ABSTRACT. Continuous and discontinuous approaches to fracture have both their advantages and drawbacks. On one side, the discontinuous approach, Griffith type or cohesive, allows for a clear separation of matter but was not designed to take into account easily triaxiality effects close to the crack faces, as well as complicate crack patterns (merging and branching).

On the other side, continuous approaches allows one to use the full blown capabilities of bulk constitutive models but face tremendous computational cost especially when process zone sizes are important. Continuous model are also not well suited if crack opening informations need to be accessed (for instance for a coupling with fluid flow).

The Thick Level Set approach (TLS) to fracture reconciles both approaches and make them usable both at the same time. The key idea is that bulk and interfacial damage are being given a related geometrical meaning (configurational type). Both models may thus combined to create a new powerful material modeling frame.

ABSTRACT. To represent groundwater-surface water interactions, lateral flows as well as the human impacts on head dynamics and the impact of capillary rise on evapotranspiration it is necessary to simulate the location and temporal variation of the groundwater table. Global-scale hydrological models have recently moved to include these processes even though data is difficult to obtain. We present an application of the well-established Morris method on a newly developed Global Gradient-based Groundwater Model 3GM to assess the sensitivity of model inputs.

ABSTRACT. We intend to introduce a proposed global wetland map in the ORCHIDEE land surface model, acting as the riparian buffer zone which receives water both from infiltration and down-flowing runoff. This will result in increased soil moisture over the wetland fraction and sometimes building a water table in the soil column, with better representation of latent heat fluxes and seasonal variations. Results will be compared to observations and also 3-D modelling results at the Little Washita watershed in the USA.

ABSTRACT. Groundwater being part of the surface continental hydrology, it is likely to impact the climate system. Therefore, it has recently been integrated to Earth System Models. CNRM-ESM2 includes a parameterization of unconfined aquifers in its river routing module, CTRIP, coupled to the land surface model ISBA. This hydrological system was validated in offline mode in previous studies. Here, we evaluate the effects of the simulated groundwater on present-day climate, and its impact on climate sensitivity in idealized climate change scenarios.

ABSTRACT. Previous studies showed when considering the groundwater dynamics in a climate model, the wet soil induced surface cooling effect can further reduce the dry season precipitation in Amazon. Further, we examine the surface cooling effects by using the idealized prescribed water table depth simulations from the I-GEM (Impact of Groundwater in Earth system Models Groundwater) project. A contradict result of the tendency of precipitation has been found, which indicates the uncertainties of the groundwater’s impacts in coupled GCMs.

ABSTRACT. In this study we propose an approach to image the spatial distribution of geochemical heterogeneities. The approach is based on distributed sensor data of water quality parameters combined with forward and inverse reactive transport modeling. The method was tested on oxidative dissolution of pyrite investigated in 1-D columns and 2-D flow-through setups.

ABSTRACT. We examine the impact of flow heterogeneities at pore scale modeling an instantaneous bimolecular chemical reaction A + B --> C in a laminar Poiseuille flow reactor. We use a reactive random walk particle tracking method which represents the reactants by particles. We are able to predict the total amount of product C formed during the reaction through the effective dispersion coefficient, which is based on the temporal evolution of the spatial moments of the plume.

ABSTRACT. Mineral dissolution rates in physically heterogeneous systems are explored using multi-component reactive transport simulations with non-linear kinetic reactions. Results from simulations at varying spatial and temporal scales demonstrate the contribution of physical heterogeneity to the observed discrepancy between laboratory- and field- measured mineral dissolution rates. Orders of magnitude deviation from laboratory-derived dissolution rates are observed in systems where preferential pathways of fast fluid velocity bound low permeability reactive zones once reactive mineral is depleted in the preferential pathways.

ABSTRACT. Dissolution of pore-scale soluble substances occurs in applications from environmental hydrogeology to CO2 storage. Development of Darcy-scale models has been widely discussed. This paper proposes an upscaling algorithm to develop large-scale models taking into account Darcy-scale heterogeneities. The theory is based on a comparison between the characteristic length-scale of the Darcy-scale heterogeneities and the length-scale of the dissolution front controlled by a Darcy-scale Damk\"{o}hler number. The obtained large-scale model is shown to confront favorably to Darcy-scale direct numerical simulations.

ABSTRACT. A domain decomposition algorithm is introduced to model the mass and energy exchanges between the geological formation and the ventilation galleries of a geological radioactive waste disposal. At each time step, the algorithm solves iteratively the nonisothermal gas liquid Darcy flow in the porous medium, the RANS gas flow in the free-flow domain, and the transport and energy equations in the free-flow domain. To speed up the convergence, Robin type boundary conditions with optimized coefficients are designed.

ABSTRACT. We present a space-time domain decomposition method for the simulation of subsurface flow, based on based on the Optimized Schwarz Waveform Relaxation algorithm, which enables local time stepping.

In order to decide when to stop the DD iterations, we have developed fully computable a posteriori error estimates that allow for the separation of the part of the error due to the space and time discretizations, and that due to the DD iterations. These error estimates are based on reconstruction techniques.

ABSTRACT. We develop a fully implicit finite volume scheme for numerical modeling of single and two-phase multicomponent flow with reactive transport in porous media. Modeling such problems leads to a highly nonlinear coupled system of PDEs to differential algebraic equations or ODEs. We have developed and integrated reactive transport modules in the DuMuX framework. Numerical results for 1D and 2D including benchmarks tests will be presented. The results have demonstrated that this approach yields physically realistic flow fields in heterogeneous media.

ABSTRACT. While finite element analysis is a well-established tool in science and engineering, traditional solver approaches become inefficient or even break down entirely for strongly heterogeneous problems and anisotropies. The topic of this talk is the GenEO spectral coarse space which can be used to either construct preconditioners that perform well in these situations or to generate efficient coarse representations in multiscale applications. A highly scalable implementation in the DUNE numerical framework will be presented alongside numerical results demonstrating its effectiveness.

ABSTRACT. In fractured rock masses, fluid flows through interconnected channels. These channels are separated by solid bridges that transmit shear stresses. When the local fluid pressure gets large enough, it generates seismogenic slips. A combined inversion of direct stress measurements and focal mechanisms of induced seismicity helps map the pore pressure magnitude. Results from an in situ experiment help demonstrate the existence of poorly connected channels. They have been confronted to in situ testing that demonstrated the role of hydromechanical coupling

ABSTRACT. We investigate the source characteristics of picoseismicity recorded during a hydraulic fracturing in-situ experiment performed in Äspö Hard Rock Laboratory, Sweden. Six stimulations resulted in induced seismicity with MW<-3.5 (sub-dm size fractures). The results of monitoring and analysis are discussed in the context of seismic injection efficiency and maximum potential magnitude

ABSTRACT. We developed an in-situ experiment of fluid injections at a 10-meter scale. Water was injected at high-pressure in a fault, in limestone at 280 m depth in the LSBB Underground Laboratory (France). Induced seismicity, as well as strains, pressure and flowrate were continuously monitored during the injections. Although plastic deformations related to fracture reactivations were observed for all injections, only a few tests generated seismicity. Comparing strain measurements and seismicity shows that more than 96% of the deformation is aseismic.

ABSTRACT. Estimating the hydro-mechanical properties of fractured rock reservoirs is critical for many engineering practices. Laboratory and numerical experiments have brought considerable insights, but in situ hydromechanical experiments are relatively uncommon. We propose a new experimental design based on the combination surface tilt and borehole strain observations in sinusoidal hydraulic tests to access both local fracture behavior and its connectivity to the fractured network at ~10m scales.

ABSTRACT. The microorganisms in groundwater rarely cover the matrix uniformly, but form colonies. This limits substrate bioavailability and biodegradation. With a high-resolution numerical model of a pore channel with bacterial colonies we simulated the transport and biodegradation of organic substrates and compared them to 1D simulations based on effective rate laws for bioavailability-limited biodegradation. We quantified bioavailability limitations and evaluated the applicability of established effective rate concepts if microorganisms are heterogeneously distributed. Effective bioavailability was reduced (> one order of magnitude).

