Geophysical Journal International

SummaryCurrent geodetic velocities show that over half (up to 10 mm/yr) of Arabia-Eurasia shortening in the west is accommodated within a relatively narrow zone across the Kura basin of Azerbaijan, in which the most prominent active structure is the Kura fold-and-thrust belt, bordering the southern margin of the Greater Caucasus. The GNSS velocities furthermore suggest equivalent amounts of north-south right-lateral shear across the eastern Kura basin along the West Caspian fault zone that is accommodating relative motion between the Kura basin and the South Caspian basin. Although destructive historical earthquakes are known to have occurred, their spread is restricted geographically and their moment release accounts for only half of the accumulated deformation. These observations can be explained by incompleteness of the historical record, that the faults fail in rare larger earthquakes, or that they slip aseismically. To distinguish between these hypotheses we produce an InSAR velocity field using Sentinel-1 SAR data to image active tectonic deformation within the Kura basin of Azerbaijan. Tectonic signals are superimposed on those relating to non-tectonic processes, including widespread mud volcano inflation that highlights the important role of fluid flow within the basin sediments. We show aseismic creep occurs on two parallel faults of the West Caspian fault zone, and infer this also on the Kura Fold and Thrust Belt from sharp gradients in velocity indicating active fold growth. Recent paleoseismic studies of the faults imaged here indicate discrete slip events, and we speculate that the creep may be episodic, perhaps triggered by deeper earthquake events or by periods of enhanced fluid mobilisation. Together, the right-lateral and left-lateral faults appear to accommodate a large-scale expulsion of the Absheron region towards the South Caspian basin, perhaps driven by gravitational potential energy contrasts.

SummaryWhile global sea level rise is a major concern for ocean-connected coastal regions, inland seas such as the Black Sea exhibit water level changes primarily governed by regional hydrological and climatic factors. Understanding the drivers of water level variability in the Black Sea is essential due to its sensitivity to river inflow, evaporation, and limited connection to the global ocean. This study, for the first time, integrates satellite altimetry and satellite gravity data from 2003 to 2023 to analyze the long-term and seasonal variations in the Black Sea water levels, as well as local sea level variability and its driving factors. The results indicate that the sea level in the Black Sea experiences a positive trend of 1.04 ± 0.39 mm/yr. This comes, however, with a negative trend on a seasonal scale during autumn (–1.14 ± 0.27 mm/yr), which contrasts with the rising trends observed in other seasons. We found that the loading deformation induced by global mass redistribution contributes to 39 per cent of the Black Sea level trend, leading to an overestimation of actual climate-induced sea level change by 0.41 mm/yr. We further found that the reduction of precipitation and river inflow from surrounding basins leads to an increase in the salinity of the Black Sea, driving the decline in steric sea level. On the other hand, it has also increased the water mass of the Black Sea, compensating for the steric sea level drop.

SummaryDetecting marsquakes is crucial for understanding the interior structure of Mars. However, the detection and analysis of such seismic events is challenging due to the strong background noise and low signal-to-noise ratios (SNR) of marsquake seismograms. Moreover, the volume of the seismic data needed to create an efficient method to detect marsquakes, and the required amplitudes of such events is limited by the low energy of the seismic. Here, we develop a new deep learning-based detection algorithm, MarsConvNet (MANet), which can fully exploit the latent information of seismic sequences by upsampling and downsampling for detecting marsquakes in the continuous seismic records. Our approach is to construct a deep encoder structure and nonlinear mapping relationship between seismic signals and arrival times of P and S waves by training on the STanford EArthquake Dataset (STEAD) from Earth. Application to the continuous waveform data from the InSight seismic data from 2019 to 2022 shows that the algorithm can detect marsquakes with weak amplitude in addition to reproducing all events detected manually by the Marsquake Service (MQS). Using our method, we detected 67 previously undetected marsquakes.

