Deep learning algorithm-enabled sediment characterization techniques to determination of water saturation for tight gas carbonate reservoirs in Bohai Bay Basin, China.

Understanding water saturation levels in tight gas carbonate reservoirs is vital for optimizing hydrocarbon production and mitigating challenges such as reduced permeability due to water saturation (Sw) and pore throat blockages, given its critical role in managing capillary pressure in water drive mechanisms reservoirs. Traditional sediment characterization methods such as core analysis, are often costly, invasive, and lack comprehensive spatial information. In recent years, several classical machine learning models have been developed to address these shortcomings. Traditional machine learning methods utilized in reservoir characterization encounter various challenges, including the ability to capture intricate relationships, potential overfitting, and handling extensive, multi-dimensional datasets. Moreover, these methods often face difficulties in dealing with temporal dependencies and subtle patterns within geological formations, particularly evident in heterogeneous carbonate reservoirs. Consequently, despite technological advancements, enhancing the reliability, interpretability, and applicability of predictive models remains imperative for effectively characterizing tight gas carbonate reservoirs. This study employs a novel data-driven strategy to prediction of water saturation in tight gas reservoir powered by three recurrent neural network type deep/shallow learning algorithms-Gated Recurrent Unit (GRU), Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), Support Vector Machine (SVM), K-nearest neighbor (KNN) and Decision tree (DT)-customized to accurately forecast sequential sedimentary structure data. These models, optimized using Adam's optimizer algorithm, demonstrated impressive performance in predicting water saturation levels using conventional petrophysical data. Particularly, the GRU model stood out, achieving remarkable accuracy (an R-squared value of 0.9973) with minimal errors (RMSE of 0.0198) compared to LSTM, RNN, SVM, KNN and, DT algorithms, thus showcasing its proficiency in processing extensive datasets and effectively identifying patterns. By achieving unprecedented accuracy levels, this study not only enhances the understanding of sediment properties and fluid saturation dynamics but also offers practical implications for reservoir management and hydrocarbon exploration in complex geological settings. These insights pave the way for more reliable and efficient decision-making processes, thereby advancing the forefront of reservoir engineering and petroleum geoscience.

NMR-based metabolomic profiling of Peganum harmala L. reveals dynamic variations between different growth stages.

Xerophytes play an active role in preventing soil denudation and desertification in arid and semi-arid areas. Peganum harmala L. (Zygophyllaceae family), a seasonally growing, poisonous and drought-tolerant plant, is widely distributed in the Xinjiang Uygur Autonomous Region and used as a traditional herbal medicine as well as, in winter, a fodder source. Previous research has focused on the pharmacological activity of isolated compounds and stress responses to growth environments. However, the metabolic profile of P. harmala and variations in its metabolites, including medicinally active and stress resistance components, have not been illustrated during different growth stages. Here, we collected plant samples in May, August, October and December. We determined the metabolic composition of methanol extracts using NMR spectroscopy, and comparisons of four growth stages were accomplished by applying statistical analysis. The results showed that vasicine, choline and sucrose were significantly elevated in samples harvested in May. Significantly higher amounts of betaine, lysine, 4-hydroxyisoleucine and proline were found in samples collected in August than in samples collected in other months, and the concentrations of phosphorylcholine, glucose, acetic acid and vasicinone were highest in December. The relationships between differential biomarkers and plant physiological states affected by diverse growth environmental factors were discussed. Our result deepened the understanding of metabolic mechanisms in plant development and confirmed the advantage of using NMR-based metabolomic treatments in quality evaluation when P. harmala is used for different purposes.

[Micro-community characteristics of vegetations in blowouts and depositional areas of Hulunbuir grassland, Inner Mongolia].

