion approach., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17337, https://doi.org/10.5194/egusphere-egu26-17337, 2026.

Estimating the stress drop (Δσ) occurring during an earthquake allows us to characterize the mechanical state of the area surrounding its source. The magnitude (Mw) and the source radius (r) of an earthquake are usually assumed to scale such that Δσ is constant. We typically refer to this assumption as self-similarity of earthquakes.

However, estimates of Δσ can vary by at least four orders of magnitude (e.g., Cocco et al., 2016, DOI: 10.1007/s10950-016-9594-4). It is an open question to what extent this variability is caused by methodological and data uncertainties, or it is a manifestation of different physical processes (Abercrombie et al., 2025, DOI: 10.1785/0120240158). This is especially true for smaller events, due to a strong correlation between source and propagation terms in the waveform modeling.

In this study, we obtain precise relative source parameters (seismic moment, corner frequency and stress drop) estimates analyzing the spectral ratios of co-located events with similar source mechanisms. This allows us to get rid of the propagation term in the waveform modeling, and to focus on the effects of the assumed source model on the stress drop.

We analyze pairs of events recorded during 2017 in North Ibaraki region (Japan) by a high-resolution temporary seismic network operated by AIST and by Hi-net stations, and during Pawnee and Prague earthquakes in Oklahoma (2011 and 2016, respectively). Overall, we explore a range of magnitudes from M = 0.7 to M = 5.8.

We find that  source model and self-similarity are not always compatible and that in general a strong correlation exists between stress drop estimates and the source parameter γ that describes the decay of source spectrum at high frequencies (γ = 2 in the  model). This emphasizes the importance of considering γ as a free parameter when modeling earthquake source spectra.

Moreover, our approach allows us to investigate potential differences among catalog and moment magnitudes through the inferred relative seismic moment estimates.  For the small events in Ibaraki, we find differences larger than 0.5 units between local magnitude (Ml) and moment magnitude, supporting previous evidence of non-linear relationships between Ml and Mw (e.g., Uchide and Imanishi, 2018, DOI:10.1002/2017JB014697). This highlights the importance of expanding Mw catalogs to smaller magnitudes, as statistical analysis of Ml catalogs may be affected by systematic biases.

This study was supported by TRANSFORM², funded by the European Commission under project number 101188365 within the HORIZON-INFRA-2024-DEV-01-01 call.

How to cite: Maille-Okada, N., Supino, M., Satriano, C., Uchide, T., and Marzocchi, W.: Relative stress drop and source parameters estimation from spectral ratio method, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18401, https://doi.org/10.5194/egusphere-egu26-18401, 2026.

Automatic monitoring of local seismicity produces events of varying quality. Some events will be poorly located, and some event solutions will not represent a real seismic event, arising only due to noise. Noise events are removed and location quality improved during manual revision, but that is not always feasible for large catalogs. In cases with >10000 events manual review is not tenable, so an automatic quality score calculation is beneficial in improving catalog quality.

Machine learning methods are a useful tool for this purpose, both for classification and calculating a quality score. We explore machine learning methods for solutions to this problem, with a special focus on feature extraction from travel-time information.

The models are evaluated on data from three seismic networks in Iceland. The dataset contains both automatic and manual solutions for a large number of earthquakes, so direct comparisons between manual and automatic solutions can be made. The manual locations can then be used as a ground truth solution that the automatic solutions attempt to approximate. The models are ranked on their classification score as well as their ability to estimate the spatial distance from the ground truth solution.

How to cite: Magnússon, R. L.: Event classification and quality assessment for local seismic events using machine learning, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19111, https://doi.org/10.5194/egusphere-egu26-19111, 2026.

Seismological studies are traditionally based on the observation of ground motions recorded by translation sensors. However, to assess a comprehensive description of any wavefield produced by seismic sources, the three components of rotation are as important as the three translation components. Due to improvements in instrumentation through the last two decades, the ground rotational motion is now an observable. Several recent publications show that 6 degrees of freedom (-dof) seismic station, recording the 3 translation and 3 rotation components of the ground motion, provides valuable information to locate events or to infer their source mechanism.

Rotation motions can be recorded directly via dedicated sensors, or numerically derived via a dense array of seismometers. The formers primarily use technologies based on fiber optic gyroscopes, liquid-based systems, or mechanical principles. However, these broadband instruments do not have the high sensitivity required to detect "weak" ground movements. Conventional sensor arrays, on the other hand, use finite differences techniques to estimate reliable indirect rotation measurements, but these are limited at high frequencies (wavelength > 4 × array aperture).

Since the end of 2025, a temporary experimental campaign is set at the Low Noise Underground Laboratory (LSBB) in Rustrel, France. Two rotational rate sensors, namely a BlueSeis 3A and a Eentec R3, are deployed in the underground galleries, jointly with five seismometers which complete the permanent seismic array. The dense seismic array co-located around the rotation sensors is used to compute array derived rotations (ADR), and validate the direct observations of rotational ground motions. Furthermore, six seismometers are deployed at the surface to form an array with an aperture of around fifty kilometers. It has been designed for long-term observations of regional and global seismicity and array processing analysis, such as beamforming techniques. This campaign allows to compare the performance of classical seismic array processing with innovative gradiometric approaches based on a single 6-dof station, focusing on detection, location, and characterization of seismic events within a common frequency band. A sensitivity study on rotation signals, in terms of instrumental conditions (array geometry, station quality) and processing parameters (signal duration, filtering), is carried out with the help of numerical full waveform modelling in order to quantify the uncertainty of the estimated source parameters. We present the benefit of such multi-component seismic wavefield recording, illustrated on several events of interest such as the recent (2025/07/29) Kamchatka event (Mw 8.8).

How to cite: Morin, M., Sèbe, O., Beucler, É., Capdeville, Y., Rouille, G., Boyer, D., Decitre, J.-B., Bremaud, V., Lallemand, C., Lepoint, F., and Govindin, G.: From Seismic Arrays to Single-Station Wavefield Gradiometry: Results from a Multi-Scale Experimental Campaign, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19637, https://doi.org/10.5194/egusphere-egu26-19637, 2026.

In regions with limited tectonic information, dense deployments of local seismic networks significantly improve the detection of low-magnitude events. The consequent enrichment of seismic catalogs is proved to have a crucial role for understanding seismotectonic processes and assessing seismic hazards. This study evaluates the performance of machine learning algorithms (MLA) for P- and S-wave picking and event association, using data recorded between April 2013 and June 2025 by the OTRIONS seismic network (FDSN code OT), operating in the Gargano Promontory (GP, Southern Italy) since 2013.

The MLA workflow consists of PhaseNet, a deep learning-based phase picker, in combination with GaMMA, an association algorithm, were employed and approximately 27,000 seismic events were detected. NonLinLoc was employed for event locations. The visual inspection confirmed that about 51% of the events were local earthquakes, while the remainder events were classified as quarry blasts, false events, or events located outside the network. The visual revision procedure was essential at this step.

Compared to the previous manual catalog (based on the STA/LTA detection algorithm) in the same area, the MLA workflow brougth to a new enriched automatic catalog. The quality assessment of the new catalog indicates that the automatic picking is reliable and confirms the OT network’s ability to detect a high rate of low-magnitude seismicity. The NonLinLoc-SSST-Coherence algorithm was also applied to better identify the structures on which seismicity is accomodated and the results suggest that NonLinLoc-SSST-Coherence has better permormances when applied to small seismic sequences than to the widespread seismicity of GP. 

From a seismotectonic perspective, the already known seismogenic layer deepening northeasternward characterizing the GP seismicity here appears for the first time to be splitted in two structures located at different depth. This study highlights the crucial role of dense local networks and MLA tools in managing and analyzing large volumes of low-energy seismic data.

How to cite: Ferreri, A. P., Panebiano, S., Satriano, C., Filippucci, M., Cecere, G., Serlenga, V., Stabile, T. A., Selvaggi, G., and Tallarico, A.: A machine-learning workflow for event detection and relocation in the Gargano Promontory (Southern Italy), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21190, https://doi.org/10.5194/egusphere-egu26-21190, 2026.

Moment tensor (MT) determination is a key component of real-time seismology, with applications in moment magnitude estimation, tsunami early warning, volcano monitoring and shake map generation. Despite its importance, the reliable inversion of MT components, especially the non-double-couple ones, presents significant challenges. Indeed, the non–DC components are highly sensitive and often exhibit large fluctuations, making reliable estimation of the full moment tensor difficult. These limitations highlight the need for robust uncertainty quantification of MTs.  To efficiently address this issue, we propose a Bayesian bootstrapping approach. The approach assumes that Signal to Noise Ratio (SNR) is fair and the velocity model is not systematically biased. The method relies on a series of weighted inversions, in which station contributions are stochastically varied using Bayesian weights. This procedure produces an ensemble of plausible MT solutions enabling statistical characterization of the inversion results (e.g., median MT, confidence intervals of ISO and CLVD components, etc.). This approach, free of the assumption of Gaussianity of data error, provides meaningful uncertainty estimates and improves the interpretability of non–double-couple components. The proposed methodology has been integrated into GISOLA, an open-source, highly efficient near–real-time MT inversion software, currently in routine operation in several seismic networks. The resulting automated operational framework handles multiple data streams in heterogeneous formats, interfaces with diverse processing modules, applies a systematic preprocessing workflow to identify the most reliable stations and corresponding signals, performs parallelized inversions, and provides robust uncertainty quantification. This enhances the reliability of source characterization in operational environments and supports more informed use of MT results in time-critical seismic monitoring.

This work is supported by TRANSFORM²  which is funded by the European Union under project number 101188365 within the HORIZON-INFRA-2024-DEV-01-01 call

How to cite: Velentzas, V., Zahradník, J., Psarakis, E., Evangelidis, C., and Sokos, E.: Bayesian bootstrapping extension of GISOLA automatic moment tensor inversion software, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21600, https://doi.org/10.5194/egusphere-egu26-21600, 2026.

In this study we introduce PHASER, a seismic event picker developed specifically for downhole microseismic applications. In recent years, DL models have been extensively explored for automating phase-picking and/or event detection for seismic data, with most applications focusing on teleseismic/regional earthquake signals from surface arrays. Surface arrays require a generalizable solution across different array geometries. In contrast, downhole arrays used to monitor industrial activities such as geothermal, CCS and hydraulic fracturing are more standardized in receiver placement, but present unique challenges and opportunities. Seismic phases arrive coherently on closely-spaced downhole geophones. However, access to downhole data is also often limited, and such data are often available only as event-based traces, and labels are often incomplete or inaccurate. To address these constraints, PHASER is trained in a multi-stage framework that remains effective and generalizable even when catalogue labels are incomplete and limited. PHASER incorporates association filtering into its training; pick probabilities are matched with their respective association probabilities based on learned source-related feature embeddings for each P- and S- arrival. By using a learned extraction threshold, PHASER avoids the manual parameter tuning typically required for pick extraction. PHASER demonstrates better continuous monitoring performance on unseen sites than existing DL phase-pickers, achieving a 6-fold performance over PhaseNet in F1 score from 0.107 to 0.584 on an out of sample test dataset.

How to cite: Leung, K., Verdon, J., and Werner, M.: PHASER: A Deep Learning Model for Real-time Downhole Microseismic Event Picking, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3133, https://doi.org/10.5194/egusphere-egu26-3133, 2026.

Traditional volcano monitoring relies on dense ground-based instrumentation (e.g., seismic, gravity, and deformation measurements), which is available for only a small fraction of the world’s volcanoes. While satellite observations can complement these measurements, estimating total erupted mass—a primary metric of eruption magnitude—remains largely a post-eruption task. Consequently, erupted mass has been quantified for only ~100 of the 1,282 known volcanoes.

These limitations motivate the use of machine learning (ML) in volcanic prediction. Unlike traditional approaches, ML can integrate diverse, globally available datasets to generate predictive insights for volcanoes with limited or no time-dependent monitoring. This capability enables proactive risk assessment and hazard planning, offering a scalable, cost-effective, and globally applicable tool for volcanic risk mitigation. By capturing complex nonlinear relationships among static geophysical, petrological, and tectonic parameters, ML allows estimation of eruption magnitude prior to or early in eruptive activity, an outcome infeasible using classical approaches.

To demonstrate this potential, we present a ML framework to forecast a volcano’s potential erupted mass using static geophysical, petrological, and tectonic characteristics, together with eruption history. The model was trained on a dataset of 914 historical eruptions from 101 volcanoes and applied to estimate erupted mass for 135 globally distributed volcanoes active between 1982 and 2024, assuming a representative eruption duration of 225 days.

This approach provides the first global-scale erupted mass estimates that do not rely on high-resolution, time-dependent monitoring data (e.g., time-lapsed gravity, deformation, or seismicity). Feature importance and permutation analyses indicate that predictions are dominated by static geophysical parameters. The most influential predictors are eruption duration, elevation, gravity, and magnetic data. Parameters with intermediate influence include Moho depth, subsurface thermal indicators (e.g., depth to the Curie isotherm at ~580 °C and the lithosphere–asthenosphere boundary at ~1330 °C), dominant rock type, last eruption, and surface heat flow. Volcano landform, eruption type and start date, host crustal type, and tectonic setting exhibit relatively minor predictive influence.

Our results demonstrate that comprehensive, globally available geophysical datasets can robustly constrain erupted mass for medium- to long-duration eruptions (>3 months), while short-duration eruptive behavior may be better captured through detailed historical eruption records.

How to cite: Mousavi, N., Fullea, J., and Mousavi, S. M.: Volcanic Eruption Mass Estimation: A Machine Learning Approach, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4281, https://doi.org/10.5194/egusphere-egu26-4281, 2026.

High-energy surface waves (ground roll) are a major source of coherent noise in land seismic data, often overlapping with reflections and degrading subsurface imaging quality. We propose an intelligent surface-wave suppression method based on a dual-constraint framework that integrates data-driven supervision with a physics-guided prior. A composite loss is constructed with (1) a data constraint in the time–space (t–x) domain, implemented as a supervised loss that compares the network output with the labeled targets, and (2) a physics constraint in the frequency–velocity (f–v) domain, where the surface-wave dispersion curve is exploited to delineate the physically plausible ground-roll region and to penalize residual energy inconsistent with the dispersion curve. We train UNet and TransUNet architectures on field datasets using this composite objective. Compared with purely data-driven training, the proposed dual-constraint loss reduces the dependence on potentially imperfect labels by enforcing dispersion-consistent behavior in the f–v domain, leading to lower residual surface-wave energy while maintaining reflection continuity. These results demonstrate that incorporating physically meaningful constraints into modern network architectures can improve robustness under imperfect supervision and enhance intelligent seismic surface-wave suppression.

How to cite: Wang, B., Shi, Y., Wen, J., and Ning, J.: A Data-Physics Dual-Constraint Framework for Intelligent Surface Wave Suppression, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4447, https://doi.org/10.5194/egusphere-egu26-4447, 2026.

How to cite: Michael, J., Moreau, L., and Malfante, M.: Deep Learning Seismic Waveforms to Predict Sea Ice Properties, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4942, https://doi.org/10.5194/egusphere-egu26-4942, 2026.

The continuous seismic monitoring provides numerous waveform signals that provide useful information about geological processes such as landslides and geohazard activities. In tectonically active zones like the Himalayan arc, mass-movement events often overlap with natural earthquakes, creating challenges for reliable event discrimination. This study presents a semi-supervised learning framework that detects landslides like debris flow, rockfall, as well as earthquakes, from continuous seismological data with the help of a small amount of labelled dataset, calculating physically interpretable attributes from waveforms. The waveform and spectrum-based 157 features were extracted from segmented seismic windows, representing temporal, spectral, energy, and morphological attributes. The workflow combines dimensionality reduction, density-based clustering, and graph-based label propagation to identify and classify seismic events. To evaluate methodological choices, two complementary studies were conducted. The model benchmarking study compared four combinations of embedding and clustering algorithms, and a parametric sensitivity analysis that investigated the influence of key hyperparameters of the embedding and clustering algorithms. Feature importance analysis using statistical and machine-learning-based techniques was integrated throughout the study to ensure physical interpretability and to identify attributes most relevant for source discrimination. In this study, three months of continuous seismological data of the Tehri region, Uttarakhand, were analysed and revealed many previously undetected events. Uniform Manifold Approximation and Projection (UMAP) for dimensionality reduction, followed by Hierarchical Density-Based Spatial Clustering (HDBSCAN) for unsupervised event grouping, provided the best performance for seismic event detection. This approach effectively identified seismic events that would be difficult to observe using conventional methods. The proposed approach is well-suited for large-scale seismic monitoring applications where labelled data are limited and provides a broad application for geohazard detection and operational seismic analysis. 

How to cite: Chakraborty, A. and Sharma, M.: Detection of Seismic Events Using Semi-Supervised Learning , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5798, https://doi.org/10.5194/egusphere-egu26-5798, 2026.

Accurate and robust seismic phase picking remains a fundamental challenge in automated earthquake monitoring, particularly under complex waveform conditions. While recent deep learning models such as PhaseNet and EQTransformer have demonstrated strong performance on commonly used benchmark datasets, their architectural design choices introduce limitations in temporal resolution, sequence modeling, and generalization to more realistic seismic scenarios.

PhaseNet adopts a convolutional U-Net structure that is effective for waveform segmentation but is constrained by a fixed receptive field and limited dynamic temporal modeling. EQTransformer, in contrast, employs an encoder–attention–decoder architecture capable of capturing long-range dependencies, yet relies on aggressive temporal downsampling to alleviate the quadratic cost of self-attention. This heavy compression can discard fine-grained temporal information and degrade phase onset precision.

In this work, we present EQMamba, a sequence modeling framework for seismic phase picking that emphasizes temporal fidelity and efficient long-range dependency modeling. The proposed architecture integrates structured state-space models with efficient linear attention mechanisms, allowing long waveform sequences to be processed with minimal downsampling. By preserving high-resolution temporal information while maintaining computational tractability, EQMamba is designed to better reflect the continuous-time and dynamical nature of seismic signals.

Beyond controlled single-event settings, this study places particular emphasis on model behavior under more realistic waveform conditions, including event superposition, amplitude imbalance, and temporal interference between phases. We introduce a revised data construction and evaluation strategy to systematically probe robustness in multi-event and mainshock–aftershock scenarios, which are often underrepresented in standard benchmarks. Model performance is analyzed not only through conventional picking metrics, but also via error distributions, phase confusion patterns, and failure modes as waveform complexity increases.

How to cite: Wang, C.-H., Wang, W.-H., and Lu, H.-H.: Toward Robust Seismic Phase Picking in Realistic Multi-Event Scenarios, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6122, https://doi.org/10.5194/egusphere-egu26-6122, 2026.

Optical images acquired by satellite are widely used to measure the displacement field caused by earthquakes co-seismic ruptures both in the near and the far field. So far, this technique relies mostly on image correlation to locate pixels from an image acquired before the event on another acquired after. However, such approach suffers from several limitations inherent in the correlation method used, such as the need for sufficient texture to make objects « recognizable » from one image to the other, or limited changes in the landscape over time, for the same reason. These limitations can lead to noisy results and even prevent any measurement from being made.
To overcome such limitations, machine learning can be used instead of correlation to train a model to compute displacement maps from a pair of images. Steady and significant progress have been made in machine learning technics, and especially for image processing and computer vision, in recent years, and they need to be adapted to our case study. First, a training dataset is carefully designed, to enable the network to learn how to measure pixel displacements in satellite images, with sub-pixel accuracy, in the most realistic way possible. Since no ground truth is available, we build synthetic examples, where a realistic and known deformation is applied to one of the images in a pair of 10-m-resolution Sentinel-2 satellite images, which originally contains no displacement. This realistic synthetic dataset is then used to feed a model.
Our network is capable of estimating a displacement field from images whose resolution differs signifiantly from that of the training dataset (for example, from a 0.5-m-resolution Pléiades image pair) and achieves results comparable to those of state-of-the-art methods, with even finer details, at both pixel and sub-pixel resolution levels. However, the ability of machine learning to overcome limitations due to landscape changes caused by time remains to be proven.

How to cite: Delorme, A., Rupnik, E., Klinger, Y., and Pierrot-Deseilligny, M.: Machine learning to compute, from optical images, the horizontal ground displacement field caused by an earthquake, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6784, https://doi.org/10.5194/egusphere-egu26-6784, 2026.

In a previous work, we established the predictive power of seismic statistical catalogue features for whole-region temporal forecasting. We here extend this framework to a spatiotemporal approach to assess localised seismic hazard in Japan and Chile. Using ensemble learning, we predict the occurrence of M>=5 earthquakes within a 15-day horizon across varying radial distances (r=3 to 24 km) to benchmark the framework's sensitivity as a proof-of-concept prior to scaling for larger magnitude hazards.

Results indicate robust predictive power, though performance is sensitive to the prediction radius. The Japan catalogue yields an AUC of 0.76 for predictions within 24 km. However, when the prediction radius is tightened to 12 km, while the model retains predictive power (AUC 0.62), the reduced performance underscores the challenge of highly localised forecasting. Crucially, we observe a distinct shift in feature importance as the spatial scale changes: parameters that track local variations in seismicity—specifically the b-value, within our feature set—rank significantly higher in localised models compared to whole-region baselines. This suggests that machine learning models can produce forecasts that reflect underlying physical fault processes.

We further present ongoing work regarding spatiotemporally overlapping predictions, testing the hypothesis that multiple alerts intersecting in both space and time indicate a compounded hazard probability. Finally, responding to the challenges of localised prediction, we introduce a novel experimental framework that augments our current statistical features by exploring additional spatial descriptors, including both deep learning representations and hand-crafted spatial features, designed to capture aspects of fault dynamics beyond standard catalogue statistics.

How to cite: Quan, W. and Gorse, D.: Leveraging the value of seismic catalogue features in building a spatiotemporal system to assess localised seismic hazard, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7938, https://doi.org/10.5194/egusphere-egu26-7938, 2026.

Phase picking in seismology is the first step of signal processing and locating a seismic event. At the beginning of seismological research, it is straightforward to manually pick P- and S-wave arrival times because the number of seismic stations is relatively small. Recently, instrumentation has improved, and the gap is lower than before. This widespread instrumentation creates problems for users who still have to pick manually, especially in national networks such as those in Türkiye. We used pre-trained deep learning models, trained on different seismic datasets, to estimate P- and S-wave arrival times for earthquakes in the Marmara Region, NW Türkiye. We compared these times with manual readings collected from the Ministry of Interior, the Disaster and Emergency Management Presidency, and the Earthquake Department (AFAD). The results indicate that PhaseNet produces arrival-time estimates that are largely consistent with expert manual readings, demonstrating its potential to substantially reduce analyst workload in large-scale seismic monitoring systems. The mean absolute error ranges from 6 to 14 seconds, and the number of total picks varies between 75,000 and 140,000 for both the P-wave and S-wave. This project has been supported by the Scientific and Technological Research Council of Türkiye (TÜBİTAK) under grant 124E294, and the results have been shared on the website (https://quakemlab.sakarya.edu.tr)

How to cite: Tezel, T., Erden, C., Arıkan, H., Küçükkara, M. Y., Yanık, K., and Utkucu, M.: QuakeMLab Phase I: Deep Learning-Based Automated Seismic Phase Picking Using PhaseNet, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8055, https://doi.org/10.5194/egusphere-egu26-8055, 2026.

Seismic waveform tomography typically uses either traditional adjoint methods to compute gradients for model updates or neural-network-based (NN-based) methods to directly predict the model. However, adjoint methods require complex analytical derivations that must be reformulated for each combination of model parameters (velocity or attenuation), wave equations (elastic or viscoelastic), and misfit functions (waveform, travel time, differential time or amplitude). Here, we replace existing adjoint methods with automatic differentiation (AD), which computes accurate gradients of wave equation-based data misfits directly without any analytical derivations. Compared with NN-based methods (e.g. PINN or neural operator), our AD-based tomography framework is fully white-box and does not require any training datasets. We demonstrate both theoretically and numerically that gradients computed with AD are identical to those from adjoint methods, regardless of the domain, wave equation, or misfit function. For a field application, we apply ambient noise differential AD tomography to data from the Southeastern Suture of the Appalachian Margin Experiment (SESAME) and obtain three 2D Love-wave shear velocity (Vsh) models. The imaged Paleozoic suture zone, Mesozoic rift basins, and Moho interface are consistent with previous studies. Our results highlight the unifying role of AD in geophysical inverse problems beyond gradient computation, with promise for broader future applications across geoscience.

How to cite: Wang, X., Li, Z., and Liu, X.: A Unified Seismic Tomography Framework Using Automatic Differentiation Applied to the Southern Appalachian Array, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9203, https://doi.org/10.5194/egusphere-egu26-9203, 2026.

Laboratory, theoretical and field data suggest that fault-zone properties should evolve during the seismic cycle as stress rises prior to failure and drops during earthquake rupture. Lab work shows systematic changes in elastic properties during the seismic cycle and that these changes can be used to predict lab earthquakes.  Recent work shows that in some cases these results are also applicable to tectonic faults. Seismic data show a clear distinction between fault zone properties pre and post mainshock, as well as post-seismic time-dependent changes in elastic properties. Here we extend these works by developing tools to distinguish seismic waves pre/post mainshock that are both predictive and physically interpretable. We train a convolutional neural network (CNN) on RGB spectrograms developed from three component seismograms recorded at seismic stations around the M6.5 2016 Norcia earthquake.  Our model can accurately distinguish foreshocks from aftershocks of the sequence. We train and test models on individual stations and also on all stations and subsets of the stations based on source-station geometry of the Norcia fault.  Models trained on the full set of stations achieve >99% accuracy for foreshock/aftershock classification and models trained on individual stations achieve higher accuracy in tests.  For each set we also performed the SHapley Additive exPlanations (SHAP) technique. We find that specific time-frequency signatures in the RGB spectrograms identify each class.  Here we extend that framework to a multi-station setting by training CNN models on spectrograms from several seismic stations surrounding the mainshock. While the multi-station model achieves high classification accuracy (about 97%), SHAP analysis reveals a substantial reorganization of feature importance, including strong station-dependent variability and a reduced contribution from aftershock-related regions.  Even for data from the reference station (NRCA), SHAP patterns differ markedly from those obtained in the single-station case, suggesting that heterogeneous training distributions alter global attribution mechanisms. To disentangle these effects, we additionally train station-wise CNN models, which achieve very high accuracy and produce more stable and physically coherent SHAP explanations. These results indicate that station-specific propagation effects play a key role in model interpretability and that caution is required when applying SHAP to models trained on spatially heterogeneous seismic datasets. The findings motivate future work toward hierarchical, region-aware, or physics-constrained interpretability frameworks.

How to cite: Marrocco, F., Magrini, M., Laurenti, L., Paoletti, G., Tinti, E., and Marone, C.: Station-Level and Network-Wide SHAP Explanation of CNN Models for Seismic Cycle Monitoring: Evidence from Norcia 2016, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10105, https://doi.org/10.5194/egusphere-egu26-10105, 2026.

Deep learning based seismic elastic inversion workflows strongly benefit from realistic synthetic angle-stack seismic data, especially where field data are limited or unavailable. This study presents a stratigraphically constrained workflow for exact Zoeppritz-based amplitude variation with offset (AVO) forward modelling using augmented well-log data. The methodology integrates P-wave velocity (Vp), S-wave velocity (Vs), and density (ρ) logs from real wells with interpreted seismic horizons to generate geologically consistent synthetic elastic models and angle-dependent seismic responses. Within each stratigraphic interval defined by seismic horizons, multivariate statistics of Vp, Vs, and ρ are estimated across all available wells, preserving intrinsic elastic parameter correlations. Synthetic wells are then generated through multivariate data augmentation conditioned to these statistics and constrained by the stratigraphic framework. The resulting well-log data are used for AVO forward modelling based on the exact Zoeppritz equations, computing angle-dependent P-wave reflection coefficients at elastic interfaces. These reflection coefficients are subsequently convolved with a seismic wavelet to generate synthetic angle stacks. The proposed workflow produces consistent sets of synthetic elastic wells logs and exact Zoeppritz-based angle-stack data, providing realistic and physically grounded training datasets for seismic elastic inversion.

How to cite: Bokhonok, O., Paulo Marchezi, J., Alexandra Delgado Blanco, L., Pereira Leite, E. P., Hartman, G., and Batezelli, A.: Stratigraphically Constrained Well-Log Data Augmentation for Angle-Stack AVO Forward Modelling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10510, https://doi.org/10.5194/egusphere-egu26-10510, 2026.

The characterization of reservoir properties from seismic data is often hindered by the scarcity and spatial bias of well-log data. To overcome these limitations, data augmentation (DA) has become essential for training robust Deep Learning models. This study presents a comparative analysis of three distinct DA approaches: Statistical Methods, Variational Autoencoders (VAE), and Generative Adversarial Networks (GAN. We synthesize well-log suites for training a UNet architecture dedicated to seismic-to-porosity prediction. Our workflow begins with real well logs, expanding the dataset through stochastic perturbations (Statistical), latent manifold sampling (VAE), and adversarial learning (GAN). To bridge the gap between 1-D well data and seismic volumes, we perform forward modeling on the augmented suites, generating synthetic seismograms via convolution with representative wavelets. A UNet-based convolutional neural network is then trained on these synthetic pairs to perform the non-linear mapping from seismic amplitudes to porosity. The performance of each method is evaluated through the geological plausibility of the generated logs and the inversion accuracy on a blind-test well. Preliminary results indicate that while statistical methods improve robustness against noise, generative models, particularly GANs, excel in capturing the multi-scale heterogeneity required for high-resolution reservoir characterization. This research demonstrates that the choice of DA is a critical geophysical decision; by integrating generative AI into the inversion workflow, we provide a scalable framework to improve porosity estimation in data-poor environments, ensuring that synthetic extensions remain grounded in petrophysical reality and stratigraphic consistency.

How to cite: Marchezi, J. P., Leite, E., Bokhnok, O., Delgado, L., Hartman, G., and Batezelli, A.: Impact of Generative AI and Statistical Data Augmentation in Synthetic Well-Log Generation for Seismic Porosity Inversion: A Comparative Study, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10548, https://doi.org/10.5194/egusphere-egu26-10548, 2026.

Deep learning has improved automated seismic phase picking in recent years. However, many pickers are trained on a single dataset and often fail to generalize when deployed on out-of-distribution (OOD) data. Variations in earthquake magnitude, propagation distance, sensor instrumentation, and ambient noise between training and target domains lead to significant performance degradation under OOD conditions. To address this challenge, we propose a new framework to reduce performance degradation under OOD conditions based on a pipeline of seismic simulation -- phase labeling -- site-conditions simulation -- transfer learning. First, we use the seismic simulation tool AxiSEM3D to establish a 3-D waveform simulation, covering local, regional, and teleseismic scales. Next, we obtain precise P- and S-phase labels by computing theoretical arrival times from velocity models and refining these onsets with traditional automatic picking algorithms, ensuring high-fidelity phase annotations. Then, based on actual site conditions, we simulate instrument responses and synthesize ambient noise constrained by PPSD analysis. This gives us station-specific, noisy waveforms that closely match real observational conditions. Finally, we employ transfer learning to fine-tune a phase picker on this specific synthetic dataset, thereby enhancing the picker's performance in new conditions. The proposed framework aims to improve the ability of deep learning models to pick phases under OOD conditions. It enables reliable performance across regional variability and instrumentation differences without large-scale manual relabeling. It also reduces the amount of real data needed for training, making it useful for small datasets and leaving more data to analyse. Overall, this work introduces a transferable methodology for seismic phase picking under distribution shift and shows that physics-informed data augmentation combined with targeted transfer learning can effectively decrease OOD performance degradation, thereby increasing the applicability of deep-learning-based phase picking.

How to cite: Jiang, P., de Laat, J. I., Wang, X., and Fadel, I.: Earthquake phase picking in out-of-distribution data conditions: Can synthetic data from earthquake simulations help?, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10617, https://doi.org/10.5194/egusphere-egu26-10617, 2026.

Earthquake catalogs are essential for monitoring natural hazards and improving our understanding of seismic processes, which depend on accurate detection and classification of both natural earthquakes and anthropogenic signals. This is especially important in intraplate regions like Norway characterized by low to moderate earthquake activity, rare impactful earthquakes and widespread anthropogenic events such as quarry blasts. Nonetheless, traditional detection workflows have struggled to keep pace with the growing number of seismic stations and temporary deployments. Recent advances in machine learning offer promising solutions for efficient detection and discrimination tasks. Here, we present preliminary results toward a fully automated seismic detection and classification system for the Norwegian National Seismic Network (NNSN).

We first evaluate pre-trained deep learning-based phase detection models PhaseNet (Zhu & Beroza 2019) and Earthquake Transformer (EQT, Mousavi et al. 2020) using a catalog of 2567 events from the NNSN bulletin from 2008 to 2025, which includes 1144 earthquakes and 1423 blasts. We find the models can detect earthquake phases with F1-scores of 0.70 and 0.67, for the PhaseNet and EQT respectively, and 0.73 and 0.70 for blasts, revealing slightly higher sensitivity to blasts than earthquakes.

Building on this, and with the goal of developing a robust deep-learning earthquake detection workflow, we set out to quality-control our classification of earthquakes and blasts. We apply a self-supervised approach based on Bootstrap Your Own Latent (BYOL, Grill et al. 2020), which learns representations by aligning augmented views of the same signal, enabling learning without relying on potentially biased labels. The only information provided to the model are the training hyperparameters and the number of classes we aim to identify (two: earthquakes and blasts). Our method achieves an F1-score of 0.93 in distinguishing blasts from earthquakes using self-supervised representations, and up to 0.94 when incorporating an additional supervised layer. Analysis of the results reveals previously misclassified events, demonstrating the effectiveness of self-supervised methods even with limited or biased labeled datasets.

Future work will focus on retraining detection models using NNSN data after BYOL based classification. Additionally, we will analyze the BYOL learned features to gain insights on the physical differences between earthquake and blast signals.

How to cite: Neves, M., Ottemöller, L., and Rondenay, S.: Exploring Deep Learning Approaches for Seismic Detection and Discrimination in Norway, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10618, https://doi.org/10.5194/egusphere-egu26-10618, 2026.

Mount Etna, unlike many volcanoes that experience prolonged calm intervals, exhibits persistent and continuous activity characterized by frequent strombolian bursts, lava fountains, and effusive events. This study aims to automatically identify recurrent and distinctive patterns in seismic signals by extracting clusters of waveforms with similar spectral characteristics using a fully data-driven, unsupervised machine learning framework, and to assess their correspondence with observed volcanic activity.

We analyzed daily seismic spectrograms from two summit seismic stations, ECPN and ECNE, spanning November 2020 to November 2021, a period encompassing both quiet intervals and two major lava fountain cycles. For dimensionality reduction and feature extraction, we employed AutoencoderZ, an encoder–decoder model with skip connections, convolutional and fully connected layers, a bottleneck latent space, and transposed convolutions. This architecture compresses inputs while preserving critical spectral features for unsupervised clustering. Extracted features are optimized using the Relative Bias metric and clustered via Deep Embedded Clustering (DEC), enabling data-driven anomaly detection and pattern recognition by clustering similar waveforms.

The resulting clusters were compared with independent observational datasets and seismic related metrics, including lava fountain records, volcano-tectonic and long-period (LP) event catalogs, and root mean square (RMS) amplitude of volcanic tremor. This comparison demonstrates the approach’s ability to uncover hidden structures in the seismic data and highlight key temporal transitions associated with underlying processes such as magma and fluid dynamics . To improve robustness and reduce potential spatial bias, analyses were conducted using both single-station and dual-station approaches, providing a more reliable characterization of seismic variability.

Overall, this study highlights AutoencoderZ’s versatility in revealing complex patterns in Etna’s seismic activity.

