Global Navigation Satellite System (GNSS) orbit and clock offset are the spatial datum of the positioning, navigation and timing services. The precision of GNSS ultra-rapid products, particularly critical for real-time applications, is significantly influenced by observational data quality and parameter configuration strategies. Among these products, the 24-h predicted satellite clock offsets demonstrate heightened sensitivity to the estimated series. This study presents a novel optimization approach for ultra-rapid clock prediction by incorporating satellite clock frequency, effectively deducing the impact of original low-quality estimation clock products. First, the epoch-wise clock offset series undergoes recursive state estimation through a constructed transition model, which dynamically updates first-order velocity and second-order acceleration clock variation components. Second, the radial basis function is employed as the kernel function of sparse modeling during training to establish a continuous model-characterizing clock offset velocity. Meanwhile, the improved clock offset is determined through a time integration strategy, where incremental clock offset values during each epoch interval are calculated and subsequently summed to obtain the refined estimation. Third, long short-term memory algorithm incorporates inter-parameter correlations to enhance clock offset prediction precision, where the gate mechanisms explicitly account for parameter dependencies during temporal feature extraction. Experimental results analyses demonstrate significant reductions in standard deviation for predicted clock offsets, achieving at least 3.4
The current study analyzes the spatial variations of seismicity and accompanying hydrological perturbations during January 2006 to October 2024, using earthquake catalog data from the Iranian Seismological Center (IRSC) and GRACE-FO satellites Equivalent Water Height (EWH) variations. Seismic complexity was estimated from Shannon entropy and the Tsallis entropy index for the description of non-extensivity. Abrupt degradation of entropy began 2.5 years before the Mw 6.3 Bandar Abbas earthquake (14 November 2021), testifying to gradual structural organization, asperity locking, and building up of stresses. The pattern reversed after the mainshock, with an increase of entropy in the first year of the postseismic period and remaining at a weaker level for the following 2.5 years, consistent with crustal relaxation and redistribution of stresses. The entropic index q = 1.71 was derived from the non-extensive analysis of the earthquake magnitude-frequency distribution in the study region. This value indicates correlated and non-equilibrium seismic behavior, consistent with previous studies in tectonically active regions. GRACE-FO-derived EWH anomalies showed similar spatial patterns, with weak changes 2.5 years before the earthquake, intensification one year prior, and regional maxima six months before the event. After the earthquake, anomalies strengthened within six months, expanded after one year, and weakened after 2.5 years, reflecting hydromechanical processes such as groundwater redistribution and fault-related fluid flow. The simultaneous evolution of entropy and EWH anomalies supports a coupled geophysical–hydrological seismic cycle in the Southern Zagros and highlights the value of integrating statistical seismology with satellite gravimetry for monitoring seismic processes in active tectonic regions.
Three-dimensional gravity data inversion is an ill-posed and non-unique problem that requires effective regularization to obtain stable and geologically meaningful subsurface density models. Although sparse regularization methods based on Lp-norms enhance structural sharpness, fixed-norm approaches—particularly L1-norm regularization—often produce overly smooth or diffuse anomalies at depth due to the limited resolving power of gravity data. To address this limitation, we introduce a sensitivity-based adaptive Lp-norm regularization framework for gravity inversion, implemented using an iterative reweighted least squares (IRLS) algorithm formulated in data space. The proposed approach spatially adapts the norm parameter p according to data sensitivity, which decreases with depth. Higher p values are assigned to shallow, well-constrained regions to promote smooth and stable solutions, while lower p values are assigned to deeper, weakly constrained regions to encourage sparsity, compactness, and sharper boundaries. This strategy is physically motivated and aims to counteract the inherent loss of resolution with depth, thereby reducing the tendency of conventional sparse norms to smear deep structures. Reformulating the IRLS algorithm in data space significantly improves computational efficiency by reducing memory requirements and computational cost, making the method well suited for large-scale three-dimensional inversions. The effectiveness of the method is demonstrated using two synthetic models of increasing complexity and a real airborne gravity gradiometer dataset from the chromite-bearing Black Thor Intrusive Complex in Ontario, Canada. Results show that the sensitivity-based adaptive Lp-norm inversion consistently outperforms conventional L1-norm regularization in resolving deep bodies, producing sharper geometries, and geologically plausible depth extents while maintaining good data fits.
