Search

We propose a physics‐informed convolutional decoder (PICD) framework for inverse modeling of heterogenous groundwater flow. PICD stands out as a direct inversion method, eliminating the need for repeated forward model simulations. The framework combines data‐driven and physics‐driven approaches by integrating monitoring data and domain knowledge into the inversion process. PICD utilizes a convolutional decoder to effectively approximate the spatial distribution of hydraulic heads, while Karhunen–Loève expansion (KLE) is employed to parameterize hydraulic conductivities. During the training process, the stochastic vector in KLE and the parameters of the convolutional decoder are adjusted simultaneously to minimize the data‐mismatch and the physical violation. The final optimized stochastic vectors correspond to the estimation of hydraulic conductivities, and the trained convolutional decoder can predict the evolution and distribution of hydraulic heads. Various scenarios of groundwater flow are examined and results demonstrate the framework's capability to accurately estimate heterogeneous hydraulic conductivities and to deliver satisfactory predictions of hydraulic heads, even with sparse measurements. Plain Language Summary: Inverse modeling refers to estimate the unknown model parameters with measurements of model responses. In groundwater flow problems, the information about subsurface formation parameters is very limited, so inverse modeling is required to inference the uncertain formation parameters with sparse measurements. Many conventional inversion methods necessitate repeated forward calculations to compare the predictions with measurements and evaluate the likelihood of different estimations, resulting in a substantial computational burden. In this work, we propose a novel physics‐informed convolutional decoder (PICD) framework, which, as a direct inversion method, can circumvent the need for multiple forward calculations during the inversion process. In addition to measurements, physical laws are leveraged to provide extra information for inversion, alleviating the dependence on data, and enforcing the predictions align with measurements as well as domain‐specific knowledge. Several groundwater flow problems are considered to validate the effectiveness of the proposed PICD framework, and satisfactory performance can be obtained. The proposed PICD framework emerges as a promising tool for efficient and informed groundwater flow inverse modeling. Key Points: A physics‐informed deep learning framework is proposed for inversion of groundwater flowInversion can be performed directly without iterative forward modelingSatisfactory inversion performance can be achieved even with sparse measurements [ABSTRACT FROM AUTHOR]

Abstract Geological fractures are mechanical breaks in subsurface rock volumes that provide important subsurface flow pathways. However, the presence of fractures can cause unwanted challenges, such as gas leakage through fractured caprocks, which must be addressed. In this study, the dynamics of enzymatically induced carbonate precipitation in rock fractures and their subsequent influence on CO2 leakage were investigated from a pore-scale perspective for the first time. This was achieved through real-time monitoring of the injection of the solution into a rock-microfluidic flow cell using optical and scanning electron microscopy. It was revealed that the main growth dynamics occur during the first three injection cycles, with growth continuing until the fracture aperture is fully closed in the 6th cycle. Based on the flow simulation, a significant reduction of up to 25% in the CO2 conductivity of the original fracture is expected even after the first treatment cycle. Future studies are suggested to explore different resolutions, testing conditions, and to conduct 3-dimensional investigations of the growth dynamics.

In this study, we investigate the combined effect of different rainfall‐runoff event types and antecedent soil moisture (ASM) on runoff processes in the headwater elementary discharge area of a small forested upland catchment. The study focuses on (i) the relationship between soil moisture thresholds and runoff generation; (ii) the combined effect of ASM and tree vicinity and (iii) the relationship between different rainfall‐runoff event types and different types of runoff (baseflow and stormflow). The results suggest that ASM has a strong impact on local runoff generation processes. Soil water content (35%–36%) threshold exceedance was related to stormflow runoff generation caused by the activation of quick preferential flow paths in the soil during storm events, especially in the upper and the deepest soil layers. At the same time, unexpected non‐linear increases in baseflow runoff ratios were documented during dry, precipitation‐free, periods and when the 31%–34% soil moisture threshold was exceeded, presumably due to the hydrological connection of farther slope areas during these conditions. Multiple stormflow periods, which exhibited the lowest runoff coefficient, were the most significant events in terms of water retention and soil water recharge due to increased vertical hydrological connectivity enabling more rapid transport to deeper soil layers. However, this rainfall type occurred least often over the study period. The important role of forest stands (individual trees) in creating spatial patterns of soil moisture and preferential infiltration paths to deeper soil layers was also confirmed. These results contribute towards a better conceptualisation of hydrological behaviour in elementary headwater discharge areas and highlight the potential dangers associated with expected increases in extreme weather events. [ABSTRACT FROM AUTHOR]

Leachate, which comes from waste landfills, contains high levels of organic and inorganic matter. Constructed Wetland technology is a suitable alternative for leachate treatment. This technology is easy to apply, relatively cost-effective, and can achieve optimal reduction results for treating and controlling leachate, thereby reducing its environmental impact. The objective of this study was to treat the BOD concentration and COD in leachate using a Constructed Wetland. Data collection was performed by conducting leachate experiments for Constructed Wetland treatment using Hippochaetes lymenalis plants based on variations in residence time and media height. There was a control group (R1) without any media or plants. The variation combination included R2 (containing 10 cm gravel and 10 cm fertile soil) and R3 (containing 5 cm gravel and 15 cm fertile soil). Residence times were set at 0, 7, 14, and 21 d. This research showed that the reactor with growth media consisting of 10 cm gravel and 10 cm fertile soil had 55% and 85% BOD and COD removal efficiencies, with final BOD concentrations and COD values of 273 mg/L and 1,321, respectively. The reactor with growth media consisting of 5 cm gravel and 10 cm fertile soil had 74% and 95% BOD and COD removal efficiency, with the final BOD concentration and COD at 159 mg/L and 432, respectively. [ABSTRACT FROM AUTHOR]

Abstract We propose a physics‐informed convolutional decoder (PICD) framework for inverse modeling of heterogenous groundwater flow. PICD stands out as a direct inversion method, eliminating the need for repeated forward model simulations. The framework combines data‐driven and physics‐driven approaches by integrating monitoring data and domain knowledge into the inversion process. PICD utilizes a convolutional decoder to effectively approximate the spatial distribution of hydraulic heads, while Karhunen–Loève expansion (KLE) is employed to parameterize hydraulic conductivities. During the training process, the stochastic vector in KLE and the parameters of the convolutional decoder are adjusted simultaneously to minimize the data‐mismatch and the physical violation. The final optimized stochastic vectors correspond to the estimation of hydraulic conductivities, and the trained convolutional decoder can predict the evolution and distribution of hydraulic heads. Various scenarios of groundwater flow are examined and results demonstrate the framework's capability to accurately estimate heterogeneous hydraulic conductivities and to deliver satisfactory predictions of hydraulic heads, even with sparse measurements.

