We investigate earthquake clustering, a prominent feature of seismic catalogs, in terms of distribution of the number of triggered events as described by a branching process (Kagan and Knopoff, Phys. Earth Planet. Inter., 1976; Saichev et al., Pure Appl. Geophys., 2005, and references therein). According to recent literature (e.g. Shebalin et al., Geophys. J. Int., 2020, and references therein), the productivity of a magnitude m event is defined as the number of triggered events of magnitude above m-Δ, where Δ is a positive default value. For a magnitude m event, we distinguish between the number of its direct descendants and the total number of its descendants, denoted respectively by the random variables v and V, both depending on Δ. Empirical analysis often testifies in favor of the identity of the type of distribution of both quantities (v and V) associated with the main event, and hypothetically is exponential. The testing or substantiation of this hypothesis is important for modeling seismicity and presents a serious challenge for seismic statistics.

In the standard Epidemic Type Aftershock Sequence – ETAS – model (Ogata, Ann. Inst. Stat. Math., 1998), the distribution of v is Poissonian. Therefore we consider the general ETAS model adapted to any distribution of v and prove that the branching structure of the model excludes the possibility of having a common distribution type (for example, Poisson or exponential) for both v and V at once  (Molchan et al., Geophys. J. Int., 2022). The second theoretical result relates to the behaviour of the tails of the productivity distribution. We show that there is a fundamental difference in tail behavior of the V-distributions for general-type clusters and for clusters with a dominant initial magnitude:  the tail is heavy in the former case and light in the latter. The real data display similar behavior. Theoretical conclusions are also illustrated through the analysis of a synthetic earthquake catalog.

How to cite: Varini, E., Molchan, G., and Peresan, A.: Theoretical analysis of the productivity of seismic events, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-1897, https://doi.org/10.5194/egusphere-egu23-1897, 2023.

The features of earthquake clusters in two contactional domains, located in Northeastern Italy and North-Central Iran, have been investigated. The tectonic and seismicity of the two study areas, namely the Alps-Dinarides junction and the Alborz regions, are controlled by the convergence between the African and Arabian plates and the Eurasia plate. Both regions are characterized by a rather complex structural setting, mainly including reverse and strike-slip faulting systems, and by moderate to high seismic activity.

The nearest-neighbor approach has been used for the identification of the earthquake clusters in the space-time-energy domain. This approach permits for a data-driven identification of clusters so that, within multi-event clusters, the features of secondary and higher orders dependent events can be explored. The investigation of seismicity in Northeastern Italy is based on bulletins compiled at the National Institute of Oceanography and Applied Geophysics (OGS) in 1977-2018, while in North-Central Iran the dataset was extracted from the catalog compiled by the Iranian Seismological Center (IRSC) for the period 1996-2022. According to preliminary analysis of the used earthquake catalogs, two corresponding regions have been identified, where a satisfactory completeness level is assessed for events with magnitude greater than 2.0. Robust values of the scaling parameters, namely the b-value and the fractal dimension of epicenters, have also been computed and are used to calculate the nearest-neighbor distances and to identify the earthquake clusters.

The results obtained in the two regions confirm that the complexity of clusters structure depends on the tectonic, structural, and geophysical properties of the area. Moreover, the complexity measures, borrowed from network theory (i.e. the Centralization and Outdegree indexes), consistently capture the complexity of the identified clusters. Besides, in both investigated regions, the results allowed us identifying two macro-areas, which are characterized by different clustering features, namely: high complexity indexes,, which indicate simple (burst-like) structure of clusters, and low complexity indexes, corresponding to complex multi-level (swarm-like) structure of clusters. Specifically, we found that "swarm-like" (high complexity) sequences are prevalent along the thrust faulting Alpine and Central-West Alborz systems, whereas "burst-like" (low complexity) sequences prevail along the strike-slip Dinaric and Central-East Alborz domains.

How to cite: Peresan, A., Talebi, M., and Zare, M.: Characterization of Earthquake Clustering in Contractional Regions, based on Nearest-Neighbor distances and Network Analysis, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-1870, https://doi.org/10.5194/egusphere-egu23-1870, 2023.

