Acta Geophysica最新文献

Seismo-acoustic and stratigraphic framework of the North İmralı Basin in the Eastern Marmara Sea has been investigated using high-resolution seismic reflection data, in order to reconstruct the evolution of the basin over the past 500 ka. Seismic data indicate that the right-lateral strike-slip İmralı Fault, which bounds the basin to the south, is the primary structural feature controlling the formation and development of the basin. The fault bends offshore of the Armutlu Peninsula and north of İmralı Island which induces a NE-SW oriented extensional regime within the basin. As a result, NW–SE trending normal faults with a right-lateral slip component have formed in the north operating as growth faults, demonstrating that the basin is actively subsiding with a counterclockwise rotational motion. Under this subsidence regime, the North İmralı Basin has developed over a negative flower structure associated with a releasing segment mechanism. Vertical displacement measurements along these faults suggest that the subsidence has persisted in a steady-state manner at a constant rate of approx. 60 cm/ka over the past 500,000 years. The four buried deltas observed in the seismic data were formed during marine regression and represent glacial-stage deltas, excluding Delta-2 which appears to be associated with sea-level rise. No delta has been observed in the past 74 ka due to an imbalance between sedimentation rates and the increasing accommodation space.

Graphical abstract

River flow is a crucial component in water resource management, environmental protection, and disaster mitigation. Also, the Mississippi River, one of the largest and most significant rivers in North America, plays an essential role in shaping the region’s infrastructure and economy. Moreover, reliable flow forecasting is essential for several key purposes, including flood prevention, optimal water resource allocation, and ecosystem preservation, all of which are critical to promoting sustainable development and enhancing disaster resilience in the region. This study aims to predict the discharge of the Mississippi River at the Memphis station for the period from 1990 to 2024 by employing hybrid models that integrate short-term long-term memory (LSTM) with random forest (RF) and neural network (CNN), considering lag intervals of 3–15 days. Various evaluation criteria were utilized to assess the accuracy of these models. The performance evaluation revealed that a three-day lag interval produced the most accurate results, with the CNN-LSTM model achieving the best performance, with NRMSE = 0.0165, at the Memphis station. Additionally, the RF-LSTM and CNN-RF-LSTM models demonstrated high accuracy in predicting daily discharge, with NRMSE = 0.0179, 0.0177, respectively. These findings have significant implications for water resource managers, providing enhanced efficiency in reducing labor costs and saving time in forecasting.

Soil improvement methods aim to enhance geotechnical properties such as strength, durability, and resistance to environmental loads. Among recent advances, microbially induced calcium carbonate precipitation (MICP) has emerged as a sustainable alternative to conventional chemical stabilizers. This study presents a novel application of Box–Behnken experimental design (BBD) for optimizing MICP parameters in sandy soils, focusing on the interaction between sand type, treatment solution molarity, and curing time. A total of 17 experimental runs were conducted under a structured response surface methodology to identify optimal conditions for maximizing unconfined compressive strength (UCS). The highest UCS value, 1297 kPa, was achieved using fine-grained sand treated with a 1.5 mol solution over 5 days. The key innovation of this research lies in the integration of statistical design techniques with microbial geotechnology, enabling efficient modeling of nonlinear interactions and minimizing experimental effort. Beyond laboratory findings, the results offer practical guidance for field-scale implementations by identifying critical parameter ranges that ensure microbial viability and performance stability. This integrated approach provides both methodological novelty and applied relevance, contributing to the advancement of bio-based soil improvement strategies in geotechnical engineering.