Quantifying flow and transport from hillslopes is vital for understanding water quantity and quality in rivers, but remains obscure because of limited subsurface measurements. Using measured hydraulic conductivity K profiles and water balance over a single year to calibrate a transmissivity feedback model for a hillslope in the East River watershed (Colorado) proved unsatisfactory for predicting flow over the subsequent years. Well-constrained field-scale K were obtained by optimizing subsurface flux predictions over years having large differences in recharge, and by including estimates of interannual transfer of excess snowmelt recharge. Water and solute exports during high snowmelt recharge occur predominantly via shallow groundwater flow through weathered rock and soil because of their enlarged transmissivities under saturated conditions. Conversely, these shallow pathways are less active in snow drought years when the water table remains deeper within the weathering zone. Hillslope soil water monitoring showed that rainfall does not infiltrate deeply during summer and fall months, and revealed water losses consistent with model ET predictions. By combining water table-dependent fluxes with pore water chemistry in different zones, time-dependent rates of solute exports become predictable. As an example, calibrated K were combined with dissolved nitrogen concentrations in pore waters to show the snowmelt-dependence of reactive nitrogen exported from the hillslope, further supporting the recent finding that the weathering zone is the dominant source of reactive nitrogen at this site. Subsurface export predictions can now be obtained for wide ranges of recharge based on measurements of water table elevation and profiles of pore water chemistry.

Ground-penetrating radar (GPR) is a rapid and non-destructive geophysical technique widely employed to detect and quantify subsurface structures and characteristics. Its capability for time lapse (TL) detection provides essential insights into subsurface hydrological dynamics, including lateral flow and soil water distribution. However, during TL-GPR surveys, field conditions often create discrepancies in surface geometry, which introduces mismatches across sequential TL-GPR images. These discrepancies may generate spurious signal variations that impede the accurate interpretation of TL-GPR data when assessing subsurface hydrological processes. In responding to this issue, this study introduces a TL-GPR image alignment method by employing the dynamic time warping (DTW) algorithm. The purpose of the proposed method, namely TLIAM–DTW, is to correct for geometric mismatch in TL-GPR images collected from the identical survey line in the field. We validated the efficacy of the TLIAM–DTW method using both synthetic data from gprMax V3.0 simulations and actual field data collected from a hilly, forested area post-infiltration experiment. Analyses of the aligned TL-GPR images revealed that the TLIAM–DTW method effectively eliminates the influence of geometric mismatch while preserving the integrity of signal variations due to actual subsurface hydrological processes. Quantitative assessments of the proposed methods, measured by mean absolute error (MAE) and root mean square error (RMSE), showed significant improvements. After performing the TLIAM–DTW method, the MAE and RMSE between processed TL-GPR images and background images were reduced by 96% and 78%, respectively, in simple simulation scenarios; in more complex simulations, MAE declined by 27–31% and RMSE by 17–43%. Field data yielded reductions in MAE and RMSE of >82% and 69%, respectively. With these substantial improvements, the processed TL-GPR images successfully depict the spatial and temporal transitions associated with subsurface lateral flows, thereby enhancing the accuracy of monitoring subsurface hydrological processes under field conditions. [ABSTRACT FROM AUTHOR]

In the case of absence of reliable information on fluctuations of levels and hydraulic parameters of aquifers within a large catchment, the method of river hydrograph decomposition can be used to assess groundwater component and other constituents of river water recharge. The difficulties and shortcomings of this method, which relies on specific schematizations and assumptions, are analysed and discussed. We propose to take into account the results of calculations of lateral groundwater inflow to the river accomplished using the numerical method based on observations at representative balance sites in order to improve the accuracy of the method of decomposition of the river hydrograph. A proper schematization should also be justified for river catchment (or its part), which describes the interaction of surface water, shallow groundwater, and artesian (deep) interlayer waters, which provides rationale for defining or neglecting groundwater inflow to river during the flood periods. The most disputable issue remains the allocation of the phase of "reverse water inflow to river banks" during flood periods. The groundwater level rise at the beginning of the flood results from the infiltration recharge to the aquifer and from inflow of water from the river. The contribution of inflow to river bank to groundwater level rise is dependent from the distance from the river to the observation well. To establish hydraulic head gradient from the river towards the adjacent groundwater aquifer at flood peak, at least two observation wells are needed, one of which should be located 3-4 m from the river water edge on the elevated part of the bank which is not flooded, and the other should be located 50-70 m from the bank towards the watershed. It is proposed that the phase of maximum groundwater inflow to the river should be adjusted based on the dates of the beginning and ending of the groundwater level decrease during the recession of the river flow during the flood period. The combined application of the described above methods showed that the use of only hydrograph decomposition method underestimates the share of groundwater recharge to the river by 2-5%. Based on the analysis of the Pivdennyi Buh River hydrographs during the 1980-2020 period, general tendencies of changes in the shallow groundwater and deep subsurface recharge of the Pivdennyi Buh River in its upper basin (upstream of the city of Khmilnyk) have been established: deep subsurface recharge prevailed over the shallow groundwater component (33.4 compared to 25.0% of the total). At the same time, the total subsurface runoff constituted on average 58%. [ABSTRACT FROM AUTHOR]

Leachate, which comes from waste landfills, contains high levels of organic and inorganic matter. Constructed Wetland technology is a suitable alternative for leachate treatment. This technology is easy to apply, relatively cost-effective, and can achieve optimal reduction results for treating and controlling leachate, thereby reducing its environmental impact. The objective of this study was to treat the BOD concentration and COD in leachate using a Constructed Wetland. Data collection was performed by conducting leachate experiments for Constructed Wetland treatment using Hippochaetes lymenalis plants based on variations in residence time and media height. There was a control group (R1) without any media or plants. The variation combination included R2 (containing 10 cm gravel and 10 cm fertile soil) and R3 (containing 5 cm gravel and 15 cm fertile soil). Residence times were set at 0, 7, 14, and 21 d. This research showed that the reactor with growth media consisting of 10 cm gravel and 10 cm fertile soil had 55% and 85% BOD and COD removal efficiencies, with final BOD concentrations and COD values of 273 mg/L and 1,321, respectively. The reactor with growth media consisting of 5 cm gravel and 10 cm fertile soil had 74% and 95% BOD and COD removal efficiency, with the final BOD concentration and COD at 159 mg/L and 432, respectively.

Urban living captivates the younger generation due to its wealth of prospects and attractive lifestyles. However, it is important to acknowledge that many familiar and unforeseen challenges accompany urban living. This study discusses the most pressing issue of the collapse of urban roads due to the formation of cavities/sinkholes on the subsurface, with scientific reasons and advanced approaches to mapping such cavities in order to reduce urban risk. Sinkhole formations are common in many cities in India, and several such incidents are being reported. Naturally formed sinkholes in lateritic deposits and urban cavities are due to improper handling of the subsurface soil. Typical reasons for the formation of urban cavities are summarized here with scientific explanations based on the density of filling and subsurface water flow. Further, scientific techniques available for identifying such cavities in advanced geophysical methods and the usual procedure for closing these cavities are presented. Regular inspection to focus on the triggering cause and comprehension of the underlying causes and technically designed treatments based on site soils can help prevent cavities from collapsing and subsequent disasters. [ABSTRACT FROM AUTHOR]

