Explore the vast range of titles available.

Porous materials, such as carbonate rocks, frequently have pore sizes which span many orders of magnitude. This is a challenge for models that rely on an image of the pore space, since much of the pore space may be unresolved. In this work, sub‐resolution porosity in X‐ray images is characterized using differential imaging which quantifies the difference between a dry scan and 30 wt% potassium iodide brine saturated images. Once characterized, we develop a robust workflow to incorporate the sub‐resolution pore space into a network model using Darcy‐type elements called microlinks. Each grain voxel with sub‐resolution porosity is assigned to the two nearest resolved pores using an automatic dilation algorithm. By including these microlinks with empirical models in flow modeling, we simulate single‐phase and multiphase flow. By fine‐tuning the microlink empirical models, we match permeability, formation factor (the ratio of the resistivity of a rock filled with brine to the resistivity of that brine), and drainage capillary pressure to experimental results. We then show that our model can successfully predict steady‐state relative permeability measurements on a water‐wet Estaillades carbonate sample within the uncertainty of the experiments and modeling. Our approach of incorporating sub‐resolution porosity in two‐phase flow modeling using image‐based multiscale pore network techniques can capture complex pore structures and accurately predict flow behavior in porous materials with a wide range of pore size. Key Points A dilation‐based algorithm adds sub‐resolution porosity, quantified by differential imaging, as microlinks to an extracted pore network Empirical models are used to characterize the flow properties of microlinks The tuned multiscale network, using permeability, formation factor, and capillary pressure, matches experimental relative permeability

We present a new pore‐scale model for multicomponent advective‐diffusive transport with coupled mineral dissolution and precipitation. Both dissolution and precipitation are captured simultaneously by introducing a phase transformation vector field representing the direction and magnitude of the overall phase change. An effective viscosity model is adopted in simulating fluid flow during mineral dissolution‐precipitation that can accurately capture the velocity field without introducing any empirical parameters. The proposed approach is validated against analytical solutions and interface tracking simulations in simplified structures. After validation, the proposed approach is employed in modeling realistic rocks where mineral dissolution and precipitation are dominant at different locations. We have identified three regimes for mineral dissolution‐precipitation coupling: (a) compact dissolution‐precipitation where dissolution is dominant near the inlet and precipitation is dominant near the outlet, (b) wormhole dissolution with clustered precipitation where dissolution generates wormholes in the main flow paths and precipitation clogs the secondary flow paths, and (c) dissolution dominant where all solid grains are gradually dissolved. In the three regimes, the proposed approach provides reliable porosity‐permeability relationships that cannot be described well by traditional macroscale models. We find that the permeability can increase while the overall porosity decreases when the main flow paths are expanded by dissolution and adjacent pore spaces are clogged by precipitation. Key Points A new pore‐scale model is developed for multicomponent reactive flow with coupled mineral dissolution and precipitation The proposed model is validated against analytical solutions and reference simulations, and then applied to realistic rocks Three dissolution‐precipitation coupling regimes with different flow patterns and porosity‐permeability relationships have been identified