The Dominican Republic, situated on eastern Hispaniola Island in the Caribbean, is subject to moderate to high seismic hazard mostly controlled by oblique convergence at the Caribbean/North American plate boundary. Offshore of the island, the North Hispaniola Trench (NHT) and Los Muertos Trough (LMT) subduction-like structures accommodate shortening, while crustal faults both onshore and offshore also take up some deformation. Historically, the Dominican Republic’s large cities, as well as those in Haiti (which shares Hispaniola) have been damaged by earthquakes, the worst of which required the population to relocate (e.g. the 1562 Santiago de los Caballeros earthquake).  Given the elevated hazard, the Dominican Republic was selected to engage in the Global Earthquake Model (GEM) Foundation coordinated USAID-funded “Training and Communication for Earthquake Risk Assessment” project, which aimed to improve earthquake risk assessment capacity in Latin American cities. This project and the collaborations that emerged were the basis for developing a seismic hazard model for the Dominican Republic.

The seismic hazard model is implemented in the OpenQuake Engine, and mostly uses GEM’s model-building tools and state-of-practice. Two main datasets were used for the seismic source characterization: a homogenized earthquake catalogue that benefited from local seismicity records contributed by the Servicio Geológico Nacional (SGN) and Universidad Autónoma de Santo Domingo (UASD), and an active faults database that combines GEM’s global database and one compiled by SGN during recent seismic hazard projects. Together, these datasets were used to constrain seismic source geometries and rates for active shallow crustal earthquakes, subduction interfaces and subduction-like thrusts, and intraslab earthquakes. Active shallow crustal sources were characterized as a combination of fault ruptures and off-fault (distributed) smoothed seismicity. Fault rupture geometries were pre-defined using the Seismic Hazard and Earthquake Rate In Fault Systems (SHERIFS) method, which allows multi-fault ruptures, incorporating epistemic uncertainty in the magnitude scaling relationship (and thus maximum magnitude), portion of earthquakes modelled on and off faults, and slip rates of two major fault systems. Additional uncertainty was considered in the assumptions used to smooth distributed seismicity rates. The NHT and LMT were also modelled using the SHERIFS method, while the other subduction sources were modelled using GEM’s more standard approaches (i.e. a single fault with complex geometry for the interface and pre-defined ruptures constrained to the intraslab volume). Two end-member magnitude frequency distributions were used for the Puerto Rico Trench interface: one assigning more moment to large magnitudes, and one obeying the Gutenberg-Richter relationship. For intraslab sources, epistemic uncertainty was captured in the assumptions for smoothing rupture probabilities according to past earthquakes. The ground motion characterization relied on residual analyses performed in past GEM projects, but replacing outdated GMPEs on subduction sources with more recent counterparts.

Hazard results generally reinforce former perceptions. In Santiago de los Caballeros, PGA reaches ~1g for 2% probability of exceedance in 50 years, controlled by the Septentrional Fault, while in the capital (Santo Domingo) PGA of ~0.5g is impacted by all tectonic region types, and includes contributions from moderate magnitude earthquakes (M_w 5-6).

How to cite: Johnson, K., Chartier, T., Pagani, M., Perez, Y., Guzmán, V., Betania Roque Quezada, M., and Yepes Estrada, C.: PSHA for the Dominican Republic, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-13313, https://doi.org/10.5194/egusphere-egu23-13313, 2023.

In any probabilistic seismic hazard analysis (PSHA), the computation of earthquake forecasting models is a fundamental step. A widely used approach is the smoothed seismicity, which uses seismic catalogs to produce earthquake forecasts in time, space, and magnitude. Early smoothed seismicity models, called fixed smoothing, used spatially uniform smoothing parameters such that the kernels were invariant to spatial variations in seismicity rate. However, recently developed adaptive smoothing methods spatially adapt the smoothing parameters according to the earthquake density. All these fixed or adaptive methods are mainly used in regions with complex seismic source characterization since they do not rely on geological, tectonic, or geodetic information, and they overcome the difficulties in characterizing and segmenting complex geological set-ups. Nevertheless, the standard smoothed seismicity approaches may not properly present the seismicity rates for complex seismotectonic areas.

