# Prof. Dr. Martin O. Saar

###### Chair for Geothermal Energy and Geofluids

Prof. Dr. Martin O. Saar
Geothermal Energy & Geofluids
Institute of Geophysics
NO F 51.2
Sonneggstrasse 5
CH-8092 Zurich Switzerland

Contact
 Phone +41 44 632 3465 Email saarm(at)ethz.ch

 Dominique Ballarin Dolfin Phone +41 44 632 3465 Email ballarin(at)ethz.ch

My research interests are in geophysical fluid dynamics of subsurface multiscale, multiphase, multicomponent, reactive fluid (groundwater, hydrocarbon, CO2) and energy (heat, pressure) transport, such as water- and CO2-based geothermal energy utilization, geologic CO2 storage, grid-scale energy storage, enhanced oil recovery, and groundwater flow. Methods include computer simulations, laboratory experiments, and field analyses.

This professorship and the associated Geothermal Energy and Geofluids (GEG) Group is generously endowed by the Werner Siemens Foundation, which is hereby gratefully acknowledged.

## Publications

### REFEREED PUBLICATIONS IN JOURNALS

 66 Ogland-Hand, J.D., J.M. Bielicki, Y. Wang, B.M. Adams, T.A. Buscheck, and M.O. Saar The value of bulk energy storage for reducing CO2 emissions and water requirements from regional electricity systems., Energy Conversion and Management, 181, pp. 674-685, 2019. AbstractThe implementation of bulk energy storage (BES) technologies can help to achieve higher penetration and utilization of variable renewable energy technologies (e.g., wind and solar), but it can also alter the dispatch order in regional electricity systems in other ways. These changes to the dispatch order affect the total amount of carbon dioxide (CO2) that is emitted to the atmosphere and the amount of total water that is required by the electricity generating facilities. In a case study of the Electricity Reliability Council of Texas system, we separately investigated the value that three BES technologies (CO2- Geothermal Bulk Energy Storage, Compressed Air Energy Storage, Pumped Hydro Energy Storage) could have for reducing system-wide CO2 emissions and water requirements. In addition to increasing the utilization of wind power capacity, the dispatch of BES also led to an increase in the utilization of natural gas power capacity and of coal power capacity, and a decrease in the utilization of nuclear power capacity, depending on the character of the net load, the CO2 price, the water price, and the BES technology. These changes to the dispatch order provided positive value (e.g., increase in natural gas generally reduced CO2 emissions; decrease in nuclear utilization always decreased water requirements) or negative value (e.g., increase in coal generally increased CO2 emissions; increase in natural gas sometimes increased water requirements) to the regional electricity system. We also found that these values to the system can be greater than the cost of operating the BES facility. At present, there are mechanisms to compensate BES facilities for ancillary grid services, and our results suggest that similar mechanisms could be enacted to compensate BES facilities for their contribution to the environmental sustainability of the system. 65 Dambly, M.L.T., M. Nejati, D. Vogler, and M.O. Saar On the direct measurement of shear moduli in transversely isotropic rocks using the uniaxial compression test, International Journal of Rock Mechanics and Mining Sciences (IJRMMS), 113, pp. 220-240, 2019. AbstractThis paper introduces a methodology for the direct determination of the shear moduli in transversely isotropic rocks, using a single test, where a cylindrical specimen is subjected to uniaxial compression. A method is also developed to determine the orientation of the isotropy plane as well as the dynamic elastic constants using ultrasonic measurements on a single cylindrical specimen. Explicit formulae are developed to calculate the shear moduli from strain gauge measurements at different polar angles. The calculation of shear moduli from these formulae requires no knowledge about Young’s moduli or Poisson’s ratios and depends only on the orientation of the isotropy plane. Several strain gauge setups are designed to obtain the shear moduli from different numbers and arrangements of strain gauges. We demonstrate, that the shear moduli can be determined accurately and efficiently with only three strain gauge measurements. The orientation of the isotropy plane is measured with different methods, including ultrasonic measurements. The results show, that the isotropy plane of the tested granitic samples slightly deviates from the foliation plane. However, the foliation plane can still determine the orientation of the isotropy plane with a good approximation. 64 Myre, J.M., I. Lascu, E.A. Lima, J.M. Feinberg, M.O. Saar, and B.P. Weiss Using TNT-NN to Unlock the Fast Full Spatial Inversion of Large Magnetic Microscopy Datasets, Earth Planets and Space, 2019. AbstractModern magnetic microscopy (MM) provides high-resolution, ultra-high sensitivity moment magnetometry, with the ability to measure at spatial resolutions better than 10^−4 m and to detect magnetic moments weaker than 10^−15 Am^2 . These characteristics make modern MM devices capable of particularly high resolution analysis of the magnetic properties of materials, but generate extremely large data sets. Many studies utilizing MM attempt to solve an inverse problem to determine the magnitude of the magnetic moments that produce the measured component of the magnetic field. Fast Fourier techniques in the frequency domain and non-negative least-squares (NNLS) methods in the spatial domain are the two most frequently used methods to solve this inverse problem. Although extremely fast, Fourier techniques can produce solutions that violate the non-negativity of moments constraint. Inversions in the spatial domain do not violate non-negativity constraints, but the execution times of standard NNLS solvers (the Lawson and Hanson method and Matlab’s lsqlin) prohibit spatial domain inversions from operating at the full spatial resolution of an MM. In this paper we present the applicability of the TNT-NN algorithm, a newly developed NNLS active set method, as a means to directly address the NNLS routine hindering existing spatial domain inversion methods. The TNT-NN algorithm enhances the performance of spatial domain inversions by accelerating the core NNLS routine. Using a conventional computing system, we show that the TNT-NN algorithm produces solutions with residuals comparable to conventional methods while reducing execution time of spatial domain inversions from months to hours or less. Using isothermal remanent magnetization measurements of multiple synthetic and natural samples, we show that the capabilities of the TNT-NN algorithm allow scans with sizes that made them previously inaccesible to NNLS techniques to be inverted. Ultimately, the TNT- NN algorithm enables spatial domain inversions of MM data on an accelerated timescale that renders spatial domain analyses for modern MM studies practical. In particular, this new technique enables MM experiments that would have required an impractical amount of inversion time such as high-resolution stepwise magnetization and demagnetization and 3-dimensional inversions. 63 Kant, M.A., E. Rossi, J. Duss, F. Amman, M.O. Saar, and P. Rudolf von Rohr Demonstration of thermal borehole enlargement to facilitate controlled reservoir engineering for deep geothermal, oil or gas systems, Applied Energy, 212, pp. 1501-1509, 2018. AbstractThe creation of deep reservoirs for geothermal energy or oil and gas extraction is impeded by insu cient stimulation. Direction and extension of the created fractures are complex to control and, therefore, large stimulated and interconnected fracture networks are di cult to create. This poses an inherent risk of un- economic reservoirs, due to insu cient heat-sweep surfaces or hydraulic shortcuts. Therefore, we present a new technique, which locally increases the cross section of a borehole by utilizing a thermal spallation process on the sidewalls of the borehole. By controlled and local enlargement of the well bore diameter, initial fracture sources are created, potentially reducing the injection pressure during stimulation, initiating fracture growth, optimizing fracture propagation and increasing the number of accessible preexisting frac- tures. Consequently, local thermal borehole enlargement reduces project failure risks by providing better control on subsequent stimulation processes. In order to show the applicability of the suggested technique, we conducted a shallow field test in an underground rock laboratory. Two types of borehole enlargements were created in a 14.5 m deep borehole, confirming that the technology is applicable, with implications for improving the productivity of geothermal, oil and gas reservoirs. 62 Ahkami, M., T. Roesgen, M.O. Saar, and X.-Z. Kong High-resolution temporo-ensemble PIV to resolve pore-scale flow in 3D-printed fractured porous media, Transport in Porous Media, pp. 1-17, 2018. AbstractFractures are conduits that can enable fast advective transfer of (fluid, solute, reactant, particle, etc.) mass and energy. Such fast transfer can significantly affect pore-scale physico-chemical processes, which can in turn affect macroscopic mass and energy transport characteristics. Here, flooding experiments are conducted in a well-characterized fractured porous medium, manufactured by 3D printing. Given steady-state flow conditions, the micro-structure of the two-dimensional (2D) pore fluid flow field is delineated to resolve fluid velocities on the order of a sub-millimeter per second. We demonstrate the capabilities of a new temporo-ensemble Particle Image Velocimetry (PIV) method by maximizing its spatial resolution, employing in-line illumination. This method is advantageous as it is capable of minimizing the number of pixels, required for velocity determinations, down to one pixel, thereby enabling resolving high spatial resolutions of velocity vectors in a large field of view (FOV). While the main goal of this study is to introduce a novel experimental and velocimetry framework, this new method is then applied to specifically improve the understanding of fluid flow through fractured porous media. Histograms of measured velocities indicate log-normal and Gaussian-type distributions of longitudinal and lateral velocities in fractures, respectively. The magnitudes of fluid velocities in fractures and the flow interactions between fractures and matrices are shown to be influenced by the permeability of the background matrix and the orientation of the fractures. 61 Kong, X.-Z., C. Deuber, A. Kittilä, M. Somogyvari, G. Mikutis, P. Bayer, W.J. Stark, and M.O. Saar Tomographic reservoir imaging with DNA-labeled silica nanotracers: The first field validation, Environmental Science &Technology, 52/23, pp. 13681-13689, 2018. AbstractThis study presents the first field validation of using DNA-labeled silica nanoparticles as tracers to image subsurface reservoirs by travel time based tomography. During a field campaign in Switzerland, we performed short-pulse tracer tests under a forced hydraulic head gradient to conduct a multisource−multireceiver tracer test and tomographic inversion, determining the two-dimensional hydraulic conductivity field between two vertical wells. Together with three traditional solute dye tracers, we injected spherical silica nanotracers, encoded with synthetic DNA molecules, which are protected by a silica layer against damage due to chemicals, microorganisms, and enzymes. Temporal moment analyses of the recorded tracer concentration breakthrough curves (BTCs) indicate higher mass recovery, less mean residence time, and smaller dispersion of the DNA-labeled nanotracers, compared to solute dye tracers. Importantly, travel time based tomography, using nanotracer BTCs, yields a satisfactory hydraulic conductivity tomogram, validated by the dye tracer results and previous field investigations. These advantages of DNA-labeled nanotracers, in comparison to traditional solute dye tracers, make them well-suited for tomographic reservoir characterizations in fields such as hydrogeology, petroleum engineering, and geothermal energy, particularly with respect to resolving preferential flow paths or the heterogeneity of contact surfaces or by enabling source zone characterizations of dense nonaqueous phase liquids. 60 Samrock, F., A.V. Grayver, H. Eysteinsson, and M.O. Saar Magnetotelluric image of transcrustal magmatic system beneath the Tulu Moye geothermal prospect in the Ethiopian Rift, Geophysical Research Letters, 2018. AbstractContinental rifting is initiated by a dynamic interplay between tectonic stretching and mantle upwelling. Decompression melting assists continental break-up through lithospheric weakening and enforces upflow of melt to the Earth’s surface. However, the details about melt transport through the brittle crust and storage under narrow rift-aligned magmatic segments remain largely unclear. Here we present a crustal scale electrical conductivity model for a magmatic segment in the Ethiopian Rift, derived from 3-D phase tensor inversion of magnetotelluric data. Our subsurface model shows that melt migrates along pre-existing weak structures and is stored in different concentrations on two major interconnected levels, facilitating the formation of a convective hydrothermal system. The obtained model of a transcrustal magmatic system offers new insights into rifting mechanisms, evolution of magma ascent, and prospective geothermal reservoirs. 