Applied Computing and Geosciences最新文献

{"title":"Explainable artificial intelligence using convolutional neural networks for sedimentary structure classification","authors":"Ammar J. Abdlmutalib,&nbsp;Korhan Ayranci,&nbsp;Umair Bin Waheed,&nbsp;James A. MacEachern","doi":"10.1016/j.acags.2026.100397","DOIUrl":"10.1016/j.acags.2026.100397","url":null,"abstract":"<div><div>Convolutional Neural Networks (CNNs) offer powerful image classification capabilities in geology, where visual interpretation of sedimentary structures is essential for reconstructing depositional environments. However, their “black-box” nature can limit trust and slow adoption in geoscientific workflows. This study presents a CNN-based framework for classifying eleven physical and biogenic sedimentary structures from core images, while treating explainability as a core objective. Three architectures (EfficientNet-B2, MobileNet-V3, and ResNet-50) were trained and evaluated on a geologically diverse annotated dataset. EfficientNet-B2 achieved the highest overall accuracy (97.7%), followed by MobileNet-V3 (96.8%) and ResNet-50 (95.8%). Blind-test evaluation on 401 unseen images showed that the CNN models retained substantial predictive capability under domain-shift conditions, with accuracies of 79.05% for EfficientNet-B2, 78.8% for MobileNet-V3, and 78.3% for ResNet-50. These results indicate good overall generalization, while the remaining errors were mainly associated with geologically similar sedimentary structures. Explainability was assessed using Grad-CAM++ and, crucially, was evaluated quantitatively using (i) deletion-based faithfulness, measuring the drop in target-class probability after masking the most salient regions, and (ii) stability under controlled perturbations (brightness changes, noise, blur, and small spatial shifts), measuring similarity between explanation maps. Across models, Grad-CAM++ consistently emphasized geologically meaningful features (e.g., pebble boundaries, inclined laminae, and biogenic textures) while de-emphasizing non-diagnostic artifacts. ResNet-50 produced the most faithful and stable explanations, showing the steepest probability decay under deletion and the highest robustness of saliency patterns under perturbations, despite slightly lower classification accuracy. Overall, these results demonstrate that high classification accuracy does not necessarily imply high explainability quality, and that explanation reliability metrics provide a practical route to more transparent and trustworthy sedimentary structure classification for real-world core analysis.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"31 ","pages":"Article 100397"},"PeriodicalIF":4.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148854238","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MorphoVolc: an open-source workflow for DEM-based morphometry and volume estimation of volcanic edifices","authors":"Marco Solinas,&nbsp;Camilla Gentili,&nbsp;Massimo Musacchio,&nbsp;Malvina Silvestri,&nbsp;Maria Fabrizia Buongiorno,&nbsp;Sergio Falcone","doi":"10.1016/j.acags.2026.100399","DOIUrl":"10.1016/j.acags.2026.100399","url":null,"abstract":"<div><div>MorphoVolc is an open-source workflow for extracting morphometric descriptors and estimating the volume of volcanic edifices from Digital Elevation Models (DEMs). The study addresses a practical limitation of many DEM-based morphometric analyses, which often rely on study-specific scripts, semi-manual procedures, and operator-dependent choices that reduce repeatability and comparability across case studies. MorphoVolc integrates DEM upload, terrain-derivative generation, boundary extraction, geometric model selection, volume estimation, and structured export within a browser-accessible and reproducible computational environment.</div><div>The workflow generates terrain derivatives, extracts base and caldera contours, computes first-order geometric and volumetric descriptors, and preserves run-based outputs as reusable artifacts. It supports interactive inspection as well as repeated and headless execution through a containerized release. The approach is demonstrated on two Japanese volcanic islands, Miyakejima and Izu-Oshima, selected as contrasting case studies because of their different overall planform geometries. In both cases, the workflow successfully produced derivative layers, boundary products, quantitative descriptors, and volumetric estimates, while runtime and memory usage remained compatible with medium-resolution DEM analysis under the tested conditions.</div><div>MorphoVolc does not propose fundamentally new geomorphometric algorithms; its main contribution lies in integrating established analytical steps into a transparent, operational, and reusable workflow. Within its current scope, it provides a practical framework for DEM-based volcanic morphometry, supporting exploratory analysis, cross-case comparison, and structured reuse of results in geoscientific workflows.