This study examines seismicity and deformation in New Zealand's North Island using data from 2014 to 2024. The North Island, shaped by the Hikurangi Subduction Zone (HSZ), the Taupo Volcanic Zone (TVZ), and the North Island Dextral Fault Belt (NIDFB), exhibits complex tectonic activity. Therefore, in this study, Global Navigation Satellite System (GNSS) data were processed to analyze the horizontal and vertical movements of the North Island relative to the Pacific and Australian plates. Additionally, earthquakes were examined based on their magnitudes, epicenters, and hypocenters. The analysis of M ≥ 4 earthquake distribution highlights a seismic quiescence in the northwestern part of the island, particularly around Auckland and the Northland Volcanic Field. According to hypocenter depths, earthquakes at 20–60 km and 60–300 km form two NE-SW trending seismic zones, suggesting distinct structural domains with a potential transition zone in the west. GNSS velocities relative to the Pacific-fixed frame are found to be consistent with the subduction rate of the HSZ, while velocities relative to the Australian-fixed frame reveal higher horizontal velocities in the north, which align with a greater subduction rate and rollback effect. In contrast, the western part of the island rotates significantly slower than the eastern part. Consequently, this study presents and evaluates deformation trends, potential transition and quiescent zones, and the recent movement of the North Island in conjunction with previous studies.

Equilibrium scour depth and morphological evolution around submerged spur dikes were investigated through a combination of laboratory experiments and numerical modeling. A series of flume experimental runs under clear-water conditions examined the influence of spur dike geometry, submergence ratio, and flow hydraulics on maximum scour depth. A new empirical equation was developed using nonlinear regression analysis, incorporating data from both the present work and Elawady et al. (Proc Hydraul Eng 45:373–378, 2001), showing improved predictive performance over existing models, including Fang et al. (Int J Sediment Res 21(2):89–100, 2006). Numerical simulations using FLOW-3D Hydro captured the flow–sediment interactions and accurately reproduced scour patterns and temporal scour development for varying submergence conditions. The accuracy was observed under all three submergence ratio conditions (SR = 0, 0.5,0.75) with statistical metrics. The findings offer valuable insights into scour mechanisms and provide a reliable framework for estimating the scour around submerged spur dikes, supporting safer and more effective hydraulic structure design. Special attention was given to instrumentation precision, and potential measurement uncertainties were considered in evaluating the reliability of both experimental and numerical outcomes.

It is one of the important problems to predict the geologic bodies ahead of the tunnel face not only for mining in the coal mines and the mines but also for the geotechnical tunnel construction related to infrastructure construction. Frequency domain induced polarization (IP) method is one of the widely applied geophysical methods in detecting anomalous bodies ahead of the tunnel face. We present a method of joint inversion of apparent resistivity and frequency domain IP data to increase the reliability of tunnel prospecting by frequency domain IP method. We determine the numbers of terms of the infinite series in formulas of forward modeling to use for joint inversion and introduce two methods to estimate the existence of the geologic body ahead of the tunnel face. Joint inversion is based on the logarithmic barrier method. Also, we demonstrate the reliability of the method and the effects of the number of current electrodes and the ratio between the length of the survey line and the distance from the tunnel face to the anomalous body on the accuracy of the method. Tests of synthetic and field data show that the method we propose has high reliability.

The Solonker suture zone (SSZ) and its surrounding areas have undergone multiple tectonic activities, including the closure of the Paleo-Asian Ocean during the Paleozoic, the closure of the Mongol–Okhotsk Ocean during the Mesozoic and the subduction of the Pacific Plate during the Mesozoic and Cenozoic. These activities have resulted in a complex deep structure in the region. Previous studies have preliminarily indicated the presence of electrical anisotropy in the lithospheric structure of the Solonker Suture Zone and its adjacent areas. To further investigate the electrical anisotropy in this region, we utilized broadband and long-period measurement data collected from the L1 profile from 2010 to 2012 and the L2 profile from 2017 to 2018. Through dimensionality and induction vector analysis of the data, the results indicate that the underground structure in this area exhibits 2D characteristics and electrical anisotropy. Subsequently, we employed the mare2dem code developed by Kerry Key to jointly invert the MT data, obtaining 2D isotropic and anisotropic resistivity models, which further confirmed the electrical anisotropy in the region. Based on the analysis of the resistivity models, it is inferred that the anisotropic origin of the Solonker Suture Zone is related to graphite and sulfide minerals present in the fold hinges, with the graphite layers possibly originating from black shales caused by the subduction of the Paleo-Asian Ocean. The analysis of the anisotropic resistivity models for the two profiles indicates that the Paleo-Asian Ocean experienced bidirectional subduction and multi-stage closure, with the closure direction being from west to east.

