Pure and Applied Geophysics

Baptista, M. A.; Miranda, J. M.; Omira, R.

doi: 10.1007/s00024-025-03689-zpmid: N/A

On the 28th February1969, a massive earthquake stroke SW Iberia and NE Morocco triggering a tsunami recorded in more than 20 tide stations. The event occurred in the SW Iberian margin, the same seismogenic area of the 1st November 1755 mega event. Several studies were developed in the last 55 years to address its earthquake mechanism and the corresponding tsunami source. In some cases, the study of the 1969 event was also the base for inferences regarding the 1755 earthquake and indirectly to give some light on tsunamigenic processes related with the SW Iberian margin. In this study, we present a comprehensive review of the tsunami data and modelling, taking advantage from the great improvement that occurred on the quality of the bathymetric data, particularly on the shallow areas close to the tide stations. We used a larger set of tide-records than previous studies. All records were digitized from the original mareograms and processed according to modern standards. We address the possible landslide triggered at the NW coast of Morocco as the explanation of the tsunami observation at Casablanca. The new dataset combining both the earthquake and the landslide sources allows a better relocation of the tsunami source, enabling a quantitative comparison of the different source scenarios that have been developed for seismological research. The simulations presented here suggest that a thrust fault of 85 km × 20 km verging to the southeast is the best candidate to be responsible for the 1969 earthquake. The trace of this deep fault follows the one of the “Horseshoe Fault”, a northwest verging structure interpreted from the multichannel seismic data. Moreover, this deep structure may be accountable for both the 1969 event and the later 12th February 2007 M6 earthquake. Even more, the “Deep Horseshoe Fault” is a strong candidate to be the source of the 1st November 1755 event up to now elusive to multiple geological and geophysical studies.

Unal, Utku; Bekler, Tolga

doi: 10.1007/s00024-024-03636-4pmid: N/A

In regions with intense seismic activity like earthquakes, rapid detection and resolution of earthquake parameters and understanding seismic activity and mechanisms are important in terms of reducing possible risks. Since this process is left to the knowledge and experience of users to a great extent in the solution stage, human errors in detection of seismic wave phase arrival times may negatively affect the reliability of model studies. In this study, machine learning, which has been successfully applied to data in various seismological fields, was applied to earthquakes occurring in the Biga Peninsula, encompassing the most complicated tectonic elements of the north-western Aegean region, has high window seismicity. Results were evaluated using the waveform database for events recorded by local (COMU—Çanakkale Onsekiz Mart University) and national (KOERI—Kandilli Observatory and Earthquake Research Institute, AFAD—Ministry of Interior Disaster and Emergency Management Presidency) seismic networks observing activity linked to tectonism in the region under consideration with the originally trained model of the PhaseNet machine learning algorithm. Data contains 918 earthquakes recorded at 118 stations from May 2020 to the end of 2021. Compared to classic methods, the machine learning model used in the study provided more accurate results for detecting P and S wave phases. Also, epicentre calculations based on machine learning algorithm appear to be in better spatial agreement with the distribution of active faults than calculations based on handpicks. Although the original model of PhaseNet has not been trained with local data from Türkiye, study shows it is possible to get meaningful results by making adjustments on the algorithm or applying signal processing techniques on the data. Study suggests that enhancing machine learning algorithm with local training data can improve phase detection accuracy and epicenter prediction in seismic studies.

Qian, Jiaqi; Zhang, Wenbo; Zheng, Ao

doi: 10.1007/s00024-024-03645-3pmid: N/A

On 5 September 2022, a Mw 6.7 earthquake struck the eastern edge of the Tibetan Plateau, providing a unique opportunity to analyze the Moxi seismic gap of the Xianshuihe fault. We investigate the source rupture process of the 2022 Luding earthquake by combining the back-projection imaging and the joint finite-fault inversion of strong-motion and teleseismic waveforms. We find that the Luding earthquake primarily occurred along conjugate faults, propagating southeastward along the Moxi segment and westward along the Mozigou segment. The aftershock distribution complements the main slip areas, signifying the incomplete stress release during the rupture of the mainshock. Coseismic slips are concentrated in the shallow brittle crust, with low-velocity anomalies at the lower edge of the aftershock area, suggesting middle and lower crustal flow is crucial in guiding coseismic rupture propagation and aftershock distribution. Moreover, the static Coulomb stress changes suggest that the majority of aftershocks occurred in areas with increased stress, underlining the triggering effect of the Luding mainshock. Our study also indicates the increased seismic hazard on neighboring faults, particularly on the northern segment of the Xianshuihe fault, and the Longmenshan and Gongga faults. The numerical simulation of strong ground motion indicates that the overall Peak Ground Velocity (PGV) distribution follows an elliptical shape, with higher PGV values primarily extending southward. Given the steep terrain variations and sedimentary factors in western China, our findings underscore the urgent need for enhanced earthquake resistance and disaster mitigation strategies in tectonically active areas.

