Geomechanics and Engineering Evaluation of Fractured Oil and Gas Reservoirs: Progress and Perspectives

1. Introduction

Unconventional oil and gas exploration and development are entering an era of interdisciplinary technological revolution. Driven by the global energy transition and carbon neutrality goals, traditional oil and gas resources are gradually depleting [1]. A key focus in petroleum geology and engineering has become how to efficiently and precisely evaluate and develop fractured reservoirs, which often exhibit complex geological conditions and strong reservoir heterogeneity [2,3,4,5]. In recent years, significant progress has been achieved in the geomechanics and engineering evaluation of fractured reservoirs, thanks to continuous advances in numerical simulation, experimental testing, seismic and well-logging techniques, and intelligent algorithms [6,7,8,9,10]. These advances provide a solid theoretical and technical foundation for the exploration and development of fractured reservoirs.

This Research Topic brings together cutting-edge research from around the world, encompassing various types of unconventional reservoirs such as shale gas, tight oil, coalbed methane, and metamorphic basement reservoirs (buried hills). The published papers integrate multi-scale and multi-disciplinary approaches in fractured and other unconventional reservoirs, aiming to advance quantitative fracture description, deep reservoir pore preservation mechanisms, innovative fracturing, and nano-EOR (enhanced oil recovery) techniques, as well as the fusion of multi-source data with intelligent algorithms. Collectively, these studies lay a robust foundation for the efficient and environmentally friendly development of unconventional resources (Figure 1).

2. Overview of Published Articles

This Research Topic comprises 15 papers that investigate a wide range of topics related to fractured and unconventional reservoirs. The study targets include shale gas, tight oil, coalbed methane, and associated geological problems. The methodologies span field geological surveys, core and laboratory analyses, numerical simulations, geophysical techniques, and artificial intelligence (AI) algorithms, reflecting a high degree of interdisciplinarity and innovation. The main contributions of the published papers are summarized as follows:

1. Expansive soil stabilization for engineering foundations (contribution 1). The authors studied the hazards of expansive soils to engineering structures and evaluated improvement methods. The results show that adding stabilizers significantly reduces the free swell ratio and swelling pressure of the expansive clay (by about 65% and 76%, respectively). Thus, a combined lime–cement–silica fume treatment effectively suppresses the volume changes in expansive soil, improving its stability.

2. Reservoir characteristics of siliceous shale (contribution 2). This study reported the characteristics and controlling factors of Upper Permian Dalong Formation siliceous shale in the western Hubei region of South China. The results revealed that abundant organic-matter-generated nanopores contribute most to the micro- and mesopores of the reservoir, and high total organic carbon (TOC) content is the primary factor promoting micro- to mesopore development and increasing methane adsorption.

3. Quantitative evaluation of tectonic fractures (contribution 3). The authors investigated the degree of natural fracture development in the Fourth Member of the Leikoupo Formation (Pengzhou area, Sichuan Basin). This research, which combines structural mechanics simulations with rock failure criteria, provides a new approach to quantitatively assessing fracture development in complex structural settings, offering insight for predicting gas accumulation in similar regions.

4. Fluid overpressure and fracture slip in induced seismicity (contribution 4). The authors explored the mechanism by which fluid overpressure induces fracture slip, based on laboratory tests related to reservoir-induced seismicity. They created granite cylinder samples with critically stressed fractures and performed cyclic hydraulic loading and shearing experiments under hydro-mechanical coupling, simulating the effects of periodic reservoir impoundment on rock fractures.

5. Sedimentary facies control on deep shale gas sweet spots (contribution 5). The authors examined how sedimentary facies characteristics affect the development of high-quality shale gas “sweet spots” in the deep Longmaxi Formation (lower Silurian) of the southern Sichuan Basin. This work reveals the controlling role of depositional environment evolution in deep shale gas sweet spot formation, providing a geological basis for predicting high-quality shale gas reservoirs.

6. Reservoir model of lacustrine volcaniclastics. Shi et al. (contribution 6) investigated a lacustrine subaqueous volcaniclastic reservoir in the Chaganhua Subsag of the Songliao Basin. They analyzed the reservoir space and petrophysical differences among tuff, resedimented tuff, and tuffaceous sandstone formed by underwater volcanic eruptions, and discussed the diagenetic evolution responsible for these differences. The study concludes that the near-vent coarse tuff facies, with their coarse grain size, plentiful glass shards, and soluble components, develop the best porosity after diagenetic alteration and are the most favorable target for hydrocarbon exploration in this lacustrine volcanic setting.

