Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement...

28
Programme of the workshop Interfaces and interface displacement in unsaturated porous media February 2-4 in Lauterbad

Transcript of Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement...

Page 1: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

Programme of the workshop

Interfaces and interface displacement in unsaturated porous media

February 2-4 in Lauterbad

Page 2: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

Wednesday, February 2

Time Speaker Title 08:30-08.45 Welcome

Pinning and Interfacial jumps (chair P. Lehmann) 08.45-09.30 Knut Jørgen Måløy

(keynote talk) Pattern formation by invasion in particle suspension

09.30-09.50 MUSIS: Franziska Moebius Linking pore scale liquid displacement dynamics with acoustic emissions during rapid interfacial jumps in porous media

09.50-10.10 Philippe Beltrame Thin liquid layers flows on a heterogeneous substrate: a dynamical systems approach

10.10-10.30 Holger Steeb The role of fluid-fluid interfaces in wave propagation phenomena in partially-saturated porous media

10.30-11.00 Coffee break Role of surfaces, layers and gaps on flow and trans port (chair J. Vanderborght)

11.00-11.45 Sjoerd van der Zee (keynote talk)

Modeling surface runoff with infiltration for flat areas

11.45-12.05 Andrea Carminati Water dynamics at the root-soil interface

12.05-12.25 Jens Eller XTM visualization of water condensation and evaporation in porous gas diffusion layers of polymer electrolyte fuel cells

12.25-14.00 Lunch break

Page 3: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

Modeling interfacial dynamics at pore scale (chair: M. Krafczyk)

14.00-14.45 Jonas Toelke (keynote talk)

Digital rock physics for the prediction of capillary pressure and relative permeabilities

14.45-15.05 Mostaghimi Peyman Permeability Estimates on micro-CT Images and Descriptions of Rock Heterogeneity

15.05-15.25 Felix Heimann Intricacies and impact of interface regularization in micro scale simulations of multi phase flow

15.25-15.45 MUSIS: Ying Wang A Lattice Boltzmann multiphase approach for pore-scale simulation of evaporation and infiltration in unsaturated soil

15.45-16.45 Poster session & Coffee break 16.45-17.30 Paul Meakin

(keynote talk) Pore scale simulation of dynamic interfaces in subsurface systems

Discussion (chair: Dani Or) 17.30-18.45 Discussion on ‘fluid interfacial displacements - challenges’ 19.00-20.00 Dinner

Interface between soil and atmosphere (chair: J. Va nderborght) 20.00-20.20 Peter Lehmann Surface-atmosphere coupling

determines evaporation dynamics from a drying porous surface

20.20-20.40 MUSIS: Klaus Mosthaf

Modeling Evaporation Influenced by Wind and Radiation

20.40-21.00 MUSIS: Marin Dimitrov Investigation of the soil moisture, soil and brightness temperature of bare soil plots with different surface structure using remote sensing and ground based measurements

Page 4: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

Thursday, February 3

Interfaces and flow instability I (chair: H.J. Voge l) Time Speaker Title 08.30-09.15 Wolfgang Durner

(keynote talk) Dynamic nonequilibrium in water flow in porous media - processes, experimental evidence, and effective modelling approaches

09.15-09.35 Milena Cislerova/Michael Snehota

Investigating the flow instabilities in soil during quasi steady state flow non-invasive methods

09.35-09.55 Horst Gerke Macroscopic representation of the interface between flow domains in a dual permeability model describing preferential flow in structured soil

09.55-10.15 Peter Germann Scrutiny of hydraulic conductivity provides scale of permeable media flow

10.15-10.45 Coffee break Interfaces and flow instability II (chair: H.J. Vog el)

10.45-11.30 Ruben Juanes (keynote talk)

Increased deep drainage in arid environments as a result of gravity fingering during infiltration

11.30-11.50 Frederike Kissling Numerical simulation of saturation overshoots in porous media with a heterogeneous multiscale method

11.50-12.10 Rony Wallach Saturation overshoot behind a 2D wetting front – observation and theory

12.10-12.30 MUSIS: Cindi Schütz Spatially Averaged Water Content in Heterogeneous Porous Media

12.30-14.00 Lunch break Can I use (trust) the model? (chair: P. Lehmann)

14.00-14.45 Mathias Schmidt (keynote talk)

Some thoughts on modeling and simulation in an industrial application context

Non-equilibrium in capillary interfaces (chair: R. Helmig) 14.45-15.05 Vahid Joekar-Niasar Non-equilibrium effects in capillary

interfaces in porous media 15.05-15.25 Paul Zegeling Adaptive grids for non-monotonic

waves in an extended BuckleyLeverett model

15.25-15.45 MUSIS: Ulli Weller Trajectories in theta - psi space 15.45-16.45 Poster session & Coffee break 16.45-17.30 Renaud Toussaint

(keynote talk) Channel formation during biphasic flow in porous media, or during fast monophasic flow and matrix hydrofracture: Consequences on the dynamic pressure-saturation and relative permeability relations

Discussion (chair: R. Helmig) 17.30-18.45 Discussion on ‘flow across material interfaces’ 19.00 Dinner 20.00-open end Posters

Page 5: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

Friday, February 4

Interface dynamics at continuum scale (chair: M. Kr afczyk) Time Speaker Title 08.30-09.15 Rudolf Hilfer

(keynote talk) Residual saturation dynamics and hysteresis in unsaturated porous media

09.15-09.35 Olaf Ippisch Reactive Multiphase Transport in the Capillary Fringe

09.35-09.55 Hector Montenegro Retarded pressure propagation due to entrapped gas below the phreatic surface

09.55-10.15 Stephan Matthai NN 10.15-10.45 Coffee break

Upscaling interface dynamics (chair: I. Neuweiler) 10.45-11.30 Marco Dentz

(keynote talk) Coarse-Graining and Non-Locality of Flow and Transport in Heterogeneous Media

11.30-11.50 Philip Binning Pseudo kinetics arising from the upscaling of geochemical equilibrium

11.50-12.10 Tycho van Noorden Upscaling of fluid flow in an unsaturated tube including inertial and capillary effects

12.30 Lunch and end of workshop

Page 6: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

Poster presentations Presenter Title Amin Ameri Effect of Salinity on Wetting Behavior of

Crude oil/Rock/Brine/Supercritical CO2 system

Shaho Bazr Arkan NN Jürgen Becker Determination of effective two-phase flow

properties using the pore morphology method

Yufei Cao NN Milena Cislerova/Michael Snehota Coupling Non-invasive Visualization

Techniques and Numerical Modeling to Investigate Air Entrapment Phenomena

Maria Datcheva NN Yabin Fan Dynamic Capillarity in Porous Media Kumar Kundan Upscaling of reactive flow involving free

boundaries Martin Leue Small-scale maps of organic matter

properties of intact preferential flow path surfaces using DRIFT spectroscopy

Babaei Masoud NN Julian Mindel NN Ali Q. Raeini

Modelling multiphase flow at pore-scale

Mathias Trojer NN Jan Vanderborght Effect of soil surface structure on soil surface

temperatures Ying Wang Musis Project P2 Franziska Moebius Musis Project P3 Ulli Weller Musis Project P5 Cindi Schütz Musis Project P6 Klaus Mosthaf Musis Project P7 Marin Dimitrov Musis Project P9