Large-scale or high-resolution geologic models usually comprise a huge number of grid blocks, which can be computationally demanding and time-consuming to solve with numerical simulators. Therefore, it is advantageous to upscale geologic models (e.g., hydraulic conductivity) from fine-scale (high-resolution grids) to coarse-scale systems. Numerical upscaling methods have been proven to be effective and robust for coarsening geologic models, but their efficiency remains to be improved. In this work, a deep-learning-based method is proposed to upscale the fine-scale geologic models, which can assist to improve upscaling efficiency significantly. In the deep learning method, a deep convolutional neural network (CNN) is trained to approximate the relationship between the coarse block of fine-scale hydraulic conductivity fields and the corresponding hydraulic heads, which can then be utilized to replace the numerical solvers while solving the flow equations for each coarse block. In addition, physical laws (e.g., governing equations and periodic boundary conditions) can also be incorporated into the training process of the deep CNN model, which is termed the theory-guided convolutional neural network (TgCNN). With the physical information considered, dependence on the data volume of training the deep learning models can be reduced greatly. Several cases of subsurface flow, with varying two-dimensional and three-dimensional structures and isotropic and anisotropic conditions, are used to evaluate the performance of the proposed deep-learning-based upscaling method. The results show that the deep learning method can provide equivalent upscaling accuracy to the numerical method, and efficiency can be improved significantly compared to numerical upscaling. [ABSTRACT FROM AUTHOR]

This study intends to investigate the influence of the rainfall condition and the stability state of the soil layer on the possible time of occurring shallow landslides by building a prediction model which combines a neural network algorithm with a clustering method based on geomorphological characteristics. Therefore, the periods of shallow landslides for different local areas with various geomorphological features in a watershed can be determined. A classification method according to ground slope, soil characteristic, and drainage area is advocated to divide the watershed into multiple local regions before training the landslide prediction model. In this study, to understand the temporal and spatial variation of unstable areas, a seepage flow model and a slope stability model are jointly adopted. The physical meaning of the machine-learning-based model for predicting the duration and the start time of shallow landslides can be preserved because the model input, namely, the dynamic cumulative unstable area, is provided by conducting the theoretical seepage flow model and analyzing the soil slope stability. The proposed approach can effectively determine the periods when shallow landslides are likely to occur for different regions suggested in this study. Present results also indicate that only some specified regions, prone to induce a significant change of total unstable grids during rains, are required to be tracked and analyzed by the proposed prediction model. According to the results of analyzing the predicted errors of the shallow landslide period (Ps) and the initial time of occurring landslide (Ts), the mean relative error can be controlled within 5.19%, and the R2 can be consistently larger than 0.889 overall. Research findings also demonstrate that the predicted periods in which shallow landslides may occur can predict approximately 82.2% of the observed landslide events in the study watershed. [ABSTRACT FROM AUTHOR]

In‐stream wood structures, such as single logs, river steps, and debris dams, are known to drive hyporheic flow, defined as the flow that goes into the subsurface region and then back to the free‐flowing surface water. The hyporheic flow plays an important role in regulating water quality and biogeochemical cycles in rivers. Here, we investigated the impact of a channel‐spanning porous log jam, representing piles of wood logs, on hyporheic flow through a combination of direct visualization and theories. Specifically, we developed a method using refractive index‐matched sediment to directly visualize the hyporheic flow around and below a porous log jam, formed by piles of cylindrical rods, in a laboratory flume. We tracked the velocity of a fluorescent dye moving through the transparent sediment underneath the log jam. In addition, we measured the water surface profile and the spatially varying flow velocity near the log jam. Our results show that the normalized log jam‐induced hyporheic flux remained smaller than 10% at Froude numbers (Fr $\mathit{Fr}$) below 0.06 and increased by a factor of five with increasing Fr $\mathit{Fr}$ at Fr>0.06 $\mathit{Fr} > 0.06$. We combined the mass and momentum conservation equations of surface flow with Darcy's equation to explain the dependency of the log jam‐induced hyporheic flux on Fr $\mathit{Fr}$. Further, we observed that at Fr>0.06 $\mathit{Fr} > 0.06$, the water surface dropped noticeably and the turbulent kinetic energy increased immediately on the downstream side of the log jam. These findings will facilitate future quantification of hyporheic flow caused by channel‐spanning porous log jams. Plain Language Summary: Log jams are trees that frequently fall and accumulate in rivers. Field surveys and numerical simulations suggest that log jams slow down the surface flow and drive hyporheic flows, which are bidirectional flows that go into the riverbed and back to the surface water. Hyporheic flows carry pollutants and nutrients and thus play a critical role in water quality and river biogeochemical cycles. Despite the importance of hyporheic flows, the quantitative characterization of log jam‐induced hyporheic flows remains incomplete. In this study, we conducted experiments in a water‐recirculating flume with a log jam model that resembles piles of wood logs commonly found in rivers. We injected a fluorescent dye into a transparent sediment bed made from hydrogel beads and visualized the flow within the sediment bed. Our experimental results show that log jams can increase the hyporheic flow rate by one order of magnitude. Further, we developed a theoretical model to explain the hyporheic flow induced by a log jam. Our experimental results and theoretical model will facilitate the evaluation of the impact of log jams on the fate and transport of nutrients and contaminants in rivers for future restoration projects. Key Points: The hyporheic flow velocity in a channel with a log jam was two orders of magnitude higher than that of a bare channelAt Froude numbers (Fr $\mathit{Fr}$) above 0.06, the nondimensional log jam‐induced hyporheic flux increased by a factor of five with increasing Fr $\mathit{Fr}$Similar to the hyporheic flux, the water surface profile and turbulence intensity changed noticeably at Fr $\mathit{Fr}$ above 0.06 [ABSTRACT FROM AUTHOR]

This study compares biochar (BCW) systems' pollutant removal effectiveness to conventional subsurface flow (CCW) in constructed wetland systems to treat textile wastewater. The two systems were identical in construction, but the biochar was 0.1 m thick over gravel and sand (maximum flow rate of 0.021 m3 h−1) as the primary medium over CCW (flow rate of 0.02 m3 h−1). The results revealed that the BCW approach was more efficient than the CCW system (pebble over sand and gravels) in removing and lowering heavy metals below thresh hold limits such as Cr, Cd, Cu, Pb, Ni, and Zn. The alkaline nature of textile water achieves neutrality in both CCW and BCW. However, BCW is more efficient due to a larger active surface area and the ability to filter out more metal and organic ions. TDS reduction efficiency in BCW was 53.07%, compared to 40.04% in CCW. Heavy metal removal was 100% in BCW at 3 to 12 h, whereas it takes 6 to 24 h in CCW (82% for Cr to 93% for Cu). The quick removal of Na from textile wastewater by BCW was reversed and achieved equilibrium in 24 h in contrast to the CCW system (> 24 h). The findings obtained at the lab scale level demonstrated that the BCW system was more effective in reducing TDS, neutralizing the alkalinity of textile wastewater, and removing heavy metals. This study strongly supports the potential application of biochar-constructed wetlands for textile wastewater treatment. [ABSTRACT FROM AUTHOR]