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"31 ","pages":"Article 100399"},"PeriodicalIF":4.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148854239","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mobile and benchtop MIR spectrometer comparison for soil salinity indicators prediction in arid and semi-arid areas","authors":"Mohamed Ghassan,&nbsp;Reda Mokere,&nbsp;Yassine Ait-Tasskourit,&nbsp;Mbarek El-Guirah,&nbsp;Abdelaziz Nilahyane,&nbsp;Issam Barra","doi":"10.1016/j.acags.2026.100398","DOIUrl":"10.1016/j.acags.2026.100398","url":null,"abstract":"<div><div>Soil salinity and alkalinity pose critical challenges to soil health, especially in arid and semi-arid regions, leading to reduced crop productivity and ecosystem degradation. Soil salinity, resulting from excessive soluble salt accumulation, is exacerbated by factors such as high temperatures, low rainfall, poor irrigation water quality, and unrestricted use of soluble fertilizers. Controlling the parameters that determine soil quality in terms of salinity and alkalinity is essential. Soil spectroscopy modeling may also offer a rapid and effective alternative for assessing these parameters, providing reliable estimates with an acceptable level of error. This study compares the predictive performance of a mobile mid-infrared (MIR) spectrometer (Alpha II) and a benchtop instrument (Tensor II) in assessing soil salinity through key indicators: pH, electrical conductivity (EC), sodium adsorption ratio (SAR), and exchangeable sodium percentage (ESP) in southern Morocco. A total of 492 soil samples were used, split into calibration (80 %) and test (20 %) sets using the Kennard-Stone algorithm, with 5-fold cross-validation applied on the calibration set to optimize model hyperparameters. Three predictive algorithms including partial least squares regression (PLSR), random forest (RF), and memory-based learning (MBL) were applied alongside various spectral preprocessing techniques to optimize prediction accuracy. Model performance was evaluated using R², RMSE, and RPD. The best results showed high accuracy for EC, SAR, and ESP, with R² values generally exceeding 0.80 (RPD &gt; 2.0), and acceptable prediction performance for pH, with R² values not exceeding 0.65 (RPD 1.4 - 1.7). MBL and PLSR consistently delivered competitive and strong predictions for all properties across both spectrometers. MBL outperformed other tested models, particularly for EC, SAR, and ESP using the benchtop instrument (R² = 0.87, 0.85, and 0.84; RPIQ = 2.90, 2.17, and 2.66, respectively) and for pH using the mobile spectrometer (R² = 0.64, RMSE = 0.25, RPIQ = 2.82). Meanwhile, the performance of RF improved significantly by using preprocessing techniques. These findings highlight the promising potential of the mobile MIR spectrometer for rapid salinity assessment, with performance comparable to the benchtop instrument under laboratory conditions, though field validation is needed to confirm its practical viability.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"31 ","pages":"Article 100398"},"PeriodicalIF":4.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148854237","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An end-to-end deep learning approach for epicentral distance and magnitude determination from single station waveforms","authors":"Pincong Zhang,&nbsp;Xiong Zhang,&nbsp;Xiao Tian","doi":"10.1016/j.acags.2026.100356","DOIUrl":"10.1016/j.acags.2026.100356","url":null,"abstract":"<div><div>Conventional single-station earthquake parameter estimation relies heavily on precise P- and S-wave arrival picking, a dependency that often compromises speed and robustness in real-time scenarios. To overcome these limitations, we propose a Deep Fully Convolutional Encoder-Decoder Network designed for the simultaneous estimation of magnitude and epicentral distance directly from three-component waveform time series. Addressing the conflict between neural network input normalization and the loss of amplitude information required for magnitude scaling, we introduce a pseudo-normalization strategy. This approach normalizes the input waveform to facilitate training while preserving the normalization factor, allowing the network to decouple waveform shape features from amplitude scaling to achieve high-precision magnitude recovery. The normalized waveforms are utilized as neural network's input to predict the epicentral distance and magnitude contribution excluding the absolute amplitude. The architecture employs stacked convolutional and pooling layers to automatically extract waveform features, culminating in a probabilistic output layer that models predictions as one-dimensional Gaussian distributions, thereby quantifying estimation uncertainty. Trained on the comprehensive INSTANCE dataset from Italy, the model utilizes a multi-time-window strategy to simulate real-time data streaming. This enables the system to generate immediate estimates upon station triggering and progressively refine predictions as the waveform evolves. Evaluation on the test set yields a Mean Absolute Error (MAE) of 0.1977 for magnitude and 1.67 km for epicentral distance. These results demonstrate that our end-to-end framework achieves rapid, reliable estimation without the need for auxiliary phase picking, offering significant potential for enhancing Earthquake Early Warning (EEW) systems.