Applied Computing and Geosciences最新文献

{"title":"Fully differentiable Lagrangian convolutional neural network for physics-informed precipitation nowcasting","authors":"Peter Pavlík ,&nbsp;Martin Výboh ,&nbsp;Anna Bou Ezzeddine ,&nbsp;Viera Rozinajová","doi":"10.1016/j.acags.2025.100296","DOIUrl":"10.1016/j.acags.2025.100296","url":null,"abstract":"<div><div>This paper presents a convolutional neural network model for precipitation nowcasting that combines data-driven learning with physics-informed domain knowledge. We propose LUPIN, a Lagrangian Double U-Net for Physics-Informed Nowcasting, that draws from existing extrapolation-based nowcasting methods. It consists of a U-Net that dynamically produces mesoscale advection motion fields, a differentiable semi-Lagrangian extrapolation operator, and an advection-free U-Net capturing the growth and decay of precipitation over time. Using our approach, we successfully implement the Lagrangian convolutional neural network for precipitation nowcasting in a fully differentiable and GPU-accelerated manner. This allows for end-to-end training and inference, including the data-driven Lagrangian coordinate system transformation of the data at runtime. We evaluate the model and compare it with other related AI-based models both quantitatively and qualitatively in an extreme event case study. Based on our evaluation, LUPIN matches and even exceeds the performance of the chosen benchmarks, opening the door for other Lagrangian machine learning models.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"28 ","pages":"Article 100296"},"PeriodicalIF":3.2,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145222153","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing SAM-based digital rock image segmentation via edge-semantics fusion","authors":"Ziqiang Wang ,&nbsp;Zhiyu Hou ,&nbsp;Danping Cao","doi":"10.1016/j.acags.2025.100292","DOIUrl":"10.1016/j.acags.2025.100292","url":null,"abstract":"<div><div>The Segment Anything Model (SAM) demonstrates strong segmentation capabilities. However, its application to digital rock images faces challenges from subtle transitions between matrix minerals and pore structures, as well as inherent heterogeneity, which result in mis-segmentation and discontinuities that affect petrophysical characterization and numerical modeling of subsurface reservoir properties. To address these challenges, we introduce ESF-SAM (Edge-Semantics Fusion-SAM), a novel approach that enhances SAM's segmentation fidelity by integrating edge and semantic features. Specifically, in ESF-SAM, semantic features from SAM's image encoder are processed through an edge decoder enhanced by progressive dilated convolutions to extract detailed structural boundaries. The resulting edge and original semantic features are adaptively fused through a dual-attention mechanism, where spatial gating attention dynamically balances their contributions across locations, and channel attention recalibrates feature importance to enrich the representation. This spatial–channel attention framework enriches feature representations, enabling targeted fine-tuning within the SAM decoder and thereby preserving global segmentation capability while significantly improving local boundary delineation in two-phase segmentation tasks. Experimental results demonstrate that ESF-SAM improves segmentation detail, leading to more accurate derivation of key rock properties such as elastic modulus and pore geometry parameters, with results that more closely align with labeled data compared to the original SAM. Trained on only a small number of annotated sandstone images, ESF-SAM effectively adapts to the target domain and exhibits strong generalization when applied to carbonate rock images without additional fine-tuning. This work exemplifies how integrating priors into foundation models can substantially enhance their applicability to complex scientific imaging tasks.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"28 ","pages":"Article 100292"},"PeriodicalIF":3.2,"publicationDate":"2025-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145159762","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Analyzing key controlling factors of shale reservoir heterogeneity in \"thin\" stratigraphic settings: A deep learning-aided case study of the Wufeng-Longmaxi Formations, Fuyan Syncline, Northern Guizhou","authors":"Ye Tao,&nbsp;Zhidong Bao,&nbsp;Fukang Ma","doi":"10.1016/j.acags.2025.100293","DOIUrl":"10.1016/j.acags.2025.100293","url":null,"abstract":"<div><div>The Wufeng-Longmaxi Formation shales are key targets for shale gas exploration, but they are often studied as a single stratigraphic unit with limited analysis of internal differences. This study combines traditional geological methods with deep learning to compare the reservoir characteristics of the Wufeng Formation, the first member of the Longmaxi Formation (Long 1), and the second member of the Longmaxi Formation (Long 2), identifying the main controlling