ABSTRACT. Pesticide leakage through soil results in critical challenges of health and environment. Yet, the mechanisms of pesticide biodegradation in soils are not fully understood. This work aims to formalize the couplings between spatial processes resulting from the saturation structure, sorption processes and microbial processes involved in biodegradation. Through two reactive transport models, we show that these couplings lead to non-linear uptake and biodegradation of pesticide by microorganisms. Strong discrepancies between initial conditions and biodegradation response can emerge from this non-linearity.

ABSTRACT. A microfluidic model will be introduced to quantify the impact of flow field heterogeneity in filtration of microbial particles (bio-colloids). Long tailing behavior and non-exponential deposition profile are observed in experimental data but not predicted in classical filtration theory (CFT). We use 2D particle tracking simulations and microfluidics experiment to investigate bio-colloids transport and filtration and we upscale the observed macroscopic filtration process in the framework of Reactive Continuous Time Random Walk (R-CTRW) model.

ABSTRACT. Denitrification by bacteria is modelled as a reaction-diffusion process, consisting of coupled reaction-diffusion equations, in a high-resolution pore space geometry obtained form X-ray micro-tomography. The model results are compared to experimental data obtained in laboratory experiments with an ensemble of artificial aggregates form sintered glass, inoculated with bacteria from two different strains capable of denitrification.

ABSTRACT. To improve the current generation of land surface models, we coupled HydroGeoSphere (HGS), a 3D fully integrated surface and variably saturated subsurface flow and transport model to the Noah-MP (Multi-Physics) LSM. The coupling method starts with Noah-MP processing the external climate forcing data (e.g. solar radiation, precipitation, wind velocity, and humidity) and provides HGS node specific net precipitation data. Conversely, the 3D HGS model replaces the internal water balance within Noah-MP by providing it soil saturation.

ABSTRACT. We analyse the similarities and the differences of the conceptual land surface model mHM and the physically based coupled surface subsurface model TerrSysMP. A special focus lies on the scaling behaviour of the spatial and the temporal resolutions.

Therefore, we simulate the Rur and the Bode catchments with both models with resolutions ranging from 200m x 200m to 1km x 1km per cell and from daily to annual timesteps.

ABSTRACT. We present two ways to overcome the computational burden associated with GW-SW models, which is a barrier to the use of parameter estimation and uncertainty analysis: i) An tool to compute the head-dependant stream boundary condition, which is a solution to obtain frugal GW-SW models; ii) a surrogate GW-SW transport model based on particle tracking, with short computation time. These approaches increase the robustness of SW-GW models prediction and make possible the use of advanced techniques to optimize decision variables

ABSTRACT. A new approach for modeling river-aquifer exchange fluxes is proposed which provides the tools to incorporate riverbed heterogeneity and anisotropy. Furthermore, it incorporates the contribution of lateral fluxes through the riverbanks. The method has been successfully applied on an artificial V-catchment model, and on a real case study of an existing gaining lowland river, the Aa River. This method can be used to improve the understanding of riverbed heterogeneity and bank seepage, and their effect on river-aquifer exchange fluxes.

ABSTRACT. An ensemble approach to couple different conceptual hydrological models with a physically based groundwater component in a modelling framework has been developed. It allows to employ the conceptual output of the rainfall runoff model as an input for transient groundwater modelling. The approach also helps to reduce the overparameterization by using the lumped conceptual recharge models with less calibrated model parameters. Furthermore, it enables the possibility to quantify the uncertainty on groundwater model results given by different recharge models.

ABSTRACT. For convenience implementation of new geochemical capabilities into flow and transport codes is often circumvented by coupling them to existing reaction codes. Here we introduce Alquimia, an API and wrapper library that provides a unified interface to the biogeochemistry routines from existing codes. We demonstrate the development with the implementation of interfaces to the geochemical capabilities of PFLOTRAN and CrunchFlow. Next, we demonstrate its use in the addition of geochemical capabilities to a newly develop flow and transport simulator, Amanzi.

ABSTRACT. This presentation introduces the approach used to implement fully unstructured grid discretization into MIN3P-THCm, a process-based numerical model designed for the investigation of subsurface fluid flow and multicomponent reactive transport in variably saturated media. The unstructured grid code, MIN3P-THCm-USG, uses a hybrid vertex-cell-centered control volume method, with options for different control volume discretization approaches, gradient reconstruction methods and cell types. The code’s capability for simulations involving complex geometry has been demonstrated through a series of verification examples.

ABSTRACT. This research aims to extend and apply a recently-developed multiscale universal interface (MUI) to dynamically couple pore-scale and Darcy-scale models (both solved using the lattice Boltzamann method, LBM) to study flow and reactive transport in porous media. The current work will address both scientific questions and multiscale methodology challenges: 1) How does the heterogeneity affect mixing-controlled reactive transport in porous media? and 2) How to best generalize hybrid multiscale methodologies in a high-performance computing environment?

ABSTRACT. Pore-scale heterogeneity in pore geometry and mineral spatial distribution largely control the interactions between minerals and fluids. In order to investigate how pore-scale processes affect the bulk properties of a representative elementary volume (REV), a pore-scale reactive transport model was developed and used to simulate geochemical reactions, flow and transport in heterogeneous pore structures. The simulation results are used to develop a framework for upscaling of rock properties - reaction rate in particular.

ABSTRACT. We propose a discontinuous modeling of chemical kinetics for general multiphase problems. This discontinuous modeling is regularized in order to solve a PDE with a continuous right-hand side. We then apply our work to academic and industrial cases.

ABSTRACT. Soil-moisture processes in the subsurface play a crucial role in the hydrological cycle and the groundwater budget. We will present a conceptual model for coupled single-phase free flow and two-phase porous-medium flow with a detailed description of the models in the free flow and in the porous medium and provide a new coupling concept for modelling coupled porous-medium and free flow.

ABSTRACT. Gas hydrate reservoirs, in the context of a future energy resource, are complex subsurface geo-systems characterized by large material heterogeneities, local anisotropies, and a large number of strongly coupled multi-physics processes, including: 1) hydrate phase change; 2) multi-phase, multi-component flows; 3) thermal effects; 4) geomechanical deformation and plastic failure; and 5) sand migration. In this talk, we present our mathematical model, decoupling strategy, and numerical solution scheme for gas-hydrate geosystems with a focus on gas production and geotechnical risks quantification.

ABSTRACT. An approach for coupled surface surface flow is presented, which is based on a diffusive wave approximation for surface and Richards' equation for subsurface flow. An operator splitting approach based on Discontinuous Galerkin discretizations is used with a coupling via appropriate boundary conditions. The numerical solver is implemented using techniques from the EXA-DUNE project to obtain a high computational efficiency.

ABSTRACT. Classically models coupling mechanical deformation and fluid flow in porous media are stated as PDEs. Recently, for fully saturated porous media, a minimization formulation of Biot's equations has been introduced yielding an alternative to the standard approach. In this work, (1) we discuss the applicability of minimization principles as modeling tool for nonlinear poromechanics, and (2) we exploit the minimization formulation and apply optimization methods in order to solve the coupled problem efficiently in a decoupled fashion.

ABSTRACT. Fluid operations at depth trigger a significant amount of seismicity and aseismic deformation, which could be interpreted as the reactivation of rapid or slow slip on preexisting fracture. Here we present results about the dynamics of slip reactivation on a planar Dieterich-Ruina rate-and-state fault model permeated by a fluid. We show how such models allow to better identify the mechanical parameters controlling the amount and the size of induced earthquakes, and the magnitude of aseismic response.

ABSTRACT. Monte Carlo simulations with a hybrid model of Enhanced Geothermal Systems (EGS) return quickly physics based answers to the question: ``Does continuing the stimulation, worth the additional induced seismicity risk?`` Here, a hybrid model is presented that employs accurate 3D numerical modeling for flow, heat transport, stochastic modeling for seismicity, allowing numerical experiments that study the effect of overpressure on induced seismicity. The proposed tool provides useful insight on long term forecasts of induced seismicity and thermal revenue.

ABSTRACT. Recent observations have shown that changes in pore pressure due to the injection of fluids from underground formations may induce earthquakes. We have developed a numerical framework with fault frictional contact described by the Dieterich–Ruina rate-and-state friction law and we model rock as a poroelastic solid. We investigate the nucleation phase earthquakes. This phase is essential, since nucleation patterns may provide the key to detect preseismic signals and estimate the magnitude of the earthquake.