On karst steep slopes, subsurface flow frequently emerges at the rock-soil interface, thereby exacerbating soil erosion. However, there have been few studies that quantitatively assess the impact of subsurface flow on soil erosion in this region. We conducted indoor artificial rainfall experiments to observe runoff and sediment yield under two conditions: with subsurface flow (SF) and without subsurface flow input (WSF). We considered various rainfall intensities (30 mm h−1, 60 mm h−1, 90 mm h−1) and slope gradients (30°, 35°, 40°). The results demonstrated that the presence of subsurface flow significantly amplified both surface runoff and sediment yield, while also intensifying the influence of rainfall intensity and slope gradient on runoff and sediment yield. Compared to WSF, under SF conditions, surface runoff increased by 1.55 to 4.43 times, and sediment yield increased by 4.60 to 8.72 times. Slope gradient played a substantial role in soil loss, while rainfall intensity primarily impacted surface runoff. Under WSF conditions, soil loss only occurred at a 40° slope gradient or under a rainfall intensity of 90 mm h−1, whereas under SF conditions, soil loss could occur at a 30° slope gradient or under a rainfall intensity of 30 mm h−1. This research provides a scientific foundation for a more profound understanding of the mechanisms driving soil erosion on karst slopes, particularly in the context of subsurface flow.

Pollution of aquatic ecosystems is mostly caused by human activities such as the laundry industry in urban areas, where often the disposal of laundry waste that is discharged directly without proper management triggers environmental pollution. There are two compounds in laundry waste that are difficult to decompose naturally in water, which can trigger environmental pollution in river ecosystems, namely sodium dodecyl benzene sulfonate (NaDBS) and sodium tripolyphosphate (STTP). Therefore, it is necessary to carry out integrated treatment measures, one of which is the Subsurface Flow Constructed Wetland model. This research aims to see the level of effectiveness of the subsurface Flow constructed wetlands model using two types of plants, namely Epipremnum aureum and Canna indica in reducing the laundry waste parameters such as Chemical Oxygen Demand (COD), Total Suspended Solid (TSS) and Detergent. This research was carried out experimentally by comparing the output of the reactor processing between the inlet and outlet. Based on this research, it can be concluded that the level of effectiveness of laundry waste treatment in the subsurface flow constructed wetland system model uses two types of plants, namely Epipremnum aureum and Canna indica can reduce the concentration of test parameters, namely COD by 76.4%, TSS by 66.2%, and Detergent by 80.9%.

Wetlands technology is one of the main sustainable and successful treatment processes. Similarly, biochar is an organic, effective, and low-cost adsorbent material used for the treatment of diverse wastewaters. The combination between wetland system and biochar, as a media, can greatly enhance the treatment efficiency. The aim of this study is to assess the performance of two horizontal flow constructed wetlands planted with Bacopa monnieri L. for the treatment of household greywater. The objectives were to investigate the raw and treated greywater characteristics, compare the removal efficiency of pollutants by using gravel bed, and biochar-gravel bed, monitor the growth and survival of the plants. Findings indicated that the simulated treatment systems were able to improve all the greywater characteristics. The wetland with biochar enhanced the removal efficiency of biological oxygen demand (BOD5), ammonia (NH3), and other parameters compared with the wetland with gravels alone.

Wetlands technology is one of the main sustainable and successful treatment processes. Similarly, biochar is an organic, effective, and low-cost adsorbent material used for the treatment of diverse wastewaters. The combination between wetland system and biochar, as a media, can greatly enhance the treatment efficiency. The aim of this study is to assess the performance of two horizontal flow constructed wetlands planted with Bacopa monnieri L. for the treatment of household greywater. The objectives were to investigate the raw and treated greywater characteristics, compare the removal efficiency of pollutants by using gravel bed, and biochar-gravel bed, monitor the growth and survival of the plants. Findings indicated that the simulated treatment systems were able to improve all the greywater characteristics. The wetland with biochar enhanced the removal efficiency of biological oxygen demand (BOD5), ammonia (NH3), and other parameters compared with the wetland with gravels alone. [ABSTRACT FROM AUTHOR]

Inverse modeling is usually necessary for prediction of subsurface flows, which is beneficial to characterize underground geologic properties and reduce prediction uncertainty. Considering the intensive computational effort required for repeated simulation runs when solving inverse problems, surrogate models can be built to substitute for the simulator and improve inversion efficiency. Deep learning models have been widely used for surrogate modeling of subsurface flow. However, a large amount of training data is usually needed to train the models, especially for constructing globally accurate surrogate models, which would bring large computational burden. In fact, the local accuracy of surrogate models in regions around the true solution of the inverse problem and the potential searching path of the solution is more important for the inversion processes. The local accuracy of surrogate models can be enhanced with active learning. In this work, active learning strategies based on likelihood or posterior information are proposed for inverse modeling, including both offline and online learning strategies. In the offline strategy, a pre‐trained model is utilized to select samples with higher likelihood, which can produce model responses closer to the observations, and then the selected samples can be used to retrain the surrogate. The retrained surrogate is further integrated with the iterative ensemble smoother (IES) algorithm for inversion. In the online strategy, the pre‐trained model is adaptively updated and refined with the selected posterior samples in each iteration of IES to continuously adapt to the solution searching path. Several subsurface flow problems, including both single‐phase groundwater flow and two‐phase (oil‐water) flow problems, are introduced to evaluate the performance of the proposed active learning strategies. The results show that the proposed strategies achieve better inversion performance than the original surrogate‐based inversion method without active learning, and the number of required simulation runs can also be reduced. Key Points: Active learning strategies based on posterior information are proposed for inverse modeling of subsurface flow problemsBoth offline and online active strategies are proposed and compared, which can locally refine the deep learning surrogatesThe proposed strategies achieve better inversion performance than the original surrogate‐based method with reduced computational costs [ABSTRACT FROM AUTHOR]

Ground-penetrating radar (GPR) is a rapid and non-destructive geophysical technique widely employed to detect and quantify subsurface structures and characteristics. Its capability for time lapse (TL) detection provides essential insights into subsurface hydrological dynamics, including lateral flow and soil water distribution. However, during TL-GPR surveys, field conditions often create discrepancies in surface geometry, which introduces mismatches across sequential TL-GPR images. These discrepancies may generate spurious signal variations that impede the accurate interpretation of TL-GPR data when assessing subsurface hydrological processes. In responding to this issue, this study introduces a TL-GPR image alignment method by employing the dynamic time warping (DTW) algorithm. The purpose of the proposed method, namely TLIAM–DTW, is to correct for geometric mismatch in TL-GPR images collected from the identical survey line in the field. We validated the efficacy of the TLIAM–DTW method using both synthetic data from gprMax V3.0 simulations and actual field data collected from a hilly, forested area post-infiltration experiment. Analyses of the aligned TL-GPR images revealed that the TLIAM–DTW method effectively eliminates the influence of geometric mismatch while preserving the integrity of signal variations due to actual subsurface hydrological processes. Quantitative assessments of the proposed methods, measured by mean absolute error (MAE) and root mean square error (RMSE), showed significant improvements. After performing the TLIAM–DTW method, the MAE and RMSE between processed TL-GPR images and background images were reduced by 96% and 78%, respectively, in simple simulation scenarios; in more complex simulations, MAE declined by 27–31% and RMSE by 17–43%. Field data yielded reductions in MAE and RMSE of >82% and 69%, respectively. With these substantial improvements, the processed TL-GPR images successfully depict the spatial and temporal transitions associated with subsurface lateral flows, thereby enhancing the accuracy of monitoring subsurface hydrological processes under field conditions.