In this study, we propose an innovative 3D approach for fixed and adaptive smoothed seismicity methods that can be advantageously exploited in all contexts with available well-constrained 3D fault models derived from high-quality seismic catalogs. This approach presents a 3D kernel in the algorithm to smooth the location of earthquakes on a spatial grid by considering the earthquake's depth and spatial coordinates. This allows the use of a three-dimensional grid built on a 3D fault model to represent the depth variations of the structure and also provide the rupture parameters. We tested our method with the Adriatic Basal Thrust (ABT) in eastern Central Italy, a regional active contractional structure with a well-constrained 3D fault model and a related high-quality location catalog.

The eastern Central Italy seismotectonic set-up is characterized by contractional active regional thrusts, such as ABT, representing the outermost and still active front of the Apennine fold-and-thrust belt and by coaxial extensional faults observable along the axis of the Apennine Chain. This complex framework shows different kinematic seismogenic sources overlapping at different depths and represents a perfect case study to test the 3D smoothed seismicity with fixed and adaptive methods. The 3D seismicity model was constructed for the ABT using a detailed catalog with completeness magnitude Mc ≥ 1.4 opportunely selected to identify ABT activity and declustered for the time-independent (Poisson) model.   We then applied the 3D kernel algorithm with the adaptive and fixed smoothed seismicity approaches to calculate the M ≥ 4.5 ABT earthquake rates. We also include a series of geological information regarding the depth, the fault strike, the dip angle, the seismogenic layers depth, and the rake of the slip direction that can be used for the seismic hazard analysis in the region. Finally, we presented the impact of the 3D smoothed seismicity model on PSHA in central Eastern Italy using OpenQuake software.

How to cite: Akinci, A., Pandolfi, C., Taroni, M., Lavecchia, G., and De Nardis, R.: Constructing a 3D Smoothed Seismicity Model for the Seismic Hazard Assessment in the Adriatic Thrust Zone, Italy, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-17432, https://doi.org/10.5194/egusphere-egu23-17432, 2023.

Site effects caused by unconsolidated sediments laying on the bedrock amplify or attenuate ground motions propagated to the surface. Site effect is unique site-by-site, and thus, makes analyzing actual attenuation characteristics of ground motions difficult. In the southern Korean Peninsula, 17 seismic stations administered by KMA and KIGAM were equipped with a pair of accelerometers; one is installed at the surface, and the other at the borehole (namely SB station). We estimate the site response functions of the stations using ambient noise data. First, the horizontal-to-vertical spectral ratios (HVSR) of the stations were calculated. Then, calibration ratios to adjust the amplitude of HVSR to that of surface-to-borehole spectral ratio (SBSR) were estimated and applied to the amplitude of HVSR. These amplitude-corrected HVSRs are used as the site response function to correct linear site effects in ground motions. To deconvolve the site effect of ground motions, we designed linear zero-phase FIR filters based on the site response functions. Then we divided the spectral amplitudes of the ground motions by the frequency response of the FIR filter. For the SB stations, site response functions of ten stations were obtained, and ground-motion data of 39 events with M_L > 2 were corrected using these site response functions. In the result, the peak ground motion (PGA) of corrected ground motions at the surface was reduced by 20-76% on average compared to raw ground motions. Comparing ground motions of the borehole and surface sensors, the corrected PGA of the surface was 1.8-4.9 times bigger than the raw PGA of the borehole. For the whole surface stations of 176, the site-response functions of 75 stations were estimated, and ground-motion data of 210 events with M_L > 3 were corrected by their site-response functions. We found that surface ground motions are deamplified to the level of borehole ground motions through the site effect correction.

How to cite: Ahn, B. S., Kang, T.-S., and Yoo, H. J.: Deconvolution of site effects in ground-motion using site response function derived from HVSR method, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-10882, https://doi.org/10.5194/egusphere-egu23-10882, 2023.