ABSTRACT. Activation of faults due to human-induced operations is becoming extremely important for risk assessment analysis and reservoir management. The variation of the loading conditions due to injection and/or extraction of fluids may cause the faults re-activation and the opening of new fractures that may generate (micro)seismic events. The aim of this work is at developing a Bayesian framework to quantify and possibly reduce the uncertainties linked to such human activities.

ABSTRACT. Hydraulic stimulation to reactivate and shear-dilate fractures can successfully enhance the permeability of geothermal reservoirs sufficiently to obtain commercial flow rates. The process involves highly coupled hydromechanical processess. We present a hybrid approach for modelling where fractures are modelled as surfaces with associated apertures in a three-dimensional domain.

Results show how the methodology can be applied to understand important mechanisms affecting permeability enhancement as well as induced seismicity. We also discuss its potential in combination with field observation data.

ABSTRACT. We introduce a microfluidic model to image and quantify the role of cell motility on bacterial dispersion and residence time. Using the bacterium Bacillus subtilis, we observe that chemotaxis in highly disordered and confined physico-chemical environment affords bacteria an increase in their ability to persistently occupy the host medium. Our findings illustrate the interplay between bacterial behavior and pore-scale disorder in fluid velocity that impacts the residence time and transport of biota in the subsurface.

ABSTRACT. The soil-bacteria interaction is fundamental for many terrestrial ecosystems processes. Some of these bacteria have the ability to uptake and express exogenous free DNA, a mechanism of gene transfer known as natural transformation. We used microfluidics devices combined with fluorescent optical micoscopy to observe the effect of different flow conditions on this mechanism. Preliminary experiments ran with engineered bacteria show that this method is very promising to study this process source of biodiversity with bacteria species from the environment.

ABSTRACT. We present field observations providing new insights on the role of fractures in the formation of a hot spot of microbial activity. We have analyzed the conditions for the formation of a microbial mat composed of iron-oxidizing bacteria at 60 meters depth in the Ploemeur fractured rock observatory. These results show that heterogeneous flowpaths, linked to the structure of fractured media, create mixing zones between iron rich water and oxygen rich water, at the origin of the microbial hot spot.

ABSTRACT. We assess the complex coupling of biological, chemical and physical processes at different scales with the help of a mechanistic modelling approach. It combines a discrete cellular automata/PDE model on the microscale to account for restructuring of soil microaggregates under the influence of geochemistry, transport and attractive forces by charges or glueing agents as EPS stemming from microorganisms. Effective parameters of interest can then be upscaled to the macroscale. Simulations complement aggregation experiments for goethite and illite with EPS.

ABSTRACT. In this study, we focus in analyzing the influence of operating conditions on the competition between 3 aggregate morphotypes in biofilm reactors: biofilms, flocs and streamers. The main objectives are: (i) to improve understanding of the mechanisms controlling the competition between the three microbial morphotypes and (ii) to analyze the dynamics of streamers formation.

ABSTRACT. Groundwater-to-atmosphere simulations are useful for closing the terrestrial water and cycle at the continental scale and interrogating anthropogenic impacts at different space and time scales. We want to understand how human water use, in this case groundwater pumping and irrigation, changes the natural terrestrial cycles over the European continent including local effects, such as changes in water table depths, evapotranspiration, and air temperature, and non-local effects, such as base flow, continental discharge and precipitation. In particular, we study whether these changes are systematic in space and time, and ultimately impact and potentially redistribute water resources across the continent. We present technical aspects of our work related to model coupling and high-performance computing technologies, and results illustrating the significant impact of human water use beyond individual watersheds.

ABSTRACT. Groundwater storage and evapotranspiration, which are important components of the water cycle and critical to national, regional and local resource management at ranging scales, are both notoriously difficult to observe. Here we present the results of an integrated, physically-based hydrologic model that spans the entire contiguous United States (CONUS) at hyper-resolution. We discuss a modern-day scenario diagnosing groundwater controls on surface fluxes and energy, and demonstrate applications of large scale physically based modeling to bridge observation and decision-making scale gaps.

ABSTRACT. Standard water management practices neglect the complex dynamics between surface and subsurface waters, making the prediction of groundwater storage in water-limited areas such as the western US highly uncertain. Although a large-scale terrestrial water balance is possible via remote sensing techniques such as GRACE, it lacks the ability to separate groundwater from surface water. This work aims to isolate groundwater from such datasets utilizing a high-resolution integrated hydrologic model by explicitly simulating individual components of a representative Sierra Nevada watershed.

ABSTRACT. The French AquiFR modeling platform gathers hydrogeological models coupled with the SURFEX land surface model. The objective is to provide forecasts of the groundwater resources for operational and climate applications. We present here the evaluation of the 1958-2017 hydrogeological reanalysis for multiple regional multilayer aquifers. The evaluation is made against selected observations in order to identify the potential to capture extreme events in long-term simulations. This reanalysis will allow to dispose of a referential state needed for operational applications

ABSTRACT. Between 2013 and 2015 California, and the already water stressed Central Valley, experienced the worst drought in its recorded history. We used the fully integrated hydrologic model, Parflow, coupled with CLM to evaluate the combined impacts of drought and groundwater extraction using multiple groundwater extraction and management scenarios. Model results were evaluated using a combination of observations and GRACE groundwater storage estimates. With this approach, we provide a method to inform sustainable groundwater use and management during severe climatic conditions.

ABSTRACT. Integrated hydrological modelling has become more and more important to bring sufficient insight in a climate change effects investigation on hydrological cycle. Nevertheless, a quantitative evaluation of the uncertainty is a challenging issue due to the long run times. As a first step, we propose to consider a pure groundwater model to identified optimal parameter estimation strategies. This will serve to save computation time and bring accessible the uncertainty analysis on the later integrated hydrological model.

ABSTRACT. Humans are having dramatic effects on the water cycle with irrigated agricultural practices. Using a process-based integrated hydrologic model, we simulate the effects of historical changes in climate and land use on streamflows and groundwater storage along with scenarios of projected climate change. We also simulate feedbacks of irrigation on climate in a WRF regional climate model, which shows that irrigation across the continental United States have significant effects on precipitation and temperatures in the eastern US.

ABSTRACT. We develop a hybrid multiscale method for mixing-controlled reactions and biofilm growth. Non-overlapping regions at pore and continuum spatial scales are coupled by a suitable method that ensures continuity of flux across the interface. Thus, regions of high reactivity where flow alteration occurs are resolved at the pore scale for accuracy while regions of low reactivity are resolved at the continuum scale for efficiency. We present results for simplified two-dimensional scenarios with transverse mixing of electron donors and acceptors.

ABSTRACT. Understanding fracture evolution is essential in many subsurface applications. Conceptually, fractured systems have been represented as being composed of fast flow paths -the fractures- and slow flow paths - the rock matrix. While reactive transport in the fractures can be better simulated with a pore scale model, a Darcy-scale continuum model may be sufficient for the matrix. Here we extend a previously developed pore-scale model based on an adaptive embedded-boundary approach to add a Darcy-scale representation of reactive porous media.

ABSTRACT. Flow through fractured media in the sub-surface is heavily dependent on microstructural information. Continuum models often eliminate features critical to accurately predicting macroscale behavior but are commonly used since resolving thousands of fractures individually is computationally intractable. We overcome this hurdle by developing compact graph representations of the fracture networks, and using machine learning algorithms to mimic the detailed physics at the microscale. The resulting workflow has been shown to achieve upto 4x computational speedup while maintaining accuracy.

ABSTRACT. We developed a method through which we can model coupled fluid-porous media dynamics such as clay swelling and fluid-induced clay transport within Digital Rock Physics. Fluid flow in these scenarios is modeled through the Darcy-Brinkman Equation, which approximates Navier-Stokes in a free fluid, and Darcy’s Law in a microporous (clay) domain. Coupling with geomechanics is achieved by incorporating porous media conservation equations (with appropriate momentum sources) into this framework. This model’s application can go beyond clay and cap rock modeling.