Zhang, Qingyun; Zhang, Jingfa; Li, Yongsheng; Li, Bingquan; Xie, Quancai; Luo, Sanming

doi: 10.1007/s00024-025-03694-2pmid: N/A

The Kumamoto earthquake is analyzed, mainly on the basis of InSAR data combined with strong earthquake and GNSS data, using a variety of joint InSAR methods and multisource data solution methods and by comprehensively considering the normalization and weighting of multisource data. The three-dimensional (3D) deformation field is determined. The results show that the joint solution with multisource data can improve the accuracy of the 3D solution deformation results to a certain extent. According to the 3D solution results, the maximum east–west deformation caused by the 2016 Kumamoto earthquake was approximately 2 m; the manifestations in the north–south direction were mainly characterized by expansion and stretching; the northwestern side subsided vertically, with a maximum subsidence of 2 m; and the southeastern side was uplifted. The horizontal deformation characteristics reveal that the earthquake was dominated by right-lateral strike-slip; the strike was NE–SW oriented, and the Futagawa fault has several normal fault properties. By analyzing the co-seismic 3D deformation field, seismogenic faults can be better understood, which provides a foundation for studying seismic mechanisms.

Liu, Naizheng; Wang, Tao; Peng, Songlin; Li, Yongdong; Cai, Ji; Fang, Guangyou

doi: 10.1007/s00024-025-03677-3pmid: N/A

We have proposed an alternative method for calculating the six parameters (three position parameters and three magnetic moment parameters) of a magnetic dipole source using extended two-dimensional orthonormal basis functions (2D-OBFs). In this method, a 2D-OBF decomposition is performed on the total-field anomaly generated by the magnetic dipole to obtain parameters defined as energy. The horizontal position estimate of the dipole is determined by identifying the peak of the energy distribution. By using peaks corresponding to two different initial vertical distance estimates (the distance from the dipole to the observation plane), the final vertical distance estimate can be analytically calculated. The magnetic moment vector is then obtained by solving the corresponding analytical equation. Thus, all six magnetic dipole parameters can be calculated simultaneously. We comprehensively demonstrated the characteristics and effectiveness of the proposed method through testing with synthetic and field data. Additionally, we conducted a comparative analysis to evaluate the similarities and differences between the proposed method and the Euler deconvolution method in field data testing. Although the 2D-OBF method requires further practical application testing, we are confident in its potential for detecting magnetic dipole sources, particularly in providing reliable initial parameter estimates for iterative optimization inversion.

El Amraoui, Oualid; Boujamaoui, Mustapha; Nait Bba, Abdellah; Rezouki, Ibtissam; Fahim, Abdelmounaim; Sahbi, Hassane; Diallo, Mamadou; Ibouh, Hassan; Mata, João; Bento dos Santos, Telmo M.; Youbi, Nasrrddine; Boumehdi, My Ahmed

Yuan, Changqing; Du, Jinsong; Gui, Jiangsong; Yin, Liang; Chen, Chao

doi: 10.1007/s00024-025-03690-6pmid: N/A

With the continually accumulated magnetic measurements and the gradually reliable global models of the lithospheric magnetic field by several advanced satellites (such as CHAMP, Swarm, CSES-1 and MSS-1), now present a requirement and also a challenge to develop realistic forward modelling methods for magnetic fields (i.e., the magnetic potential and its derivatives) that take the curvature of the Earth into account. The spatial discretization by a set of elementary tesseroids is generally utilized to approximate the complex magnetized source in spherical domain by the principle of superposition and saturate the source volume without “holes”. Since there is no analytic solution for magnetic fields of the tesseroid (except for the special points located on the polar axis), the numerical solution is the efficient way, where the Gauss–Legendre quadrature (GLQ) is usually employed. However, the required computation becomes notably time-consuming when the geometric sizes of the tesseroids are very large or the distances between the tesseroids and the observation points are very close, that is, the distance-to-size ratio (DSR) is quite small. Moreover, in an actual application, the DSRs vary with relative distances between source locations and observation points and hence are often non-uniform. Therefore, in order to reduce the computational time while maintaining a desired accuracy (i.e., relative percentage error) of each observation point, an efficient forward modelling scheme is employed. The key point of this scheme is the adoption of a new simple and efficient adaptive subdivision method. It is an equidistant subdivision method based on the longest side length, rather than recursion or stacking. By comparing the number of subdivided tesseroids, this method demonstrates its ability to avoid over-subdivision and perform more efficient calculations compared to the recursive method, because it adopts a new priori termination condition for subdivision rather than the traditional posteriori way. We obtain the required DSRs with errors of 0.1% and 0.01% through numerical simulation. At the same time, we package this scheme and release the open-source forward calculation software written by the C++ programming. Then, the analytical solution of the global homogeneous spherical shell using Runcorn’s theorem is utilized to test our newly proposed method. As a practical application, the impacts of Earth’s curvature on forward modelling of the magnetic fields are investigated.