This work investigates how the uncertainty in the soil parameters influences the frequency content of the earthquake accelerograms that are expected to occur at a specific site. To this end, the Clough-Penzien power-spectrum is parametrized using random variables to describe the stochastic soil parameters and an artificial set of accelerograms is generated using the Spectral-Representation method. Subsequently, the equations of motion of a single degree-of-freedom oscillator are solved numerically for each earthquake scenario and the response spectrum is extracted. By comparing the derived response spectrum with the one obtained from considering deterministic soil parameters, useful conclusions are drawn pertaining to structural design and analysis.

How to cite: Michailidis, A.: Uncertainty quantification of seismic structural response due to randomness in the soil properties, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-17557, https://doi.org/10.5194/egusphere-egu23-17557, 2023.

At present, seismic prediction technology is still not available. Scientific decision-making and rescue after an earthquake are the main means of mitigating the immediate consequences of earthquake disasters. If earthquake emergency response level, fatalities, and economic losses can be estimated rapidly and quantitatively, this estimation will provide timely, scientific guidance to government organizations and relevant institutions to make decisions on earthquake relief and resource allocation, thereby reducing potential losses and more conducive to the implementation of social activities such as post-disaster reconstruction and reinsurance. To achieve this goal, a rapid earthquake disaster loss estimation method is proposed herein, based on a combination of physical simulations and empirical statistics. The numerical approach was based on the three-dimensional curved grid finite difference method (CG-FDM), implemented for graphics processing unit (GPU) architecture, to rapidly simulate the entire physical propagation of the seismic wavefield from the source to the surface for a large-scale natural earthquake over a 3-D undulating terrain. Simulated seismic intensity data were used as input for the earthquake disaster loss estimation model to estimate the fatality, economic loss, and emergency response level. The estimation model was developed by regression analysis of the data on human loss, economic loss, intensity distribution, and population exposure from the composite damaging earthquake catalog. We used the 2021 Ms 6.4 Yangbi earthquake as a study case to provide estimated results. The number of fatalities estimated by the model was in the range of 0–10 (five expected fatalities). The most probable economic loss range was 1–10 billion RMB (the expected economic loss was 4.862 billion RMB). Therefore, Level IV earthquake emergency response plan should have been activated (the government actually overestimated the damage and activated a Level II emergency response plan). The local government finally reported three deaths and 3.32 billion RMB economic losses during this earthquake, which is consistent with the model predictions.

How to cite: Li, Y., Zhang, Z., Wang, W., and Xin, D.: Rapid Estimation of Earthquake Disaster Loss Based on Physical Simulation and Empirical Statistics—A Case Study of the 2021 Yangbi Earthquake, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-7690, https://doi.org/10.5194/egusphere-egu23-7690, 2023.

The b-value in the frequency-magnitude relation is one of the fundamental seismological parameters to define an assemblage of earthquakes. This study focuses on the use of b-value as a precursor to know the chances of occurrence of major earthquake events in Sikkim and adjoining Himalayas. A catalogue containing 9192 earthquakes (M ≥ 0.30) recorded across Nepal, Sikkim and Bhutan Himalayas during 1980-2022 is considered for the present study. The study area has been divided into three blocks encompassing Nepal, Sikkim and Bhutan segments and the b-values are computed. The results show variations in b-value across these three blocks which might be associated with the differential stress pattern across the regions. Further, we map the spatial variation of frequency-magnitude distribution by dividing the entire region into 0.1⁰ x 0.1⁰ grids. The grids with less than 30 earthquake events are excluded to ensure the reliability of our results. The results suggest that the entire segment belongs to a high stressed region. The lowest b-values are mostly observed in Sikkim and western Nepal, reflecting the possible zones of future earthquakes.

How to cite: Maitreyi, , Singh, C., Singh, A., Uthaman, M., Dutta, A., Kumar, G., and Kumar Dubey, A.: b- value mapping in Sikkim and adjoining Himalayas, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-5879, https://doi.org/10.5194/egusphere-egu23-5879, 2023.