Page 7: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

Abstracts Session: Pinning and Interfacial jumps Pattern formation by invasion in particle suspension Knut Jørgen Måløy Here we study the patterns that form by injection of air in a suspension of particles in a Hele Shaw cell. By varying the injection rate over several orders of magnitude, we uncover several new morphologies and dynamic responses ranging from stick slip bubbles at low rate via intermediate states to destabilized viscous fingers at high rate. We determine how the pattern formation emerges as a result of the interplay between the governing forces, where the nonlinear frictional response from a granular media that jams and yields, alternating between solid- and fluid like behaviour is central to the observed dynamics. Linking pore scale liquid displacement dynamics with acoustic emissions during rapid interfacial jumps in porous media Franziska Möbius and Dani Or The dynamics of fluid displacement processes such as drainage and imbibition are of interest for hydrology, petroleum, and chemical engineering applications. The motion of the interfacial front seems smooth on macroscopic scale but in fact result from numerous rapid pore scale jumps associated with rapid energy release. The discontinuity at the interfacial front and the characteristics in motion are only poorly described but are of importance for macroscopic phase distribution and transport properties. Experimental pore scale observations indicate a complex dynamic behavior depending on porous media properties and boundary conditions. We investigated fluid displacement processes in porous media with acoustic emission technique, pressure measurements, and high-speed imaging, focusing on pore scale displacement processes occurring during passage of interfacial fronts. Systematic displacement measurements have shown that boundary conditions, geometry, and liquid properties affect interface dynamics and interactions with neighboring pores. A simplified model representing hydraulically connected pores based on force balance and conservation of mass including capillary forces, viscous loss and inertia was developed to gain further knowledge on the dynamic behavior and discover the source of fast energy release resulting in acoustic emission. The model demonstrate how flow regimes and geometry influence interfacial dynamics and provide physical links with acoustic emission signals detected during interfacial jumps. Thin liquid layers flows on a heterogeneous substrate: a dynamical systems approach Philippe Beltrame A ’pinning’ effect of the advancing or receding front may cause intermittent flows in unsaturated media. This phenomenon is due to the heterogeneities of the substrate as topographic roughness and/or chemical variations. The present contribution aims at better understanding the contact line dynamics and the resulting flows using a dynamical systems approach. The considered system is a free-surface thin liquid layer on a substrate striped by defects and the gravity drives the flow. The defects are modeled using a spatially variable wettability leading to an evolution equation of the profile height. This approach enables to determine the onset of depinning and to investigate the dynamics of the motion after

Page 8: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

depinning. Recently, a specific numerical code has been developed to perform this task for a 3D system [1]. As results, one finds two stable steady-states, (i.e. static liquid retained by "pinning" effect): liquid ridges (continuous liquid front) and a discrete pattern of drops. Their stability domain has been determined in the parameter plane. The mechanism of drop depinning is related to bifurcations of homoclinic kind resulting in an intermittent flow. It consists of two distinct evolution phases that take place on distinct timescales: the drop slowly stretches away from the defect but once it breaks away it slides over to the next defect [2]. This stick-slip motion agrees well with experimental observations. Ridge steady-state may depin as previously leading to a continuous moving front. However, there is another mechanism due to the transverse instability resulting in a formation of fingers. Finally, we show for large flux that the stick-slip motion becomes a stationary rivulet flow. Dynamical systems tools is powerful to produce a fairly complete picture of film flows on a striped substrate and then to determine the mean mass fluxes and its spatial distribution. References [1] Ph. Beltrame and U. Thiele. Time integration and steady-state continuation for 2d lubrication equations. SIAM J. Appl. Dyn. Syst., 9:484–518, 2010. [2] Ph. Beltrame, P. Hänggi, and U. Thiele. Depinning of three-dimensional drops from wettability defects. Euro. Phys. Lett., 86:24006, 2009. The role of fluid-fluid interfaces in wave propagation phenomena in partially-saturated porous media Holger Steeb In the present contribution, we present a three-phase multi-scale model describing wave propagation phenomena in partially saturated porous media. The effect of capillary forces on attenuation and phase velocities is investigated in the frequency range. On the one hand, the model is able to describe two continuous fluid phases, and, on the other hand the present approach models a continuous non-wetting phase and a discontinuous wetting phase. Furthermore, resonance effects of single liquid bridges and/or liquid clusters are captured with miscellaneous eigenfrequencies taking into account a viscoelastic restoring force (pinned oscillations and/or sliding motion of the contact line). As one of the aim is to study frequency-dependent attenuation due to fluid oscillations and attenuation with respect to wave-induced flow, we discuss dispersion relations of reservoir rocks in the seismic range and compare the results with well-established limits, i.e. the biphasic poroelastic Biot model. In addition, we will show that for the quasi-static limit case the results of the model are identical with the phase velocity obtained by the Gassmann-Wood limit. Session: Role of surfaces, layers and gaps on flow and transport Modeling Surface Runoff with Infiltration for Flat Areas Sjoerd E.A.T.M. van der Zee Surface runoff in fields arises when the rainfall rate exceeds the infiltration rate. The excess water that does not infiltrate, will be ponding and flow through relatively low lying parts of the local microtopography. This flow increases the hydrological connectivity of fields. An increased level of hydrological connectivity leads to a higher surface runoff flux at the field boundaries. We investigated the functional hydrological connectivity of synthetical elevation fields with varying statistical properties. For this purpose, we developed an object-oriented ponding and redistribution model to which Philip's infiltration model was coupled. The

Page 9: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

connectivity behaviour is determined by the presence of depressions with a large area and spatial organization of microtopography in rills or channels. The presence of microdepressions suppresses the effect of the spatial variation of infiltration properties. Connectivity behaviour of a field with a varying spatial distribution of infiltration properties can be predicted by transforming the unique connectivity function that was defined for a designated microtopography. Water dynamics at the root-soil interface Andrea Carminati Water supply to roots is essential for plant life. To sustain root water uptake, a continuous liquid phase has to be maintained at the interface between soil and roots. Gaps between soil and roots may interrupt the liquid-phase continuity across the soil-root interface, acting as capillary barriers for the water flow. Additionally, due to the radial geometry of the flow to roots and the non-linearity of the soil hydraulic conductivity, a drop in water potential and water content is expected to occur next to the roots when soil dries. Such a drop in water content may limit water and nutrient uptake by roots. How plants can overcome these mechanical and hydraulic flow resistances at the root-soil interface? Recent experiments with neutron radiography showed that during transpiration the water content next to roots was larger than in the bulk soil. Immediately after rewetting, the picture reversed and the soil next to roots remained markedly dry. During the following days the water content next to roots increased, exceeding that of the bulk soil. These unexpected water dynamics were caused by the distinct, hysteretic and time-dependent hydraulic properties of the soil region in the immediate vicinity of the roots, the rhizosphere. The hydraulic properties of the rhizosphere are controlled by mucigel exuded by roots and micro-organisms embedding root hairs and root debris. Mucigel is composed of polymeric substances that have a high water holding capacity. When roots take up water, mucigel remains hydrated maintaining the rhzisosphere relatively wet compared to the bulk soi, with potential benefit for nutrient and water uptake. When water supply from soil becomes limited, mucilage shrinks and the rhizosphere dries up too. Additional measurements showed that after drying the rhizosphere became temporarely hydrophobic, because of lipids present in mucilage. The temporal hydrophobicity of the rhizosphere limits water transfer between soil and roots. This may be a mechanism that plants use to hydraulically disconnect the roots from dry soils, while young roots growing in wet soils take up water from more favorable conditions. XTM visualization of water condensation and evaporation in porous gas diffusion layers of polymer electrolyte fuel cells J. Eller, Roth, J., M. Stampanoni, M., Wokaun, A., Büchi, F. N. The gas diffusion layer (GDL) in polymer electrolyte fuel cells (PEFC) connecting the flow field plate and the catalyst layer has to fulfill - besides chemical and mechanical stability - two antagonistic tasks: high permeability in the void for access of the gases to the catalyst layers and removal of product water, and high conductivity for electric charge and heat conduction in the solid. The electrochemically produced water occupies part of the porous structure it strongly affects and limits the mass transport in gas phase and influences the current distribution and fuel cell performance. Only x-ray tomographic microscopy (XTM) is capable to reveal the 3D micro-scale liquid water distribution and the solid structure simultaneously with a pixel size of about 1 µm. Two different setups at the TOMCAT beamline of the Swiss Light Source were used to study condensation and evaporation of liquid water within the porous GDL structures. Time resolved data of GDL samples with different PTFE loadings were collected. The PTFE

