There are mainly two sedimentary types of the 2nd member of Jurassic Sangonghe Formation (J₁s₂) in the Qianshao area: shallow-water delta and sandy debris flow. In order to accurately understand the characteristics and further identify the difference of sedimentary microfacies, the characteristics of the grain size curve are analyzed. The results show that the cumulative probability curve of sandstone particle size in the research area mainly includes two basic types: tractive current type and sandy debris flow type. The tractive current type is mainly developed in the shallow-water delta front. Among them, the inner front subaqueous distributary channel microfacies develop two patterns: tri-segment pattern and one bouncing segment-one suspension segment-one complex transitional zone pattern. The transitional section is widely developed. There are two patterns in the subaqueous distributary channel microfacies of the outer front (bi-segment pattern and one bouncing segment-one suspension segment-one transitional zone pattern) and two patterns in the estuary bar microfacies (bi-bouncing segment-one suspension segment-one transitional zone pattern and one bouncing segment-one suspension segment-one transitional zone pattern). The main tractive current characteristics are higher bouncing component content and slope and lower suspended component content. The sand debris flow type mainly develops two patterns: one bouncing segment-one suspension segment-one transitional zone pattern and one bouncing segment-one suspension segment-one complex transitional zone pattern with lower jump component content and slope and higher suspended component content.

Predicting the remaining time before the next earthquake based on seismic signals generated in a laboratory setting is a challenging research task that is of significant importance for earthquake hazard assessment. In this study, we employed a mel spectrogram and the mel frequency cepstral coefficient (MFCC) to extract relevant features from seismic signals. Furthermore, we proposed a deep learning model with a hierarchical structure. This model combines the characteristics of long short-term memory (LSTM), one-dimensional convolutional neural networks (1D-CNN), and two-dimensional convolutional neural networks (2D-CNN). Additionally, we applied a stacking model fusion strategy, combining gradient boosting trees with deep learning models to achieve optimal performance. We compared the performance of the aforementioned feature extraction methods and related models for earthquake prediction. The results revealed a significant improvement in predictive performance when the mel spectrogram and stacking were introduced. Additionally, we found that the combination of 1D-CNN and 2D-CNN has unique advantages in handling time-series problems.

From 5000 m underground to the surface, there is a multi-layer hydrocarbon reservoir under the Wuerhe nose uplift in the Junggar Basin. Light oil, heavy oil, oil sand, and solid bitumen are found in Permian to Cretaceous strata. The normal crude oil present under heavy oil and solid bitumen reservoir can easily ignored by explorers. To effectively exploit the petroleum and bitumen mineral resources in the Junggar Basin, geochemical characteristics of crude oils in the different layers were analyzed. It is concluded that the crude oils and bitumen minerals came from Permian source rocks of alkaline lacustrine facies. Combined with tectonic movement analysis, two stages of accumulation occurred in research area. During the Indosinian Tectonic movement, the crude oil generated from Permian source rocks first migrated upwards along large faults and then accumulated in the Permian, Triassic, and Jurassic reservoirs. The crude oil of the Jurassic reservoir was seriously biodegraded and a high abundance of 25-norhopane was detected. At the end of the Yanshan movement, small normal faults were developed to connect the oil in the Triassic reservoirs to the surface. The light components of the oil in the fault system quickly volatilized and left solid bitumen minerals in the faults. Due to the plugging effect of “bitumen plug”, the oil and gas in the lower part cannot migrate upward and be damaged. Therefore, light oil-heavy oil–oil sand and bitumen minerals were formed from the bottom to the top. The research result will also have a guiding significance for oil and gas exploration in the northwest margin of the Junggar Basin.

