Arabian Journal of Geosciences最新文献

This study investigated the performance of machine learning models in predicting the FS and slip surface. The models considered include support vector machine (SVM), random forest (RF), and extreme gradient boosting (XGBoost) algorithms. The slope stability analysis data for training of machine learning algorithms were obtained through the limit equilibrium method. This includes various scenarios of dry and homogeneous slope cases, encompassing a range of slope geometries (height (H), slope ratio (v/h)), and soil shear strength parameters (soil unit weight (γ), cohesion (c), friction angle (ϕ)). According to the evaluation using Taylor’s chart metrics, including standard deviation, correlation determination (R²), and root-mean-square error (RMSE), the XGBoost algorithm demonstrated the best performance. Additionally, employing the SHapley Additive exPlanations (SHAP) methodology revealed the significance order of variables as v/h > H > c > ϕ > γ for the factor of safety (FS) and H > v/h > c > ϕ > γ for the slip surface.

The Adigrat Sandstone Formation, representing the siliciclastic assemblage of the Lower Mesozoic succession, underlies the Gohatsion Formation in the Blue Nile Basin. Despite its stratigraphic and geological significance, the impact of diagenesis on the porosity value of the Adigrat sandstone remains poorly understood. By analyzing a detailed field description of two stratigraphic logs, along with the associated thin sections and additional correlated well data, the environment of deposition has been interpreted. The petrographic analysis was carried out to 30 sandstone samples collected from five selected outcrops, and complemented by the two stratigraphic columns. The main diagenetic processes affecting the Adigrat sandstone porosity were compaction, cementation, mineral dissolution, replacement, authigenesis, and recrystallization. The framework grain and cement relationship suggests an early quartz cement precipitation, followed by partial or intense calcite and hematite development in some samples as the second cementation phase. Feldspar alteration to lath-shaped kaolinite clusters causes kaolinite to act as a pore-lining and pore-filling cement, thereby reducing porosity. Conversely, the fracture and dissolution of some samples enhanced the fluid storage capacity. The estimated existing optical porosity (EOP) varies between 1 and 8%, with a mean value of 5%, of which 70% of the samples possess catenary and cul-de-sac porosities. Based on petrographic analysis, the sandstone is mineralogically categorized as sub-mature to mature. These findings significantly contribute to understanding the diagenetic evolution of the Adigrat Sandstone Formation, providing valuable insights for reservoir characterization and exploration strategies in the Blue Nile Basin (BNB).

Drought is one of the most destructive environmental hazards posing negative economic and social consequences. The country of Iran, which is located in the dry and semi-arid belt, is involved in much damage caused by drought every year, which makes it necessary to investigate. In this study, an attempt was made to investigate the frequency (number of occurrences) of severe and extreme droughts in the future. We considered of monthly averaged precipitation of 10 climate models of the fifth report of the Intergovernmental Panel on Climate Change (IPCC) for the basic period (1976–2005) and future periods (2020–2049, 2070–2099) under two scenarios (RCP4.5, 8.5). Using a new method, the difference between the average monthly precipitation of the models in the base period with the observed data and the inverse of the difference of each model was divided by the sum of the inverse of all models in each month (WP). Next, the average monthly precipitation of each model in the future period and the corresponding scenario was divided into the base period of that model (PCF), and at the end, for each month, the amount of WP was multiplied by the PCF of each model and their sum was obtained ((Delta P)). The value (Delta P), which is a 12-month time series, is introduced to the lars-wg model as a scenario file, and this model builds precipitation data based on this file. In the following, using the 12-month SPI index, according to the SPI index classification (values between − 1.5 and − 2 as severe drought and greater than − 2 as extreme drought), the total number of events in which the 12-month SPI (during the examined period in each station) being placed in the severe and extreme category was calculated. The estimation of error indices, especially R_Sqr (0.95), on average shows the accuracy of the combined weighted method and the Lars-Wg model in simulating precipitation. Also, the result presented in box plots shows an increase in the frequency of severe and extreme droughts in most of the country’s stations. Except group 3 (Southwestern and Western regions of the country), where the frequency of severe drought has decreased, in other groups, especially group 4 (60% on average), there is an obvious increase. The frequency of extreme drought in areas with good rainfall in the western and northern half of the country (especially groups 3 and 4) has declined (86% on average), while extreme events has decreased in group 1 with low rainfall. Considering that these areas are the main agricultural poles in the country, increasing the frequency of extreme drought can create harmful economic, social, and environmental consequences.

The primary intention of this study is to explore the ability of convolutional neural networks (CNNs) for forest classification using Compact Polarimetric (CP) data. Due to the phenomenal performance of the CNNs, more and more studies have tended to apply CNN-based methods to classify polarimetric synthetic aperture radar (PolSAR) images. In this study, three strategies were applied for this purpose. The first strategy involved designing and applying a CNN-based network to the Full Polarimetry (FP) mode of RADARSAT-2 C band, the simulated CP modes, and the reconstructed Pseudo Quad (PQ) modes. The results of these different modes were then compared with each other. In the second strategy, we compared the outcomes obtained from the first strategy with those from the Wishart classifier and the support vector machine (SVM) used in previous studies. Finally, the last strategy combined the CP modes to improve the classification outcomes further. Results showed that the CNN network outperformed other methods by using the CP modes for forest classification, and combining π/4 and DCP_L modes provided higher overall accuracy.