Page 10: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

loading is used to adjust the wetting properties of the untreated hydrophilic GDL carbon fiber structures. Session: Modeling interfacial dynamics at pore scale Digital rock physics for the prediction of capillary pressure and relative permeabilities J. Tölke M.Z. Kalam, A. Grader, Y. Mu, G. De Prisco, C. Baldwin, N. Derzhi and A. Nur Ingrain's digital rock physics lab computes the physical properties and fluid flow characteristics of oil and gas reservoir rocks like shales, carbonates, tight gas sands and oil sands. Digital rock physics (DRP) is referred to as the combination of advanced 3D imaging techniques such as X-ray computed tomography (CT) scanning or focused ion beam scanning electron microscopy (FIB-SEM), segmentation algorithms to create a digital representation of the rock and advanced numerical methods like the Finite Element Method (FEM) and the Lattice Boltzmann Method (LBM) for electrical, elastic and fluid flow properties of the rocks. We discuss a multi-scale approach to compute relative permeabilities and capillary pressure saturation relationship for real rocks. Images of the rock samples’ structure are obtained utilizing X-ray CT or FIB-SEM in resolutions ranging from 600 micro-meters to 3 nano-meters. The sample is analyzed using a descending scale of image resolution along with physical sub-sampling [5,6]. The descending size of scanning leads to an increased resolution of the three-dimensional digital core. The low-resolution/large field-of-view images guide decisions about the location and size of higher resolution/smaller-field-of-view scans and physical sub-samples. This process allows consistent registration of samples and images of different size and resolution, i.e. we know where the high-resolution volume is located inside the low-resolution images. The images are processed (segmented) to identify locations in the rock occupied by various minerals and pores. The result of this process is a digital rock, i.e. a 3-D matrix where each cell is either a solid, or a pore, and is assigned elastic properties and conductivity accordingly to the mineral or fluid occupying the corresponding location in the rock sample. We use the LBM [1,2] to simulate the flow of single or multiple fluids through the digital rock where the 3- D pore structure forms the grid system to compute fluid transport properties: absolute permeabilities, relative permeabilities, and capillary pressures [4]. A practical approach which is guided by the scan imagery on different scales is used to transfer the information to larger scales and to compute the composite and/or upscaled properties. We present different examples and the comparison of the results obtained by digital rock physics to direct measurement data show a good to reasonable prediction. Some deviations are present and discussed. References: [1] Frisch, U., D. d’Humieres, B. Hasslacher, P. Lallemand, Y. Pomeau and J. P. Rivet, Lattice gas hydrodynamics in two and three dimensions, Complex Syst., 1 (1987), 649-707. [2] Ahrenholz, J. Tölke, P. Lehmann, A. Peters, A. Kaestner, M. Krafczyk and W. Durner: Prediction of capillary hysteresis in porous material using lattice Boltzmann methods and comparison to experimental data and a morphological pore network model, Advances in Water Resources, 31: 1151–1173 (2008) [3] Grader, A., Kalam, M. Z., Toelke, J., Mu, Y., Derzhi, N., Baldwin, C., Armbruster, M., Al Dayyani, T., Clark, A., Al Yafei, G. B., and Stenger, B: A Comparative study of DRP and laboratory SCAL evaluations of Carbonate cores, SCA, Halifax, Oct. 2010 [4] Toelke, J., C. Baldwin, Y. Mu, N. Derzhi, Q. Fang. A. Grader, J. Dvorkin: Computer simulations of fluid flow in sediment: From images to permeability, The Leading Edge, (2010) 29, 1040-1048. [5] AS. Grader, Ingrain, A. B.S. Clark, ADCO, Taha Al-Dayyani, ADCO, and Amos Nur, Ingrain: Computations of Porosity and Permeability of Sparic Carbonate Using Multi-scale CT

Page 11: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

Images, International Symposium of the Society of Core Analysts, Noordwijk aan Zee, The Netherlands, 27-30 September, 2009 [6] J. Dvorkin, N. Derzhi, Q. Fang, A. Nur, B. Nur, A. Grader, C. Baldwin, H. Tono and E. Diaz: From Micro to Reservoir Scale: Permeability from Digital Experiments, The Leading Edge; December 2009; v. 28; no. 12; p. 1446-1452 Intricacies and impact of interface regularization in micro scale simulations of multi phase flow Felix Heimann Pore scale simulations of multi phase flow in porous media present a promising approach in the development and verification of continuum scale models as well as in the understanding of the underlying processes of flow phenomena like hysteresis or the peculiarities of the capillary fringe. Modern numerical models usually apply some kind of regularization to the sharp interfaces separating the immiscible fluids, thereby corrupting the interface conditions of the pressure and velocity fields. As the simulation within interesting and thus complicated domains involves a vast computational effort, the resolution of the spatial discretization is bound to be severely limited by the available hardware. This may impede a thorough convergence analysis; hence compromise the reliability of methods with - in principle - solid convergence properties. We present a discretization scheme for multi phase flow which allows the solution of the Navier Stokes equations within multiple transient domains. This unfitted discontinuous Galerkin method achieves a highly accurate approximation of the flow field irregularities at the interface. Based on this method, we compare different treatments for the numerical intricacies of the sharp interface and their impact on the pressure and velocity fields. A Lattice Boltzmann multiphase approach for pore-scale simulation of evaporation and infiltration in unsaturated soil Ying Wang, Manfred Krafczyk and Benjamin Ahrenholz In recent decades the lattice Boltzmann method (LBM) attracted much attention for DNS multiphase simulations especially in porous media due to its simplicity and efficiency. In the context of project MUSIS (Multi-Scale Interfaces in Unsaturated Soil), the impact of porous properties on the infiltration and evaporation are analyzed in the lab as well as numerically. For that purpose we extend an isothermal lattice Boltzmann multiphase approach to reproduce laboratory infiltration and evaporation measurements. The interplay of various forces in porous media on the pore scale, e.g. the interaction between capillary pressure and viscous dissipation, can be deduced from characteristic length and time scales to determine the duration of the first stage evaporation. The drying process in the first stage is characterized by vapor exchange with the atmospheric boundary layer flow and corresponds to a relatively high and constant drying rate. The second stage of evaporation involves vapor diffusion in the pores, where the drying rate drops steeply to a low value in comparison to the first stage. The isothermal multiphase approach of the lattice Boltzmann method models the transient evaporation mechanism in a 3D porous medium obtained from a CT-scan. The resulting dynamics is compared to evaporation experiments carried out by Dr. P. Lehmann et al. for both stages. In addition, the non-equilibrium infiltration process is also investigated using the lattice Boltzmann method. The comparable MSF (MultiStepFlux) experiments are performed by Dr. U. Weller et al. By applying equivalent boundary conditions of the MSF experiments on an X-ray CT scanned geometry, the simulated fluid and gas distribution will be explored to identify non-equilibrium dynamics, which corresponds to an interrupted water potential and saturation relationship. In addition, the LBM approach is validated by computing the unsaturated

Page 12: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

conductivity at various water potentials, which is compared to the laboratory measurements. In the future, we focus on the improvement of the multiphase model in order to include inertia and thermal effects. Furthermore, GPGPU computing will be used to accelerate the LBM simulations to accelerate parameter studies and allow an in depth sensitivity analysis. Pore scale simulation of dynamic interfaces in subsurface systems Paul Meakin, Hai Huang, Alexandre Tartakovsky and Anders Malthe-Sorenssen Dynamic interfaces associate with multiphase fluid flow, liquid-vapor-solid contact lines, dissolution and precipitation, and fracturing play an important in the behavior of subsurface systems. Both particle methods such as smoothed particle hydrodynamics and dissipative particle dynamics, and continuum methods with interface capturing/tracking have been used successfully to simulate the dynamics of interfaces in porous media, fractured media and fractured porous media. Despite these advances, the accurate and reliable simulation of subsurface processes at the pore scale, under a wide range of conditions, is still very difficult because of the wide range of relevant time scales and length scales. Some recent advances in the pore level simulation of multiphase flow in geometrically complex systems will be presented, and remaining challenges such as the coupling between fluid flow and geomechanics and the coupling between contact line dynamics and fluid-fluid interface dynamics will be discussed. Session: Interface between soil and atmosphere Surface-atmosphere coupling determines evaporation dynamics from a drying porous surface Peter Lehmann Prediction of drying rates from porous media remain a challenge due to the interactions between energy and atmospheric conditions (radiation, humidity, temperature, air velocity) and porous medium properties often resulting in abrupt transitions and complex dynamics. Surface wetness has an important role on the control of the rates of vapor transfer from wet pores across atmospheric viscous sub-layer next to surface for which a pore scale analysis has been developed. To quantify the mass coupling between soil and atmosphere, the model should be able to simulate the variation of surface water content in the course of evaporation and to link it to the transport properties of the surface. We have constructed a pore scale model based on the work of Suzuki and Maeda [1968] calculating the change of mass exchange coefficient due to the change of water content under different boundary condition. Linking such a model to an algorithm for redistribution of evaporated mass based on the interface curvature and check for the invasion in each time step provides the possibility for the simulation of a transient evaporation process starting from a fully saturated surface and dried out to the stage II of evaporation showing a continuous decrease in evaporation flux reminiscent of internal diffusion control process while capillary flow continuously supplies the vaporization surface. The results are consistent with the experiment conducted in the wind tunnel under controlled conditions for air velocity and provide the necessary building block for the field scale studies of the interaction between partially wet terrestrial surfaces and atmosphere during evaporation. Modeling Evaporation Influenced by Wind and Radiation Klaus Mosthaf