59 Amann, F., V. Gischig, K. Evans, et al., A. Kittilä, S. Wiemer, M.O. Saar, S. Löw, Th. Driesner, H. Maurer, and D. Giardini The seismo-hydro-mechanical behaviour during deep geothermal reservoir stimulations: open questions tackled in a decameter-scale in-situ stimulation experiment, Solid Earth, pp. 115-137, 2018. AbstractIn this contribution, we present a review of scientific research results that address seismo-hydromechanically coupled processes relevant for the development of a sustainable heat exchanger in low-permeability crystalline rock and introduce the design of the In situ Stimulation and Circulation (ISC) experiment at the Grimsel Test Site dedicated to studying such processes under controlled conditions. The review shows that research on reservoir stimulation for deep geothermal energy exploitation has been largely based on laboratory observations, large-scale projects and numerical models. Observations of full-scale reservoir stimulations have yielded important results. However, the limited access to the reservoir and limitations in the control on the experimental conditions during deep reservoir stimulations is insufficient to resolve the details of the hydromechanical processes that would enhance process understanding in a way that aids future stimulation design. Small-scale laboratory experiments provide fundamental insights into various processes relevant for enhanced geothermal energy, but suffer from (1) difficulties and uncertainties in upscaling the results to the field scale and (2) relatively homogeneous material and stress conditions that lead to an oversimplistic fracture flow and/or hydraulic fracture propagation behavior that is not representative of a heterogeneous reservoir. Thus, there is a need for intermediate-scale hydraulic stimulation experiments with high experimental control that bridge the various scales and for which access to the target rock mass with a comprehensive monitoring system is possible. The ISC experiment is designed to address open research questions in a naturally fractured and faulted crystalline rock mass at the Grimsel Test Site (Switzerland). Two hydraulic injection phases were executed to enhance the permeability of the rock mass. During the injection phases the rock mass deformation across fractures and within intact rock, the pore pressure distribution and propagation, and the microseismic response were monitored at a high spatial and temporal resolution. 58 Rossi, E., M.A. Kant, C. Madonna, M.O. Saar, and Ph. Rudolf von Rohr The Effects of High Heating Rate and High Temperature on the Rock Strength: Feasibility Study of a Thermally Assisted Drilling Method, Rock Mechanics and Rock Engineering, 51/9, pp. 2957-2964 , 2018. AbstractIn this paper, the feasibility of a thermally assisted drilling method is investigated. The working principle of this method is based on the weakening effect of a flame-jet to enhance the drilling performance of conventional, mechanical drilling. To investigate its effectiveness, we study rock weakening after rapid, localized flame-jet heating of Rorschach sandstone and Central Aare granite. We perform experiments on rock strength after flame treatments in comparison to oven heating, for temperatures up to 650 $$^{\circ}$$C and heating rates from 0.17 to 20 $$^{\circ}$$C/s. The material hardening, commonly observed at moderate temperatures after oven treatments, can be suppressed by flame heating the material at high heating rates. Our study highlights the influence of the heating rate on the mechanism of thermal microcracking. High heating rate, flame treatments appear to mostly induce cracks at the grain boundaries, opposed to slow oven treatments, where also a considerable number of intragranular cracks are found. Herewith, we postulate that at low heating rates, thermal expansion stresses cause the observed thermal cracking. In contrast, at higher heating rates, thermal cracking is dominated by the stress concentrations caused by high thermal gradients. 57 Hobé, A., D. Vogler, M.P. Seybold, A. Ebigbo, R.R. Settgast, and M.O. Saar Estimating Fluid Flow Rates through Fracture Networks using Combinatorial Optimization, Advances in Water Resources, 122, pp. 85-97, 2018. AbstractTo enable fast uncertainty quantification of fluid flow in a discrete fracture network (DFN), we present two approaches to quickly compute fluid flow in DFNs using combinatorial optimization algorithms. Specifically, the presented Hanan Shortest Path Maxflow (HSPM) and Intersection Shortest Path Maxflow (ISPM) methods translate DFN geometries and properties to a graph on which a max flow algorithm computes a combinatorial flow, from which an overall fluid flow rate is estimated using a shortest path decomposition of this flow. The two approaches are assessed by comparing their predictions with results from explicit numerical simulations of simple test cases as well as stochastic DFN realizations covering a range of fracture densities. Both methods have a high accuracy and very low computational cost, which can facilitate much-needed in-depth analyses of the propagation of uncertainty in fracture and fracture-network properties to fluid flow rates. 56 Mikutis, G., C.A. Deuber, L. Schmid, A. Kittilä, N. Lobsiger, M. Puddu, D.O. Asgeirsson, R.N. Grass, M.O. Saar, and W.J. Stark Silica-encapsulated DNA-based tracers for aquifer characterization, Environmental Science & Technology, 52, pp. 12142-12152, 2018. AbstractEnvironmental tracing is a direct way to characterize aquifers, evaluate the solute transfer parameter in underground reservoirs, and track contamination. By performing multitracer tests, and translating the tracer breakthrough times into tomographic maps, key parameters such as a reservoir’s effective porosity and permeability field may be obtained. DNA, with its modular design, allows the generation of a virtually unlimited number of distinguishable tracers. To overcome the insufficient DNA stability due to microbial activity, heat, and chemical stress, we present a method to encapsulated DNA into silica with control over the particle size. The reliability of DNA quantification is improved by the sample preservation with NaN3 and particle redispersion strategies. In both sand column and unconsolidated aquifer experiments, DNA-based particle tracers exhibited slightly earlier and sharper breakthrough than the traditional solute tracer uranine. The reason behind this observation is the size exclusion effect, whereby larger tracer particles are excluded from small pores, and are therefore transported with higher average velocity, which is pore size-dependent. Identical surface properties, and thus flow behavior, makes the new material an attractive tracer to characterize sandy groundwater reservoirs or to track multiple sources of contaminants with high spatial resolution. 55 Leal, A.M.M., D.A. Kulik, W.R. Smith, and M.O. Saar An overview of computational methods for chemical equilibrium and kinetic calculations for geochemical and reactive transport modeling, Pure and Applied Chemistry, 89/5, pp. 597-643, 2017. AbstractWe present an overview of novel numerical methods for chemical equilibrium and kinetic calculations for complex non-ideal multiphase systems. The methods we present for equilibrium calculations are based either on Gibbs energy minimization (GEM) calculations or on solving the system of extended law of mass-action (xLMA) equations. In both methods, no a posteriori phase stability tests, and thus no tentative addition or removal of phases during or at the end of the calculations, are necessary. All potentially stable phases are considered from the beginning of the calculation, and stability indices are immediately available at the end of the computation to determine which phases are actually stable at equilibrium. Both GEM and xLMA equilibrium methods are tailored for computationally demanding applications that require many rapid local equilibrium calculations, such as reactive transport modeling. The numerical method for chemical kinetic calculations we present supports both closed and open systems, and it considers a partial equilibrium simplification for fast reactions. The method employs an implicit integration scheme that improves stability and speed when solving the often stiff differential equations in kinetic calculations. As such, it requires compositional derivatives of the reaction rates to assemble the Jacobian matrix of the resultant implicit algebraic equations that are solved at every time step. We present a detailed procedure to calculate these derivatives, and we show how the partial equilibrium assumption affects their computation. These numerical methods have been implemented in Reaktoro (reaktoro.org), an open-source software for modeling chemically reactive systems. We finish with a discussion on the comparison of these methods with others in the literature. 54 Luhmann, A.J., B.M. Tutolo, B.C. Bagley, D.F.R. Mildner, W.E. Seyfried Jr., and M.O. Saar Permeability, porosity, and mineral surface area changes in basalt cores induced by reactive transport of CO2-rich brine, Water Resources Research, 53, pp. 1-20, 2017. AbstractFour reactive flow-through laboratory experiments (two each at 0.1 mL/min and 0.01 mL/min flow rates) at 150°C and 150 bar (15 MPa) are conducted on intact basalt cores to assess changes in porosity, permeability, and surface area caused by CO2-rich fluid-rock interaction. Permeability decreases slightly during the lower flow rate experiments and increases during the higher flow rate experiments. At the higher flow rate, core permeability increases by more than one order of magnitude in one experiment and less than a factor of two in the other due to differences in preexisting flow path structure. X-ray computed tomography (XRCT) scans of pre- and post-experiment cores identify both mineral dissolution and secondary mineralization, with a net decrease in XRCT porosity of ∼0.7%–0.8% for the larger pores in all four cores. (Ultra) small-angle neutron scattering ((U)SANS) data sets indicate an increase in both (U)SANS porosity and specific surface area (SSA) over the ∼1 nm to 10 µm scale range in post-experiment basalt samples, with differences due to flow rate and reaction time. Net porosity increases from summing porosity changes from XRCT and (U)SANS analyses are consistent with core mass decreases. (U)SANS data suggest an overall preservation of the pore structure with no change in mineral surface roughness from reaction, and the pore structure is unique in comparison to previously published basalt analyses. Together, these data sets illustrate changes in physical parameters that arise due to fluid-basalt interaction in relatively low pH environments with elevated CO2 concentration, with significant implications for flow, transport, and reaction through geologic formations. 53 Luhmann, A.J., B.M. Tutolo, C. Tan, B.M. Moskowitz, M.O. Saar, and W.E. Seyfried, Jr. Whole rock basalt alteration from CO2-rich brine during flow-through experiments at 150°C and 150 bar, Chemical Geology, 453, pp. 92-110, 2017. AbstractFour flow-through experiments at 150 °C were conducted on intact cores of basalt to assess alteration and mass transfer during reaction with CO2-rich fluid. Two experiments used a flow rate of 0.1 ml/min, and two used a flow rate of 0.01 ml/min. Permeability increased for both experiments at the higher flow rate, but decreased for the lower flow rate experiments. The experimental fluid (initial pH of 3.3) became enriched in Si, Mg, and Fe upon passing through the cores, primarily from olivine and titanomagnetite dissolution and possibly pyroxene dissolution. Secondary minerals enriched in Al and Si were present on post-experimental cores, and an Fe2O3-rich phase was identified on the downstream ends of the cores from the experiments at the lower flow rate. While we could not specifically identify if siderite (FeCO3) was present in the post-experimental basalt cores, siderite was generally saturated or supersaturated in outlet fluid samples, suggesting a thermodynamic drive for Fe carbonation from basalt-H2O-CO2 reaction. Reaction path models that employ dissolution kinetics of olivine, labradorite, and enstatite also suggest siderite formation at low pH. Furthermore, fluid-rock interaction caused a relatively high mobility of the alkali metals; up to 29% and 99% of the K and Cs present in the core, respectively, were preferentially dissolved from the cores, likely due to fractional crystallization effects that made alkali metals highly accessible. Together, these datasets illustrate changes in chemical parameters that arise due to fluid-basalt interaction in relatively low pH environments with elevated CO2. 