The southeastern coast of Bangladesh is forming the backbone of Bangladesh’s Blue Economic Zone, where the shoreline types are constantly changing. This research examines the classification of Bangladesh’s southeast coast, changes in shoreline types, and shoreline dynamics from 1990 to 2020. Field investigations, data from Google Earth, satellite images, and statistical analysis have all been carried out, where the transition matrix, sedimentation-erosion, LRR, EPR, and NSM are all used to examine the shifting of coastlines. According to the findings, the overall length of the investigated coastline in 1990 was 295.64 km, with only 11.12 km of human-induced coastline, but the total length of the coastline in 2020 was 281.38 km, with 67.39 km of human-induced coastline. Natural coasts include bedrock, beach, estuary, mangrove, and muddy coastlines; human-induced coastlines include salt fields, constructions, revetment and seawall, ship breaking, and manmade forest coastline. Approximately 60% of the sandy, muddy, and bedrock coastline has been transformed into seawalls, salt fields, and construction shorelines between 1990 and 2020. The changing intensity of coastal length in the study area is highest during 2010–2020, with a value of 0.28%. The analysis shows that over the last 30 years, the study area has lost around 1.06 km² per year and gained 2.35 km² per year, for a total net increase of 38.74 km². Human activities are hastening the process of coastline change, making it critical to protect healthy coastal ecosystems.

This work explores a complex system consisting of two separate layers: an upper layer consisting of a self-reinforced medium and a bottom layer made up of a half-space, especially a dry sandy medium. The surroundings that Rayleigh waves travel through are these two layers. Analytical solutions for the self-reinforced layer and the dry sandy half-space have been methodically derived using the variables separation method approach. After that, dispersion relation of the system has been found within a predetermined range. The computational capacity of MATHEMATICA software has allowed for the quantitative illustration of some important Rayleigh wave features. The characteristics give a thorough knowledge of wave behaviour in such layered material and include phase velocity, group velocity, and wave attenuation. The results draw special attention to the important ramifications for a range of real-world applications, especially in the areas of military infrastructure and coastal marine constructions. The design and stability of foundations subjected to wave-induced forces in marine structures depend heavily on an understanding of Rayleigh wave behaviour. The knowledge gained can be applied to military settings to improve the resilience and lifespan of structures by designing them to withstand the impact of vibrations and waves. Thorough investigation and findings of this study deepen our understanding of wave mechanics in layered media and provide important insights for engineering applications where complicated geological features and wave interactions are crucial. Emphasizing the wider significance and application of the research findings, this study not only increases theoretical knowledge but also offers helpful guidance for the design and analysis of structures in difficult situations. The study ends with conclusions and an outlook on possible future research directions.

The deep mining industry commonly encounters significant challenges, including large displacements of the roof in the gob-side entry retaining, and large deformations and failures of the filling body, which seriously compromise the structural safety and reliability of mines. In this study, we focus on a specific engineering case of the 8318 working face from the Xinzhou coal mine, employing the methodology of roof cutting for pressure relief and goaf filling with gangue reinforcement. Through a comprehensive approach involving numerical analysis, theoretical derivation, and experimental validation, the key parameters of the gob-side entry retaining by roof cutting, and the cooperative load-carrying mechanism of gangue reinforcement in gob-side entry retaining by cutting roof were systematically investigated. The numerical simulations and underground mine pressure monitoring data demonstrated that the optimum roof cutting height is 10 m at an optimum roof cutting angle of 8°. The proposed cooperative load-carrying technology of gob-side entry retaining by roof cutting significantly reduced the peak vertical stresses in the centre of the roof and the filling body by 22.6% and 43.4%, respectively. Furthermore, individual pillar stress levels showed notable reductions, with the maximum working stress and average stress decreasing by 34.2% and 47.8%, respectively. The practical implementation of our study offers valuable guidance in the control of surrounding rock in deep mining, thereby contributing significantly to the advancement in the field of surrounding rock support control.