Page 13: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

Evaporation from partially saturated soils includes a variety of complex interacting processes and depends on different parameters of the soil and the fluids. Moreover, it is often influenced by external factors like wind and radiation. Usually, the porous media and the free flow domain are considered separately. To examine and capture the interactions, a new continuum concept for the coupling of porous media which contain two fluid phases and two components with a free flow domain (wind) was developed and implemented into the numerical simulator DuMux. The special focus is on the application on evaporation processes from soils. The relevant coupling conditions and balance equations are presented. The ideas, how to capture the relevant processes (transfer of mass, momentum and energy) are elaborated. Measurements from a corresponding evaporation experiment are shown and compared to the results of the numerical model. Investigation of the soil moisture, soil and brightness temperature of bare soil plots with different surface structure using remote sensing and ground based measurements Marin Dimitrov The structure of the soil surface layer plays an important role for the water distribution in the soil and water exchange between the soil and atmosphere. Measurements of small field scales are important for the estimation of the soil moisture distribution in different depths and underneath different soil structures. Three measurement plots (ploughed, seedbed and compacted plot) in a silt loam bare soil with different type of soil structure were prepared with a tractor before the campaign started. In each plot, soil moisture and temperature sensors were installed at 2, 5, 10 and 30 cm depth. In addition, the surface temperature and the brightness temperature of the three plots were monitored with an infrared (IR) and L-Band (1.4 GHz) radiometer, respectively. First results indicate considerable differences in the soil moisture between the different plots. The measured volumetric soil moisture at 2 and 5 cm of the ploughed plot was the lowest throughout the experimental period and varied the most spatially. In the seedbed plot and in the compacted plot, the measured soil moisture was very similar. This indicates dynamic soil structure and soil hydraulic properties. Over the whole campaign, the compacted plot showed lowest soil surface (skin) temperature, which indicates a higher loss of latent heat due to evaporation. Similar temperature was measured in the ploughed and the seedbed plots, but the ploughed plot showed higher variability. A first comparison between surface soil moisture content that was retrieved from L-Band radiometer measurements and in-situ measurements showed a good correlation if temporal changes in soil surface roughness are considered for the interpretation of the brightness temperatures measured with L-Band. Obvious changes of the brightness temperature are observed after high precipitations. The measured soil moisture and brightness temperature data were coupled in different types of hydrological models, inverse models and retrieval procedures. Existing dielectric mixing and radiative tranfer models were used for this coupling during the modeling process. The experiment itself, as well as, the modelling strategy will be presented and discussed in detail. Session: Interfaces and flow instability I Dynamic nonequilibrium in water flow in porous media - processes, experimental evidence, and effective modelling approaches Durner, Wolfgang

Page 14: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

The contribution gives an overview on the various phenomena that cause "dynamic nonequilibrium" of water flow in soils. Starting from historic observations on the macro-scale, the current perceptions and effective modeling approaches will be discussed. It will be hypothesized that the key processes are limited air-phase permeability, local heterogeneity of soil properties, and the thermodynamics of air-water-interface redistributions. These processes act in different pressure head regions and on different time scales. The need for precision measurements that are designed to quantify dynamic effects will be stressed. Investigating the flow instabilities in soil during quasi steady state flow non-invasive methods M. Snehota, Sobotkova, V., Jelinkova, A., Pohlmeier, P., Vontobel and Cislerova, M. In structured soils the preferential flow may be accompanied by a temporal variation of (quasi)saturated hydraulic conductivity. These effects are often ascribed to changing volume of the entrapped air and they are not considered in standard theory. We were able to reproduce variation of (quasi)steady state flow during ponded infiltration experiment on small undisturbed sample of coarse sandy loam and to image the process by neutron radiography (NR) and tomography (NT). Two main flow irregularities effects were detected during recurrent ponding experiment: (1) outflow rates during the first infiltration run, which was started in dry soil, were gradually decreasing, while the boundary condition remained unchanged during the experiment, and (2) the flux rate at the beginning of second infiltration was even lower that at the end of first run and remained relatively steady for the rest of the infiltration run. Series of NT images of the sample taken during quasi steady state stage of the first infiltration run showed air trapping in many of large pores and cavities in the sample. Furthermore, many of entrapped air bubbles increased in volume during the course of the first infiltration run. Further entrapped air redistribution has been detected during the second run. Similar experiment has been conducted with concurrent imaging by magnetic resonance imaging (MRI) and relaxation mapping (Jelínková et al, submitted to Organic Geochemistry). Computed x-ray tomography of soil samples was also conducted to obtain information about the inner structure of soil samples. The experimental results support the earlier hypothesis (Snehota et al., 2010) that the effect of the gradual decrease of the flow rates is caused by entrapped air redistribution and gradual build-up of bubbles in preferential pathways. The air comes probably from the soil matrix where residual air encapsulated during the primary fast breakthrough of gravitational water near the edges of large pores is being gradually replaced by water attracted to fine pores by strong capillary forces. The trapped air may thus restrict the preferential flow pathways and cause overall lower infiltration and outflow flux rates. Macroscopic representation of the interface between flow domains in a dual permeability model describing preferential flow in structured soil Horst H. Gerke In structured soil, water and solute movement through the network of interconnected larger pores, such as root and worm channels or the inter-aggregate cracks, may be considerably faster than in the bulk soil matrix. The dual-permeability model approach allows for local non-equilibrium in pressure head and solute concentration between preferential flow (PF) and soil matrix (SM) domains in case of relatively (i) large matrix blocks, (ii) small SM hydraulic conductivity, (iii) small interfacial area, and (iv) hydraulic resistances at the interface restricting the inter-domain mass transfer. However, these parameters are not well-defined, and have mostly been obtained empirically or by fitting model results to experimental data.

Page 15: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

The objective of this presentation is to discuss methods to independently determine mass transfer parameters that are related to soil structural properties. The theoretically derived mass transfer parameters are describing the geometry (shape and size) and the hydraulic conductivity and diffusivity of soil matrix blocks as macroscopic parameters. The transfer term parameters have been evaluated for idealized geometries and compared with equivalent 2-dimensional models for simple geometries. For naturally structured soils with complex geometries, an extension was proposed based on the relation between surface area and volume of the matrix. From X-ray computed tomography data of undisturbed soil cores, surface areas (S) and volumes (V) of the matrix are calculated for different bulk density thresholds. Comparisons for two differently structured soils suggest the S/V-relations to be characteristic for particular structures. For a sensitivity analysis the soil structure related parameters have been roughly estimated from data of soil protocols. Mass transfer was more sensitive to a hydraulic and diffusive resistance possibly caused by clay-organic aggregate coatings than to geometric parameters. Problems remain including the identification and properties of the interface area that is effectively involved in the inter-domain mass transfer. Scrutiny of hydraulic conductivity provides scale of permeable media flow Peter F. Germann Darcy's law separates the flow-driving forces from the flow-restraining forces. The former are forces with potential, while the latter are without potential. Flow-driving forces are usually derived from the spatial gradient of the fluid's potential energy that may be composed of gravity, capillarity, pressure and other components like temperature and salinity. Flow-restraining forces are summarized in the factor of proportionality K (L.A.Richards, 1931; Physics 1:318-333) that is referred to as hydraulic conductivity. Regardless of the underpinning concept of potentials, the accompanying parameters, relationships and scales, there always emerges a factor of proportionality K of some sort. However, its physical interpretation relies strongly on the specific concept. Reducing the flow-driving forces to ubiquitous gravity allows for a basic look at K that thus is composed of viscosity (i.e., the dissipation of momentum), the proportion of water that actively participates in flow (i.e., mobile water content), and the specific surface area per unit volume of the permeable medium onto which momentum dissipates. It is speculated that this specific surface area also participates quantitatively in the exchange of solutes, particulates, and heat during miscible displacement. We derived the parameters directly from hundreds of time series of water content variations during infiltration into field soils and sand models. The resulting frequency distributions of mobile water, minimal widths of flow paths, and depth-time relationships of wetting offer process scales of infiltration and transport in various permeable media. Session: Interfaces and flow instability II Numerical simulation of saturation overshoots in porous media with a heterogeneous multiscale method Frederike Kissling We consider the infiltration into a homogeneous porous media. Thereby different patterns can form. On the one hand the solutions can consists of a planar front, otherwise instabilities can occur and fingers build out. Both cases have in common that saturation overshoots can appear (DiCarlo, 2004; Schütz&Neuweiler, 2010). We are interested in solutions which contain saturation overshoots. Neglecting gravitational forces and assuming the fluids to be immiscible and incompressible, the governing equations are given by mass balance laws. For the capillarity-free case these

