Although conventional reservoirs dominate the Bohai Basin, China, a new type of sandstone reservoir also exists in the Dongpu depression that has a low matrix porosity (tight) in which natural fractures govern both permeability and porosity. These fractured sandstones are located on a structurally modified buried hill underlying Paleogene mudstones, and are truncated along an angular unconformity. The fractured sandstone oils of the Triassic Liujiagou, Heshanggou, and Ermaying Formations are derived from the Paleogene Shahejie Formation, which reached peak oil generation and expulsion during the Oligocene to early Miocene (32.8–15.6 Ma). Gas was generated primarily during the Paleogene from Carboniferous and Permian coals. Petrographic evidence suggests that oil and gas emplacement followed the compaction and cementation of the Triassic sandstone reservoirs. Fluid inclusion evidence and burial history analysis suggest that fractures developed before oil emplacement but may have coincided with peak gas generation, which suggests that oil and gas mainly migrated and accumulated in fractures.

Paleogene saline lacustrine carbonate rocks are important fractured reservoirs in the western Qaidam Basin. Core data show that most fractures are small, steeply dipping faults; bedding-plane slip faults; and subvertical opening-mode fractures. Other fractures are diagenetic in origin. Fracture occurrence and abundance patterns are controlled by lithology, bed thickness, and proximity to larger faults. Fractures are generally filled with calcite, gypsum, or glauberite (Na2Ca[SO4]2); the degree of fracture filling determines the effectiveness of fractures as fluid conduits and the distribution of high-quality reservoirs. Open fractures not only provide the main pathways for fluid flow, but also enhance the free fluid index and the free fluid saturation measured by nuclear magnetic resonance and determine the potential production rates of tight carbonate reservoirs. The open fractures are parallel to and occur near faults, and many do not coincide with the present-day direction of the maximum horizontal compressive stress.

Natural fractures are the key factors controlling the enrichment of shale oil. It is of great significance to clarify the distribution of natural fractures to guide the selection of sweet spots for shale oil. Taking the Qing-1 Member shale oil reservoir in the northern Songliao Basin, China as an example, a new method considering the factors affecting fracture distribution was proposed to quantitatively predict the structural fractures. And the effect of natural fractures on shale oil enrichment was discussed. Firstly, the types and characteristics of fractures in shale oil reservoirs are characterized by using core and outcrop data. Combined with the experimental analysis, the influences of fault, mechanical stratigraphy, mineral composition and content, TOC, and overpressure on fracture intensity were clarified. Then, the number and density of fractures are quantitatively predicted according to the power-law distribution of fault length. Next, geomechanical simulation and fracture prediction were carried out on the model which was established with comprehensive consideration of the influencing factors of fracture distribution. Finally, the fracture distribution is evaluated comprehensively based on above prediction. The prediction results in this work are consistent with the core measurements.

Using the conventional fracture parameters is difficult to characterize and predict the complex natural fractures in the tight conglomerate reservoirs. In order to quantify the fracture behaviors, a fractal method was presented in this work. Firstly, the characteristics of fractures were depicted, then the fracture fractal dimensions were calculated using the box-counting method, and finally the geological significance of the fractal method was discussed. Three types of fractures were identified, including intra-gravel fractures, gravel edge fractures and trans-gravel fractures. The calculations show that the fracture fractal dimensions distribute between 1.20 and 1.50 with correlation coefficients being above 0.98. The fracture fractal dimension has exponential correlation with the fracture areal density, porosity and permeability and can therefore be used to quantify the fracture intensity. The apertures of micro-fractures are distributed between 10 μm and 100 μm, while the apertures of macro-fractures are distributed between 50 μm and 200 μm. The areal densities of fractures are distributed between 20.0 m·m⁻² and 50.0 m·m⁻², with an average of 31.42 m·m⁻². The cumulative frequency distribution of both fracture apertures and areal densities follow power law distribution. The fracture parameters at different scales can be predicted by extrapolating these power law distributions.