ABSTRACT. Accurate modeling of Moving Contact Line (MCLs) at low Capillary numbers requires honoring the interfacial jump conditions and capturing the multiscale physics. We use a level-set coupled with Navier-Stokes equations (LSNS). The LSNS model captures MCLs in the presence of a thin film that covers the solid surface. We use a low-dimensional film model with disjoining pressure. Several validation cases are presented. We also discuss the effect of roughness on film stability at low Capillary numbers.

ABSTRACT. The present work is based on the study of concentration-flow relationship from high frequency measurements, made in ORACLE Observatory (Paris, France) using a mixture equation, which is based on the separation of the hydrographs (runoff, subs-surface and groundwater). Thanks to the high frequency (measurements made every 30 minutes) the C-Q ratio is much more accurate than those made from point measurements (e.g., each week or month), allowing us to better understand the hydrogeochemical processes of the catchment.

ABSTRACT. Critical zone is a complex system where water cycle occurs. Understanding at several scales hydro(geo)logical processes will allow a sustainable water management. Here, we assess the informative content of pluridisciplinary measured data on the Ploemeur observatory (France). From several years of measured hydraulic heads, deformations, concentrations and discharges, we refine our understanding. First we model piezometry, then, other components have been added. Finally, bringing back parameters set exploration to the theoretical framework permits to extract information about processes.

ABSTRACT. Quantifying water and element fluxes, mixing of different water bodies and the impact of ecological processes is essential to understand the ecosystem dynamics and resilience. Here, we show how high spatial resolution of dissolved gas data can decipher the spatial heterogeneity of water sources and biogeochemical processes. These innovative experiments implemented on a classified coastal wetland downstream a hydrogeological observatory (H+ Ploemeur) allowed to localise and quantify groundwater discharges and biogeochemical reactivity hotspots associated with a deep concentration of cyanobacteria.

ABSTRACT. The West-African critical zone is undergoing strong climatic (e.g. the observed intensification of extreme precipitation) and anthropogenic changes (e.g. +3% of demographic rate, inducing rapid land use changes) which already affect coupled hydrological systems as surface water/groundwater. Managing large hydrosystems call for integrated models to analyze present changes and anticipate future ones. We construct a high resolution (1 km²) regional scale physically-based model using ParFlow-CLM that is evaluated using AMMA-CATCH observation database which covers 3 supersites with contrasted environments.

ABSTRACT. In seasonal flooding wetlands,the strong coupling between vegetation and biological, hydrological, meteorological & physicochemical elements leads to important feedbacks between ecosystem processes and environmental changes.In this research, a typical flooding process was captured in one typical freshwater wetlands in China. The ecohydrological response of a continuous vegetation transect in wetlands was investigated during a typical flooding process in 2017. Simulated Results show distinct difference among plant types and are concordant with the in-situ monitoring data in the transect.

ABSTRACT. We develop novel watershed-characterization methodology to quantify complex watershed systems across scales, using advanced sensing, inversion, and machine learning approaches. We integrate multi-scale multitype datasets of surface geophysics (e.g. electrical, seismic), airborne electromagnetic survey, airborne LiDAR and satellite/UAV images. By characterizing heterogeneous properties over the watershed, we aim to develop the new ‘Digital Watershed’ concept for model parameterization and validation of hydrological and biogeochemical simulations.

ABSTRACT. Mineral dissolution and growth rate variability is one of the most important unresolved problems in modern geochemistry. We approach this problem by using Kinetic Monte Carlo (KMC) techniques as a hypothesis testing tool. We developed KMC models that allow to vary the crystal surface reactivity at will. Comparison of KMC results with experimental data provide insight into reaction mechanisms of phyllosilicates, quartz and calcite dissolution and identify sources of rate variability.

ABSTRACT. Experimental samples of basaltic composition were created in laboratory conditions at various temperatures. This allowed in-situ observations of crystals using X-ray tomography beam lines at Spring-8. Analyses of crystals properties in the synthetic magma were rendered possible through automatic processing using software for CT treatment. Crystals were modeled in 3D using the numerical platform 3DEXPERIENCE© developed by Dassault Systèmes (DS). The 3D imaging facilitated the study of crystals shapes, texture and interaction between mineral species and magma.

ABSTRACT. A common theme in fundamental examinations of adsorption, wetting, nanoparticle aggregation, and nanofluidics is the difficulty of accounting for both short- and long-range interactions. In particular, computational models largely rely on mean-field theories that neglect short-range interactions and treat the fluid as a viscous dielectric continuum. Here, we discuss how molecular dynamics simulations of mineral-water interfaces, mineral-water-CO2 wetting, and nanoparticle interaction in liquid water can provide insight into the relative importance of short- and long-range interactions at mineral-water interfaces.

ABSTRACT. A microfluidic cell was designed to investigate kinetically controlled dissolution/precipitation reactions in confined spaces. Strontium chloride solution and sodium sulphate solutions were allowed to diffuse across of the cell resulting in the precipitation of celestine. The growth of the celestine crystals were monitored by optical microscopy. Pore scale modelling based on Lattice Boltzmann method was used to estimate concentrations and fluid velocities at the fluid-solid growing interface and to gain an insight of the underlying observed processes.

ABSTRACT. In this presentation we will highlight the complex interplays of mineralogical, chemical and microstructural characteristics of clay materials that are ultimately responsible for a remarkable array of macro-scale properties such as specific adsorption, high swelling pressure, semi-permeable membrane properties, and non-Fickian diffusional behavior.

ABSTRACT. Upscaling mineral dissolution kinetics experiments to geological timescales is complicated by the observation that mineral dissolution rates decrease with time. Here, we related time-resolved wollastonite dissolution rate to the physicochemical evolution of the amorphous silica-rich layers which developed on each surface. Some faces were quickly passivated, because of progressive densification of the surface layer, which limits the transport of reactive species at the fluid-mineral interface. Accounting for these gradual changes appears essential to better reproduce temporal decline ofmineral weathering rate.

ABSTRACT. Probabilistic inversion within a multiple-point statistics framework can be intractable for high-dimensional problems. We therefore introduce a new training-image based simulation and inversion approach for complex geologic media. It relies on a deep, spatial generative adversarial network (SGAN) which defines a (very) low-dimensional parameterization. This allows for efficient probabilistic (or deterministic) inversion using state-of-the-art Markov chain Monte Carlo (MCMC) methods. Synthetic case studies involving 2D steady-state flow and 3D transient hydraulic tomography are used to demonstrate our proposed SGAN-based inversion.

ABSTRACT. Remediation predication under uncertainty for a case study of a contaminated site in Israel is considered. A Monte Carlo approach is adopted and various indicator realizations of the aquifer structure were generated. Uncertainty regarding the plume distribution prior to remediation was evaluated using a forward predictive approach, i.e. reconstructing the plume shape and hydraulic head measurements. Remediation simulations showed that generating a reliable initial plume reduced the uncertainty in estimating remediation efficiency by more than 35%.

ABSTRACT. We present a general and flexible Bayesian approach using uncertainty multipliers to simultaneously analyse the input and parameter uncertainty of a fully distributed groundwater flow model with consideration of the heteroscedasticity of the groundwater level error. New heteroscedastic error model for groundwater level included to consider the heteroscedasticity. The results confirm that input uncertainty is the dominant source of uncertainty in the groundwater flow prediction and use of groundwater recharge and abstraction multipliers leads to physically more realistic results.

ABSTRACT. The posterior population expansion (PoPEx) is a new technique that was developed to invert discrete parameter fields using a given prior distribution. The main characteristics of this approach is that it allows dealing with categorical variables. Solving an inverse problem in such discrete cases usually is very difficult time consuming. The main advantage of PoPEx is that it is more effective than existing Markov Chain Monte-Carlo approaches while it estimates accurately the level of uncertainty.

ABSTRACT. Integrated models are increasingly run at catchment or larger scales as the demand for regional models increases. Computational resources for these complex models limits traditional parameter calibrations. Here we present a suite of sensitivity analyses in a 255km2 mountainous domain that varies hydraulic conductivity and mannings n over orders of magnitude at 1km and 100m resolution. We develop a methodology to downscale/upscale K and n in complex domains by incorporating slope into effective K and n parameters.