Page 16: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

laws lead in the fractional flow formulation to a first-order nonlinear evolution equation with unknown phase saturation and a nonlinear flux function. A coupled Darcy-type pressure velocity equation closes the system. This system can have multiple weak solutions. There is a whole family of solutions involving so-called nonclassical transitional waves. The above mentioned saturation overshoots can be identified as nonclassical waves. We are interested in the unique physically relevant solution which is selected as the sharp-interface limit of solutions of regularized equations. For the regularized equation capillary pressure effects have to be added, which are not only modeled by the equilibrium pressure alone, but should involve a rate-dependent contribution (Hassanizadeh&Gray, 1993). In general saturation overshoots cannot be approximated numerically by monotone schemes for conservation laws since they converge against the classical solution. On the other hand it is too expensive to solve the regularized equation on the whole domain with fine mesh and above all the solution of the regularized equation is only relevant close to a transitional wave. Apart from a transitional wave it is adequate to use the non-regularized equation (Kissling&Rohde, 2010). We will give an overview about saturation overshoots appearing in two-phase flow in porous media. Moreover we present an efficient new mass-conserving numerical method which belongs to the class of the Heterogeneous Multiscale Methods originally introduced by E&Engquist (2003). For the regularized equation we use a special solver on a small microscale space-time domain whereas we use a standard Finite Volume scheme for the non-regularized equation. The additional information which we gain from the regularized equation serves as update with the help of a special flux function. Depending on the special flux function we track the front and only use next to the front the regularized equation. We will demonstrate the efficiency of the new method on the basis of numerical examples in 1D and 2D including saturation overshoots. Saturation overshoot behind a 2D wetting front – observation and theory Rony Wallach, Xiong, Yunwu, Furman, Alex Saturation overshoot behind a 2D wetting front was observed for sandy soils. Water was supplied, at different application rates, on the soil surface as a point source. The saturation overshoot was observed for the lower water application rate at the infiltration stage and for all tested water application rates at the redistribution stage. To the best of our knowledge, this is the first report of saturation overshoot behind the vertical wetting front of an advancing plume. The formation of 2D saturation overshoot was ascribed to the simultaneous horizontal and vertical flux components in the 2D plume and the difference between the water-entry and air-entry pressures in the sandy soil. The difference in the pressures was assumed to result from the contact angle, with a static value measured at 30o but a dynamic value that was actually higher. The distribution and thickness of the saturation overshoot layer, at both infiltration and redistribution stages, were water-flux dependent Spatially Averaged Water Content in Heterogeneous Porous Media Cindi Schütz Water infiltration into the unsaturated zone is an important process for groundwater recharge. Infiltration fronts can be unstable, when the porous medium is coarse, has a homogeneous structure and when it is dry. This so-called fingering was analyzed in great detail and it cannot be modeled with standard approaches like the Richards equation. Due to the fact that natural soil systems are not homogeneous, it is important to analyze the interaction between unstable infiltration fronts and soil structure. We carried out experiments with heterogeneous structures in a 30 cm high and 60 cm wide Hele-Shaw cell. We

Page 17: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

determine the influence of structures on horizontally averaged water content and test the stabilization of the infiltration front due to fine material inclusions. Session: Can I use (trust) the model? Some thoughts on modeling and simulation in an industrial application context Matthias Schmidt The adoption of modeling & simulation practices has proven critical to improve innovation productivity at P&G. As and example, understanding and optimizing fluid flow in absorbing porous media is critical for developing superior consumer hygiene products such as paper towels, wipes, feminine pads and baby diapers. Fluid flow models have proven very valuable for the development of Pampers based on multi-year investments in the fundamental understanding of transport mechanisms, numerical simulation tools, computation infrastructure and lab methods for both input data generation as well as validation of the models. The complexity of the models requires strong cooperation of a wide range of specialists and practitioners across P&G and with key cooperation partners. In addition, the application to real-time business decision requires models to be trusted by the decision makers and simulation results available with fast feedback. We present some experience and best practices for collaboration that have advanced both capability and capacity for the development and application of these models. Session: Non-equilibrium in capillary interfaces Non-equilibrium effects in capillary interfaces in porous media Vahid Joekar-Niasar Capillary pressure-saturation curve is the major constitutive law in two-phase flow formulation in porous media. However, there is a hysteresis between Pc-S curves obtained for drainage and imbibitions. Hassanizadeh and Gray (1993) proposed that capillary pressure, saturation and interfacial area have an interplay to create a unique surface independent from drainage and imbibition. In the other words, there is no hysteresis between Pc-S-anw surfaces created under drainage and imbibitions conditions. This concept is also of practical use in interfacial mass transfer mechanisms such as evaporation, CO2 sequestration, etc. Up to now, this conjecture has been analyzed under equilibrium conditions. In this study we analyze this theory not only under equilibrium but also non-equilibrium conditions. In this regard, pore-scale modeling and micro-model experiments have been employed. We employ a novel dynamic pore-network simulator for two-phase flow (DYPOSIT). DYPOSIT can provide detailed information on fluids and interfaces configuration under non-equilibrium and equilibrium, drainage and imbibitions conditions for different viscosities. We analyze the uniqueness of the surfaces obtained under different equilibrium and non-equilibrium conditions and provide the new insights to how the dynamics effects can be identified at the macro scale. Trajectories in theta - psi space U. Weller and H.J. Vogel Most soil hydrological applications are based on the description of water content and water

Page 18: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

potential at the continuum scale, and the central parameters are θ and ψm. However, the relation of the two is not as unique as it is generally assumed. Especially under transient conditions the θ(ψm) curve deviates substantially from an equilibrium derived state. The reasons for this are twofold: first, the description of the water content as a relative saturation is in itself an averaging process. The underlying assumption is thus that the inner state of the regarded space is in a well defined and equilibrated condition. If the regarded compartment of the soil shows an inner heterogeneity then the inner state may be not equilibrated, and therefore the macroscopic perception will exhibit some deviation from the static equilibrium. A second source of a dynamic correlation of θ and ψm is the continuity of the phases at a given water saturation. Both the water phase and the air phase have to be continuous in order to enable an equilibrated relation between water content and energetic state. These connectivities depend on the water content but also on the preceding dynamics of the water phase. We studied trajectories within the θ – ψm space in response to the dynamic of the water content (expressed as dθ ), both in numerical as dt in laboratory experiments. The results show a clear dependency of the trajectories on the dynamic and the water content. Our goal is to derive a quantitative description of these trajectories and to implement this concept for numerical simulations. Channel formation during biphasic flow in porous media, or during fast monophasic flow and matrix hydrofracture: Consequences on the dynamic pressure-saturation and relative permeability relation Renaud Toussaint, Michael Niebling, Jean Schmittbuhl, Gerhard Schaefer, Knut Jorgen Maloy, Eirik G Flekkoy, Ken Tore Tallakstad, Grunde Lovoll, Mihailo Jankov, Yves Méheust We study various types of biphasic flows in quasi-two-dimensional transparent porous models. These flows imply a viscous wetting fluid, and a lowly viscous one. The models are transparent, allowing the displacement process and structure to be monitored in space and time. In drainage situations, we study how the geometry of the invader can be explained, and how it gives rise to apparent dynamic capillary effects. We show how these can be explained by viscous effects on evolving geometries of invading fluid [1]. Next, in stationary biphasic flows, we study the relationship between the macroscopic pressure drop (related to relative permeability) and the average flow rate, and how this arises from the cluster size distribution of the lowly viscous fluid [2]. We will also present how dynamic effects change the invasion geometry, the pressure/saturation relationship and the level of residual saturation for a biphasic flow - hence modifying the preferred transport path. Influence of viscous fingering on dynamic saturation-pressure curves in porous media Next, we study drainage in such models, and investigate the relationship between the pressure field and the morphology of the invading fluid. This allows to model the impact of the saturation changes in the system over the pressure difference between the wetting and non wetting phase. We show that the so-called dynamic effects referred in the hydrology literature of experimentally measured capillary pressure curves might be explained by the combined effect of capillary pressure along the invasion front of the gaseous phase and pressure changes caused by viscous effects. A detailed study of the structure optically followed shows that the geometry of the invader is self-similar with two different behaviors at small and large scales: the structure corresponds to the ones of invasion percolation models at small scales (capillary fingering structures with fractal dimension D=1.83), whereas at large scales, viscous pressure drops dominate over the capillary threshold variations, and the structures are self-similar fingering structures with a fractal dimension corresponding to Dielectric Breakdown Models (variants of the DLA model), with D ' 1.5. The cross-over scale is set by the scale at which capillary fluctuations are of the order of the viscous pressure drops. This leads physically to the fact that cross-over scale between the two fingering