SummaryA new approach for the parallel forward modelling of transient electromagnetic (TEM) fields is presented. It is based on a family of uniform-in-time rational approximants to the matrix exponential that share a common denominator independent of the evaluation time points. The partial fraction decomposition of this family is exploited to devise a fast solver with high parallel efficiency. The number of shifted linear systems that need to be solved in parallel does not depend on the number of required time channels nor the spatial discretization. We also argue that similar parallel efficiency gains can be expected when solving the inverse TEM problem.

AbstractDesert strata consist of dune and interdune deposits, and they have different magnetic properties. Contrasting magnetic properties of these two types of strata is a foundation for extracting paleoclimate information from dune-interdune strata in these eco-fragile regions, but few have done so. Here we compared magnetic properties between dune and interdune strata from the late Miocene desert deposits in the northwestern Tarim Basin. We found that interdune strata displayed enhanced magnetic properties mainly due to content increase of stable single-domain (SSD) and small pseudo single-domain magnetite (PSD) within chlorite layers. But magnetic enhancement in interdune strata does not correspond with content increase of superparamagnetic (SP) ferrimagnetic minerals. Our high-resolution transmission electron microscope results suggest pedogenic SSD and small PSD magnetite form within chlorite layers, and we propose weathering of chlorites resulted in high concentration of Fe (II) and Fe (III) ions within layers, facilitating the formation of SSD and small PSD magnetite. By contrast, dry climate in the Tarim Basin results in micropore spaces between mineral particles having limited concentration of Fe (II) and Fe (III) ions, inhibiting SP magnetite formation. This pattern is different from Chinese loess-paleosols sequence, where magnetic enhancement is accompanied with content increase of both SP and SSD or small PSD ferrimagnetic mineral content. Therefore, magnetic enhancement pattern differences (whether magnetic enhancement is accompanied with SP magnetite formation) may inform climate conditions. To infer environmental variations in arid regions using magnetic parameters, reliance on proxies that indicate the relative content of ferrimagnetic grains or content of magnetic grains with stable remanence is essential.

SUMMARYSeismology has advanced significantly with ambient noise interferometry, enhancing the extraction of surface waves for detailed Vs imaging. However, direct Vp measurements remain critical for geological and engineering applications. Guided P waves offer potential for Vp insights, yet their detection from ambient noise is challenging, especially in urban environments. This study explores the feasibility of using underground urban seismic noise, specifically tunnel traffic, for guided P wave extraction and subsurface imaging. We introduce an adaptive workflow for preprocessing and enhancing guided P wave signals, applied to data from the Zijingang Tunnel beneath the campus of the Zhejiang University. Our refined data processing workflow significantly improves the quality of retrieved empirical Green’s functions of guided P waves. The joint inversion of surface waves and guided P waves provides high-resolution Vs and Vp profiles, revealing detailed subsurface structures. We also examine the impacts of source depth and type on guided P wave retrieval. We observed distinct differences in urban ambient noise characteristics between daytime and nighttime, suggesting that daytime traffic noise, with higher frequency content, is more suitable for high-frequency guided P wave retrieval. Our results demonstrate the effectiveness of using underground urban seismic noise for guided P wave extraction, highlighting the potential applications of urban ambient noise imaging in geotechnical engineering, urban planning, and seismic hazard assessment.

SUMMARYAirborne natural source electromagnetics (EM) provide spatially densely gridded, frequency-dependent inter-site transfer function estimates referring airborne vertical and horizontal magnetic field data to the horizontal components of a ground site. Common processing schemes for airborne natural source EM data are conceptually based on approaches developed for ground-based magnetotelluric sounding curve estimation and yield statistically independent transfer function estimates for selected times windows and separately for each component of the airborne recording. However, the magnetic field is spatially smooth and, moreover, its components are mutually dependent in space. Here, we propose to exploit spatial potential field relations directly in the processing workflow to refine transfer functions. The self-consistent response function estimator is applied after a first statistical estimation of transfer functions and supports the determination of a spatially smooth and self-consistent set of vertical and horizontal magnetic transfer functions that reflects physical constraints. Self-consistency is enforced by fitting sources distributed within an equivalent sheet to a preliminary set of transfer function estimates determined on a statistical basis. Smoothness is controlled by the depth to the equivalent sources. Once estimated, the distribution of vertical and horizontal magnetic transfer functions can be predicted from the equivalent sheet at any point in the air half-space. Additionally, the scheme allows for the identification of anomalous parts of the horizontal fields at the ground site. The procedure is demonstrated on a large-scale field data set from Gobabis (Namibia). We observe that transfer functions derived by using the equivalent source layer interpolator improve processing results when compared to purely statistically determined transfer function averages.

