Dielectric permittivity measurements are used across various geoscience-related applications for the purpose of determining soil moisture content, characterizing soil and rock properties, and monitoring geoprocesses. However, their accuracy is often compromised by the diverse composition of natural brines. A key aspect of interpreting dielectric permittivity measurements is having a reliable model for the saturating brine solution. It is unclear whether the existing models, typically based on single-salt NaCl solutions, fail to account for diverse ionic mixtures. We address this gap by developing a comprehensive dielectric model from an extensive experimental dataset of eight salts (i.e., NaCl, KCl, CaCl 2 , MgCl 2 , K 2 CO 3 , K 2 SO 4 , LiCl, NaBr). This dataset includes more than 200 measurements of both single salt solutions and complex mixtures. The results demonstrate that for brines dominated by Na + , Ca 2+ , and Cl - ions, a simple NaCl-based model can be used in practical applications in the absence of extended ionic analysis when the salt concentration is below 100 kilo parts per million (kppm). However, at salt concentrations above this threshold, the permittivity of the aqueous salt solution needs to be determined by considering the specific concentration of each constituent ion. For diverse multi-component brines, we propose a linear mixing rule that predicts static permittivity with a mean absolute percentage error of 0.7%. This work provides an experimentally grounded framework for bulk brine permittivity under controlled laboratory conditions (20.2 ± 0.1 °C, atmospheric pressure), serving as a foundational baseline for calibration standards used in hydrology, geosciences, and petroleum engineering applications.

Hydrogen sulfide (H 2 S) is a relatively common component in hydrocarbon fields, where it may be mixed with hydrocarbon oil or gas in proportions of up to 50% or more. Such hydrocarbons are often described as ‘sour’. The H 2 S primarily originates from thermochemical sulfate reduction associated with evaporites, although biogenic pathways may apply in some cases. Hydrocarbon fields with the highest concentrations of H 2 S often remain undeveloped, representing already-discovered resources that could support the transition towards a lower-carbon economy. Meanwhile, hydrogen – recognized as a critical element of the energy transition – can be obtained from H 2 S currently by several energy consuming processes. A new subsurface engineering concept introduced here combines the rehabilitation of stranded sour hydrocarbon resources via H 2 S removal with the production of potentially economic amounts of hydrogen. The proposed approach removes H 2 S from the hydrocarbons as they are passed through a subsurface iron-bearing ‘scavenging’ reservoir. Reactions between the sour hydrocarbons and the iron minerals in this reservoir convert H 2 S to solid iron sulfide (pyrite) releasing hydrogen gas during the process. Sweetened hydrocarbons, hydrogen or both can then be produced. Subsurface removal of H 2 S and sequestering of sulfur from known stranded hydrocarbons avoids the cost and risk of surface-based H 2 S facilities, as well as exploration costs for new hydrocarbons in pristine locations. Here we term hydrogen produced from H 2 S in this way ‘amber hydrogen’, an addition to the hydrogen colour spectrum that can also be applied to hydrogen produced from H 2 S by any method.

By constructing and dynamically simulating a three-dimensional geological model of a hydrocarbon prospect in the North Sea, several development strategies were tested to assess how hydrocarbon fields in sand injectite systems could be optimized. Significant reductions in water production and a moderate increase in oil can be achieved if a well is placed at a shallow depth in a sand-injectite system within an interval where sand injectites are not normally resolved on seismic data, net-to-gross (N/G) is low (typically less than 0.1) and stand-off to the oil–water-contact (OWC) is maximized. The low N/G of such wells challenges the paradigm that a successful development well must have a high proportion of net sand. While the cumulative volume of oil production may be similar to a development well placed closer to an OWC in a seismically mapped drill target, the cumulative volume of water produced may be significantly reduced. In the simulations presented here a broad K v / K h range was used to test the gross vertical connectivity uncertainty within the sand injectite system. A sweet spot exists at a K v / K h of around 0.1 – above that value water breakthrough and cut were accelerated, while below that sand injectite connectivity and water flood were choked back. The gross architecture and flow potential of both seismically mapped and non-seismically resolved sand injectites should be considered. Scenarios may exist where lateral sweep within a seismically mapped sand injectite is enhanced due to reduced vertical connectivity within the non-seismically resolved sand injectites.