How to cite: Abed, W., Zali, Z., Sciotto, M., Cocina, O., Cannata, A., Picozzi, M., Martínez-Garzón, P., Vuan, A., Saraò, A., and Sugan, M.: Understanding Volcanic Seismic Patterns with Unsupervised ML Clustering: Application at Mount Etna volcano, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11584, https://doi.org/10.5194/egusphere-egu26-11584, 2026.

The exploration and characterization of complex continuous seismological datasets remain challenging, particularly for highly active and/or noisy environments. Recently, several artificial intelligence based approaches have been proposed to facilitate the analysis of seismological data, either by characterizing detected events or continuous streams. Among these, we introduced an image-based self-supervised learning framework to explore continuous seismic records without requiring prior labeling or event detection. However, image-based representations may result in a loss of information, as they are derived transformations of the original raw seismic time series and may impact the discrimination of seismic events.

In this study, we adapted a self-supervised learning based clustering workflow to operate directly on multichannel seismic time series. The main challenge when using self-supervised contrastive learning approaches with time series is adapting the data augmentation techniques to ensure sufficient transformation without losing the physics contained in the seismological records. We leveraged the contrastive learning framework to analyse two months of continuous records from Ocean Bottom Seismometers deployed near the Fani Maoré submarine volcano, using data augmentation strategies consistent with seismological records, such as channel masking and window masking in the time and frequency domains. The model was trained using four-channel time series derived from the raw data (three-component seismometer and one hydrophone) using 60 s sliding windows with a 50% overlap, enabling the network to learn meaningful latent representations of the data. Clustering was then performed directly within the learned latent space, allowing the identification of distinct signal groups. Applied to the Fani Maoré dataset, this approach revealed several families of clusters, including very rare and previously undocumented events likely associated with the activity of the Fani Maoré submarine volcano.

How to cite: Rimpot, J., Retailleau, L., Saurel, J.-M., Hibert, C., Malet, J.-P., Forestier, G., and Weber, J.: A Time-Series-Based Self-Supervised Learning Approach for the Exploration of Complex Seismological Datasets: Application to the Fani Maoré Submarine Volcano, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13291, https://doi.org/10.5194/egusphere-egu26-13291, 2026.

Submarine volcanism represents one of the dominant forms of magmatic output on Earth, yet our understanding of the underlying eruptive processes remains limited compared to subaerial systems. The remote locations and lack of direct visual observations often obscure the dynamics of these eruptions, where the interaction with the hydrostatic load and phase changes (steam/water) creates distinct physical regimes. In the absence of near-field observations, far-field hydroacoustic records—propagated over thousands of kilometers in the SOFAR channel—provide a critical, high-temporal-resolution window into these deep-sea events. However, a significant challenge lies in processing the high volume of continuous data to categorize signals into physically meaningful regimes without relying on manual classification or subjective thresholds.

In this study, we present Spec2Vec, a novel framework for the unsupervised classification of hydroacoustic time series, applied to the major 2021 Fukutoku-Oka-no-Ba shallow submarine eruption. Current unsupervised approaches in geo-acoustics often rely on decomposition methods like Non-Negative Matrix Factorization (e.g., SPECUFEX), Independent Component Analysis (ICA), the scattering transform, or latent representations from neural network auto-encoders. While effective, these methods can be computationally intensive or result in "black box" features lacking direct physical intuition. In contrast, Spec2Vec utilizes Hilbert space-filling curves to topologically map 2D time-frequency representations into a 1D sequence, strictly preserving multi-scale locality. From this linearized stream, we extract a compact set of entropy and scaling features. This approach captures the "texture" of the spectrogram—the specific arrangement of energy in time and frequency—more uniquely and efficiently than standard modal image features. The resulting feature space is fast to compute and highly interpretable, bridging the gap between raw acoustic data and physical source mechanics.

We evaluate this feature set by applying it to ten days of continuous hydroacoustic data from the Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO) International Monitoring System (IMS), capturing the main eruption sequence as well as pre- and post-eruptive phases. Through unsupervised learning, Spec2Vec automatically organizes the complex acoustic stream into coherent clusters. To validate the physical interpretability of these clusters, we correlate the unsupervised classes with independent eruptive proxies, including satellite-derived lightning data, plume height, and mass eruption rates. Furthermore, we inject synthetic acoustic source models—simulating single bubble oscillations, turbulent jets, bubble plumes, hydroacoustic earthquakes, explosions, and volcanic tremor—into the dataset to map clusters to specific source mechanisms.

Our results offer a rare, data-driven characterization of the Fukutoku-Oka-no-Ba eruption, identifying distinct phases of jetting and tremor that align with atmospheric observations. This demonstrates that Spec2Vec serves not merely as a feature generation tool, but as a generalizable engine for the automated discovery of physical processes in complex geophysical time series. This approach holds significant potential for scaling the analysis of global hydrophone datasets, enabling the systematic distinction and quantification of eruption rates and processes across the global submarine volcanic inventory.

How to cite: Swar, S., Mittal, T., and Olugboji, T.: Decoding the Dynamics of the 2021 Fukutoku-Oka-no-Ba Submarine Eruption via Interpretable Spectral-Hilbert Representations (Spec2Vec), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14499, https://doi.org/10.5194/egusphere-egu26-14499, 2026.

Earthquake catalogues are derived from continuous seismic recordings through signal detection, phase picking, association, location, and magnitude estimation, but pervasive noise still limits reliable automation, especially for small events and noisy stations, and often requires manual review. Building on the demonstration that deep-learning denoising can be applied to continuous data to improve network-wide earthquake monitoring (Dahmen et al., 2026), we advance toward operational deployment by (i) implementing denoising within the SeisComP ecosystem (Helmholtz Centre Potsdam, 2008), (ii)  evaluating on larger continuous datasets, and (iii) systematically comparing multiple denoising approaches and testing monitoring-driven methodological refinements.

We train and compare multiple denoising models using a dedicated, curated training and benchmarking dataset composed of earthquake signals and noise recordings from Switzerland and its border regions, with event waveforms pre-cleaned to enhance label quality. Denoised waveforms are then propagated through an end-to-end monitoring workflow spanning signal detection, continuous waveform denoising, phase picking with arrival-time uncertainty estimation and peak amplitude estimation, and final catalog generation. The performance is benchmarked with monitoring-relevant metrics such as signal detection capability, waveform fidelity, phase-pick quality, and the reliability of amplitude estimation, thereby quantifying, for each denoiser, the trade-offs and improvements relative to standard digital filters and relative to applying common phase pickers to raw versus denoised data.

A case study using continuous data in realistic settings shows that catalogues based on denoised data can contain significantly more detected events with more associated phase picks, improved location quality, and more reliable magnitude estimates than catalogs derived from raw data, ultimately extending catalogue depth toward smaller magnitudes while preserving reliability.

This work is carried out within TRANSFORM², funded by the European Commission under project number 101188365 within the HORIZON-INFRA-2024-DEV-01-01 call.

References:

Dahmen, N., J. Clinton, M.-A. Meier, and L. Scarabello, 2026, Toward Operational Earthquake Seismogram Denoising, Bull. Seismol. Soc. Am., XX, 1–23, doi: 10.1785/0120250198

Helmholtz Centre Potsdam (2008). The SeisComP seismological software package, GFZ Data Services.

How to cite: Dahmen, N., Clinton, J., Meier, M.-A., and Scarabello, L.: Towards Operational Earthquake Data Denoising, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17302, https://doi.org/10.5194/egusphere-egu26-17302, 2026.

Understanding earthquake rupture directivity is crucial for constraining source physics and improving seismic hazard assessment. In an ideal setting, one would analyze far-field seismograms free from subsurface scattering, travel-time uncertainties, and ambient noise, enabling direct inference of rupture directivity from the observed waveforms. In practice, however, recorded seismograms are strongly influenced by path effects, site responses, and additive noise, while station azimuthal coverage is often sparse and uneven. These limitations significantly complicate directivity analysis, particularly for 1) low-magnitude noisy events and 2) earthquakes with complex rupture processes, for which simple source–path deconvolution models are inadequate.

We employ a conditional Diffusion Transformer (DiT) to learn the dependence of far-field seismograms on the P-ray take-off direction. The DiT is trained on measured far-field seismograms from multiple earthquakes with moment magnitudes Mw ≥ 6.0. Once trained, the model generates virtual far-field seismograms conditioned on specified ray azimuth and take-off angle, while holding an empirical realization of the path effects fixed by conditioning on a reference observed seismogram. This enables controlled experiments in which variations attributable to source directivity can be examined independently of path-induced variability. In this sense, our approach closely mimics the idealized setting in which far-field seismograms vary only with source directivity and are free from complex path effects. In other words, this generative framework enables us to isolate and examine variations in the wavefield that are attributable solely to source directivity, while holding path effects constant. We demonstrate that this approach is particularly effective for earthquakes with complex multi-episode moment release. For all events considered, the generated wavefields vary smoothly with take-off direction, indicating physical consistency. Importantly, the DiT training is self-supervised, requiring neither synthetic earthquake simulations nor explicit correction for path effects. The proposed framework provides a scalable and physically consistent tool for investigating earthquake directivity and rupture complexity across a wide range of magnitudes.

How to cite: Gupta, P. and Bharadwaj, P.: Understanding the directivity of earthquakes by generating virtual far-field seismograms conditioned on the P-ray take-off direction  , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18002, https://doi.org/10.5194/egusphere-egu26-18002, 2026.

In modern seismology, the rapid and accurate characterization of seismic sources following an earthquake is a fundamental task. Most observatories currently compute moment tensor solutions for events exceeding specific magnitude thresholds to ensure reliability. For instance, the National Institute of Geophysics and Volcanology (INGV) provides routine solutions for moderate to large events (Mw​≥3.5) using established methods (Scognamiglio et al., 2009). However, characterizing smaller events remains challenging due to low signal-to-noise ratios and the need of modeling high-frequency waveforms that requires detailed knowledge of the velocity model, which is rarely available.

Machine learning (ML) techniques have emerged as powerful tools to address these limitations, particularly in improving the prediction of first-arrival seismic wave polarities. These ML-derived polarities can be effectively integrated into traditional frameworks to compute robust focal mechanisms. In this study, we implement a workflow that bridges deep learning polarity predictions with standard focal mechanism estimation techniques, focusing on the tectonic setting of the Italian Peninsula.

Our methodology consists of two primary stages. First, we trained a Convolutional Neural Network (CNN) for polarity prediction using the INSTANCE catalog (Michelini et al., 2021), which provides the high-quality, manually reviewed data essential for supervised learning. Second, we validated the model’s performance by analyzing approximately 4,700 earthquakes that occurred in Italy between January 1 2021, and January 1 2025, with magnitudes below 4.5.

To benchmark our results, we selected a subset of earthquakes with existing Time Domain Moment Tensor (TDMT) solutions (Scognamiglio et al., 2006). Using the polarities predicted by the CNN, we computed focal mechanisms using the SKHASH code (Skoumal et al., 2024). The accuracy of these solutions was then evaluated against the TDMT catalog using Kagan angle analysis (Kagan, 1991) to quantify the rotation between double-couple sources.

How to cite: Tavani, F., Scognamiglio, L., Artale Harris, P., and Meier, M.-A.: Source Characterization via Deep Learning: Integrating Polarity Prediction and Focal Mechanism Estimation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18532, https://doi.org/10.5194/egusphere-egu26-18532, 2026.

The STAR-E Project that aims to develop state-of-the-art information science techniques, including artificial intelligence, applicable in seismology is going on in Japan, supported by the Ministry of Education, Culture, Sports, Science and Technology (MEXT). We introduce our various activities in the SYNTHA-Seis, which is one of the research teams in the STAR-E Project. We have been developing deep learning models to detect earthquake signals in seismic continuous waveforms, such as for phase-picking (Tokuda and Nagao, 2023; Katoh et al., 2025; Gerardo Mendo et al., submitted) and for P-wave polarity determination with UQ using the Monte Carlo dropout method (Katoh et al., 2025). We have also been developing methods to extract waveform features of low-frequency tremors (LFTs), such as a template matching technique to extract LFTs waveforms (Gerardo Mendo et al., 2025), a deep learning model to detect LFTs in historical paper records obtained by mechanical seismograms more than fifty years ago (Kaneko et al., 2023), and a deep learning technique to acquire a stochastic differential equation expression of LFTs (Kusui et al., 2025). We also discuss the future direction of AI seismology in Japan.

How to cite: Nagao, H., Mendo Pérez, G. M., Katoh, S., and Kusui, T.: Deep Learning Models for Seismic Continuous Data Analysis Developed in STAR-E Project, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18544, https://doi.org/10.5194/egusphere-egu26-18544, 2026.

Estimating ground‐motion intensity at individual seismic stations is a fundamental task in seismology, with direct implications for seismic hazard assessment.
Ground‐motion intensity measures (IMs) such as Peak Ground Acceleration (PGA), Peak Ground Velocity (PGV), and Spectral Acceleration (SA) at selected periods are commonly used to quantify shaking severity and relate it to building structural response and seismic hazard.

Here, we present an AI-driven framework for predicting multiple IMs at the station level, building on previous graph-based approaches for the Italian seismic network.
Starting from the INSTANCE dataset, we applied filters on event magnitude (≥ 3) and waveform quality, resulting in 3076 events recorded across 565 stations. For waveform analysis, we used a 10-second time window starting 1 second before the first P-wave arrival, balancing prediction speed and accuracy.

Seismic waveforms are first encoded using a pre-trained PhaseNet model to extract compact temporal representations. Spatial dependencies are modeled with a masked Graph Convolutional Network (GCN) based on Delaunay triangulation, which links each station to its nearest neighbors while avoiding long or crossing edges. This structure allows identification of border stations (at mesh edges) and coastal stations (within 15 km of the coastline). A binary mask distinguishes these nodes, helping the model account for areas with high azimuthal gap, which can make IMs estimation more challenging.

Before the final prediction layer, the model concatenates the maximum waveform amplitude across stations with event metadata predicted by a fine-tuned LLM for magnitude and location estimation.
This enables joint exploitation of temporal waveform features, network geometry, and global event information.

The framework predicts PGA, PGV, and SA at periods of 0.3, 1, and 3 s at all stations. We obtained preliminary results for multiple configurations. The baseline model served as reference, which included waveform representations and the GCN but not LLM metadata, border features, or weighted loss. Including LLM-derived metadata consistently improved performance across all regions and parameters, reducing relative errors by 14% in Southern Italy and over 27% in Northern Italy. The addition of border and coastal features provided only minor improvements, confirming that metadata was the main factor driving gains.

Applying a weighted loss emphasizing stations closer to the epicenter further improved predictions, particularly in Northern and Southern Italy. In Central Italy, where network coverage is denser, improvements were smaller, suggesting that local contributions were already well captured. For the best configuration (LLM metadata and weighted loss), global mean absolute errors across stations were 0.511 (PGA), 0.423 (PGV), 0.514 (SA0.3), 0.447 (SA1.0), and 0.397 (SA3.0), demonstrating the model’s predictive accuracy.

Overall, these preliminary results show that combining waveform features, network topology, and LLM-informed event metadata can substantially enhance station-level IMs estimation, achieving better results in challenging conditions (north or south Italy) with respect to previous similar approaches. This method has potential for rapid earthquake characterization and early warning, where timely and accurate ground-motion predictions are essential.

How to cite: Bassani, A., Trappolini, D., Scifoni, A., Poggiali, G., Tinti, E., Galasso, F., Michelini, A., and Marone, C.: AI Framework for Ground-Motion Prediction Across the Italian Seismic Network, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18748, https://doi.org/10.5194/egusphere-egu26-18748, 2026.

Microseismic source localization is a key diagnostic in stimulated reservoirs, supporting spa-
tiotemporal tracking of fracture activation, stress transfer, and operational risk. Distributed Acoustic
Sensing (DAS) provides dense strain-rate observations along fiber-optic cables, but the resulting data
volume and strong near-well heterogeneity motivate localization workflows that are both fast and
physically constrained. We present an Eikonal-regularized U-Net Fourier Neural Operator (U-FNO)
that predicts full first-arrival traveltime fields for a given source location in a known 2-D velocity
model. The architecture combines Fourier-domain operator learning to capture long-range kinematic
structure with a multiscale encoder–decoder to recover spatial detail. Training is guided by an
Eikonal-consistency loss, complemented by source anchoring and a non-negativity constraint to
encourage physically admissible solutions. We benchmark U-FNO against a vanilla FNO baseline
and fast-marching traveltime solutions across velocity models of increasing complexity (smooth
gradient, Marmousi, and Utah FORGE). In the FORGE model, U-FNO reduces traveltime RMSE
by up to 97% relative to the baseline and reaches comparable misfit in up to 50% fewer training time.
Field transferability is assessed using 15 DAS-recorded events from the Utah FORGE microseismic
catalogue. Models are fine-tuned on four events and evaluated on the remaining events. U-FNO converges
within 2 minutes (versus 45 minutes for FNO) and reduces the mean location error from 33.71 m to
29.28 m. These results indicate that physics-regularized neural operators with multiscale structure can
deliver accurate, scalable, near-real-time localization for high-volume DAS monitoring in complex
geothermal settings.

How to cite: Al-Qadasi, B. and Bin Waheed, U.: Physics-Informed U-Net Fourier Neural Operator for DAS Microseismic Localization at Utah FORGE, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20316, https://doi.org/10.5194/egusphere-egu26-20316, 2026.

Indonesia is located in one of the most seismically active regions in the world and is monitored by more than 550 broadband and short-period seismic sensors. On average, around 40,000 earthquakes with magnitudes greater than 2 occur each year. However, this number of earthquakes with magnitudes larger than 2 has only been observable in recent years due to the significant expansion of the seismic network, for example in 2025. In earlier years, the recorded number of earthquakes was significantly lower, with the magnitude of completeness (Mc) reaching only about 4. Completing earthquake catalogs in a region is extremely important for revealing detailed main and secondary fault structures, which is essential for improved earthquake and tsunami hazard assessment. Recently, earthquake event recognition and phase picking using machine learning (ML) have proven highly successful in detecting smaller-magnitude earthquakes that are often overlooked by conventional methods such as STA/LTA. This study presents high-resolution earthquake catalogs generated using ML-based earthquake detection. The ML algorithms were pre-trained across various regions, tectonic settings, and environmental conditions using data from the last decade of combined BMKG and temporary seismic networks across Indonesia. We compare a three-catalog framework consisting of ML-derived, real-time, and analyst-reviewed catalogs. The results show that ML-based detection identifies significantly more earthquakes at lower magnitudes, with Mc reaching approximately 2, compared to real-time processing and human-reviewed catalogs. Using a recently published ML-based focal mechanism model, our results also show a substantially larger focal mechanism catalog, including many events with magnitudes smaller than 5 that are often not reviewed in conventional processing. This study demonstrates the importance of ML-based earthquake detection in improving the efficiency and completeness of earthquake detection and highlights its strong potential for operational integration into Indonesia’s seismic monitoring systems.

How to cite: Simanjuntak, A. V. H., Warnana, D. D., Pranata, B., Supendi, P., Daryono, D., Mai, M., Riama, N. F., and Palgunadi, K. H.: Completing Indonesia Earthquake Catalog for Better Earthquake and Tsunami Hazard Assessments, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20854, https://doi.org/10.5194/egusphere-egu26-20854, 2026.

Accurate delineation of subsurface lithological and structural characteristics is essential for applications ranging from groundwater exploration to environmental geophysics. Traditional single-modality inversion techniques often suffer from resolution trade-offs and ambiguity in petrophysical interpretation. In this work, we propose a novel multimodal deep learning framework for the joint inversion of Ground Penetrating Radar (GPR) and Vertical Electrical Sounding (VES) data, enabled through physics-informed Siamese neural network architecture. This network is explicitly designed to learn shared subsurface representations by encoding signal-specific features via dual 1D convolutional pathway, which are subsequently fused into a common latent embedding. From this shared space, the network predicts three key geophysical outputs: (i) layer-wise resistivity, (ii) normalized thicknesses , and (iii) geological model classification from eight lithological types.

To ensure physical consistency between modalities, the architecture incorporates an empirical dielectric–resistivity relationship derived from soil physics literature as a physics-informed regularization loss, coupling the inferred resistivity profile with dielectric behavior. The resistivity head uses a Huber loss on log-transformed outputs to reduce the effect of noise and outliers, while the thickness head is stabilized with Batch Normalization and dropout layers to prevent over fitting. A multi-class cross-entropy loss is used for geological classification, and a joint loss function ensures simultaneous optimization across modalities.

Training is conducted on a synthetically generated dataset comprising 24,000 4-layer models, covering diverse resistivity-thickness scenarios and geological facies. A dedicated subset includes thin-layer configurations, simulating challenging cases where GPR contributes enhanced resolution beyond VES capabilities. The network achieves a classification accuracy of 95.4%, a resistivity RMSE of 1.76 Ω·m, and thickness RMSE of 1.85 m on unseen validation data, validating its predictive performance. An ablation study with three independent random seeds (42, 123, 2025) confirms the network’s stability and generalizability.

Visual comparison of predicted vs. true resistivity and thickness profiles exhibits strong structural alignment, even in geologically complex models. Further, input-output attention diagnostics and multimodal fusion behavior reveal interpretable latent correlations between GPR and VES responses.

This work introduces a scalable and domain-aware inversion framework that learns geophysical realism, respects petrophysical coupling laws, and demonstrates potential for field-deployable AI-assisted subsurface mapping. The integration of empirical physics, attention mechanisms, and synthetic realism places this methodology at the frontier of modern geophysical inversion strategies.

How to cite: Gogoi, H., Sengupta, P., Hazarika, C., and Dutta, A. D.: Joint Inversion of GPR and VES Data Using Physics-Guided Siamese Networks, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21987, https://doi.org/10.5194/egusphere-egu26-21987, 2026.

Seismic full waveform inversion (FWI) is a powerful technique that uses seismic waveform data to generate high resolution images of the Earth's interior. However, significant uncertainties exist in FWI solutions due to imperfect acquisition geometries, inherent noise in the data, and the nonlinearity of the forward problem. Probabilistic Bayesian FWI estimates the family of all possible model solutions and quantifies their uncertainties by calculating the so-called posterior probability density function (pdf) of model parameter values of interest. In a linearised framework, the posterior pdf can be represented as a Gaussian distribution centred around the maximum a posteriori (MAP) solution, and the associated uncertainties are described by an a posteriori covariance matrix derived from the inverse Hessian matrix. Recent advancements have introduced nonlinear methods, such as variational inference, to solve Bayesian FWI problems efficiently. Their solutions quantify full uncertainties including those created by the nonlinearity of the problem. In this study, we apply both linearised and fully nonlinear methods to 2D acoustic Bayesian FWI problems. In particular, we use a physically structured variational inference algorithm for the nonlinear case, in which a transformed Gaussian distribution is optimised to approximate the full posterior pdf, such that the results can be compared fairly with those from the linearised, Gaussian-based method. We also employ an independent nonlinear variational algorithm – Stein variational gradient descent – for validation. The results show that while both linearised and nonlinear methods adequately recover the posterior mean models, they exhibit significantly different posterior uncertainty structures, especially at layer interfaces, due to the linearisation of wave physics. In addition, we show that linearised uncertainties are inaccurate since they can not fit observed waveform data and they yield biased estimates of inferred meta-properties such as volumes of geological bodies. This work therefore justifies the application of fully nonlinear inversion methods in Bayesian FWI if accurate uncertainty estimates are needed.

How to cite: Zhao, X. and Curtis, A.: Uncertainty Quantification in Full Waveform Inversion: Linearised versus Fully Nonlinear Methods, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-780, https://doi.org/10.5194/egusphere-egu26-780, 2026.

Tomography models of the inner core commonly assume a model of anisotropy that is transversely isotropic with a fast direction parallel to the Earth's rotation axis.  However, decades of debate have proposed various forms of this anisotropy, including the presence of a distinct innermost inner core or rotations of the direction of the fast axis.  These assumputions have yet to be directly compared to determine which is best supported by the available data.  Bayesian model comparison via the Bayesian Evidence provides this assessment but has historically been difficult to calculate, particularly in high-dimesional settings, and thus largely ignored in seismic tomography. New machine learning-based techniques can now be used to estimate the Evidence with greater stability and less uncertainty. In this work I demonstrate various methods, including the Savage-Dickey Density Ratio, the learnt harmonic mean estimator and trans-conceptual sampling, applied to the comparison of inner core anisotropy tomography models.

How to cite: Marignier, A., Lambert, B., Sambridge, M., and Koelemeijer, P.: Bayesian Model Comparison of Inner Core Anisotropy Models, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1022, https://doi.org/10.5194/egusphere-egu26-1022, 2026.

Solving inverse problems allows for the estimation of system properties or model parameters that cannot be measured directly. However, the models arising from various experts differ more than their individual uncertainty estimates might suggest [1]. A crucial reason for this is the combined impact of many small subjective choices undertaken during the inversion procedure. This includes a) the selected subset of data, b) the model space parametrisation, c) the type of forward model, d) the chosen numerical and optimisation methods, and e) regularisation.

In this work, we present an adapted Bayesian inference method that explicitly incorporates these subjective choices – collectively referred to as the control parameters – as a random variable to obtain estimates of ensemble statistics. It can be shown that, while the ensemble model m^e may be computed simply by averaging over N individual models m_i (i = 1, 2, ..., N), the ensemble covariance C^e consists of a sum of two terms,

The first term represents the mean of individual posterior covariances C_i, and the second term represents the variance of the mean models.

The theoretical developments are illustrated with a novel small-scale "Community Monte Carlo" experiment, where a group of experts was asked to select suitable regularisation (tuning) parameters to obtain a solution to a linear straight-ray tomography problem. The regularisation parameters include, e.g., the data prior standard deviation σ_D and the model prior standard deviation σ_M.

Crucially, the computation of ensemble estimates reveals that the average of individual covariances – the first term in eq. (1) – dominates the ensemble covariance. This is due to individuals favouring smaller values of σ_M, resulting in similar-looking models that deviate minimally from the prior, and larger data errors σ_D, leading to comparatively large posterior covariance matrices tending towards the prior model covariance.

Our proof-of-concept suggests that the field of seismic tomography should not strive for consensus among models, which risks condensing the ensemble and producing overly optimistic uncertainty estimates. Instead, the diversity of expert-derived models can be seen as an opportunity for "Community Monte Carlo," emphasising the need to actively explore a broader range of plausible subjective choices and rigorously quantify their effect on model uncertainty.

References:

[1] Andreas Fichtner, Jeroen Ritsema, Solvi Thrastarson; A high-resolution discourse on seismic tomography. Proc. A 1 August 2025; 481 (2320): 20240955. https://doi.org/10.1098/rspa.2024.0955

How to cite: Kaplunov, N. and Fichtner, A.: A Community Monte Carlo Approach for Quantifying Subjectivity-Driven Ensemble Uncertainty in Inverse Problems , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1642, https://doi.org/10.5194/egusphere-egu26-1642, 2026.

Over the past several decades Trans-dimensional Bayesian sampling has been widely applied in the geosciences. Most implementations have used the Reversible-jump Markov chain Monte Carlo (Rj-McMC) algorithm. This approach allows sampling across variably dimensioned model parameterizations and hierarchical noise models. Due to practical limitations Reversible-Jump is restricted to cases where the number of free parameters changes in a regular sequence, usually by addition or subtraction of a single variable. Furthermore, jumps between model dimensions rely on bespoke mathematical transformations that are only valid within a particular parametrization class. As a result, the range of model classes that can be practically considered is limited, and McMC balance conditions must be rederived for each class of problem. A framework for Trans-conceptual Bayesian sampling, which is a generalization of trans-dimensional sampling, is presented. Trans-C Bayesian inversion allows exploration across a finite, but arbitrary, set of conceptual models, i.e. ones where the number of variables, the type of model basis function, nature of the forward problem, and even assumptions on the class of measurement noise statistics, may all vary independently.

A key feature of the new framework is that it avoids parameter transformations and thereby lends itself to development of automatic McMC algorithms, i.e. where the details of the sampler do not require knowledge of the parameterization details. Algorithms implementing Bayesian conceptual model sampling are illustrated with examples drawn from geophysics, using real and synthetic data. Comparison with reversible-jump illustrates that trans-C sampling produces statistically identical results for situations where the former is applicable, but also allows sampling in situations where trans-D would be impractical, including asking the data to choose between competing forward models.

How to cite: Sambridge, M., Valentine, A., and Hauser, J.: Trans-Conceptual Inversion: Bayesian Inference with Competing Assumptions, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2663, https://doi.org/10.5194/egusphere-egu26-2663, 2026.

Inference problems within the geosciences vary considerably in terms of size and scope, ranging from the detection of changepoints in 1D time/depth models, to the construction of complex 3D or 4D models of the Earth. Solving an inverse problem typically requires fusing various classes of data, each associated with its own forward model. The choice of an appropriate inference method is itself not obvious. An investment of much time effort and is required in software development and education. Many researchers have developed bespoke inversion and parameter estimation algorithms tailored to their specific needs. Associated software is then typically bespoke to the particular application, often requiring significant investment by new researchers to master with minimal documentation. This is entirely understandable as generalisation and ongoing support of inference codes requires significant time and effort that is frequently beyond the primary objectives of the research. As a result the all important experimentation often required to choose an appropriate inversion method for a new data set or domain, is often not practical. Furthermore design choices made in existing software implementations often dictate those by subsequent researchers and influence the scientific direction taken.

The Common Framework for Inference, CoFI, is an open source project which aims to capture inherent commonalities present in all types of inverse problems, independent of the specific methods employed to solve them. CoFl is codifies the definition of an inference problem and then provides an interface to reliable and sophisticated third-party packages, such as SciPy and PyTorch, to tackle inverse problems across a broad range. The modular and object-oriented design of CoFI, supplemented by a comprehensive suite of tutorials and practical examples, ensures its accessibility to users of all skill levels, from experts to novices. This not only has the potential to streamline research and promote best practice but also to support education and STEM training. This poster gives an overview of CoFl through domain relevant examples, from optimisation to probabilistic sampling.  With a focus on CoFI’s modular approach we hope to foster collaboration centred around interaction by expanding the set of inference algorithms and domain-relevant examples.

How to cite: Sambridge, M., He, J., Chaivannacoopt, K., Hauser, J., Koch, M., Magrini, F., Marignier, A., and Valentine, A.: CoFI - The Common Framework for Inference: A software platform for experimentation, education and application of geophysical inversion tools, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3644, https://doi.org/10.5194/egusphere-egu26-3644, 2026.

A fundamental challenge in inverse problems is non-uniqueness: many models may fit a given data set exactly, or within observational uncertainty. A common remedy is the introduction of additional constraints encoding prior beliefs about the true model. Such explicit regularization mechanisms typically receive the most attention in inverse-problem research. However, it is well known that inversions may also be influenced by implicit sources of regularization. Discretization is perhaps the most prominent example, often introducing unintended and opaque prior assumptions.

Discretizations are commonly adopted for computational convenience and to avoid the theoretical complexities associated with models defined as functions. This practice, if done too early, can obscure fundamental questions concerning probabilities, regularity, and boundary behavior of the model. Although discretization may appear to eliminate these difficulties, it in fact makes choices for us that are often left unexamined.

In this contribution, we demonstrate that for linear(ised) problems in seismology, the undiscretized formulation can be treated rigorously using well-established theoretical tools. This perspective exposes hidden assumptions embedded in standard inversion workflows and allows prior choices to be made explicit and transparent. Although discretization is unavoidable in practice, we show that how and when it is introduced plays a crucial role, both for ensuring correct convergence and for computational efficiency.

Rather than attempting a fully rigorous solution of infinite-dimensional inverse problems—which can be expensive—we focus instead on probabilistic linear inference. Unlike classical inversion, linear inference targets specific properties of the model rather than a particular model realization. These quantities of interest are exactly representable in finite-dimensional spaces without discretizing the model itself. As a result, our framework delivers complete and mathematically consistent answers at reduced computational cost. We illustrate the proposed approach with synthetic inversion and inference examples in 1D.

How to cite: Mag, M. A., Al-Attar, D., Koelemeijer, P., and Zaroli, C.: Think first, discretize later, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5013, https://doi.org/10.5194/egusphere-egu26-5013, 2026.

The Earth’s mid-mantle (800–1200 km depth) hosts enigmatic seismic discontinuities whose physical origins remain debated. Competing hypotheses attribute these features to either thermal anomalies, such as partial melting due to volatile transport, or compositional heterogeneities associated with ancient subducted crust. However, distinguishing between these scenarios remains a challenge; standard imaging techniques often fail to robustly resolve the polarity of weak seismic reflections amidst noise and reverberations that contaminate mid-mantle reflections. Consequently, previous global surveys relying on linear time-domain stacking have yielded only a fragmented perspective, where the global connectivity, topography, and distinct physical origin of mid-mantle discontinuities remain debated. To address these limitations, we present a new global imaging framework that integrates curvelet-based wavefield separation and deconvolution, with probabilistic array processing. Rather than relying on traditional linear stacking, we develop "probabilistic vespagrams" that rigorously account for uncertainties in signal coherence and wavelet estimation. This approach allows us to distinguish robust structural features from processing artifacts. We apply this workflow to a global dataset of SS and PP precursors to construct a probability map of mid-mantle discontinuities. By systematically quantifying the likelihood of positive (eclogitic/compositional) versus negative (thermal/melt) impedance contrasts globally, we aim to resolve the global distribution of mid-mantle heterogeneities and determine the relative dominance of compositional stratification versus partial melting by water transport in controlling deep-Earth dynamics

How to cite: Carr, S. and Olugboji, T.: Earth’s Mid-mantle via Probabilistic Array Imaging , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7770, https://doi.org/10.5194/egusphere-egu26-7770, 2026.

Geoscientists often solve inverse problems to estimate values of parameters of interest given relevant data sets. Bayesian inference solves these problems by combining probability distributions that describe uncertainties in both observations and unknown parameters, and we require that the solution provides unbiased uncertainty estimates in order to inform evidence- or risk-based decisions. It has been known for over a century that employing different, but equivalent parametrisations of the same information can yield conditional probabilities that are mathematically inconsistent, a property referred to as the BK-inconsistency. Recently this inconsistency was shown to invalidate the solutions to physical problems found using several well-established methods of Bayesian inference. This talk explores the extent to which this inconsistency affects solutions to common geophysical problems. We demonstrate that changes in parametrisations result in inconsistent conditional prior probability densities, even though they represent exactly the same prior information. These inconsistent prior distributions can change Bayesian posterior solutions dramatically across various geoscientific problems including seismic impedance inversion, surface wave dispersion inversion, and travel time tomography, using real and synthetic data. Significantly different posterior statistics are obtained, including for maximum a posteriori (MAP) solutions, mean estimates, standard deviations, and full posterior distributions. Given that deterministic inversion is often equivalent to finding the MAP solution to specific Bayesian problems (the mathematical equations to be solved are identical), the BK-inconsistency also results in inconsistent solutions to deterministic inverse problems. Indeed, we show that solutions can potentially be designed, simply by changing the parametrisation. This study highlights that a careful rethinking of Bayesian inference and deterministic inversion may be required in physical problems, and we present one possible consistent method of solution.

How to cite: Curtis, A., Mosegaard, K., and Zhao, X.: Reverend Bayes, we have a problem - and a solution, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11966, https://doi.org/10.5194/egusphere-egu26-11966, 2026.

Reliable evaluation of three-dimensional seismic velocity models is critical for wave-propagation simulations and seismic hazard assessment, yet remains challenging in tectonically complex regions where source-related uncertainties and strong lateral heterogeneity limit conventional validation approaches. Here we present a propagation-centered framework for velocity-model evaluation based on reverse-time source-point gathers constructed from ambient-noise–derived surface waves.