factors of differences. We found that: (1) The Wufeng Formation primarily develops siliceous shale lithofacies (S), mixed siliceous shale lithofacies (S-2), and clay siliceous shale lithofacies (S-3). Long 1 develops mixed siliceous shale lithofacies (S-2) and clay siliceous shale lithofacies (S-3), while Long 2 exhibits clay and siliceous mixed shale lithofacies (M-2) and siliceous clay shale lithofacies (CM-1). (2) The YOLO-v8 model demonstrates higher accuracy in shale pore type detection than the YOLO-v10 model, with a maximum mAP of 78.9 %. Using the YOLO-v8 model, it was found that S, S-2, and S-3 lithofacies are dominated by dissolution pores and organic pores with larger specific surface areas and porosities, while CM-1 and M-2 lithofacies are characterized by dissolution pores with smaller specific surface areas and porosities. (3) Based on evaluation indicators such as TOC content, BET surface area, porosity, brittleness index, and gas content, S and S-2 are classified as Class I lithofacies, S-3 as Class II lithofacies, and M-2 and CM-1 as Class III lithofacies. The main controlling factor for the heterogeneity of the shale reservoirs in the study area is lithofacies.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"28 ","pages":"Article 100293"},"PeriodicalIF":3.2,"publicationDate":"2025-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145159761","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using machine learning classifiers together with discrimination diagrams for validation of rock classification labels","authors":"Malte Mues ,&nbsp;Dennis Kraemer ,&nbsp;David M. Ernst Styn","doi":"10.1016/j.acags.2025.100288","DOIUrl":"10.1016/j.acags.2025.100288","url":null,"abstract":"<div><div>Rock classification based on chemical components is a common task in the geochemical domain. Literature recommends the Total Alkali and Silica (TAS) discrimination diagram for classifying igneous volcanic rocks by the sum of Na<span><math><msub><mrow></mrow><mrow><mn>2</mn></mrow></msub></math></span>O and K<span><math><msub><mrow></mrow><mrow><mn>2</mn></mrow></msub></math></span>O in relation to SiO<span><math><msub><mrow></mrow><mrow><mn>2</mn></mrow></msub></math></span> contents. This paper comparatively applies the TAS diagram and machine learning classification techniques to a collection of volcanic rocks from the <span>GEOROC</span> database. The study demonstrates a mismatch between the rock type labeled by experts in the database and rock types assigned by the TAS diagram. Despite this discrepancy, the experiments show that support vector machines are particularly promising for building decision systems for rock classification. Random forests, multi-layer perceptrons and K nearest neighbors are less suitable as rock classifiers in the study.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"28 ","pages":"Article 100288"},"PeriodicalIF":3.2,"publicationDate":"2025-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145120571","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automating fault detection in seismic data: integrating image processing with deep learning","authors":"Ahmad Ashtari","doi":"10.1016/j.acags.2025.100286","DOIUrl":"10.1016/j.acags.2025.100286","url":null,"abstract":"<div><div>Fault interpretation in seismic images is crucial for identifying fluid accommodation and flow migration pathways in the oil and gas industry. Several algorithms have been developed to calculate seismic attributes which help identify faults. Despite these advancements, challenges still remain in fault interpretation due to the complexity of fault networks, noise, and quality of seismic data. Hybrid seismic attributes extracted through artificial neural networks can enhance fault interpretation. In the case of neural network-based approaches used for geological feature extraction, picking precise samples for training neural networks is vital. In this study, an innovative method based on the Shi–Tomasi corner detection algorithm has been introduced to automatically pick fault samples on seismic data to be used as input to deep neural networks to predict faults. The method has been tested on two field seismic images that were acquired at different surveys. The field examples indicate that the trained neural networks could give a precise and clear estimation of faults with different azimuths. This proves the proposed sampling method can effectively provide a high-quality training data set for deep neural networks to automatically predict faults from seismic data.