Soils irrigated with phosphorus (P)-rich wastewater can increase the risk of P losses from land to water. We investigated if wastewater-irrigated soils can leach P through the soil profile and enrich groundwater. Soils were sampled annually for 20 years to 7.5 cm depth and for one year to 2.2 and 15 m. A mass balance showed that topsoils >200 mg L−1 Olsen P (and <50% anion storage capacity, ASC) were no longer accumulating P and had very high concentrations of water and CaCl2-extractable P (designed to indicate loss to runoff and leaching, respectively). Samples in five paddocks showed substantial P leaching to 2.2 m, while sampling to 15 m showed leaching to 4.5 m depth. Deeper layers, especially in the 15 m deep cores, had a coarse texture and low capacity to sorb P, which caused high P concentrations (1.9 mg L−1) at one groundwater site. Based on these data preventing soils from reaching the thresholds of 200 mg L−1 Olsen P and 50% ASC is necessary to prevent excessive P losses. [ABSTRACT FROM AUTHOR]

Surface flow (SF) and subsurface flow (SSF) are important hydrological processes occurring on slopes, and are driven by two main factors: rainfall intensity and slope gradient. To explore nitrogen (N) migration and loss from sloping farmland in the Miyun Reservoir, the characteristics of total nitrogen (TN) migration and loss via SF and SSF under different rainfall intensities (30, 40, 50, 60, 70, and 80 mm/h) and slope gradients (5°, 10°, and 15°) were studied using indoor stimulated rainfall tests and mathematical models. Nitrogen loss via SF and SSF was found to increase exponentially and linearly with time, respectively, with SSF showing 14–78 times higher loss than SF. Under different rainfall intensities, SSF generally had larger TN loss loading than SF, thereby indicating that SSF was the main route for TN loss. However, the TN loss loading proportion via SF increasing from 14.03% to 35.82% with increasing rainfall intensity is noteworthy. Furthermore, compared with the measurement data, the precision evaluation index Nash-Suttcliffe efficient (NSE) and the determination coefficient (R2) of the effective mixing depth model in the numerical simulation of TN loss through SF in the sloping farmland in the Miyun Reservoir were 0.74 and 0.831, respectively, whereas those of the convection-dispersion equation for SSF were 0.81 and 0.811, respectively, thus indicating good simulation results. Therefore, this paper provides a reference for studying the mechanism of N migration and loss in sloping farmland in the Miyun Reservoir. [ABSTRACT FROM AUTHOR]

[Objective] Rock fragments are ubiquitous in soil, especially in arid and semi-arid regions. They have a considerable influence on various subsurface hydrological processes. Understanding the impact of rock fragments on these processes is crucial for improving water resource management, ecosystem restoration and sustainable development. [Method] This paper reviews the research over the past decades on the influence of rock fragments on subsurface hydrological processes, highlighting the challenges and limitations we currently face and proposing potential research perspectives. Specifically, we analyze the effects of characteristic parameters of rock fragments, such as their coverage, content, and size, on water infiltration, surface runoff and subsurface flow, preferential flow and soil evaporation. We also review progress in mathematical models used to simulate and quantify the effects of rock fragments on subsurface hydrological processes. [Result] Our review reveals that the mechanisms underlying the influence of rock fragments on subsurface hydrological processes remain obscure, and current methods are unable to capture the complex causal relationship between rock fragments and soil hydrological processes. Areas such as quantification of rock fragment heterogeneity and its association with soil hydrological parameters, multi-scale modelling of soil hydrological processes are still in the infant stage and require further work. [Conclusion ] We propose that developing innovative technologies and methodologies, and improving understanding of the underlying mechanisms are particularly important in future research. Specifically, future research should focus on characterizing rock fragments in deep soil and their influence on hydrological process and formation of preferential flow, methods to link gravel content to hydrological information at different scales. Improving research in these areas will advance our in-depth understanding of the impact of rock fragments on soil hydrological processes, helping develop sustainable water resource management in arid and semi-arid regions. [ABSTRACT FROM AUTHOR]

This study investigated the effects of hydrological, agroecological, and soil-hydrophysical factors on the vertical subsurface flow within the vadose zone and groundwater (GW) recharge on the sand massifs in the Don Region. This involves issues of adapting lysimeter studies to field observations in investigations of the moisture transport dynamics in light-textured soils. Techniques for reconstruction of the main hydrophysical characteristics and the scaling were employed to simulate moisture transport under various initial and boundary conditions. Laboratory investigations were performed at the Hydrological Complex of the Federal Scientific Centre of Agroecology of the Russian Academy of Sciences; field studies were conducted on the Eterevsky sand massif in Volgograd Oblast. The reconstructed hydrophysical parameters and HYDRUS-1D program revealed the effects of different soil–plant conditions on GW recharge. The open sands have an inflow to GW throughout the year, which ensures a stable supply to the river systems. The grounds occupied by herbaceous plants and pine plantations on soils with low water holding capacity, are the supplementary source of GW recharge. The volume of the gravity flow from the vadose zone to GW was established to depend entirely on the biomass of the herbaceous plants on the sands and the quality (bonitet) of the forest plantations. [ABSTRACT FROM AUTHOR]

Runoff generated by heavy rainfall in natural catchment area and caused by ditch clogging in artificial structure increases the instability of the slope. The effects of runoff are usually neglected in traditional rainfall-induced slope failure analysis. In order to consider the effects of runoff on the slope instability, this study attempts to simulate the runoff, infiltration and slope instabilities simultaneously on both wide/narrow area. For this purpose, this study firstly proposes a coupled model of surface flow, subsurface flow, and soil mechanics based on Diffusion wave equations, Richards’s equation, Green-Ampt infiltration capacity model, and local factor of safety approach in wide area simulation. Next, snowmelt and surface grass layer are additionally considered in narrow area simulation. Finally, the proposed coupled model is applied to two slopes at Nissho Pass and Kuromatsunai in Hokkaido, Japan. The slope instabilities assessment approach provides an effective method for simulating runoff and slope instabilities. In wide area, the distribution map of the factor of safety has significant implications for precisely determining the dangerous spots and accurately releasing the warning information. In narrow area, the slope disaster points caused by the ditch clogging are extracted and effect of grass layer on slope stability assessment is demonstrated. The construction and maintenance of the drainage system can be re-designed to ensure all the surface flow caused by the precipitation drain out smoothly.

