Abstract Earthquake prediction remains a formidable challenge due to its theoretical and practical complexities. The multifactorial nature of earthquakes leads to diverse anomalies, which are potential precursors. However, the intricate earthquake process and limited knowledge of the Earth's crust structure restrict the accuracy of these predictions. This study introduces an advancements using machine learning and deep learning methods, notably the Kora 3 and Kora 4 algorithms, to identify key earthquake features. We employed LSTM and RNN deep learning algorithms to predict earthquakes of magnitude M≤4.3 without temporal data. Our methodology was applied to the 2022 earthquake monitoring in the Fergana depression, demonstrating significant advancements in seismic event prediction.

Abstract The marine and coastal environments of the Scotia Sea regions in the Southern Atlantic Ocean and Antarctica are vulnerable to the potentially disastrous effects from seismic activity in the Scotia Arc. This paper presents a magnetogravimetric study of the Scotia Plate for tsunami characterization. The influence of earthquakes on the Geomagnetic Field (GMF) is investigated using data from INTERMAGNET network observatories.  A tectonic model from gravimetric data is evaluated using gravity data from NOAA and seismic refraction data from Lamont-Doherty Earth Observatory oceanic surveys. The study also assesses the impact on water level (WL) measured at 6 tide gauge stations in the region obtained from the Intergovernmental Oceanographic Commission (IOC). The WL records collected are filtered and analyzed to identify tsunamis at each station. Cross Wavelet Transform (XWT) is applied, and a frequency analysis of the GMF is conducted to identify specific frequencies during seismic events. A 2D tectonic model is constructed for the North Scotia Ridge using gravimetric and seismic data to characterize structural boundaries that may be activated during seismic events. The results reveal anomalous frequencies in the GMF horizontal component frequency analysis during the November 25, 2013 earthquake, those results shown great data correlation with [1;1.5] hour periods from different observatories in the study area. Gravimetric modeling delineates faults activated during seismic activity and edges of structures potentially activated due to the margin transcurrent and compressional nature. WL anomalies up to 1.30 m are obtained following earthquakes with a Richter scale magnitude greater than 8 Mw. The tsunami propagation speed in the study area averaged 460 km/h, consistent with the expected speed for those depths, except for Puerto Argentino (PA), which exceeded them by 50%.

Abstract In the central portion of the Arabia-Eurasia collision zone, the Tehran domain is positioned at a transitional boundary between seismotectonic zones of the Central Iranian lowland (to the south) and the Alborz highland (to the north). Consequently, numerous destructive seismic events have occurred in this active tectonic domain. This study delves into the tectonic geomorphology of the region within its northern highland domain, specifically focusing on the hanging wall of the E-striking north-dipping North Tehran fault (NTF) zone. Our findings in this northern domain emphasize several prominent topographic scars as significant co-seismic features. These include huge landslides, rockfalls, rock avalanches, and offset geomorphic surfaces and could be present as the main indirect co-seismic morphological features. Within this seismically active region, the extensive dimensions of these geomorphic pieces of evidence can indicate the seismic potential of the Tehran Region to experience really strong earthquakes (i.e. M>7.5). These results contrast with the previous Maximum Credible Earthquake (MCE) magnitude estimated for the Tehran Region (i.e. M~7.2) through different approaches in Seismic Hazard Assessments (SHAs). Consequently, the previous SHAs of the Tehran Region might have underestimated the seismic risk, and therefore, it is necessary to conduct an updated and complementary deterministic SHA based on the more detailed seismogenic geological features in this crucial area. This new approach can be employed in comparable active tectonic regions worldwide to assess existing SHAs.

Abstract This paper presents a comparative study for two couples of very strong seismic events-Bulgarian Kresna-Kroupnic seismic source activated during 4th April, 1904 (magnitudes 7.2 and 7.8) and those in Turkey 6th February 2023, Gaziantep and Kahramanmaraş seismic source (magnitudes 7.8 and 7.5) called “doublets”. The comparison includes main geophysical and seismological parameters and some social and post events consequences and characteristics. The study reveals similarities in the power and huge differences in the consequences of these seismic events occurred in different geodynamic and environmental conditions. Other studies are considered [20, 21] related to strong seismic events and their consequences in other seismic active regions. Comparative tables are developed for easier and visual following of the similarities and differences, and detailed comparison and explanations are given in discussion to try to explain why these differences existed. Conclusions about the effects (social and practical issues) are derived and they could be useful in case of actual retrospective analysis for seismic zoning and for the future assessment of seismic hazard and risk.

Abstract Fault plane solutions for a group of 104; 4.0 ≤ Mw ≤ 7.1 earthquakes between January 1979 and December 2016, extracted from the Global Centroid Moment Tensor Project catalog. Were used to investigate the regional tectonic stress regime of the Gulf of Guinea region. The idea is to validate the theory of membrane tectonics put forward by Freeth (1977) [1] in which the tectonic of the Gulf of Guinea and the sub-Sahara West Africa region were described based on Freeth (1977) [1] . The tectonic of the Gulf of Guinea and the sub-Sahara West Africa region are based on the movement of the African plate, we emphasized the use of rigorous statistical tests to decide on the quality and variability of the earthquake focal mechanisms (FMSs) utilized for the stress tensor inversion analysis. To constrain our analysis, we have applied both the Algorithm of Michael and Gauss technique in our stress tensor inversion analysis of FMS obtained from the region, and the results are found to be coherent and in good agreement with each other. Both Michael (1984) [2] and Zalohar and Vrabec (2007) [3] techniques show that the regional tectonic stress regime of the Gulf of Guinea and the sub-Sahara West Africa is extensional, which is in good agreement with the work of Freeth (1977) [1] . However, our investigation concluded that the orientation of the extensional stress regime is the same as the orientation of the movement of the African plate, which is towards the Euro-Asia plate.

Abstract “Living in fear of Nigeria biggest Earthquake” is a sub-heading of the Punch daily newspaper of Nigeria, dated, 21st of August, 2016, reported the earthquake/tremor recently witnessed in the ancient town of Saki, accompanied by a series of aftershock events that lasted for about three (3) months (March/May, 2016) southwest Nigeria [1] . Similarly, roughly five (5) years later, another series of earthquakes/tremors occurred again in Saki town, which was reported by an online news vendor named “Ripples Nigeria” dated September 8, 2021, under the sub-heading of “earth tremor rocks Saki in Oyo state” [2] . However, these events were not captured nor recorded by any of the functional seismological stations in Nigeria. The nearest seismological station located at the Obafemi Awolowo University (OAU) Ile-Ife, also failed to capture the events. We therefore seek to examine the instrumentation of the Nigerian National Network of Seismographic Stations. This is necessary to understand the functionalities and the capabilities of the deployed seismometers in each of the seismic stations. Therefore, we evaluated the bandpass limit of the seismic wave frequency for the respective seismic stations in Nigeria through the computation of the amplitude-frequency response curve, phase response curve, and count to cm/sec.