Empirical Green’s functions are retrieved from ambient-noise cross correlations and reverse-time propagated under candidate velocity models to virtual source locations. If a velocity model adequately captures the kinematic characteristics of wave propagation, back-propagated wavefields from different azimuths and offsets refocus coherently near the zero-time reference. Systematic time shifts in the reverse-time source-point gathers indicate kinematic inconsistencies and reveal velocity biases. We quantify this behavior using the arrival-time deviation, Δt, and analyze its dependence on period, offset, and azimuth.

We apply this framework to evaluate four recently developed three-dimensional S-wave velocity models in the Sichuan–Yunnan region, southwest China. The results reveal clear model-dependent patterns in back-propagated arrivals across multiple period bands. At short periods (5–10 s), the back-propagated arrivals are more scattered in time and exhibit stronger directional variability, indicating that the tested velocity models still have limited accuracy in representing shallow structures and local-scale heterogeneity. In contrast, at longer-period waves (15–45 s), the back-propagated wavefields refocus more coherently at the virtual source, with arrival times clustering closer to zero, suggesting that the models are able to reproduce the large-scale characteristics of wave propagation more consistently. Consistent trends observed across multiple virtual source locations highlight both regional-scale performance differences and azimuth-dependent kinematic biases among the tested models.

The proposed reverse-time source-point gather approach offers a source-robust and physically intuitive perspective for velocity-model evaluation. By emphasizing kinematic self-consistency of wave propagation rather than detailed waveform matching, this framework complements existing evaluation methods and provides a flexible tool for diagnosing the strengths and limitations of three-dimensional velocity models in structurally complex regions.

How to cite: Li, T., Li, X., Wang, X., Zhang, W., Liu, Y., and Yao, H.: A reverse-time source-point gather framework for evaluating 3-D crustal velocity models using ambient-noise surface waves, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21245, https://doi.org/10.5194/egusphere-egu26-21245, 2026.

Full-waveform inversion (FWI) is capable of providing high-resolution Earth models, but quantifying uncertainty in these models remains a challenging and costly endeavour. The computational limitations of FWI mean that practical uncertainty estimates are necessarily based on highly incomplete information. This makes significant the difference between aggressive approaches, which systematically underestimate uncertainty, and conservative approaches, which systematically overestimate it. Here, we investigate an approach for inexpensive, conservative uncertainty quantification for high-dimensional FWI problems [1].  

This uncertainty quantification strategy is based on truncated singular value decomposition of the inverse problem Hessian. It takes as input a set of model and gradient pairs, which can, but do not have to be the inversion update history. This machinery can be used for both standard deviation estimation and hypothesis testing, using a targeted nullspace shuttling approach. In addition to its flexibility,  comparatively low cost and large-problem scaling, a key advantage of this approach is its conservativism; it provides a guarantee that the estimated uncertainty is greater than that which would be achieved with a full-rank Hessian estimate.

[1] Keating, S., Zunino, A., & Fichtner A, 2026. A comparison of rank-reduction strategies for uncertainty estimation in full-waveform inversion. Accepted for publication in Geophysical Journal International

How to cite: Keating, S., Zunino, A., and Fichtner, A.: Rank-reduction based standard deviation estimation and shuttling for FWI, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21968, https://doi.org/10.5194/egusphere-egu26-21968, 2026.

Over the past two decades, seismic hazard modeling has advanced along two complementary frontiers: empirical probabilistic frameworks, which systematically capture uncertainty through statistical inference, and physics-based simulation platforms, which directly compute ground motions from the governing equations of wave propagation. This project seeks to unify these two worlds by developing an end-to-end integration between OpenQuake and CyberShake, thereby creating a new generation of seismic hazard models that are globally extensible, probabilistically complete, and physically consistent. CyberShake has been under active development for more than a decade, demonstrating its robustness and scientific maturity through extensive implementations in California. It performs a physics-based probabilistic seismic hazard analysis (PSHA), replacing traditional empirical Ground Motion Prediction Equations (GMPEs) with full 3D numerical simulations of seismic wave propagation. Built upon the UCERF2/3 Earthquake Rupture Forecasts, CyberShake computes hazard curves directly from synthetic seismograms generated via Strain Green’s Tensors and thousands of stochastic rupture variations. This approach enables non-ergodic, site-specific hazard estimation and has set a global benchmark for high-fidelity hazard computation. However, its application has remained geographically limited: both the ERF and 3D velocity models were designed specifically for California, requiring extensive datasets that are rarely available elsewhere. Conversely, OpenQuake, developed by the Global Earthquake Model (GEM) Foundation, provides a fully open-source, Python-based framework for probabilistic seismic hazard and risk analysis. It serves as the computational backbone of large-scale hazard models such as the European Seismic Hazard Model 2020 (ESHM20), which integrates decades of regional expertise into a unified and statistical representation. OpenQuake provides a complete probabilistic framework to build Earthquake Rupture Forecasts (ERFs) that combine declustered catalogs, background seismicity, and multi-branch logic trees, ensuring a balanced and uncertainty-aware representation of regional tectonics. Furthermore, its ecosystem extends seamlessly to vulnerability and exposure modules, enabling the translation of hazard into actionable risk assessments and resilience planning.

This project will establish a direct pipeline from OpenQuake’s event-based results to the generation of an ERF compatible with CyberShake’s simulation framework, ensuring moment–rate consistency. By doing so, it will enable CyberShake simulations to be performed for regions beyond California, extending its use to Europe based on the knowledge contained in the ESHM20. The first pilot region is Istanbul, Turkey, a densely populated metropolis located near the western termination of the North Anatolian Fault. Our initial results show that the workflow is already functioning at the prototype level: we have developed a unified 3D velocity model for the Istanbul region by combining available tomographic models with local datasets; generated preliminary event-based rupture catalogs from ESHM20 using OpenQuake; and demonstrated early convergence behavior in hazard curves, indicating that the rupture sampling strategy is statistically robust. These initial results demonstrate the feasibility of the integration approach and indicate that the essential elements needed for a CyberShake-ready ERF are already in place.

How to cite: Riaño Escandon, A. C., de la Puente, J., Danciu, L., and Callaghan, S.: Adapting CyberShake for Europe using OpenQuake-Derived Earthquake Rupture Forecasts, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-540, https://doi.org/10.5194/egusphere-egu26-540, 2026.

The northern branch of the North Anatolian Fault (NAF), the Main Marmara Fault (MMF), constitutes one of the most critical seismic hazards in the Eastern Mediterranean. This system currently hosts an ~120-km seismic gap bounded by the Mw 7.4 1912 Ganos and Mw 7.4 1999 İzmit earthquakes, and most recently accommodated the Mw 6.2 April 23, 2025 Marmara Sea Earthquake. The 2025 event ruptured the Kumburgaz segment, a key structural transition zone between the partially creeping Central Marmara Basin to the west and the fully coupled Çınarcık Basin to the east. Given the ~260-year seismic quiescence along this region of the MMF, understanding how the 2025 earthquake, together with the 1912 and 1999 events, has modified the regional stress field is essential for evaluating the likelihood and characteristics of a future large Marmara Sea earthquake.

In this study, we construct three complementary quasi-static block models to quantify stress evolution along the MMF: (1) a cumulative coseismic stress transfer model incorporating the 1912, 1999, and 2025 earthquakes; (2) a coseismic model isolating the effects of the 2025 rupture; and (3) an interseismic loading model constrained by GNSS observations. The two models enable a comparative assessment of static Coulomb stress changes on adjacent fault segments, illuminating how recent and historical ruptures collectively influence present-day stress accumulation patterns.

Building upon the quasi-static results, we generate new 3D dynamic rupture simulations using a 1D crustal velocity structure for the nonplanar multi-segment MMF, explicitly incorporating interseismic stress loading, coseismic stress perturbations, and the partially creeping behavior of the MMF. We further benchmark these new simulations against our earlier dynamic models that assumed a homogeneous velocity structure to evaluate the sensitivity of rupture dynamics to crustal heterogeneity and initial stress conditions.

Our integrated modeling framework reveals that, during a potential future large Marmara earthquake, rupture is likely to propagate westward through multiple MMF segments, while arresting near the eastern entrance of the İzmit Fault. New segmented rupture patterns are also observed as a result of using a 1D crustal structure instead of a homogeneous medium, together with the inclusion of coseismic stress transfer. The findings offer important insights into post-2025, post-1999, and post-1912 stress redistribution, fault-segment interactions, and rupture cascade potential across the Marmara region. Collectively, this work advances the scientific basis for earthquake hazard assessment in one of the world’s most densely populated and tectonically active metropolitan corridors.

How to cite: Korkusuz Öztürk, Y., Konca, A. Ö., and Meral Özel, N.: Integrated Stress Evolution and Multi-Segment Rupture Dynamics of the Main Marmara Fault After the 2025 Mw6.2 Marmara Sea Earthquake, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1665, https://doi.org/10.5194/egusphere-egu26-1665, 2026.

Reservoir-induced seismicity (RIS) is a critical concern in geo-engineering, arising from the coupled interactions among in-situ stress, fluid flow, and fault mechanics, associated with reservoir impoundment. Improving our understanding of earthquake dynamics is therefore essential for elucidating the dynamics of rupture processes at RIS. In particular, understanding fault reactivation and the transition from quasi-static aseismic slip to dynamic rupture is crucial, as the nucleation phase may provide valuable information for detecting pre-seismic signals and estimating earthquake magnitudes.

We develop a novel two-dimensional, fully coupled poro-visco-elasto-dynamic finite-element model (implemented in COMSOL) to simulate RIS under reservoir impoundment in extensional tectonic settings. The porous medium is represented as a Kelvin–Voigt poro-visco-elastic solid to capture elastic deformation and intrinsic damping, while inertial effects are included to resolve rupture dynamics and seismic wave propagation. The fault is modeled as  non-penetrating surfaces enforced using an augmented Lagrangian contact formulation and governed by rate-and-state friction, where fault deformations are tolerated by using a virtual thin layer capability.

Model results show that when frictional and hydromechanical conditions permit fault reactivation, slip may become unstable and transition into a coseismic event, with rupture propagating along the fault in asymmetric two–crack-tip–like slip pattern emanating from the hypocenter. Rupture propagation speed is higher in the stiffer rock than in the softer one. Preferential flow induced by the reservoir impoundment forces the rupture nucleation earlier. Porosity and permeability of the fault damage zone decrease with depth (higher than that of the ambient rock at the upper part of the fault), providing the conduit for fluid flow over the fault and promoting longer rupture lengths at RIS.

These findings highlight the critical role of mechanical and hydraulic properties in controlling nucleation and rupture processes in RIS, with important implications for the design and management of reservoir impoundment.

How to cite: Zhou, X. and Katsman, R.: Reservoir Induced Seismicity Modelled Using a Fully Coupled Poro-Visco-Elasto-Dynamic Model with Frictional Contact and Rate-and-State Dependent Friction: Dynamics of Spontaneous Coseismic Rupture, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2242, https://doi.org/10.5194/egusphere-egu26-2242, 2026.

Precursory signals preceding large earthquakes are commonly attributed to the acceleration of localized slip during rupture nucleation, yet their spatial expression in the surrounding medium remains poorly constrained. Here, we model the evolution of off-fault strain during earthquake nucleation governed by rate-and-state friction. Our results show that strain accumulates gradually during the early nucleation phase and then accelerates sharply, exceeding a threshold of ε ~ 10^-7—comparable to natural strain levels and detectable by modern strainmeters and geodetic instruments—tens to hundreds of days before instability, depending on the uncertainty in the characteristic slip distance D_c and effective normal stress σ_eff. Approximately 0.7 times the nucleation duration prior to failure, the strained region (ε > 10^-7) extends to distances exceeding one nucleation length away from the fault and spans most of its entire length. We further show that σ_eff  controls both the magnitude and spatial distribution of strain, whereas D_c primarily influences the spatial extent of the strained region. Assuming a representative value for the sensitivity of seismic velocity changes to strain (η ≈ 10⁴), the predicted strain amplitudes correspond to ~0.1%-100% changes in seismic velocity, well above the detection limits of ambient-noise monitoring. A comparison between strain footprints and seismic wavelengths further suggests that analysis of short-period noise (T = 0.1 - 1 s) would be most favorable for identifying these precursory signals. Together, these findings directly link nucleation theory to observable field-scale precursors and provide a physics-based framework for precursor identification in natural fault systems.

How to cite: Zhang, L., Ampuero, J.-P., and Romanet, P.: Linking off-fault strain to rate-and-state friction nucleation: Implications for monitoring precursory velocity changes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2376, https://doi.org/10.5194/egusphere-egu26-2376, 2026.

Physics-based approaches are increasingly recognized as essential for improving seismic hazard assessment, however, no fully physics-based probabilistic seismic hazard analysis (PSHA) exists for the Italian territory. This gap is particularly relevant in the Po Plain area in northern italy, where deep sedimentary deposits strongly amplify seismic waves and prolong shaking, even for moderate-magnitude events. In this context, broadband ground-motion simulations represent a key requirement for capturing both long-period basin effects and high-frequency scattering. In this study, we generate synthetic seismograms spanning the engineering-relevant 0.1–10 Hz bandwidth using a hybrid approach that combines deterministic low-frequency (<1 Hz) simulations with stochastically generated high-frequency (1–10 Hz) ground motion. The low-frequency component (<1 Hz) is computed using the SPECFEM3D Cartesian code, which implements the spectral element method to solve the full seismic wave equation in complex 3D media. A central goal of this work is the validation of the 3D MAMBo velocity model (Molinari et al., 2015). We test the model using several earthquakes and compare its performance against alternative candidate 1D and 3D velocity models, highlighting the critical role of a detailed 3D representation of basin geometry and major velocity discontinuities. The synthetic seismograms are quantitatively evaluated using time–frequency misfit and goodness-of-fit metrics. Our results show that the 3D characterization significantly improves the agreement with observed waveform shapes and durations, and they provide a foundation for future refinement of the regional velocity model. The resulting broadband synthetics are suitable for seismic-hazard analysis and engineering applications in the densely populated and economically important Po Plain. Overall, this study outlines a pathway toward fully physics-based probabilistic seismic hazard analysis (PSHA) in northern Italy, grounded on validated 3D structure and physics-based broadband ground-motion simulations.

How to cite: Saturnino, C., De Siena, L., and Molinari, I.: Toward physics-based PSHA study in northern Italy: 3D velocity model validation and broadband seismic signals synthesis., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2980, https://doi.org/10.5194/egusphere-egu26-2980, 2026.

The 2016 Mw7.8 Kaikoura earthquake presents one of most challenging natural events to model dynamically, with up to 21 faults involved in the full rupture, according to geological measurements of surface rupture ( e.g. Litchfield et al. 2018). However, most studies using static displacement observations do not resolve individual fault activation and their temporal connectivity at some parts of the fault range (e.g. Hamling et al. 2017), as suggested by the near-source strong motion data (REF). A more recent complete aftershock catalog provides improved seismological constraints on the rupture kinematics, offering new insights into the fault geometry and faulting mechanisms (Chamberlain et al. 2021). These advances motivate a re‑examination of the mysterious multi-fault rupture with complete seismological observation and physics-based dynamic rupture modeling for to better understand the governing mechanisms of multi-fault ruptures.

Compared to kinematic source inversions, dynamic modeling is a powerful numerical tool to compute realistic cases of earthquake occurrence due to complex ruptures. Yet, for earthquakes involving multiple interacting faults, even state-of-the-art dynamic models can lead to fundamentally different physical interpretations. On one hand, the corresponding dynamic modeling setup heavily depends on prior knowledge of the full system geometry, as well as on the stress-state and velocity model of the medium. On the other hand, due to the nonlinear nature of the problem, several models can produce similar results. Results show that for relatively simple ruptures, involving one or two fault planes, the solution is stable. However, when the rupture involves several faults, even minor changes to the dynamic setup result in instability and non-uniqueness of the solution.

To gain insight into how such extreme fault complexity controls rupture evolution, we perform the dynamic modeling of the 2016 Mw7.8 Kaikoura earthquake using the open-access SeisSol package. We use the New Zealand 3D velocity model and compare two different geometries. The first geometry uses the NZ Community Fault Model, while the second is based on a previously published rupture model (Ando & Kaneko 2018). For the first geometry, we analyze whether the rupture actually used secondary faults to continue its path, or if subsequent rupture was triggered by the generated wavefield. For the second geometry, we investigate the impact of rupture bifurcation onto two faults and assess whether this process generates identifiable seismic phases in the wavefield.

We analyze both dynamic scenarios using near-field and regional strong-motion records, which are expected to capture hidden features of the rupture. We further compare the simulated rupture evolution with previously published high-resolution earthquake catalogs to identify rupture patterns and evaluate potential changes in the stress field before and after the event. Our results highlight both the strengths and inherent ambiguities of dynamic rupture modeling for complex multi-fault earthquakes and provide new constraints on the physical processes governing the Kaikoura rupture.

How to cite: Caballero-Leyva, E., Li, D., Ando, R., and Benites, R.: Impact of Fault Geometry in dynamic modeling simulations: The case of the 2016 Mw7.8 Kaikoura., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3216, https://doi.org/10.5194/egusphere-egu26-3216, 2026.

This study investigates rupture directivity effects on source spectra of small-magnitude earthquakes in Central Italy, based on a dataset comprising 18,994 waveforms from 656 shallow crustal events recorded between 2008 and 2018. The Generalized Inversion Technique (GIT) is employed to isolate frequency-dependent source characteristics. Apparent Source Spectra (AppSS) exhibit clear azimuthal variations, indicating the presence of directivity effects, particularly in events associated with higher standard deviations. The source spectra are analyzed using multiple empirical models, allowing for the estimation of seismic moment and stress drop for 138 events. Model performance is evaluated through residual analysis across a frequency range of 0.5–25 Hz. Our findings indicate that the ω² source model fitting on the plateau (ωest²) provides a better fit to the observed spectra for the selected events in the dataset. Comparison with previous studies confirms the reliability of the spectral estimates and modeling approach. For the two selected events, spatial maps of ground motion are presented, offering valuable insights into the regional variability of shaking. The study results underscore the importance of incorporating rupture directivity in ground motion models, thereby reinforcing the robustness of empirical predictive approaches and their relevance for improving seismic hazard assessments.

How to cite: Xhafaj, E., Vitrano, L., Pacor, F., Sgobba, S., and Lanzano, G.: Characterizing Earthquake Rupture Directivity Using Apparent Source Spectra: A Case Study from Central Italy, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3583, https://doi.org/10.5194/egusphere-egu26-3583, 2026.

UrgentShake is an urgent computing system developed by OGS (National Institute of Oceanography and Applied Geophysics) for the rapid generation of physics-based ground shaking scenarios. It employs a distributed architecture across High-Performance Computing (HPC) and cloud infrastructures to perform numerical simulations in near real-time, providing reliable estimates of ground motion following significant seismic events in Northeastern Italy, thereby supporting decision-making by emergency management authorities.

Although primarily designed for rapid response to earthquakes, UrgentShake’s flexible architecture also makes it suitable for non-real-time applications, such as Civil Protection exercises and risk analyses. In these contexts, a single realization of a specific seismic source is not sufficient; instead, a suite of plausible scenarios is needed to define median, minimum and maximum estimates of ground shaking and potential impacts.

To address this need, a feasibility study was conducted to demonstrate the potential integration of UrgentShake with CyberShake, a physics-based platform for seismic hazard modeling that simulates many rupture scenarios. CyberShake simulations for a representative earthquake scenario were performed using the Graves and Pitarka stochastic rupture generator and the Anelastic Wave Propagation code on HPC resources at the Barcelona Supercomputing Centre. By generating multiple independent source realizations with varying nucleation points, fault geometries and rupture characteristics, this proof of concept illustrates how source-related uncertainties can be incorporated into UrgentShake to produce robust ground shaking scenarios. These scenarios can support Civil Protection training and preparedness activities while enabling physics-based damage assessments to inform risk analyses.

How to cite: Zuccolo, E., Zamora, N., and Scaini, C.: UrgentShake for Scenario-Based Ground Motion Simulations: Integrating Multiple Source Realizations with CyberShake, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4273, https://doi.org/10.5194/egusphere-egu26-4273, 2026.

Fault geometrical complexity is a first-order controlling factor on the extent of strike-slip fault surface rupture and earthquake magnitude, and step-over represents a key type of such complexity. The Banquan pull-apart basin along the Tanlu fault zone provides a natural example to investigate how tectonically evolved fault geometry influences dynamic rupture propagation across step-overs. We construct a 3-dimensional fault model that incorporates Y-shaped negative flower structure, connecting faults, and a sedimentary layer within the extensional step-over. The shallow fault geometry is constrained by surface geological observations, and the deep fault structure is informed by analogue experiments of pull-apart basin formation. Spontaneous coseismic dynamic rupture simulations are performed to examine the rupture behavior under these fault geometries. Our results show that when stress perturbation associated with stopping phases at the main fault termination is insufficient to trigger rupture on the secondary fault directly, the presence of connecting faults can act as a bridge to facilitate rupture propagation across the step-over. A deeper connecting fault can generate a stress shadow on the secondary fault, inhibiting local rupture propagation and potentially behaving as a barrier on the secondary fault, whereas shallow connecting faults have little influence on the rupture process. These findings provide insights into rupture jumping behavior in step-overs with similar fault structures and extend the existing interpretation of step-over triggering based on stopping phases with planar fault geometries. 

How to cite: Lu, Z. and Hu, F.: Effects of tectonic evolution informed fault geometry on dynamic rupture propagation across step-overs: A case study of the Banquan pull-apart basin, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4321, https://doi.org/10.5194/egusphere-egu26-4321, 2026.

Aseismic creep is widely recognized to influence earthquake rupture, but whether its role remains stationary in different earthquakes is poorly understood. In this study, we integrate GNSS/InSAR observations along the Xianshuihe fault in eastern Tibet and identify six aseismic creeping sections, which have been partially or fully involved in historical earthquakes. The creep exhibits spatiotemporal transient behavior. Using interseismic fault locking as a constraint, we performed 3D dynamic rupture simulations of the Xianshuihe fault. We demonstrate that aseismic creep exerts a dual role in earthquake rupture. On the stabilizing side, creeping sections terminate rupture propagation, with earthquakes that nucleate and are absorbed within the creeping zones further reinforcing their function as stable rupture barriers. Conversely, under favorable local stress conditions and modulated by transient aseismic slip migration and hypocenter location, creeping sections could promote rupture propagation, rendering their impact on rupture non-stationary in different earthquakes. These findings provide a plausible explanation for the pronounced variability of rupture segmentation and cascading on the geometrically simple Xianshuihe fault, and highlight the importance of incorporating both stabilizing and destabilizing effects of aseismic creep into seismic hazard assessments.

How to cite: Li, Y. and Shan, X.: Uncovering the stabilizing and destabilizing roles of aseismic creep in earthquake rupture, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4671, https://doi.org/10.5194/egusphere-egu26-4671, 2026.

Material interfaces play crucial role in forming seismic wavefield in local surface sedimentary structures and resulting free-surface motion. Multiple reverberations between the free surface and sediment-bedrock interface can lead to resonant amplifications and generation of local surface waves, and consequently to strong site effects of earthquakes.

It is therefore important to properly implement material interfaces in numerical modelling of seismic wave propagation and seismic motion. This has been well known for some time, and several approaches have been developed in variety of numerical methods.

The finite-difference (FD) method is still dominant method in numerical investigations of site effects of earthquakes. It applies relatively simple discretization in space to the material parameters and discretization in space and time to wavefield variables. Therefore, consequences of discretization must be analyzed in time, space, frequency and wavenumber domains.

Interestingly enough, the least attention has been paid to the wavenumber domain. Mittet (2017) and Moczo et al. (2022) recently demonstrated that, due to spatial discretization, a model of the medium must be wavenumber-limited by a wavenumber k smaller than the Nyquist wavenumber. Mittet (2021) and Valovcan et al. (2024) proved that the wavefield (numerically simulated or exact) in a medium limited by wavenumber k can only be accurate up to half this wavenumber. This has significant consequence for practical FD modelling of motion in realistic models of local structures.

We numerically demonstrate a perfect and unprecedented sub-cell resolution (capability to sense the position of interface within a grid cell) of FD modelling based on the wavenumber-limited medium using a finite spatial low-pass filter. The finding that it is possible to use a finite-length filter for wavenumber limitation of the medium is of key importance for the next development of the concept in terms of computational efficiency in modelling site effects.

How to cite: Kristek, J., Valovcan, J., Moczo, P., Kristekova, M., Mittet, R., and Galis, M.: Accuracy of the Finite-difference Modeling of Seismic Motion – Wavenumber Limitation of Medium, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5554, https://doi.org/10.5194/egusphere-egu26-5554, 2026.

The Gulf of Aqaba (GoA) fault system constitutes the southernmost segment of the Dead Sea Transform Fault (DSTF) and forms a ~180 km long left-lateral strike-slip plate boundary separating the Arabian Plate from the Sinai microplate.  As the most seismically active region of the Red Sea, the GoA has hosted multiple large historical earthquakes and poses significant seismic hazards to surrounding coastal communities. Increasing tourism activity and the infrastructural giga-project NEOM of the Kingdom of Saudi Arabia in the vicinity of the GoA, highlight the need for advanced seismic hazard assessment (SHA). However, the offshore nature of the fault system and limited availability of observational data complicate the efforts. 

To assess earthquake potential and seismic hazard in the region, we construct multiple realizations of three-dimensional, multi-segment fault models representing alternative configurations of the GoA fault system. We constrain variations in 3D fault geometry with  recent high-resolution multibeam imaging and local seismicity, while explicitly accounting for uncertainties in seismogenic depth, initial stress conditions, and fault roughness. Incorporating off-fault plasticity along with realistic topography and bathymetry, we perform dynamic rupture simulations with varying hypocenter locations to investigate mechanically plausible rupture scenarios and the resulting ground motions in the GoA. Our physics-based simulations show that all considered model uncertainties, especially the fault geometry, prestress condition and hypocenter location, can strongly influence rupture dynamics, cascading, and segment interactions, determining how and if rupture propagates across the multi-segment GoA fault system. Beyond characterizing earthquake potential on individual fault segments, the simulations indicate that events as large as Mw 7.6 are possible if rupture extends along the full north–south length of the fault system. The resulting synthetic ground motions show attenuation properties consistent with empirical ground motion models, but display highly heterogeneous spatial patterns, including strong rupture-directivity effects during subshear propagation and pronounced off-fault Mach-front amplification for supershear rupture that significantly enhance ground shaking in coastal communities along this narrow gulf. These results underscore the substantial seismic hazard posed by large, dynamically complex earthquakes in the Gulf of Aqaba region and highlight the value of physics-based simulations in enhancing and complementing seismic hazard assessments.

How to cite: Li, B. and Mai, P. M.: Physics-based assessment of earthquake potential and ground motions in the Gulf of Aqaba, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5925, https://doi.org/10.5194/egusphere-egu26-5925, 2026.

Earthquake rupture propagation speed is an essential source factor that largely controls hazard and risk. However, measuring rupture speeds of natural earthquakes is often challenging and ambiguous. Near-fault seismic waveforms (recorded within several km) are believed to have high capability for resolving rupture process. In this study, we probe the feasibility of using near-fault data signatures to directly infer rupture speeds in continental strike-slip earthquakes.

To thoroughly understand near-fault features, we synthesize the near-fault seismic waves for kinematic source models on a strike-slip fault under different rupture speeds in a 3D medium. We identify the dependence of velocity waveform and particle motion on rupture speed in both amplitude and shape. In addition, we compare our results with the analytical solution with steady-state constant rupture speed. The discrepancies between the kinematic model and the analytical model indicate the contribution of radiation from different configurations. With inspecting the near-fault dataset of eight M>7 strike-slip earthquakes, we find that instead of dealing with the velocity waveforms with multiple high-frequency spikes, the features of the particle motion shape are easier to identify. Then we apply the particle-motion-based criterion to identify signatures associated with supershear, subshear, and other complexities such as multiple rupture fronts and initial-stage rupture phase. Our study highlights the further application of near-fault seismic data in studying earthquake sources.

How to cite: Yao, S., Yang, H., Bhat, H., and Aochi, H.: Rupture Speed Signatures of Near-fault Particle Motion in Large Strike-slip Earthquakes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7716, https://doi.org/10.5194/egusphere-egu26-7716, 2026.

The Ryukyu subduction zone offshore eastern Taiwan possesses significant seismogenic potential, exemplified by the 1920 M8 earthquake. However, even to date, the scarcity of near-field data leaves the ground motion characteristics of such mega-earthquakes poorly constrained, posing a threat to seismic hazard assessments. To estimate potential ground motions in inland eastern Taiwan from future mega-earthquakes, we simulated an M8 scenario earthquake using characterized source models (CSMs) based on the "Recipe" procedure (Irikura and Miyake, 2011). We employed a 3-D finite-difference method to conduct 1,728 full-waveform simulations, incorporating kinematic fault-rupture parameters, including rupture directivity, rupture speed, source time function, and asperity distribution, along with two recent tomographic velocity models and topography. Synthetic waveforms generated at 4,950 virtual stations (about 1.5 km spacing) were analyzed using RotD50 spectral accelerations (SA) at 1, 3, and 5 s. Detailed analysis highlights two notable characteristics of the dataset: first, rupture speed and directivity primarily govern the spatial variability and intensity of ground motions; second, tests demonstrate that utilizing a Gaussian source time function with periods of 2, 5, and 9 s yields optimal performance for assessing SA at 1.0, 3.0, and 5.0 s, respectively. We further calculated non-ergodic terms based on the CH20 GMM (Chao et al., 2020). The patterns clearly delineate northeastern Taiwan's geological domains: high values in the Ilan area (SA 1.0 s) and Longitudinal Valley (SA 1.0, 3.0, 5.0 s), and low values in the Coastal Range. These patterns mirror the crustal velocity structure, highlighting the dominance of path effects over relatively weak source effects. Consequently, our extensive simulation datasets provide a foundation for refining current GMMs and facilitate the transition toward non-ergodic seismic hazard assessments, thereby improving the accuracy of ground motion predictions for future mega-earthquake scenarios in the region.

How to cite: Hsieh, M.-C., Sung, C.-H., and Yang, Y.-C.: 3-D Seismic Wave Simulations for Non-Ergodic Ground Motion Modeling: Source and Path Variability in an M8 Ryukyu Subduction Scenario, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8902, https://doi.org/10.5194/egusphere-egu26-8902, 2026.

Geological observations show that fault surfaces are complex at both large scales (fault segmentation) and small scales (surface roughness). These geometric complexities strongly influence earthquake rupture behaviour, including slip, rupture speed, rise time, and peak slip velocity. Understanding how these rupture parameters are related to each other is essential for improving understanding of earthquake rupture physics and for developing synthetic rupture models that reproduce realistic dynamic behaviour within kinematic frameworks. Although earlier studies have examined these correlations, the effect of small-scale fault roughness is still not well understood. Therefore, this study focuses on understanding how fault roughness affects correlations among rupture parameters.

To address this problem, we use the dynamic rupture dataset of Mai et al. (2018), which includes twenty-one rupture models with different roughness realizations, roughness amplitudes, and hypocentre locations. Because dynamic slip-velocity functions have complex shapes, we simplify them by fitting the regularized Yoffe function proposed by Tinti et al. (2005). From these fits, we extract key kinematic parameters. We then examine correlations among eight parameters: slip, peak slip velocity, acceleration time, rise time, rupture speed, strike, dip, and rake.

Our results show that slip is positively correlated with rise time, but it does not show clear correlations with other rupture or geometry parameters. Peak slip velocity is negatively correlated with both acceleration time and rise time, and positively correlated with rupture speed. Importantly, as fault roughness increases, the correlation between peak slip velocity and rupture speed becomes weaker. Acceleration time is also negatively correlated with rupture speed, and this correlation also decreases with increasing fault roughness. In contrast, the geometry parameters strike and dip do not show significant correlations with any rupture parameters. Overall, fault roughness mainly affects the relationships between only two pairs of rupture parameters, whereas the correlations among other parameter pairs are not strongly affected.

Our findings provide important constraints for developing synthetic rupture models that can generate realistic high-frequency seismic radiation consistently with radiation of dynamic ruptures propagating on rough faults.

How to cite: Vyas, P. K. and Galis, M.: Influence of Fault Roughness on Earthquake Rupture Parameters Correlations, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9624, https://doi.org/10.5194/egusphere-egu26-9624, 2026.

Sequences of Earthquakes and Aseismic Slip (SEAS) simulations focus on km scale models and consider all phases of faulting from aseismic slip to earthquake nucleation, propagation and termination. Many codes exist that use different approaches to tackle SEAS simulations; however, solutions designed to leverage the potential of GPUs to parallelize and speed up the simulation are limited (although recent examples are emerging such as PyQuake3D). In this work, we propose a GPU parallelized SEAS quasi-dynamic solver written in Julia adopting the Spectral Boundary Integral Method (SBIM). The SBIM approach is optimal for GPUs as it is more memory efficient in respect to other mesh-based solvers, thus enabling to efficiently run high resolution simulations with around 10 million nodes on the fault plane. We rearrange the rate-and-state equations to solve for the slip rate and adopt a slightly modified Newton-Raphson algorithm for root finding. We introduce elastic bulk by using an analytical stress-slip relationship in the Fourier/wavenumber domain. Most of the operations that are carried out in the solver are element wise and thus can be run in parallel on GPUs, significantly cutting down on computation time as the domain resolution increases. While FFT is inherently not fully parallelizable, GPU kernels are available to efficiently perform Fourier transforms on GPUs. Moreover, by using the FFT algorithm, the numerical complexity for calculating the stress is reduced from O(N²) to O(NlogN). To verify the correctness of our solver, we use the BP4-QD benchmark and show comparable results with other outputs hosted by SCEC. We then measure the runtime of the solver on CPU and 2 NVIDIA GPUs, the RTX4060 8GB and the A100 40GB, and show a x5 to x16 speedup for simulations depending on the GPU. Finally, we run the BP4-QD problem on the A100 GPU, decreasing the indicated node spacing and Dc values by an order of magnitude. This simulation yielded 36 events of Mw > 7 and 181 events of Mw between 5 and 6, showing emergence of complexity. Moreover, we observe that the earthquake’s nucleation points are distributed along the edges of the rate-weakening patch. The smaller events are mostly concentrated on the four corners and the two sides parallel to the slipping direction while the bigger events are distributed more uniformly all around the border.

How to cite: Benedetti, G. and Heimisson, E. R.: Accelerating and scaling up SEAS simulations using GPUs in Julia, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9901, https://doi.org/10.5194/egusphere-egu26-9901, 2026.

Earthquake cycle modeling has enabled to reproduce the full spectrum of slip rates observed along fault segments, and refine our understanding of seismic cycle dynamics. However, key parameters controlling the occurrence of shallow slow slip events (SSE) such as those observed along strike-slip fault segments remain unclear, due to rare worldwide observations and the lack of long-lasting observations covering all phases of the seismic cycle. Here, we apply rate and state friction quasi-dynamic 1D models to explain the ensemble of observations along the Izmit segment of the North Anatolian Fault in Türkiye. This fault segment ruptured in 1999 with the magnitude 7.6 Izmit earthquake, and has been since then widely studied, providing constraints on most of the phases of the seismic cycle, from mainshock amplitude and recurrence times to afterslip logarithmic decay, and the occurrence of shallow SSEs. GNSS, InSAR and creepmeters geodetic data associated with seismological and paleoseismological data enable to describe the cumulative displacement during all phases of the seismic cycle. The comparison between model predictions and the observational time scales led to an optimal set of frictional models. First, the mainshocks maximum slip of ~6 m and return times of about ≥200 yrs are explained by an unstable seismogenic layer below 5.5 km depth with a thickness of 9.5 km and with frictional parameters a-b of about -0.004. The decadal afterslip, well constrained by a pair of campaign GNSS stations located on both sides of the fault, is mostly due to a stable layer located between 5.5 and 1.3 km depth, the lower limit being compatible with the aftershocks sequence limit. We compared model slip predictions and GNSS time series by computing Green's functions for a layered elastic half space medium. Model parameters for this intermediate layer explaining the observed relaxation time have frictional parameters a-b and critical distance of about 0.005 and 8 km, respectively. Finally, a shallow layer from the surface to 1.3 km depth with either a gradient of frictional parameters with depth or constant negative frictional parameters is needed to generate shallow SSEs 20 yrs after the main earthquakes. The shallow layer depth extent being compatible with the Izmit Quaternary sedimentary basin may suggest a key role of the sediments frictional properties to allow a velocity weakening behavior. Models with a gradient of apparent frictional properties throughout the basin may suggest the importance of pore-pressure variations as a function of the fault gouge depth.