Kerogen wettability can significantly affect the preferential movement of fluids in organic-rich mudrocks as kerogen constitutes a significant fraction of mudrock volumes. In previous publications, the determination of the wettability of kerogen and organic-rich mudrocks was typically achieved using contact angle measurements through the sessile drop method, which might not be considered to be a ground truth quantitative measure of wettability. This method also requires pellets of kerogen to create a surface for the contact angle to be measured. No standardized procedure exists for making pellets under stress and with saturating fluid to replicate reservoir conditions. In this paper, we introduced a novel method for quantifying the wettability of kerogen, collected from different organic-rich mudrock formations (i.e. formations A, B and C), as a function of thermal maturity using adsorption isotherms. We compared the results from the adsorption isotherm experiments with contact angle measurements. Results demonstrated a reduction in water adsorption capacity as thermal maturity increases. The cumulative amount of adsorbed water (within the range of 0–60% relative humidity) in the isolated kerogen sample decreased by 75.3 and 91.8% for formations B and C, respectively, compared with formation A. Kerogen from formation A formed a 50° air/water contact angle, whereas kerogen fromformation B formed a 111° air/water contact angle and kerogen from formation C formed a 109° air/water contact angle.

The Clair Field is thought to contain the largest hydrocarbon accumulation on the UK continental shelf, with the principal reservoir comprising low matrix-permeability, naturally fractured, Devonian clastic red-beds of the Lower Clair Group (LCG). The pre-2020 development plan on Clair Ridge focused on achieving production rates via targeting natural fractures. However, recent challenges with open fracture prediction (from early-water-breakthrough to low productivity wells) has led to a renewed focus on understanding the matrix. This study summarises the predictability of matrix character across the LCG, specifically to (i) optimise well placement (matrix property trends), (ii) understand completion requirements (rock strength trends), and (iii) improve pre-drill deliverability predictions and well sequences. The study integrated a range of reservoir quality analysis with sedimentology, core observations, petrophysical log, and heavy minerals. Analysis unveiled a relationship between the resulting sedimentological characteristics (facies), diagenetic overprint, primarily calcite cement, and the contrasting styles of reservoir quality across the LCG units which denote fluctuating climatic conditions. Furthermore, constraining spatial trends in clay mineral volumes (source of calcite cement) within a resolved depositional model has produced a pre-drill predictive capability in cement volumes and facies (matrix character). The availability of a predictive model for matrix character has increased confidence in well performance prediction and impacted well sequencing, reservoir targeting and completionsstrategies. Together with a successful production-increasing drilling strategy, this has increased production on Clair Ridge from 40 mboed (thousand barrels of oil equivalent per day) to 80 mboed in 2021 alone.

Accurate porosity prediction is essential for reliable reservoir characterization in data-limited and heterogeneous formation. Traditional approaches generally have a difficulty handling the inherent complexities and uncertainties of well log data. This study applies and compares three machine learning (ML) approaches, including Artificial Neural Network optimized with Levenberg-Marquardt (ANN-LM), Random Forest (RF), Fuzzy Logic (FL) along with a baseline Multiple Linear Regression (MLR) model, to estimate total porosity from standard geophysical well-logs in three wells from the Mazalai Gas Field (MGF), Kohat Basin, Pakistan. The models utilize sonic, neutron porosity, bulk density, and gamma ray as input parameters. The ANN-LM model was trained using backpropagation and K-fold cross-validation. RF was implemented as an ensemble of decision trees with feature ranking, FL employed Gaussian membership functions in ten bins, and MLR served as a baseline linear method. Model performance was evaluated using the coefficient of determination (R²) and root mean square error (RMSE). ANN-LM showed the strongest generalizability and robustness, achieving R² = 0.99 and RMSE = 3.5 pu by effectively minimizing errors in complex, nonlinear and heterogenous data. RF and FL performed reasonably well achieving R 2 equal to 0.89 and 0.85 respectively, but showed reduced generalization to unseen data. MLR demonstrated the lowest performance acquiring R 2 =0.82. Additionally, A Taylor diagram analysis revealed that ANN-LM provided the most accurate and statistically consistent predictions, closely matching the reference data. These results show machine learning, especially well-optimised neural networks, greatly improves porosity prediction from logs, strengthening reservoir evaluation and development planning in MGF-like settings.

Significant progress has been achieved in Paleogene hydrocarbon exploration in the Pinghu Slope Belt of the Xihu Depression, but the mechanisms by which fault-caprock configurations control differential accumulation remain unclear. Using 3D seismic interpretation, geochemical data, fault activity analysis, and fluid-inclusion geochronology, this study investigates the multistage evolution of the Pinghu Fault (F1) and its coupling with caprock development in governing hydrocarbon migration. The results indicate that the Pinghu Slope Belt has excellent source rock conditions, and the geochemical characteristics of oil and gas suggest that hydrocarbon accumulation is characterized by near-source hydrocarbon charging. F1 evolved from multiple isolated segments into a unified fault plane through lateral and dip linkage, followed by late-stage segmented reactivation. Source rocks reached peak hydrocarbon generation by the end of the Miocene, with two key charging events at ∼15 Ma (local) and ∼5 Ma (regional). Pre-Late Miocene dip linkage and subsequent reactivation provided critical vertical migration pathways during peak generation. Although thick mudstone caprocks occur in the Pinghu Formation, faulting has disrupted their continuity. Analysis of fault-caprock configurations shows that seal integrity is lost when residual thickness falls below 63.6 m, while fault throws under 100 m reduce accumulation potential near faults. Under a multiphase tectonic background, the coupled fault-caprock sealing capacity plays a critical role in hydrocarbon migration and vertical distribution. This provides important insights for predicting exploration targets in faulted basins.