Seismic signals exhibit characteristics that are neither harmonic nor transient, making it essential to have both time and frequency resolution at the same time in their time–frequency representation. Developing a tool with high-resolution time–frequency representation and reversibility is crucial. Although synchrosqueezing methods improve frequency resolution, they do not enhance time resolution. In this paper, we introduce a novel approach known as the multisynchrosqueezing optimized S-transform (MSS-OST), which utilizes an optimized S-transform to enhance time resolution before improving frequency resolution. By combining this approach with the multisynchrosqueezing transform, which enhances frequency resolution by redistributing time–frequency coefficients along the frequency axis, the proposed method is positioned to generate a time–frequency representation characterized by superior time and frequency resolutions. To validate the effectiveness of our proposed method, we present results obtained from synthetic and real field data. Furthermore, we evaluate the performance of the proposed method through comparative analyses with conventional and state-of-the-art approaches, including short-time Fourier transform (STFT), S-transform (ST), optimized S-transform (OST), multisynchrosqueezing transform (MSST), multisynchrosqueezing S-transform (MSS-ST), time-reassigned multisynchrosqueezing S-transform (TMSS-ST), and multisynchrosqueezing generalized S-transform (MSS-GST). The results indicate that our method produces a sparser time–frequency representation compared to the aforementioned techniques. Moreover, the innovative sparse representation excels in precisely identifying the location of gas reservoirs through time–frequency AVO analysis, outperforming other utilized methods.

Measurements of the fractal dimensions and entropy of the spatial distribution of seismic sources are seen to correlate with measurements of the Gutenberg–Richter b-value, but, although this parameter is related to the fractal dimension of the rupture areas distribution and to the entropy of the magnitude distribution, there is no specific mechanism to relate it to epi- or hypocentral source distributions. A plausible relation between the b-value and the spatial source distributions is proposed and tested through Monte Carlo simulation on a cellular automaton model based on the premise that the probability of an earthquake occurring at a particular point in space is proportional to the stress at that point. Results showing the appropriate correlations are robust and not critically dependent on the values of the parameters of the model.

Precise fault boundary delineation is crucial for oil and gas exploration and development. Meanwhile, artificial intelligence algorithms have achieved remarkable results in fault detection tasks. In particular, the application of attention mechanisms in deep learning has significantly improved feature extraction in seismic interpretation. However, relying on a single type of attention module may cause the network to concentrate too narrowly on certain training features, which can lead to overfitting and hinder generalization. Although integrating multiple models can improve overall performance and robustness, it incurs a relatively high computational cost. In response to the above problems, we propose a novel two-branch network architecture that adds a separate encoding and decoding path and introduces a new lightweight attention mechanism on the basis of the existing U-Net structure. Two separate encoding and decoding branches can be flexibly learned to obtain different fault characteristics during training, thereby reducing implicit prediction errors and improving the generalization ability and stability of the model. This study used the FaultSeg3D synthetic dataset for training, with input features being three-dimensional seismic amplitude volumes and output being corresponding fault probability annotations. Validation was then performed on two publicly available field seismic datasets: the Netherlands offshore F3 block and the Kerry-3D dataset. The experimental results show that this mechanism enhances the representation of fault continuity. The two-branch network architecture focuses on a variety of typical fault characteristics. In particular, in the initial training iterations, it accelerates the convergence of multiple evaluation metrics compared to each branch used independently. After applying the complementary advantages of the two network branches, the combined network more effectively identifies key faults in real seismic data and enhances the representation of detailed fault features.