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"30 ","pages":"Article 100356"},"PeriodicalIF":3.2,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148184077","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":"Anisotropic marble quality estimation using orientation-aware neural networks","authors":"Agni Patra ,&nbsp;Ioannis Kapageridis ,&nbsp;Charalampos Albanopoulos ,&nbsp;Francis Pavloudakis ,&nbsp;Argyro Asvesta","doi":"10.1016/j.acags.2026.100358","DOIUrl":"10.1016/j.acags.2026.100358","url":null,"abstract":"<div><div>Reliable estimation of marble quality is a key prerequisite for effective quarry design, production scheduling, and economic assessment. In ornamental stone deposits, quality attributes such as veining intensity, colour variation, and textural patterns exhibit pronounced directional dependence, meaning that the apparent quality of a block or sample is strongly influenced by cutting orientation. Drillhole data are acquired along multiple trajectories in three-dimensional space, and the resulting observations implicitly encode this directional dependence, which is commonly neglected in conventional spatial estimation workflows. To address this limitation, this study introduces an orientation-aware machine learning approach for the prediction of marble product proportions based on neural networks. The proposed framework integrates spatial coordinates with explicit orientation descriptors, represented through trigonometric encodings of azimuth and dip, alongside multi-scale Random Fourier Feature transformations that enable the network to capture complex, anisotropic spatial patterns. The model is trained on composite drillhole data and subsequently deployed to generate predictions across a three-dimensional block model. Permutation-based interpretability analysis indicates that orientation-related inputs contribute substantially to predictive performance, in many cases exceeding the importance of spatial coordinates alone. These results demonstrate that explicit incorporation of orientation information enables the model to learn anisotropic quality behaviour that would otherwise remain unresolved, leading to more accurate and operationally meaningful predictions. The methodology is illustrated through a case study from a marble quarry in northeastern Greece, highlighting its practical relevance for resource evaluation and production planning.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"30 ","pages":"Article 100358"},"PeriodicalIF":3.2,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148184078","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":"FEM-aligned weak-form physics-informed neural networks with Laplace–Beltrami geometry encoding for linear elastic surrogate modeling","authors":"Cheng Hsin Liu,&nbsp;Ahmad S. Abushaikha","doi":"10.1016/j.acags.2026.100354","DOIUrl":"10.1016/j.acags.2026.100354","url":null,"abstract":"&lt;div&gt;&lt;div&gt;Reliable prediction of subsurface deformation is central to geo-engineering applications such as geological CO&lt;span&gt;&lt;math&gt;&lt;msub&gt;&lt;mrow&gt;&lt;/mrow&gt;&lt;mrow&gt;&lt;mn&gt;2&lt;/mn&gt;&lt;/mrow&gt;&lt;/msub&gt;&lt;/math&gt;&lt;/span&gt; storage and underground hydrogen storage. We propose a finite-element–aligned physics-informed neural network (FEM–PINN) for small-strain linear elasticity. Unlike conventional strong-form PINNs that minimize PDE residuals at collocation points, the FEM–PINN minimizes the total potential energy — a variational (weak-form) objective — that mirrors the structure, numerical quadrature, and post-processing of classical finite element methods (FEM). Geometry is encoded using a compact Laplace–Beltrami eigenbasis computed on the analysis mesh, while training minimizes the total potential energy evaluated consistently at FEM Gauss points. Essential boundary conditions are enforced through smooth architectural gating complemented by light penalty regularization.