Characterizing physical properties of faults, such as their transmissibility, is crucial for performing predictive numerical simulation of subsurface flows, such as those encountered in petroleum engineering and remediation of subsurface contamination. This paper provides a complete investigation of the inverse problem for fault transmissibility in subsurface flow models, under appropriate assumptions on fault structure. In particular, the following aspects are considered: 1) fault modeling and well-posedness of the forward problem; 2) finite element (FEM) discretizations of the forward problem and their rigorous a priori convergence analysis; 3) Well-posedness of the Bayesian inverse problem, FEM discretization of the infinite dimensional Bayesian inverse formulation, and its rigorous a priori analysis. Moreover, computation of the maximum a posteriori (MAP) point via fast inexact Newton-conjugate gradient optimization and a Laplace approximation of the Bayesian posterior are also presented. Numerical results illustrate the use of the proposed fault model in forward and inverse problems for subsurface flows in two dimensional domains with multiple faults. • A priori error analysis of finite element methods for a faulted subsurface flow model. • Well-posedness of the Bayesian inverse problem for fault transmissibility. • FEM discretization of the infinite dimensional Bayesian inverse problem with analysis. • Maximum a posteriori computation via a fast inexact Newton method. • Numerical results illustrating forward and inverse problems of fault models. [ABSTRACT FROM AUTHOR]

Slope farmland is prone to soil erosion, especially in sub/tropical regions. However, our understanding of near-surface hydrology characteristics and their controlled factors in red soil sloping farmland remains limited. Here, we conducted simulated rainfall experiments to assess the impact of rainfall pattern, straw mulching, and soil structure on near-surface hydrological processes of red soil sloping farmland of southern China. Results showed that: (1) short duration-high intensity rain caused greater surface runoff and sediment production than did long duration-low intensity rain, whereas the variation pattern of subsurface flow exhibited the opposite trend; (2) tillage behavior could weaken the surface runoff intensity and promote the development of subsurface flow; (3) straw mulching increased the water infiltration rate and associated subsurface flow production (increased by 1.33~12.71 times), and thus reduced the surface runoff production (reduced by 99.68~100%). These findings highlight the crucial roles of rainfall pattern and straw mulching in regulating the spatial distribution pattern of rainwater and suggest that straw mulching can effectively reduce soil erosion via accelerating water infiltration and subsurface flow form in slope farmland of soil erosion in southern China. [ABSTRACT FROM AUTHOR]

The optimization of well locations and controls is an important step in the design of subsurface flow operations such as oil production or geological CO 2 storage. These optimization problems can be computationally expensive, however, as many potential candidate solutions must be evaluated. In this study, we propose a graph network surrogate model (GNSM) for optimizing well placement and controls. The GNSM transforms the flow model into a computational graph that involves an encoding-processing-decoding architecture. Separate networks are constructed to provide global predictions for the pressure and saturation state variables. Model performance is enhanced through the inclusion of the single-phase steady-state pressure solution as a feature. A multistage multistep strategy is used for training. The trained GNSM is applied to predict flow responses in a 2D unstructured model of a channelized reservoir. Results are presented for a large set of test cases, in which five injection wells and five production wells are placed randomly throughout the model, with a random control variable (bottom-hole pressure) assigned to each well. Median relative error in pressure and saturation for 300 such test cases is 1-2%. The ability of the trained GNSM to provide accurate predictions for new (geologically similar) permeability realizations is demonstrated. Finally, the trained GNSM is used to optimize well locations and controls with a differential evolution algorithm. GNSM-based optimization results are comparable to those from simulation-based optimization, with a runtime speedup factor of 36. Much larger speedups are expected if the method is used for robust optimization, in which each candidate solution is evaluated on multiple geological models. • Graph-network-based surrogate model developed for subsurface flow optimization. • Surrogate predicts pressure and saturation states for different well configurations. • Well controls (bottom-hole pressures) can also be varied. • Multistage multistep training strategy applied to enable predictions through time. • Accurate state predictions for different geological models achieved. [ABSTRACT FROM AUTHOR]

A deep learning model can perform efficient uncertainty quantification (UQ) for reservoir flow with uncertain model parameters, but usually requires large amounts of training data to ensure accuracy. However, the cost of obtaining large amounts of data is prohibitive, and the performance will deteriorate if sufficient training data is lacking. Alternatively, more interpretable neural networks with embedded physical laws have recently been used to solve partial differential equations as well as to solve UQ problems. This approach has received a lot of attention due to its low data volume requirements and its adherence to the laws of physics during the training process. In this paper, we propose a theory-guided framework based on a bilevel programming model with hard constraints to embed physical meaning in the model. Theory guided Lagrange programming neural network (TGLPNN) combines the method of Lagrange programming neural network approach where physical laws such as stochastic partial differential equations and boundary conditions are incorporated into the training process of a convolutional neural network. At the same time, the upper-level variables are iteratively optimized. The method based on Lagrange programming neural network inherently embeds physical laws in the network. Practical applications have shown that TGLPNN can provide higher prediction accuracy compared to state-of-the-art physics-driven methods and improved efficiency compared to numerical methods. • Improving embedding theory guidance using Lagrange Programming neural network. • Optimizing hyperparameters using Meta-Weight-Net. • Effective neural network approach to subsurface flow problems. [ABSTRACT FROM AUTHOR]

This study presents a hybrid Finite Volume – Finite Element (FV-FE) model that describes the coupled surface-subsurface flow processes occurring during furrow irrigation and fertigation. The numerical approach combines a one-dimensional description of water flow and solute transport in an open channel with a two-dimensional description of water flow and solute transport in a subsurface soil domain, thus reducing the dimensionality of the problem and the computational cost. The modeling framework includes the widely used hydrological model, HYDRUS, which can simulate the movement of water and solutes, as well as root water and nutrient uptake in variably-saturated soils. The robustness of the proposed model was examined and confirmed by mesh and time step sensitivity analyses. The model was theoretically validated by comparison with simulations conducted with the well-established model WinSRFR and experimentally validated by comparison with field-measured data from a furrow fertigation experiment conducted in the US.

Physics-informed neural network (PINN) models are developed in this work for solving highly anisotropic diffusion equations. Compared to traditional numerical discretization schemes such as the finite volume method and finite element method, PINN models are meshless and, therefore, have the advantage of imposing no constraint on the orientations of the diffusion tensors or the grid orthogonality conditions. To impose solution positivity, we tested PINN models with positivity-preserving activation functions for the last layer and found that the accuracy of the corresponding PINN solutions is quite poor compared to the vanilla PINN model. Therefore, to improve the monotonicity properties of PINN models, we propose a new loss function that incorporates additional terms which penalize negative solutions, in addition to the usual partial differential equation (PDE) residuals and boundary mismatch. Various numerical experiments show that the PINN models can accurately capture the tensorial effect of the diffusion tensor, and the PINN model utilizing the new loss function can reduce the degree of violations of monotonicity and improve the accuracy of solutions compared to the vanilla PINN model, while the computational expenses remain comparable. Moreover, we further developed PINN models that are composed of multiple neural networks to deal with discontinuous diffusion tensors. Pressure and flux continuity conditions on the discontinuity line are used to stitch the multiple networks into a single model by adding another loss term in the loss function. The resulting PINN models were shown to successfully solve the diffusion equation when the principal directions of the diffusion tensor change abruptly across the discontinuity line. The results demonstrate that the PINN models represent an attractive option for solving difficult anisotropic diffusion problems compared to traditional numerical discretization methods. [ABSTRACT FROM AUTHOR]