Page 19: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

dimensions, goes like the inverse of the capillary number. This study utilizes these geometrical characteristics of the viscous fingers forming in dynamic drainage, to obtain a meaningfull scaling law for the saturation-pressure curve at finite speed, i.e. the so-called dynamic capillary pressure relations. We thus show how the micromechanical interplay between viscous and capillary forces leads to some pattern formation, which results in a general form of dynamic capillary pressure relations. By combining these detailed informations on the displacement structure with global measures of pressure, saturation and controlling the capillary number Ca, a scaling relation relating pressure, saturation, system size and capillary number is developed. By applying this scaling relation, pressure-saturation curves for a wide range of capillary numbers can be collapsed. Steady-State Two-Phase Flow in Porous Media: Statistics and Transport Properties First, in stationary flow with a control of the flux of both fluids, we show how the pressure drop depends on the flow rate. We will show that the dynamics is dominated by the interplay between a viscous pressure field from the wetting fluid and bubble transport of a less viscous, nonwetting phase. In contrast with more studied displacement front systems, steady-state flow is in equilibrium, statistically speaking. The corresponding theoretical simplicity allows us to explain a data collapse in the cluster size distribution of lowly viscous fluid in the system, as well as the scaling law between the observed relative permeability and the average injection rate. Channeling of the porous medium itself induced by fast flow Eventually, a last channeling effect will be illustrated in hydrofracture, a technique often used for reservoir stimulation to increase its capacity. Using laboratory reduced setup in Hele Shaw cells and simulations, we will show how the fast injection of a fluid can lead to channeling and hydrofracture, with a characteristic distance between channels depending on the injected fluid compressibility, viscosity, and the porosity of the porous medium. Bibliographie [1] Lovoll, G., M. Jankov, K.J. Maloy, R. Toussaint, J. Schmittbuhl, G. Schaefer and Y. M´eheust, Influence of viscous fingering on dynamic saturation-pressure curves in porous media, TIPM (2010) [2] Tallakstad, K.T., H.A. Knudsen, T. Ramstad, G. Løvoll, K.J. M°aløy, R. Toussaint and E.G. Flekkoy , Steady-state two-phase flow in porous media: statistics and transport properties, Phys. Rev. Lett. 102, 074502 (2009). doi:10.1103/PhysRevLett.102.074502 Session: Interface dynamics at continuum scale Residual saturation dynamics and hysteresis in unsaturated porous media Rudolf Hilfer The concepts of relative permeability and capillary pressure are crucial for the accepted traditional theory of two phase flow in porous media. Recently a theoretical approach was introduced that does not require these concepts as input [1],[2],[3] Instead it was based on the concept of hydraulic percolation of fluid phases. The presentation will describe this novel approach. It allows to predict residual saturations and local spatiotemporal changes between imbibition and drainage during two phase immiscible displacement [1],[2],[3], [4],[5],[6]. References [1] R. Hilfer, Capillary Pressure, Hysteresis and Residual Saturation in Porous Media Physica A, vol. 359, pp. 119, (2006) [2] R. Hilfer, Macroscopic Capillarity and Hysteresis for Flow in Porous Media, Physical Review E, vol. 73, pp. 016307, (2006). [3] R. Hilfer, Macroscopic capillarity without a constitutive capillary pressure function, Physica A, vol. 371, pp. 209, (2006)

Page 20: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

[4] R. Hilfer, Modeling and Simulation of Macrocapillarity, in: P. Garrido et al. (eds.) Modeling and Simulation of Materials vol. CP1091, pp. 141, American Institute of Physcis, New York, 2009. [5] R. Hilfer and F. Doster, Percolation as a basic concept for macroscopic capillarity, Transport in Porous Media, vol. 82, pp. 507, (2010) [6] F. Doster and P. Zegeling and R. Hilfer, Numerical solutions of a generalized theory for macroscopic capillarity, Physical Review E, vol. 82, pp. 036307, (2010) Reactive Multiphase Transport in the Capillary Fringe Olaf Ippisch The capillary fringe is a region where there is a very intensive phase interaction. A rising groundwater level leads to trapping of gas bubbles, which slowly dissolve. According to resent experiments in Tübingen a quickly dropping groundwater level leads to a higher dissolution of oxygen due to enhanced phase mixing. Hysteresis can not be neglected in (often coarse textured) aquifers. A macroscopic model for reactive multiphase multicomponent transport in the capillary fringe is presented. The model is based on a new splitting approach to reduce numerical dispersion. Hysteresis is incorporated using the Modified Dependent-Domain Model of Mualem (1984). There are still many unanswered questions on macroscopic description of gas phase dissolution and gas transport close to saturation, which are discussed but remain unanswered. Retarded pressure propagation due to entrapped gas below the phreatic surface Héctor Montenegro The presence of even small volumes of entrapped gas, for example as a result of phreatic surface fluctuations, can have substantial effects on pressure propagation owing to the contrast in compressibility between gaseous and fluid phase. Further the storage coefficient, traditionally assumed to be constant below the phreatic surface, considerably increases and becomes pressure dependent. A physically based macroscopic model for gaseous phase distribution from the unsaturated zone throughout the zone of gas occlusions below the phreatic surface was elaborated. The advantage of the proposed model rests in the straightforwardness of its incorporation within saturated-unsaturated flow modeling. However the implications at the transition from free to entrapped gaseous phase are yet not understood. The approach was evaluated via a two-phase model (TOUGH) and tested against laboratory and field observations. Application of dimensional analysis yielded a combination of parameters (rate of pressure change, the soil’s saturated hydraulic conductivity and the gas phase saturation at the phreatic surface) as a dimensionless group characterizing the relevance of retarded pressure propagation. Some examples illustrate, that under the presence of entrapped gas the retarded pressure propagation can be crucial in the assessment of safety in geotechnical engineering practice. Session: Upscaling interface dynamics Pseudo kinetics arising from the upscaling of geochemical equilibrium Philip Binning and Mike Celia

Page 21: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

Multicomponent contaminant transport models in groundwater are typically based on assumptions of local geochemical equilibrium on the grid scale. However, in heterogenous systems there may be significant coupling between transport processes and geochemical equilibrium at smaller than grid block scale. Here various pore- and fieldscale examples are considered to illustrate the impact of transport processes on assumptions of geochemical equilibrium. In each example the flow length scales required to reach equilibrium are calculated. It is shown that these can be as large as many meters at the pore scale and kilometers at field scales. The influence of heterogeneity in the distribution of the reactive zones is assessed for the pore-scale example, and it is shown that patchiness of reactive zones within a pore increases equilibration length, with the length and density of reactive zones, pore radius, and diffusion coefficient all playing a role in the equilibration length. When constructing models of field-scale problems it may not be reasonable to apply geochemical equilibrium, and it may be necessary to explicitly couple pore-scale and fieldscale models in a multiscale simulation. A field-scale example is also shown to illustrate that the upscaling of geochemical equilibrium poses a significant practical problem because we usually do not know the spatial location and distribution of geochemically active sites, and this information is essential input to geochemical transport models. Upscaling of fluid flow in an unsaturated tube including inertial and capillary effects Tycho Van Noorden We consider a 3D axisymmetric single fluid flow in a vertical tube with small radius, and we focus on the case when the fluid has one axisymmetric free interface. The fluid flow is described by the Navier-Stokes equations with surface tension. We use an asymptotic expansion for the unknowns (pressure, velocity and interface position) in the radius of the tube, and the scaling of the parameters is such that the Reynolds number is kept constant. In this way we obtain upscaled equations for the Darcy velocity, the pressure and the saturation. The lowest order terms of the expansions result in a rather standard Richards equation. Next we look at the first order corrections to these equations and we obtain an upscaled equation for the saturation including inertial effects.