Inversion of Self-Potential (SP) data poses a significant challenge in exploration geophysics due to the inherent non-linearity and presence of multiple local optima. This study presents the first application of Harris Hawks Optimization (HHO) algorithm for quantitative interpretation of SP data based on a dipole model characterized by polarization intensity ( k ) and pole positions ( x_1 , z_1 , x_2 , z_2 ). Unlike conventional optimization methods, HHO’s unique two-phase mechanism—dynamically transitioning between exploration and exploitation through soft and hard besiege strategies—provides superior capability in escaping local optima, a critical advantage for the highly non-linear SP inverse problem. Validation on synthetic data (noise-free and 5–25 R^2=0.9725 at 15 x_1 as the most influential parameter (normalized sensitivity = 1.0). Field validation on SP data from Surda copper mine, India, yielded excellent agreement (RMS = 5.69 mV, R^2=0.9982 ), with interpretations consistent with regional geology and previous studies. These results establish HHO as a robust, accurate, and practically viable tool for SP data inversion in mineral exploration.
The main objective of this study is to improve and refine the understanding of the deep structural framework of the Saouaf, Sisseb El Alem, and Echougafia basins using seismic and gravity data. A residual gravity map was derived from the Bouguer anomaly map, allowing the identification of an architecture composed of high-density contrast zones (positive anomalies) and low-density zones (negative anomalies), separated by gravity gradients of varying amplitudes. This gravity pattern corresponds to resistant structural domains bounded by major faults that separate them from subsiding areas. Three-dimensional inversion and two-dimensional forward gravity modeling, integrated with geological outcrop data, made it possible to reconstruct the main structural features. Euler deconvolution was applied and combined with seismic reflection data to generate a comprehensive structural model, confirming several faults previously identified by seismic method and revealing additional lineaments trending NE–SW, N–S, NW–SE, and E–W.
The Van Gölü Fault was recently identified after the occurrence of two devastating earthquakes in the district of Van, Eastern Anatolia, Türkiye. This significant fault has not been included in the literature because of large volcanic cover in this region. The aforementioned earthquakes have stimulated geological and geophysical surveys around Lake Van, despite the limited amount of seismic, borehole and sedimentological data. There are apparent, E–W trending magnetic anomalies along a slightly curved line crossing the northern part of the lake (indicating the existence of a fault). The main goal of this study was to display the location of this fault using various data processing methods for aeromagnetic anomalies. Based on these processes, the first and second derivatives, analytic signal, total horizontal derivative, tilt, and theta angle maps were constructed to clearly indicate the fault. Considering the geographical location of the fault, it is named as the “Van Gölü Fault” in this study. The Nemrut and Süphan volcanic mountains together with the strong magnetic anomalies to the east of Lake Van are all aligned on the Van Gölü Fault trend and they likely originated from the magmatic-volcanic emplacements related to this recently reactivated fault.
Hydrocarbon reservoir imaging plays a crucial role in the oil and gas exploration industry. Seismic data processing is commonly employed as the primary imaging tool, but interpreting the data in complex geological settings can present challenges. To supplement seismic imaging, magnetotelluric (MT) data offers valuable information on the electrical conductivity of subsurface structures. MT data is highly effective for differentiating the components of a hydrocarbon system characterized by varying resistivity and boundaries; however, it encounters challenges in thinly layered formations, which are often easier to identify through seismic data. In these contexts, the unconstrained inversion of MT data is inherently non-unique. There has been a growing interest in enhancing hydrocarbon reservoir imaging through the cooperative inversion of seismic and MT data which combines information from multiple methods to improve the estimation of subsurface rock properties. To implement this approach, seismic reverse time migration (RTM) result is first utilized to establish appropriate initial boundaries for the MT sharp boundary inversion scheme, resulting in robust and reliable MT seismic-oriented modeling outcomes. To illustrate the feasibility of using seismic imaging techniques as constraints for the MT sharp boundary inversion, synthetic hydrocarbon trapping systems are simulated and investigated to disclose the efficiency of the proposed method.
The generation mechanism for the geomagnetic field perturbation caused by a tsunami wave over land in a coastal region is studied. The sources of perturbation are electric currents in the marine environment and in the ionosphere. The current in conducting seawater arises as a result of its movement in the geomagnetic field during the propagation of a tsunami wave. The current in the ionosphere is caused by the movement of ionospheric plasma under the action of an acoustic-gravity wave radiated into the atmosphere due to the vertical displacement of the seawater surface in a tsunami wave. A model has been constructed describing geomagnetic disturbances caused by a tsunami wave and an acoustic-gravity wave, including the decay of a tsunami wave when the wave front reaches the coastline. The model makes it possible to explain the propagation of geomagnetic perturbations on land over long distances from the coastline. Based on this model, geomagnetic perturbations are calculated depending on time and distance to the coastline. The effects produced by currents in the marine environment and ionosphere are compared.