Due to the difficulties in estimating groundwater recharge and cross‐boundary nature of many aquifers, estimating groundwater recharge at large scale has been called upon. Process‐based models as well as data‐driven models have been established to meet this need. Meanwhile, with the advent of explainable artificial intelligence (XAI) methods, data‐driven machine learning models can take advantage of enhanced explainability while keeping the strength of high flexibility. In this study, an ensemble neural network model was built to check the suitability of the model to predict groundwater recharge and the possibility to gain new insights from large data set. Recent large inputs of groundwater recharge data and additional input for the Arabian Peninsula collated in this study were fed to the model with multiple predictors related to climatology considering seasonality, soil and plant characteristics, topography, and hydrogeology. The model showed higher performance (adjusted R2: 0.702, RMSE: 193.35 mm yr−1) than a recent global process‐based model in predicting groundwater recharge. Using XAI methods as individual conditional expectations and Shapley Additive Explanation interaction values, the model behavior was analyzed and possible linear and non‐linear relationships between the predictors and the groundwater recharge rate were found. Long‐term averaged precipitation and enhanced vegetation index showed non‐linear relationships with groundwater recharge rate, while slope, compound topographic index, and water table depth showed low importance to the model results. Most model behaviors followed the domain knowledge, while multi‐correlation between predictors and data skewness hindered the model from learning. Plain Language Summary Estimating groundwater recharge rates at a large scale has been an important task among hydrologists. Both process‐based models and data‐driven models have been used for this purpose. Despite their high flexibility and high performance, there has been criticism over data‐driven models, especially machine‐learning models, that the result of the models are difficult to explain. However, new analysis tools called explainable artificial intelligence (XAI) can help explain the model results. In this study, a machine‐learning model (ensemble neural network model) has been built at global scale to check if the model can estimate groundwater recharge rates and to check if the model's behavior explained by XAI can give new insights into the processes. Our model shows higher performance compared to a recent global process‐based model. XAI tools are used to explain how the model predicted the groundwater recharge rates. Long‐term averaged precipitation and enhanced vegetation index show high sensitivity and high importance in predicting groundwater recharge rates, while topographical factors related to slope, curvature, and depth to the groundwater aquifer show low sensitivity and importance. Key Points Estimating groundwater recharge rates at global scale using an ensemble neural network model with 5541 observations and 20 predictors XAI can quantify the sensitivity and importance of each predictor, showing non‐linearities with long‐term precipitation and vegetation index Predictions show higher accuracy than the current process‐based model, with most behaviors measured by XAI aligning with domain knowledge

Biofilms in porous media significantly impact solute transport, beyond their role in reducing permeability through bioclogging. Experimental evidence has shown that biofilms can induce anomalous transport behaviors such as increased dispersion and pronounced tailing. These effects arise from the structural heterogeneity of the biofilm and the development of internal convective pathways. Despite being mostly composed of water, biofilms exhibit reduced effective diffusivity due to their complex microstructure. To capture these effects, we develop an original pore‐scale transport model combining the micro‐continuum approach with Random Walk Particle Tracking. Our simulations show that biofilm permeability, effective diffusivity, and spatial heterogeneity strongly influence solute breakthrough times, highlighting the critical role of biofilm structure in shaping complex transport behavior in porous systems.

Particle transport in subsurface porous media under multiphase flow conditions is widely concerned in many practical applications. Previous studies have focused on retention behaviors and interfacial effects, ignoring the unique role of pronounced rheological effect under dilute conditions. Here, we investigate how accumulation effect reshapes microgel particle transport and immiscible displacement process driven by concentration‐sensitive viscosity. As a foundation, a mixture‐rheology two‐fluid model is developed and combined with color‐gradient lattice Boltzmann method for modeling complex particulate multiphase flow. The consistency between simulation results and microfluidic experiments confirms the validity of our model in capturing accumulation phenomena. Results in heterogeneous dual‐permeability structures reveal the two‐way coupling between particle accumulation and interfacial evolution. Particle accumulation can be enhanced at higher injection concentrations and larger particle sizes, leading to the formation of filter‐cake‐like structures despite the absence of clogging effects. Capillary resistance further weakens the driving force for particle migration, intensifying local accumulation compared to suspension flow. The non‐uniform concentration distribution contributes to flow rate reallocation via diversion effects, producing variable displacement patterns under varying conditions. Results in disordered media exhibit a similar trend as in the dual‐permeability model but with more significant accumulation. The dramatic reduction in nonaqueous phase saturation by sweeping efficiency improvement indicates the promising application potential of such accumulation. Our findings deepen the understandings of particle transport in porous media with implications for manipulation of immiscible displacement. Key Points An improved two‐fluid model based on multiphase lattice Boltzmann method is applied to model particle‐water‐nonaqueous phase flow Particle lagging driven by pore‐to‐throat variations under concentration‐sensitive viscosity reshapes particle transport in porous media Significant flow diversion can be achieved by accumulation of microgel particles during multiphase displacement