7. Viscoelastic creep behavior of Paleozoic shales. Wilczyński et al. (contribution 7) conducted triaxial creep experiments to study the viscoelastic deformation of Paleozoic shales from the Baltic Basin in Poland. They developed a relatively simple method to determine the parameters of the Burgers rheological model for the shales. This work offers a new quantitative approach to obtaining shale creep parameters, which is valuable for improving unconventional reservoir geomechanical models and enhancing the effectiveness of fracturing stimulation.

8. Paleo-stress and present stress in coalbed methane reservoirs (contribution 8). Focusing on the Dahebian Block of the Liupanshui Coalfield in western Guizhou, China, this study examined the paleo-tectonic stress, current in situ stress field, and their implications for coalbed methane (CBM) development. The study suggests that fracture density, together with the current stress state (which governs fracture opening or closure), controls the permeability of the coal reservoir, thus impacting CBM extraction.

9. Fracture zones in metamorphic basement reservoirs. Lu et al. (contribution 9) performed a detailed characterization of internal fracture zones within an Archean metamorphic buried-hill reservoir in the Bozhong Depression (Bohai Bay Basin). This work highlights the decisive role of fracture zones in forming high-productivity reservoirs in such crystalline basement settings, providing valuable insights for the evaluation of hydrocarbon potential in ancient basement formations.

10. Pulsed plasma stimulation experiments. Khalaf et al. (contribution 10) conducted an experimental study on pulsed plasma stimulation (plasma pulse fracturing) and matched the results with simulations. They designed a laboratory-scale plasma fracturing apparatus and carried out multiple tests on limestone and sandstone samples, analyzing how factors such as confining pressure, rock type, discharge energy, and electrode wire length affect fracture generation. This research provides experimental validation and modeling insight for a novel stimulation technology aimed at enhancing production from tight reservoirs.

11. AI-based reservoir facies classification from logs. Ye et al. (contribution 11) proposed a method for reservoir facies classification in fractured-vuggy carbonate reservoirs using well-logging data and artificial intelligence. Adopting an unsupervised machine learning approach, they introduced an improved density-sensitive fuzzy c-means clustering algorithm (FCM-DSDFP) to automatically classify logging curves, combined with principal component analysis (PCA) for noise reduction and dimensionality reduction. This demonstrates the potential of intelligent algorithms to enhance subsurface facies analysis in complex carbonate reservoirs.

12. Molecular deposition film for enhanced oil recovery. Shao and Chen (contribution 12) reported a new technique to improve the recovery of low-pressure tight oil reservoirs using molecular deposition film (MDF) technology. This study highlights the potential of nanoscale surface modification to increase recovery in tight reservoirs, suggesting that adjusting the physico-chemical properties of the oil–water interface can partially replace or complement conventional EOR methods.

13. Pore preservation in ultra-deep shale. Wang et al. (contribution 13) conducted an in-depth study on pore evolution and preservation mechanisms in an ultra-deep shale gas reservoir of the Wufeng–Longmaxi Formation in the eastern Sichuan Basin. These findings suggest that overpressure, aided by hydrocarbon generation, is a crucial mechanism for pore preservation in ultra-deep shales.

14. Strike-slip faults and hydrocarbon sweet spots. Liu et al. (contribution 14) investigated the identification of strike-slip faults in a tight carbonate gas reservoir (Middle Permian Maokou Formation, southern Sichuan Basin) and their influence on gas accumulation. This implies that fault intersections, often associated with increased fracture density and permeability, could serve as key exploration targets for tight gas reservoirs.

15. Multi-stage stress evolution and fracture prediction. Wang et al. (contribution 15) performed a comprehensive study on the evolution of tectonic stress and regional fracture prediction in the concealed, coal-bearing Tucheng Syncline of western Guizhou, China. This work provides a better understanding of the multi-phase tectonic history and its impact on fracture development in coal-bearing strata, offering guidance for CBM exploration and safe extraction in structurally complex settings.