Stationary ground-based geomagnetic measurements, carried out at geomagnetic observatories, are used in a large amount of geoscience and environmental studies. With the widening of the scope of researches dealing with geomagnetic data and their analysis, the requirements for the initial data quality, their operational acquisition, analysis, processing and storage also increase. This stimulates the need for various upgrades of the existing geomagnetic observatories, as well as for the installation of the new ones. The following article summarizes the main achievements in the upgrade of the Russian segment of geomagnetic observatories, highlights the existing problems and their new solutions and provides some examples of their implementation in the installation and maintenance of geomagnetic observatories; also some recent results based on observatory data processing and analysis are mentioned.
The Ahar–Varzaghan doublet earthquakes (Mw 6.4 and Mw 6.2) in northwestern Iran offer a unique natural laboratory for examining the interplay between seismotectonic processes and hydrological variations across an entire seismic cycle. In this study, we integrate high-resolution spatiotemporal analyses of b-value derived from the Iranian Seismological Center catalog with GRACE Equivalent Water Height (EWH) observations (2006–2024) to assess stress–fluid interactions before and after the mainshock. The pre-seismic stage is characterized by a steady decline in b-value from approximately 0.85 to 0.55 over the 3 years leading up to the event, reflecting increasing differential stress and progressive fault locking. During the same interval, cumulative EWH anomalies exhibit a pronounced mass deficit, reaching approximately − 27 cm and peaking 2–6 months before rupture. This negative anomaly is consistent with poroelastic compaction or fluid withdrawal driven by compressive loading in the near-field of the impending rupture. Following the mainshock, the system evolves in the opposite manner: b-value sharply increases to 0.72–1.02 within 1 year, forming a high b-value zone (> 0.9) surrounding the rupture area, indicative of stress release and widespread microfracturing. In contrast, EWH anomalies remain strongly negative, reaching approximately − 20 cm in the epicentral region, suggesting that fluid drainage persists despite the enhanced fracture permeability generated by the earthquakes. This sustained inverse relationship between b-value and EWH anomalies implies a decoupling between mechanical recovery and hydrological re-equilibration. The results highlight the complementary roles of these parameters: b-value effectively captures short-term stress evolution, whereas EWH anomalies reflect longer-term mass redistribution.
In this study, the potential of artificial neural networks (ANN) for hybrid geoid modelling was investigated, and then ANN was compared with traditional surface models. A mountainous region in central France (Auvergne test-bed) was selected as the study area due to its complex topography and availability of high-resolution gravimetric geoid model as well as homogenously distributed GNSS-levelling benchmarks. The gravimetric geoid model was adapted to the national vertical datum adopting both classical surface models (4-, 5-, and 7-parameter) and feed-forward back-propagation ANN models. The performance of the models was evaluated using Root Mean Square Errors derived from both training and test datasets. The ANN-based model achieved the best test accuracy (2.11 cm), outperforming the best parametric model (2.34 cm). While ANN models offer higher accuracy, they require greater expertise and computational resources. The numerical results highlight the potential of ANN as a viable alternative for geodetic height transformation and exhibit their effectiveness even in a mountainous area.
The Hangzhou–Shaoxing–Taizhou high-speed railway passes below the Xinchang Coal Mine, where the goaf is deeply buried. To scientifically evaluate the impact of the Xinchang Coal Mine goaf on the under-construction Hangzhou–Shaoxing–Taizhou high-speed railway and ensure the safety of engineering construction and railway operation, a comprehensive investigation of the Xinchang Coal Mine goaf was carried out using a combined method of geological survey, drilling, and geophysical exploration. Ground surveys identified two adit entrances (P1, P2) and a mine ventilation shaft (S1). Five transient electromagnetic (TEM) survey lines were arranged above the coal mine with a point spacing of 5 m. The TEM data were finely processed, and the inversion parameters were verified. Borehole verification was conducted to validate the inferred abnormal areas, ensuring reliable results. A total of 18 goaf anomalies were identified. Among them, anomalies C1, C5, C8, and C11 are close to the tunnel, anomaly C16 is located directly above the tunnel, and the remaining anomalies are situated farther away from the tunnel. No water inflow was detected by either geophysical exploration or drilling, confirming the absence of mine water. Finally, the stability of the goaf was evaluated. Calculation and analysis indicate that the duration of surface movement in the goaf was shorter than the mine’s operation period, with a maximum subsidence of approximately 3.99 m in the surface deformation zone. The coal mine goaf is located 39 m above the tunnel roof, which is greater than the tunnel’s safe roof thickness of 8.88 m. Additionally, the tunnel surrounding rock is dense basalt with relatively intact rock mass. It is concluded that the Xinchang Coal Mine goaf has a relatively minor impact on the Hangzhou–Shaoxing–Taizhou high-speed railway.