52 Myre, J.M., E. Frahm, D.J. Lilja, and M.O. Saar TNT-NN: A Fast Active Set Method for Solving Large Non-Negative Least Squares Problems, Procedia Computer Science, 108C, pp. 755-764, 2017. AbstractIn 1974 Lawson and Hanson produced a seminal active set strategy to solve least-squares prob- lems with non-negativity constraints that remains popular today. In this paper we present TNT-NN, a new active set method for solving non-negative least squares (NNLS) problems. TNT-NN uses a different strategy not only for the construction of the active set but also for the solution of the unconstrained least squares sub-problem. This results in dramatically improved performance over traditional active set NNLS solvers, including the Lawson and Hanson NNLS algorithm and the Fast NNLS (FNNLS) algorithm, allowing for computational investigations of new types of scientific and engineering problems. For the small systems tested (5000 × 5000 or smaller), it is shown that TNT-NN is up to 95× faster than FNNLS. Recent studies in rock magnetism have revealed a need for fast NNLS algorithms to address large problems (on the order of 105 × 105 or larger). We apply the TNT- NN algorithm to a representative rock magnetism inversion problem where it is 60× faster than FNNLS. We also show that TNT-NN is capable of solving large (45000 × 45000) problems more than 150× faster than FNNLS. These large test problems were previously considered to be unsolvable, due to the excessive execution time required by traditional methods. 51 Walsh, S.D.C., N. Garapati, A.M.M. Leal, and M.O. Saar Calculating thermophysical fluid properties during geothermal energy production with NESS and Reaktoro, Geothermics, 70, pp. 146-154, 2017. AbstractWe investigate how subsurface fluids of different compositions affect the electricity generation of geothermal power plants. First, we outline a numerical model capable of accounting for the thermophysical properties of geothermal fluids of arbitrary composition within simulations of geothermal power production. The behavior of brines with varying compositions from geothermal sites around the globe are then examined using the model. The effect of each brine on an idealized binary geothermal power plant is simulated, and their performances compared by calculating the amount of heat exchanged from the fluid to the plant’s secondary cycle. Our simulations combine (1) a newly developed Non-linear Equation System Solver (NESS), for simulating individual geothermal power plant components, (2) the advanced geochemical speciation solver, Reaktoro, used for calculation of thermodynamic fluid properties, and (3) compositional models for the calculation of fluid-dynamical properties (e.g., viscosity as a function of temperature and brine composition). The accuracy of the model is verified by comparing its predictions with experimental data from single-salt, binary-salt, and multiple-salt solutions. The geothermal power plant simulations show that the brines considered in this study can be divided into three main categories: (1) those of largely meteoric origin with low salinity for which the effect of salt concentration is negligible; (2) moderate-depth brines with high concentrations of Na+ and K+ ions, whose performance is well approximated by pure NaCl solutions of equivalent salinity; and (3) deeper, high-salinity brines that require a more detailed consideration of their composition for accurate simulation of plant operations. 50 Leal, A.M.M., D. Kulik, G. Kosakowski, and M.O. Saar Computational methods for reactive transport modeling: An extended law of mass-action, xLMA, method for multiphase equilibrium calculations, Advances in Water Resources, 96, pp. 405-422, 2016. AbstractWe present a numerical method for multiphase chemical equilibrium calculations based on a Gibbs energy minimization approach. The method can accurately and efficiently determine the stable phase assemblage at equilibrium independently of the type of phases and species that constitute the chemical system. We have successfully applied our chemical equilibrium algorithm in reactive transport simulations to demonstrate its effective use in computationally intensive applications. We used FEniCS to solve the governing partial differential equations of mass transport in porous media using finite element methods in unstructured meshes. Our equilibrium calculations were benchmarked with GEMS3K, the numerical kernel of the geochemical package GEMS. This allowed us to compare our results with a well-established Gibbs energy minimization algorithm, as well as their performance on every mesh node, at every time step of the transport simulation. The benchmark shows that our novel chemical equilibrium algorithm is accurate, robust, and efficient for reactive transport applications, and it is an improvement over the Gibbs energy minimization algorithm used in GEMS3K. The proposed chemical equilibrium method has been implemented in Reaktoro, a unified framework for modeling chemically reactive systems, which is now used as an alternative numerical kernel of GEMS. 49 Tutolo, B.M., D.F. Mildner, C.V. Gagnon, M.O. Saar, and W.E. Seyfried Nanoscale constraints on porosity generation and fluid flow during serpentinization, Geology, 44/2, pp. 103-106, 2016. AbstractField samples of olivine-rich rocks are nearly always serpentinized—commonly to completion—but, paradoxically, their intrinsic porosity and permeability are diminishingly low. Serpentinization reactions occur through a coupled process of fluid infiltration, volumetric expansion, and reaction-driven fracturing. Pores and reactive surface area generated during this process are the primary pathways for fluid infiltration into and reaction with serpentinizing rocks, but the size and distribution of these pores and surface area have not yet been described. Here, we utilize neutron scattering techniques to present the first measurements of the evolution of pore size and specific surface area distribution in partially serpentinized rocks. Samples were obtained from the ca. 2 Ma Atlantis Massif oceanic core complex located off-axis of the Mid-Atlantic Ridge and an olivine-rich outcrop of the ca. 1.1 Ga Duluth Complex of the North American Mid-Continent Rift. Our measurements and analyses demonstrate that serpentine and accessory phases form with their own, inherent porosity, which accommodates the bulk of diffusive fluid flow during serpentinization and thereby permits continued serpentinization after voluminous serpentine minerals fill reaction-generated porosity. 48 Leal, A.M.M., D.A. Kulik, and M.O. Saar Enabling Gibbs energy minimization algorithms to use equilibrium constants of reactions in multiphase equilibrium calculations, Chemical Geology, 437, pp. 170-181, 2016. AbstractThe geochemical literature provides numerous thermodynamic databases compiled from equilibrium constants of reactions. These databases are typically used in speciation calculations based on the law of mass action (LMA) approach. Unfortunately, such LMA databases cannot be directly used in equilibrium speciation methods based on the Gibbs energy minimization (GEM) approach because of their lack of standard chemical potentials of species. Therefore, we present in this work a simple conversion approach that calculates apparent standard chemical potentials of species from equilibrium constants of reactions. We assess the consistency and accuracy of the use of apparent standard chemical potentials in GEM algorithms by benchmarking equilibrium speciation calculations using GEM and LMA methods with the same LMA database. In all cases, we use PHREEQC to perform the LMA calculations, and we use its LMA databases to calculate the equilibrium constants of reactions. GEM calculations are performed using a Gibbs energy minimization method of Reaktoro — a unified open-source framework for numerical modeling of chemically reactive systems. By comparing the GEM and LMA results, we show that the use of apparent standard chemical potentials in GEM methods produces consistent and accurate equilibrium speciation results, thus validating our new, practical conversion technique that enables GEM algorithms to take advantage of many existing LMA databases, consequently extending and diversifying their range of applicability. 47 Buscheck, T.A., J.M. Bielicki, T.A. Edmunds, Y. Hao, Y. Sun, J.B. Randolph, and M.O. Saar Multifluid geo-energy systems: Using geologic CO2 storage for geothermal energy production and grid-scale energy storage in sedimentary basins, Geosphere, 12/3, pp. 1-19, 2016. AbstractWe present an approach that uses the huge fluid and thermal storage capac ity of the subsurface, together with geologic carbon dioxide (CO 2 ) storage, to harvest, store, and dispatch energy from subsurface (geothermal) and surface (solar, nuclear, fossil) thermal resources, as well as excess energy on electric grids. Captured CO 2 is injected into saline aquifers to store pres – sure, generate artesian flow of brine, and provide a supplemental working fluid for efficient heat extraction and power conversion. Concentric rings of injection and production wells create a hydraulic mound to store pressure, CO 2 , and thermal energy. This energy storage can take excess power from the grid and excess and/or waste thermal energy and dispatch that energy when it is demanded, and thus enable higher penetration of variable renewable en – ergy technologies (e.g., wind and solar). CO 2 stored in the subsurface func – tions as a cushion gas to provide enormous pressure storage capacity and displace large quantities of brine, some of which can be treated for a variety of beneficial uses. Geo thermal power and energy-storage applications may generate enough revenues to compensate for CO 2 capture costs. While our ap – proach can use nitrogen (N 2 ), in addition to CO 2 , as a supplemental fluid, and store thermal energy, this study focuses on using CO 2 for geothermal energy production and grid-scale energy storage. We conduct a techno-economic assess ment to determine the levelized cost of electricity using this approach to generate geothermal power. We present a reservoir pressure management strategy that diverts a small portion of the produced brine for beneficial con – sumptive use to reduce the pumping cost of fluid recirculation, while reducing the risk of seismicity, caprock fracture, and CO 2 leakage. 46 Tutolo, B.M., X.-Z. Kong, W.E. Seyfried Jr., and M.O. Saar High performance reactive transport simulations examining the effects of thermal, hydraulic, and chemical (THC) gradients on fluid injectivity at carbonate CCUS reservoir scales, International Journal of Greenhouse Gas Control, 39, pp. 285-301, 2015. AbstractCarbonate minerals and CO2 are both considerably more soluble at low temperatures than they are at elevated temperatures. This inverse solubility has led a number of researchers to hypothesize that injecting low-temperature (i.e., less than the background reservoir temperature) CO2 into deep, saline reservoirs for CO2 Capture, Utilization, and Storage (CCUS) will dissolve CO2 and carbonate minerals near the injection well and subsequently exsolve and re-precipitate these phases as the fluids flow into the geothermally warm portion of the reservoir. In this study, we utilize high performance computing to examine the coupled effects of cool CO2 injection and background hydraulic head gradients on reservoir-scale mineral volume changes. We employ the fully coupled reactive transport simulator PFLOTRAN with calculations distributed over up to 800 processors to test 21 scenarios designed to represent a range of reservoir depths, hydraulic head gradients, and CO2 injection rates and temperatures. In the default simulations, 50 °C CO2 is injected at a rate of 50 kg/s into a 200 bar, 100 °C calcite or dolomite reservoir. By comparing these simulations with others run at varying conditions, we show that the effect of cool CO2 injection on reservoir-scale mineral volume changes tends to be relatively minor. We conclude that the low heat capacity of CO2 effectively prevents low-temperature CO2 injection from decreasing the temperature across large portions of the simulated carbonate reservoirs. This small thermal perturbation, combined with the low relative permeability of brine within the supercritical CO2 plume, yields limited dissolution and precipitation effects directly attributable to cool CO2 injection. Finally, we calculate that relatively high water-to-rock ratios, which may occur over much longer CCUS reservoir lifetimes or in materials with sufficiently high brine relative permeability within the supercritical CO2 plume, would be required to substantially affect injectivity through thermally-induced mineral dissolution and precipitation. Importantly, this study shows the utility of reservoir scale-reactive transport simulators for testing hypotheses and placing laboratory-scale observations into a CCUS reservoir-scale context. 