ABSTRACT. A question that should be addressed at the early stage of model development: What is the optimum level of complexity for the purpose of the study? The complexity cursor should be adjusted for the model to be sufficiently faithful to the processes governing the predictions of interest, but sufficiently simple for model analysis methods to remain practical. The approach will be illustrated with practical case-studies. These examples highlight the benefits of choosing the relevant level of model complexity.

ABSTRACT. In the Aconcagua upper basin some thermal springs outflow spatially related to the Pocuro Fault. The outcrops related to this fault show the presence of several hydrothermal minerals filling fractures suggesting a hotter fluid circulation in the past. To better understand the groundwater circulation and fluid evolution in this system we have used Reactive Transport Models. Three different scenarios have been run by means of Crunchflow software the mineralogical, hydrogeological and geochemical set was implemented to reproduce the field observations.

ABSTRACT. We propose a new and efficient approach to model and calibrate shallow groundwater flow at regional scale and its contribution to water cycle. The hillslope is set as the basic element, where flow is resolved in 1D under the Boussinesq approximation. The model is validated on baseflow analytical solutions. We also show that baseflow analysis offer the opportunity to represent heterogeneity and calibrate the model at basin scale with river discharge data.

ABSTRACT. Studying both past and future long term evolutions of the groundwater resource helps pointing out how extreme could be the future. Past evolution of the groundwater resource are difficult to assess due to the short observation period. Similar methods are used to treat past and future : climate modeling, downscaling method, and hydrogeological modeling. However, observation availability in the past help driving and assessing the XXth groundwater evolution. Results obtained with the Aqui-FR modeling in France will be shown

ABSTRACT. The use of foams for subsurface remediation takes advantage of their peculiar flow properties, which we study using a two-dimensional setup in which the foam structure and bubble velocities are monitored. We uncover a rich phenomenology, where larger bubbles follow preferential paths of higher velocities, the flow is locally intermittent under stationary forcing, and bubble fragmentation leads to an irreversible evolution of the foam's bubble size distribution. We decipher the theoretical link between that evolution and individual bubble fragmentation events.

ABSTRACT. We have investigated how spatial variability in rainfall within a given catchment, as for example seen in mountain environments affected by orographically-controlled precipitation, affects the relationship between the characteristics of rainfall variability and those of the water discharge in the river. For this, we have applied to data from Taiwan a model developed by Doulatyari et al. (2017) that uses properties of the surface network and topography to relate spatial variability in rainfall to spatial variability in discharge.

ABSTRACT. In this paper, we model the flow of nonreactive tracers in karst reservoirs by coupling the continuity equation, the Brinkman’s equation to the Advection-Diffusion-Adsorption equation. The flow of fluid is modelled as single-phase, slightly-compressible flow by solving the continuity equation and the Brinkman’s equation simultaneously to obtain the pressure and velocity distribution in the entire reservoir. Using the computed velocity distribution, the Advection-Diffusion-Adsorption equation is then solved numerically to model the transport of the tracers through the reservoir.

ABSTRACT. Resolving the nitrogen sources in river discharge and the “hot moments” of nitrogen transformations at river catchment is crucial in attempts to reduce nitrogen export. Transient time information can be utilized to find these „hot moments“. We use particle tracking of stable water isotopes within a hydrological model to retrieve water travel times and use the information to analyze nitrogen transport and to spot relevant times of nitrate formation within a mixed land use catchment (10 km²).

ABSTRACT. Commonly, Darcy scale models are adopted to investigate the removal of gaseous waste products within electrolysis cells. In contrast, we model the two-phase flow within cell channels on the bubble scale. To this end, direct numerical simulations are conducted by adopting smoothed particles dynamics. Results concerning limitations and parametrizations of Darcian scale models will be presented. Moreover, channel flow properties are detected which affect the channels’s gas uptake from porous cell components.

ABSTRACT. We present a mathematical model of hydro-mechanical interaction based on the Biot system in a fractured rock. The fracture can be void or filled by an elastic material. We perform dimension reduction of the equations in the fracture and obtain averaged flow and elasticity equations coupled with the equations in the rock through interface conditions. For nearly incompressible materials a mixed formulation of the linear elasticity is used. We further study stability of finite element formulations and of iterative splittings.

ABSTRACT. The mouse brain vascular system is composed of a space-filling capillary network connected to branched quasi-fractal arterioles and venules. We investigate the distribution and correlation properties of blood flow velocities from numerical simulations in this system. Flow is solved from a 1D non-linear model taking account of the complex rheological properties of blood flow.

The network structural complexity imparts broad and spatially correlated Lagrangian velocity distributions, leading to power law transit time distributions. The origins of this behavior are studied.

ABSTRACT. We present a new flexibly designed, fully distributed nitrate transport and removal model (mHM-Nitrate) at catchment scale. The model was developed mainly based on the mesoscale Hydrological Model (mHM) and the Hydrological Predictions for the Environment (HYPE) model. The mHM-Nitrate model was tested in the Selke catchment (Central Germany), which is characterized by heterogeneous physiographic and land-use conditions, using adequate observed hydrological and nitrate data at three nested gauging stations. Long term (1997-2015) daily simulations showed that the model well reproduced the seasonal dynamics of biweekly nitrate observations in forested, agricultural and urban areas. High-frequency measurements (2010-2015) were additionally used to validate model performance of simulating short-term changes in stream-water concentrations that reflect changes in runoff partitioning and event-based dilution effects.

ABSTRACT. We present the results of the hydrogeochemical modeling of the geochemical compositions of spring and borehole waters from the Ringelbach catchment (Vosges Mountains, France). The simulations, using the hydrogeochemical code KIRMAT, were performed through two different water pathways in different types of rocks and make it possible to account for not only the geochemical differences between the spring and borehole waters but also the geochemical variations observed in waters in both contexts.

ABSTRACT. A pumping test conducted in a faulted formation provides consistent scalings of transmissivity and storativity over a few hundred meters. Drawdown analysis shows persistent reductions of transmissivity in two structures. The linear dependency of transmissivity to storativity indicates a fractured damage zone while the cubic dependency identifies well-defined faults. The latter demonstrates the validity of Poiseuille flow at a rarely investigated scale. Well-designed pumping tests and scaling analysis thus provide essential information on flow-bearing structures for site characterization and modeling.

ABSTRACT. The increased intensity and severity of droughts within the Amazon Basin region has emphasized questions of vulnerability and resilience of tropical forests to water limitation. In this work, we utilize a three-dimensional model of root water uptake to explore and quantify the contributions of tree root systems to individual and community drought resilience. The model is informed by data collected during the 2015-2016 El Niño drought period from the Tapajós National Forest, Pará, Brazil.

ABSTRACT. Already groundwater abstraction often exceeds groundwater recharge, leading to groundwater depletion. Here we explore the limits to global groundwater consumption for the first time by evaluating effects of head declines on natural groundwater baseflows and groundwater accessibility. We show that negative effects on groundwater baseflows are often experiences first and also for regions not know for significant depletions or head declines. Regionally the increasing level of groundwater stress will be an important factor in future economic development and socio-economic environment.

ABSTRACT. We report the use of 1 and 2D reactive transport simulations to quantify the development of weathering profiles in a dual flow domain fractured argillaceous bedrock. The models uniquely quantify weathering patterns across a variably saturated hillslope within the Eel River Critical Zone Observatory in Mendocino County, California. The validity of the models is demonstrated by comparison to novel direct measurements of solute concentrations from aqueous samples extracted throughout the fractured vadose zone.

ABSTRACT. We investigate how fracture wall roughness impacts the flow of a shear-thinning fluid. Numerical simulations of flow in 3D geological fractures are carried out by solving a modified Navier-Stokes equation incorporating the Carreau viscous-shear model. The objective is to investigate how varying the correlation length impacts the flow behavior, for different degrees of closure, and how this behavior diverges from Newtonian fluids. The results from the 3D simulations are also compared to 2D simulations based on the lubrication theory.

ABSTRACT. Understanding the effects of the viscosity of fluids on fault mechanics can shed light on the seismicity in engineering reservoirs induced by the injection of pressurized fluids. Here, we investigate precursory patterns and we explore general stability criteria for earthquake nucleation in presence or not of viscous fluids, using a rotary shear apparatus in laboratory experiments.