Residence time of water flow is an important factor in subsurface media to determine the fate of environmental toxins and the metabolic rates in the ecotone between the surface stream and groundwater. Both numerical and lab-based experimentation can be used to estimate the residence time. However, due to high variability in material composition in subsurface media, a pragmatic model set up in the laboratory to trace particles is strenuous. Nevertheless, the selection and inclusion of input parameters, execution of the simulation, and generation of results as well as post-processing of the outcomes of a simulation take a considerable amount of time. To address these challenges, an automated particle tracing method is developed where the numerical model, i.e., flow and reactive transport code, MIN3P, and MATLAB code for tracing particles in saturated porous media, is used. A rectangular model domain is set up considering a fully saturated subsurface media under steady-state conditions in MIN3P. Streamlines and residence times of the particles are computed with a variety of seeding locations covering the whole model surface. Sensitivity analysis for residence time is performed over the varying spatial discretization and computational time steps. Moreover, a comparative study of the outcomes with Paraview is undertaken to validate the automated model (R2 = 0.997). The outcome of the automated process illustrates that the computed residence times are highly dependent on the accuracy of the integration method, the value of the computational time step, ∆t, spatial discretization, stopping criterion for the integration process of streamlines, location, and amount of seed points. The automated process can be highly beneficial in obtaining insights into subsurface flow dynamics with high variability in the model setup instead of laboratory-based experimentation in a computationally efficient manner. [ABSTRACT FROM AUTHOR]

Preferential flow (PF) through soil and regolith results in a rapid vertical and lateral water movement within the profile. This study focused on quantifying PF in contrasting forested hillslopes of the Hailuo Valley, located on China's east Tibetan Plateau. Quantifying PF in this region is challenging since the underlying matrix is complex, with shallow soils, thick underlying saprolite, and the combined effect of moraines and weathered residual material. We developed new methods that integrated dye tracing with a high‐frequency soil moisture monitoring network to characterise PF pathways and quantity PF frequency on contrasting forested hillslopes (coniferous and broadleaved). Dye tracing experiments showed that soil‐root interfaces in the upper soil layer (10–50 cm) and soil‐rock interfaces in the deeper soil layer (50–100 cm) were the primary PF pathways and that large rocks strongly influence the percolation depth of water. The high‐frequency soil moisture network revealed that the mean PF frequency was 67.8% and 71.7% in the coniferous and broadleaved forests, respectively. The frequency of PF in the deeper soil layer increased from upslope to the downslope locations on both forested hillslopes, highlighting the tendency for PF to occur downslope. In addition to matrix conditions (e.g., stony saprolite soil), the total amount and maximum intensity of precipitation (as throughfall in this study) were identified as factors that control PF occurrence. In contrast, the initial soil water conditions (at 10, 50 and 100 cm depth) were insignificant in predicting PF occurrence. Results of this field‐based study highlight that PF is a ubiquitous and critical subsurface flow mechanism that regulates rainfall‐runoff relationships in forests of the Tibetan Plateau, underscoring the need to consider PF in hillslope hydrological modelling. Highlights: Combined dye tracing and soil moisture monitoring network were applied to explore PF on forested hillslopes.Root networks and rock fragments constituted the main PF paths.PF frequency increased from upslope to downslope in two forest types.Initial soil moisture conditions showed no controlling effect on PF occurrence. [ABSTRACT FROM AUTHOR]

Understanding the interaction between competing fluids in the pore space of rocks is key for predicting subsurface flow and trapping, such as with CO2 in a saline aquifer. These processes occur over a large span of timescales (from seconds to thousands of years), and length scales (from microns to kilometers). Understanding the link between these temporal and spatial scales will enable us to interpolate between observations made at different resolutions. In this work we explore the temporal scales present during macroscopically steady‐state multiphase flow in a porous carbonate rock using differential pressure measurements acquired over a period of 60 min. Nitrogen and brine were injected simultaneously into a sample 5 mm in diameter and 21 mm in length. We observe a cascade of timescales in the pressure differential that is, a continuous range of frequencies, with lower frequencies having greater amplitudes. We demonstrate a scaling of the spectral density with frequency of S ∼ 1/f2, or red noise, to describe the dynamics. This scaling is independent of the flow rate of the fluids or the fraction of the flow taken by water. This red, or Brownian, noise indicates a stochastic process where pressure fluctuations are seen throughout the pore space, resulting in intermittent filling of pores over a wide range of time‐scales, from seconds to minutes in these experiments. The presence of red noise suggests self‐organized critically, with no characteristic time or length scale. Key Points: Pressure fluctuations are linked to fluid flow dynamics in the pore spaceThe pressure fluctuations span a range of frequencies. The spectral density scales with frequencyRed noise can be used to describe the dynamics [ABSTRACT FROM AUTHOR]

Surface flow (SF) and subsurface flow (SSF) are important hydrological processes occurring on slopes, and are driven by two main factors: rainfall intensity and slope gradient. To explore nitrogen (N) migration and loss from sloping farmland in the Miyun Reservoir, the characteristics of total nitrogen (TN) migration and loss via SF and SSF under different rainfall intensities (30, 40, 50, 60, 70, and 80 mm/h) and slope gradients (5°, 10°, and 15°) were studied using indoor stimulated rainfall tests and mathematical models. Nitrogen loss via SF and SSF was found to increase exponentially and linearly with time, respectively, with SSF showing 14–78 times higher loss than SF. Under different rainfall intensities, SSF generally had larger TN loss loading than SF, thereby indicating that SSF was the main route for TN loss. However, the TN loss loading proportion via SF increasing from 14.03% to 35.82% with increasing rainfall intensity is noteworthy. Furthermore, compared with the measurement data, the precision evaluation index Nash-Suttcliffe efficient (NSE) and the determination coefficient (R2) of the effective mixing depth model in the numerical simulation of TN loss through SF in the sloping farmland in the Miyun Reservoir were 0.74 and 0.831, respectively, whereas those of the convection-dispersion equation for SSF were 0.81 and 0.811, respectively, thus indicating good simulation results. Therefore, this paper provides a reference for studying the mechanism of N migration and loss in sloping farmland in the Miyun Reservoir.