How to cite: Doubre, C., Estelle, N., Baptiste, R., Matt, W., and Yoshihiro, K.: Postseismic and shallow slow slip events on the Izmit segment of the North Anatolian Fault controlled by depth-dependent frictional variations, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10178, https://doi.org/10.5194/egusphere-egu26-10178, 2026.

This paper presents the quantification of site-city-interaction (SCI) effects on the dynamic response of buildings and free-field motion using domain reduction method. The simulated physics based broadband near-fault ground motion due to Mw6.5 strike-slip earthquake is being utilized to excite the site-city models using domain reduction method. The two step, domain reduction method, is utilized to reduce the exorbitant computational memory and speed as well as measures have been taken to preserve the ground motion characteristics. A building is incorporated in numerical grid as a building block model (BBM) and its dimension, different modes of vibrations and damping are as per the real building. The dynamic response of site-city models is simulated using both the pulse and non-pulse type motions. The analysis of simulated results reveals that the SCI study using realistic earthquake ground motion has caused a reduction of response of building and free field motion in a wide frequency bandwidth as well as its fundamental frequency. An increase of these reductions has been obtained with decrease of building-damping, fundamental frequency and impedance contrast between the BBM and the underlying sediment. A considerable difference in SCI effects is obtained when site-city model is excited with pulse and non-pulse type near-fault ground motions. Detailed study is carried out in order to find out the terms and conditions under which SCI is beneficial to all the buildings of the city.

How to cite: malik, S. and Narayan, J. P.: Quantification of site city interaction effect on Response of building in near fault region using Domain Reduction Method, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10205, https://doi.org/10.5194/egusphere-egu26-10205, 2026.

The Tianzhu seismic gap is an important segment of the Haiyuan fault system. In recent decades, earthquakes have occurred on most fault segments within this region, whereas the Jinqianghe–Maomaoshan fault has not experienced a major earthquake for an extended period. Given that this fault segment is widely regarded as having elevated potential seismic hazard, we conduct three-dimensional dynamic rupture and strong ground motion simulations using the curved grid finite difference method.To effectively constrain model input parameters, interseismic locking coefficients and slip deficit distributions inverted from InSAR and GPS observations are used to impose physically based constraints on the heterogeneous initial stress conditions along the fault. Simulation results indicate that the spatial distribution of locked regions plays a critical role in controlling rupture extent. Under locking-constrained conditions, scenario earthquakes with moment magnitudes of Mw 7.3–7.4 and maximum slip of approximately 5.5 m are generated. Further analyses show that larger accumulated slip deficits tend to promote higher earthquake magnitudes, whereas the surface seismic intensity does not exhibit a monotonic response to slip deficit.These results suggest that the Jinqianghe–Maomaoshan fault segment may be associated with elevated potential seismic hazard.

How to cite: Ren, B., Wang, W., and Qiu, H.: Dynamic Rupture and Ground Motion Simulations of Potential Earthquake on the Tianzhu Seismic Gap, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10413, https://doi.org/10.5194/egusphere-egu26-10413, 2026.

Seismic simulations are essential for ground motion characterization and seismic hazard mitigation. However, achieving accurate seismic modelling requires highly refined computational grids, which impose severe memory and computational challenges. Traditional seismic solvers based on single-precision floating-point 32-bit (FP32) arithmetic, suffer from excessive memory consumption, low-memory access efficiency and limited computational efficiency. In contrast, half-precision floating-point 16-bit (FP16) halves memory usage and effectively doubles memory access efficiency, making it attractive for large-scale seismic simulations. However, direct application of FP16 to classical elastic wave equations is challenging due to overflow and underflow caused by the wide dynamic range of physical variables. In this work, we reformulate the elastic wave equations by introducing three dimensionless scaling constants, Cv, Cs, and Cp, and derive an FP16-based elastic wave equation. Furthermore, we provided a practical strategy for determining these constants based on the source time function, ensuring that velocity and stress variables remain within the representable range of FP16. To maintain FP32-level accuracy, a mixed-precision strategy using “FP16 storage and FP32 arithmetic” is adopted. From a computational perspective, we further exploit the Scalable Vector Extension (SVE) on ARM architectures to accelerate stencil-based computations. However, effectively combining FP16 with SVE introduces additional challenges, including stencil restructuring for vectorization and data layout mismatches arising from “FP16 storage and FP32 arithmetic”. To overcome these challenges, this study develops three complementary seismic solvers on the ARM architecture: an FP16-based solver, an SVE-accelerated solver, and an FP16–SVE hybrid solver that integrates memory efficiency with vectorized computation. All three solvers are implemented, systematically validated, and benchmarked using both synthetic test cases and real earthquake simulations. Numerical results demonstrate near-identical agreement with a reference FP32 solver across diverse seismic scenarios. In particular, the FP16–SVE hybrid solver reduces memory consumption by approximately 50% and achieves up to a threefold speedup, delivering more than a 2.3× acceleration in real-world earthquake simulations. These results highlight the strong potential of the proposed FP16–SVE approach for enabling large-scale, high-efficiency, and near-real-time seismic simulations and earthquake hazard assessment on ARM-based platforms.

How to cite: Wang, W., Ren, B., Zheng, J., and Zhang, Z.: A High-Efficiency and Low-Storage Algorithm for Seismic Simulation Using Half Precision and Scalable Vector Extension on ARM Platforms, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11804, https://doi.org/10.5194/egusphere-egu26-11804, 2026.

The Statewide California Earthquake Center (SCEC) sequence of earthquake and aseismic slip (SEAS) group has regularly developed and published benchmarks along the years to follow recent development in the modeling of sequences of aseismic slip and earthquakes, as well as progress in numerical methods. These benchmarks, as well as the results from the different groups are publicly available at: https://strike.scec.org/cvws/seas/. It provides both reference solutions for code verification and a framework for systematic comparison of different modeling approaches. Every group is welcomed to join this Community driven comparison.

Recent efforts have focused on adding physics to better reproduce earthquake cycle by considering 2-dimensional fault (Jiang et al., 2022), improve our understanding of fluid injection processes (Lambert et al., 2025), and the effect of free surface and dipping fault (Erickson et al., 2023).

To follow up these developments, the SCEC SEAS group is designing two new benchmarks, that will be released to the community soon:

[1] A generalization of our benchmark about fluid injection (BP6) from a 2-dimensional domain to 3-dimensional domain. In this benchmark pore fluid diffuses along a 2-dimensional fault, modifying the effective normal traction  through one-way hydromechanical coupling. The fluid is injected for 10 hours on a rate strengthening faut and then shut off. The benchmark is designed to admit an analytical formulation for pore fluid diffusion while avoiding numerical singularities that may occurred with point source injection. For this reason, fluid is injected along a Gaussian profile.

[2] An updated version of our previous dipping fault benchmark (BP3), in a two-dimensional medium with a free surface. Previous version assumed that the normal traction along the fault was constant.  This is obviously a strong assumption, because the normal traction should increase with depth. However, this has proven to be difficult to simulate numerically as the system is going stiffer with lower normal traction. This benchmark therefore aims at providing a more realistic simulation of a dipping fault with a free surface by introducing depth dependent normal traction while also testing the ability of different numerical code to circumvent the problem of stiffness. This benchmark will be a joint benchmark with CRESCENT (Cascadia Region Earthquake Science Center).

This contribution will present the design of these forthcoming benchmarks and will provide an opportunity for the community to discuss about future benchmarks and directions for SEAS code comparison efforts.

How to cite: Romanet, P., Dunham, E., Erickson, B., Kim, T., Lambert, V., and Thakur, P.:  The future Community-Driven SCEC SEAS Code Comparisons for [1] Three-Dimensional fluid injection and [2] a Two-Dimensional dipping fault with variable normal traction., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12099, https://doi.org/10.5194/egusphere-egu26-12099, 2026.

Seismic waves originate from dynamic rupture propagation in faults. Seen from afar, faults are analogous to shear cracks, and their rupture can be analysed using the tools of fracture mechanics. However, a closer look reveals that faults can also be considered locally as a tribosystem, i.e. as a layered structure which accommodates deformation thought localized shearing in a thin granular layer of fault gouge. These two scales are equally important but are difficult to handle simultaneously in simulations.

In this communication, we propose a novel numerical model where this challenge is addressed. The gouge layer is represented using the Discrete Element Method, where each micrometric gouge grain (about 1 million of them in the present case) is explicitly represented and submitted to Newtonian dynamics, based on the forces it receives from its contacting neighbours. This layer is 2 mm-thick, and is confined between two continuum regions simulated using an explicit Meshfree Method. They receive the elastic properties of country rock, and are prestressed in the normal and tangential directions in order to bring the gouge layer just below its peak strength. The resulting fault system has a total length of 64 cm.

A labquake is then triggered from the central point of the fault, and the weakening rheology of the gouge layer allows it to propagate along two rupture fronts, which exhibit specific properties inherited from the frictional response and structure of the gouge. Inclined Riedel bands spontaneously develop at quasi-periodic intervals in the granular layer, and both rupture fronts propagate by leaps when successively activating slip in these structures. They both transition to a supershear regime after a certain sliding distance.

This model allows for the first time to observe the behaviour and response of the gouge layer as it endures the propagation of a rupture front. Localization patterns and granular complexity render the rupture irregular and heterogeneous, but a moving average in time in the frame of the crack tip allows to recover stress concentrations and slip velocity patterns which are consistent with the Linear Elastic Fracture Mechanics predictions. Il allows to relate gouge frictional response and rupture dynamics without the need to prescribe an arbitrary friction law or to rely on separation of scales.

How to cite: Mollon, G. and Casas, N.: Dynamic rupture of a gouge layer in a meter-sized labquake: a coupled numerical model, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12115, https://doi.org/10.5194/egusphere-egu26-12115, 2026.

Slow slip events (SSEs) are usually observed in elongated transitional zones between the seismogenic and creeping regions of the subduction zones, with the potentials to trigger large subduction earthquakes. Geodetic observations of SSEs in the Cascadia subduction zone (Michel et al., 2019) reveal contrasting complexities of rupture segmentations in the northern and southern segments separated by 44°N, with the northern segment preferring longer ruptures and the southern part preferring shorter ruptures. However, it remains unclear what mechanisms control the observed contrasting rupture segmentations of SSEs. Additionally, understanding the mechanisms behind the rupture complexities of SSEs can provide physical insights into the processes governing characteristic or complex earthquakes. Here, we conduct numerical simulations of SSE cycles along an elongated fault with a finite width W , which is governed by the rate-and-state friction with velocity-strengthening. We find that the rupture complexities of SSEs on a fault – classified as characteristic ruptures, complex ruptures, or creeping – depend on two non-dimensional ratios  L_nuc/W and L_c/W, where L_nuc is the critical nucleation length and L_c is the critical cohesive zone length. When L_nuc/W  is larger than 0.5, the fault keeps creeping and cannot produce any SSEs, which is consistent with previous theoretical predictions of 0.5 to 1. In addition, we find that runaway characteristic ruptures are enabled if the fault satisfies the energy balance condition between the energy release rate G₀ and the fracture energy G_c,, G₀ = G_c, derived from the three-dimensional theory of dynamic fracture mechanics that accounts for finite rupture width (Weng and Ampuero, 2022). If G₀ < G_c, ruptures prefer to arrest in a short distance and form complex events. This work proposes that a wide spectrum from creeping to characteristic ruptures is controlled by two length ratios in the framework of fracture mechanics, providing new physical insights into the mechanisms of SSEs.

References:

Michel, S., Gualandi, A., & Avouac, J.-P. (2019). Similar scaling laws for earthquakes and Cascadia slow-slip events. Nature, 574(7779), 522–526. https://doi.org/10.1038/s41586-019-1673-6

Weng, H., & Ampuero, J.-P. (2022). Integrated rupture mechanics for slow slip events and earthquakes. Nature Communications, 13(1). https://doi.org/10.1038/s41467-022-34927-w

How to cite: Shi, Y. and Weng, H.: Rupture Complexities of Slow Slip Events Controlled by Fault Friction Mechanics, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12196, https://doi.org/10.5194/egusphere-egu26-12196, 2026.

The main objective of this study is to identify clusters of seismic records with similar Fourier Amplitude Spectrum (FAS) shapes that can be associated with different tectonic domains, path attenuation properties, and site effects in Southern Italy. The analyzed dataset consists of FAS of S-wave windows, computed in the 0.5–25 Hz frequency range from accelerometric and velocimetric records available from EIDA and ITACA for 1349 events and 502 stations, with focal depths up to about 40 km.

We analyzed residuals between empirical FAS-based ground-motion models (GMMs), using ITA18 as reference, and observed spectral amplitudes through a mixed-effects regression framework. This allows us to decompose the total residuals into systematic contributions due to source (between-events term, δBe), path (systematic differences in attenuation, δWes), and site (site-to-site term, δS2S) effects, which are then grouped into clusters.

For the source terms δBe, four clusters are identified. Two of them are particularly interesting: one shows systematic amplification with increasing frequency, while the other shows systematic deamplification at high frequencies. The spatial distribution of the corresponding events highlights the Gargano and southeastern Sicily as regions characterized by amplified spectral amplitudes, whereas northeastern Sicily and the Aeolian area exhibit deamplified amplitudes. Additional insights are obtained by examining the dependence of these clusters on magnitude and focal depth; this analysis reveals that one of the source-related clusters is composed exclusively of shallow events (depth ≤ 10 km), which display distinctive spectral behaviors in specific crustal and volcanic domains.

For the path residuals δWes, four clusters are also recognized, revealing systematic differences in wave propagation across distinct crustal structures. The systematic site terms δS2S are grouped into three clusters: one identifies stations largely unaffected by significant soil amplification, while the other two show, respectively, systematic amplification and deamplification across the whole frequency band, with the clearest separation at intermediate frequencies (about 3–8 Hz).

These results provide a regional framework for ground-motion characterization in Southern Italy, supporting the identification of reference stations and of areas with distinct source and attenuation properties. This work is preparatory to future large-scale and local-scale Generalized Inversion Technique (GIT) studies aimed at the characterization of ground motion for shallow-crustal events and at the definition of key input parameters for earthquake simulations. In particular, the source-related clusters associated with volcanic areas reveal spectral features that deviate from classical ω² source models, pointing to processes likely controlled by complex fluid–rock interactions.

How to cite: Morasca, P., D'Amico, M. C., and Spallarossa, D.: Cluster Analysis of Fourier Amplitude Spectra Residuals for Ground Motion Characterization in Southern Italy., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12340, https://doi.org/10.5194/egusphere-egu26-12340, 2026.

Earthquakes and tsunamis occur on a timescale of seconds and are experienced by humans as sudden devastating disasters. However, the tectonic systems that determine where they occur are shaped over millions of years. Deformation in subduction zones is characterized by visco-elasto-plastic interactions between the accretionary prism featuring splay faults, subducting and overriding plate, asthenosphere, and free surface. To understand the present-day seismicity, earthquake cycle, and splay faulting in particular, these deformation processes need to be considered across all time scales. However, numerical models have not been able to resolve the dynamics across both tectonic and earthquake time scales.

We present a novel numerical modeling technique that simulates fully dynamic earthquake sequences and slow slip events in a subduction zone described by a visco-elasto-plastic rheology. Faults form and evolve spontaneously according to heterogeneous, temperature-dependent material parameters and the local stress field during both the initial 4 million years of subduction and the subsequent seismic phase. We employ an invariant formulation of rate- and state-dependent friction and adaptive time stepping to fully resolve all phases of the seismic cycle.

We generate events covering the slip spectrum from aseismic creep to earthquakes with slip rates in the order of m/s and tens of meters of slip. We find that events are largely characteristic despite the potential for deviating rupture paths in the subduction channel. We find that splay faults need to be sufficiently weak to be activated during a megathrust earthquake, since they cannot accumulate stress over time because velocity-strengthening afterslip relaxes their stresses. Dynamic triggering of a splay fault can lead to an early arrest of the megathrust rupture. Such short-term effects alter the long-term deformation compared to a purely geodynamic model by increasing the importance of one splay fault over others. We also observe that trapped seismic waves significantly change the slip distribution in a similar manner as has been found using a dynamic rupture model.

We conclude that our model successfully combines aspects of established geodynamic models and dynamic rupture models, providing a missing link between the long-term and the short-term. When applying this modeling approach to a complex continental setting, the interaction of multiple faults results in further complexities such as clustering. This highlights the potential and versatility of the method for a wide range of tectonic settings.

How to cite: Koelzer, A., de Vos, M., Gerya, T., and van Dinther, Y.: Bridging the Gap Between Millions of Years and Milliseconds: Modeling Earthquake Sequences, Slow Slip, and Splay Fault Rupture in Subduction Zones, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13053, https://doi.org/10.5194/egusphere-egu26-13053, 2026.

Physics-based earthquake wave propagation and ground motion simulations rely critically on three-dimensional seismic velocity models as inputs. These models may originate from seismic tomography, empirical regional compilations, geological constraints, or hybrid modelling approaches, and are commonly treated as deterministic representations of the subsurface. However, all such velocity models are affected by substantial epistemic uncertainty arising from limited data coverage, modelling assumptions, and methodological choices, and often disagree in overlapping regions. Neglecting this uncertainty obscures how variability in Earth structure propagates into simulated wavefields and ground motion estimates, limiting the interpretability and robustness of physics-based seismic hazard assessments.

We present a probabilistic framework to account for velocity model variability in physics-based ground motion predictions. Rather than selecting a single preferred velocity model, we represent model uncertainty through the fusion of multiple, spatially overlapping velocity models using scalable Gaussian process (GP) regression. Our approach treats existing velocity models as spatially correlated observations of an underlying velocity field and infers a continuous probability distribution that captures both shared structural features and model disagreement. The GP formulation thus preserves spatial coherence across scales and provides an interpretable description of uncertainty in terms of spatial covariance, characteristic length scales, and amplitude variability. This enables the generation of ensembles of physically plausible velocity model realisations for use in wave propagation solvers, thereby producing ground motion predictions that explicitly reflect velocity model uncertainty.

Using our framework and realistic 3D seismic velocity models in a regional case study, we generate an ensemble of velocity model realisations and propagate them through physics-based earthquake simulations. We show that uncertainty in velocity structure alone can produce substantial variability in simulated wavefields and predicted ground motions, even when all other aspects of the simulation are held fixed. These results highlight the sensitivity of physics-based ground motion estimates to uncertain subsurface structure and motivate the need to explicitly incorporate velocity model uncertainty in physics-based earthquake simulations.

While demonstrated here for seismic velocity models, the framework can readily incorporate additional geophysical parameters relevant to earthquake wave propagation, such as density and attenuation. This provides a practical route for incorporating epistemic Earth model uncertainty into physics-based seismic hazard assessment.

How to cite: Scivier, S. A., Koelemeijer, P., Mag, A. M., and Nissen-Meyer, T.: From uncertain velocity models to ensemble-based ground motion simulations, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13780, https://doi.org/10.5194/egusphere-egu26-13780, 2026.

Earthquake nests are defined as volumes of intense intermediate-depth seismicity which are isolated from any surrounding seismic activity. The high seismic activity within these earthquake nests occurs continuously and thus sets them apart from other seismic sequences such as earthquake swarms or aftershocks. These intermediate-depth earthquakes cannot be explained by the same causes as shallow earthquakes. Instead, they are often linked to slab detachment (e.g. in the Hindu Kush).

To constrain the conditions at which these large intermediate-depth earthquakes occur, numerical models are required to better understand their tectonic environment. Here, we use two-dimensional thermomechanical models with a nonlinear visco-elasto-plastic rheology were to determine the deformation state and the controlling mechanisms of the detachment process.

In this study, we focus on the question how the viscosity ratio (η_lith /η_lc) between the lithosphere and the lower crust and the depth d_lc to which lower crust may have been subducted influence the subduction process. Both is poorly constrained for the Hindu Kush. To this end, we varied the viscosity ratio η_lith /η_lc between 0.01 and 1000 and the subduction depth of the lower crust d_lc between 160 km and 240 km. We obtained detachment depths ranging from 110 km to 470 km, which fall within the range of the Hindu Kush earthquake nest, extending up to 280 km. The deformation behaviour from the 264 models can be classified into five different regimes based on stress, strain rate, detachment depth, and coupling between subducting and overriding plate. The five regimes represent the dependency of the detachment depth (d_det) to its viscosity ratio (η_lith /η_lc). Detachment in regime two is enhanced via shear heating and detachment in the other regimes occurs via necking. The relationship between lower crustal depth and detachment depth varies by model category. This variability reflects the complex influence of the “lubrication effect” of a weak lower crust and the limitation of subduction depth governed by its rheological properties.

How to cite: Weiler, T., Piccolo, A., Spang, A., and Thielmann, M.: Effects of the lower crust on slab detachment – a case study in the Hindu Kush, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14664, https://doi.org/10.5194/egusphere-egu26-14664, 2026.

The amount of geodetic surface displacement observations from GNSS and InSAR has been growing in recent years yet exploring the model space of corresponding sub-surface deformation remains a complicated and computationally expensive exercise. This is especially the case when there is more than one source and is further complicated when there is a variety in source types, such as combinations of on-fault slip and off-fault mantle flow.  While analytical solutions exist for a variety of deformation types within elastic half-spaces (such as fault slip, tensile dislocation, volumetric strain, expansion/contraction) the optimization of source parameters beyond single source models is computationally burdensome due to the need to extensively search with forward passes of the numerical solutions.  In most kinematic modelling exercises, the strategy is to assume geometries of sources and solve for magnitude parameters in inversions or to let a Finite Elements simulation evolve from a starting static displacement.  Furthermore, there is no effective way to blindly discover the number of sources along with their respective modes of deformation.

Here we demonstrate a solution to these problems that uses surrogate cuboid anelastic deformation sources and sparsity. Cuboid surrogates, that are trained on analytical solutions of anelastic deformation in a half-space, provide a versatile parametrization capable of approximating a wide range of deformation styles - from volumes to faults - by collapsing the thickness towards a near-planar geometry.  Once trained, the model can be run in inversion mode so that parameters of the source, such as centroid, length, width, depth, and strain tensor can be optimized by means of a back-propagated loss between the measured surface displacement and surrogate model prediction.  Multiple sources can be added trivially, and a sparse solution found with an approximately sparse optimization strategy.

By replacing repeated forward evaluations with a trained surrogate model, the proposed framework enables rapid optimization directly from observed deformation fields without the need for assuming the types of deformations or number of sources. This combination of a flexible cuboid-based source representation and efficient surrogate modelling offers a practical route towards scalable discovery of sub-surface deformation features.

How to cite: Çökerim, K. and Bedford, J.: Versatile Surrogate Inversion of Deformation Sources, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14752, https://doi.org/10.5194/egusphere-egu26-14752, 2026.

Understanding the physical mechanisms governing aftershock patterns and their evolution in fault networks is crucial for interpreting seismic catalogues and improving physics-based seismic hazard assessment. Here, we develop a mechanics-based modeling framework based on the discrete fracture network approach to explicitly simulate mainshock rupture, coseismic stress changes, and aftershock generation in complex 3D fault networks. The fault system that we model comprises a primary strike-slip fault surrounded by a network of thousands of secondary faults with sizes following a power-law distribution. Dynamic rupture nucleates within a localized patch on the primary fault and propagates spontaneously at a sub-Rayleigh speed, producing a M_w 7.6 mainshock. The model captures aftershock triggering driven by radiated seismic waves and/or permanent stress redistribution, and quantifies their combined effect using Coulomb failure stress changes. Fault slip is governed by a linear slip-weakening friction law, where the critical slip distance is varied over orders of magnitude to explore its influence on breakdown-zone size, fracture-energy dissipation, and rupture propensity on secondary faults. The simulations capture key emergent characteristics of aftershock sequences: spatially, aftershocks cluster within positive Coulomb stress lobes and are suppressed within stress shadows, with additional localization near fault intersections; statistically, the cumulative frequency–magnitude distributions follow Gutenberg–Richter scaling over a broad magnitude range. Importantly, the synthetic catalogues consistently exhibit a two-branch frequency–magnitude scaling behavior, in which the lower-magnitude branch is dominated by partial ruptures and premature arrest, whereas the higher-magnitude branch corresponds to self-sustained ruptures whose moment magnitudes scale with fault area and are therefore more strongly constrained by fault network geometry. We further show that the transition between these regimes is governed by fault criticality and fracture energy dissipation, providing an alternative mechanics-based explanation for the commonly observed roll-off in frequency–magnitude distribution. Overall, our framework mechanically connects fault network structure and rupture dynamics to explain aftershock statistics, enabling physics-based interpretation of seismic catalogues and supporting improved seismic hazard assessment.

How to cite: Pan, W., Zhang, Z., and Lei, Q.: Mechanics-based simulation of aftershock sequences in complex 3D fault networks, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15115, https://doi.org/10.5194/egusphere-egu26-15115, 2026.

Modeling earthquake rupture dynamics often requires stochastic approaches to address the impracticality of obtaining analytical solutions for asymmetric many-body systems. Following Langevin's approach, we propose a stochastic dynamic model for the earthquake rupture process, where complexity in degrees of freedom is reduced by introducing a random force to account for uncertainties in fault plane heterogeneity and structural collisions. In this coarse-grained framework, the random term captures unresolved heterogeneity and interactions at a macroscopic system scale; it does not assert that rupture at the scale of specific fault patches is inherently random. Treating the tectonic process as a Coulomb friction process allows this Langevin equation to be viewed as a stochastic variant of Newton’s second law, attributing physical significance to the sample paths.

However, applying a zero-dimensional (0-D) stochastic framework to complex faulting raises a critical conceptual challenge: can a model lacking explicit spatial dimensions reproduce the highly heterogeneous energy distribution observed in nature? Intuition suggests that the exponential slip distribution derived from a 0-D process may not exhibit tail behavior sufficient to satisfy the standard asperity criterion, where a small fraction of the fault area releases most of the seismic energy. To validate the physical basis of the model, we first examine the spectral properties of the synthetic velocity fluctuations. Results demonstrate that the model output is not arbitrary white noise; rather, the velocity spectra exhibit a Lorentzian form characterized by a single corner frequency. This spectral structure indicates that system memory is governed by a characteristic timescale determined by the load ratio, reflecting a competition between frictional dissipation (which erases memory) and external driving (which sustains motion).

Furthermore, we evaluate the steady-state slip distribution derived from the corresponding Fokker–Planck equation against empirical scaling relations for asperities. Adopting the criterion which defines an asperity as regions where slip exceeds 1.5 times the average, and using squared slip as an upper-bound proxy for energy release under elastic loading, we calculate the theoretical energy concentration. The model predicts that the top ∼22% of the statistical "area" contributes ∼81% of the total energy. This theoretical prediction lies within the 20–30% range observed empirically for asperity area fractions. These findings suggest that the concentration of energy in asperities can emerge from stochastic frictional dynamics, arising from the exponential tail of the slip distribution without explicit modeling of spatial heterogeneity.

How to cite: Wu, T.-H. and Chen, C.-C.: Emergence of asperity-like energy concentration in a stochastic Langevin framework, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15721, https://doi.org/10.5194/egusphere-egu26-15721, 2026.

The excitation and propagation of multiple wave types, including seismic waves, ocean acoustic waves, and tsunamis triggered by earthquakes within the oceanic wavefield, constitute a problem of substantial scientific and practical challenge for both fundamental geophysical understanding and hazard assessment. While various numerical approaches have been proposed to model these full-coupled wavefields, the role of realistic seafloor topography in modulating wave propagation remains underexplored.

We present a novel earthquake-tsunami coupled simulation approach based on the spectral-element method (SEM), leveraging its robustness and accuracy in representing arbitrary fluid-solid interface geometries. The approach is quantitatively validated through comparisons of simulated permanent seafloor deformation and sea-surface displacement time series with benchmark finite-difference method (FDM) solutions, yielding an excellent correlation coefficient of 0.998 and negligible errors. Furthermore, we construct two distinct numerical models: one incorporating realistic seafloor topography and another assuming an idealized flat seafloor to investigate the effects of bathymetry on oceanic wavefield. Our analyses reveal that complex bathymetry profoundly alters the propagation of both seismic and tsunami waves, modifying amplitudes, arrival times, and spatial distribution patterns. By systematically separating the contributions of the overlying seawater and the underlying seafloor topography, we clarify their individual influences on the composite oceanic wavefield. We also investigate how variations in earthquake source location affect wave propagation waves, underscoring the necessity of accurate bathymetric representation for offshore events.

This SEM-based earthquake-tsunami coupling framework offers a robust tool for comprehensively understanding the oceanic wavefield under gravity and holds considerable promise for advancing earthquake and tsunami risk evaluation, especially when combined with seismological observational data.

How to cite: Hou, X., Zhang, L., and Xu, Y.: Coupled simulation of earthquake and tsunami by spectral-element method and effects of bathymetry on oceanic wavefield, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15953, https://doi.org/10.5194/egusphere-egu26-15953, 2026.

Rotational ground motions have recently emerged as an important and independent observable in seismology, driven by advances in rotational seismometers and the growing availability of high-quality rotational datasets. These observations provide new insights towards understanding near-source and near-surface wave propagation beyond traditional translational measurement. To model rotational components, several analytical approaches have been proposed in the recent past. However, these formulations are typically restricted to idealized source representations and simplified Earth models, limiting their applicability to realistic geological settings accounting for three-dimensional complexities.

Finite-element modeling techniques provide a powerful alternative by enabling the simulation of seismic wavefields in complex media by incorporating heterogeneous velocity structures, layered stratigraphy, surface topography, and finite-fault earthquake sources. Despite this capability, commonly used ground motion simulation codes have not yet been adapted to compute rotational ground motions. In this study, we extend the spectral finite-element code SPECFEM3D to internally compute and output rotational ground motions alongside conventional translational components. The numerical implementation is validated against analytical solutions for two benchmark cases: (i) a homogeneous half-space and (ii) a three-layered velocity model, demonstrating excellent agreement in both amplitude and waveform characteristics. Following validation, the modified code is used to simulate rotational ground motions for a range of realistic scenarios, including layered representations of the subsurface and finite-fault source models. These simulations are used to investigate the generation and propagation characteristics of rotational motions and to examine their spatiotemporal relationship with translational ground motions. Differences in amplitude and propagation behavior between rotational and translational components are particularly analyzed in the present work.

Finally, we assess the potential implications of rotational ground motions for earthquake engineering by evaluating their relative amplitudes and propagation patterns under different source and structural conditions. The results provide a framework for identifying the source characteristics and conditions under which rotational components of ground motion may become significant and potentially influence structural response. These findings contribute to an improved understanding of whether, and under what circumstances, rotational ground motions should be considered in seismic analysis and earthquake-resistant design practice.

Keywords: Rotational ground motions, Seismic wave propagation, Numerical modeling, Earthquake engineering

How to cite: Dhabu, A. C., Hejazi Nooghabi, A., and Hadziioannou, C.: Propagation Characteristics of Rotational Ground Motions in Layered Earth Media, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17597, https://doi.org/10.5194/egusphere-egu26-17597, 2026.

The Eastern Adriatic region has been historically affected by strong destructive earthquakes, including the M6.4 1667 Dubrovnik earthquake, the M6.6 1905 Shkodra event, and the M6.4 2019 Durrës earthquake. Some of those destructive events are associated with the Scutari-Pec Fault System (Albania). This tectonic structure extends sub-parallel to the coastline, in the SW-NE direction, through the Dinaride-Hellenide transition. This fault system corresponds to a compressive and transform fault system near the Adriatic Sea that changes the tectonics to an extensional regime towards the East. The distribution of focal mechanisms  of microseismicity recorded in the region (Serpelloni et al, 2007) evidences the complex tectonics (Grund et al., 2023). 

As part of the German SPP project DEFORM, we plan to simulate 3D dynamic earthquake scenarios to study the rupture propagation of large earthquakes across the Scutari-Pec Fault System. We apply the open-source SeisSol code to generate synthetic seismograms up to frequencies of 2 Hz. We will specifically investigate the effect of variability of locking depth as a crucial parameter for determining the earthquake potential of the fault system. Ground motion for dynamic rupture scenarios with characteristics similar to the destructive reported events will be  estimated, in particular for the most populated cities located within 50 km of the central fault system. This presentation is a first step toward these goals and  aims to provide relevant information for such simulations, which may complement seismic hazard assessment in the region.

How to cite: Abril, C. and Gabriel, A.: Towards physic-based ground-motion simulations for the Scutari-Pec Fault System, Eastern Adria, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18857, https://doi.org/10.5194/egusphere-egu26-18857, 2026.

Reliable physics-based seismic hazard assessment (PB-SHA) requires basin velocity models that accurately reproduce key characteristics of observed seismic wave propagation, which is critical for predicting ground-motion scenarios in complex sedimentary basins. We present a validation study of a three-dimensional basin model with depth-dependent P- and S-wave velocity profiles of the Sulmona Basin (central Italy), developed to represent basin-scale structures relevant for physics-based ground-motion simulations.

The model is implemented in the spectral-element code SPECFEM3D and evaluated through direct comparison of observed and synthetic seismograms at the available stations within the basin. Simulations are performed for selected regional earthquakes, with synthetic waveforms filtered to match the target frequency range of the model. Waveform misfit is quantified using the Pyflex framework, allowing an objective assessment of phase arrival times, waveform similarity, and amplitude differences across multiple stations.

The results show that the model reproduces basin-controlled wave-propagation characteristics, including waveform duration and spatial variability of ground-motion amplitudes. Amplitude variability and waveform agreement primarily reflect the depth-dependent velocity structure and 3D basin geometry, while localized misfits reflect unresolved features and the limited number and spatial coverage of recording sites.

Overall, this validation provides a first quantitative assessment of the Sulmona Basin velocity model, forming a foundation for subsequent work towards physics-based seismic hazard assessment and scenario modelling.

How to cite: May, J. B., Kastelic, V., Carafa, M. M. C., de Nardis, R., and Casarotti, E.: Validation of a 3-D Basin Velocity Model for Physics-Based Seismic Hazard Assessment: The Sulmona Basin, Central Italy, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21376, https://doi.org/10.5194/egusphere-egu26-21376, 2026.

This study investigates the influence of surface-water-level fluctuations on seismicity in the upper crust, using the historic Dead Sea as a natural laboratory.  We apply a validated 2D poro-elasto-plastic coupling model in COMSOL Multiphysics. The model integrates coupled hydro-mechanical processes, including pore-pressure evolution, plastic strain localization, and permeability changes, to capture the interaction between surface loading and fault stability. Given reported challenges in capturing hydro-mechanical coupling and scaling behaviour in natural systems using the rate-and-state friction (RSF) formulation, this study adopts an alternative modelling framework that does not explicitly incorporate RSF. The study focuses on applying the model to reconstruct earthquake occurrence patterns associated with Dead Sea water-level variations over the past two millennia. Results demonstrate a strong correlation between relatively rapid water-level changes and increased seismic activity, highlighting the critical role of hydrological forcing in earthquake triggering. These findings provide new insights into reservoir-induced seismicity and underscore the importance of incorporating surface water dynamics into seismic hazard assessment.