Shi, Ke; Liu, Shuang; Jian, Xiange; Xu, Feng; Mao, Youping; Liu, Xianxin

doi: 10.1007/s00024-025-03675-5pmid: N/A

The surface magnetic anomaly provides excellent horizontal resolution, the three-component borehole magnetic anomaly offers excellent vertical resolution, and the aeromagnetic anomaly contains valuable information about deeper and larger magnetic sources. The advantages features of the three types of data can be combined to improve inversion resolution and reduce the nonuniqueness through the joint inversion. Currently, research primarily focuses on borehole–surface joint inversion, with the aeromagnetic anomaly seldom integrated into the joint inversion system, resulting in underutilization of its rich information. Furthermore, existing research findings on borehole–surface joint inversion primarily offer qualitative insights into the improvement in vertical resolution due to borehole magnetic anomalies. However, further investigation is required to understand the varying degrees of improvement brought about by different borehole quantities, positions, and distributions. To fully exploit the advantages of the three data and achieve higher inversion resolution, we propose a 3D joint inversion algorithm incorporating borehole, surface, and airborne magnetic anomalies. Through synthetic model experiments, we initially assess the actual enhancement brought by the aeromagnetic anomaly on inversion quality and discover effective ways to leverage its advantages. Subsequently, we investigate the degree of improvement in inversion quality resulting from different combinations of boreholes, summarizing optimal borehole selection methods. Finally, we apply the algorithm to a real mining area, validating its practicality by comparing the inversion results with drilled rock cores. Our research indicates that the proposed method yields inversion results with both high horizontal and vertical resolution, faithfully representing the physical properties of shallow-small and deep-large magnetic sources.

Zhao, Bairu; Li, Houpu; Zhang, Henglei

doi: 10.1007/s00024-025-03683-5pmid: N/A

Three-dimensional (3D) magnetic inversion plays a critical role in magnetic exploration by providing information about the spatial location, geometric shape, and distribution of physical parameters of anomalous bodies. The size of the model space subdivision determines the inversion resolution, while there is currently no consensus regarding the selection of subdivision accuracy for model space, which hampers its practical application. By discussing the relationship between subdivision accuracy and the effectiveness of 3D inversion, this study aims to provide a basis for the selection of the size of the model space subdivision. Multiple sets of theoretical magnetic models are used for 3D inversion with different size of the model space subdivisions, and the correlation coefficient between the inverted magnetization model and the theoretical magnetization model is used to evaluate the inversion accuracy. The results showed that the inversion accuracy continuously increases as the model subdivision spacing decreases to 0.5 times of the observed data spacing, and further improvement in the subdivision accuracy affect slightly the inversion accuracy. Therefore, it is suggested that the model subdivision spacing for inversion could be half of the observed data spacing. The applications of the model tests and field magnetic data from a mining area demonstrated that the inverted magnetization obtained using the size of the model space subdivision of 0.5 times the observed data spacing are significantly better than those obtained using the general 1 time the observed data spacing.

Sharma, Swati; Chakraborty, Alolika; Borah, Kajaljyoti

doi: 10.1007/s00024-025-03668-4pmid: N/A

The lithospheric structure beneath the west African craton (WAC) is presented in this study to understand the oldest parts of the continent. The shear velocity structure, associated composition (felsic/intermediate/mafic) and nature of crust transition (sharp or flat) at depth provide the link between the age these oldest parts of the continents formed and reworked in, formulating models of their evolution. This study estimates the crust and uppermost mantle velocity structure using joint inversion of the Rayleigh wave group velocity dispersion and receiver functions data calculated from 8 broadband seismic stations. The results show a significant variation of crustal properties in the Precambrian WAC. The shear wave velocity (Vs) at depth reveals a 42–47 km thick crust of felsic-to-intermediate composition near the boundary of the Precambrian old lithosphere and ~ 39 km thick felsic crust in the cratonic interior. The thick crust near the margins is also synchronous with a thick (~ 10–21 km) lower crust layer with high Vs (4.0–4.3 km/s). Contrarily, the thin crust accommodates a thin (~ 4–6 km) high Vs lower crust layer in the cratonic interior. This high Vs layer is often interpreted as the laminated base of the crust, providing insight into the thickness of the Precambrian lithosphere. Its presence as a thick layer at the base in the Proterozoic crust, or its absence (or thinning) in the Archean crust, is linked with the evolution of the continents. This high-velocity base is dissolved, reworked and delaminated over time, forming a thin felsic stabilised crust. We also observed higher uppermost mantle Vs in WAC, similar to the other Precambrian cratons (≥ 4.5 km/s).