Tourism expansion can put pressure and deterioration on natural resources particularly in vulnerable riverside areas. Because of water pollution scenario could negatively impact the number of visitors, wastewater treatment before discharging becoming a necessary process for resort enterprises. However, lack of sewerage network and treatment system in Thailand poses the need to have an affordable and efficient onsite wastewater treatment system. In this study, a modified subsurface flow constructed wetland (MSF-CW) system was applied to treat a resort wastewater that had high and fluctuating organic contents. The system performance, investigated during 7 months, revealed that chemical oxygen demand (COD), biochemical oxygen demand (BOD), total kjeldahl nitrogen (TKN), total phosphorus (TP) and suspended solids (SS) removal efficiencies were 81%, 84%, 30%, 40% and 78%, respectively, and all meeting the Thai building effluent standards. The treated effluent qualities were also compared to the guidelines for water reuse, then applied for landscape irrigation at the resort to promote water conservation and sustainable use. Based on the analytical results, the hydraulic retention time (HRT) of this system should not be less than 5 days to maintain a satisfactory efficiency. From the long-term monitoring period, this system showed easy operation with low maintenance cost. The benefits of the MSF-CW application in this study are not only for wastewater treatment but also for green area creation, useful for the resort to promote eco-tourism and sustainability. The constructed wetland (CW) system is, therefore, strongly suggested as a facility for small or medium-size resorts, hotels and homestays to achieve water pollution abatement and promoting eco-sustainable tourism. [ABSTRACT FROM AUTHOR]

This paper presents the results of laboratory tests that allowed us to determine the effect of the soil surface inclination and its initial moisture content on the formation of overland and subsurface runoff. The experiments were carried out for the soil that is commonly present in the southern part of Poland, including the Outer Carpathians. The results of these measurements served as a reference for overland runoff calculations using the Richards model, simplified Green–Ampt model, and the empirical model (MSME). The results of the measurements showed that, for low-permeable soil, overland runoff is the dominant form. It was shown that a slope in the range of 2.5–5.0% does not have a significant effect on the amount of overland runoff, but affects its dynamics. The measurements also showed that the starting time and amount of overland runoff are strictly associated with the initial soil moisture content. High soil moisture content in the period preceding the onset of rainfall causes faster generation and an increase in overland runoff, which is caused by the saturation of the surface layer of the soil. This mechanism was confirmed by the results of calculations using the Richards model and measurements of the electrical resistance of the soil. Theoretical calculations showed that the results of the runoff calculations using the Richards and Green–Ampt models are strongly dependent on the hydraulic properties of the soil adopted for the analysis. It was also demonstrated that the modified MSME model satisfactorily estimates the amount of overland and subsurface runoff, but requires parameter calibration based on existing hydrological data. [ABSTRACT FROM AUTHOR]

In gravity irrigation, how water is distributed in the soil profile makes it necessary to study and develop methodologies to model the process of water infiltration and redistribution. In this work, a model is shown to simulate the advancing front in border irrigation based on the one dimensional equations of Barré de Saint-Venant for the surface flow and the equation of Green and Ampt for the flow in a porous medium. The solutions were obtained numerically using a finite difference Lagrangian scheme for the surface flow and the Raphson method for the subsurface flow. The model was validated with data obtained from the literature from an irrigation test and its predictive capacity was compared with another model and showed excellent results. The hydrodynamic parameters of the soil, necessary to obtain the optimal irrigation discharge, were obtained through the solution of the inverse problem using the Levenberg–Marquardt optimization algorithm. Finally, the results found here allow us to recommend that this model be used to design and model border irrigation, since the infiltration equation uses characteristic parameters of the physical soil. [ABSTRACT FROM AUTHOR]

Flame spread over liquid fuel is an evitable process during the development of spilling liquid fires. The scale of the combustion zone may be expanded by the spreading fire, making the fire uncontrollable. Thus, the flame spread is a major threat to the process safety of the usage of liquid fuels. An experimental apparatus is designed to study the flame spread over n‐butanol under the effect of the gas step that is built by the low‐heat conductivity marbles. The flame morphology, flame resident time, and movement of subsurface liquid convection are measured under various widths and heights of the marble steps. The resident time is affected by the coupling effects of the viscous resistance of step and the heat loss to unburned fuel. The flame height is decreased by the width of the step but enhanced by the height of the step. The convective heat transfer through subsurface flow and flame radiation to the unburned fuel behind the step is calculated. The convective heat transfer rate accounts for more than 90% of the total heat transfer rate. This current finding provides a theoretical basis for the safety application and emergency management of liquid fuels. [ABSTRACT FROM AUTHOR]

Constructed wetlands as a treatment system are widely explored in different climate conditions and established to be effective in pollution removal from water environment. This study aims to demonstrate the performance of pilot-scale subsurface flow constructed wetland for storm water treatment in Latvia. The catchment basin was located in a farmyard of agricultural area and storm water was collected from the impermeable pavements. Storm water was accumulated in an open pond and periodically pumped above the filter part of the subsurface flow constructed wetland. Grab samples were collected once or twice per month at the inlet and outlet of the treatment system during a period of 73 months from year 2014 to 2020. Water quality parameters as nitrate nitrogen (NO3-N), ammonium nitrogen (NH4–N), total nitrogen (TN), orthophosphate phosphorus (PO4-P), and total phosphorus (TP), total suspended solids (TSS), biochemical oxygen demand (BOD5) and chemical oxygen demand (COD) were monitored. Water level at the inlet structure was automatically measured and flow rate was calculated based on the Manning equation for partially filled circular pipes. Results showed the reduction of average concentrations for all parameters during the study period. However, in some sampling cases concentrations increased at the outlet of the treatment system and can be explained by influencing factors of farming and maintenance. The treatment efficiency of NO3-N, NH4-N, TN, PO4-P, TP, TSS, BOD5 and COD concentrations was 17 %, 68 %, 55 %, 78 %, 80 %, 57 %, 80 % and 74 %, respectively. The study site demonstrated a potential to improve water quality in the long term.

• Effective method is proposed for exploring the fundamental mechanism of SSF. • Soil temperature is proved to be the most fundamental factor for SSF pathways. • The patterns of vertical and lateral source for SSF are analyzed and figured out. The impact of the freeze–thaw process on the active layer is reflected in the changed subsurface flow (SSF) process in cold alpine regions. Identifying sources and pathways of SSF in the freeze–thaw process is critical but difficult, and the related dominant factors and mechanisms are still unknown. In this paper, the effective identification and analysis of SSF are promoted based on field sampling data from the thawing (June) to freezing (September) period of 2022 in the Qinghai Lake basin on the northeastern Qinghai-Tibetan Plateau. By the proposed method with a high sampling frequency and refined sampling spatial scale, the sources and pathways of SSF are clearly identified. The results are as follows: (1) The soil temperature is considered the most fundamental factor affecting the SSF pathways, it influences water infiltration to the deep layer and the effect is extended to the saprolite and weathered bedrock layers. (2) Thawing promotes water to infiltrating into deep layer. 30 cm soil water contributes the most to SSF (2 %–86 %) in the thawing period, while the contribution difference of the water from the 30 cm, 60 cm, and 90 cm layers is small (ranging from 32 %–33 %, 24 %–26 %, and 32 %–35 %, respectively) in the thawed period. (3) Meanwhile, the soil water from different slope positions contribute differently to SSF, and the SSF from deep soil layer is transit in prolong paths and depths. It is caused by the out-of-sync water transit process in the hillslope. With continuing climate warming, we propose that the differences in the water sources of SSF across soil layers may decrease, while the differences in the transit processes of SSF across soil layers may increase. [ABSTRACT FROM AUTHOR]