How to cite: Belferman, M. and Agnon, A.:  Hydro-Mechanical numerical Modeling of Water-Level-Induced Seismicity: Insights from Historic Dead Sea Fluctuations, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-22999, https://doi.org/10.5194/egusphere-egu26-22999, 2026.

The French Massif Central (FMC) hosts a complex intraplate volcanic system that is probably influenced by deep mantle processes, Variscan inheritance, and Cenozoic rifting. To better understand the crustal and mantle structures of the FMC, a consortium of 4 French laboratories – ISTerre Grenoble, LMV Clermont-Ferrand, GET & IRAP Toulouse – has set up the MACIV project (2023–2027) based on multiscale seismic experiments combining regional-scale and dense local deployments. The regional scale is covered by 2 temporary networks of 100 broadband stations spanning the whole FMC for a duration of 3–4 years. The large-scale XP array of 35 stations complements the permanent networks to achieve homogeneous coverage with ~35 km spacing. It is France’s contribution to the European AdriaArray project. The XF network of 65 stations includes 3 quasi-linear profiles (N–S, E–W, NW–SE) that cross major volcanic areas and Variscan structures with inter-station spacing of 5–20 km.

In September 2025, the broadband arrays were supplemented by a month-long deployment of 2 dense arrays of 624 three-component short-period nodes (5 Hz) across an 80 km x 100 km area. The instruments were deployed in two nested networks : an ultra-dense network (inter-station distance 0.5–1.5 km) on an aperiodic grid covering the recent volcanoes of Chaîne des Puys and Mont-Dore/Sancy massif, and a larger-scale regular grid network (inter-station distance 3.5–7 km).

All MACIV data are or will be distributed by SI-S EPOS-France. Waveforms of the XP array are openly available in real-time since the acquisition started in 2023. XF and nodal data will be publicly available after July 1, 2026.

We will present an overview of the project objectives, the experimental setup designed to optimize performance and cost efficiency, as well as the innovative tools developed for dense network deployment, the data acquisition strategy, and preliminary results.

How to cite: Aubert, C., Scheiblin, G., Chevrot, S., Cluzel, N., Mordret, A., Pauchet, H., Paul, A., Roussel, S., Shapiro, N., Souriot, T., and Sylvander, M. and the MACIV-NODES Team: Probing lithosphere and volcanoes of the French Massif Central using multiscale seismic experiments: the MACIV project, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7345, https://doi.org/10.5194/egusphere-egu26-7345, 2026.

Distributed Acoustic Sensing (DAS) based on fibre-optic technology is increasingly used in seismic monitoring and subsurface imaging, thanks to its dense spatial sampling, robustness, and operational flexibility. Despite its growing adoption, a detailed comparison with conventional borehole seismic sensors under controlled conditions remains essential to better understand the respective strengths, limitations, and complementarities of the two technologies.

This study presents a controlled Vertical Seismic Profiling (VSP) experiment carried out in the PITOP2 well at the PITOP geophysical testing site¹, operated by OGS and part of the ECCSEL-ERIC facilities². The experiment was performed within a Transnational Access framework funded by the Geo-INQUIRE project³. The acquisition was specifically designed to compare the seismic response of a non-permanent deployment of a borehole DAS fibre-optic cable with that of a co-located cemented array of three-component borehole geophones.

The seismic source consisted of a Vibroseis Minivib operating in P-mode, with three surface energization points located at offsets of 25, 50, and 75 m from the well. The DAS measurements were acquired using two different interrogators (Silixa iDAS and Carina®), enabling the assessment of DAS response consistency across interrogation systems. The borehole DAS cable and the Carina® interrogator were purchased thanks to PNRR ITINERIS⁴ funding.

The experiment benefits from the well-characterized subsurface conditions and permanent infrastructure available at PITOP, minimizing geological uncertainties and allowing the focus to be placed on instrumental and acquisition-related effects.

The results of this study enabled to draw technical and scientific evaluations on the experimental design, acquisition parameters and comparison strategy, outlining the analysis of waveform characteristics, frequency content, signal-to-noise ratio, and depth-dependent response.

All datasets of this experiment are openly available in SEG-Y format through the PITOP data portal¹, in compliance with FAIR data principles, supporting transparency, reproducibility, and reuse by the scientific and industrial communities

The experiment builds upon previous geophysical investigations conducted at PITOP, as summarized in the recent publication “Geophysical exploration case histories at the PITOP geophysical test site – A key facility in the ECCSEL-ERIC consortium”⁵, further demonstrating the role of PITOP as a reference facility for testing emerging seismic technologies.

Acknowledgements:

Geo-INQUIRE is funded by the European Commission under project number 101058518 within the HORIZON-INFRA-2021-SERV-01 call.

ECCSEL ERIC was established by the European Commission implementing decision of 9 June 2017 (EU) 2017/996.

References:

• OGS PITOP website: pitop.ogs.it - Transnational Access, SeisDAS Project 2025, EXP1, DOI: 10.13120/0DTM-JX93)
• ECCSEL ERIC PITOP page: https://eccsel.eu/catalogue/facility/?id=126
• Geo-INQUIRE website: https://www.geo-inquire.eu/
• EU - Next Generation EU Mission 4, Component 2 - CUP B53C22002150006 - Project IR0000032 – ITINERIS - Italian Integrated Environmental Research Infrastructures System
• Geophysical exploration case histories at the PITOP geophysical test site – A key facility in the ECCSEL-ERIC consortium: an overview. Bellezza et al., Bulletin of Geophysics and Oceanography, 2025

How to cite: Travan, A., Meneghini, F., Bellezza, C., Lucerna, L., Bernardi, P., Barison, E., and Schleifer, A.: Comparison of borehole seismic recordings from Distributed Acoustic Sensing and conventional 3C geophones: a VSP experiment at the PITOP testing site (OGS-ECCSEL), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7458, https://doi.org/10.5194/egusphere-egu26-7458, 2026.

Accelerating mass loss from the Greenland Ice Sheet is strongly modulated by meltwater-driven changes of bed conditions in ice dynamics, but these processes remain poorly constrained due to limited high-resolution observations of glacier hydrological systems. Constraints on basal properties can be obtained from borehole measurements and active reflection seismic surveys; however, these methods are inherently limited by their localized spatial coverage and are time intensive. As a result, they are poorly suited to monitoring the subglacial hydrological system, which evolves over seasonal timescales and across spatial scales larger than can be practically sampled. Thus, dense seismic node arrays offer a powerful alternative, enabling passive, high-resolution imaging and continuous monitoring of englacial and subglacial processes over broad areas and extended time periods.

Here, we first present a dense seismic array experiment covering approximately 2.5 km² we conducted in the ablation zone of Isunnguata Sermia, West Greenland. The experiment comprised 82–117 autonomous seismic nodes deployed during one-month long monitoring periods in spring 2023 and in summer and fall 2024, complemented by multi-week surface Distributed Acoustic Sensing (DAS) measurements conducted in 2024. We assess data quality using power spectral densities and ambient noise cross-correlations to detect sensor tilt and GPS timing desynchronization, which are critical issues for data quality in remote and highly dynamical glacial environments such as the Greenland Ice Sheet.

Then, we demonstrate we can use natural icequakes to perform seismic reflection analysis for bed mapping and interface property evaluation. We locate numerous natural seismic events with meter-scale resolution using Matched Field Processing (MFP) beamforming applied to array data in the 4–6 Hz frequency band. These events reveal a large population of near-surface sources associated with crevassing that generate both surface and body waves. We extract body waves and enable event stacking following a two-step synchronization procedure first synchronizing signals in the 4–6 Hz band dominated by surface waves, and subsequently refining timing in the 40–80 Hz band where P waves dominate. By sorting and stacking waveforms from these synchronized natural sources into common midpoint gathers, we identify high signal-to-noise ratio reflected P waves from the ice–bedrock interface at offsets of up to 1 km, significantly greater than those typically achieved in traditional active reflection seismic surveys on glaciers. This extended offset range makes the passive approach better suited for estimating basal conditions using classical amplitude-versus-offset (AVO) analyses due to the stronger dependence of reflectivity to bed properties at large incidence angles.

Finally, following a workflow analogous to active reflection seismology, we derive a two-dimensional map of ice thickness beneath the array with a vertical resolution of approximately 15 m, comparable to that of conventional active surveys. This ice-thickness model will serve as a critical constraint for future high-resolution surface-wave tomography, enabling improved imaging of glacier structure and basal conditions at depth.

How to cite: Paris, N., Gimbert, F., Roux, P., Livingstone, S. J., Doyle, S. H., Michel, A., Sole, A. J., Lecointre, A., Pinzon-Rincon, L., Hillers, G., Courbis, R., and Barruol, G. and the SLIDE-REASSESS team: High-Resolution Glacier Bed Imaging with Passive Seismology Using a Novel Dense Seismic Array at Isunnguata Sermia, West Greenland, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10595, https://doi.org/10.5194/egusphere-egu26-10595, 2026.

The Artius broadband node represents a transformative innovation in seismic instrumentation, designed to bridge the gap between traditional broadband seismometers and popular nodal systems. While broadband seismometers offer unparalleled sensitivity and frequency range, their cost and complexity often limit large-scale and dense deployments. Conversely, geophones provide cost-effective solutions for high-frequency applications but lack sensitivity to low-frequency seismic signals, which are critical for many research and monitoring purposes. Artius provides a cost-effective compromise, delivering the increased sensitivity and a true broadband frequency range at an economic price point.

Designed by Güralp Systems, Artius integrates a compact, highly sensitive broadband seismometer with an environmentally sealed anodised aluminium enclosure, ensuring optimal performance and robustness across diverse geophysical applications. Boasting a response of 30 seconds to 200 Hz, Artius greatly outperforms geophone-based systems while still being perfectly suited to rapid temporary deployments where it can be either pushed or staked into the ground and connected to an external power supply. Artius pushes the limits of versatility, facilitating real-time data monitoring, as well as passive data collection. Artius has an onboard SEEDlink server, compatible with all standard seismological monitoring techniques, truly setting it apart from anything on the current market.

Artius is designed to be docked into an eight-node capacity docking station for data validation and mass data download. The docking station also serves as a ‘huddle’ system for configuration and testing prior to deployment, ensuring each node is performing optimally. The Artius nodes are intended to be deployed in large arrays, perfect for passive seismology, ambient noise studies, and earthquake investigations.

How to cite: Calver, J., Lindsey, J., Kerkenyakova, A., Watkiss, N., Kitka, K., Hill, P., and Restelli, F.: Artius: A revolutionary broadband node to enable passive seismology, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11433, https://doi.org/10.5194/egusphere-egu26-11433, 2026.

Large earthquakes involve processes that occur primarily at seismogenic depths of 5-10 km or more and that are thus difficult to observe in detail with traditional seismic networks. The FaultScan project establishes the first long-term, dense nodal array experiment by acquiring continuous data at 300 stations along the San Jacinto Fault (SJF, South California) at the Piñon Flat Observatory for 2.5 years (April 2022-Nov. 2024).

We report on the experiment logistics, data quality and describe the characteristics of the High-Frequency (HF, >1 Hz) back-ground noise with a focus on incoherent wind generated noise, train traffic tremors and far distant coherent urban noise. We illustrate how these HF coherent noise sources recorded mostly as body-waves can be used to track velocity changes across the SJF by applying seismic interferometry between the array and nearby permanent seismic stations.

We further describe how much slant-stacking increases the level of signal to noise ratio for the detection of small earthquakes along the San Jacinto Fault and show how newly detected events improve our description of foreshock/aftershock patterns. Our search for tectonic tremors along the San Jacinto Fault turns up empty but we observe tremor-like signals, mostly T-phases coming exclusively from the Tonga subduction and one intriguing T-phase like sequence originating from the October 2023, offshore Japan volcanic crisis.

How to cite: Brenguier, F., Higueret, Q., Mordret, A., Sheng, Y., Vernon, F., Hollis, D., Aubert, C., Pinzon-ricon, L., Vignon-livache, M., Lavoué, F., Capes, C., Pineda, M., Hobkirk, C., Kakiuchi, Y., and Ben-zion, Y.: The FaultScan Long-Term Dense Nodal Array to Study the San Jacinto Fault (Southern California), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12341, https://doi.org/10.5194/egusphere-egu26-12341, 2026.

Distributed Acoustic Sensing is a transformative technology rapidly advancing in seismology. It enables recording high-resolution seismic data in remote areas and is especially beneficial for seafloor sensing, where standard seismic instruments are complex to operate in real time.

Over the past two years, the Geological Survey of Denmark and Greenland (GEUS) has conducted continuous and campaign-based Distributed Acoustic Sensing (DAS) acquisitions on terrestrial and submarine fiber-optic cables, exploring the potential of DAS as a complement to the national seismic monitoring infrastructure. Using two ASN C-band interrogators deployed in both permanent and mobile configurations, GEUS has collected tens of TB of DAS data, with applications ranging from earthquake monitoring and onshore active seismic experiments to the detection of anthropogenic and marine activities.

This exploration constitutes a steep learning curve for GEUS in field logistics (how to access dark fibers, how to trench our own fibers, ...), as well as in data acquisition, processing, and archiving. Here, we present lessons learned along the way from fieldwork operations and observations of regional and teleseismic earthquakes, quarry blasts, submarine explosions, controlled naval experiments, active seismic tests, and so on. 

Regional earthquakes recorded on submarine cables demonstrate DAS's sensitivity to S-waves even when P-wave signal-to-noise ratios are low. In contrast, teleseismic events (e.g., the M6.5 Jan Mayen earthquake) reveal complex wavefields that are strongly modulated by bathymetry and sedimentary structures. Comparative analyses across multiple cables highlight significant variations in signal quality and frequency content linked to cable type, installation conditions, and seafloor environment. Automatic phase picking using PhaseNet shows promising results, particularly when combining multiple pre-trained models, but also exposes key limitations related to strain-rate measurements, coupling variability, and absolute timing errors caused by the lack of GPS synchronization. Anthropogenic signals, including quarry blasts, anchor drops, vessel traffic, and rare submarine passages offshore, and active and passive seismic acquisition onshore, illustrate both the detection capabilities and the sensitivity to acquisition parameters.

Overall, this two-year dataset demonstrates that DAS can significantly enhance seismic and environmental monitoring, while also identifying critical technical challenges, data volume management, timing accuracy, and site-dependent coupling that must be addressed before DAS can be fully integrated into our operational seismological workflows.

How to cite: Mordret, A., Fønss Jensen, E., Larsen, T., Voss, P., Dahl-Jensen, T., Rinds, N., and Sólheim, L.: Two years of DAS acquisitions at GEUS, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12427, https://doi.org/10.5194/egusphere-egu26-12427, 2026.

The Norwegian Caledonides are well mapped at the surface, but their depth extent and subsurface geometry are poorly known. Renewed interest in mineral occurrences within the Caledonian nappes comes with the need to better understand the three-dimensional geometry of mineral bearing units and their tectonic history. A recurring discussion entails whether the nappe stack in central Norway forms a synclinal or anticlinal geometry.

To address these questions, we conducted an ambient noise tomography (ANT) survey aimed at imaging the 3D subsurface structure of the Caledonian nappes in central Norway. ANT provides a comparatively inexpensive and flexible method to investigate crustal structure, allowing us to distinguish high-velocity mineral bearing units from surrounding lower-velocity metasedimentary layers as well as the topography of the underlying crystalline basement. 

The survey was carried out in October 2025 over a three-week period and involved the deployment of 300 vertical component Sercel DFU seismic accelerometer nodes. A 3D-array of ca. 15x20 km, with about 0,6-1 km inter-station spacing was supplemented with a NW-SE trending profile of 30 nodes. The fieldwork in the rugged subarctic terrain was carried out by 3-4 teams of 2 people for the installation and retrieval of the array, for 5-6 days for each operation. About a third of the array was deployed by car along gravel roads, the rest was deployed either by e-bikes along large hiking and tractor trails or on foot for small hiking trails and off-trail sites. The planning and identification of the site access modes was paramount for the efficiency of the deployment. We used the Google Earth phone app as field deployment tool to automatically build the metadata of the array based on geo-localized pictures of the nodes and their serial numbers in the field, aided by an Optical Character Recognition algorithm. 

While the 2D profile cannot resolve lateral structures, its full extension and orientation toward the north-west and thereby toward the main ambient seismic noise sources of the North Atlantic should allow us to resolve the velocity contrast between basement and nappes and the nature of their geometry. 

We present the first results from both the 3D array and the 2D profile, providing new constraints on the depth extent and internal structure of the Caledonian nappes and contributing to the ongoing discussion of their subsurface geometry. 

How to cite: Gradmann, S., Mordret, A., Saalmann, K., Bredemeier, B., Flem, B., Nesgaard, A., Fauskanger, A., Yanev, M., Moe, A., and Kvithyll Eriksen, J.: Studying the Norwegian Caledonides with ambient noise tomography: fieldwork lessons and preliminary results, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12768, https://doi.org/10.5194/egusphere-egu26-12768, 2026.

In recent years, Distributed Acoustic Sensing (DAS) has emerged as a powerful technology in seismology, enabling the acquisition of high-resolution seismic data using optical fibers as sensors. As the number of DAS experiments continues to grow and DAS interrogators become able to record along ever longer fiber-optic cables, the volume of generated data is rapidly increasing, creating significant challenges for long-term archiving and efficient data access. These challenges include not only the storage of very large datasets—often on the order of hundreds of terabytes—but also the ability to fast random access and process subsets of the data.

Currently, the DAS ecosystem is dominated by proprietary data formats defined by individual vendors. While HDF5 is increasingly adopted as a more open alternative, it presents strong limitations regarding scalability in multi-threaded and multi-process environments. In contrast, modern data container formats such as Zarr and TileDB offer native support for parallel I/O and flexible storage backends, ranging from local file systems to on-premise and cloud-based object storage.

In this contribution, we present a comparative evaluation of these modern data formats for DAS applications, focusing on performance, scalability, and usability. We discuss the latest results obtained from the activities of the Geo-INQUIRE* project and assess the feasibility and potential benefits of their adoption for the long-term management and analysis of DAS datasets.

* Geo-INQUIRE is funded by the European Union (GA 101058518)

How to cite: Quinteros, J., Rodriguez Tribaldo, V., Wollin, C., and Strollo, A.: Modern Data Containers for Scalable Archiving and Access of Distributed Acoustic Sensing Data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12779, https://doi.org/10.5194/egusphere-egu26-12779, 2026.

In addition to permanent geophysical observation networks, temporary field measurements are an important component of solid Earth research. In seismology in particular, there has been a steady increase in the number of measuring devices used within one deployment. This is mainly due to the fact that dense networks (as opposed to individual stations with large distances between them) allow wave fields to be recorded in their entirety enabling new processing methods and higher-resolution subsurface imaging.

Since the maintenance of such large numbers of devices required for dense networks is not a side issue, instrument pools are necessary to supply devices for the academic community. One of the largest instrument pools in Europe is the Geophysical Instrument Pool Potsdam (GIPP), which is operated by the GFZ (gipp.gfz.de). The GIPP provides geophysical and geodetic measurement technology (e.g., recorders and sensors) for temporary active seismic, passive seismological, electromagnetic, and GNSS experiments. The equipment is supplied for usage at universities and research institutes worldwide and free of charge for non-commercial experiments. We team up with our partners in Europe through the ORFEUS Mobile Pools Service Management Committee. Together we are working on improving the cooperation between the major European instrument pools and offering services that facilitate access to instruments, also within the EPOS-ON project.

Over the past 30 years, we have supported more than 500 geophysical field experiments. The data of these experiments is archived at GFZ and is made available to the public, e.g., via the GEOFON repository. The majority of the seismological experiments consists of fewer than 50 stations, but the number of large networks (with up to 500 devices) is increasing. These large networks are realized either in the form of rolling arrays or as temporary installations (LARGE-N), sometimes in collaboration with other providers.

In this presentation, we give an overview of the latest deployments, our data management approach, the challenges that we are facing due to the high demand of LARGE-N experiments, and technological advances in our equipment, particularly in robust node-type field recorders. Thereby, we want to discuss the challenges and potentials of such pools, making them best setup for future research.

How to cite: Wawerzinek, B., Haberland, C., Ritter, O., Männel, B., and Krawczyk, C. M.: Geophysical Instrument Pool Potsdam (GIPP): Enabling large-scale seismic projects for 30 years – experiences and challenges, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18050, https://doi.org/10.5194/egusphere-egu26-18050, 2026.

Distributed Acoustic Sensing (DAS) has emerged as a promising technology for large-scale, passive monitoring of acoustic sources in the marine environment. In this work, we investigate the feasibility of using DAS to track and localize surface vessels based on their continuously emitted signals. Ship-generated acoustic signals span a wide frequency range and are often characterized by periodic or quasi-periodic narrowband tonal components. Depending on the coupling conditions of the fiber cable, these signals may exhibit limited spatial coherence across the DAS array thereby reducing the reliability of conventional localization techniques, such as time-difference-of-arrival (TDOA) approaches. To address these challenges, we apply a combination of signal-to-noise ratio (SNR) enhancement techniques and coherent signal processing methods to optimize signals. We further implement different localization techniques to assess their limitations and strengths.

To quantitatively assess the performance of the proposed methods, DAS-based localization results are fused and compared with Automatic Identification System (AIS) data. This enables evaluation of localization accuracy, tracking consistency, and uncertainty as a function of vessel range, signal strength, and environmental conditions. By quantifying these limitations and associated uncertainties, this study provides practical insight into the operational capabilities and constraints of DAS for maritime surveillance and monitoring applications.

How to cite: Bülow Pedersen, H., Heiselberg, H., Bostani Nezhad, K., Heiselberg, P., and Aalling Sørensen, K.: Localization and Tracking of Incoherent Ship Signals Using Distributed Acoustic Sensing, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19172, https://doi.org/10.5194/egusphere-egu26-19172, 2026.

Distributed Acoustic Sensing (DAS) is a novel earth observation. Existing telecommunication cables, named dark fiber, are usually utilized as part of DAS network. Due to its inherent feature, high-frequency ambient motion is recorded along the length of the fiber optic cable for long distances. Since the dark fiber cables were placed without given care certain issues namely  ground coupling, the data acquired are excessively noisy compared to the conventional sensors particularly on land. However, the cables lay on the ocean bottom prone to less urban noise source. On the other hand, tha traces from the the marine traffic and water ocean motions are apparen in the DAS time series. This study exhibits the potential of DAS data in monitoring the marine traffic near the shoreline in addition to earth observations. A fiber optic cable with NEC interrogator has been studying since May 2024. A long term data in marine traffic with regular and irregular events were significantly emerged from the huge data with proper detection techniques. 

How to cite: Dindar, A. A., Tunç, S., Kato, A., Özel, N. M., Ergün, T., Zhang, J., and Kaneda, Y.: Utilizing DAS data in Earth Observation and Marine Traffic Monitoring, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21518, https://doi.org/10.5194/egusphere-egu26-21518, 2026.

Accurate horizontal orientation of three-component seismic sensors remains a persistent weakness in field deployments, despite its direct impact on modern seismological observables and advanced processing methods. While orientation is routinely performed using magnetic compasses for practicality, the effective accuracy achieved in the field is often insufficient and leads to systematic errors that propagate into receiver functions, amplitude and polarization analyses, propagation direction estimates, waveform separation, and moment tensor inversion. Large-scale surveys of permanent and temporary networks report frequent misorientations at the multi-degree level, with a worldwide dispersion typically on the order of several degrees, and a significant fraction of stations exhibiting errors well above 4–10°.

This contribution reviews current practices used to transfer the North direction to seismic stations, with emphasis on the real causes of misorientation in operational conditions: magnetic declination handling (including temporal drift of the magnetic pole), local magnetic disturbances (geology, infrastructure, reinforced concrete, metallic objects, and even the station itself), and the intrinsic limitations of visual alignment and reference transfer. Post-installation orientation estimation techniques, although valuable, often remain an additional and fragile processing step, not consistently preserved as station metadata and not always correctly exploited by end users.

A specific focus is placed on the orientation accuracy required to enable reliable array-derived rotation (ADR), which explicitly links field alignment to scientific performance. Building on published uncertainty analyses, we highlight that station misalignment is among the dominant contributors to ADR uncertainty, and that replacing magnetic compasses by gyrocompass-based procedures can reduce uncertainty by about one order of magnitude while extending the usable wavelength band toward longer periods—thereby unlocking higher-value array analyses.

To address this gap between scientific needs and field reality, we introduce the NJORD product line, designed specifically for seismic station orientation, from cost-effective solutions to higher-performance instruments compatible with static deployments and field. Among these, the Årian true-north finder is a compact, handheld static optical gyrocompass enabling rapid operation (alignment time < 4 minutes), one-day field autonomy, and an export-free approach requiring no aiding sensors (no GNSS, no magnetics), with a stated heading performance of 0.9° seclat RMS.

Årian is intented to shift field-deployment practice, enabling high-quality station orientation to become routine rather than exceptional, improving both data quality and operational efficiency at network scale.

How to cite: Guattari, F., Bercy, A., Gaillot, A., Ponceau, D., and Leray, V.: Orientation to True North for Seismic Stations: Review of Field Practices and New Instruments to Improve Accuracy and Operational Efficiency, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-22593, https://doi.org/10.5194/egusphere-egu26-22593, 2026.

The Geological Survey of Denmark and Greenland (GEUS) has recently expanded and modernised its land seismic equipment portfolio to support high-resolution seismic imaging and monitoring for research, public-sector tasks, and industry collaboration. This contribution presents the current land seismic capabilities at GEUS, acquired during late 2023 and early 2024, and highlights their applicability across a broad range of near-surface and crustal-scale investigations.

The source capability is centred on two modern vibroseis trucks (INOVA UV2), providing peak forces of up to 115 kN and a broad operational frequency range from below 1 Hz to 400 Hz. The units comply with EU Stage V emission standards, enabling environmentally responsible seismic acquisition in both urban and rural settings. Their flexibility makes them suitable for applications ranging from high-resolution near-surface surveys to deeper structural imaging and monitoring.

On the receiver side, GEUS operates an extensive wireless nodal system comprising 1200 Sercel WING digital field units equipped with one-component MEMS accelerometers. The system offers high timing accuracy via GPS synchronisation, bandwidth up to 400 Hz, and long battery autonomy (up to 50 days), allowing efficient large-scale 2D and 3D deployments as well as long-term passive or active monitoring campaigns. Additional sensors can be leased to further upscale acquisition geometries when required.

For rapid, high-resolution profiling, GEUS also maintains a 200 m landstreamer system (SeisMove) equipped with 100 three-component MEMS sensors at 2 m spacing. With bandwidths up to 800 Hz and sub-millisecond sampling options, this system is particularly well-suited for urban surveys, infrastructure studies, and near-surface characterisation where speed and data quality are critical.

These nodal capabilities are complemented by two ASN OptoDAS C01-S Distributed Acoustic Sensing (DAS) interrogators, enabling the use of standard fibre-optic cables as dense, multi-kilometre seismic sensor arrays. The DAS systems significantly extend the receiver portfolio by allowing rapid deployment on existing dark fibers, ultra-dense spatial sampling, and continuous monitoring. They are particularly well-suited for infrastructure monitoring, near-surface studies, and emerging applications in geothermal energy and CO₂ storage surveillance. GEUS also possesses 4 km of its own armoured fiber-optic cable with 4 individual 9/125 μ single-mode OS1-grade fibers for local, high-signal-to-noise ratio deployments.

Together, these complementary systems position GEUS to deliver state-of-the-art land seismic imaging and monitoring solutions, supporting research in seismicity monitoring, groundwater, geohazards, infrastructure, geothermal energy, and CO₂ storage, as well as national and international collaborative projects.

How to cite: Keiding, M., Sólheim, L., Mordret, A., Voss, P., Fønss Jensen, E., Larsen, T., Dahl-Jensen, T., and Rinds, N.: Land-seismic capabilities at GEUS, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-23075, https://doi.org/10.5194/egusphere-egu26-23075, 2026.

We report on the operations and organizational structure of the mobile Finnish seismic instrument pool FINNSIP (https://finnsip.fi) that is jointly owned and operated by five Finnish academic and research institutions. The pool infrastructure was funded by the Research Council of Finland through the FLEX-EPOS project under the FIN-EPOS umbrella. Funding supported the build-up phase from 2021 to 2024, after which the pool continues operating as a national research infrastructure.

The seismic instrumentation includes 46 Güralp broadband seismometers, 4 Güralp accelerometers, 1166 Geospace three-component short-period geophones-digitizers pairs, 71 SmartSolo self-contained three-component short-period geophone units, and 50 Geospace digitizers with cellular connection. This makes it probably the largest coherent mobile seismic instrument pool in Europe in the public sector. Instrument pools, when coupled with efficient data storage and transmission and powerful computing resources, provide strong support for research activities of various institutions. The mobile Finnish seismic instrument pool actively engages with ORFEUS/EIDA and the Geo-INQUIRE project, contributing to the development of community solutions for data discovery and accessibility. Nevertheless, even in developed countries, it remains challenging for a single institution to acquire and maintain a sufficiently large mobile pool of instruments and ensure sustainable data production and distribution. Here we report on the pool’s governance structure, project management, and the challenges encountered in the daily operations. We discuss example of domestic and international collaborative projects of temporary deployments to enhance data-driven subsurface and environmental applications. We report statistics of deployments for active or passive experiments that can range from a few days up to a few years. Over the past 5 years, the pool has supported more than 40 projects and generated more than 110 TB of raw and curated data. Short-period sensors are used in most projects, and a quarter involves deployments of more than 500 nodes, highlighting the demand and interest in large-number nodal deployments.

How to cite: Hillers, G., Courbis, R., Heinonen, S., Luhta, T., Mäkinen, J., Kozlovskaya, E., Moisi, K., Leveinen, J., Ding, Y., Komminaho, K., Oksanen, T., and Vänskä, J.: Operations and Organization of the Mobile Finnish Seismic Instrument Pool FINNSIP, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-23154, https://doi.org/10.5194/egusphere-egu26-23154, 2026.

Underwater passive acoustics enables the detection of a wide range of sounds from diverse sources, including seismic T-waves. To locate these events, hydrophone arrays are deployed, requiring arrivals to be identified across multiple synchronized sensors. While manual picking is reliable, it is labor-intensive, time-consuming, and prone to human bias. Automated picking using neural networks significantly increases the number of detected events but introduces challenges such as false positives. The next challenge is to associate detections of the same events to locate their source by trilateration.

The recently published TAPAAS algorithm (Raumer et al., 2025) generates candidate associations for these detections. However, among the produced candidates, there is a substantial number of overlapping and subsets of detections leading to different event locations. Relying solely on least-squares cost or RMSE is insufficient to address these conflicts. To overcome this, we developed a systematic approach combining relocation and a graph-based review of conflicting events, which we validated using synthetic event datasets.

When applied to real data from OHASISBIO and CTBTO arrays of hydrophones in the Indian Ocean, our method successfully built comprehensive catalogs of events. In addition, with a sufficient number of hydrophones (> 4), it is possible to check and improve the instrument synchronization. Generally, autonomous instruments are synchronized with a GNSS clock at deployment and upon recovery, allowing to measure the clock drift (~1-2s/yr); but when hydrophones run out of battery before recovery, the final time information is lost (i.e. drift unknonw). We also found that, sometimes, real-time hydrophone clocks are reset. Our approach allowed us to identify and determine these unknown clock drifts or clock offsets, then correct the arrival-times, and iteratively improve the trilateration.

Raumer et al. (2025). Geochem., Geophys., Geosyst. https://doi.org/10.1029/2025GC012572

How to cite: Safran, R., Raumer, P.-Y., Bazin, S., Briais, A., and Royer, J.-Y.: A Graph-Based Approach for Resolving Conflicts in Underwater Acoustic Event Association for Systematic and Improved Trilateration, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3816, https://doi.org/10.5194/egusphere-egu26-3816, 2026.

In April 2024, a major submarine eruption occurred at the I1 segment of the Southeast Indian Ridge (SEIR), near Amsterdam Island in the southern Indian Ocean (see Royer et al., EGU26-GD5.1 and Olive et al., EGU26-GD5.1). Luckily, the OHA-GEODAMS Autonomous Hydrophone (AuH) network had been deployed a few weeks before the event and efficiently monitored the eruptive event. This work presents results from this hydroacoustic monitoring together with an innovative automatic cataloging pipeline that allowed to obtain a full spatio-temporal coverage of the event.

Thanks to the low attenuation of low-frequency hydroacoustic waves, hydroacoustics is known to be an efficient way to monitor earthquake swarms. In this case, the GEODAMS network efficiently monitored the eruptive swarm and detected both direct (P- and S-phases) and indirect (T-phases) seismic waves, as well as H-waves generated by interactions between lava and seawater.

During the first weeks of hydroacoustic activity, more than 500 T-waves and 200 H-waves were detected and their sources located. This enabled a precise relocation of the early swarm of strong, teleseismically-recorded earthquakes (Mw~5) that had been registered in the ISC and GCMT catalogs. It also enabled the detection of smaller events that had been missed by land stations. However, this initial analysis relied on manual annotation, a process that is particularly time-consuming and hinders the possibility to build reliable and near-exhaustive catalogs over extended time periods. To overcome this limitation, a fully automated cataloging pipeline was developed (Raumer et al., 2024 & 2025), enabling systematic detection, association, and location of hydroacoustic signals associated with the swarm. This methodology enabled us to estimate key parameters of the eruptive dynamics, such as the precise timing of sequential dyking events, and subsequent lava outpouring, which was later revealed by diachronous seafloor mapping. Moreover, the seismological catalog showed to be complementary with vertical deformation measurements (detailed in Ballu et al., EGU26-SM3.2).

Beyond its own geodynamical significance, the 2024 I1 eruption constitutes a relevant case study to demonstrate the strong potential of an automated approach for hydroacoustic earthquake monitoring. Future applications of this generic methodology should enable the extraction of geodynamical insights from past and potentially large-scale AuH observatories.

Raumer et al. (2024). Seismica. doi: 10.26443/seismica.v3i2.1344

Raumer et al. (2025). Geochem., Geophys., Geosys. doi: 10.1029/2025GC012572

How to cite: Raumer, P.-Y., Royer, J.-Y., Sara, B., Lise, R., Jean-Arthur, O., Valérie, B., Anne, B., and Edgar, L. and the OHA-GEODAMS Scientific party: Making Hydrophones Speak: Semi-Automated Hydroacoustic Monitoring of a Seafloor Spreading Event, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5145, https://doi.org/10.5194/egusphere-egu26-5145, 2026.

Researchers continue to intensify their focus on the seafloor to gain a deeper understanding of the Earth's structure, tectonic processes, and potential hazards through the acquisition of ocean-bottom seismic (OBS) data. However, the unique challenges of deep-sea environments require innovative, purpose-built engineering solutions and robust manufacturing techniques to safeguard data quality, data completeness and system reliability, while meeting scientific objectives and optimizing ease of deployment. A range of cabled and autonomous ocean bottom sensing solutions is now available, supporting the global community’s study of underwater ground motion, its dynamic properties, and natural or triggered events on the seafloor.

This poster presentation provides a comprehensive overview of the engineering challenges in the domain of OBS platforms, highlighting the advancements in technology and capability solutions that the Nanometrics team has developed to address those challenges for various configurations and use cases. With proven technologies such as integrated kinematic gimbals for levelling at all landing tilt angles, an integrated MEMS gyrocompass for precise orientation, and designs certified for deployment depths of up to 6,000 m, this poster will demonstrate that seamless multidisciplinary data collection across diverse marine environments is now more accessible than ever before. Recent technological advances include customer deliveries of both SMART cable seismic instrumentation and an integrated Cabled OBS observatory, expanding options to support a wide range of application-driven sensing instruments and dataloggers. This continuous innovation aims to facilitate further understanding of the dynamic properties in these challenging deep ocean environments.