&lt;/div&gt;&lt;div&gt;The framework outputs FEM-grade quantities, including nodal displacements, Gauss-point strains and stresses, von Mises stress, and total potential energy, enabling direct one-to-one comparison with a finite-element baseline. Three benchmarks are examined. A uniaxial compression test is used as a verification case, yielding relative &lt;span&gt;&lt;math&gt;&lt;msup&gt;&lt;mrow&gt;&lt;mi&gt;L&lt;/mi&gt;&lt;/mrow&gt;&lt;mrow&gt;&lt;mn&gt;2&lt;/mn&gt;&lt;/mrow&gt;&lt;/msup&gt;&lt;/math&gt;&lt;/span&gt; displacement errors of approximately 7% and energy gaps of 0.2–3.5% across all cases. A second, large-scale prismatic block with free lateral boundaries and a fixed base demonstrates robustness under boundary conditions representative of geomechanical compression scenarios, with displacement errors in the range of 6%–8% and near-unity parity with FEM across all displacement components. A third, quasi-two-dimensional domain with mixed directional boundary conditions is verified, yielding relative &lt;span&gt;&lt;math&gt;&lt;msup&gt;&lt;mrow&gt;&lt;mi&gt;L&lt;/mi&gt;&lt;/mrow&gt;&lt;mrow&gt;&lt;mn&gt;2&lt;/mn&gt;&lt;/mrow&gt;&lt;/msup&gt;&lt;/math&gt;&lt;/span&gt; displacement errors of approximately 8%.&lt;/div&gt;&lt;div&gt;Stress and strain fields are recovered consistently at Gauss points. A sensitivity study over the eigenmode budget &lt;span&gt;&lt;math&gt;&lt;mi&gt;m&lt;/mi&gt;&lt;/math&gt;&lt;/span&gt; confirms that LBO encoding is essential for variational consistency: the coordinate-only case (&lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;m&lt;/mi&gt;&lt;mo&gt;=&lt;/mo&gt;&lt;mn&gt;0&lt;/mn&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;) yields an energy gap of 137%, whereas &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;m&lt;/mi&gt;&lt;mo&gt;=&lt;/mo&gt;&lt;mn&gt;24&lt;/mn&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt; achieves 0.33%, the best among all configurations tested. Displacement errors vary non-monotonically with &lt;span&gt;&lt;math&gt;&lt;mi&gt;m&lt;/mi&gt;&lt;/math&gt;&lt;/span&gt;, consistent with the non-convex optimization landscape at a fixed training budget. By aligning weak-form training, geometry representation, and numerical integration with FEM practice, the proposed FEM–PINN provides a physically interpretable and practical surrogate for geomechanics workflows and offers a natural pathway toward hydro–mechanical coupling via augmented variational formulations.&lt;/div&gt;","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"30 ","pages":"Article 100354"},"PeriodicalIF":3.2,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148184138","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":"Convolutional neural networks for wildfire spread and intensity prediction","authors":"Maryam Moradpour,&nbsp;Pankaj Kumar,&nbsp;Gholam Ali Hoshyaripour","doi":"10.1016/j.acags.2026.100355","DOIUrl":"10.1016/j.acags.2026.100355","url":null,"abstract":"<div><div>Wildfires significantly impact ecosystems, human health, infrastructure, and the climate, making accurate prediction of fire behavior and its effects critical. Traditional physics-based models simulate fire-atmosphere interactions in detail but are computationally expensive and limited in real-time applications. In addition, uncertainties in input parameters and simplified combustion representations can reduce their reliability in forecasting wildfire-driven emissions and plume dynamics. On the other hand, empirical and statistical models are computationally efficient but often lack the ability to capture the nonlinear and coupled processes that drive wildfire spread.</div><div>This study presents a deep learning approach using a convolutional neural network (CNN) to predict wildfire dynamics under varying environmental conditions of wind, fuel, and atmospheric stability. The model is trained on a high resolution Weather Research and Forecasting (WRF) model coupled with the SFIRE (WRF-SFIRE) simulation dataset and predicts the temporal evolution of wildfire spread, represented through ground-level heat flux (GHF) fields as an indicator for fire intensity and progression. Model performance is evaluated using root mean square error (<span><math><mrow><mi>R</mi><mi>M</mi><mi>S</mi><mi>E</mi></mrow></math></span>) of 14.3 kW/m<span><math><msup><mrow></mrow><mrow><mn>2</mn></mrow></msup></math></span>, mean absolute error (<span><math><mrow><mi>M</mi><mi>A</mi><mi>E</mi></mrow></math></span>) of 6.6 kW/m<span><math><msup><mrow></mrow><mrow><mn>2</mn></mrow></msup></math></span>, correlation coefficient (<span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span>) of 84%, and the Structure–Amplitude–Location (SAL) method for spatial verification.</div><div>Results show that the CNN effectively reproduces the spatial and temporal evolution of wildfire dynamics, closely aligning with reference simulations across diverse conditions. By accurately capturing fire spread and intensity patterns at much lower computational cost, the proposed approach demonstrates the potential of deep learning to complement existing fire modeling frameworks and to support faster, scalable forecasting of wildfire behavior.