ABSTRACT. The local rate of stretching of an imaginary filament in a heterogeneous flow is linked to the rate of dilution of a conservative scalar. By linking the rate of dilution to the reactivity for fast reactions, we arrive at a method to understand how the heterogeneous structure affects locally the reaction rates. While yielding identical particle trajectories, the residual divergence, small yet finite, causes significant deviations in the rate of stretching, and thus, the prediction of the reactivity.

ABSTRACT. Continental scale hydrology models at high spatial resolution are increasing in application. We compare models developed for the continental US: ParFlow-CONUS using the integrated model ParFlow and a configuration of the National Water Model using the National Center for Atmospheric Research, Weather Research and Forecasting hydro extension package. Accurately representing large domains remains a challenge considering the difficult task of representing complex hydrologic processes, computational expense, and extensive data needs. Intercomparing both models helps disentangle process, parameter, and formulation differences.

ABSTRACT. Mineral dissolution rates measured in natural systems have long been observed to be lower than those measured in well-mixed reactors by 3–6 orders of magnitude. This work develops a general rate law for magnesite dissolution in heterogeneous media under variable flow velocity and length scale conditions through the connections between rate constants at well-mixed reactors, spatial heterogeneity, and timescales in both reactive zones and the whole domain.

ABSTRACT. The scope of this study is to develop a forecast hydrological model at the field scale, able to simulate water level and soil tension in the roots zone for water table control operations. In this work, the finite element CATHY (CATchment Hydrology) model associated with sequential data assimilation with an ensemble Kalman filter (EnKF) method will be used to simulated the soil water dynamics and perform model calibration in real-time. The study is conducted in cranberry fields located in Canada.

ABSTRACT. Production and transport of oil drops in micro as well as millimetric channel is studied. Two simple model porous media were considered. First, made with a capillary tube with a controlled constriction in the middle, modeling pore - throat system. The second, is a bifurcation, which models local distribution at pore scales. Both systems were are wetted by oil. The size and distribution, splitting and coalescence of drops were studied as a function of water and oil flow rates.

ABSTRACT. The evolution of groundwater system geochemistry in a fractured Canadian Shield crystalline rock setting is explored through a series of numerical experiments in which both model hydraulic and transport properties are varied. The degree of importance of both matrix and fracture zone hydraulic properties to the geochemical evolution of the groundwater system is demonstrated through the use of performance measures such as groundwater age, mean life expectancy, and salinity distributions in both the rock mass and discrete fracture zone network.

ABSTRACT. Contaminant migration through clays mainly occurs by diffusion and is affected by sorption. Experimental diffusion data often shows that cation diffusion coefficients are higher than diffusion coefficients of water tracers. In order to account for that discrepancy a surface diffusion model is set up in the reactive transport code FLOTRAN. The model accounts for a partial mobility of sorbed cations. First results for Cs diffusion in Opalinus Clay show that the model is able to describe the experimental data consistently.

ABSTRACT. We aim to provide and integrate field and modelling approaches to overcome the conceptual gap between the hyporheic zone and the river corridor across different spatial and temporal scales. This is indispensable for understanding and quantifying biogeochemical turnover. Numerical models simulating both the hyporheic and stream system will allow quantifying exchange fluxes and processes across the different spatial and temporal scales. These models will be based on a high-resolution LIDAR-UAV streambed and will integrate tracer -and UAV- based data.

ABSTRACT. In this work we estimate the base flow time constant at global scale by using an analytical solution of the Boussinesq equation for a homogeneous, isotropic, linear reservoir that depends on the effective porosity, transmissivity, drainage density and slope. Different literature values of hydrogeological parameters were tested, and the results that are closest to references from river discharge observations were evaluated in the groundwater reservoir of the ORCHIDEE Land surface Model.

ABSTRACT. In chemical weathering environments the balance between reactivity and transport govern the extent of fluid-rock interactions. In such systems, the common method of treating isotope fractionation using irreversible rates fails to capture the dynamic balance between both kinetic and equilibrium fractionation and transient shifts between them as a function of transport. A transient model involving coupled fluid and solid phase isotopic compositions is used to describe silicon isotopic equilibration during amorphous silica precipitation with applicability to interpretation of weathering fluxes.

ABSTRACT. This study was performed to examine the coupled hydro-mechanical process during hydraulic fracturing in a typical shale rock formation, embedding a system of well-behaved, horizontal plate-like natural fractures. The Stimulated Reservoir Volume modelling technique was integrated with the finite element approach to simulate a 3-D poroelastic porous matrix. The computations were done through a quasi-transient simulations for eight different hydro fractures radius. The corresponding numerical method was verified versus the benchmark model of Wendland and Himmelsbach.

ABSTRACT. In geothermal exploration, the potential to locate thermal anomalies is compromised in some “blind” geological contexts (lack of thermal information at depth and of surface manifestations). We propose an approach to help locating thermal anomalies based on a groundwater flow model and a geomechanical model. The approach incorporates a thorough structural analysis to build the models geometry (Discrete Fracture Network) and a cross-analysis of models results to estimate areas where the physical conditions are expected to favor thermal anomalies.

ABSTRACT. Transport properties of rock formations are dictated by the behaviour of fracture networks. Noting that the effect of electrical properties of fracture walls on the flow dynamics remains an open question with potentially significant implications, here, we explore the coupled electro-hydrodynamic transport in the same using a generalized version of lubrication theory assuming specified hydraulic head, electrical potential and concentration across the fracture. The resulting system of equations is solved using an iterative Finite-Volume-Method for realistic aperture fields.

ABSTRACT. This study focus on the impacts of a shallow water table in the climate model of the IPSL. A sensitivity experiment was performed by carry out global simulations with different fractions of each grid-cell occupied by a prescribed saturated zone below 1m. We highlighted a non-linear response of the near surface climate due to the coupling between land and atmosphere models, with strongest variations where shallow water table fractions are weak.

ABSTRACT. The mechanics of cellular systems is complex, in part because the materials themselves are difficult to characterize, and in part because there can be biological feedbacks between biofilm properties and applied stresses. We present a first effort for formally homogenizing the mechanics of such systems using the method of volume averaging. Our starting point is a simplified model containing cells, an extracellular matrix, and a constitutive law describing how cells respond to induced stresses.

ABSTRACT. We present a novel experimental system based on a fully instrumented 3D printed fluidic micro-bioreactor. We demonstrate how, combined with X-ray microtomography, this system can quantify biofilm distribution in pore-space for different growth conditions. 3D printing further allows for porous structures with known properties (porosity, permeability and pore size) to be precisely replicated. This makes it possible to vary one parameter (e.g. flow rate and nutrient concentrations) and to reveal its influence on the biofilm architecture inside the porous network.

ABSTRACT. Solute transport within catchments is controlled by how the critical zone selects water from storage for discharge or evapotranspiration. This study investigates the storage selection of the Southern Sierra Critical Zone Observatory, which we constrained by a novel combination of cosmogenic radioactive and stable isotopes: tritium, sodium-22, sulfur-35 and oxygen-18. The derived water storage and travel times are related to the weathering rates and measured solutes in the catchment.

ABSTRACT. Hard rock aquifers are known as highly compartmentalized structures. The chemical composition of groundwater reflect this compartimentalization including anoxic groundwater with long residence time in the lower section. These deep groundwater have recorded long-term climatic changes. Exploitation induces the propagation of modern pollutants downward and increases mineral dissolution and biogeochemical reactivity such as denitrification. Long term monitoring in the H+ network have shown the limits of anthropogenic influences in these aquifers.

ABSTRACT. To obtain an explicit insight into the soil hydro-thermal consistency for land surface modelling over the Tibetan Plateau (TP), in situ and laboratory measurements of over 40 soil property profiles were obtained across three dominant climate zones (i.e. arid, semi-arid, and semi-humid). Parameterization schemes for porosity, soil hydraulic and thermal properties were investigated. Using the superior parameterization schemes, the uncertainties of five existing soil datasets and their derived soil hydraulic and thermal properties over the TP were quantified.

ABSTRACT. We aim at proposing a new bedload transport model without closure relationship. Our model is deduced from a fluid-like description of the sediment layer. It is obtained by performing simultaneously the Shallow-Water approximation and the diffusive limit in the Navier-Stokes equations. Dierent scalings of the viscosity coecient allow to obtain an equation for the sediment velocity with or without viscous term. Numerical experiments illustrate the novelty of the model.