In order to clarify the provenance of the 1st member of the Zhongjiangou formation in Wudun sag, Dunhuang basin, the structural attributes, weathering and sedimentary characteristics of the provenance area were analyzed by means of element geochemistry, so as to determine the differences of sediment sources in different well areas. The results show that the higher the Al₂O₃ and K₂O contents, the higher the enrichment of large ion lithophile elements and high field strength elements, while the iron and magnesium elements are relatively deficient, and there are characteristics of medium degree differentiation of light and heavy rare earth elements in Well XC1 and Well D2. The lower the Al₂O₃ content and the higher the SiO₂ content, a loss of large ion lithophile elements and high field strength elements are observed, while the ferrophilic magnesium elements show serious loss, as shown in the characteristics of the high degree of differentiation of light and heavy rare earth elements in Well D1. In the UCC-normalized element spidergrams, the trend of Well XC1 and Well D2 is similar, which is different from that of well D1, indicating that the sediments of Well XC1 and Well D2 come from the same provenance area, while the sediment of Well D1 comes from a different provenance area. The provenance area of Well XC1 and Well D2 shows strong tectonic activity and strong weathering, while the provenance area of well D1 exhibits relatively weak tectonic activity and weathering. Combined with previous research results, Wudun sag is mainly characterized by a faulted lacustrine basin controlled by the southern boundary fault in the Jurassic layer. Therefore, the sediments of Well XC1 and Well D2 mainly come from the southern Sanweishan uplift provenance area, with strong tectonic activity; the sediments of Well D1 mainly come from the northern Beishan provenance area, with relatively weak tectonic activity.

To promote oil and gas exploration of the Carboniferous formation in the Shibei sag, the northeastern margin of Junggar Basin, recently drilled rocks from Well ZB6 with typical seismics were characterized. Through systematic core observation, the identification marks of seismites were described, a vertical sequence of seismites was established, and its oil and gas geological significance was analyzed. The results show that the seismites have typical identification marks, such as soft-sediment deformation structures (including five typical marks: liquefied stone vein, liquefied crinkled deformation structure, ball–pillow structure, flame structure and load cast, water release structure and liquefied breccia), brittle fracture structures (including three typical marks: seismic fractures, synsedimentary microfractures and seismic fracture rock) and special rock types, such as seismic grain-supported conglomerates. The stratigraphic succession reconstructed in Well ZB6 was characterized, from base to top, by (1) a basal non-seismic interval; (2) a seismic interval made up of a grain-supported conglomerate level, brittle fracture level, soft-sediment deformation level; and (3) a non-seismic interval. The discovery of seismites has oil and gas geological significance for improving reservoir performance and forming favorable source–reservoir–cap assemblages. The research describes the new reservoir genetic type and exploration direction of the Carboniferous formation in the Shibei sag (China), which has important guiding significance for the next step of oil and gas exploration.

Shallow-water delta front deposits represent important hydrocarbon reservoir bodies in continental lacustrine basins, and reservoir quality differentiation among varied sedimentary microfacies assemblages remains a core constraint on exploration evaluation. Taking the second member of the Jurassic Sangonghe Formation in the Moxizhuang area, central Junggar Basin, as a case study, this paper classifies sedimentary microfacies combination patterns and clarifies their genetic mechanisms of reservoir differentiation. Four patterns are identified through systematic core observation: superimposed subaqueous distributary channel (SC), truncated subaqueous distributary channel (TC), superimposed mouth bar (SMB), and upper channel and lower residual bar (CMB). Their formation conditions and reservoir features are further compared via hydrodynamic index calculation and blue epoxy-impregnated thin-section analysis. The results indicate that SC and TC patterns are widely developed in the study area, while SMB and CMB patterns are relatively limited. Hydrodynamic intensity decreases in the order of TC > CMB > SC > SMB, jointly controlled by base-level fluctuations and facies positions. SC reservoirs, dominated by rigid lithic fragments with good sorting, weak cementation, and intense dissolution, have the highest porosity and permeability with weak heterogeneity, constituting the most favorable reservoir type. TC reservoirs maintain relatively high porosity and permeability but show strong heterogeneity. SMB and CMB reservoirs, dominated by ductile lithic fragments with strong compaction and cementation, generally have poor petrophysical properties and are classified as unfavorable reservoirs. Rather than identifying entirely new microfacies architectural types, the core contribution of this study is the establishment of a quantitative genetic linkage between depositional hydrodynamic regimes, petrographic diagenesis, and reservoir quality differentiation in shallow-water delta fronts. These findings provide geological support for hydrocarbon exploration and favorable reservoir prediction in analogous lacustrine delta settings.