45 Tutolo, B.M., A.J. Luhmann, X.-Z. Kong, M.O. Saar, and W.E. Seyfried Jr. CO2 sequestration in feldspar-rich sandstone: Coupled evolution of fluid chemistry, mineral reaction rates, and hydrogeochemical properties, Geochimica Et Cosmochimica Acta, 160, pp. 132-154, 2015. AbstractTo investigate CO2 Capture, Utilization, and Storage (CCUS) in sandstones, we performed three 150 °C flow-through experiments on K-feldspar-rich cores from the Eau Claire formation. By characterizing fluid and solid samples from these experiments using a suite of analytical techniques, we explored the coupled evolution of fluid chemistry, mineral reaction rates, and hydrogeochemical properties during CO2 sequestration in feldspar-rich sandstone. Overall, our results confirm predictions that the heightened acidity resulting from supercritical CO2 injection into feldspar-rich sandstone will dissolve primary feldspars and precipitate secondary aluminum minerals. A core through which CO2-rich deionized water was recycled for 52 days decreased in bulk permeability, exhibited generally low porosity associated with high surface area in post-experiment core sub-samples, and produced an Al hydroxide secondary mineral, such as boehmite. However, two samples subjected to ?3 day single-pass experiments run with CO2-rich, 0.94 mol/kg NaCl brines decreased in bulk permeability, showed generally elevated porosity associated with elevated surface area in post-experiment core sub-samples, and produced a phase with kaolinite-like stoichiometry. CO2-induced metal mobilization during the experiments was relatively minor and likely related to Ca mineral dissolution. Based on the relatively rapid approach to equilibrium, the relatively slow near-equilibrium reaction rates, and the minor magnitudes of permeability changes in these experiments, we conclude that CCUS systems with projected lifetimes of several decades are geochemically feasible in the feldspar-rich sandstone end-member examined here. Additionally, the observation that K-feldspar dissolution rates calculated from our whole-rock experiments are in good agreement with literature parameterizations suggests that the latter can be utilized to model CCUS in K-feldspar-rich sandstone. Finally, by performing a number of reactive transport modeling experiments to explore processes occurring during the flow-through experiments, we have found that the overall progress of feldspar hydrolysis is negligibly affected by quartz dissolution, but significantly impacted by the rates of secondary mineral precipitation and their effect on feldspar saturation state. The observations produced here are critical to the development of models of CCUS operations, yet more work, particularly in the quantification of coupled dissolution and precipitation processes, will be required in order to produce models that can accurately predict the behavior of these systems. 44 Garapati, N., J.B. Randolph, and M.O. Saar Brine displacement by CO2, energy extraction rates, and lifespan of a CO2-limited CO2-Plume Geothermal (CPG) system with a horizontal production well, Geothermics, 55, pp. 182-194, 2015. AbstractSeveral studies suggest that CO2-based geothermal energy systems may be operated economically when added to ongoing geologic CO2 sequestration. Alternatively, we demonstrate here that CO2-Plume Geothermal (CPG) systems may be operated long-term with a finite amount of CO2. We analyze the performance of such CO2-limited CPG systems as a function of various geologic and operational parameters. We find that the amount of CO2 required increases with reservoir depth, permeability, and well spacing and decreases with larger geothermal gradients. Furthermore, the onset of reservoir heat depletion decreases for increasing geothermal gradients and for both particularly shallow and deep reservoirs. 43 Tutolo, B.M., A.T. Schaen, M.O. Saar, and W.E. Seyfried Jr. Implications of the redissociation phenomenon for mineral-buffered fluids and aqueous species transport at elevated temperatures and pressures, Applied Geochemistry, 55, pp. 119-127, 2015. AbstractAqueous species equilibrium constants and activity models form the foundation of the complex speciation codes used to model the geochemistry of geothermal energy production, extremophilic ecosystems, ore deposition, and a variety of other processes. Researchers have shown that a simple three species model (i.e., Na+, Cl?, and NaCl(aq)) can accurately describe conductivity measurements of concentrated NaCl and KCl solutions at elevated temperatures and pressures (Sharygin et al., 2002). In this model, activity coefficients of the charged species (e.g., Na+, K+, Cl?) become sufficiently low that the complexes must redisocciate with increasing salt concentration in order to meet equilibrium constant constraints. Redissociation decreases the proportion of the elements bound up as neutral complexes, and thereby increases the true ionic strength of the solution. In this contribution, we explore the consequences of the redissociation phenomenon in albite–paragonite–quartz (APQ) buffered systems. We focus on the implications of the redissociation phenomenon for mineral solubilities, particularly the observation that, at certain temperatures and pressures, calculated activities of charged ions in solution remain practically constant even as element concentrations increase from <1 molal to 4.5 molal. Finally, we note that redissociation has a similar effect on pH, and therefore aqueous speciation, in APQ-hosted systems. The calculations and discussion presented here are not limited to APQ-hosted systems, but additionally apply to many others in which the dominant cations and anions can form neutral complexes. 42 Adams, B.M., T.H. Kuehn, J.M. Bielicki, J.B. Randolph, and M.O. Saar A comparison of electric power output of CO2 Plume Geothermal (CPG) and brine geothermal systems for varying reservoir conditions, Applied Energy, 140, pp. 365-377, 2015. AbstractIn contrast to conventional hydrothermal systems or enhanced geothermal systems, CO2 Plume Geothermal (CPG) systems generate electricity by using CO2 that has been geothermally heated due to sequestration in a sedimentary basin. Four CPG and two brine-based geothermal systems are modeled to estimate their power production for sedimentary basin reservoir depths between 1 and 5km, geothermal temperature gradients from 20 to 50°Ckm-1, reservoir permeabilities from 1×10-15 to 1×10-12m2 and well casing inner diameters from 0.14m to 0.41m. Results show that CPG direct-type systems produce more electricity than brine-based geothermal systems at depths between 2 and 3km, and at permeabilities between 10-14 and 10-13m2, often by a factor of two. This better performance of CPG is due to the low kinematic viscosity of CO2, relative to brine at those depths, and the strong thermosiphon effect generated by CO2. When CO2 is used instead of R245fa as the secondary working fluid in an organic Rankine cycle (ORC), the power production of both the CPG and the brine-reservoir system increases substantially; for example, by 22% and 20% for subsurface brine and CO2 systems, respectively, with a 35°Ckm-1 thermal gradient, 0.27m production and 0.41m injection well diameters, and 5×10-14m2 reservoir permeability. 41 Luhmann, A.J., M. Covington, J. Myre, M. Perne, S.W. Jones, C.E. Alexander Jr., and M.O. Saar Thermal damping and retardation in karst conduits, Hydrology and Earth System Sciences, 19/1, pp. 137-157, 2015. AbstractWater temperature is a non-conservative tracer in the environment. Variations in recharge temperature are damped and retarded as water moves through an aquifer due to heat exchange between water and rock. However, within karst aquifers, seasonal and short-term fluctuations in recharge temperature are often transmitted over long distances before they are fully damped. Using analytical solutions and numerical simulations, we develop relationships that describe the effect of flow path properties, flow-through time, recharge characteristics, and water and rock physical properties on the damping and retardation of thermal peaks/troughs in karst conduits. Using these relationships, one can estimate the thermal retardation and damping that would occur under given conditions with a given conduit geometry. Ultimately, these relationships can be used with thermal damping and retardation field data to estimate parameters such as conduit diameter. We also examine sets of numerical simulations where we relax some of the assumptions used to develop these relationships, testing the effects of variable diameter, variable velocity, open channels, and recharge shape on thermal damping and retardation to provide some constraints on uncertainty. Finally, we discuss a multitracer experiment that provides some field confirmation of our relationships. High temporal resolution water temperature data are required to obtain sufficient constraints on the magnitude and timing of thermal peaks and troughs in order to take full advantage of water temperature as a tracer. 40 Tutolo, B.M., A.J. Luhmann, X.-Z. Kong, M.O. Saar, and W.E. Seyfried Jr. Experimental observation of permeability changes in dolomite at CO2 sequestration conditions, Environmental Science and Technology, 48/4, pp. 2445-2452, 2014. AbstractInjection of cool CO2 into geothermally warm carbonate reservoirs for storage or geothermal energy production may lower near-well temperature and lead to mass transfer along flow paths leading away from the well. To investigate this process, a dolomite core was subjected to a 650 h, high pressure, CO2 saturated, flow-through experiment. Permeability increased from 10–15.9 to 10–15.2 m2 over the initial 216 h at 21 °C, decreased to 10–16.2 m2 over 289 h at 50 °C, largely due to thermally driven CO2 exsolution, and reached a final value of 10–16.4 m2 after 145 h at 100 °C due to continued exsolution and the onset of dolomite precipitation. Theoretical calculations show that CO2 exsolution results in a maximum pore space CO2 saturation of 0.5, and steady state relative permeabilities of CO2 and water on the order of 0.0065 and 0.1, respectively. Post-experiment imagery reveals matrix dissolution at low temperatures, and subsequent filling-in of flow passages at elevated temperature. Geochemical calculations indicate that reservoir fluids subjected to a thermal gradient may exsolve and precipitate up to 200 cm3 CO2 and 1.5 cm3 dolomite per kg of water, respectively, resulting in substantial porosity and permeability redistribution. 39 Luhmann, A.J., X.-Z. Kong, B.M. Tutolo, N. Garapati, B.C. Bagley, M.O. Saar, and W.E. Seyfried Jr. Experimental dissolution of dolomite by CO2-charged brine at 100oC and 150 bar: Evolution of porosity, permeability, and reactive surface area, Chemical Geology, 380, pp. 145-160, 2014. AbstractHydrothermal flow experiments of single-pass injection of CO2-charged brine were conducted on nine dolomite cores to examine fluid–rock reactions in dolomite reservoirs under geologic carbon sequestration conditions. Post-experimental X-ray computed tomography (XRCT) analysis illustrates a range of dissolution patterns, and significant increases in core bulk permeability were measured as the dolomite dissolved. Outflow fluids were below dolomite saturation, and cation concentrations decreased with time due to reductions in reactive surface area with reaction progress. To determine changes in reactive surface area, we employ a power-law relationship between reactive surface area and porosity (Luquot and Gouze, 2009). The exponent in this relationship is interpreted to be a geometrical parameter that controls the degree of surface area change per change in core porosity. Combined with XRCT reconstructions of dissolution patterns, we demonstrate that this exponent is inversely related to both the flow path diameter and tortuosity of the dissolution channel. Even though XRCT reconstructions illustrate dissolution at selected regions within each core, relatively high Ba and Mn recoveries in fluid samples suggest that dissolution occurred along the core’s entire length and width. Analysis of porosity–permeability data indicates an increase in the rate of permeability enhancement per increase in porosity with reaction progress as dissolution channels lengthen along the core. Finally, we incorporate the surface area–porosity model of Luquot and Gouze (2009) with our experimentally fit parameters into TOUGHREACT to simulate experimental observations. 