3. Discussion

3.1. Advances in Fractured Reservoir Research

Fractured reservoirs have long been both a crucial and challenging focus in oil and gas exploration. Traditionally, geologists relied on outcrop and core observations to qualitatively describe fractures. Today, however, numerical modeling and intelligent sensing are elevating our ability to quantitatively characterize fracture networks. Several papers in this Research Topic achieved important progress in understanding fracture formation mechanisms, predicting fracture distribution, and assessing the impact of fractures on reservoirs. Overall, for fractured oil and gas reservoirs, developing quantitative methods for fracture characterization and prediction remains a hot and difficult topic [11,12,13,14]. Importantly, accounting for multi-phase tectonic events and stress superposition effects is key to understanding fracture development.

Currently, a common approach is to integrate multi-source data for fracture analysis. This involves combining information from outcrop studies, cores, thin sections, conventional well logs, formation imaging logs, and 3D seismic interpretation, supplemented by advanced remote sensing techniques (e.g., drone-based imaging). By analyzing these datasets across multiple scales, researchers can quantitatively describe fracture networks from micro- to macro-scale and investigate the structural and non-structural factors that control fracture development. An effective strategy is to establish a workflow where macro-scale geological analysis and micro-scale experimental tests inform each other for fracture timing and genesis identification. Macro-scale analysis includes reconstructing tectonic evolution, and examining fracture orientations and cross-cutting relationships across outcrops, cores, and thin sections. Micro-scale tests may involve rock acoustic emission experiments, geochemical analysis of fracture fill (carbon–oxygen isotopes, fluid inclusions), electron spin resonance (ESR) dating, and apatite fission-track analysis. By correlating these scales and data, one can determine fracture chronologies and their driving mechanisms. Building on this foundation, predictive modeling of fractures has advanced considerably. Techniques such as paleo-stress field numerical simulation (calibrated with structural geology and rock failure criteria), seismic multi-attribute analysis for fracture indicators, and AI-driven interpretation of remote sensing data are employed to achieve multi-scale fracture prediction.

In recent years, the application of remote sensing technology in evaluating fractured reservoirs has gained increasing attention. For example, satellite-based synthetic-aperture radar (SAR) and interferometric SAR (InSAR), high-resolution optical imagery, and airborne or ground-based LiDAR can help identify surface lineaments, fault zones, and subtle deformation related to subsurface fractures. These remote sensing methods, when integrated with subsurface data, provide a powerful means to infer fracture patterns and in situ stress fields over broad areas.

3.2. Novel Methods for Deep Reservoir Characterization

Deep unconventional reservoirs (deeply buried shales, ultra-deep carbonates, volcanic reservoirs, etc.) have long been difficult to evaluate due to their great depths and complex genesis. A clear trend is that multi-scale, multi-method comprehensive characterization has become the baseline strategy for deep reservoirs [15,16,17].

Researchers are combining a wide array of advanced experimental techniques to scrutinize everything from microscopic pore structures to macroscopic geological processes, thus achieving a full-spectrum characterization. Another notable development is the recognition and definition of new types of unconventional reservoirs and their unique characteristics. In this Research Topic, several special reservoir types are examined. For example, lacustrine subaqueous volcaniclastic reservoirs represent a hybrid sedimentary–volcanic reservoir type. Another example is metamorphic basement buried-hill reservoirs in which the crystalline rock is fractured: despite the crystalline nature, extensive fracture zones within the buried hill can store significant hydrocarbons, making them a hot exploration target in recent years. For instance, in ultra-deep settings, identifying abnormal pore pressure zones becomes crucial, and for volcanic reservoirs, quantifying diagenetic porosity (such as that from devitrification) is necessary for proper assessment. Equally important for deep reservoirs is a thorough understanding of their genetic and evolutionary mechanisms [18,19]. Deep reservoirs’ present-day properties are often the cumulative result of multiple geological processes over time. Only by unraveling the formation and evolution history can we understand why a deep reservoir has good or poor quality today.

3.3. Multi-Source Remote Sensing and Unconventional Energy Coupling

“Multi-source remote sensing” in the context of energy geoscience is a broad concept. It includes not only surface remote sensing (e.g., satellite imagery, aerial geophysics) for extracting geological and environmental information, but also subsurface sensing in a geophysical sense (seismic surveys, well logs, electromagnetic detection), as well as various sensor-derived data streams. Integrating multi-source remote sensing data with unconventional oil and gas exploration and production is an emerging trend in fractured reservoir evaluation.