How to cite: Jusko, M., Allardice, S., Somerville, T., Bainbridge, G., and Perlin, M.: Ocean-Bottom Seismology: Next-Generation Technology Solutions for Marine Monitoring, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5534, https://doi.org/10.5194/egusphere-egu26-5534, 2026.

Amphibian geophysical experiments are transforming our ability to image Earth’s interior beneath the oceans, yet only a limited number of deployments have been designed at the scale and density required to address specific scientific questions from the crust to the mantle. In this invited talk, I synthesize results from two large ocean-bottom seismometer (OBS) experiments, at Santorini and the Galápagos, that demonstrate a range of science questions, spatial scales, depth ranges, and physical targets.

To image magma transport throughout the entire crust at an arc volcano, the PROTEUS experiment deployed a uniquely dense amphibian short-period array of 89 OBS and 65 land stations at the Santorini-Kolumbo volcanic system. The design combined active-source and passive seismic observations to resolve crustal magma plumbing and volcanic structure at kilometer-scale resolution. The data revealed the depth, volume, and melt extent of shallow magma accumulation beneath the Santorini caldera and demonstrated that caldera-collapse structures formed during the Late Bronze Age Plinian eruption continue to control present-day magma recharge, including the 2011-2012 and 2024-2025 inflation episodes. Imaging further showed how the evolution of regional extension during the Neogene to Quaternary formed a complex of faulted basins.  This tectonic system has interacted with magmatism, influencing magma pathways, storage geometry, and anisotropic crustal structure.

A key advance was the discovery of a small, high–melt-fraction magma chamber beneath the adjacent Kolumbo volcano at ~2–4 km depth using full-waveform inversion; an approach enabled by the first use of dense source–receiver coverage at a volcanic system. The smaller size of this eruptible magma reservoir meant it could not be detected using traditional travel-time tomography. More recent tomography leverages the large aperture of the array to extend the velocity structure to greater depths, revealing a mid-crustal magma storage region (8-15 km depth) laterally offset from both Santorini and Kolumbo, as well as deep accumulation of mafic–ultramafic material that thickens the crust beneath Santorini. This crustal framework is central to interpreting the February 2025 seismic swarm, whose migration pattern is consistent with a substantial magma intrusion interacting with extensional faulting.

In contrast, the Marine IGUANA experiment was designed to address mantle-scale questions, including plume-ridge interaction and lithosphere-asthenosphere coupling beneath the Galápagos. These questions require broad spatial coverage and long-duration recordings of natural earthquakes, motivating a 15-month deployment of 53 broadband OBS spanning a ~650 × 800 km region around the archipelago and the nearby Galápagos Spreading Center. Preliminary results reveal low-velocity mantle structures in both P- and S-waves associated with hot plume material, high-velocity material consistent with foundering lithosphere, and plume transport to both the eastern and western Galápagos spreading centers. In addition, spatial variations in seismic anisotropy indicate mantle flow due to absolute plate motion that is modified by the mantle plume.

Together, these experiments illustrate how science-driven experimental design enables insights into processes ranging from crustal magma systems to mantle dynamics, highlighting the power of dense, large-aperture amphibian OBS arrays to address Earth science questions across different scales.

How to cite: Hooft, E. and the PROTEUS team & Marine IGUANA team: Imaging Crustal Magma Systems and Mantle Dynamics with Large Amphibian OBS Arrays, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6002, https://doi.org/10.5194/egusphere-egu26-6002, 2026.

The SEASMO (SEismo-Acoustic Submarine Mediterranean Observatory) platform represents a major technological advancement in deep-sea multidisciplinary monitoring. Situated in the Ionian Sea at a depth of 3,450 meters, approximately 80 km SE offshore from Portopalo di Capo Passero, Sicily, the observatory was successfully deployed in October 2024 as part of the “Marine Hazard” project. Having now surpassed its first year of continuous operation, SEASMO has established itself as a cornerstone of real-time scientific data acquisition and environmental risk mitigation within the Mediterranean basin.

This sophisticated facility is the result of a strategic collaboration between the National Institute for Nuclear Physics (INFN) and the National Institute of Geophysics and Volcanology (INGV). The platform leverages the world-class deep-sea neutrino telescope KM3NeT/IDMAR infrastructure, utilizing two primary 100 km electro-optical cables to provide robust connectivity. This high-bandwidth link connects a submarine network of junction boxes equipped with Remotely Operated Vehicle (ROV) mateable wet connectors to an onshore laboratory, ensuring a stable power supply and instantaneous telemetry for uninterrupted data transfer.

A key feature of SEASMO is its complete integration into global and national monitoring frameworks. It is officially registered in the International Federation of Digital Seismograph Networks (FDSN) database under the station code MHPPL, operating as part of the KM3NeT Seismic Network (K3). Beyond its registration, the station’s data stream is directly integrated into the INGV seismic monitoring system, significantly extending the reach of the national seismic network into the offshore environment. Furthermore, the observatory is currently being incorporated into the operational workflow of the Tsunami Alert Center (CAT-INGV), providing essential offshore data that enhances the accuracy and timeliness of natural hazard alerts for the region. This latter is particularly important given that some of the most damaging historical earthquakes in the Mediterranean realm were generated by offshore active faults located close to the Sicilian coastline, including the “1693 Val di Noto earthquake” and the “1908 Messina earthquake” (M > 7). Nonetheless, the observatory is also located close to the poorly constrained Ionian Subduction Zone, whose potential to generate subduction-related megathrust earthquakes remains unclear.

The scientific payload of the station is specifically designed for high-resolution environmental and geodynamic characterization. Central to its mission is a 120-second broadband Ocean Bottom Seismometer (OBS) and a high-resolution hydrophone, capable of detecting frequencies with a 12.8 kHz bandwidth down to  1 Hz. This combination allows for the precise monitoring of seismic activity and the characterization of both natural and anthropogenic acoustic sources, providing critical insights into the impact of human activities on marine ecosystems. Additionally, the suite includes a Conductivity, Temperature, and Depth (CTD) sensor for tracking physical-chemical shifts in the water column, as well as detecting  high-resolution temporal  sea-level anomalies.

Throughout its operation, SEASMO has demonstrated the reliability of its automated data-processing routines, providing 24/7 real-time monitoring of the deep Mediterranean and offering a unique window into its geodynamics and acoustic soundscape, with considerable implications for marine science and hazard prevention.

How to cite: Sciré Scappuzzo, S. and the SEASMO Team: Introducing SEASMO: SEismo-Acoustic Submarine Mediterranean Observatory, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7119, https://doi.org/10.5194/egusphere-egu26-7119, 2026.

Distributed Acoustic Sensing (DAS) enables broadband strain measurements with exceptional spatial and temporal resolution, effectively transforming fibre-optic cables into dense sensor arrays. Its ability to operate continuously and cost-effectively in challenging environments makes DAS a powerful tool for geophysical, seismological, and oceanographic research. However, the lack of adequate transfer functions linking DAS strain measurements to true ground motion remains a key limitation for its reliable use in quantitative seismic analysis.

Project ECHO (Earthquakes, Currents, Hydroacoustics, and Oceanography) addresses this limitation by establishing local, empirically derived transfer functions for the GeoLab fibre. By deploying a suite of collocated reference instruments along the cable track, the project will characterize the contributions of ground motion, temperature, pressure, and oceanographic processes to the measured DAS strain, and also quantify cross-sensitivities between these parameters.

To constrain strain and ground motion, two ocean-bottom seismometers equipped with additional pressure and temperature sensors will be deployed at water depths exceeding 2000 m, complemented by a land-based seismic station. Oceanographic variability will be monitored using acoustic Doppler current profilers (ADCPs) and conductivity, temperature, depth (CTD) sensors. Hydrophones will serve as an independent reference for calibrating the DAS response to underwater acoustic signals, including marine mammal vocalizations and anthropogenic sound sources, with complementary confirmation of low frequency signals from the ocean-bottom seismometers.

Expected outcomes include segment-specific DAS transfer functions, a DAS-derived seismic catalogue, detection and classification of marine mammals from their vocalizations, the development of DAS-based proxies for sea-state variability, and an assessment of the feasibility of shear-wave splitting analysis given the existing cable geometry. Together, these results will advance understanding of regional seismicity, ocean dynamics from the coastal zone to the deep basin, and underwater acoustics, while substantially enhancing the scientific utility and applicability of DAS in Earth and marine sciences.

The richness of the multi-parameter datasets acquired within ECHO create significant opportunities for interdisciplinary collaboration, providing a unique framework for joint studies across seismology, physical oceanography, marine ecology, and fibre-optic sensing. This integrated approach is intended to foster collaboration within the broader scientific community and to stimulate the development of novel methodologies and applications beyond the immediate scope of the project.

All datasets generated by ECHO will be released under F.A.I.R. principles at the conclusion of the project.

This work is supported by the ECHO project (DOI: 10.54499/2024.13655.PEX), ARDITI - Agência Regional para o Desenvolvimento da lnvestigação, Tecnologia e lnovação, the SUBMERSE EU project (GA 101095055), and FCT, I.P./MCTES through national funds (PIDDAC): LA/P/0068/2020 (DOI:10.54499/LA/P/0068/2020), UID/50019/2025 (DOI: 10.54499/UID/50019/2025), UID/PRR2/50019/2025 (DOI: 10.54499/UID/PRR/50019/2025).

How to cite: Loureiro, A., Schlaphorst, D., Corela, C., Matias, L. M., Peliz, Á., Ferreira, R., Pereira, A., and Reis, J.: ECHO project: Multi-parameter calibration of the GeoLab DAS fibre, Madeira, Portugal, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7689, https://doi.org/10.5194/egusphere-egu26-7689, 2026.

Vertical ground deformation is a key parameter to document magmatic or tectonic processes occurring below the surface, such as magma movement, tectonic strain build-up or seismic rupture. When occurring underwater, vertical deformation can be quantified using ocean bottom pressure gauges. Although these gauges have an intrinsic resolution sufficient to document sub centimeter movements, the identification and precise quantification of vertical displacements face several challenges due to instrumental drift modeling, ocean dynamics or instrumental artefacts or orientation related sensitivity.

In 2024 and 2025, in the framework of the GEODAMS project (see related presentations Royer et al. and Olive et al., in session EGU26-GD5.1), we carried out repeated bathymetric surveys and deployed an A0A gauge - a bottom pressure recorder with a built-in drift controlling system - along with networks of seafloor acoustic transponders and moored hydrophones (Raumer et al., EGU26-SM3.2). This experiment allowed us to detect and characterize a major magmato-tectonic event which, luckily, occurred 2 months after the initial bathymetric survey and instrument installation. The detailed description of this unique event will be given in the presentation by Royer et al. (EGU26-GD5.1).

Our presentation will focus on the interpretation of the recorded pressure variations to derive a chronicle of vertical deformation, before, during and after the magmato-tectonic event. Although the instrument recorded a spectacular total subsidence close to 4 meters, the precise quantification of the deformation through the submarine eruption requires a precise modeling of the instrumental drift and changes in calibration parameters, which may be affected by likely changes in gauge orientation induced by the seafloor deformation.

How to cite: Ballu, V., Testut, L., Dausse, D., Royer, J.-Y., Olive, J.-A., Tranchant, Y.-T., Joyard, R., Laurent, A., Bazin, S., Retailleau, L., Briais, A., Raumer, P.-Y., and Lenhof, E. and the OHA-GEODAMS: Challenges and rewards: a chronicle of vertical displacements during a seafloor spreading event at the Southeast Indian Ridge, from a seafloor pressure A0A recorder, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9061, https://doi.org/10.5194/egusphere-egu26-9061, 2026.

The İzmit Basin, located at the easternmost part of the Sea of Marmara, constitutes a key segment of the North Anatolian Fault system where the main fault strand bifurcates into northern and southern branches. This structurally complex transition zone is characterized by interacting fault segments and distributed deformation, making it a critical area for investigating microseismic activity.  The dataset analyzed in this study was acquired by an ocean-bottom seismometer (OBS) deployment in the İzmit Basin. Eight OBS stations equipped with 4.5 Hz geophones were deployed in September 2023 and recovered in July 2024, providing approximately 10 months of continuous three-component seismic recordings. The instruments recorded data at a sampling rate of 100 samples per second, enabling the detection of local microearthquakes with magnitudes down to approximately M 0.2.

Initial earthquake locations reveal dense microseismic activity distributed along both the northern and southern branches of the North Anatolian Fault, as well as within intervening fault segments and fracture zones connecting these branches. These preliminary patterns highlight active deformation within the basin but exhibit significant scatter in both epicentral location and hypocentral depth, primarily due to velocity-model uncertainties and sediment effects inherent to offshore OBS observations. Reliable earthquake locations, in both horizontal and vertical dimensions, are critical for resolving fault-specific seismicity patterns and for investigating depth-dependent variations in seismic parameters, such as b-values. To improve location accuracy, events were refined using VELEST (Kissling et al., 1994), which iteratively optimizes a one-dimensional P- and S-wave velocity model and relocates earthquakes by minimizing travel-time residuals.

The initial VELEST velocity models display substantial scatter at shallow depths, particularly within the upper 0–5 km, reflecting strong sediment-related velocity uncertainties typical of offshore OBS observations. Following iterative relocation and velocity model optimization, the final VELEST solutions show clear convergence of both P- and S-wave velocities. The most pronounced improvement is observed within the 3–12 km depth range, corresponding to the main seismogenic layer. At depths around 10 km, the optimized models yield Vp values of approximately 5.8–6.2 km/s and Vs values of approximately 3.3–3.6 km/s, consistent with previous studies in the Marmara region and indicating a physically meaningful, data-driven velocity structure. The VELEST-based iterative velocity model optimization substantially reduces velocity uncertainty within the seismogenic depth range (approximately 5–15 km), leading to improved hypocentral depth and overall location reliability. The refined locations delineate coherent patterns of microseismic activity along the fault branches and connecting structures, providing a high-resolution view of seismic deformation in the İzmit Basin.

How to cite: Ergün, Dr. T., Meral Özel, N., Yamamoto, Y., Takahashi, N., Polat, R., Teoman, U. M., Turhan, F., Dindar, A. A., Kaneda, Y., and Aitaro, K.: High Resolution Microseismicity in the İzmit Basin-Sea of Marmara from OBS Data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10836, https://doi.org/10.5194/egusphere-egu26-10836, 2026.

Real-time seismic monitoring infrastructure requires permanent power and data transmission and therefore largely relies on land-based seismological stations. As a result, regional seismic monitoring of marine regions remains challenging with respect to event detection thresholds and hypocentral location accuracy. Particular challenges in seismic event monitoring in Northern Germany are the occurrence of both natural and anthropogenic events in the southern Baltic Sea and therefore the discrimination of event source types and, for tectonic events, hypocentral depth determination of earthquakes in the wider Tornquist zone region. Beyond that, seismic monitoring gained in recent years importance as an additional tool to monitor critical infrastructure as well as controlled and uncontrolled detonations of unexploded ordnance (UXO).

Recent advances in both offshore infrastructure and seismic offshore instrumentation systems provide a realistic opportunity to facilitate deployments of continuous, real-time ocean-bottom seismometers to significantly improve monitoring capabilities in offshore regions.

In summer 2025, we deployed a three component, cabled broad band Ocean Bottom Seismometer (Trillium Compact OBS) at the Darss Sill in the western Baltic Sea. At the site, the permanent MARNET monitoring station, operated by the Leibniz Institute of Baltic Sea Research in Warnemünde on behalf of the Federal Maritime and Hydrographic Agency (BSH) provides the infrastructure for a prototype deployment of a real-time OBS system. Seismic data is transmitted in near real-time to the data center at Kiel University and included in the real-time monitoring system. Over the past months, the OBS installation has been continuously improved, to enhance the data quality. The MARNET station also records oceanographic, biological and meteorological parameters that can be used to improve our understanding of ocean generated microseism with in situ data. While models and theories exist for microseism generation in the deep ocean, its generation mechanisms in coastal areas is still an open question.

We present first data observations and correlations and discuss challenges and opportunities for future offshore seismological monitoring in Northern Germany coastal region.

How to cite: Wiesenberg, L., Weidle, C., Mars, R., Schönke, M., Uhlmann, S., and Meier, T.: A cabled, real time ocean bottom seismometer in the western Baltic Sea, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11722, https://doi.org/10.5194/egusphere-egu26-11722, 2026.

Autonomous free-fall ocean-bottom seismometers (OBS) offer flexible deployment and redeployment options. The Güralp Aquarius operates at any orientation without a gimbal and can wirelessly transmit state-of-health (SOH) and seismic data to the surface via an integrated acoustic modem. This enables monitoring and partial real-time data transmission without offshore cabling, reducing logistical complexity while maintaining data accessibility. These capabilities make Aquarius well suited to OBS pool operations, including the National Facility for Seismic Imaging in Canada, one of the most recent large-scale OBS pools to become operational worldwide. In contrast, cabled observatory systems provide continuous, high-resolution real-time data via direct connections to onshore infrastructure. The Güralp Orcus is a compact underwater seismic station integrating a broadband seismometer and strong-motion accelerometer in a single package. The slimline Güralp Maris offers additional flexibility, using the same omnidirectional sensor as Aquarius and supporting deployment on the seabed or within narrow-diameter subsea boreholes. Both systems are deployed globally within multidisciplinary observatories, including the Neptune array operated by Ocean Networks Canada.

SMART Cables represent a promising pathway to expanding cabled ocean observatory networks at significantly reduced cost. By combining seismology, oceanography, and telecommunications within a single system, large-scale monitoring networks can be developed through shared logistics and funding across industries. Güralp has demonstrated this approach through a successful wet demonstration in the Ionian Sea, conducted in collaboration with the Istituto Nazionale di Geofisica e Vulcanologia (INGV), representing the first practical deployment of this technology. Future projects will leverage low-power, low-volume sensor and data acquisition designs to support both commercial SMART cable initiatives and science-driven observatories.

How to cite: Lindsey, J., Watkiss, N., Calver, J., Kerkenyakova, A., Kitka, K., Hill, P., and Restelli, F.: Güralp Ocean Bottom Monitoring Solutions: Autonomous Nodes, Cabled Observatories and SMART Cables, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11873, https://doi.org/10.5194/egusphere-egu26-11873, 2026.

It is well known that seismic and electromagnetic experiments provide complementary information regarding lithology and fluid content. There are many published examples where marine seismic and electromagnetic experiments have been conducted on the same geological target and their results combined to provide new insights, and a few where such pairs of experiments have been conducted on coincident profiles. For both techniques, normally the most time-consuming and therefore most expensive component is seabed instrument deployment and recovery. In 2023 we conducted a novel experiment that involved acquisition of both seismic and electromagnetic data using seafloor instruments equipped with both types of sensor that therefore only needed to be deployed once. We deployed 35 Scripps instruments equipped with orthogonal horizontal electrodes, orthogonal horizontal coil magnetometers and 14 instruments from the UK Ocean Bottom Instrumentation Facility equipped with orthogonal horizontal electrodes, three-component fluxgate magnetometers, hydrophones and short-period three-component geophone packages. These instruments were deployed for c. two weeks at c. 4-km intervals along a transect across the continent-ocean transition southwest of the UK. During this time, we shot two different airgun sources and then conducted a controlled source electromagnetic (CSEM) experiment by towing Southampton’s deep-towed electromagnetic transmitter along the transect. Our transect coincided with a pre-existing seismic reflection profile collected with a 10-km streamer as part of an Irish government project (covering part of the transect) and a new profile with a 2-km streamer acquired during a test cruise by the National Oceanography Centre in 2024 (covering the remainder of the transect).

The resulting rich dataset allows a variety of analyses, some of which are only made possible by the multiphysics acquisition. Electromagnetic sensors are normally located by acoustic triangulation; these locations can be improved by using the airgun shots. The fluxgate magnetometers can be used as compasses to orient both the electromagnetic sensors and the horizontal geophones. At crustal level, the seismic and CSEM experiments allow us to obtain coincident P-wave velocity and resistivity models at similar resolution. Magnetotelluric data provide constraints on the mantle lithosphere and the lithosphere-asthenosphere boundary. Our seismic experiment is not designed to image at these depths, but there are some constraints from teleseismic events such as the 8^th September Morocco earthquake, which is well recorded across our array. Additional constraints may come from ambient noise cross-correlation of hydrophone data, which extend to 10-12 s period. Airgun shots and the Morocco earthquake are well recorded by magnetometer channels and our electromagnetic transmitter is well recorded by geophone channels; these coupled signals may yield further complementary information about Earth structure.

How to cite: Minshull, T., Bayrakci, G., Constable, S., and Ivey, K. and the UK Ocean Bottom Instrumentation Facility: On the value of multiphysics instrumentation on the seafloor , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12651, https://doi.org/10.5194/egusphere-egu26-12651, 2026.

Macquarie Island is a small sliver of uplifted oceanic crust and mantle lying between New Zealand and Antarctica, near the middle of the Macquarie Ridge Complex (MRC) - a transpressional plate boundary that divides the Australian and Pacific plates. Evidence for subduction initiation has previously been found at both extremities of the MRC, yet subsurface information on its central portion near Macquarie Island remains limited. In this study, we extract teleseismic waveform data from ocean bottom seismometers (OBSs) deployed around Macquarie Island between October 2020 and November 2021 in conjunction with a small number of temporary land stations and the permanent MCQ station in order to perform teleseismic tomography across the region. We apply a denoising scheme (ATaCR) to help extract as much usable data as possible from the noisy OBS recordings to produce a denoised dataset, which we intend on using to image subcrustal features in order to better understand the nature of this plate boundary.

How to cite: Li, Y., Rawlinson, N., Lü, C., O'Hara, T., Winder, T., Tkalcic, H., Coffin, M., and Stock, J.: Investigating the nature of the Australian/Pacific plate boundary beneath Macquarie Island using teleseismic tomography, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13418, https://doi.org/10.5194/egusphere-egu26-13418, 2026.

In northern Macaronesia in the Atlantic the Azores, Canaries and Madeira are three mantle upwelling surface expressions. While Madeira and the Canaries are located on old oceanic crust as intraplate hotspots (~135 Ma and 150-180 Ma), the Azores result from interaction between a hotspot and the Mid-AtlanticRidge and sit on younger crust (<50 Ma). Mantle upwellings can influence the subsurface discontinuity structure, including shallow crustal contributions, in an area that reaches beyond the locations of the islands into the offshore oceanic subsurface. Therefore, an observation of seismic crust and upper mantle properties in the whole region is key to understanding mantle dynamics and its effect on volcanism.

Since seismic stations are mostly found on land, offshore seismic measurements are more challenging and globally sparse. From 2021 to 2022 an OBS network was deployed in the oceanic region of northern Macaronesia by the UPFLOW project. This deployment enables, for the first time in this region, the investigation of discontinuity variations over a broad offshore area and their comparison with land-station observations from the islands.

With P-S receiver functions we obtain high-resolution point measurements of the Moho and further intercrustal and upper mantle discontinuities beneath 43 stations. We use frequencies of 0.4 to 3 Hz, but add higher frequency ranges to investigate the robustness. Furthermore, we compare frequency- and time-domain deconvolution techniques. Moho depths vary between 5 to 12 km, in some places over short length-scales, which could be linked to melt generation and composition changes. The crustal structure is more complex around the Azores region, reflecting the complex dynamics around the mantle upwellings. Oceanic noise levels, shallow subsurface structure, and thick sediments – particularly in the southeastern part of the region – pose additional challenges by causing receiver-function polarity flips and less well-defined converted-phase arrivals, which complicate interpretation.

This contributes to projects AMULETO (2022.06660.CEECIND) and GEMMA (DOI:10.54499/PTDC/CTA-GEO/2083/2021). This work is supported by FCT, I.P./MCTES through national funds (PIDDAC): LA/P/0068/2020 - https://doi.org/10.54499/LA/P/0068/2020 , UID/50019/2025,  https://doi.org/10.54499/UID/PRR/50019/2025, UID/PRR2/50019/2025.

How to cite: Schlaphorst, D., Silveira, G., Ferreira, A., Dias, N., and Miranda, M.: Moho Discontinuity Structure using Data from Land Stations and a Broad Atlantic OBS Network – Potential and Challenges, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13683, https://doi.org/10.5194/egusphere-egu26-13683, 2026.

Transfer function techniques are commonly used to remove tilt and compliance noise from broadband ocean bottom seismometers (BBOBS), particularly at periods longer than 10 s. Temporary deployments and cabled observatory installations utilizing BBOBS have increased in the past several years and involved a variety of different organizations globally. Furthermore, techniques utilizing non-traditional approaches, such as short-period ambient noise and the expanded use of horizontal-component seismic data, have also expanded. Here we present updates to the Automated Tilt and Compliance Removal (ATaCR) package, designed to help users both assess and remove noise from BBOBS projects. In particular, we have improved flexibility and assessment metrics to better meet the variety of approaches necessary for modern analyses. Notable additions include: (1) summary figures assessing tilt properties; (2) improved phase analyses to discriminate between microseismic and other sources of noise; (3) expanded flexibility to incorporate a variety of data and metadata, such as quality control metrics, and instrument handedness; (4) generalized TF removal algorithm to allow exploration of non-traditional correction sequences; (5) transfer function workflows for ambient noise imaging.

How to cite: Janiszewski, H., Russell, J., Audet, P., and Wu, Y.: Updates to and New Applications of the ATaCR Package for Tilt and Compliance Removal, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14832, https://doi.org/10.5194/egusphere-egu26-14832, 2026.

As part of the ERC (European Research Council)-funded UPFLOW project (2021–2027), we conducted the largest passive seafoor seismic experiment to date in the Atlantic, focusing on the Azores-Madeira-Canary Islands region, a unique setting with multiple unresolved upwellings. Between June 2021 and August 2022, we deployed 50 and recovered 49 ocean bottom seismometers (OBSs) across a ~1,000×2,000 area with ~110–160 km spacing. The data reveal a wealth of good quality seismic signals, enabling detailed imaging of mantle upwellings and opening avenues for interdisciplinary research in marine biology and oceanography.

We present our ongoing UPFLOW 3-D mantle model series, which uses a combination of massive global seismic datasets (millions of body wave travel-times, multimode surface wave dispersion data) with tens of thousands of measurements from UPFLOW's OBS waveforms. Various modelling approaches are used ranging from computationally efficient ray theory-based global inversions to finite-frequency and waveform approaches. We invert for a range of model parameters including shear- and P-wave speed, as well as for radial anisotropy.

Our images show complex 3-D structures from the upper mantle to the lowermost mantle. We also observe lateral links between low velocity anomalies beneath the Canary, Azores and Madeira Islands associated with positive radial anisotropy anomalies, which possibly indicate plume ponding and horizontal mantle flow. Moreover, low-velocity anomalies right beneath the Azores appear not to extend into the lower mantle; instead, they seem to spread laterally and connect to the lower mantle beneath the center of our study region. We discuss the resolution of our images and well as their scope and geodynamical implications.

How to cite: Ferreira, A. M., Harris, K., Witek, M., Chang, S.-J., Veisi, M., Tsekhmistrenko, M., and Miranda, M.: UPFLOW 3-D mantle tomography from crust to core, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15079, https://doi.org/10.5194/egusphere-egu26-15079, 2026.

The Lower St. Lawrence Seismic Zone, a paleorift zone in Quebec, is one of the most seismically active areas in eastern Canada. It underlies the St. Lawrence Estuary, which is itself an important habitat for endangered baleen whales. Between 2023 and 2025, we carried out two deployments of an amphibious seismic network with eight broadband ocean-bottom seismometers (OBS) and four coastal stations to monitor earthquakes and baleen whales. The OBS were deployed via free-fall; we present the required preliminary analysis including 1) clock-drift calculation using ambient noise cross-correlation, 2) tilt estimation using transfer functions, and 3) horizontal-component orientation using passing ships as a noise source. 

We apply machine learning pickers in combination with probabilistic earthquake phase association to construct an earthquake catalogue. Initial results indicate a detection rate of roughly twice that of the Canadian National Earthquake Database, although some detections may be rock blasts. Fin and blue whale calls are monitored using their characteristic internote intervals. An increase in vocal activity of both species is observed between the first (2023-24) and second (2024-25) deployments, despite the second being shorter in duration and missing the typically active month of October. In terms of per-OBS averages, fin whale activity increased from 63 active days to 97, and the number of detected calls per active day increased from 270 to 960, whereas blue whales remained more consistent in terms of active days (83 to 76) but the number of calls per active day increased from 106 to 173.

How to cite: Sirati, E., Plourde, A., Liu, Y., Chien, J., Darbyshire, F., Nedimović, M., and Zhang, M.: The Lower St. Lawrence Seismic Zone Ocean Bottom Seismometer Deployment, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15093, https://doi.org/10.5194/egusphere-egu26-15093, 2026.

The Queen Charlotte Triple Junction (QCTJ) is a complex plate boundary offshore British Columbia, Canada, linking the Cascadia subduction zone, the Queen Charlotte transform fault, and the Juan de Fuca Ridge through the Revere–Dellwood fault (RDF) system. This region accommodates marked changes in plate motion and deformation style over short spatial scales, yet its offshore structure and seismicity remain poorly constrained due to distant land-based seismic coverage. The Pacific Coast Seismic Assessment for Faults and Earthquakes (PACSAFE) is a multi-year Canadian ocean-bottom seismometer (OBS) program designed to resolve this plate-boundary transition through dense, multi-deployment offshore monitoring.

During Leg 1 (October 2023-July 2024), 26 broadband OBS from the National Facility for Seismological Investigations (Dalhousie University) were deployed across the QCTJ and adjacent segments of the RDF and Explorer Ridge, and continental shelf break. During the deployment, two of these instruments located along the continental shelf break as well as one on the RDF prematurely released from the sea floor.  All remaining 23 instruments were successfully recovered. Clock drift corrections were applied by the NFSI technical team, then orientations of all instruments were computed using local and teleseismic waveforms. We then applied a deep-learned based approach to phase picking, utilizing the OBSTransformer on the 3-component data, resulting in ~1.23 million P phases and ~3.11 million S phases. These arrivals were associated and located using the maximum likelihood approach, to generate a catalog of ~11,000 events. These events were then reduced using additional quality control parameters, and double-difference relocated using both pick times and waveform cross correlations. The resulting relocated catalog contains more than 5,000 of the most robustly relocated events.

The catalog reveals dense, previously unobserved microseismicity that delineates near-vertical fault strands, fault-perpendicular seismicity lineations associated with the Revere-Dellwood transform and multiple seismicity strands associated with the northern Explorer ridge and transform system. Seismicity extends from near the seafloor to ~20 km depth, with most activity concentrated between 5 and 10 km. We provide new observations with unprecedented constraints on deformation and plate-boundary partitioning within the QCTJ. Ongoing analyses of focal mechanisms, seismicity, and tectonic context will further refine models of seismic and tsunami hazard for Canada’s Pacific margin.

How to cite: Schaeffer, A. and Bostock, M. and the PACSAFE Team: High-resolution seismicity and fault imaging of the Queen Charlotte Triple Junction region from the PACSAFE Leg1 ocean-bottom seismometer network, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15321, https://doi.org/10.5194/egusphere-egu26-15321, 2026.

Seafloor sediment layers strongly influence seismic signals recorded by ocean-bottom seismometers through reverberations, velocity reduction, and waveform amplification. These effects can significantly bias seismic observations, limiting investigations of the oceanic crust and mantle. While direct drilling and active-source seismic surveys provide robust constraints on sediment structure, they are not always feasible in areas instrumented solely with passive seafloor seismometers. In this study, we estimate Rayleigh-wave ellipticity from both ambient seismic noise and earthquake recordings using a polarization-based H/V approach that isolates elliptically polarized Rayleigh waves. Rayleigh-wave ellipticity derived from OBS data shows clear correlations with water depth and sediment thickness. The combined ellipticity curves are inverted using the Neighbourhood Algorithm to constrain crustal shear-wave velocity structure beneath the OBS stations. Our inversion results indicate sedimentary cover thicker than ~2 km beneath the Madeira region, closer to the continent, whereas relatively thin sediment layers are observed near the Azores region. The resulting crustal thickness and shear-wave velocity models across the Azores–Madeira–Canaries region provide a useful reference for future seismic investigations in this region, including studies based on the UPFLOW data.

How to cite: Kim, T., Ferreira, A. M. G., Jones, G. A., and Chang, S.-J.: Constraining shallow S-wave velocity structure beneath the Azores–Madeira–Canaries region from Rayleigh-wave ellipticity analysis using UPFLOW data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15400, https://doi.org/10.5194/egusphere-egu26-15400, 2026.

Seafloor quartz-resonant pressure gauges manufactured by Paroscientific have long been used to measure vertical seafloor deformation, yet the gauge-specific drift characteristics continue to hinder the precise identification of geodetic and oceanic signals. Recent self-calibrating pressure gauge designs address this limitation by housing an internal reference pressure standard that can be isolated from ambient seawater. Scheduled valve operations switch between ambient and reference pressures, thereby enabling in situ drift calibrations that are isolated from oceanographic variability and seafloor deformation. We evaluate four designs of self-calibrating pressure gauges. The University of Washington (UW) A-0-A, commercial Sonardyne Fetch AZA, and commercial RBR BPRzero use the internal pressure of the instrument housing, measured by a barometer, as a reference. In contrast, the Scripps Institution of Oceanography Cabled Self-Calibrating Pressure Recorder (CSCPR) uses a piston-cylinder system to generate a controlled reference pressure near ambient pressure reading. A-0-A, Fetch AZA, and CSCPR instruments are deployed at 1500 m depth on Axial Seamount at the Central Caldera site of the Ocean Observatories Initiative Regional Cabled Array. Additionally, a Fetch AZA is deployed at 400 m depth on the Barkley Canyon, and a BPRzero is deployed at 2200 m depth on the Endeavour segment on the Ocean Networks Canada NEPTUNE cabled observatory.

 At Axial Seamount, the instruments are within 50 m of one another and are adjacent to a conventional pressure gauge in a bottom pressure and tilt (BOTPT) instrument that has been deployed since 2014 and is well aged, with a small drift rate inferred from repeated mobile pressure recorder surveys. Assuming ocean-derived pressure fluctuations and volcanic deformation are spatially coherent across all sensors, each self-calibrating gauge is evaluated by (i) comparing its data with the BOTPT and other gauges to quantify post-calibration residuals, and (ii) assessing internal consistency for the UW A-0-A and CSCPR, which have two pressure gauges inside each unit. Our comparison shows that the self-calibrating pressure gauges generally agree within ±1.0 hPa/year (or 1.0 cm/year of water column height change) over multiple years of deployment. The one exception is the UW A-0-A system. Early in two deployments (2019-2022 and 2024-present), its two gauges are inconsistent with one another and other instruments by up to several hPa, but this bias diminishes within a year, and the records converge. We are evaluating the cause of this transient behavior by analyzing A-0-A calibration sequences. At Barkley Canyon and the Endeavour segment, we evaluate Fetch AZA and BPRzero in the same manner as at Axial Seamount, using several co-sited gauges within 3 km. Both commercial gauges agree with the independent co-sited gauge within ±1.0 hPa/year after applying drift corrections. Comparing data from co-sited sensors enables us to investigate the subtle features of each sensor's performance. All designs demonstrate the potential to reduce instrumental drift to less than 1.0 cm/year. Further evaluation across a wider range of ambient conditions and deployment configurations is warranted.

How to cite: Dobashi, Y., Wilcock, W., Manalang, D., Smith, K., Dentoni, L., Zumberge, M., Sasagawa, G., Cook, M., Sullivan, C., Nooner, S., Chadwick, W., Beeson, J., Heesemann, M., and Schlesinger, A.: Evaluation of Self-Calibrating Pressure Gauges for Seafloor Geodesy: Instrument Comparison at Axial Seamount, Barkley Canyon, and the Endeavour Segment, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15728, https://doi.org/10.5194/egusphere-egu26-15728, 2026.