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"30 ","pages":"Article 100355"},"PeriodicalIF":3.2,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148184139","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":"Data transformation for geostatistical simulation of grades of correlated metals in a complex deposit","authors":"Tshimangadzo Ndou ,&nbsp;Emmanuel John M. Carranza","doi":"10.1016/j.acags.2026.100359","DOIUrl":"10.1016/j.acags.2026.100359","url":null,"abstract":"<div><div>Geostatistical simulation methods are important in mineral resources estimation. However, applying such methods to simulate the grades of metals that are strongly correlated within a deposit remains challenging because it requires preserving the statistical properties and inter-variable dependencies. For this challenge, multivariate data transformations, such as minimum/maximum autocorrelation factors (MAF) and projection pursuit multivariate transform (PPMT), are widely applied in the field of geosciences to handle the global inter-variable correlations before independent simulation. Therefore, it is important to evaluate existing multivariate transformation methods in independent simulation framework to determine how these transformations behave statistically when used within independent simulation framework. This study evaluated the performance of PPMT and MAF within the turning bands simulation method (TBSIM) using Fe and Al₂O₃ data from the Carajás deposits in Brazil, where these variables exhibit strong negative Pearson correlation. The assessments focused on the ability of each transformation to reproduce the univariate statistical characteristics and evaluate inter-variable dependency after back-transformation. Performance was evaluated using summary statistics, histogram reproduction, global inter-variable correlation, direct variogram, spatial cross-correlation, and computational efficiency between Fe and Al₂O₃. However, due to the use of TBSIM, global inter-variable correlations and spatial cross-correlation were not preserved after simulation and back-transformation. MAF reproduced the statistical characteristics of Fe more consistently, particularly its mean, whereas PPMT showed better reproduction of Al₂O₃ but less reliable performance for Fe. Overall, the findings indicate that although both MAF and PPMT transformations can reproduce certain univariate properties, preserving multivariate dependency remains challenging within the TBSIM framework.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"30 ","pages":"Article 100359"},"PeriodicalIF":3.2,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148184142","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":"Quadrilateral mesh-based reactive transport modeling in non-orthogonal random fracture-matrix systems","authors":"Danyang Su ,&nbsp;Kerry T.B. MacQuarrie ,&nbsp;K. Ulrich Mayer","doi":"10.1016/j.acags.2026.100353","DOIUrl":"10.1016/j.acags.2026.100353","url":null,"abstract":"<div><div>Reactive transport models (RTM) are critical tools for understanding complex fluid-rock interactions in fractured rock systems. However, applying RTM in such systems remains challenging due to the wide disparity of scales spanning from millimeters to kilometers. While existing RTM effectively address fracture-matrix interactions at smaller scales, few can handle irregular fracture networks on larger scales while explicitly incorporating matrix processes. To bridge this gap, we introduce a novel discrete fracture-matrix (DFM) reactive transport model based on the MIN3P code, utilizing an anisotropic quadrilateral mesh. This approach enables coarser discretization along fractures (advection-dominated zones) and refined discretization perpendicular to fractures (diffusion-dominated zones), significantly improving computational efficiency without compromising accuracy. The model's capabilities are demonstrated through simulations of conservative tracer transport and dissolved oxygen migration in fractured crystalline rock at both intermediate (hundred-meter) and large (kilometer) scales. Comparative analyses with traditional triangular mesh methods reveal that the proposed approach achieves comparable or superior accuracy while drastically reducing computational demands. The model's ability to efficiently simulate large-scale fractured rock systems makes it a powerful tool for applications such as assessing geochemical stability in crystalline rock with extensive fracture networks.</div></div>","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"30 ","pages":"Article 100353"},"PeriodicalIF":3.2,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148184135","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":"Corrigendum to “Separation of P- and S-waves on shallow subsurface using transfer learning” [Appl. Comput. Geosci. (2025) 100307]","authors":"Jian Li,&nbsp;Jinchao Xing,&nbsp;Lujun Wei,&nbsp;Chuankun Li,&nbsp;Xiaoyan Li","doi":"10.1016/j.acags.2026.100349","DOIUrl":"10.1016/j.acags.2026.100349","url":null,"abstract":"","PeriodicalId":33804,"journal":{"name":"Applied Computing and Geosciences","volume":"30 ","pages":"Article 100349"},"PeriodicalIF":3.2,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148241580","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}