High-quality source rocks exert primary control on hydrocarbon accumulation in continental lacustrine basins, particularly for tight sandstone oil reservoirs. The identity of the principal source rocks for Chang 7 tight oil in the Yanchang Formation, Ordos Basin, and their control mechanisms on accumulation and enrichment remain uncertain. This study integrates gas chromatography–mass spectrometry (GC–MS) analysis of mudstones, oil shales, and crude oils with hydrocarbon generation–expulsion simulations, abnormal pressure calculations, and fluid inclusion trapping pressure reconstruction to establish oil–source correlations, characterize primary source rocks, and define controlling accumulation mechanisms. Results demonstrate that: (1) Chang 7 tight oil was sourced predominantly from Chang 7 oil shales in a "lower generation–upper reservoir" configuration. (2) These oil shales are laterally extensive with substantial thickness (>15 m), high organic matter abundance (average TOC 11.36 wt.%), excellent kerogen quality (Types I and II₁), and moderate thermal maturity (average Ro 0.85%). (3) Cumulative hydrocarbon generation intensity averaged 159.46 × 10⁴ t/km², providing abundant material for tight oil accumulation. (4) Source–reservoir pressure differentials (SRPD) averaging 15.58 MPa provided the necessary driving force for efficient oil charging into tight reservoirs. (5) High-quality oil shale distribution directly governs tight oil distribution, with transitional zones between hydrocarbon generation centers and high-pressure domains representing optimal enrichment fairways. These findings clarify the fundamental role of lacustrine oil shales in tight oil systems and provide practical guidance for exploration in analogous continental basins.

Shale gas, a clean energy source with large reserves and wide distribution, is gaining global attention. The Nanpanjiang Basin, in the southern part of Yangtze Block, is a strategic area for marine shale-gas exploration, with the Tian'e region as a key target. Field investigations and previous studies have confirmed the distribution of Lower Devonian Tangding shale in the Nanpanjiang Basin. This study, using organic geochemistry, X-ray diffraction and scanning electron microscopy, analysed the geochemical characteristics of source rocks, shale reservoir properties, gas content and preservation conditions. The Tangding shale is 100–250 m thick, with burial depths of 2100–4200 m. The total organic carbon (TOC) values of the shales exceed 2.0 wt%, comprising mainly kerogen types Type II 1 –II 2 and high- to over-mature organic matter, indicating excellent source-rock potential. The shales contain a high percentage of brittle minerals, with well-developed pore spaces and adsorption capacities, suggesting a good shale-gas reservoir. A relatively high clay mineral content, along with strong compaction and cementation, enhances the shale's self-sealing capacity, ensuring good preservation conditions for shale gas. The gas content is relatively high, indicating significant shale-gas accumulation. Multi-episode tectonic movements have significantly influenced shale-gas preservation. Compared with typical shale-gas accumulation conditions in other basins, the Tangding shales in the Tian'e area offer favourable conditions for shale-gas accumulation, making the northwestern part of the Tian'e area an important target zone for shale-gas exploration in the Nanpanjiang Basin.

Dielectric permittivity mixture models often assume simplified rock geometries, limiting their accuracy in rocks with complex pore structures. Systematically evaluating the influence of pore geometry, grain shape and grain size on model performance for water-saturation assessment is experimentally challenging and thus largely untested. Frequency-domain dielectric permittivity simulations, however, provide a means to effectively model these geometrical influences at the pore scale. Therefore, this paper aims to: (1) investigate the influence of grain geometry (size, shape and alignment) on dielectric permittivity using synthetic samples; and (2) evaluate the mixture model performance in assessing water saturation in synthetic and actual rocks. We performed frequency-domain simulations in the frequency range of 10 Hz–5 GHz. The dielectric permittivity dispersion significantly increased as grains flattened (i.e. the aspect ratio increased). The frequency-domain simulations conducted over the range of 10 MHz–5 GHz showed that grain size had a negligible impact on permittivity above 10 MHz. We observed that the relative permittivity in the z direction decreased with an increased aspect ratio of the grains. Simulations suggested that directional permittivity measurements can enhance grain-shape characterization. The unique contribution of this paper is the comprehensive quantification of the impacts of grain size, shape and alignment on the dielectric permittivity. Conducting such an investigation is challenging and almost impossible in the core-scale domain.