Oceanic transform faults (OTFs) are a major type of plate boundary on Earth. They are segmented into seismic and aseismic sections, with large earthquakes clustering on individual seismic segments. Some OTF segments even produce quasi-periodic large earthquakes, a behavior that is rarely observed on continental faults. Despite their remote locations, OTFs therefore provide a unique natural laboratory for understanding earthquake processes and faulting mechanisms.

Because OTFs are located on the seafloor, their internal structure and the ways in which they accommodate plate motion remain poorly understood. Over the past two decades, several ocean-bottom seismometer (OBS) experiments have been conducted along OTFs. Microseismicity recorded by these experiments has revealed important and sometimes surprising properties, including the behavior of aseismic sections and deep-penetrating seismicity that has changed our understanding of OTF rheology. To further resolve fault structure and slip behavior, we need earthquake catalogs with both high-resolution locations and high completeness, so that we can interpret fault geometry and identify transient slip processes.

In recent years, machine-learning phase pickers have been increasingly applied to OBS data, greatly improving the efficiency of earthquake detection. However, major challenges remain in building high-quality catalogs, including uncertainties in phase picking, limitations of location algorithms, and the effects of station spacing and velocity models. In this study, we systematically evaluate key components of the data-processing workflow that control catalog quality, including the performance of different phase pickers, the impact of station coverage and velocity models, and the behavior of different earthquake location methods. We also discuss strategies for building multi-tier catalogs that balance location accuracy and catalog completeness, providing datasets that are suitable for both structural interpretation and studies of fault slip behavior.

How to cite: Gong, J.: Microseismicity Detection and Location along Oceanic Transform Faults, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18632, https://doi.org/10.5194/egusphere-egu26-18632, 2026.

The global transition toward renewable energy has accelerated the deployment of offshore wind farms, where reliable mapping of the shallow subsurface is crucial for wind-turbine foundation design and installation. Detecting shallow structures and their lateral variability in marine sediments helps reduce geotechnical uncertainties related to subsoil heterogeneity, which strongly influence construction risks, performance, and costs. Among geophysical methods, Direct Current (DC) resistivity surveys are effective for lithological and structural characterization; however, their offshore application remains strongly limited by two main factors: current leakage into the highly conductive seawater column and the resulting reduction in apparent resistivity (ρₐ) sensitivity to seabed targets.

We present a marine acquisition enhancement technique that deploys an electrically insulating sheet directly above the electrode cable placed on the seafloor—a previously patented concept—to restrict vertical current leakage and promote lateral current diffusion along the seabed, thereby increasing electrical interaction with subsurface formations.

A numerical parametric study was conducted using the finite element method (COMSOL Multiphysics), followed by a comparison between insulated marine models and equivalent terrestrial reference models without seawater. The analysis investigated the effects of: (1) seawater depth (Wd) = 1–60 m, (2) seawater-to-subformation resistivity contrast (Cr) = 1.5–100, (3) Wenner electrode spacing (a) = 2–30 m, and (4) insulation width (L) = 1–100 m, symmetrically covering 30 electrodes with 2 m spacing (Fig. 1a).

For Cres = 1.5, conventional marine acquisition without insulation produces large ρₐ errors relative to the terrestrial reference (Fig. 1b), exceeding 20% at small spacings (a = 2–6 m) in shallow water (Wd = 1 m), and remaining between 50% and 60% for most spacings when Wd increases to 28–60 m. The insulating sheet significantly enhances sensitivity: in shallow water (Wd = 1 m), long sheets (L = 60–100 m) as well as intermediate coverage (L = 30 m) reduce ρₐ errors to less than 2% over the entire spacing range (a = 2–30 m), closely reproducing the terrestrial response. Shorter sheets (L ≤ 10 m) still reduce errors to below 10% at intermediate to large spacings.

As water depth increases (Wd = 28–60 m), resistivity recovery becomes partial: even long sheets reduce ρₐ errors to approximately 10% at a = 30 m, compared to more than 50% without insulation. The effectiveness of leakage control also decreases geometrically at larger spacings, where errors tend to stabilize or slightly increase. Furthermore, when Cres increases to 10, relative errors rise for all sheet lengths, reaching approximately 15% at a = 30 m even for L = 60–100 m. This behavior is attributed to stronger lateral current diffusion within the subsurface, which diminishes the influence of insulation on current pathways.

This comparative analysis confirms that seafloor insulation systematically improves ρₐ sensitivity relative to conventional marine acquisition, particularly for larger insulation coverage. The proposed technique provides an effective solution for enhancing shallow structural detection in offshore DC resistivity surveys and offers quantitative guidelines for optimized survey design in offshore wind-turbine foundation site investigations.

How to cite: tartoussi, N., Palma Lopes, S., and Leparoux, D.: Enhancement of Apparent Resistivity Sensitivity in Offshore DC Surveys Using Local Seafloor Insulation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19552, https://doi.org/10.5194/egusphere-egu26-19552, 2026.

In the last years, fibre optic sensing methods, in particular Distributed Acoustic Sensing (DAS), have been experimentally demonstrated to be suitable for monitoring Earth System parameters in submarine cables. The SUBMERSE project (SUBMarinE cables for ReSearch and Exploration) aims to develop blueprints for using telecommunication fibre optic cables as sensors by attaching fibre optic interrogators at selected landing stations, also building a data infrastructure for both temporary storage of full resolution data and permanent archival of reduced data sets.

We analyse data from interrogating the East/West oriented Ionian Submarine System cable from both end points, i.e., Preveza, Greece, and Crotone, Italy, along the same fibre. This cable is located to the north of the Kefalonia Transform Zone. This fault zone marks the western termination of the Hellenic subduction system and is one of the most active seismic zones in Greece, with large damaging earthquakes above M > 6 occurring every few years on average.

In addition to acquiring the full large dataset, decimated channels (~100 in each case) acted as virtual seismic stations offshore, acquired at the NOA datacenter for realtime monitoring purposes. We explored various approaches to automated phase picking and magnitude determination on a reduced data set as well as the full resolution data. We also consider other test sites on the Ellalink cable branches extending from Sines in southern Portugal and from Madeira.

In order to support these and other acquisitions, we have developed a range of tools that can be deployed at future sites. We have enabled real-time streaming of DAS data following the standard Seedlink protocol, which allows straightforward integration into existing workflows at earthquake observatories. We have developed an automated, machine-learning-based algorithm for analysing earthquake waveforms and assembled a benchmark data set of earthquake recordings from DAS cables worldwide with labels of P and S arrival times that can serve to further refine machine learning and other automated analysis approaches. Finally, leveraging the Xdas platform (Trabattoni et al. 2025), we have extended the popular SeisBench platform (Woollam et al, 2022) for machine learning in seismology with the ability to efficiently process dense DAS datasets with algorithms/machine learning models operating across either single or multiple channels.

How to cite: Tilmann, F., Evangelidis, C., Fountoulakis, I., Xiao, H., Münchmeyer, J., Heinloo, A., Strollo, A., Hillmann, L., Morten, J. P., Poggi, V., Parolai, S., Loureiro, A., Custodio, S., and Atherton, C.: Establishing continuous seismic monitoring by DAS interrogation of submarine telecommunication cables in Europe with the SUBMERSE consortium: tools and use cases, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19750, https://doi.org/10.5194/egusphere-egu26-19750, 2026.

The North Sea is transitioning from a hydrocarbon province to a wind-turbine and CO2-sequestration hotspot. The latter activity needs assurance that the CO2 is kept within the envisioned subsurface containers. Microseismic monitoring is one of the methods to track the movement of the injected CO2. Additionally, the North Sea experiences both tectonic earthquakes and events that are related to gas production. To sufficiently detect, locate and characterize the different events, onshore sensors do not suffice. The seismic network thus needs to be expanded into the sea. The dynamic marine environment, characterized by shallow water, active sand dynamics, and diverse marine life, makes it unfavourable for standard seismological deployments.

In this work, we report a series of experiments toward deploying a seismometer in the Dutch North Sea. The experiments were conducted at shallow water depth (~2m) near the NIOZ harbour at Texel Island and at ~10m depth in the Wadden Sea. We have chosen to deploy a broadband seismometer, so that the acquired data is not only useful for local monitoring, but also for crustal studies using teleseismic earthquakes and teleseisms. We tested two designs, one with a sea-bottom seismometer and the other with a posthole one. Furthermore, we developed a full prototype for a stand-alone station setup, which has been tested in the Wadden Sea. The sea-bottom seismometer performed well at long periods > 10s, while the posthole version showed a higher signal-to-noise ratio at shorter periods, making it more stable for local seismicity detection and localisation. Due to active sand dynamics, the sea-bottom sensor showed a higher temporal variation with the sensors' masses, requiring mass balancing more frequently than the posthole version. The posthole sensor remained clean, whereas the sea-bottom sensor acted as a reef for all kinds of marine life.

How to cite: Fadel, I., Akinremi, S., de Laat, J. I., Schmidt-Aursch, M. C., Thomas, C., and Ruigrok, E.: Towards permanent seismological monitoring in the Dutch North Sea: Progress and early results, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20259, https://doi.org/10.5194/egusphere-egu26-20259, 2026.

Developed after the breakup of Gondwana in the Early Cretaceous, the southeastern margin of Brazil comprises a large, economically important hydrocarbon province that includes two of the world’s most prolific offshore sedimentary basins: Santos and Campos. At the same time, this passive margin hosts the country's main offshore seismic zone, characterised by frequent small-magnitude earthquakes and occasional larger events (M ≥ 4.8), occurring on average every 20-25 years, mostly within the extended and submerged continental crust. The larger events include the 1955 Vitória earthquake (mb 6.1), the 1939 Tubarão earthquake (mb 6.0), the 1972 Campos earthquake (M 4.8), the 1990 Porto Alegre earthquake (mR 5.2), and the 2008 São Vicente earthquake (mR 5.2). Due to insufficient instrumentation, limited to a few distant and unevenly distributed coastal stations of the Brazilian Seismographic Network (RSBR), and the low frequency of larger events, both the origin of this seismic activity and the risk it poses to key offshore infrastructure remain poorly understood. To better understand the processes related to the seismic activity in southeastern Brazil, the Brazilian Seismographic Network at the Sea Project (RSBR-Mar) intends to improve coverage by deploying: i - six temporary broadband coastal stations; ii - five Ocean Bottom Seismometers (OBSs); and iii - eight Mobile Earthquake Recorder in Marine Areas by Independent Divers (MERMAIDs) between São Paulo and Espírito Santo states. To date, two temporary land stations have been installed in Espírito Santo in June 2025. During a cruise mission between late September and early October 2025, we deployed all five OBSs and eight MERMAIDs. The OBSs will be recovered by September 2026, after one year of data acquisition, at which time the batteries will be replaced and the OBSs redeployed for an additional year. MERMAIDs periodically surface and establish satellite communication, allowing them to send important data back to us, collected during each operational cycle. To date, we have collected 50 waveforms detected by the MERMAIDs. Although MERMAIDs were originally designed to record high-quality P-waveforms for teleseismic tomography, rather than to detect small local events, we will explore this possibility by requesting data stored in their one-year internal storage. If successful, MERMAIDs could improve the localisation of earthquakes recorded in continuous records from land stations and OBSs. First, we plan to automatically detect and locate earthquakes in continuous seismograms from land stations and OBSs using machine-learning-based algorithms (e.g., LOC-FLOW, MALMI). These preliminary locations allow us to identify which data windows to request for each MERMAID. Then, we can relocate the small earthquakes with the acoustic data. Finally, relocation methods may help us to delineate possible seismogenic offshore structures (e.g., HypoDD).

How to cite: Leite Neto, G., Nascimento, A., Coelho, D., Fontes, S., Maurício, Í., Alfonzo, A., Chaves, C., Bianchi, M., do Nascimento, A., França, G., and Rocha, M.: The RSBR-Mar Project: Monitoring Offshore Small-Magnitude Earthquakes in Southeastern Brazil with Marine and Land-Based Instrumentation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21055, https://doi.org/10.5194/egusphere-egu26-21055, 2026.

The Blanco transform fault system (BTFS) represents an evolving transform plate boundary in the Northeast Pacific Ocean. Its seismic behavior was captured with a dense network of 54 ocean-bottom seismometers (OBS) operated for one year. We created a high-resolution earthquake catalog based on different machine-learning onset pickers. The high-resolution seismicity catalog has 12,708 events outlining the current deformation and stress release. Seismicity indicates seismic and aseismic fault patches or segments, as well as complex along-strike and off-axis deformation, step-overs, and internal faulting within pull-apart basins. Along simple linear fault strands, earthquakes are localized within 2 km of the seafloor expression of the fault. By applying cross-correlation techniques, we identified bursts and repeaters along the BTFS. Most bursts have interevent times of less than five minutes. Repeaters are predominantly found in the west of the BTFS and along the Gorda Depression, and to a smaller extent beneath the eastern Blanco Ridge. Slip rates estimated from repeaters exceed the geological slip rate by approximately 4 times, suggesting small seismic patches that release their slip every ~4 years. Along the BTFS, fault coupling varies between fully locked and creeping. Local earthquake tomography shows elevated vp/vs values exceeding 2, suggesting significant serpentinization from seawater entering the transform faults, the oceanic crust, and the mantle. The study shows how modern machine learning pickers applied to OBS data yield essential insights into the physics of faulting along major plate boundary faults in time and space, including the partitioning of deformation between seismic and aseismic slip.

How to cite: Lange, D., Ren, Y., and Grevemeyer, I.: Seismicity, Repeating Earthquakes,and Tomographic Imaging of the Blanco Transform Fault System, Northeast Pacific, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21326, https://doi.org/10.5194/egusphere-egu26-21326, 2026.

The Macquarie Ridge Complex (MRC), situated at the the boundary between the Australian, Macquarie, and Pacific plates south of New Zealand, is presently understood to be a primarily transform boundary that originated as a mid-ocean ridge. Although the northern (Puysegur) and southern (Hjort) sections of the MRC are considered to be initiating subduction, the tectonic activity along its central portions (Macquarie and McDougall segments) remains ambiguous.

Macquarie Island is situated on the central segments of the MRC. The region is characterized by exceptionally rugged topography, a feature that can indicate the incipient subduction. Furthermore, the MRC has produced some of the most powerful intra-oceanic strike-slip earthquakes on record. Such significant seismic events in this area pose a potential tsunami hazard. Consequently, despite its isolation, detailed geophysical studies are warranted, which led to our deployment of an integrated network of ocean bottom seismometers (OBSs) and seismometers on the island (Tkalčić et al., 2020; 2021). The primary aim of our research is therefore to employ seismological methods to advance the comprehension of the tectonic development of the Australian-Macquarie-Pacific plate boundary.

From 2020 to 2021, we installed a network comprising five land-based stations and 27 ocean bottom seismometers on and around Macquarie Island along the MRC in the Southern Ocean. Utilizing data from the successfully retrieved OBSs and island stations, including the permanent station MCQ, we applied an adjoint waveform tomography technique to surface waves (5-20 s period) extracted from ambient seismic noise. This process, conducted over five iterations, yielded a 3-D model of S-wave velocity for the crust and uppermost mantle. Our starting 3-D model incorporated accurate bathymetry, a water layer, and an optimized 1-D velocity structure. For the seismic wavefield simulations required in the inversion, we employed the spectral element method with Specfem3D_Cartesian (Komatitsch and Tromp, 1999). The resulting S-wave velocity model shows a marked velocity increase at crustal and uppermost mantle depths, between 7 and 12 kilometers. The presence of relatively high S-wave velocities (>3.8 km/s) in the shallow lithosphere aligns with upper mantle rocks being located at unusually shallow depths along the ridge. The extensive distribution of this high-velocity material suggests that the uppermost lithosphere has not undergone significant deformation during the process of obduction.

Reference
Tkalčić, H., Eakin, C., Rawlinson, N., Coffin, M. F., & Stock, J. (2020) Macquarie Ridge [Data set]. AusPass: The Australian Passive Seismic Server. https://doi.org/10.7914/SN/3F_2020

Tkalčić, H., Eakin, C., Coffin, M. F., Rawlinson, N. & Stock, J. (2021) Deploying a submarine seismic observatory in the Furious Fifties, Eos, 102, https://doi.org/10.1029/2021EO159537 

Komatitsch, D., & Tromp, J. (1999). Introduction to the spectral element method for three-dimensional seismic wave propagation. Geophysical Journal International, 139(3), 806-822. https://doi.org/10.1046/j.1365-246x.1999.00967.x

How to cite: Tkalčić, H., Wei, Z., and Pham, T.-S. and the MRC Team: The Central Macquarie Ridge Complex in 3D from Adjoint Waveform Tomography Using Ambient Seismic Noise, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21714, https://doi.org/10.5194/egusphere-egu26-21714, 2026.

Strong-motion flat-files form the backbone of ground-motion modelling and seismic hazard assessment, yet India has long lacked a uniformly processed, comprehensive strong-motion database aligned with international standards. The study addresses this critical gap by developing a systematically processed ground-motion flat-file for earthquakes recorded across the Indian subcontinent, particularly for the Himalayan region where seismic hazard remains high and strong-motion data are sparse. The compiled flat-file includes 778 manually processed accelerograms from 195 earthquakes spanning the time period 1986-2018. These events, with moment magnitudes M_w ≥ 2.0 and epicentral distances R_epi < 600 km, were recorded at 254 seismic stations. The diversity of source-to-site configurations captured in this dataset enhances its applicability for developing regionally representative GMMs and for examining spatial variations in ground-motion characteristics. The waveform processing followed a consistent step-by-step protocol involving baseline correction, tapering, filtering, windowing and signal-to-noise ratio. The resulting flat-file contains a comprehensive suite of Intensity Measures (IMs) including Peak Ground measures (PGA, PGV, PGD), Spectral Acceleration (SA), Fourier Amplitude Spectrum (FAS), Effective Amplitude Spectrum (EAS), Arias intensity (AI), Cumulative absolute velocity (CAV), Significant Duration (SD), Acceleration Spectrum Intensity (ASI), Velocity Spectrum Intensity (VSI), and Characteristic Intensity (Ic). The reliability of the processed IMs was validated through residual analysis of FAS against an empirical model. As the first uniformly processed strong-motion flat-file for India that includes both horizontal and vertical components, this dataset provides a much-needed foundation for advancing ground-motion modelling and seismic hazard assessment in the region. Overall, this flat-file significantly strengthens the database, evaluates attenuation behaviour, conducting parametric and near-field ground-motion studies, and supporting site-specific seismic hazard assessments across the Indian region.

How to cite: Sharma, S., Omkar, O., Mannu, U., and Bora, S.: A Uniformly Processed Strong-Motion Flat-File for Crustal Earthquakes across the Indian Region, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-406, https://doi.org/10.5194/egusphere-egu26-406, 2026.

We introduce SeedPSD, a web service designed to visualize precomputed Power Spectral Densities (PSDs) as PDFs and spectrograms. SeedPSD was initially deployed at the Epos-France EIDA node and has since been deployed across 12 nodes as of 2025.

PSDs are routinely employed to characterize seismological station sites and perform data quality control. Furthermore, seismic noise is now widely used for tomography, crustal monitoring, and investigating ocean-solid Earth coupling. An efficient means of retrieving or plotting spectrograms and PDFs for a given site is valuable for identifying stations that have recorded specific noise events (such as oceanic storms) and, more broadly, for understanding the noise wavefield.

Looking ahead, we are considering proposing jointly with Earthscope a new FDSN standard for a PSD web service. This service would enable users to access PSDs, as well as PSD-derived spectrograms and Probability Density Functions (PDFs), computed from continuous seismic data. This initiative is supported by EIDA, EarthScope, and Geo-INQUIRE.

How to cite: Stehly, L., Panay, S., Bollard, P., Schaeffer, J., and Pedersen, H.: SeedPSD: a web service for analyzing seismological station noise levels, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5057, https://doi.org/10.5194/egusphere-egu26-5057, 2026.

Modern seismic networks have taken upon the crucial task of identifying geological risks to our increasingly dense civilization. Unfortunately, risk prevention and mitigation are still pending issues in many regions, specially in developing countries prone to geological disasters. This lack of proper monitoring infrastructure has been traditionally caused by the elevated costs of seismic equipment, mainly sustained due to commercial interests for its use in the oil&gas industry.

With the advent of the Open Seismometer design with performance comparable to state-of-the-art broadband stations, there is now the possibility to vastly increase seismic monitoring density in previously underrepresented regions. However, there were still unavoidable complexities regarding sensor deployment and infrastructure requirements that in practice limited their use to experts and large organizations.

We propose a new approach for a public Open Seismometer Network that simplifies the steps needed to deploy these optimized stations. The technical complexities of seismic infrastructure can be minimized now there is a base open-source seismometer design, as calibration is similar between units, and improvements are easily transferred back to all stations. Also, by leveraging open-source seismic software and P2P internet communication protocols it is possible to provide a distributed and scalable service without depending on a single server. And additionally, the transition to modern seismic formats such as Mseed3 and the cost of maintaining legacy Mseed2 streams can be simplified and reduced through an efficient data distribution and network design.

Building on open-source seismic solutions such as RingServer, combined with a novel P2P architecture, can finally provide a unified Seedlink service that transparently englobes a network of interconnected and independent servers. This enables field recovery of seismometer data in a reliable way, and at the same time can provide the real-time waveforms to a large body of users.

In today’s big-tech world, where it seems every technological solution is pushed to become a centralized subscription service, we believe it is specially crucial to make a stand and design a truly open seismic research environment. For this, the Open Seismometer Network aims to bridge the gap between novel low-cost seismic instrumentation and effective seismic networks.

How to cite: García-Saura, C. and Méndez-Chazarra, N.: The Open Seismometer Network as a Collaborative P2P Infrastructure that is Scalable, Distributed and Robust, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6008, https://doi.org/10.5194/egusphere-egu26-6008, 2026.

Subbotin Institute of Geophysics of the National Academy of Sciences of Ukraine is currently undertaking efforts to rebuild and modernize the national seismic network, aiming to improve earthquake monitoring and data quality across the country. With the support of international collaborations, including the ORFEUS Integration Grant and the SNEMU project funded by the U.S. Department of Energy, several seismic stations have recently been upgraded, and new stations have been installed. Within the framework of the ORFEUS Integration Grant, five seismic stations in the Carpathian region were modernized, and waveform data from these upgraded stations have been integrated into the European Integrated Data Archive (EIDA). In parallel, four new seismic stations were installed as part of the SNEMU project, expanding the national network coverage. Data from these newly deployed stations are openly available through the EarthScope Data Management Center.

As part of this modernization process, we are implementing routine seismic data processing and event analysis using the SeisComP software framework. This work includes waveform quality control, phase picking, event localization, and the development of a preliminary earthquake catalogue based on data from the upgraded stations. We present first results from a pilot dataset processed using recordings from the newly modernized network. Although the resulting catalogue represents an initial stage of processing and includes a limited number of detected events, it demonstrates the operational capability of the upgraded infrastructure and provides a first overview of recent seismicity in Ukraine.

Until recently, seismic data acquisition within the IGPH network was operational, while systematic data processing was limited due to the ongoing situation in the country. Data from upgraded stations were primarily archived without routine analysis. Prior to 2022, earthquake detection at IGPH relied on continuous manual processing using outdated software. As part of the current modernization efforts, we have initiated the transition toward a modern, automated seismic data processing workflow for earthquake detection and cataloging, with SeisComP selected as the core processing framework. In parallel, further expansion of the seismic network through the installation of additional stations is planned for the coming years, although specific timelines remain uncertain.

Subbotin Institute of Geophysics acknowledges funding support from the Data Integration Grant (ORFEUS, Geo-INQUIRE, Grant Agreement 101058518). Instruments and technical support were provided by GFZ German Research Centre for Geosciences and GIPP-GEOFON, GaiaCode and Université Côte d’Azur, CNRS, Laboratoire Géoazur. The SNEMU project is implemented in partnership with the Science and Technology Center of Ukraine, the U.S. Department of Energy, Lawrence Livermore National Laboratory, Michigan State University, and the EarthScope Consortium. T. Amashukeli is supported by the Philipp Schwartz Initiative for Researchers at Risk.

How to cite: Amashukeli, T., Farfuliak, L., Haniiev, O., Kuplovskyi, B., Petrenko, K., and Levon, D.: Upgrading the Ukrainian seismic network: first results of data processing and earthquake monitoring, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7421, https://doi.org/10.5194/egusphere-egu26-7421, 2026.

ORFEUS (Observatories and Research Facilities for European Seismology, www.orfeus-eu.org; orfeus.readthedocs.io; forum.orfeus-eu.org) is a non-profit foundation that coordinates the collection, archival, and distribution of seismic waveform (meta)data, services and products based on international standards. It serves a broad community of seismological data users, on behalf of the Euro-Mediterranean seismic networks and monitoring agencies (orfeus.readthedocs.io/en/latest/governance.html). ORFEUS core domains comprise: (i) the European Integrated waveform Data Archive (EIDA; orfeus-eu.org/data/eida), providing access to raw seismic waveform data and basic station metadata; (ii) the European Strong-Motion databases (orfeus-eu.org/data/strong), offering automatically/manually processed waveforms, advanced station/site metadata, and associated products ; and iii) the European Mobile Instrument Pools (orfeus-eu.org/data/mobile), facilitating access to seismic instrumentation for temporary deployments. Currently, ORFEUS services distribute waveform data from more than  33,000 stations, including DAS deployments and dense temporary regional experiments (eg., orfeus.readthedocs.io/en/latest/adria_array_main.html), with an emphasis on FAIR principles, open access, and high data quality. ORFEUS services constitute a core component of EPOS (www.epos-eu.org/tcs/seismology) and are seamlessly integrated into the EPOS Data Access Portal (www.ics-c.epos-eu.org). Access to data and products relies on state-of-the-art information and communication technologies, with a strong emphasis on web services (www.orfeus-eu.org/data/eida/webservices; https://esm-db.eu/webservices) enabling programmatic interaction. ORFEUS promotes transparent data policies and licenses and acknowledges the indispensable contribution of data providers. Ongoing activities focus on further development of existing services and on facilitating access to massive and multidisciplinary datasets through collaboration with global and regional initiatives, including the FDSN (www.fdsn.org) and EarthScope (www.earthscope.org),  as well as  through support from EC-funded projects (e.g., www.geo-inquire.eu). ORFEUS implements community-oriented services that include software and travel grants, a sustained training/outreach programme of webinars and workshops (www.orfeus-eu.org/other/workshops), and editorial initiatives supporting best practices in seismological data use and dissemination.

How to cite: Cauzzi, C., Crawford, W., D'Amico, S., Danecek, P., Evangelidis, C., Haberland, C., Kiratzi, A., Mascandola, C., Poggi, V., Roumelioti, Z., Schaeffer, J., Sigloch, K., Sleeman, R., and Strollo, A.: ORFEUS Seismological Data Resources and Community Services for the Euro-Mediterranean Region and Beyond, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12355, https://doi.org/10.5194/egusphere-egu26-12355, 2026.

Operational seismic networks are required to deliver reliable, low-latency data and timely earthquake information while managing heterogeneous instrumentation, long-term data archiving, and evolving network and operational demands. Beyond waveform acquisition, network operators require comprehensive visibility of network state-of-health (SOH), latency, and data continuity in order to maintain catalogue completeness and rapid response capabilities.

We present the Güralp Data Centre (GDC), a data acquisition and archiving platform designed to support operational network monitoring by providing centralized access to real-time and archived seismic data alongside integrated SOH analysis tools. GDC aggregates SEEDlink data streams from distributed stations and provides time-based data interrogation, long-term performance metrics, and automated SOH reporting to support both day-to-day operations and long-term network assessment. GDC can be deployed either on Güralp-hosted cloud infrastructure or within a customer-managed cloud environment, allowing users to balance redundancy, scalability, and data control requirements. When accessed through the Güralp Discovery platform, GDC enables automated instrument registration, remote firmware and configuration updates across networks, and traffic-light dashboards for rapid assessment of station health, latency, and outages. We discuss how these capabilities address common challenges in network seismology, including remote station management, near-real-time data availability, and long-term network performance monitoring, and consider lessons learned for improving the reliability and responsiveness of seismic monitoring systems.

How to cite: Kerkenyakova, A., Lindsey, J., Calver, J., Watkiss, N., Kitka, K., Hill, P., and Restelli, F.: Güralp Data Centre: Cloud-Based Data Acquisition and State-of-Health Monitoring for Seismic Networks, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12662, https://doi.org/10.5194/egusphere-egu26-12662, 2026.

The seismological community is producing ever more datasets, and datasets themselves increase in size and complexity. To support the community in FAIRly archiving this data, seismological data centers must find solutions for the storage and distribution of such large and diverse data. Different aspects related to the FAIRness of these datasets are being considered, among others during the GeoInquire project: appropriate metadata to describe experiments, access control for both data and metadata, and improved data formats for raw data. For the first two topics, two proposals are currently under review by FDSN Working Groups, and a positive decision is expected in the coming months. Access control is particularly important for distributed acoustic sensing (DAS) data, given the sensitivity of the detailed cable location information.

Concerning improved raw data storage, one possible way forward is to combine cloud storage, already in place at several data centers, with asynchronous services that provide direct links to data. Ideally, this will allow users to load specific segments of data from the cloud storage using high-level languages like Python and massively parallelize their processing. In such a setting, analysis-ready, cloud-optimized data might provide advantages over traditional miniSEED archives; previous studies also suggest advantages over the HDF5 formats commonly output by distributed acoustic sensing (DAS) devices and currently used to store the majority of DAS data.

In this contribution, we report on ongoing collaborative work to systematically evaluate cloud-optimized formats on commercial and on-premise (university or institute) cloud storage services to evaluate their usability for archival, distribution and analysis-ready access to large datasets. We tested I/O performance and storage aspects in Zarr, tileDB and Apache Iceberg on AWS and self-hosted S3 buckets. We will report on test results and a first scientific use case that utilizes data on an on-premise cloud. We will also compare challenges and opportunities of these storage solutions for DAS and for large-N nodal data.

The zarr format is already used in the Earth Science community and, combined with rich metadata and the xarray library, turns out to provide very user-friendly access and data slicing for DAS data. The TileDB format provides similarly good access and slicing, but is less well known in the Earth Science community and requires careful engineering of data ingestion and maintenance. With this presentation, we aim to provide updates on the ongoing collaboration, show first usage examples for scientific workflows, and to stimulate discussion about future seismological data archives.

How to cite: Schaeffer, J., Lecointre, A., Ermert, L., Hamilton, A., and Quinteros, J.: Testing cloud-optimized formats for future data archival & distribution, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12951, https://doi.org/10.5194/egusphere-egu26-12951, 2026.

Orientation of a seismic sensor is a key item of metadata required for analysis, however this is often inaccurate or missing from stations where the operator cannot directly access the sensor to see its orientation, for example, in boreholes or on the ocean bottom.  Nanometrics has developed a unique, patented system incorporating a miniature MEMS gyrocompass in the seismometer to determine its orientation with an accuracy of +/-0.5 degrees, which is generally not achievable by other methods.  Software in the data recorder automatically queries this orientation and incorporates it in the channel metadata.  Seismometers incorporating this gyrocompass also have self-leveling functionality, so the tilt of the data is known as well as the orientation.  This North-finding capability was initially developed for ocean bottom use and is now being integrated into down-hole systems using the Trillium 120 Borehole or Posthole seismometer with the Centaur data recorder.

This self-orienting seismic system requires only a single software command to find its orientation and automatically incorporate it in the metadata.  It removes any concern about the accuracy of the orientation, since the gyrocompass measures the direction of the Earth’s axis of rotation, which is the definition of geographic North.  This is an improvement over a magnetic compass, which measures magnetic, not geographic, North and may not accurately measure magnetic North due to disturbances from nearby magnetic material, such as the borehole casing.  Furthermore, this North-finding capability eliminates the complex operations currently used to control the orientation of borehole sensors, such as installing a “bishop’s hat” fixture with expensive equipment or cross-correlating seismic data between the borehole sensor and a surface sensor.  The orientation determined by the gyrocompass will automatically accompany the seismic data to the data center, so it will always be available to the user.  Optionally, a rotation transform can be applied in the Centaur data recorder, to supply data that is already aligned to North.

How to cite: Uhlmann, S., Jusko, M., Bainbridge, G., Perlin, M., and Allardice, S.: Solving the Down-hole Installation Problem: A Self-Orienting Seismic System, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13508, https://doi.org/10.5194/egusphere-egu26-13508, 2026.

As cross-disciplinary science becomes increasingly critical to understanding geophysical phenomena, a multidisciplinary approach is essential for integrating instrumentation and ensuring reliable and efficient data acquisition for successful scientific outcomes.

The scientific community requires adaptable solutions for the co-location of diverse sensor types. Deploying such instruments in remote, volatile environments while ensuring reliable, continuous data acquisition presents additional challenges. The complexity and cost associated with deploying, operating and maintaining remote stations are significantly increased if using multiple independent sensors, each with dedicated acquisition infrastructure. Recent efforts, such as the European Plate Observing System, aim to address this by integrating multidisciplinary geophysical applications into unified and efficient deployments.

Modern seismic data loggers, such as the Nanometrics Centaur Gen5, support integration of a wide range of sensing elements, while maintaining ultra-low power consumption, precise timing, local data storage and reliable real-time data transmission. Enhanced capabilities regarding customization and edge computing allow the implementation of functionality tailored to meet specific monitoring objectives for unique station configurations.

A case study is presented for a multidisciplinary geophysical monitoring station that leverages these capabilities to enable comprehensive, reliable and efficient data collection. The multidisciplinary station configuration and end-to-end data pipeline, from remote sensing to science doorstep in the data center, are discussed.

How to cite: Allardice, S., Jusko, M., and Perlin, M.: Multidisciplinary Stations: A Next Generation Tool Kit for Geoscience, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13809, https://doi.org/10.5194/egusphere-egu26-13809, 2026.

Co-locating weak-motion seismometers with strong-motion accelerometers enables monitoring of seismicity at all scales, from the largest earthquakes to background-level microtremors.  However, a combined analysis depends on having comparable data from both instruments, installed at the same depth, accurately aligned in the same package, and precisely calibrated so they can produce equivalent data, i.e. the same ground motion velocity or acceleration signals after response correction.

We present data from recent earthquake sequences in the Hualien region of Taiwan, captured by dual downhole sensors (Cascadia Slim Posthole) in the Downhole Seismic Observation Network of Taiwan CWA.  In this example the seismometer and accelerometer signals match within 0.5% on average after response correction, allowing for the synthesis of a combined data stream with an unprecedented dynamic range of 220 dB.  Algorithms for optimal combination of the data are discussed and demonstrated.

This processing also enables a new quality assurance metric for calibration accuracy.  Previously, it has only been possible to verify this by running a calibration test procedure, typically no more often than once a year, since the test process is laborious and interrupts normal data collection.  However, in analyzing data from a dual instrument, the response-corrected amplitude ratio of the strong and weak-motion data streams can be continuously measured and reported as a state-of-health metric, to verify that the two instruments are operating correctly and measuring ground motion with the same accuracy in terms of sensitivity and frequency response.

How to cite: Catalano, N., Jusko, M., Bainbridge, G., Perlin, M., Somerville, T., and Allardice, S.: Processing Strong and Weak Motion Data from a Combined Instrument to Verify Calibration and Maximize Dynamic Range, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14367, https://doi.org/10.5194/egusphere-egu26-14367, 2026.

How to cite: Joo, H., L. Evans, P., Heinloo, A., Quinteros, J., and Strollo, A.: Station Explorer: the new interactive EIDA web tool for seismic data discovery , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16737, https://doi.org/10.5194/egusphere-egu26-16737, 2026.