AbstractThis study investigates the mantle transition zone (MTZ) beneath Central and Eastern Europe using a 3D Common Conversion Point migration of P-to-S receiver functions derived from a dense regional seismic network. The analysis focuses on the major seismic discontinuities at ∼410 km, ∼520 km, and ∼660 km depth to assess their depth variations, continuity, and implications for past and ongoing geodynamic processes.Our results reveal significant spatial variations in the thickness and topography of the MTZ. In the Western Alps and central Pannonian Basin, the MTZ is thickened up to ∼280 km, deviating from the global average of ∼250 km. This thickening is attributed to the presence of stagnant slab material in the transition zone, suggesting a long-lasting influence of past subduction, particularly of the Adria Plate and Vrancea slab. In the Carpathians and Dinarides, the 410 km discontinuity is uplifted to depths as shallow as ∼400 km, while the 660 km discontinuity reaches depths of ∼670–680 km in regions affected by subducted lithosphere, further supporting the presence of cold slab remnants.Additionally, the 520 km discontinuity—often intermittent or absent in global studies—is clearly imaged in many parts of the region, and found at variable depths ranging between ∼510 and ∼540 km. These depth anomalies suggest the presence of compositional heterogeneities and thermal variations within the MTZ, possibly linked to subducted oceanic crust or recycled lithospheric material.Evidence for mantle upwellings is also observed, particularly beneath the Pannonian Basin, where low-velocity anomalies near the 410 km discontinuity are consistent with small-scale plumes or thermal anomalies. These may be associated with post-subduction processes or intraplate volcanism. Importantly, the Alpine slab itself is not clearly detected in the transition zone, indicating that it may have already sunk below the MTZ or is not well-coupled to the upper mantle structure imaged by receiver functions.By providing new constraints on the structure of the upper mantle and its transition zone, this study refines existing models of regional tectonic evolution. Our findings emphasize the interplay between surface tectonics and deep mantle dynamics and demonstrate that the observed MTZ features preserve a strong geodynamic imprint of both past subduction and intraplate processes across the Alpine-Carpathian-Pannonian-Dinarides region.

SUMMARYThe temporal variations in seismic wave velocity provide critical insights into the sources and physical mechanisms underlying diverse geophysical processes. While traditional approaches rely on measuring coda wave traveltime shifts to estimate velocity changes, the increasing availability of dense seismic networks has shifted attention toward ballistic waves for seismic velocity monitoring. Current methodologies for measuring ballistic wave time shifts predominantly employ the wavelet-transform technique, which, despite its proven reliability for coda wave analysis, introduces nonnegligible biases in ballistic wave monitoring due to the spectral leakage effect. To address this limitation, we propose a novel frequency-domain approach that estimates time shifts at each frequency, leveraging the characteristics of invariant phase shifts of ballistic waves along lag time. This method offers enhanced computational efficiency and simplicity in phase shift measurements compared to the time-frequency domain analysis. The phase velocity change is subsequently determined through a linear regression of phase time shifts along the offset. Synthetic tests validate the superior stability and accuracy of our method in estimating ballistic wave phase velocity changes. We further apply this approach to extract surface wave relative phase velocity changes from field data. Our results bring a robust and efficient method for measuring relative phase velocity changes in ballistic wave seismic monitoring.