Firstly, integrating surface and subsurface observations across scales is crucial for unconventional resource exploration. Secondly, multi-disciplinary data fusion is another form of “multi-source remote sensing”. This refers to integrating data from different domains—geological, geophysical, geochemical, etc.—to solve complex problems. In this Research Topic, many papers fuse different data types to draw conclusions (for example, combining seismic attributes with log and core data, or integrating stress modeling with microseismic monitoring). Thirdly, coupling real-time remote monitoring with unconventional field operations is a promising direction. The development of unconventional oil and gas often involves dynamic operations like hydraulic fracturing, fluid injection, or CO₂ sequestration, which can have environmental effects such as surface subsidence, induced seismicity (micro-earthquakes), and potential groundwater contamination.

3.4. Intelligent Algorithms in Energy Geoscience

Intelligent algorithms—primarily artificial intelligence (AI) and machine learning (ML) techniques—are rapidly being incorporated into geoscience research [19,20]. Several works in this Research Topic demonstrate the potential of AI for pattern recognition, parameter optimization, and predictive analysis in geological contexts, representing a broader trend toward digitalization and intelligent workflows in energy geoscience.

Firstly, AI greatly improves the efficiency and objectivity of pattern recognition in geosciences. Traditional geological analysis often depends on expert interpretation of images, logs, and other data to recognize patterns (such as facies changes, fractures, or sedimentary structures). AI, on the other hand, excels at learning hidden patterns from large datasets and can automate the recognition process. Secondly, AI can be used for modeling and optimizing complex processes. Geological and reservoir processes in unconventional resources are often highly non-linear and involve many parameters. Traditional physics-based models (while essential) sometimes struggle to capture all the nuances or require simplifications. Machine learning models can serve as a complementary tool by using data-driven approaches to approximate complex processes like fluid flow in fractured media, pressure depletion, or enhanced recovery mechanisms. Moreover, AI algorithms (such as evolutionary algorithms or reinforcement learning) can assist in optimizing engineering designs—for example, finding the best combination of fracture spacing and pumping schedule in a hydraulic fracturing operation to maximize recovery. Thirdly, AI has the ability to uncover correlations and controlling factors that traditional analysis might miss. Unconventional oil and gas systems are complex, with vast amounts of data coming from different sources (geochemistry, petrophysics, microseismic, production history, etc.). AI techniques, particularly those in the realm of big data analytics and deep learning, have the potential to mine these datasets for patterns—for example, identifying which combination of geological features and operational parameters results in the best well performance. It is important to note, however, that applying AI in the energy domain also faces challenges. Often, the available training datasets are limited (for example, only a modest number of wells in a new shale play), which makes it difficult for complex models to generalize. Additionally, many AI models, especially deep learning ones, act as “black boxes” with limited interpretability—which means geoscientists may be hesitant to trust their predictions without understanding the reasoning. Therefore, a collaboration between AI and human experts—an “AI + geologist” synergy—is crucial.

4. Conclusions

This Research Topic gathers 15 papers that advance the geomechanics and engineering evaluation of fractured oil- and gas-bearing reservoirs. The contributions fall into four themes. (i) Fracture quantification: Coupling paleo-stress reconstruction, finite element modeling, and multi-scale imaging delivers macro-to-micro mapping of fracture networks. (ii) Deep reservoir porosity preservation: New evidence shows that diagenetic–tectonic coupling, hydrocarbon-generation overpressure, and devitrification jointly safeguard pore space in ultra-deep shales and basement highs. (iii) Engineering innovations: Experiments and simulations confirm that pulsed-plasma stimulation, nano-fluid flooding, and molecular-film agents markedly enhance flow in tight formations. (iv) Data fusion and AI: An improved density-sensitive fuzzy C-means algorithm and deep learning-based fault detection illustrate how AI, InSAR, LiDAR, seismic, and logging data converge to predict fractures and optimize production.

Looking forward, further progress can be achieved by the following means: (1) multi-source data fusion—combining satellite remote sensing and big data analytics to build dynamic 3D/4D models of unconventional systems; (2) ultra-deep and novel reservoirs—linking lab simulations with field pilots to validate mechanisms such as pore preservation and innovative stimulation; (3) intelligent geoscience—developing explainable, generalizable AI frameworks trained on high-quality datasets so that AI becomes a trusted assistant.