Determining the geoid gravimetrically poses challenges which are complex mathematical computation and variability representation of the Earth, particularly when precision is necessary. This study aimed to compare the geoid determination methods namely: least squares collocation (LSC), Stokes integral, and Hotine integral methods. Shuttle Radar topography mission (SRTM), gravity anomaly, gravity disturbance, Earth gravity model 2008 (EGM2008), and global navigation satellite systems with levelling (GNSS/Levelling) data were used for the purpose of geoid determination methods. The LSC method was applied using the RCR technique. Whereas Stokes and Hotine integrals methods were computed based on KTH approach with essential corrections, including topographic, atmospheric, ellipsoidal, and downward continuation refinements. All three approaches yielded consistent geoid models with validation against benchmarks produced standard deviations of 0.069 m for LSC, 0.062 m for Stokes, and 0.061 m for Hotine. The findings demonstrate that the LSC provides reliable geoid approximations, Stokes and Hotine integrals attain marginally higher precision. This study highlights the significance of integrating airborne gravity data with global geopotential models and determined the gravimetric geoid model to contribute the development of an accurate nationwide vertical reference for Ethiopia.
Groundwater supplies more than 70% of water demand in SSA regions, but is increasingly under threat from climate change and population growth. Conventional groundwater models are data-intensive and not appropriate for Sub-Saharan Africa (SSA) contexts. This systematic review assesses Machine Learning-Augmented Groundwater Models (ML-AGMs) such as Random Forest, XGBoost, and LSTM algorithms to predict groundwater stress zones with minimal data inputs and satellite-based climate data. Several peer-reviewed articles (2015-2025) were analyzed to assess the efficacy, applicability, and scalability to SSA regions. Findings reveal that ML-AGMs provided excellent prediction potential (RMSE error range: 0.356-0.372 m; R-squared: 0.503-0.679) with minimal data inputs. This represents an important innovation where data-scarce groundwater modeling is possible and dynamic. Google Earth Engine and QGIS enhance spatiotemporal analysis and improve communication with stakeholders. However, this review also highlights limitations of ML-AGM concerning data quality, applicability issues, and ethical considerations. Further research recommendations include hybridized models and inclusion of socioeconomic factors. ML-AGM and applications provide scalable and enhanced Sustainable Development Goal (SDG) 6 implementation. However, issues and limitations persist with aquifer diversity and the equitable application of groundwater stress assessments.
In this study, a low-Earth orbit (LEO) constellation of 160 LEO satellites was simulated, and simulated LEO observations from American regional stations were used to investigate regional augmented precise point positioning (PPP). For reference networks with scales of 75 km and 165 km, the low-order surface model (LSM), the distance-based linear interpolation method (DIM), and the modified linear combination model (MLCM) were respectively used to interpolate atmospheric delays. For both scales of the reference networks, the root mean square errors (RMSEs) of the inter-satellite single-difference (SD) ionospheric slant delay residuals and the tropospheric zenith wet delays (ZWDs) residuals interpolated by the three methods were respectively less than 2.00 cm and 0.25 cm, and the accuracy of the atmospheric delays interpolated by the three methods was comparable. Within the 75‑km and 165‑km reference networks, using the interpolated atmospheric delays to constrain undifferenced and uncombined (UDUC) PPP improved the positioning performance of both the float and fixed solutions, and also augmented the ambiguity resolution (AR) performance of the fixed solutions. The enhancement effects of the three methods were comparable, and the enhancement effect decreased slightly as the scale of the reference network increased. Among the three methods, LSM performed slightly better than DIM and MLCM in the up direction, since the LSM used in this study is an interpolation model that accounts for the elevation factor.