ABSTRACT. A high resolution data set on discharge, precipitation, suspended sediment concentration as well as quantified sediment source contributions obtained with fingerprinting in two-meso scale Mediterranean catchments is used in combination with a distributed numerical model in order to assess the impact of connectivity on suspended sediment fluxes. The hydro-sedimentary model is set up in Iber, a 2-D hydrodynamic model that solves the St. Venant equations and is completed with a soil infiltration module and an erosion model.

ABSTRACT. The timescales and mechanisms by which rivers evacuate landslide deposits are poorly known, but are critical for predicting geomorphic hazards and determining the role of extreme events on landscape evolution. Here, we use a numerical model of river evolution accounting for the morphodynamic feedbacks between river transport capacity and changes in river geometry, initially perturbed by the potentially large landslide deposits. Our simulations evidence the previously unrecognized role of river dynamic narrowing in significantly accelerating the removal of landslide deposits.

ABSTRACT. The Multilayer-HySEA model including non-hydrostatic effects has been used to perform all the benchmarking problems dealing with laboratory experiments proposed in the workshop that was organized at Texas A&M University - Galveston, on January 9-11, 2017 by NTHMP. The aim of this presentation is to show some of the latest numerical results obtained with the Multilayer-HySEA (non-hydrostatic) model in the framework of this validation effort.

ABSTRACT. Topo-bathymetric lidar can now map water surface and river bathymetry at very high accuracy (5-10 cm), resolution (> 10 pts/m²) and over large areas (> 10 km). We use a reduced complexity 2D flood model and knowledge of the discharge to invert for spatial patterns of the Manning friction coefficient in various rivers. Systematic spatial variations are observed. The relationship to bed lidar roughness is explored to evaluate if bed friction can be computed from topo-bathymetric lidar data.

ABSTRACT. The Adaptive Hydraulics (AdH) software suite is a state-of-the-art modeling system developed by ERDC, USACE. AdH is an adaptive and implicit finite element suite containing models for 2D and 3D shallow water (SW) dynamics and salt, sediment and general transport. This presentation will focus on the formulation and implementation of the AdH SW 3D stand alone baroclinic model and 2D/3D SW coupling. Validation results will be shown for both stand-alone and coupled results.

ABSTRACT. Groundwater table behaviour in shallow aquifers is often more dynamic than one would expect from monthly groundwater level observations. Moreover, spatial patterns of groundwater flow can vary considerably throughout the seasons, and may even be influenced by single rainfall and corresponding recharge events. We investigate the typical temporal and spatial groundwater table behaviour at the ONDRAF/NIRAS Mol/Dessel site, and quantify the daily spatial groundwater table dynamics over a period of 15 years, based on monthly time series.

ABSTRACT. In this work, the infiltration of chlorinated solvents underground is modelled using a dual porosity single-permeability multiphase transport formulation. The model considers a fractured, clay-rich vadose zone overlying a sandy aquifer. The results predict that once the infiltration is stopped, most contaminant is already settled down at the bottom of aquifer, remaining mainly at the dead-end of fractures in the vadose zone. At longer times, back-diffusion drives the release of contaminant to the aquifer.

ABSTRACT. An analytical solution for interpreting dipole-tests in heterogeneous media is presented. It is shown how the solution can be used for interpreting field tests in view of geostatistical aquifer characterization on some illustrative examples. The analytical solution for the tracer breakthrough curve at the pumping well is developed by considering a perfectly stratified formation and associating the shape of the BTC with the log-conductivity variance. Solutions are presented for resident/flux proportional injection and for instantaneous/continuous solute injections.

ABSTRACT. In this study we use the integrated surface subsurface model HydroGeoSphere for simulating the generation of surface saturation within a small (45 ha) forested catchment. We compare the simulated surface saturation patterns and dynamics of several locations distributed over the catchment with repeated field mappings of surface saturation at these locations. The comparison demonstrates the potential of the model to reproduce the spatial and temporal variability of surface saturation and helps to identify the underlying surface saturation generation processes.

ABSTRACT. Integrated hydrological modeling systems simulate complex interactions between groundwater, surface water flows, vegetation and atmosphere under heterogeneous conditions. Therefore, the developed codes are often not easy portable and difficult to be handled. Moreover, the application such models to large domains or long time series requires significant computational efforts. We present our recent software re-engineering efforts of the GEOtop model to create a robust and computationally efficient scientific software package open to the hydrological community, easily usable by researchers and experts.

ABSTRACT. A novel mechanistic model of tropical leaf-phenology, reconciling observations of dry season greening and water stress in Amazonia, is presented. The model accounts for leaf development, synchronized dry season litterfall and the effect of leaf quality and quantity on leaf photosynthetic capacity. Simulation results from 32 sites in the Amazon basin over a 15-year period reveal that leaf phenology explains carbon fluxes seasonality but the impact on evapotranspiration is limited due to compensatory effects of canopy-atmosphere decoupling and LAI dynamics.

ABSTRACT. In this study rainfall-runoff-soil loss relationship were modeled for three prominent vegetation types in tree line zone of Central Himalaya in India. Nine experimental plots (10mx5m) were delineated on three prominent vegetation types, viz; Coniferous forests, Krumholz, and alpine grassland. Using stochastic and soft computing models, an empirical relationship was developed to simulate rainfall-runoff- soil loss and hence to estimate soil erosion from tree line vegetation in central Himalaya; these findings will lead towards devising better conservation and management strategies.

ABSTRACT. The induced seismicity at the Enhanced Geothermal System of Soultz-Sous-Forêts in France has been related to both fluid pressure increase during stimulation and aseismic creeping movements. The fluid-induced seismic events often exhibit a high degree of similarity and the mechanism at the origin of these repeated events is thought to be associated with slow slip processes where asperities rupture several times. We investigate the sources of such events to decipher the link between multiplets and aseismic slip.

ABSTRACT. To understand hydro-mechanical couplings between fluid and fault rocks during earthquake rupture, we performed laboratory earthquakes at conditions representative of Enhanced-Geothermal-Reservoirs (Stress, Fluid Pressure, Lithology). We observe that fluid pressure level controls dynamic fault heating and weakening trough fluid's phase transitions which can act as heat buffers. Our experimental observations, supported by microstructures and simplified models of temperature rise in the fault led to an understanding of the weakening mechanisms during dynamic earthquake rupture in presence of fluids.

ABSTRACT. Large typhoons or earthquakes trigger numerous landslides and produce large volumes of sediments that are exported by rivers in years to decades. This surface unloading could induce stress changes at depth and trigger shallow seismicity. We use the QDYN numerical model to investigate the effects of time-dependent variations of stress rate on faults activity. Our results show a step-like increase of earthquake frequency following a normal stress decrease, which is most significant if river transport occurs within a few years.

ABSTRACT. In this work, we extend some previous analyses of mixing an reaction under transient conditions to the simple case of Taylor dispersion and reaction in a tube. This geometry has provided an archetype for understanding dispersion in porous media for more than 70 years, and here we consider that analysis for the case of reacting fluids. In particular, we examine how the configuration of the initial condition influences the resulting effective rate of reaction.

ABSTRACT. Direct numerical simulations of pore-scale flow and transport in porous media require the knowledge of pore-space topology. Real porous systems, e.g. rocks, pose additional challenges since they usually exhibit multi-modal distributions in physical and chemical properties. These hierarchical media cannot be approached by a single continuum formulation. We discuss both the approximation of sequential homogenization approaches and propose an algorithm of downscaling followed by segmentation to reconstruct the unresolved pore space from XCT images of natural polydisperse geological porous media.

ABSTRACT. Simulating reactive transport at the large scale requires upscaled frameworks where local incomplete mixing must be acknowledged. Departing from continuous time random walk theory, we extend classical Kinetic Monte Carlo theory to account for arbitrary inter-reaction times representing incomplete mixing, providing a generalized Gillespie algorithm. Our approach efficiently simulates generalized Michaelis—Menten kinetics and time-nonlocal kinetics exhibiting weak ergodicity breaking. We introduce a hybrid framework that models transport through a Lagrangian random walk and local-scale reactions through our generalized framework.