38 Adams, B.M., T.H. Kuehn, J.M. Bielicki, J.B. Randolph, and M.O. Saar On the importance of the thermosiphon effect in CPG (CO2-Plume geothermal) power systems, Energy, 69, pp. 409-418, 2014. AbstractCPG (CO2 Plume Geothermal) energy systems use CO2 to extract thermal energy from naturally permeable geologic formations at depth. CO2 has advantages over brine: high mobility, low solubility of amorphous silica, and higher density sensitivity to temperature. The density of CO2 changes substantially between geothermal reservoir and surface plant, resulting in a buoyancy-driven convective current – a thermosiphon – that reduces or eliminates pumping requirements. We estimated and compared the strength of this thermosiphon for CO2 and for 20 weight percent NaCl brine for reservoir depths up to 5 km and geothermal gradients of 20, 35, and 50 °C/km. We found that through the reservoir, CO2 has a pressure drop approximately 3–12 times less than brine at the same mass flowrate, making the CO2 thermosiphon sufficient to produce power using reservoirs as shallow as 0.5 km. At 2.5 km depth with a 35 °C/km gradient – the approximate western U.S. continental mean – the CO2 thermosiphon converted approximately 10% of the energy extracted from the reservoir to fluid circulation, compared to less than 1% with brine, where additional mechanical pumping is necessary. We found CO2 is a particularly advantageous working fluid at depths between 0.5 km and 3 km. 37 Tutolo, B.M., X.-Z. Kong, W.E. Seyfried Jr., and M.O. Saar Internal consistency in aqueous geochemical data revisited: Applications to the aluminum system, Geochimica et Cosmochimica Acta, 133, pp. 216-234, 2014. AbstractInternal consistency of thermodynamic data has long been considered vital for confident calculations of aqueous geochemical processes. However, an internally consistent mineral thermodynamic data set is not necessarily consistent with calculations of aqueous species thermodynamic properties due, potentially, to improper or inconsistent constraints used in the derivation process. In this study, we attempt to accommodate the need for a mineral thermodynamic data set that is internally consistent with respect to aqueous species thermodynamic properties by adapting the least squares optimization methods of Powell and Holland (1985). This adapted method allows for both the derivation of mineral thermodynamic properties from fluid chemistry measurements of solutions in equilibrium with mineral assemblages, as well as estimates of the uncertainty on the derived results. Using a large number of phase equilibria, solubility, and calorimetric measurements, we have developed a thermodynamic data set of 12 key aluminum-bearing mineral phases. These data are derived to be consistent with Na+ and K+ speciation data presented by Shock and Helgeson (1988), H4SiO4(aq) data presented by Stefánsson (2001), and the Al speciation data set presented by Tagirov and Schott (2001). Many of the constraining phase equilibrium measurements are exactly the same as those used to develop other thermodynamic data, yet our derived values tend to be quite different than some of the others’ due to our choices of reference data. The differing values of mineral thermodynamic properties have implications for calculations of Al mineral solubilities; specifically, kaolinite solubilities calculated with the developed data set are as much as 6.75 times lower and 73% greater than those calculated with Helgeson et al. (1978) and Holland and Powell (2011) data, respectively. Where possible, calculations and experimental data are compared at low T, and the disagreement between the two sources reiterates the common assertion that low-T measurements of phase equilibria and mineral solubilities in the aluminum system rarely represent equilibrium between water and well-crystallized, aluminum-bearing minerals. As an ancillary benefit of the derived data, we show that it may be combined with high precision measurements of aqueous complex association constants to derive neutral species activity coefficients in supercritical fluids. Although this contribution is specific to the aluminum system, the methods and concepts developed here can help to improve the calculation of water–rock interactions in a broad range of earth systems. 36 Garapati, N., J.B. Randolph, J.L. Valencia Jr., and M.O. Saar CO2-Plume Geothermal (CPG) Heat Extraction in Multi-layered Geologic Reservoirs, Energy Procedia, 63, pp. 7631-7643, 2014. AbstractCO2-Plume Geothermal (CPG) technology involves injecting CO2 into natural, highly permeable geologic units to extract energy. The subsurface CO2 absorbs heat from the reservoir, buoyantly rises to the surface, and drives a power generation system. The CO2 is then cooled and reinjected underground. Here, we analyze the effects of multi-layered geologic reservoirs on CPG system performance by examining the CO2 mass fraction in the produced fluid, pore-fluid pressure buildup during operation, and heat energy extraction rates. The produced CO2 mass fraction depends on the stratigraphic positions of highly permeable layers which also affect the pore-fluid pressure drop across the reservoir. 35 Buscheck, T.A., J.M. Bielicki, M. Chen, Y. Sun, Y. Hao, T.A. Edmunds, M.O. Saar, and J.B. Randolph Integrating CO2 Storage with Geothermal Resources for Dispatchable Renewable Electricity, Energy procedia, 63, pp. 7619-7630, 2014. AbstractWe present an approach that uses the huge fluid and thermal storage capacity of the subsurface, together with geologic CO2 storage, to harvest, store, and dispatch energy from subsurface (geothermal) and surface (solar, nuclear, fossil) thermal resources, as well as energy from electrical grids. Captured CO2 is injected into saline aquifers to store pressure, generate artesian flow of brine, and provide an additional working fluid for efficient heat extraction and power conversion. Concentric rings of injection and production wells are used to create a hydraulic divide to store pressure, CO2, and thermal energy. Such storage can take excess power from the grid and excess/waste thermal energy, and dispatch that energy when it is demanded, enabling increased penetration of variable renewables. Stored CO2 functions as a cushion gas to provide enormous pressure-storage capacity and displaces large quantities of brine, which can be desalinated and/or treated for a variety of beneficial uses. Geothermal power and energy-storage applications may generate enough revenues to justify CO2 capture costs. 34 Randolph, J.B., M.O. Saar, and J.M. Bielicki Geothermal energy production at geologic CO2 sequestration sites: Impact of thermal drawdown on reservoir pressure, Energy Procedia, 37, pp. 6625-6635, 2013. AbstractRecent geotechnical research shows that geothermal heat can be efficiently mined by circulating carbon dioxide through naturally permeable rock formations — a method called CO2 Plume Geothermal — the same geologic reservoirs that are suitable for deep saline aquifer CO2 sequestration or enhanced oil recovery. This paper describes the effect of thermal drawdown on reservoir pressure buildup during sequestration operations, revealing that geothermal heat mining can decrease overpressurization by 10% or more. 33 Gottardi, R., P.-H. Kao, M.O. Saar, and Ch. Teyssier Effects of permeability fields on fluid, heat, and oxygen isotope transport in extensional detachment systems, Geochemistry, Geophysics, Geosystems, 14/5, pp. 1493-1522, 2013. Abstract[1] Field studies of Cordilleran metamorphic core complexes indicate that meteoric fluids permeated the upper crust down to the detachment shear zone and interacted with highly deformed and recrystallized (mylonitic) rocks. The presence of fluids in the brittle/ductile transition zone is recorded in the oxygen and hydrogen stable isotope compositions of the mylonites and may play an important role in the thermomechanical evolution of the detachment shear zone. Geochemical data show that fluid flow in the brittle upper crust is primarily controlled by the large-scale fault-zone architecture. We conduct continuum-scale (i.e., large-scale, partial bounce-back) lattice-Boltzman fluid, heat, and oxygen isotope transport simulations of an idealized cross section of a metamorphic core complex. The simulations investigate the effects of crust and fault permeability fields as well as buoyancy-driven flow on two-way coupled fluid and heat transfer and resultant exchange of oxygen isotopes between meteoric fluid and rock. Results show that fluid migration to middle to lower crustal levels is fault controlled and depends primarily on the permeability contrast between the fault zone and the crustal rocks. High fault/crust permeability ratios lead to channelized flow in the fault and shear zones, while lower ratios allow leakage of the fluids from the fault into the crust. Buoyancy affects mainly flow patterns (more upward directed) and, to a lesser extent, temperature distributions (disturbance of the geothermal field by ~25°C). Channelized fluid flow in the shear zone leads to strong vertical and horizontal thermal gradients, comparable to field observations. The oxygen isotope results show δ18O depletion concentrated along the fault and shear zones, similar to field data. 32 Walsh, S.D.C., and M.O. Saar Developing extensible lattice-Boltzmann simulators for general-purpose graphics-processing units, Communications in Computational Physics, 13/3, pp. 867-879, 2013. AbstractLattice-Boltzmann methods are versatile numerical modeling techniques capable of reproducing a wide variety of fluid-mechanical behavior. These methods are well suited to parallel implementation, particularly on the single-instruction multiple data (SIMD) parallel processing environments found in computer graphics processing units (GPUs). Although recent programming tools dramatically improve the ease with which GPUbased applications can be written, the programming environment still lacks the flexibility available to more traditional CPU programs. In particular, it may be difficult to develop modular and extensible programs that require variable on-device functionality with current GPU architectures. This paper describes a process of automatic code generation that overcomes these difficulties for lattice-Boltzmann simulations. It details the development of GPU-based modules for an extensible lattice-Boltzmann simulation package – LBHydra. The performance of the automatically generated code is compared to equivalent purposewritten codes for both single-phase,multiphase, andmulticomponent flows. The flexibility of the new method is demonstrated by simulating a rising, dissolving droplet moving through a porous medium with user generated lattice-Boltzmann models and subroutines. 31 Kong, X.-Z., and M.O. Saar DBCreate: A SUPCRT92-based program for producing EQ3/6, TOUGHREACT, and GWB thermodynamic databases at user-defined T and P, Computers and Geosciences, 51, pp. 415-417, 2013. AbstractSUPCRT92 is a widely used software package for calculating the standard thermodynamic properties of minerals, gases, aqueous species, and reactions. However, it is labor-intensive and error-prone to use it directly to produce databases for geochemical modeling programs such as EQ3/6, the Geochemist’s Workbench, and TOUGHREACT. DBCreate is a SUPCRT92-based software program written in FORTRAN90/95 and was developed in order to produce the required databases for these programs in a rapid and convenient way. This paper describes the overall structure of the program and provides detailed usage instructions. 30 Kong, X.-Z., and M.O. Saar Numerical study of the effects of permeability heterogeneity on density-driven convective mixing during CO2 dissolution storage, Int. J. Greenhouse Gas Control, 19, pp. 160-173, 2013. AbstractPermanence and security of carbon dioxide (CO2) in geologic formations requires dissolution of CO2 into brine, which slightly increases the brine density. Previous studies have shown that this small increase in brine density induces convective currents, which greatly enhances the mixing efficiency and thus CO2 storage capacity and rate in the brine. Density-driven convection, in turn, is known to be largely dominated by permeability heterogeneity. This study explores the relationship between the process of density-driven convection and the permeability heterogeneity of an aquifer during CO2 dissolution storage, using high-resolution numerical simulations. While the porosity is kept constant, the heterogeneity of the aquifer is introduced through a spatially varying permeability field, characterized by the Dykstra-Parsons coefficient and the correlation length. Depending on the concentration profile of dissolved CO2, we classify the convective finger patterns as dispersive, preferential, and unbiased fingering. Our results indicate that the transition between unbiased and both preferential and dispersive fingering is mainly governed by the Dykstra-Parsons coefficient, whereas the transition between preferential and dispersive fingering is controlled by the permeability correlation length. Furthermore, we find that the CO2 dissolution flux at the top boundary will reach a time-independent steady state. Although this flux strongly correlates with permeability distribution, it generally increases with the permeability heterogeneity when the correlation length is less than the system size. 29 Luhmann, A.J., X.