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"28 ","pages":"Article 100286"},"PeriodicalIF":3.2,"publicationDate":"2025-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145108300","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Neural network inversion of seismic wave velocities for vadose zone water content profile","authors":"Quentin Didier,&nbsp;Victor Sauvage,&nbsp;Léna Pellorce,&nbsp;Rémi Valois,&nbsp;Slimane Arhab,&nbsp;Arnaud Mesgouez","doi":"10.1016/j.acags.2025.100285","DOIUrl":"10.1016/j.acags.2025.100285","url":null,"abstract":"<div><div>Accurate estimation of water saturation in the vadose zone is crucial for hydrological and agricultural applications. However, traditional seismic inversion methods often struggle with non-linearity and sensitivity to noise, limiting their generalisation capabilities. To address this challenge, we propose a neural network-based inversion approach that allows the assessment of the vertical distribution of water saturation from compressional (<span><math><msub><mrow><mi>v</mi></mrow><mrow><msub><mrow></mrow><mrow><mi>P</mi></mrow></msub></mrow></msub></math></span>) and shear (<span><math><msub><mrow><mi>v</mi></mrow><mrow><msub><mrow></mrow><mrow><mi>S</mi></mrow></msub></mrow></msub></math></span>) wave velocities. Specifically, our architecture is based on a regression model and incorporates an autoencoder layer to improve robustness against noise. As a result, this enhances its ability to invert complete saturation profiles and increases its adaptability to complex hydrological conditions. Furthermore, the model demonstrates strong performance across varying water table depths, with low error metrics and high resilience to input noise with a RMSE of 3.34 × 10⁻² and a R<span><math><msup><mrow></mrow><mrow><mn>2</mn></mrow></msup></math></span> of 0.978 for 5% noise. Our current approach has been trained exclusively on noisy synthetic data. We plan to validate it in the near future against experimental field data we have recorded for an agricultural soil. Overall, this study establishes a foundation for future applications of deep learning in hydrogeophysical inversion and underscores the need for validation with real-world data.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"27 ","pages":"Article 100285"},"PeriodicalIF":3.2,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145044083","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"GPU-accelerated simulation of steady-state flow and particle transport in discrete fracture networks","authors":"Tingchang Yin ,&nbsp;Teng Man ,&nbsp;Pei Zhang ,&nbsp;Sergio Andres Galindo-Torres","doi":"10.1016/j.acags.2025.100284","DOIUrl":"10.1016/j.acags.2025.100284","url":null,"abstract":"<div><div>Fracture networks in the subsurface can serve as the primary pathway for fluid flow, allowing for solute transport. This process is critical to various real-world applications, including resource extraction and contaminant migration in fractured rocks. We develop an open-source code called <em>cuDFNsys</em> to simulate flow and transport in discrete fracture networks (DFNs). Our code uses the mixed hybrid finite element method to solve the hydraulic head and velocity fields in DFNs, and the particle tracking method to simulate the movement of solute plumes. The GPU parallelization accelerates the generation of DFNs, identification of intersections between fractures, determination of elementary matrices, and motion of particles. We use several benchmarks to verify the accuracy of flow and transport simulation in <em>cuDFNsys</em>. Dispersion in a DFN is used to demonstrate examples of particle tracking. Performance analyses demonstrate that our code is well-suited for Monte Carlo iterations of DFN simulations, enabling physicists and geoscientists to study critical phenomena and phase transitions in fracture networks using percolation theory.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"27 ","pages":"Article 100284"},"PeriodicalIF":3.2,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144988473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Simwave: A finite difference simulator for acoustic waves propagation","authors":"Hermes Senger ,&nbsp;Jaime Freire de Souza ,&nbsp;João Baptista Dias Moreira ,&nbsp;Keith Jared Roberts ,&nbsp;Roussian di Ramos Alves Gaioso ,&nbsp;Emílio Carlos Nelli Silva ,&nbsp;Edson Satoshi Gomi","doi":"10.1016/j.acags.2025.100283","DOIUrl":"10.1016/j.acags.2025.100283","url":null,"abstract":"<div><div>Simwave is an open-source software package for wave simulations in 2D or 3D domains. It solves the constant and variable density acoustic wave equation with the finite difference method and has support for domain truncation techniques, several boundary conditions, and the modelling of sources and receivers given a user defined acquisition geometry. The architecture of Simwave is designed for applications with geophysical exploration in mind. Its Python front-end enables straightforward integration with many existing Python scientific libraries for the composition of more complex workflows and applications (e.g., migration and inversion problems). Its back-end is implemented in C, enabling performance portability across a range of computing hardware and compilers including both CPUs and GPUs. Simwave also provides non-optimized versions of the algorithms, which can be used as benchmarks for high-performance computing systems, serving as a proxy application for actual production solvers used by the geophysical exploration industry for the identification of Oil and Gas reservoirs.