By using traditional sampling methods, the micro-communities of vegetations in fixed, semi-bare, and bare blowouts of Hulunbuir grassland were investigated, and the investigation data were statistical analyzed. The results showed that the vegetation coverage decreased in the order of fixed blowout, semi-bare blowout, and bare blowout, and was lower than that of the primary vegetation Form. Stipa grandis. Potentilla acaulis and Kengia squarrosa were the dominant species in fixed blowout, with the coverage being 5%; while P. acaulis and Carex sp. were the dominant species in semi-bare blowout, with the coverage being 2%. The dominant species in depositional areas of semi-bare blowout were P. acaulis, K. squarrosa, Agropyron cristatum, and Thymus mongolicus, and the coverage was 4%. The dominant species on the southwest slope of bare blowout was Agriophyllum pungens. The middle depositional area of bare blowout was also occupied by A. pungens (coverage 4.7%), and the edge of it was dominated by A. cristatum (coverage 2.7%), Carex sp. (coverage 2.6%), and T. mongolicus (coverage 1.7%) from the edge of the depositional area to primary grassland. The mean species importance value in fixed, semi-bare, and bare blowouts was 12.64%, 13.38%, and 20.08%, while that in the depositional area of semi-bare blowout and in the middle and edge of bare blowout was 12.55%, 40.18%, and 11.15%, respectively.

[Physico-chemical characteristics of ambient particles settling upon leaf surface of six conifers in Beijing].

The study on the density of ambient particles settling upon the leaf surface of six conifers in Beijing, the micro-configurations of the leaf surface, and the mineral and element compositions of the particles showed that at the same sites and for the same tree species, the density of the particles settling upon leaf surface increased with increasing ambient pollution, but for various tree species, it differed significantly, with the sequence of Sabina chinensis and Platycladus orientalis > Cedrus deodara and Pinus bungeana > P. tabulaeformis and Picea koraiensis. Due to the effects of road dust, low height leaf had a larger density of particles. The density of the particles was smaller in summer than in winter because of the rainfall and new leaf growth. The larger the roughness of leaf surface, the larger density of the particles was. In the particles, the overall content of SiO2, CaCO3, CaMg(CO3,), NaCl, 2CaSO4 . H2O, CaSO4 . 2H2O and Fe2O3 was about 10%-30%, and the main minerals were montmorillonite, illite, kaolinite and feldspar. The total content of 21 test elements in the particles reached 16%-37%, among which, Ca, Al, Fe, Mg, K, Na and S occupied 97% or more, while the others were very few and less affected by sampling sites and tree species.

Physicochemical characteristics of ambient particles settling upon leaf surfaces of urban plants in Beijing.

Particulate pollution is a serious health problem throughout the world, exacerbating a wide range of respiratory and vascular illnesses in urban areas. Urban plants play an important role in reducing particulate pollution. Physicochemical characteristics of ambient particles settling upon leaf surfaces of eleven roadside plants at four sites of Beijing were studies. Results showed that density of particles on the leaf surfaces greatly varied with plant species and traffic condition. Fraxinus chinensis, Sophora japonica, A ilanthus altissima, Syringa oblata and Prunus persica had larger densities of particles among the tall species. Due to resuspension of road dust, the densities of particles of Euonymus japonicus and Parthenocissus quinquefolia with low sampling height were 2-35 times to other taller tree species. For test plant species, micro-roughness of leaf surfaces and density of particles showed a close correlation. In general, the larger micro-roughness of leaf surfaces is, the larger density of particles is. Particles settling upon leaf surfaces were dominantly PM, (particulate matter less than 10 microm in aerodynamic diameter; 98.4%) and PM25 (particulate matter less than 2.5 microm in aerodynamic diameter; 64.2%) which were closely relative to human health. Constant elements of particles were C, O, K, Ca, Si, Al, Mg, Na, Fe, S, Cl and minerals with higher content were SiO2, CaCO3, CaMg(CO3)2, NaCI and 2CaSO4 x H20, SiO2. CaCO3 and CaMg(CO3)2 mainly came from resuspension of road dust. 2CaSO4 x H20 was produced by the reaction between CaCO3 derived from earth dust or industrial emission and SO2, H2SO4 or sulfate. NaCl was derived from sea salt.