-Z. Kong, B.M. Tutolo, K. Ding, M.O. Saar, and W.E. Seyfried Jr. Permeability reduction produced by grain reorganization and accumulation of exsolved CO2 during geologic carbon sequestration: A new CO2 trapping mechanism, Environmental Science and Technlogy, 47/1, pp. 242-251, 2013. AbstractCarbon sequestration experiments were conducted on uncemented sediment and lithified rock from the Eau Claire Formation, which consisted primarily of K-feldspar and quartz. Cores were heated to accentuate reactivity between fluid and mineral grains and to force CO2 exsolution. Measured permeability of one sediment core ultimately reduced by 4 orders of magnitude as it was incrementally heated from 21 to 150 °C. Water-rock interaction produced some alteration, yielding sub-?m clay precipitation on K-feldspar grains in the core’s upstream end. Experimental results also revealed abundant newly formed pore space in regions of the core, and in some cases pores that were several times larger than the average grain size of the sediment. These large pores likely formed from elevated localized pressure caused by rapid CO2 exsolution within the core and/or an accumulating CO2 phase capable of pushing out surrounding sediment. CO2 filled the pores and blocked flow pathways. Comparison with a similar experiment using a solid arkose core indicates that CO2 accumulation and grain reorganization mainly contributed to permeability reduction during the heated sediment core experiment. This suggests that CO2 injection into sediments may store more CO2 and cause additional permeability reduction than is possible in lithified rock due to grain reorganization. 28 Covington, M., A.F. Banwell, J. Gulley, and M.O. Saar Quantifying the effects of glacier conduit geometry and recharge on proglacial hydrograph form, Journal of Hydrology, 414-415, pp. 59-71, 2012. AbstractThe configuration of glacier hydrological systems is often inferred from proxy data, such as hydrographs, that are collected in proglacial streams. Seasonal changes in the peakedness of hydrographs are thought to reflect changes in the configuration of the subglacial drainage system. However, the amount of information that proglacial hydrographs contain about drainage system configurations depends critically on the degree to which the drainage systems modify recharge hydrographs. If the drainage system does not modify recharge hydrographs, then proglacial hydrographs primarily reflect the recharge conditions produced by supraglacial inputs. Here, we develop a theoretical framework to determine the circumstances under which glacier drainage systems can modify recharge hydrographs and the circumstances under which recharge pulses pass through glaciers unchanged. We address the capability of single conduits, simple arborescent conduit networks, and linked cavity systems to modify diurnal recharge pulses. Simulations of discharge through large sets of such systems demonstrate that, unless large reservoirs or significant constrictions are present, the discharge hydrographs of simple glacial conduit systems are nearly identical to their recharge hydrographs. Conduit systems tend not to modify hydrographs because the changes in storage within englacial and subglacial conduit networks on short time scales are typically small compared to their ability to transmit water. This finding suggests that proglacial hydrographs reflect a variety of factors, including surface melt rate, surface water transfer, and subglacial water transfer. In many cases the influence of suglacial processes may be relatively minor. As a result, the evolution of proglacial hydrographs cannot be used unambiguously to infer changes in the structure or efficiency of englacial or subglacial hydrological systems, without accurate knowledge of the nature of the recharge hydrograph driving the flow. 27 Covington, M.D., A.J. Luhmann, C. Wicks, and M.O. Saar Process length scales and longitudinal damping in karst conduits, Journal of Geophysical Research – Earth Surface, 117, F01025, 2012. Abstract[1] Simple mathematical models often allow an intuitive grasp of the function of physical systems. We develop a mathematical framework to investigate reactive or dissipative transport processes within karst conduits. Specifically, we note that for processes that occur within a characteristic timescale, advection along the conduit produces a characteristic process length scale. We calculate characteristic length scales for the propagation of thermal and electrical conductivity signals along karst conduits. These process lengths provide a quantitative connection between karst conduit geometry and the signals observed at a karst spring. We show that water input from the porous/fractured matrix is also characterized by a length scale and derive an approximation that accounts for the influence of matrix flow on the transmission of signals through the aquifer. The single conduit model is then extended to account for conduits with changing geometries and conduit flow networks, demonstrating how these concepts can be applied in more realistic conduit geometries. We introduce a recharge density function, ϕR, which determines the capability of an aquifer to damp a given signal, and cast previous explanations of spring variability within this framework. Process lengths are a general feature of karst conduits and surface streams, and we conclude with a discussion of other potential applications of this conceptual and mathematical framework. 26 Alexander, S.C., and M.O. Saar Improved characterization of small u for Jacob pumping test analysis methods, Ground Water, 50/2, pp. 256-265, 2012. AbstractNumerous refinements have been proposed to traditional pumping test analyses, yet many hydrogeologists continue to use the Jacob method due to its simplicity. Recent research favors hydraulic tomography and inverse numerical modeling of pumping test data. However, at sites with few wells, or relatively short screens, the data requirements of these methods may be impractical within physical and fiscal constraints. Alternatively, an improved understanding of the assumptions and limitations of Theis and, due to their widespread usage, Jacob analyses, leads to improved interpretations in data-poor environments. A fundamental requirement of Jacob is a “small” value of u = f(r2/t), with radial distance, r, and pumping time, t. However, selection of a too stringent (i.e., too low) maximum permissible u-value, umax, results in rejection of usable data from wells beyond a maximum radius, rmax. Conversely, data from small radii, less than rmin, where turbulent- and vertical-flow components arise, can result in acceptance of inappropriate data. Usage of drawdown data from wells too close to the pumping well, and exclusion of data from wells deemed too far, can cause unrealistic aquifer transmissivity, permeability, and storativity determinations. Here, data from an extensive well field in a glacial-outwash aquifer in north-central Minnesota, USA, are used to develop a new estimate for umax. Traditionally quoted values for umax range from 0.01 to 0.05. Our proposed value for Jacob distance-drawdown analyses is significantly higher with umax up to 0.2, resulting in larger allowable rmax-values and a higher likelihood of inclusion of additional wells in such pumping test analyses. 25 Covington, M.D., A.J. Luhmann, F. Gabrovsek, M.O. Saar, I. Willis, and C.M. Wicks Mechanisms of heat exchange between water and rock in karst conduits, Water Resources Research, 47, W10514/10, 2011. Abstract[1] Previous studies, motivated by understanding water quality, have explored the mechanisms for heat transport and heat exchange in surface streams. In karst aquifers, temperature signals play an additional important role since they carry information about internal aquifer structures. Models for heat transport in karst conduits have previously been developed; however, these models make different, sometimes contradictory, assumptions. Additionally, previous models of heat transport in karst conduits have not been validated using field data from conduits with known geometries. Here we use analytical solutions of heat transfer to examine the relative importance of heat exchange mechanisms and the validity of the assumptions made by previous models. The relative importance of convection, conduction, and radiation is a function of time. Using a characteristic timescale, we show that models neglecting rock conduction produce spurious results in realistic cases. In contrast to the behavior of surface streams, where conduction is often negligible, conduction through the rock surrounding a conduit determines heat flux at timescales of weeks and longer. In open channel conduits, radiative heat flux can be significant. In contrast, convective heat exchange through the conduit air is often negligible. Using the rules derived from our analytical analysis, we develop a numerical model for heat transport in a karst conduit. Our model compares favorably to thermal responses observed in two different karst settings: a cave stream fed via autogenic recharge during a snowmelt event, and an allogenically recharged cave stream that experiences continuous temperature fluctuations on many timescales. 24 Randolph, J.B., and M.O. Saar Impact of reservoir permeability on the choice of subsurface geothermal heat exchange fluid: CO2 versus water and native brine, Geothermal Resources Council (GRC) Transactions, 35, pp. 521-526, 2011. AbstractAbstract Geothermal systems utilizing carbon dioxide (CO 2) as the subsurface heat exchange fluid in naturally porous, permeable geologic formations have been shown to provide improved geothermal heat energy extraction, even at low resource temperatures. Such systems, termed CO 2 Plume Geothermal (CPG) systems, have the potential to permit expansion of geothermal energy utilization while supporting rapid implementation through the use of existing technologies. Here, we explore CPG heat extraction as a function of reservoir permeability and in comparison to water and brine geothermal heat extraction. We show that for reservoir permeabilities below 2 x 10 -14 m 2 , CO 2 -based geothermal provides better electric power production efficiency than both water-and brine-based systems. Impact of reservoir permeability on the choice of subsurface geothermal heat exchange fluid: CO2 versus water and native brine (PDF Download Available). Available from: https://www.researchgate.net/publication/266220008_Impact_of_reservoir_permeability_on_the_choice_of_subsurface_geothermal_heat_exchange_fluid_CO2_versus_water_and_native_brine [accessed Jun 7, 2017]. 23 Randolph, J.B., and M.O. Saar Combining geothermal energy capture with geologic carbon dioxide sequestration, Geophysical Research Letters, 38, L10401, 2011. Abstract[1] Geothermal energy offers clean, renewable, reliable electric power with no need for grid-scale energy storage, yet its use has been constrained to the few locations worldwide with naturally high geothermal heat resources and groundwater availability. We present a novel approach with the potential to permit expansion of geothermal energy utilization: heat extraction from naturally porous, permeable formations with CO2 as the injected subsurface working fluid. Fluid-mechanical simulations reveal that the significantly higher mobility of CO2, compared to water, at the temperature/pressure conditions of interest makes CO2 an attractive heat exchange fluid. We show numerically that, compared to conventional water-based and engineered geothermal systems, the proposed approach provides up to factors of 2.9 and 5.0, respectively, higher geothermal heat energy extraction rates. Consequently, more regions worldwide could be economically used for geothermal electricity production. Furthermore, as the injected CO2 is eventually geologically sequestered, such power plants would have negative carbon footprints. 22 Davis, M.A., S.D.C. Walsh, and M.O. Saar Statistically reconstructing continuous isotropic and anisotropic two-phase media while preserving macroscopic material properties, Physical Review E, 83, 026706, 2011. AbstractWe propose a method to generate statistically similar reconstructions of two-phase media. As with previous work, we initially characterize the microstructure of the material using two-point correlation functions (a subset of spatial correlation functions) and then generate numerical reconstructions using a simulated annealing method that preserves the geometric relationships of the material’s phase of interest. However, in contrast to earlier contributions that consider reconstructions composed of discrete arrays of pixels or voxels alone, we generate reconstructions based on assemblies of continuous, three-dimensional, interpenetrating objects. The result is a continuum description of the material microstructure (as opposed to a discretized or pixelated description), capable of efficiently representing large disparities in scale. Different reconstruction methods are considered based on distinct combinations of two-point correlation functions of varying degrees of complexity. The quality of the reconstruction methods are evaluated by comparing the total pore fraction, specific surface area of the percolating cluster, pore fraction of the percolating cluster, tortuosity, and permeability of the reconstructions to those of a set of reference assemblies. Elsewhere it has been proposed that two-phase media could be statistically reproduced with only two spatial correlation functions: the two-point probability function (the probability that two points lie within the same phase) and the lineal path function (the probability that a line between two points lies entirely within the same phase). We find that methods employing the two-point probability function and lineal path function are improved if the percolating cluster volume is also considered in the reconstruction. However, to reproduce more complicated geometric assemblies, we find it necessary to employ the two-point probability, two-point cluster, and lineal path function in addition to the percolating cluster volume to produce a generally accurate statistical reconstruction. 