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"27 ","pages":"Article 100283"},"PeriodicalIF":3.2,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145026991","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automatic seismic fault detection and surface construction","authors":"Xin Liu ,&nbsp;Xingyu Zhu ,&nbsp;Xupeng He ,&nbsp;Yuzhu Wang","doi":"10.1016/j.acags.2025.100287","DOIUrl":"10.1016/j.acags.2025.100287","url":null,"abstract":"<div><div>This paper proposes an effective approach for automatically building the fault model based on the 3D seismic images via two steps of automatic seismic fault detection and fault surface construction. Automatic seismic fault detection is performed to automatically classify the seismic image into two phases of fault and background using a slightly revised deeplabv3_resnet50 architecture with pretrained parameters provided by PyTorch. The output of the automatic seismic fault detection is a binary image contains fault and background, where one fault may be separated into different fault segments, or several faults are connected with each other which need further distinguish. To reassemble these detected fault segments and construct the fault surface model, four steps are implemented including:1) a morphological workflow is used to separate all connected faults into separated fault segments; 2) the moving least square (MLS) method is used to fit each fault segments as a smooth, one-voxel thickness surface; 3) the weighted principle component analysis (WPCA) method is applied to calculate the normal vector of each surface voxel to judge whether two or more adjacent segments should be combined in one fault surface; 4) MLS method is applied again to fit all surface segments from one fault as an unique fault surface. The final output of the proposed method provides a fault model with well-defined, cleanly separated, labeled fault surfaces that is competent for structure modelling.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"27 ","pages":"Article 100287"},"PeriodicalIF":3.2,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145044135","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Estimating aboveground biomass using environmental covariates and a machine-learning approach in the Lower Brazos River Basin, Texas, USA","authors":"Birhan Getachew Tikuye,&nbsp;Ram Lakhan Ray","doi":"10.1016/j.acags.2025.100289","DOIUrl":"10.1016/j.acags.2025.100289","url":null,"abstract":"<div><div>Forest ecosystems play a pivotal role in global carbon sequestration, serving as essential carbon sinks for climate change mitigation, while also providing a range of ecosystem services such as seed dispersal, pollination, pest control, and habitat provisioning. This study aimed to estimate aboveground biomass density (AGBD) using environmental covariates and a machine learning approach from the Global Ecosystem Dynamics Investigation Light Detection And Ranging (GEDI-LiDAR) in the Lower Brazos River Watershed, Texas, USA. Specifically, GEDI Level 4A data from the National Aeronautics and Space Administration (NASA) Oak Ridge National Laboratory Distributed Active Archive Center (ORNL DAAC) was integrated with Landsat-9 Operational Land Imagery (OLI) and Shuttle Radar Topographic Mission (SRTM) data to enhance predictive accuracy for AGBD. Spectral indices, including the Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI), were derived from Landsat 9 to support AGBD prediction. Three machine learning models, such as Multivariate Adaptive Regression Splines (MARS), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost), were deployed, with performance assessed using the coefficient of determination (R²), root mean square error (RMSE), and mean absolute error (MAE). Among the models, XGBoost achieved the highest predictive accuracy (R² = 0.43, RMSE = 31.03, MAE = 22.49). The modelling indicated that longitude, latitude, moisture stress indices (MSI), and digital elevation model (DEM) are among the critical predictors for AGBD. The mean AGBD across the watershed was estimated at 72.3 Mg ha^-1, corresponding to a total biomass of approximately 66.6 million tons. Evergreen forests showed the highest AGBD values at 110 Mg ha^-1, while cultivated lands averaged 33 Mg ha^-1. These findings highlight the effectiveness of integrating environmental covariates with machine learning to estimate AGBD from GEDI LiDAR across diverse ecosystems. This approach provides a robust tool for advancing carbon management and climate change mitigation efforts, while also supporting data-driven conservation planning in both forested and agricultural landscapes.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"27 ","pages":"Article 100289"},"PeriodicalIF":3.2,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145104692","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}