Page 22: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

Poster presentations Effect of Salinity on Wetting Behavior of Crude oil/Rock/Brine/Supercritical CO2 system Amin Ameri, Bruining, Hans, Shojaikaveh, Narjes, Rudolph, Susanne J., Wolf, Karl-Heinz Wettability is considered a crucial factor that affects fluid flow and fluid distributions in reservoir. Wettability is strongly dependent on the reservoir fluid properties, the injected fluid composition, the rock surface chemistry, and the reservoir rock-fluid interactions. Although there exist a considerable amount of experimental and modeling studies on miscible and immiscible CO2 flooding the interfacial interactions between crude oil/brine/ock/CO2 at reservoir conditions have received little consideration in spite of the fact that it has a significant influence on oil production and CO2 storage. The main purpose of this study is to investigate the effect of brine salinity on the wettability behavior of crude oil/brine/rock/CO2 system at reservoir conditions. Synthetic reservoir brines are prepared by dissolving NaCl in double distilled water with different salinities varying between 5000-30000 ppm (mg/L). Experiments were performed in a pendant drop cell that has been modified to measure contact angle. A pendant drop shape analysis (PDSA) is applied to extract contact angle data (Figure 1). In the experiments, a rock slab at residual oil saturation is mounted into a see-through windowed high pressure cell. Having filled the cell with brine, a CO2 bubble is released to the system by means of a specifically designed needle delivery tool to form a bubble on the rock surface inside the cell. A high resolution digital camera is then used to capture images successively to measure dynamic and equilibrium contact angles. The electron microprobe analysis (EMPA) was used to evaluate compositional changes that occur both at and within the rock surface. This technique is a powerful tool for the study of heterogeneous systems for which previous analysis was possible only on an average basis or by indirect methods. According to the results, salinity increases the contact angle which means that the surface becomes less brine wet. In the range of pressure and salinity examined in this work, the system changes from strongly brine-wet to intermediate wet. The contact angle never exceeds 90° even at high pressures and elevated temperatures. Post-exposure electron microprobe analysis of a polished cross-section of the surface revealed that the rock slab was covered in some areas by a relatively thick carbon rich layer (Figure 2b). In other areas a uniform ball-shaped mineral deposition is detected on the surface (Figure 2c). From the present study, it can be concluded that CO2 can alter the wetting properties of rock during CO2 EOR or even CO2 storage processes. This could be attributed to the formation of mineral and organic deposit layers on the rock surface during CO2 injection. Determination of effective two-phase flow properties using the pore morphology method Jürgen Becker The need to optimise a porous layer with respect to its two-phase flow properties arises in various industrial applications; fuel cell gas diffusion layers, filter media and hygiene products are typical examples. To find the material with the best pore structure various structural designs have to be compared. Thus the need for computer simulations determining capillary pressure curves and relative permeabilities arises. In this talk we present methods to determine these two-phase flow properties numerically. The capillary pressure curve is determined numerically efficient with the pore morphology method. Relative permeability can be calculated using a combination of pore morphology and single-phase flow simulations: the pore morphology method determines the distribution of air and water. Then, using this distribution, the saturation dependent permeability is determined with comparatively low numerical cost. We will apply this method to various

Page 23: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

fibrous structures and present and compare the results of different imbibition and drainage models available in the pore morphology method Coupling Non-invasive Visualization Techniques and Numerical Modeling to Investigate Air Entrapment Phenomena Milena Cislerová The instability of steady state flow rates and the tracer breakthrough in undisturbed soil sample of coarse sandy loam during repeated ponded infiltration-outflow experiment have been studied by means of magnetic resonance imaging (MRI). Multiple-slice images yielding three-dimensional (3D) information from main stages of the first and of the second recurrent infiltration runs were acquired together with 3D relaxivity data. Breakthrough curves of a solute applied in each particular infiltration run were obtained analyzing the outflow concentrations. In addition, the local breakthrough curves were evaluated from MR images for a series of voxels. The selected voxels, used as observation points, lay on two different flow pathways where the flow instabilities were detected. Extent and location of highly porous regions were as well inspected by the computed tomography. It was found that the flowing volumes during the first and the second infiltration run changed due to the air encapsulation. The analyses of MRI measurements in selected observation points revealed the air bubble formation in particular preferential pathways leading to the 60% decrease of steady state infiltration rates in consecutive infiltration runs. It was accompanied by the re-routing of solute transport trajectories. The data set is used to simulate transient water flow and transport in 3D heterogeneous porous system. The applied model solves Richards' equation for water flow and the advection-dispersion equation for transport of solute. The simulation results are compared with simultaneously measured water flow and MRI data showing a reasonable agreement. Dynamic Capillarity in Porous Media Yabin Fan We consider a pseudo-parabolic equation modeling two-phase flow in porous media, where a dynamic capillary pressure term is included. To simplify the model, constant dynamic factor "tau" is considered. We investigate the traveling wave solution to the equation, and extend the result obtained previously for linear higher order terms to nonlinear and degenerate cases. The existence of such solutions is proved and non-classical traveling wave solutions (sharp TW) are introduced with proof and numerical examples. Upscaling of reactive flow involving free boundaries Kundan Kumar We consider the transport of solute particles in a reactive flow under dominant transport conditions for the variable geometry in the case of a thin strip. We derive the upscaled model (1-D) from the 2-D model using perturbation methods and compare the 1-D model with direct 2-D computations numerically. Further, we use rigorous homogenization techniques for a specific choice of reaction rate, namely crystal precipitation and dissolution in periodic porous medium (fixed geometry case). This is a joint work with T L van Noorden, I S Pop (Department of Mathematics and Computer Sciences, TU Eindhoven, Netherland) and Maria Neuss-Radu (IWR, University of Heidelberg, Germany).

Page 24: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

Small-scale maps of organic matter properties of intact preferential flow path surfaces using DRIFT spectroscopy Martin Leue The interaction of percolating water and reactive solutes with the soil matrix is mostly restricted to surfaces of preferential flow paths in structured soils. Surfaces of flow paths are often formed by cracks, decayed root channels or worm burrows that may have clay-organic coatings (i.e., cutanes) in which the outermost layer is formed by organic matter (OM). The composition of OM finally controls wettability, and sorption and transfer properties. However, the local distribution of OM-properties along such surfaces is largely unknown to date since the coatings are relatively thin and vulnerable. The Fourier transformed infrared spectroscopy in diffuse reflectance mode (DRIFT) was applied to determine 1 mm x 1 mm maps of OM functional group ratios (i.e., CH/CO-ratios, OH bands) of intact structural surfaces from Luvisol soil samples. However, DRIFT signals are affected by the micro-topography of intact surfaces. Both the illumination and reflectance are affected by the measurement point position (e.g., height, slope, aspect) on the sample surface and by the geometric arrangement of light source, sample, and detector. The objective of this paper was to analyze and develop a correction method for the effects of the surface topography on the signal by using gypsum block with defined shapes as an experimental model system. One part of gypsum block surface was coated with humic acid (HA). For both, the coated and the uncoated sides of the gypsum block, DRIFT mapping illustrated the wave length- and material-specific relief effect on the reflectance signal’s intensity. To compensate such relief effects, (i) a reflectance-to-height regression was developed using a digital terrain model (DTM) that corrected measured reflectance signal intensity, (ii) a signal intensity calibration was related to each specific wave number’s reflectance, and (iii) the corrections were used for spectral ratios calculations. The methods were tested using data from an intact soil sample surface and found applicable for analyzing intact structural surfaces. The interpretation of DRIFT measured OM composition maps of preferential flow path surfaces was improved. The local distributions of the OM properties at structural surfaces indicate yet unknown implications for preferential flow and transport especially for reactive solutes. Modelling multiphase flow at pore-scale Ali Q. Raeini Pore-scale modelling is a supplement to core analysis techniques in understanding single and multiphase fluid flow through porous media, and also in obtaining the macroscopic properties of porous media. The aim of this study is to develop a computationally efficient method which can be used in direct numerical simulation of multiphase flow, on pore space of experimentally obtained images of porous media. Incompressible Navier-Stokes equation were discretized by the widely-used finite volume (FVM) method. The volume of fluid (VOF) method were used for modelling the motion of fluid interfaces. A sharp surface force (SSF) model is devised based on the continuous surface force (CSF) model, and applied in the numerical method in a consistent manner. The solution algorithm was modified and a correction term was introduced to develop a stable solution algorithm, free of any non-physical velocities, for direct numerical simulation of multiphase flow in presence of high surface tension forces and complex solid boundaries. This model will be used to analyse the complex processes controlling multiphase flow in porous media and to predict the relative permeability and capillary pressure curves under different flow patterns.