Evidence suggests that novel enzyme functions evolved from low‐level promiscuous activities in ancestral enzymes. Yet, the evolutionary dynamics and physiological mechanisms of how such side activities contribute to systems‐level adaptations are not well characterized. Furthermore, it remains untested whether knowledge of an organism's promiscuous reaction set, or underground metabolism, can aid in forecasting the genetic basis of metabolic adaptations. Here, we employ a computational model of underground metabolism and laboratory evolution experiments to examine the role of enzyme promiscuity in the acquisition and optimization of growth on predicted non‐native substrates in Escherichia coli K‐12 MG1655. After as few as approximately 20 generations, evolved populations repeatedly acquired the capacity to grow on five predicted non‐native substrates—D‐lyxose, D‐2‐deoxyribose, D‐arabinose, m‐tartrate, and monomethyl succinate. Altered promiscuous activities were shown to be directly involved in establishing high‐efficiency pathways. Structural mutations shifted enzyme substrate turnover rates toward the new substrate while retaining a preference for the primary substrate. Finally, genes underlying the phenotypic innovations were accurately predicted by genome‐scale model simulations of metabolism with enzyme promiscuity. Synopsis Computational modeling of underground metabolism, laboratory evolution and omics analyses reveal that enzyme promiscuity can play a major role during adaptation to new growth environments and indicate that the genes underlying the phenotypic innovations can be predicted. Enzyme promiscuity can confer a fitness benefit in novel growth environments and open routes for achieving innovative growth states. Mutation events which enable growth on non‐native carbon sources can be structural or regulatory in nature and single mutation events related to a promiscuous activity can be sufficient to support growth while some cases require multiple mutations. Metabolic network analysis and constraint‐based modeling can predict adaptation to non‐native carbon sources through promiscuous enzyme activities. Laboratory evolution can be used to select for enzymes with structural mutations enabling an improved substrate affinity for a non‐native carbon source. Graphical Abstract Computational modeling of underground metabolism, laboratory evolution and omics analyses reveal that enzyme promiscuity can play a major role during adaptation to new growth environments and indicate that the genes underlying the phenotypic innovations can be predicted.

This study compares two widely used approaches for modeling soil moisture (SM) infiltration in mesoscale hydrology: the one‐dimensional Richards equation (1D RE), which governs vertical flux exchange but is nonlinear and computationally demanding, and the infiltration capacity (IC) scheme, which is simpler and restricts SM movement to the downward direction. A major challenge in implementing the RE is the estimation of effective parameters at the typical model resolution (hundreds to thousands of meters), as the equation was originally developed for finer scales. To address this, we conducted experiments using the mHM model with Multiscale Parameter Regionalization (MPR) to parameterize both RE and IC approaches. The RE parameterization relied on three distinct pedo‐transfer functions (PTFs). Parameters were estimated across 201 basins in Germany and evaluated using streamflow data at multiple spatial resolutions, as well as in situ SM observations from 46 sites (0–25 cm) and 42 sites (25–60 cm and 0–60 cm). Results show that both mHM‐IC and all mHM‐RE variants perform comparably in streamflow prediction. The use of MPR enables the spatial transferability of PTF‐based parameters. Owing to its two‐way flux capability, the mHM‐RE variant better captures SM dynamics, particularly in deeper soil layers. Although the IC scheme often leads to saturation at depth, it still provides strong performance in capturing SM anomalies. Overall, the study demonstrates that with appropriate parameterization, the RE approach can yield transferable parameters and robust simulations of both streamflow and soil moisture states.