ABSTRACT. Fast high volume infiltration (FHVI) is a technique to inject high volume of water into subsurface in a short span of time. During periods of heavy rainfall or dewatering of sites of construction large volume of water is available which can effectively be used in future if stored. The conventional infiltration techniques cannot achieve FHVI. The present study mathematically models the FHVI process. The model results are compared with that of a numerical model and also an experimental field study.

ABSTRACT. Groundwater flooding is a critical issue in relatively flat coastal areas like along the coasts of Normandy (Western France). Starting from field observations, we build a numerical groundwater flow model of a coastal aquifer near Bréville-sur-Mer (Normandy, France). We show that while tidal variations have a limited influence, the mean sea level rise may induce significant groundwater table elevations. Local topographical depressions through their capacity to offer natural drains may significantly alleviate flood vulnerability and present key structures.

ABSTRACT. CO2-induced reactions in fractured mudstone include mineral dissolution/precipitation, clay swelling reactions, and CO2-exsolution promoted mineral precipitation; the progress of these reactions controls fracture opening versus self-sealing behavior. We investigate the controls on fracture opening versus self-sealing for synthetic vertical single rough walled fractures with self-affine fracture surfaces, using 2.5D reactive transport modeling simulations. The impact of mineralogy, aperture distribution, fluid residence time and poro-elastic effects on fracture opening versus self-sealing behavior is investigated for single and two phases flows.

ABSTRACT. To estimate the performance of enhanced geothermal systems (EGS) an accurate characterization of the created fracture networks is crucial. One of the most relevant parameters for heat extractions is the fracture surface area. However, it is also one of the major uncertainties in the characterization of fractured reservoirs. Here, we perform numerical simulations to investigate the influence of heterogeneous aperture distribution on fracture surface area and its implication for the results of tracer tests.

ABSTRACT. A challenge in modelling reactive transport in discrete fractures is how to conceptualize the matrix-fracture interface. We present an approach that solves reactive transport in discrete fractures and the porous matrix independently but couples them through mass transfer terms. Its implementation in iCP (interface Comsol-Phreeqc) is validated against analytical and numerical solutions for flow and transport. We apply it to model calcite precipitation in fractures driven by mixing between granitic waters and pore water from concrete leaching.

ABSTRACT. This work presents a discrete fracture model on the basis of a cell-centered finite volume approach, where the fractures are modeled as lower-dimensional geometries. The scheme allows for an explicit evaluation of matrix-fracture transfer fluxes and is able to represent jumps in solution and fluxes across the fractures. First, we examine the performance of the model for complex flow processes on simple geometries by comparing the results to an equi-dimensional discretization. Subsequently, it is applied to complex fracture networks.

ABSTRACT. Understanding solute transport in fractured rocks is of particular importance in many applications. This paper focused on the dispersion process in rough single fractures under non-Darcian flow conditions. Non-Darcian flow existed in both smooth and rough single fractures and were best described by the Forchheimer equation.The ADE model failed to capture the long-tailing of breakthrough curves (BTCs). Instead, the CTRW model could better explain BTCs in both smooth and rough fractures, especially in capturing the long-tailing feature

ABSTRACT. Natural nutrients fluxes and solute concentrations needed for bacteria metabolism may be highly variable in space and intermittent in time. Reactive hotspots induce biogeochemical reaction kinetics that differ significantly from homogeneous model environments. We develop microfluidic devices to produce single and double concentration gradients, allowing for quantitative monitoring of the bacteria's spatial distribution and early-stage biofilm development. We investigate the influence of chemical and velocity gradients on biogeochemical reactions kinetics, biomass production, and we develop growth models.

ABSTRACT. An integrated 3D catchment-scale groundwater-stream-lake model was developed and calibrated using PEST. The aim was to derive the water budgets and lake water residence times (T) for 13 lakes in two neighboring catchments. Water stable isotopes were used to derive a 18O – T relationship in order to investigate if 18O in lakes can be used to estimate residence time. The results are encouraging, but work best for T < 2-3 years.

ABSTRACT. Subsurface flow patterns exert a significant influence on solute transport and biogeochemical processes for a broad range of scales. Consequences on solute turnover and chemical weathering is usually addressed by residence time distribution approaches. However, understanding streamline spatial patterns impact and the effect of mixing across streamlines on reaction kinetics remains an outstanding challenge. Here, we investigate the impact of surface topography on subsurface streamline patterns and analyse their influence on mixing and reaction processes for different biogeochemical reactions.

ABSTRACT. We explore the dynamics of transit times and subsurface mixing in the small agricultural catchment Schäfertal, Central Germany. We use a 3D fully coupled surface-subsurface hydrological model to simulate water flow, and particle tracking to track flow paths and calculate the transit times of water parcels. The catchment-scale transit time distributions, residence time distributions and fractional StorAge Selection functions are computed and evaluated in terms of the transient, seasonal nature of median transit/residence times and discharge selection preferences.

ABSTRACT. We consider porous media applications that contain evolving solid phases. Deriving models that describe such structural changes at a variety of scales is essential to understanding the intimate linking between structure and function. With reference to the soil's heterogeneity, we aim to develop a mathematical model at the pore scale, perform its upscaling to transfer our model from the small scale to the macroscale and investigate the resulting model numerically.

ABSTRACT. The paramount role of pore fluid pressure during earthquake nucleation and dynamic rupture deserves further comprehension. Earthquake ruptures in poroelastic media encompass complex phenomena, including stick-slip frictional instabilities, thermal processes and hydromechanical couplings. A numerical simulation that couples fault poromechanics, rock poroelasticity and inertial effects is required in order to understand the elastodynamic issues on the rupture of induced earthquakes. Our numerical model includes fully-coupled hydromechanical, frictional and dynamic features; analyses the effects of both poroelasticity and the viscous material properties on the characterization of injection-induced earthquakes. We assume a rate-and state-frictional law for the fault and the Kelvin-Voigt constitutive model for the rock. We simulate the whole earthquake sequence, both the interseismic and dynamic rupture phases, and quantify the differences between the rupture events whether the dynamic approach is considered or not. So we elucidate the key question of whether quasi-dynamic models are able to faithfully reproduce the main features of induced earthquakes obtained in dynamic simulations. The viscoelastic constitutive relation for the solid domain implies a dissipative term that affects the overall response and allows us to simulate the physical process of seismic wave attenuation. Furthermore, viscous dissipation avoids spurious high-frequency oscillations during wave propagation. The inclusion of inertial terms enables the model to account for the incremental fluctuations of both pore pressures and the stress field during dynamic rupture, which may shed light on the mechanisms that control the dynamic triggering on nearby faults.

ABSTRACT. We propose an original algorithm for a compositional flow modeling in porous media with rock-fluid interactions using two different space and time discretization: one mesh, as usual for the pressure equation and a much finer one for the chemical reactions. The interest of this scheme is that the calculation of the flow on the high resolution grid is done solving a local problem on each coarse grid. This step is easily solved on parallel machines. The transport of species are done on the high resolution mesh using an explicit scheme which is also easy to solve on a parallel machine. The time step and the well indexes for the transport on the high resolution mesh is of course adapted to this explicit scheme. Two examples will illustrate this algorithm: one for an injection of CO2 in a carbonate reservoir and the second one will address reactions during the injection of an acidic water for improving well injectivity again in a carbonate reservoir.

ABSTRACT. A gridded water budget model was developed to compute the groundwater recharge with a resolution of 8 km and a daily time step. Three different water budget methods were included to assess the uncertainty associated to the effective rainfall parameterization method. We then proposed a global solution to split the effective rainfall between runoff and infiltration, relying on a GIS built parameter called IDPR. A linear relationship between the BFI and the spatial average IDPR index was proved for the sedimentary basins. It thus allowed calibrating the IDPR index in terms of infiltration capacity. Finally, groundwater direct recharge could be calculated and annual average recharge maps were generated.

ABSTRACT. We study statistical properties of a Bingham fluid in macroscopic porous media generated with a log-normal distribution. We find multi-scale properties characterised by exponents. We also find an unexpected flow rate as a function of the relative pressure. We explore the exponents and give an explanation for this specific flow regime.