Environmental protection requirements and the need to increase the proportion of oil recovered by secondary methods have led to the rise in popularity of microbial enhanced oil recovery (MEOR) techniques. Usually, MEOR requires the use of indigenous strains of microorganisms residing in wells, as they are adapted to local conditions. However, for some wells and fields, such as the Boca de Jaruco field in Cuba, information about the oilfield microorganisms and their properties is extremely limited. One of the properties crucial for the successful implementation of MEOR in fields is the ability of indigenous strains to produce biosurfactants. The aim of the present study is to evaluate the ability of six bacterial isolates obtained from the Boca de Jaruco field in Cuba to produce biosurfactants. The isolates capable of utilizing oil as their sole carbon source were identified as Bacillus subtilis (strains CC21, CC23, CC31, and CC32), Bacillus licheniformis (strain CC33), and Aeromonas veronii (strain CC22). It was determined that all isolates can tolerate temperatures between 30 and 60 °C, salinity ranging from 0.5 to 10.0% NaCl, and pH levels between 6 and 9. Regarding their ability to produce biosurfactants, assessed using the drop collapse method, oil-spreading method, emulsification activity test, and surface tension measurement, the isolates ranked as follows: A. veronii CC22 > B. subtilis CC21 = B. subtilis CC31 > B. subtilis CC23 = B. subtilis CC32 > B. licheniformis CC33. The biosurfactants produced were stable in the presence of 1.7 to 20.0% NaCl, irrespective of temperature (30 or 70 °C). However, substituting 20% of the NaCl with CaCl₂ resulted in destabilization of the biosurfactants produced by all investigated isolates, with destabilization levels averaging up to 32% at 70 °C.

The reservoir fluid PVT model is necessary to all types of hydrodynamic modelling (field development, well flow, well test, basin modelling, etc.). The PVT model, when not properly tuned, can result in significant inaccuracies in calculating PVT properties and field production of volatile oil and gas-condensate systems. The process of tuning the reservoir fluid PVT model is a complex and time-consuming task. Various methods, such as regression and machine learning (ML), have been employed for reservoir oil PVT model tuning; however, a definitive approach has not yet been identified. This paper introduces a novel and efficient step-by-step approach for developing and tuning reservoir fluid PVT which enables engineers to tune PVT models much faster than before. The new proposed approach can assist in the initialisation of a PVT model by employing effective methods for initial data pre-processing. Furthermore, it can accurately reproduce the results obtained from field measurements and basic laboratory studies conducted on representative samples, in a model using a cubic equation of state. Tuning the PVT model enables the reliable modelling of the PVT properties of all five types of reservoir fluids (black oil, volatile oil, gas condensate, wet gas, dry gas) in various applications; the applications include the design and monitoring of field development, multiphase flow calculations in wells and field pipelines, and basin modelling. It is possible to algorithmise and automate the application of this approach in specialised software. This study considered eight Russian reservoir oil and gas-condensate systems, for which the PVT models were tuned, using the proposed approach. The comparison between proposed approach and other tuning methods in modern PVT simulators (PVTi, PVTsim, Multiflash, PVT Designer) is shown in the article. These examples show the effectiveness of the proposed approach.

The Koldam site is located in the Himachal Lesser Himalaya in the vicinity of the main boundary thrust. The area near Koldam is seismically active, with earthquakes ranging in size from mild to major. Earthquakes result in the death and destruction of people and property. It is vital to research the features and nature of earthquake sources. This paper examines an analysis of 45 local earthquakes that were observed in the Himachal Himalaya region during the period from June 2014 to May 2019. The seismic moment, source radius, and stress drop are among the source parameters that are computed for earthquakes. These source parameters are computed using the hypocenter parameters, corner frequency (({f}_{text{c}})), and low-frequency spectral level (({Omega }_{text{o}})) with the help of Seisan software following Brune’s source model which describes the displacement amplitude spectrum as the physical process that releases energy at the source. This study is used to monitor and interpret the characteristics of the regional seismicity. The seismic moment ranges between 7.94 × ({10}^{10}) and 1.25 × ({10}^{15}) N-m with magnitudes between 1.6 and 4. The source radius is found to vary from 122.3 to 427.2 m. The stress drops of most of the events vary from 0.45 to 74.92 bar except for two events which have stress drops of 130.56 bar and 175.06 bar, respectively. Stress drops of 19 events range from 0.45 to 10.83 bar with M_o between 1 × 10¹² and 8 × 10¹² N-m, stress drops of 18 events range from 2.5 to 39.41 bar with M_o between 1 × 10¹³ and 8 × 10¹³ N-m, and stress drops of 6 events range from 18.08 to 74.92 bar with M_o between 1 × 10¹⁴ and 4 × 10¹⁴ N-m, respectively. Stress drops exhibit an increasing tendency up to a focal depth of 20 km, beyond which they show a decreasing pattern that appears to be associated with the strength of the crust. It appears that below 20 km, the strength of the upper crust decreases based on the variation of the maximum stress drop with focal depth. A scaling relationship has been established between the source parameters and the magnitudes for the region. A scaling law ({M}_{text{o}}) = 3.85 × ({10}^{15}{f}_{text{c}}^{-3.068}) has been developed between the seismic moment and corner frequency, for the region. This law almost agrees with that of the Kameng region (({M}_{text{o}}) = 2.0 × ({10}^{15}{f}_{text{c}}^{-3.34})) of the Arunachal Lesser Himalaya, the Bilaspur region (({M}_{text{o}}) = 2.0 × ({10}^{15}{f}_{text{c}}^{-3.03})) of the Himachal Lesser Himalaya, and the Garhwal Himalaya (({M}_{text{o}}) = 3.0 × ({10}^{16}{f}_{text{c}}^{-3.0})).