Earthquake early warning has been a longstanding effort of the Kandilli Observatory and Earthquake Research Institute (KOERI), beginning in 1998 with the establishment of the Istanbul Earthquake Rapid Response and Early Warning System. Building on this experience and benefiting from advances in seismic instrumentation and data transmission technologies, KOERI has recently developed a new-generation early warning system
(EEWS) that is currently operational across Türkiye; however, given  the high seismic hazard associated with the North Anatolian Fault and its offshore segments in the Sea of Marmara, the present current station density, spatial coverage, and communication latency characteristics, the generation of reliable and timely public warning messages is presently feasible primarily in the Marmara region.
Earthquake detection, location, and magnitude estimation are performed using the Virtual Seismologist algorithm developed within the TRANSFORM project (funded by the European Union under project number 101188365; Cua, 2005; Cua and Heaton, 2007; Cua et al., 2009). Alerts are issued once seismic signals are recorded at a minimum of four stations. System performance is evaluated using the 2 October 2025 Mw 5.0 Marmara Sea earthquake as a case study. The first early warning alert was issued 8.4 s after the earthquake origin time, providing more than 20 s of warning for the Anatolian side of Istanbul prior to the onset of strong ground shaking. The installation of real-time seafloor seismic stations in the Sea of Marmara is therefore expected to substantially reduce detection times and further enhance the overall effectiveness of the Earthquake Early Warning System (EEWS) in the Marmara Region. These results demonstrate the capability of the KOERI-EEWS to deliver timely alerts and highlight its potential to enhance seismic risk mitigation and the protection of critical infrastructure.

How to cite: Meral Özel, N., Ergintav, S., Turhan, F., Konca, A. Ö., Aksarı, D., and Ergün, T.: Earthquake Early Warning System for the Marmara Region, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17285, https://doi.org/10.5194/egusphere-egu26-17285, 2026.

Seismic monitoring in areas of low natural seismicity is often complicated by anthropogenic signals which can have waveform characteristics similar to small magnitude earthquakes. In Ireland, quarry blasts make up the majority of signals detected by the Irish National Seismic Network (INSN), increasing analyst workload and making it difficult to produce reliable seismic catalogues.

Here we present an automated classification workflow for seismic events based on supervised machine learning methods that is suitable for operational use in discriminating between earthquakes, quarry blasts, and false detections. Instead of relying solely on waveform data, the classifier uses as input a combination of features derived from seismic waveforms plus event information, such as source parameters (e.g. depth, origin time, magnitude, number of phases) and location (e g. distance to the nearest known quarry).

Gradient-boosting classifiers, including XGBoost and CatBoost, were trained on a catalogue of more than 7,000 labelled events. Synthetic oversampling and hyperparameter optimisation were used to enhance robustness and reduce overfitting, and combining probabilistic outputs from multiple models was also explored. Our results show that by incorporating event information along with waveform features the accuracy and reliability of the classification was improved, with the final model achieving accuracies of more than 99% for quarry blasts and 95% for earthquakes, based on testing with unseen event data.

The trained classification tool has now been integrated into the INSN processing environment (SeisComP), and is currently being used to provide classification information for real-time alerts of automatically triggered events, as well as assisting in manual analysis. Although this work relies on the use of network-specific information, it demonstrates a transferable approach that can be used to integrate data-driven classifiers into operational geophysical monitoring systems. It also highlights the effectiveness of modern machine-learning techniques, supports the development of next-generation seismic data services and provides a practical example of how such tools can augment and complement traditional seismic monitoring workflows.

How to cite: Smith, P., Grannell, J., Moelhoff, M., and Bean, C.: Real-time classification of Irish seismic events using supervised machine learning, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18191, https://doi.org/10.5194/egusphere-egu26-18191, 2026.

The detection of nanoseismicity (very tiny earthquakes sometimes associated with small cracks in rock, also called acoustic emissions) is an important area of research aiding in the understanding of geophysical processes, hazard detection, material failure and human-driven nanoseismicity. The high frequency and attenuation of nanoseismicity require high-frequency monitoring within metres of the source to capture the event. This has made them difficult to monitor in conditions outside of small-scale lab experiments, in which failure is intentionally induced. The development of distributed acoustic sensing (DAS) as a new tool for seismic monitoring, however, has increased the feasibility of investigating such signals in the field due to its high temporal and spatial resolution. Manual picking of these events, while possible, is impractical for long-term deployments and for time-critical applications such as stability monitoring, which limits the utility of the technology. Automation of the detection of nanoseismic events within such data is therefore essential for the long-term processing of DAS data and real-time processing of data for use in stability monitoring.  

We have developed a pipeline for the automated extraction of nanoseismic events from DAS data, using a new, simple ratio technique called Spatial Short-Term Average (SSTA). The pipeline takes an input of DAS data and generates a series of windows within the data containing information about high amplitude signals relating to nanoseismicity.  

Using the automatically detected events, we labelled the windows to train a series of machine learning models to classify the different signals. Once trained, we evaluated the performance of the various models to select the most effective method for processing the collected data. The best performing models will then be tested at scale with the resulting classified dataset being plotted spatially along the length of the deployment to identify patterns of activity across space and time. 

How to cite: Seager, D., Johnson, J., Bie, L., De La Iglesia, B., and Milner, B.: Automatic detection and classification of Nanoseismicity in Distributed Acoustic Sensing data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-211, https://doi.org/10.5194/egusphere-egu26-211, 2026.

Optical interferometry on submarine fiber-optic telecommunication cables offers a transformative opportunity for offshore geohazard monitoring by providing continuous measurements of seafloor perturbation at useful intervals over trans-oceanic distances (Marra et al., 2022). We analyze a southwest Pacific subset of data from a section of the Southern Cross NEXT cable connecting Auckland (New Zealand) to Alexandria (Australia). Using only cable-based measurements, we image the seismic rupture kinematics of the 17 December 2024 Mw 7.3 Vanuatu earthquake, the largest seismic event recorded on this cable since its installation.

We analyze measurements of a section of cable more than 1,000 km in length and comprising 18 inter-repeater spans including the section that runs roughly parallel to the Vanuatu subduction zone and the adjoining section extending southward toward New Zealand. The earthquake produces clear and coherent arrivals in the optical frequency deviation recorded across multiple spans, with well-defined signatures visible in both time series and spectrograms. We first extract earthquake-related strain signals in the 0.1-0.3 Hz frequency band and apply the Multiple Signal Classification (MUSIC) back-projection technique to recover the source-time evolution of the rupture. The inferred rupture is predominantly bilateral and consistent with the USGS finite-fault solution, confirming that interferometric submarine cables can function as effective regional seismic arrays for rapid characterization of offshore earthquakes.

These results further demonstrate the capability of submarine fiber-optic cables to image earthquake rupture processes using high-frequency strain signals, providing valuable monitoring coverage, especially in instrumentally sparse regions such as the southwest Pacific. By resolving rupture kinematics directly, cable-based observations offer a pathway toward improved tsunami early-warning strategies that rely less on empirical magnitude–scaling relations, which are uncertain for large earthquakes. Planned upgrades of the interrogating laser will allow the performance of this approach to be assessed at lower frequencies, where cable-based observations may provide direct constraints on tsunami propagation and other long-period geophysical processes.

How to cite: A. Naeini, A., Fry, B., Marra, G., Tamussino, M., Grand, J., D. Eccles, J., van Wijk, K., Veverka, D., and Pandit, R.: Optical Interferometry-based seafloor cable Measurements for Rupture Imaging and Tsunami Signal Analysis in the Southwest Pacific, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-893, https://doi.org/10.5194/egusphere-egu26-893, 2026.

We present a physics-based point source earthquake early warning system using distributed acoustic sensing (DAS) data. All core modules of the system are based on physical principles of wave propagation, and models that describe the earthquake source and far-field ground motion. The detection-location algorithm is based on time-domain delay-and-sum beamforming, and the magnitude estimation and ground motion prediction are performed using analytical equations based on the Brune omega squared model. We demonstrate the performance of the system in terms of magnitude estimation and ground motion prediction, and in terms of real-time computational feasibility using local 3.1 ≤ M ≤ 3.6 earthquakes. This DAS early warning system allows for fast deployment, circumventing some calibration phases that require gathering local DAS earthquake data before the system becomes operational.

How to cite: Lior, I. and Ben Zeev, S.: Physics-based earthquake early warning using distributed acoustic sensing, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1594, https://doi.org/10.5194/egusphere-egu26-1594, 2026.

As a vehicle approaches the fiber-optic cable, the distributed acoustic sensing (DAS) records a broadband strain rate, which corresponds to propagating seismic waves at high frequencies (>1Hz) and to quasi-static strain fields at low frequencies (<1Hz). However, characterizing the subsurface media through quasi-static deformations remains challenging. Here, we propose a new method for imaging shallow urban subsurface structures using quasi-static strain waveforms, measured with fiber-optic cables. This technique utilizes the quasi-static waveform of a single DAS channel to generate a local 1D velocity model, thereby enabling high-resolution imaging of the underground using thousands of densely packed channels. We employed the Markov Chain Monte Carlo (MCMC) inversion strategy to investigate the depth range of inversion using car-induced quasi-static waveforms. The synthetic data demonstrates that the quasi-static strain field generated by a standard small car moving over the ground enables detailed imaging of structures at depths from 0 to 10 meters. Additionally, we conducted field experiments to measure the 2D shear-wave velocity model along a highway using quasi-static strain waveforms generated by a four-wheeled small car. The velocity structure we obtained is closely aligned with that derived from the classical surface-wave phase-velocity inversion. This consistency indicates that the inversion depth range is comparable to the simulation results, which confirms the applicability of this method to real data. In the future, we anticipate using the city's extensive fiber-optic communication network to record quasi-static deformations induced by various types of vehicles, thereby enabling imaging of the urban subsurface at a citywide scale. This will provide valuable insights for the design of urban underground infrastructure and for assessing urban hazards and risks.

How to cite: Tang, L., Bertrand, E., Stutzmann, E., Bonilla Hidalgo, L. F., Mohammed, S. A., Gélis, C., Hok, S., Lehujeur, M., Leparoux, D., Gugole, G., and Durand, O.: Quasi-static waveform inversion from DAS observations, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3915, https://doi.org/10.5194/egusphere-egu26-3915, 2026.

A seismic observation using Distributed Acoustic Sensing (DAS) using seafloor cable can provide spatially high-density data for a long distance in marine areas. A seafloor seismic and tsunami observation system using an optical fiber cable off Sanriku, northeastern Japan was deployed in 1996. Short-term DAS measurements were sporadically repeated since February 2019 using spare fibers of the Sanriku system (Shinohara et al., 2022). A total measurement length is approximately 100 km.  It has been concluded that measurement with a sampling frequency of 100 Hz, a ping rate of 500 Hz, gauge length of 100 m, and a spatial interval of 10 m is adequate for earthquake and tsunami observation.  From March 2025, we started a continuous DAS observation to observe seismic activity. When the continuous DAS observation was commenced, we developed quasi real time data transmission system through the internet. Because a DAS measurement generates a huge mount of data per unit time and capacity of internet is limited, decimation for spatial direction is adopted. In addition, data format is converted from HDF5 to conventional seismic data exchange format in Japan (win format). An interrogator generates a HDF5 file every 30 seconds. After the file generation, the telemetry system reads the HDF5 file, and decimates data for spatial domain. Then, the data format is changed to the win format and the data are sent to the internet. In other words, data transmission is delayed for a slightly greater than 30 seconds. Data with the win format can be applied to various seismic data processing which has been developed before. To locate a hypocenter using DAS data, seismic phases in DAS data must be identified. To evaluate performance of hypocenter location using DAS records, arrival times of P- and S-waves were picked up on the computer display for local earthquakes. Every 100 channel records on DAS data and data from surrounding ordinary seismic stations were used. Location program with absolute travel times and one-dimensional P-wave velocity structure was applied. Results of location of earthquakes were evaluated by mainly using location errors. Errors of the location with DAS data were smaller than those of the location without the DAS data. Increase of arrival data for DAS records seems to be efficient to improve a resolution. However, picking up signals for all channels (seismic station) manually are costly due to a large number of channels. To expand the location method, an improved automatic pick-up program using evaluation function from conventional seismic network data by seismometers for DAS data (Horiuchi et al., 2025) was applied to the DAS data obtained by the Sanriku system. As a result, arrivals time of P, S and converted PS waves can be precisely identified with high resolution. We have a plan to locate earthquakes using all DAS channels (seismic stations)  and surrounding ordinary marine and land seismic stations.

How to cite: Shinohara, M., Fukushima, S., Uehira, K., Asano, Y., Tanaka, S. S., and Otsuka, H.: Seismic data telemetry system and precise hypocenter location for distributed acoustic sensing observation using seafloor cable off Sanriku, Japan, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4163, https://doi.org/10.5194/egusphere-egu26-4163, 2026.

Combining traditional seismic networks with Distributed Acoustic Sensing (DAS) to record ground-motion on telecommunications cables provides new opportunities to study small earthquakes with unprecedented spatial and temporal resolution. Here we present a detailed study of an earthquake sequence offshore northwest of Kefalonia island, Greece that began in March 2024 and returned to background levels by November–December. The sequence was recorded by both a permanent seismic network for its duration and by DAS on a fiber-optic telecommunications cable between 1 - 15 August 2024.  The two-week DAS dataset provides continuous strain measurements along ~15 km of optical fiber between northern Kefalonia and Ithaki during a period that captured elevated seismic activity. Combining seismic station and DAS data reveals distinct physical features of the sequence that are not observable with seismic stations alone, including details of mainshock-aftershock clustering and well-resolved source spectra at frequencies of up to ~50 Hz for M < 3 events. The signal-to-noise-ratio > 3 at frequencies of up to 50 Hz observed on DAS waveforms for a representative group of events suggests consistency with typical earthquake stress-drop values that range from 1-10 MPa. It further suggests that DAS data may be used to augment detailed studies of microearthquake source parameters.

We apply semblance-based detection to DAS waveforms and manually inspect 5,734 earthquakes that occurred within ~50 km of the fiber to build an initial earthquake catalog. We then combine DAS and seismic-station data to locate 284 events with high signal-to-noise ratios and compute their local magnitudes with seismic station data to create a detailed subset of the initial catalog. We apply waveform cross-correlation to offshore DAS data for events in the detailed catalog to associate unlocated detections with template events and estimate relative magnitudes from amplitude ratios and further enhance the detailed catalog. This approach adds an additional 2,496 earthquakes (2,780 events in total) with assigned locations and magnitudes and leads to an enhanced catalog with completeness magnitude M_c = -0.5. Most earthquakes (2,718 of 2780) cluster within a ~5 km radius approximately 10 km offshore of northwestern Kefalonia and exhibit local rates exceeding 100 events per hour.

Our enhanced catalog provides a detailed spatiotemporal record of seismicity in a region with limited station coverage and demonstrates the effectiveness of integrating DAS with seismic networks for earthquake monitoring of active seismic sequences. Furthermore, it resolves details of mainshock–aftershock clustering that would have otherwise likely have been erroneously classified as swarm-like with standard monitoring, highlighting how observational resolution influences the interpretation of the physics driving earthquake sequences.

How to cite: Harrington, R. M., Bocchini, G. M., Bozzi, E., Roth, M. P., Gaviano, S., Pascucci, G., Grigoli, F., Biondi, E., and Sokos, E.: Using a hybrid seismic and Distributed Acoustic Sensing (DAS) network to study microseismicity in high spatiotemporal resolution offshore of Kefalonia Island, Greece , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4254, https://doi.org/10.5194/egusphere-egu26-4254, 2026.

The first commercially available fibre-optic Distributed Acoustic Sensing (DAS) system, Cobolt, was released in 2004, with early uptake driven by applications in perimeter security, pipeline monitoring, and upstream oil and gas operations. Although these deployments demonstrated the disruptive potential of DAS, it is only within the past five years that the geoscience community has widely embraced the technology, exploiting its ability to deliver continuous, high-fidelity measurements with exceptional spatial and temporal resolution.

Historically, commercially available DAS systems were optimised for industrial monitoring rather than scientific metrology. As a result, key requirements of geoscience applications—such as quantitative accuracy, extreme sensitivity, extended range, and robustness in challenging environments—were not primary design drivers. This situation is now changing rapidly as geoscience applications mature and expand. This contribution reviews the principal performance characteristics that define the suitability of modern DAS systems for geoscience research and examines how recent technological developments are addressing these needs.

Five performance parameters are of particular importance. First, the transition from amplitude-based, qualitative DAS to phase-based, quantitative systems has enabled true strain-rate and strain measurements suitable for metrological applications. Second, instrument sensitivity has improved by several orders of magnitude, with contemporary systems achieving pico-strain-level detection along standard telecom fibre. Third, measurement range—ultimately limited by available backscattered photons in pulsed DAS—has been extended beyond 150 km through the adoption of spread-spectrum interrogation techniques. Fourth, spatial resolution continues to improve, with gauge lengths of ≤1 m and sampling intervals of ≤0.5 m now routinely achievable, and further reductions anticipated. Finally, dynamic range remains a critical consideration for high-amplitude signals such as earthquakes; however, reductions in gauge length provide a clear pathway to mitigating cycle-skipping limitations, supporting the future use of DAS in Earthquake Early Warning (EEW) systems.

Alongside raw performance, the ability to quantify and compare DAS system capabilities has become increasingly important. Industry-led efforts have resulted in well-defined test methodologies and performance metrics, providing a common framework for objective evaluation of DAS instruments used in scientific studies.

Practical deployment considerations are also shaping system design. Reduced size, weight, and power (SWaP) enable operation in remote and hostile environments, while improved reliability, passive cooling, and environmental sealing facilitate long-term field installations. These advances are particularly relevant to emerging marine and subsea applications, where low-power, marinised DAS systems are required for seabed deployment.

Finally, the growing complexity of DAS instrumentation places increasing emphasis on software. Automated configuration, intuitive user interfaces, and integrated edge-processing capabilities are becoming essential to ensure that non-specialist users can reliably extract high-quality scientific data.

Together, these developments signal a transition in DAS from an industrial monitoring tool to a mature geoscience instrument, with continued innovation expected to further expand its role across solid-Earth, cryospheric, and marine research over the coming decade.

How to cite: Hill, D.: DAS design features critical to geoscience applications, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4295, https://doi.org/10.5194/egusphere-egu26-4295, 2026.

How to cite: Bajad, S., Bowden, D., Bharadwaj, P., Fern, E. J., Fichtner, A., and Edme, P.: Coherent Source Subsampling of Seismic Noise for Distributed Acoustic Sensing in the Swiss Alps, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4413, https://doi.org/10.5194/egusphere-egu26-4413, 2026.

Distributed Acoustic Sensing (DAS) provides dense, meter-scale ground-motion measurements along fiber-optic cables. However, developing ground-motion models (GMMs) from DAS data is challenging because observations are controlled by DAS-specific factors such as cable coupling, orientation, and channel correlation. In this study, we present the first regional, partially non-ergodic DAS-based GMM that explicitly identifies and quantifies cable-related contributions to ground-motion variability. We analyze strain-rate data from a 400-channel DAS array at the Milun campus in Hualien City, Taiwan, compiling peak strain rates and Fourier amplitudes (0.1–10 Hz) from 77 regional earthquakes (3<M<7, 45<R<170 km). Building on classical seismometer-based GMMs, we extend the variability framework to account for (1) cable coupling influenced by installation and environment types, (2) cable orientation, and (3) channel correlation inherent to DAS measurement principles and array geometry. Channel correlation is modeled using Matérn kernels parameterized by along-fiber and spatial proximity distances. The resulting DAS-based GMM shows magnitude-distance scaling comparable to classical models, while decomposing variability into physically interpretable components. Cable coupling emerges as a dominant broadband source of within-event variability, whereas orientation effects capture repeatable, frequency-dependent earthquake source radiation patterns. Modeling channel correlation significantly reduces channel-related standard deviations, demonstrating that treating DAS channels as independent observations biases uncertainty estimates. Overall, our results show that DAS-derived ground motions require a fundamentally different variability framework than that of classical GMMs, highlighting the importance of deployment metadata and correlation modeling. This approach provides a statistical and physical foundation for next-generation seismic hazard assessments using dense fiber-optic sensing.

How to cite: Lin, C.-R., von Specht, S., and Cotton, F.: What Controls Variability in DAS Earthquake Observations? Implications for Ground-Motion Models, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4603, https://doi.org/10.5194/egusphere-egu26-4603, 2026.

Monitoring fin whale (Balaenoptera physalus) vocalizations is of significant scientific importance and practical value for marine ecology, hydroacoustics, and geophysics. Conventional monitoring approaches, such as hydrophone arrays, ocean-bottom seismometers (OBS), and satellite tagging, are limited by sparse spatial coverage, potential biological disturbance, and high costs. Distributed acoustic sensing (DAS) is an emerging technology that utilizes submarine optical cables as dense acoustic arrays, providing opportunities for large-scale, high-resolution monitoring of whale vocalizations. Here, we reveal the wavefield features of fin whale vocalizations by integrating DAS observational data combined with numerical simulations. Three distinct features—Insensitive response segment (IRS), high-frequency component loss, and acoustic notch—were identified in the observed wavefield. DAS response analysis via ray-acoustic modeling indicates that the length of the IRS is positively correlated with the vertical source-to-cable distance, while the gauge length is responsible for the high-frequency loss in Type-B calls. Furthermore, wavefield simulations using the spectral-element method (SEM) demonstrate that the acoustic notches represent transitions between transmission zones of waterborne multipath waves entering the seafloor, exhibiting high sensitivity to the seafloor P-wave velocity, water depth, and topography. These findings not only enhance our understanding of the DAS-observed wavefields, but also highlight the potential of utilizing DAS and acoustic notches for ocean environmental parameter estimation.

How to cite: Wang, Q.: Revealing the Wavefield Features of Fin Whale Vocalizations Observed by Distributed Acoustic Sensing, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4625, https://doi.org/10.5194/egusphere-egu26-4625, 2026.

This study aims to develop a system for the identification of vessels, seismic events, and volcanic activity through analysis of the spatiotemporal characteristics of wavefields recorded by distributed acoustic sensing (DAS) using a submarine fiber-optic cable. DAS provides unprecedented spatial coverage and resolution, making it highly suitable for monitoring dense wavefield variations and anthropogenic activities, whereas traditional seismometers remain indispensable for quantitative seismic analysis and low-frequency observations. In this study, continuous DAS records acquired from a submarine fiber-optic cable located in the northeastern offshore region of Taiwan near Guishan Island, an active volcano. This region experiences frequent seismic activity due to the northwestward subduction of the Philippine Sea Plate beneath the Eurasian Plate. In addition, the passage of the Kuroshio Current, a warm ocean current, brings abundant fish resources, resulting in frequent activities of fishing vessels and whale-watching boats. Event detection is first carried out using the recursive short-time-average/long-time-average (STA/LTA) method which uses two time windows with different durations and computes the average signal amplitude within each window. When a signal arrives, the average amplitude within a short time window changes rapidly, thereby increasing the ratio of the short-time average to the long-time average. An event is detected when this ratio exceeds a predefined threshold and manual secondary inspected. However, low signal-to-noise ratios (SNR) can significantly reduce the sensitivity of STA/LTA-based detection, leading to missed events. To overcome this problem, signal processing adjustments were applied to enhance detection performance. To validate the detection performance, the detected ship-related events were compared with records from the Automatic Identification System (AIS), while earthquake events identified from the DAS data were compared with the earthquake catalog of Taiwan Seismological and Geophysical Data Management System (GDMS). Subsequently, a regression analysis of catalog magnitudes against hypocentral distance and maximum DAS-recorded amplitude was applied to determine the minimum detectable earthquake magnitude. The proposed framework demonstrates the potential of DAS as a complementary tool for offshore geophysical and maritime monitoring, providing a basis for future studies on vessel tracking, seafloor topography, and earthquake monitoring.

How to cite: Wei, Y. J. and Chan, C. H.: Application of Distributed Acoustic Sensing to Detect and Identify of Vessels and Natural Events in the Northeastern Offshore Region of Taiwan, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4712, https://doi.org/10.5194/egusphere-egu26-4712, 2026.

Ice streams are major contributors to ice sheet mass loss and critical regulators of sea level change. Despite their important, standard viscous flow simulations of ice stream deformation and evolution have limited predictive power, mostly because our understanding of the involved processes is limited. This leads, for instance, to widely varying predictions of sea level rise during the next decades.

Here we report on a Distributed Acoustic Sensing experiment conducted in the borehole of the East Greenland Ice Core Project (EastGRIP) on the Northeast Greenland Ice Stream. For the first time, our observations reveal a brittle deformation mode that is incompatible with viscous flow over length scales similar to the resolution of modern ice sheet models: englacial ice quake cascades that are not being recorded at the surface. A comparison with ice core analyses shows that ice quakes preferentially nucleate near volcanism-related impurities, such as thin layers of tephra or sulfate anomalies. These are likely to promote grain boundary cracking, and appear as a macroscopic form of crystal-scale wild plasticity. A conservative estimate indicates that seismic cascades are likely to produce strain rates that are comparable in amplitude to those measured geodetically, thereby bridging the well-documented gap between current ice sheet models and observations.

How to cite: Fichtner, A., Hofstede, C., Kennett, B., Svensson, A., Westhoff, J., Walter, F., Ampuero, J.-P., Cook, E., Zigone, D., Jansen, D., and Eisen, O.: Englacial ice quake cascades in the Northeast Greenland Ice Stream - Observations and implications of ice stream dynamics, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5156, https://doi.org/10.5194/egusphere-egu26-5156, 2026.

We present a back-projection based earthquake location method tailored to Distributed Acoustic Sensing (DAS) arrays, using short overlapping fiber segments and a combined P–S framework to reliably locate local earthquakes. A 66km quasi-linear telecommunication fiber in Israel was repurposed as a DAS array. We analyzed several local earthquakes with varying source–array geometries. We divided the fiber into overlapping 5.4 km segments and back-projected P- and S-wave strain-rate recordings using a local 1D velocity model over a regional grid of potential earthquake locations. Each grid point is assigned with P- and S-phase semblance, and the corresponding phase-specific origin times, associated with the timing of maximum semblance. Segment-specific P- and S-phase semblance maps and the difference between P and S origin times were combined through a weighting scheme that favors segments with spatially compact high-semblance regions. The objective is maximizing both P- and S-wave semblance and minimizing P- and S-wave origin time discrepancies. Results for the analyzed earthquakes reveal robust constraints on both azimuth and epicentral distance from the fiber, and demonstrate the ability to mitigate DAS-related artifacts associated with broadside sensitivity and reduced coherency. We demonstrated the potential of the approach for real-time earthquake location and showed its performance when only P-wave recordings are available, underscoring the method’s potential for future DAS-based earthquake early warning implementation.

How to cite: Noy, G., Ben Zeev, S., and Lior, I.: Earthquake Location using Back Projection with Distributed Acoustic Sensing with Implications for Earthquake Early Warning, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5259, https://doi.org/10.5194/egusphere-egu26-5259, 2026.

Etna is the largest, most active and closely monitored volcano in Europe,
making it a crucial study region for volcanology and geohazard assessment. In early
July 2019, a 1.5 km fibre-optic cable was deployed near the summit of Mount Etna
and interrogated for two months. The cable was divided into four main segments, two
of which point towards different active crater areas. Temporary seismic broadband
stations and infrasound sensors were also deployed along the cable. During the
experiment, three distinct eruptive events were recorded. The first two events are
characterised by a large number of explosions in the active crater area, together with
an increase in background tremor activity. The third event is characterised by a larger
increase in background tremor, but almost no explosions.

The continuous recordings are analysed in the frequency-wavenumber domain,
which reveals the features of the background tremor activity and the stacked transient
signals, such as explosions. During the first two eruptive events, the stack of
explosive sources is characterised by a non-dispersive arrival, travelling with
different apparent velocities along each segment, and a non-linear ground response up
to 25 Hz. These segments can be used as an antenna to estimate an average back-
azimuth for the explosions, which come from the same crater area during both
eruptive events.

Outside of the three eruptive events, the background tremor features two slow
dispersion modes, both well resolved on the raw recordings. The slowest mode is
affected by gauge-length attenuation at higher frequencies, due to its short
wavelength, but remains detectable up to 27 Hz, with group velocities as low as 170
m/s. These observations showcase the utility of simple, rectilinear geometries in
deployments despite their known shortcomings, such as in location procedures. For
known source regions, such as volcanoes, a well-oriented segment can leverage
continuous activity to record the incoming wavefield and extract dipersion curves
without the need to perform cross-correlations, simplifying the workflow.

How to cite: Latorre, H., Diaz-Meza, S., Jousset, P., Ventosa, S., Ugalde, A., Currenti, G., and Bartolomé, R.: Spectral analysis of background and transient signals at Mount Etna using rectilinear fibre-optic segments, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5274, https://doi.org/10.5194/egusphere-egu26-5274, 2026.

Distributed Acoustic Sensing (DAS) enables unused fiber optic cables in existing telecommunication networks, known as dark fibers, to function as dense arrays of virtual seismic receivers. Seismic waves generated by human activities and recorded by dense sensor networks provide an abundant, high-frequency energy source for high-resolution, non-invasive imaging of the urban subsurface. This approach enables detailed characterization of near-surface soils, sediments, and shallow geological structures with minimal surface impact, supporting applications such as groundwater management, site response and seismic amplification analysis, seismic hazard assessment, geothermal development, and urban planning. However, extracting coherent seismic signals from complex urban noise is challenging due to uneven source distribution, uncertain fiber deployment conditions, and variable coupling between the fiber and the ground. In particular, high-frequency range signals (e.g., above 4 Hz), needed to resolve shallow subsurface structures, are particularly difficult to recover. Two strategies can be used to address some of these challenges, by discarding poor quality seismic noise segments or by focusing on particularly favorable noise sources. In this study, we adopt the second approach and use vibrations generated by passing vehicles, particularly trains which are energetic sources that contain valuable high frequency information . Capturing and exploiting the seismic waves generated by these vehicles offers unique opportunities for efficient and high resolution urban seismic imaging.

We present an enhanced ambient noise interferometry workflow designed to exploit noise sources that are particularly favorable to the fiber geometry, i.e. transient and strong sources occurring at the edge of the fiber segment to be analyzed. The workflow is applied to traffic-dominated seismic noise recorded on a dark fiber deployed along a major urban road in Berlin, Germany. First, we select short seismic noise segments that contain signals from passing trains and then apply a frequency–wavenumber filter to isolate the targeted train-generated surface waves while suppressing other wavefield contributions. The filtered data is then processed using a standard interferometric approach based on cross-correlations to retrieve coherent seismic phases from ambient noise, producing virtual shot gathers. Finally, Multichannel Analysis of Surface Waves is applied to derive one dimensional velocity models. This workflow targeted on specific transient sources reduces computational cost while enhancing dispersion measurements particularly at higher frequencies. By stacking the responses from tens of tracked vehicles, enhanced virtual shot gathers can be obtained and inverted to improve shallow subsurface models. This can be achieved with only a few hours of seismic noise recording, which is challenging using conventional ambient noise interferometry workflows.

How to cite: Ehsaninezhad, L., Wollin, C., Rodríguez Tribaldos, V., and Krawczyk, C.: Enhancing High-frequency Ambient Noise for shallow subsurface imaging using urban ambient noise DAS recordings, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5880, https://doi.org/10.5194/egusphere-egu26-5880, 2026.

Distributed Acoustic Sensing (DAS) offers a promising approach for dense seismic recording in urban environments by repurposing existing telecommunication infrastructure. Athens presents an ideal setting for such an approach, as Greece is one of the most seismically active countries in Europe, and the Athens metropolitan area — home to nearly four million inhabitants — lies within a geologically complex basin whose vulnerability was demonstrated by the destructive 1999 Mw 5.9 Parnitha earthquake. Seismic hazard assessment requires accurate subsurface velocity models, but acquiring the data to build them in dense urban areas remains challenging.

We present results from a multi-fiber DAS experiment conducted in Athens, Greece, from 16 May to 30 June 2025, using four telecommunication fibers provided by the Hellenic Telecommunications Organisation (OTE). Two Sintela ONYX interrogators simultaneously interrogated the four fibers, which fan out from an OTE building with lengths of approximately 24, 38, 42, and 48 km, providing extensive azimuthal coverage of Athens. This makes the study one of the largest urban DAS campaigns ever performed.

Data were acquired in two configurations, a lower spatial resolution mode optimised for earthquake recording (~26 days) and a higher resolution mode for ambient noise interferometry (~19 days). To detect seismic events, we applied bandpass filtering followed by phase-weighted stacking across channels to enhance coherent arrivals. An STA/LTA (short-time average/long-time average) trigger was then used to identify candidate events. During the acquisition period, the National Observatory of Athens (NOA) recorded 2,645 events across the broader seismic network, of which 548 were detected on at least one fiber (368, 343, 328, and 322 on fibers 1–4, respectively). Detection capability depends on distance and magnitude — we achieve near-complete detection within ~20 km, while many events of ML ≥ 2 were recorded at distances exceeding 200 km. The array also captured small local events absent from the NOA catalogue, likely corresponding to local seismicity below the detection threshold of the sparser regional network. Characterising this unobserved local seismicity is one of the objectives of ongoing work.

For events within 50 km of the interrogator site, we pick P- and S-wave arrivals to constrain body-wave travel times. These picks are used to locate events in the NOA catalogue, which enables us to compare with network-derived hypocentres and allows us to assess potential improvement from the dense DAS coverage, before applying the approach to smaller events detected only by DAS. The travel-time data will also serve as input for 3D eikonal traveltime tomography to image subsurface velocity structure beneath metropolitan Athens.

How to cite: Almarzoug, M., Bowden, D., Melis, N., Edme, P., Bogris, A., Smolinski, K., Rigaux, A., Lohan, I., Simos, C., Simos, I., Deligiannidis, S., and Fichtner, A.: Multi-fiber Distributed Acoustic Sensing for Urban Seismology in Athens, Greece, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6600, https://doi.org/10.5194/egusphere-egu26-6600, 2026.

Traditional tsunami early warning systems (TEWS) are typically expensive, have limited real-time availability, require continuous maintenance, and involve long deployment times. The SAFE project aims to overcome these limitations by developing a new tsunami warning technology based on Distributed Acoustic Sensing (DAS), leveraging existing seafloor fiber optic cables. This approach offers continuous 24/7 monitoring, near-zero maintenance, faster response times, and ease of installation. The project includes contributions ranging from the development of a novel Chirped-pulse DAS interrogator (HDAS) with improved low-frequency performance to a novel post-processing software to obtain tide height from the measured seafloor strain and automatic detection and confirmation of a tsunami wave. All this has been implemented in a friendly user interface and is undergoing final evaluation by the tsunami warning authority in the NE Atlantic (the Instituto Português do Mar e da Atmosfera, IPMA).  

The validation is currently ongoing using the ALME subsea cable, which connects Almería and Melilla across the Alboran Sea. The interrogator has demonstrated the ability to detect swell waves with a maximum error of 20 cm in the deep sea and a post-processing response time of less than 90 seconds. It is expected that slower tsunami waves will yield more precise estimations of wave height.

Importantly, the technology could also successfully detect the 5.3 Mw earthquake near Cabo de Gata, Spain, on July 14, 2025, at a distance of only 40 km from the epicenter without major saturation. The extremely large dynamic range of the interrogator (approximately 10 times larger than a usual phase system) enables the system to monitor large-magnitude earthquakes without signal clipping. The SAFE system is capable of delivering critical seismic and hydrodynamic data within 5 minutes of an event, supporting early tsunami detection and rapid response.

How to cite: Preciado-Garbayo, J., A. Ramirez, J., Godino-Moya, A., Canudo, J., Gella, D., Garcia, J. M., Xie, Y., Ampuero, J. P., and Gonzalez-Herraez, M.: SAFE - Tsunami early warning system using available seafloor fiber cables with Chirped-pulse DAS, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6949, https://doi.org/10.5194/egusphere-egu26-6949, 2026.