Gravitational methods offer a cost effective and robust approach for delineating subsurface structures, particularly in tectonically complex or volcanically covered terrains where other geophysical tools are limited. This study develops and applies an integrated forward-inversion workflow to investigate concealed mass anomalies in Unai geothermal field, located along the western flank of the Narmada-Son Lineament, southern Gujarat. The methodology combines analytical forward modeling of buried spherical body, polynomial detrending to remove regional trends, sinusoidal correction for oscillatory residuals and a damped Gauss-Newton inversion scheme with Tikhonov regularization to recover geologically plausible parameters. Synthetic testing of the inversion framework yielded parameters estimated with < 1
Recent advancements in remote sensing technologies have enabled the acquisition of large volumes of high-resolution spatial data. However, the classification of complex datasets remains challenging using traditional methods. This study investigates the potential of Mask R-CNN and its variants for advancing the classification of satellite imagery with diverse resolutions and classes. The key advantages of the model, including transferability across regions, high accuracy, and rapid processing when trained on sufficiently large datasets, offer a novel solution beyond traditional methodologies. Extensive experiments were performed on novel datasets containing 1515 images and 18,811 annotated labels generated from Sentinel-2 and WorldView-3 satellite images. In instance segmentation tasks for urban area detection on Sentinel-2 imagery, Mask R-CNN F1 scores of 0.81 and 0.79 were achieved using ResNet-101 and ResNet-50 backbones, respectively. For the multi-class land cover instance segmentation (five classes) of Sentinel-2 data, the model attained an F1 score of 0.75. Finally, building segmentation on WorldView-3 imagery yielded a score of 0.84. The results demonstrate that the proposed methodology can effectively classify satellite images with diverse characteristics and resolutions. This study provides a methodological baseline for potential geographical adaptability, enabling the model to process images across diverse contexts, which is a critical step toward scalable remote sensing applications. The findings highlight the model’s transfer learning capacity and multi-scale data processing proficiency, positioning it as a versatile tool for tasks ranging from land cover mapping to urban monitoring.
Hydraulic fracturing can improve the permeability of shale reservoirs, and accurately acquiring post-fracturing fracture geometric parameters is critical for optimizing subsequent development plans and enhancing hydrocarbon recovery efficiency. This study presents a monitoring method using three electromagnetic parameters based on a single-transmitter and double-receiver coil system in a horizontal borehole. By establishing a finite-element fracture geoelectric model, the study investigates the influence of coil source distance and tool eccentricity on the three electromagnetic parameters, as well as their response characteristics under different fracture parameters. Results show that increasing the coil source distance broadens the response peak width of the three electromagnetic parameters; tool eccentricity has a more significant effect on the parameters at small source distances. The three electromagnetic parameters exhibit distinct resolution capabilities for fracture geometry: The real-part amplitude ratio offers the best resolution for fracture dip angles, the phase difference is most sensitive to changes in fracture length, and the apparent conductivity is most sensitive to variations in fracture width. Additionally, co-axial monitoring of transmitter and receiver coils cannot distinguish fractures with complementary dip angles, whereas when the two receiver coils are tilted during monitoring, all three electromagnetic parameters can effectively distinguish such fractures. This method leverages the differential sensitivity of the three electromagnetic parameters to fracture dip angle, length, and width, selecting the most suitable electromagnetic parameter for precise evaluation according to different fracture parameters and verifying results using the remaining two parameters. It provides a new solution for evaluating shale fracturing effects.
Accurate determination of hydrodynamic parameters holds immense significance in achieving sustainable development objectives and ensuring comprehensive water resources management. This research aims to assess the performance of machine learning models in estimating specific yield using hydrogeological data and pumping test data in the single well system for the unconfined alluvial aquifers of Iran’s central plateau. Artificial neural network, gene expression programming, adaptive neuro-fuzzy inference system, fuzzy logic, least square support vector machine, and group method of data handling were used to estimate specific yield. In order to accomplish this objective, data from pumping tests and hydrogeological evaluations conducted on 83 wells located in the central plateau of Iran were gathered and normalized. The available data was utilized to generate different combinations as input for the models. Subsequently, 70
This study introduces an innovative binary search method for calculating intersection point discrepancies in marine gravity and magnetic survey lines, significantly enhancing the efficiency and accuracy of data processing. The discrepancies at the intersections of survey lines are critical for evaluating the quality of geophysical measurements and serve as essential data for network adjustments and compensation. Traditional methods often suffer from high computational complexity or slow processing speeds, which hinders the rapid and accurate identification of all intersection points. To address these limitations, we propose a binary search approach that efficiently narrows down the search range, significantly reducing the number of data points required for intersection detection. Our method combines mathematical formulas for initial intersection detection with an iterative binary search process, allowing for quick localization of intersection points. The experimental results demonstrate the effectiveness of this approach, achieving a 100