The significant risk associated with fault reactivation often necessitates slip tendency analyses for effective risk assessment. However, such analyses are challenging, particularly in large areas with limited or absent reliable stress measurements and where the cost of extensive geomechanical analyses or simulations is prohibitive. In this paper, we propose a novel approach using a physics‐informed neural network that integrates stress orientation and satellite displacement observations in a top‐down multi‐scale framework to estimate two‐dimensional slip tendency analyses even in regions lacking comprehensive stress data. Our study demonstrates that velocities derived from a continental scale analysis, combined with reliable stress orientation averages, can effectively guide models at smaller scales to generate qualitative slip tendency maps. By offering customizable data selection and stress resolution options, this method presents a robust solution to address data scarcity issues, as exemplified through a case study of the South Australian Eyre Peninsula. Plain Language Summary Fault reactivation poses significant risks, often requiring slip tendency analyses for thorough risk assessment. Yet, such analyses face challenges, especially in large areas lacking reliable stress measurements or where extensive geomechanical analyses are too costly. Our paper suggests a new method using a physics‐based neural network. This approach combines compressive direction and satellite displacement observations to estimate slip tendencies in two dimensions, even where stress data is lacking. Our study shows that by using displacements from a continental scale analysis and reliable averages of compressive directions, we can guide models to create smaller‐scale maps indicating where faults are more likely to reactivate. This method allows for customizable data selection and stress resolution, offering a strong solution to data scarcity issues. We demonstrate its effectiveness through a case study of South Australia's Eyre Peninsula. Key Points Physics‐based neural networks allow two‐dimensional slip tendency analyses without prior full‐stress information A multi‐scale approach provides required displacement constraints when inferring full stresses from global navigation satellite system (GNSS) and stress orientation data We present a new application for GNSS data that would welcome more stations, even in seismically stable areas

Mixing‐limited reactions in unsaturated porous media are controlled by complex pore‐scale processes arising from air and water phases coexistence. Decreasing water saturation increases flow heterogeneity, creating preferential flow paths and dead‐end regions (DER) that alter solute distribution and reaction efficiency. Transmitting pores (TP) enhance mixing via interface deformation driven by stretching and shrinking. Conversely, DER act as low‐velocity traps, contributing to mixing through diffusion and delayed reactant release. A unified understanding of their distinct roles in mixing interface evolution and upscaled reaction rates remains limited. Using high‐resolution multiphase flow simulations, we investigate how water saturation influences mixing interface evolution across Péclet numbers. We develop a two‐compartment model that separately accounts for interface deformation in TP and solute trapping in dead‐end regions. We show that, even under unsaturated conditions, the mixing interface deformation within TP eventually plateaus once a balance between stretching and diffusion is reached. In contrast, interface segments in DER are governed by the dynamic interplay between the generation of new trapped segments and the decay of existing ones. This controls the late‐time behavior of interface length, which continues to grow until it reaches saturation. Our framework reproduces the observed mixing dynamics and provides a simple expression linking reaction rate to the total mixing interface length. The results demonstrate that under low saturation, the prolonged elongation of the interface substantially enhances reaction rates, highlighting the critical role of saturation‐driven heterogeneity in reactive transport.

The pore‐scale interfacial dynamics including main‐meniscus flow and corner flow usually occurs in heterogeneous porous media and significantly affects the macroscopic multiphase flow process. Numerical research on the competition between main‐meniscus flow and corner flow remains challenging due to the ambiguity in pore‐scale interfacial dynamics and the complexity of upscaling these processes to porous media, given the substantial spatial and temporal scale differences. In this study, we proposed a critical capillary number (Cac$C{a}_{c}$ ) by considering the interplay of local capillary and viscous forces, which predicts transition from main‐meniscus flow‐dominated processes into corner flow‐dominated processes during the strong imbibition. The Cac$C{a}_{c}$criterion was integrated into a dynamic network model to translate pore‐scale interfacial dynamics into multiphase flow patterns in porous media. The forced imbibition in heterogenous porous media under different Ca were simulated and compared to microfluidic experimental data. The comparison indicates that the dynamic competition between main‐meniscus flow and corner flow has a vital impact on displacement behaviors predicted by pore‐scale modeling, and our dynamic network model accurately captures the interfacial dynamics observed in microfluidic experiments. Moreover, the impact of interfacial dynamics on macroscopic multiphase flow pattern and displacement efficiency in heterogeneous porous media were addressed during strong imbibition under various viscosity ratios and capillary numbers. The phase diagram manifests a monotonic effect of viscosity ratio on displacement efficiency at high Ca due to the dominance of viscous fingering. A non‐monotonic effect of viscosity ratio is revealed at low Ca, which is ascribed into competition between corner flow and main‐meniscus flow. This study highlights the gap in the existing models of interfacial dynamics at pore scale, and provide an effective upscaling approach to investigate the multiphase flow in porous media.