21 Randolph, J.B., and M.O. Saar Coupling carbon dioxide sequestration with geothermal energy capture in naturally permeable, porous geologic formations: Implications for CO2 sequestration, Energy Procedia, 4, pp. 2206-2213, 2011. AbstractCarbon dioxide (CO2) sequestration in deep saline aquifers and exhausted oil and natural gas fields has been widely considered as a means for reducing CO2 emissions to the atmosphere as a counter-measure to global warming. However, rather than treating CO2 merely as a waste fluid in need of permanent disposal, we propose that it could also be used as a working fluid in geothermal energy capture, as its thermodynamic and fluid mechanical properties suggest it transfers geothermal heat more efficiently than water. Energy production and sales in conjunction with sequestration would improve the economic viability of CO2 sequestration, a critical challenge for large-scale implementation of the technology. In addition, using CO2 as the working fluid in geothermal power systems may permit utilization of lower temperature geologic formations than those that are currently deemed economically viable, leading to more widespread utilization of geothermal energy. Here, we present the results of early-stage calculations demonstrating the geothermal energy capture potential of CO2-based geothermal systems and implications of such energy capture for the economic viability of geologic CO2 sequestration. 20 Saar, M.O. Review: Geothermal heat as a tracer of large-scale groundwater flow and as a means to determine permeability fields, special theme issue on Environmental Tracers and Groundwater Flow, editor-invited peer-reviewed contribution, Hydrogeology Journal, 19, pp. 31-52, 2011. AbstractA review of coupled groundwater and heat transfer theory is followed by an introduction to geothermal measurement techniques. Thereafter, temperature-depth profiles (geotherms) and heat discharge at springs to infer hydraulic parameters and processes are discussed. Several studies included in this review state that minimum permeabilities of approximately 5 × 10−17 < kmin <10−15 m2 are required to observe advective heat transfer and resultant geotherm perturbations. Permeabilities below kmin tend to cause heat-conduction-dominated systems, precluding inversion of temperature fields for groundwater flow patterns and constraint of permeabilities other than being 19 Myre, J., S.D.C. Walsh, D.J. Lilja, and M.O. Saar Performance analysis of single-phase multiphase, and multicomponent lattice-Boltzmann fluid flow simulations on GPU clusters, Concurrency and Computation: Practice and Experience, 23, pp. 332-350, 2010. AbstractAbstract The lattice-Boltzmann method is well suited for implementation in single-instruction multiple-data (SIMD) environments provided by general purpose graphics processing units (GPGPUs). This paper discusses the integration of these GPGPU programs with OpenMP to create lattice-Boltzmann applications for multi-GPU clusters. In addition to the standard single-phase single-component lattice-Boltzmann method, the performances of more complex multiphase, multicomponent models are also examined. The contributions of various GPU lattice-Boltzmann parameters to the performance are examined and quantified with a statistical model of the performance using Analysis of Variance (ANOVA). By examining single- and multi-GPU lattice-Boltzmann simulations with ANOVA, we show that all the lattice-Boltzmann simulations primarily depend on effects corresponding to simulation geometry and decomposition, and not on the architectural aspects of GPU. Additionally, using ANOVA we confirm that the metrics of Efficiency and Utilization are not suitable for memory-bandwidth-dependent codes. Copyright © 2010 John Wiley & Sons, Ltd. 18 Dasgupta, S., M.O. Saar, R.L. Edwards, C.-C. Shen, H. Cheng, and C.E. Alexander Jr. Three thousand years of extreme rainfall events recorded in stalagmites from Spring Valley Caverns, Minnesota, Earth and Planetary Science Letters, 300, pp. 46-54, 2010. AbstractAnnual layer analysis in two stalagmites collected from Spring Valley Caverns, southeastern Minnesota, reveals hydrological response of the cave to extreme rainfall events in the Midwest, USA. Cave-flooding events are identified within the two samples by the presence of detrital layers composed of clay sized particles. Comparison with instrumental records of precipitation demonstrates a strong correlation between these cave-flood events and extreme rainfall observed in the Upper Mississippi Valley. A simple model is developed to assess the nature of rainfall capable of flooding the cave. The model is first calibrated to the last 50-yr (1950–1998 A.D.) instrumental record of daily precipitation data for the town of Spring Valley and verified with the first 50 yr of record from 1900 to 1949 A.D. Frequency analysis shows that these extreme flood events have increased from the last half of the nineteenth century. Comparison with other paleohydrological records shows increased occurrence of extreme rain events during periods of higher moisture availability. Our study implies that increased moisture availability in the Midwestern region, due to rise in temperature from global warming could lead to an increase in the occurrence of extreme rainfall events. 17 Walsh, S.D.C., and M.O. Saar Interpolated lattice-Boltzmann boundary conditions for surface reaction kinetics, Physical Review E, 82, 066703, 2010. AbstractThis paper describes a method for implementing surface reaction kinetics in lattice Boltzmann simulations. The interpolated boundary conditions are capable of simulating surface reactions and dissolution at both stationary and moving solid-fluid and fluid-fluid interfaces. Results obtained with the boundary conditions are compared to analytical solutions for first-order and constant-flux kinetic surface reactions in a one-dimensional half space, as well as to the analytical solution for evaporation from the surface of a cylinder. Excellent agreement between analytical and simulated results is obtained for a wide range of diffusivities, lattice velocities, and surface reaction rates. The boundary model’s ability to represent dissolution in binary fluid mixtures is demonstrated by modeling diffusion from a rising bubble and dissolution of a droplet near a flat plate. 16 Randolph, J.B., and M.O. Saar Coupling geothermal energy capture with carbon dioxide sequestration in naturally permeable, porous geologic formations: A comparison with enhanced geothermal systems, Geothermal Resources Council (GRC) Transactions, 34, pp. 433-438, 2010. AbstractGeothermal energy offers clean, consistent, reliable electric power with no need for grid-scale energy storage, unlike wind and solar renewable power alternatives. However, geothermal energy is often underrepresented in renewable energy discussions and has considerable room for growth. New technology and methods will be critical for future investment, and rapid implementation of new techniques will be critical in ensuring geothermal energy plays a significant role in the future energy landscape world – wide. Here, we discuss a novel approach with the potential to permit expansion of geothermal energy utilization while supporting rapid implementation through the use of exist – ing technologies: geothermal heat use in naturally porous, permeable geologic formations with carbon dioxide as the working heat exchange fluid. 15 Walsh, S.D.C., and M.O. Saar Macroscale lattice-Boltzmann methods for low-Peclet-number solute and heat transport in heterogeneous porous media, Water Resources Research, 46, W07517, 2010. Abstract[1] This paper introduces new methods for simulating subsurface solute and heat transport in heterogeneous media using large-scale lattice-Boltzmann models capable of representing both macroscopically averaged porous media and open channel flows. Previous examples of macroscopically averaged lattice-Boltzmann models for solute and heat transport are only applicable to homogeneous media. Here, we extend these models to properly account for heterogeneous pore-space distributions. For simplicity, in the majority of this paper we assume low Peclet number flows with an isotropic dispersion tensor. Nevertheless, this approach may also be extended to include anisotropic-dispersion by using multiple relaxation time lattice-Boltzmann methods. We describe two methods for introducing heterogeneity into macroscopically averaged lattice-Boltzmann models. The first model delivers the desired behavior by introducing an additional time-derivative term to the collision rule; the second model by separately weighting symmetric and anti-symmetric components of the fluid packet densities. Chapman-Enskog expansions are conducted on the governing equations of the two models, demonstrating that the correct constitutive behavior is obtained in both cases. In addition, methods for improving model stability at low porosities are also discussed: (1) an implicit formulation of the model; and (2) a local transformation that normalizes the lattice-Boltzmann model by the local porosity. The model performances are evaluated through comparisons of simulated results with analytical solutions for one- and two-dimensional flows, and by comparing model predictions to finite element simulations of advection isotropic-dispersion in heterogeneous porous media. We conclude by presenting an example application, demonstrating the ability of the new models to couple with simulations of reactive flow and changing flow geometry: a simulation of groundwater flow through a carbonate system. 14 Covington, M., C.M. Wicks, and M.O. Saar A dimensionless number describing the effects of recharge and geometry on discharge from simple karst aquifers, Water Resources Research, 45, W11410, 2009. Abstract[1] The responses of karstic aquifers to storms are often used to obtain information about aquifer geometry. In general, spring hydrographs are a function of both system geometry and recharge. However, the majority of prior work on storm pulses through karst has not studied the effect of recharge on spring hydrographs. To examine the relative importance of geometry and recharge, we break karstic aquifers into elements according to the manner of their response to transient flow and demonstrate that each element has a characteristic response timescale. These fundamental elements are full pipes, open channels, reservoir/constrictions, and the porous matrix. Taking the ratio of the element timescale with the recharge timescale produces a dimensionless number, γ, that is used to characterize aquifer response to a storm event. Using sets of simulations run with randomly selected element parameters, we demonstrate that each element type has a critical value of γ below which the shape of the spring hydrograph is dominated by the shape of the recharge hydrograph and above which the spring hydrograph is significantly modified by the system geometry. This allows separation of particular element/storm pairs into recharge-dominated and geometry-dominated regimes. While most real karstic aquifers are complex combinations of these elements, we draw examples from several karst systems that can be represented by single elements. These examples demonstrate that for real karstic aquifers full pipe and open channel elements are generally in the recharge-dominated regime, whereas reservoir/constriction elements can fall in either the recharge- or geometry-dominated regimes. 