Page 25: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

Effect of soil surface structure on soil surface temperatures Jan Vanderborght Soil surface temperatures of three different soil plots: a ploughed plot, a seedbed plot, and a compacted seedbed plot were measured with an IR camera. The surface roughness and the soil bulk density of the plots were regularly measured during the monitoring campaign. The differences in soil surface structure led to clear differences in soil surface temperature. Using simulations with a coupled heat and water flow model, Hydrus-1D, the effect of different parameters and variables on the soil surface energy balance was evaluated. Considered variables/parameters were: the albedo, the soil bulk density and its impact on hydraulic and thermal soil properties, and the roughness of the soil surface and its impact on the momentum and sensible heat transfer between the soil surface and the atmosphere.

Page 26: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

List of participants Benjamin Ahrenholz, Institute for Computational Modeling in Civil Engineering, TU

Braunschweig, Germany

Amin Amerighasrodashti, Department of Geotechnology, Section of Petroleum Engineering, TU Delft, The Netherlands

Shaho Bazr Afkan, Lehrstuhl Reservoir Engineering, Montanuniversität Loeben, Austria

Jürgen Becker, Fraunhofer ITWM, Kaiserslautern, Germany

Philippe Beltrame, Université d'Avignon, France

Cas Berentsen Department of Geotechnology, TU Delft, The Netherlands

Philip Binning, Technical University of Denmark (TDU), Denmark

Yufei Cao, Deptartment of Hydromechanics and Modeling of Hydrosystems, Universität Stuttgart, Germany

Andrea Carminati, Helmholtz Center for Environmental Research - UFZ, Department of Soil Physics, Halle, Germany

Milena Císlerová, CTU in Prague, Civil engineering, Deptartment of Irrigation Drainage and Landscape Engineering, Czech Republic

Maria Datcheva, Ruhr-University Bochum, Mechanics-Continuum Mechanics, Germany

Marco Dentz, Institute of Environmental Assessment and Water Research (IDÆA-CSIC), Barcelona, Spain

Marin Dimitrov, Forschungszentrum Jülich, Agrosphere, IBG-3, Germany

Wolfgang Durner, TU Braunschweig, Germany

Jens Eller, Paul Scherrer Institut, Electrochemistry Laboratory, Villigen, Switzerland

Yabin Fan, Eindhoven University of Technology, Department of Mathematics and Computer Science, The Netherlands

Hannes Flühler, ETH Zürich, Switzerland

Horst H Gerke, Leibniz Centre for Agricultural Landscape Research (ZALF), Institute of Soil Landscape Research, Germany

Peter Germann, University of Bern, Switzerland

Adrien Grenier Upstream Technology Development, Procter and Gamble, Frankfurt, Germany

Felix Heimann, IWR, Heidelberg University, Germany

Rainer Helmig, Deptartment of Hydromechanics and Modeling of Hydrosystems, Universität Stuttgart, Germany

Rudolf Hilfer, Universität Stuttgart, Germany

Olaf Ippisch, IWR, Heidelberg University, Germany

Joaquin Jimenez, Geosciences Rennes UMR 6118 CNRS, Universite de Rennes I, France

Page 27: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

Vahid Joekar-Niasar, Evironmental hydrogeology group, Utrecht University, The Netherlands

Ruben Juanes, Massachusetts Institute of Technology (MIT) Cambridge, USA

Frederike Kissling, Institut für Angewandte Analysis und Numerische Simulation, Universität Stuttgart, Germany

Manfred Krafczyk, Institute for Computational Modeling in Civil Engineering, TU Braunschweig, Germany

Kundan Kumar, TU Eindhoven, The Netherlands

Peter Lehmann, Soil and Terrestrial Environmental Physics, ETH Zürich, Switzerland

Martin Leue, Leibniz Centre for Agricultural Landscape Research (ZALF), Institute of Soil Landscape Research, Germany

Knut Jørgen Måløy, University of Oslo, Norway

Babaei Masoud, Imperial College London, United Kingdom

Stephan Matthäi, Lehrstuhl Reservoir Engineering, Montanuniversität Loeben, Austria

Paul Meakin, Iadho National Laboratory, USA

Julian Mindel, Lehrstuhl Reservoir Engineering, Montanuniversität Loeben, Austria

Franziska Möbius, Soil and Terrestrial Environmental Physics, ETH Zürich, Switzerland

Héctor Montenegro, VAW Federal Waterways Engineering and Research Institute, Karlsruhe, Germany

Klaus Mosthaf, Deptartment of Hydromechanics and Modeling of Hydrosystems, Universität Stuttgart, Germany

Insa Neuweiler, Institute for Fluid Mechanics and Environmental Physics in Civil Engineering, Leibniz Universität Hannover, Germany

Ehsan Nikooee Evironmental hydrogeology group, Utrecht University, The Netherlands

Dani Or, Soil and Terrestrial Environmental Physics, ETH Zürich, Switzerland

Ali Q. Raeini, Imperial College London, United Kingdom

Mathias Schmidt Upstream Technology Development, Procter and Gamble, Frankfurt, Germany

Cindi Schütz, Institute for Fluid Mechanics and Environmental Physics in Civil Engineering, Leibniz Universität Hannover, Germany

Narjes Shojaikaveh, Department of Geotechnology, Section of Petroleum Engineering, TU Delft, The Netherlands

Michael Snehota, CTU in Prague, Civil engineering, Deptartment of Irrigation Drainage and Landscape Engineering, Czech Republic

Fritz Stauffer, IfU, ETH Zurich, Switzerland

Holger Steeb, Ruhr-University Bochum, Mechanics-Continuum Mechanics, Germany

Jonas Tölke, Ingrain, Huston TX, USA

Page 28: Interfaces and interface displacement in unsaturated ... · Interfaces and interface displacement in unsaturated porous ... Steeb The role of fluid-fluid ... defect but once it breaks

Renaud Toussaint, University of Strasbourg, CNRS, France

Mathias Trojer, Lehrstuhl Reservoir Engineering, Montanuniversität Loeben, Austria

Tycho Van Noorden, TU Eindhoven, The Netherlands

Jan Vanderborght, Forschungszentrum Jülich, Agrosphere, IBG-3, Germany

Sjoerd van der Zee, Wageningen University, The Netherlands

Hans-Jörg Vogel, Helmholtz Center for Environmental Research - UFZ, Department of Soil Physics, Halle, Germany

Wolfgang Wachter Deutsche Forschungsgemeinschaft (DFG), Bonn, Germany

Rony Wallach, Hebrew University of Jerusalem, Israel

Ying Wang, Institute for Computational Modeling in Civil Engineering, TU Braunschweig, Germany

Ulrich Weller, Helmholtz Center for Environmental Research - UFZ, Department of Soil Physics, Halle, Germany

Paul Zegeling, University of Utrecht, Department of Mathematics, The Netherlands