AbstractSpectral electrical impedance tomography (sEIT) has attracted increasing interest in hydrogeology, biogeosciences, agriculture, and environmental studies. However, broadband sEIT measurements, particularly at frequencies above 50 Hz, have long been challenging due to electromagnetic (EM) coupling effects. Recent advances in instrumentation, data correction, and filtering have improved sEIT measurements at higher frequencies, yet many of these developments rely on a customized system with distributed amplifiers. Extending these advancements to more universally applicable methods is necessary, as sEIT measurements are often acquired using systems with centralized multiplexers. This study aims to bridge this gap by developing model-based data correction methods to mitigate EM coupling effects in sEIT measurements acquired with such a set-up. For this, the differences in EM coupling effects between measurements with a centralized multiplexer and distributed amplifiers were discussed, and the required correction methods in case of a centralized multiplexer were developed. The effectiveness of the developed corrections was tested using sEIT measurements acquired with a centralized multiplexer. A dataset obtained using distributed amplifiers and corrected using previously developed approaches served as a reference. Finally, inversion results of all datasets were compared. It was shown that cable capacitance dominates the capacitive coupling in the sEIT measurements acquired with a centralized multiplexer when coaxial cables are used. Improvements were observed after each correction step using the developed methods. It was concluded that broadband sEIT imaging results can be obtained using measurements with a centralized multiplexer and coaxial cables using the proposed data correction and filtering methods.

SummaryThe dynamic Stern layer (DSL) petrophysical model can be used to interpret field induced polarization data and can be applied to both magmatic (volcanic and igneous) rocks and sediments. Thanks to it, field-scale tomograms of conductivity and normalized chargeability can be transformed into tomograms of porosity, Cation Exchange Capacity (CEC), and temperature. Furthermore, kilometer-scale galvanometric induced-polarization surveys are nowadays doable thanks to the recent development of independent stations measuring the primary and secondary electrical fields. This approach reduces capacitive and inductive coupling effects inherent to systems based on long cables and allow for deeper investigations. We apply here this combined methodology (novel equipment and revised petrophysical model) to a geothermal prospect located at Mashyuza, Republic of Rwanda, in Eastern Africa. At this site, the rifting activity led to the occurrence of an extensional regime favoring the occurrence of a rising thermal plume at the intersection between two faults. The existence of this plume is expressed at the ground surface by the presence of a hot spring at a temperature of ∼52°C (a well nearby provides a temperature of ∼65°C). A time-domain large scale induced polarization survey is performed. The current source signal is produced by a VIP-5000 squared signal injector and injection current values range from 1 A to 4 A, with stacking of 1s on-off signals ranging from 100 to 300 stacks to improve the signal-to-noise ratio. The size of the 3D array is ∼1.5 km and allows a tomography down to a depth of ∼300 meters. The data are inverted with the deterministic least-square technique, penalizing the roughness of the resulting tomograms. The conductivity and normalized chargeability tomograms are combined to get the temperature, porosity, and CEC distributions. The temperature distribution is consistent with the temperature of the hot spring and well. The results are interpreted in terms of ground water flow pattern and dilution of the mineralized thermal water with the fresher surface meteoric water. The survey images a rising plume of warm water from a depth of at least 300 m along intersecting fracture systems.

SummaryTianshan Mountains in Central Asia are one of the largest and most active orogenic belts in the world, characterized by complex structures and strong seismic activity. In this paper, we use recently updated GNSS data to self-consistently estimate the slip rates of major faults in Tianshan region via the elastic block model. Our results indicate that crustal deformation in Tianshan region is predominantly manifested as crustal shortening, regulated by foreland thrust belts and intermontane basin boundary faults. The shortening rate decreases from 14.2 ± 3.4 mm/yr in the west to approximately 3 mm/yr in the east. By estimating the seismic moment accumulation rates of the major seismic belts and comparing them with the historical earthquake catalog, we identify six seismic belts with significant seismic moment deficits. This indicates a potential risk for earthquakes exceeding magnitude 7, including the Kash fault, Keping fault, the Maidan fault zone, and the North Tianshan seismic belt. The Nalati seismic belt exhibits a relatively small seismic moment deficit, indicating the potential of earthquakes in the magnitude range of 6 to 7. In contrast, Qiulitage fault, the West Tianshan seismic belt and the Manas fold-and-thrust belt show a moment surplus, suggesting a low likelihood of strong earthquakes occurring in the near future. This study provides critical data and theoretical support for the prediction and risk assessment of seismic activity in Tianshan region.