13 Walsh, S.D.C., M.O. Saar, P. Bailey, and D.J. Lilja Accelerating geoscience and engineering system simulations on graphics hardware, Computers and Geosciences, 35/12, pp. 2353-2364, 2009. AbstractMany complex natural systems studied in the geosciences are characterized by simple local-scale interactions that result in complex emergent behavior. Simulations of these systems, often implemented in parallel using standard central processing unit (CPU) clusters, may be better suited to parallel processing environments with large numbers of simple processors. Such an environment is found in graphics processing units (GPUs) on graphics cards. This paper discusses GPU implementations of three example applications from computational fluid dynamics, seismic wave propagation, and rock magnetism. These candidate applications involve important numerical modeling techniques, widely employed in physical system simulations, that are themselves examples of distinct computing classes identified as fundamental to scientific and engineering computing. The presented numerical methods (and respective computing classes they belong to) are: (1) a lattice-Boltzmann code for geofluid dynamics (structured grid class); (2) a spectral-finite-element code for seismic wave propagation simulations (sparse linear algebra class); and (3) a least-squares minimization code for interpreting magnetic force microscopy data (dense linear algebra class). Significant performance increases (between 10×× and 30×× in most cases) are seen in all three applications, demonstrating the power of GPU implementations for these types of simulations and, more generally, their associated computing classes. 12 Walsh, S.D.C., H. Burwinkle, and M.O. Saar A new partial-bounceback lattice-Boltzmann method for fluid flow through heterogeneous media, Computers and Geosciences, 35/6, pp. 1186-1193, 2009. AbstractPartial-bounceback lattice-Boltzmann methods employ a probabilistic meso-scale model that varies individual lattice node properties to reflect a material’s local permeability. These types of models have great potential in a range of geofluid, and other science and engineering, simulations of complex fluid flow. However, there are several different possible approaches for formulating partial-bounceback algorithms. This paper introduces a new partial-bounceback algorithm and compares it to two pre-existing partial-bounceback models. Unlike the two other partial-bounceback methods, the new approach conserves mass in heterogeneous media and shows improvements in simulating buoyancy-driven flow as well as diffusive processes. Further, the new model is better-suited for parallel processing implementations, resulting in faster simulations. Finally, we derive an analytical expression for calculating the permeability in all three models; a critical component for accurately matching simulation parameters to physical permeabilities. 11 Walsh, S.D.C., and M.O. Saar Magma yield stress and permeability: Insights from multiphase percolation theory, Journal of Volcanology and Geothermal Research, 177, pp. 1011-1019, 2008. AbstractMagmas often contain multiple interacting phases of embedded solid and gas inclusions. Multiphase percolation theory provides a means of modeling assemblies of these different classes of magmatic inclusions in a simple, yet powerful way. Like its single phase counterpart, multiphase percolation theory describes the connectivity of discrete inclusion assemblies as a function of phase topology. In addition, multiphase percolation employs basic laws to distinguish separate classes of objects and is characterized by its dependency on the order in which the different phases appear. This paper examines two applications of multiphase percolation theory: the fi rst considers how the presence of bubble inclusions in fl uences yield stress onset and growth in a magma’s crystal network; the second examines the effect of bi-modal bubble-size distributions on magma permeability. We fi nd that the presence of bubbles induces crystal clustering, thereby 1) reducing the percolation threshold, or critical crystal volume fraction, φ c , at which the crystals form a space-spanning network providing a minimum yield stress, and 2) resulting in a larger yield stress for a given crystal volume fraction above φ c . This increase in the yield stress of the crystal network may also occur when crystal clusters areformed due toprocesses otherthanbubble formation, such as heterogeneouscrystallization, synneusis, and heterogeneity due to deformation or fl ow. Further, we fi nd that bimodal bubble size distributions can signi fi cantly affect the permeability of the system beyond the percolation threshold. This study thus demonstrates that larger-scale structures and topologies, as well as the order in which different phases appear, can have signi fi cant effects on macroscopic properties in multiphase materials. 10 Walsh, S.D.C., and M.O. Saar Numerical Models of Stiffness and Yield Stress Growth in Crystal-Melt Suspensions, Earth and Planetary Science Letters, 267/1-2, pp. 32-44, 2008. AbstractMagmas and other suspensions that develop sample-spanning crystal networks undergo a change in rheology from Newtonian to Bingham flow due to the onset of a finite yield stress in the crystal network. Although percolation theory provides a prediction of the crystal volume fraction at which this transition occurs, the manner in which yield stress grows with increasing crystal number densities is less-well understood. This paper discusses a simple numerical approach that models yield stress in magmatic crystalline assemblies. In this approach, the crystal network is represented by an assembly of soft-core interpenetrating cuboid (rectangular prism) particles, whose mechanical properties are simulated in a network model. The model is used to investigate the influence of particle shape and alignment anisotropy on the yield stress of crystal networks with particle volume fractions above the percolation threshold. In keeping with previous studies, the simulation predicts a local minimum in the onset of yield stress for assemblies of cubic particles, compared to those with more anisotropic shapes. The new model also predicts the growth of yield stress above (and close to) the percolation threshold. The predictions of the model are compared with results obtained from a critical path analysis. Good agreement is found between a characteristic stiffness obtained from critical path analysis, the growth in assembly stiffness predicted by the model (both of which have approximately cubic power-law exponents) and, to a lesser extent, the growth in yield stress (with a power-law exponent of 3.5). The effect of preferred particle alignment on yield stress is also investigated and found to obey similar power-law behavior. 9 Edwards, R.A., B. Rodriguez-Brito, L. Wegley, M. Haynes, M. Breitbart, D.M. Petersen, M.O. Saar, S.C. Alexander, E.C. Alexander Jr., and F. Rohwer Using pyrosequencing to shed light on deep mine microbial ecology, BMC Genomics, 2006. Abstract Background Contrasting biological, chemical and hydrogeological analyses highlights the fundamental processes that shape different environments. Generating and interpreting the biological sequence data was a costly and time-consuming process in defining an environment. Here we have used pyrosequencing, a rapid and relatively inexpensive sequencing technology, to generate environmental genome sequences from two sites in the Soudan Mine, Minnesota, USA. These sites were adjacent to each other, but differed significantly in chemistry and hydrogeology. Results Comparisons of the microbes and the subsystems identified in the two samples highlighted important differences in metabolic potential in each environment. The microbes were performing distinct biochemistry on the available substrates, and subsystems such as carbon utilization, iron acquisition mechanisms, nitrogen assimilation, and respiratory pathways separated the two communities. Although the correlation between much of the microbial metabolism occurring and the geochemical conditions from which the samples were isolated could be explained, the reason for the presence of many pathways in these environments remains to be determined. Despite being physically close, these two communities were markedly different from each other. In addition, the communities were also completely different from other microbial communities sequenced to date. Conclusion We anticipate that pyrosequencing will be widely used to sequence environmental samples because of the speed, cost, and technical advantages. Furthermore, subsystem comparisons rapidly identify the important metabolisms employed by the microbes in different environments. 8 Christiansen, L.B., S. Hurwitz, M.O. Saar, S.E. Ingebritsen, and P.A. Hsieh Seasonal seismicity at western United States volcanic centers, Earth and Planetary Science Letters, 240, pp. 307-321, 2005. AbstractWe examine 20-yr data sets of seismic activity from 10 volcanic areas in the western United States for annual periodic signals (seasonality), focusing on large calderas (Long Valley caldera and Yellowstone) and stratovolcanoes (Cascade Range). We apply several statistical methods to test for seasonality in the seismic catalogs. In 4 of the 10 regions, statistically significant seasonal modulation of seismicity (> 90% probability) occurs, such that there is an increase in the monthly seismicity during a given portion of the year. In five regions, seasonal seismicity is significant in the upper 3 km of the crust. Peak seismicity occurs in the summer and autumn in Mt. St. Helens, Hebgen Lake/Madison Valley, Yellowstone Lake, and Mammoth Mountain. In the eastern south moat of Long Valley caldera (LVC) peak seismicity occurs in the winter and spring. We quantify the possible external forcing mechanisms that could modulate seasonal seismicity. Both snow unloading and groundwater recharge can generate large stress changes of > 5 kPa at seismogenic depths and may thus contribute to seasonality. 7 Saar, M.O., M.C. Castro, C.M. Hall, M. Manga, and T.P. Rose Quantifying magmatic, crustal, and atmospheric helium contributions to volcanic aquifers using all stable noble gases: Implications for magmatism and groundwater flow, Geochemistry Geophysics Geosystems, 6/3, 2005. Abstract[1] We measure all stable noble gases (He, Ne, Ar, Kr, Xe) in spring waters in the Oregon Cascades volcanic arc and in eastern Oregon, USA. We show that in order to estimate magmatic helium (He) contributions it is critical to simultaneously consider He isotopic ratios, He concentrations, and mixing of He components. Our component mixing analysis requires consideration of all measured noble gases but no other elements and is particularly insightful when strong dilution by air-saturated water has occurred. In addition, this approach can allow distinction between crustal and magmatic He components and facilitates their identification in deep groundwaters that have been diluted by near-surface water. Using this approach, we show that some cold springs on the eastern flanks of the Oregon Cascades exhibit He isotopic ratios that indicate significant magmatic He contributions comparable to those observed in thermal springs on the western flanks. Furthermore, while these magmatic He contributions are largest in deep groundwaters near the Cascades crest, greater magmatic excess He fractions than may be inferred from He isotopic ratios alone are present in all (deep) groundwaters including those at larger distances (>70 km) from the volcanic arc. We also suggest that excess He and heat discharge without dilution by air-saturated water may be restricted to spring discharge along faults. 6 Saar, M.O., and M. Manga Depth dependence of permeability in the Oregon Cascades inferred from hydrogeologic, thermal, seismic, and magmatic modeling constraints, Journal of Geophysical Research, 109/B4, B04204, 2004. Abstract[1] We investigate the decrease in permeability, k, with depth, z, in the Oregon Cascades employing four different methods. Each method provides insight into the average permeability applicable to a different depth scale. Spring discharge models are used to infer shallow (z < 0.1 km) horizontal permeabilities. Coupled heat and groundwater flow simulations provide horizontal and vertical k for z < 1 km. Statistical investigations of the occurrences of earthquakes that are probably triggered by seasonal groundwater recharge yield vertical k for z < 5 km. Finally, considerations of magma intrusion rates and water devolatilization provide estimates of vertical k for z < 15 km. For depths >0.8 km, our results agree with the power law relationship, k = 10−14 m2 (z/1 km)−3.2, suggested by Manning and Ingebritsen [1999] for continental crust in general. However, for shallower depths (typically z ≤ 0.8 km and up to z ≤ 2) we propose an exponential relationship, k = 5 × 10−13 m2 exp (−z/0.25 km), that both fits data better (at least for the Cascades and seemingly for continental crust in general) and allows for a finite near-surface permeability and no singularity at zero depth. In addition, the suggested functions yield a smooth transition at z = 0.8 km, where their permeabilities and their gradients are similar. Permeabilities inferred from the hydroseismicity model at Mount Hood are about one order of magnitude larger than expected from the above power law. However, higher permeabilities in this region may be consistent with advective heat transfer along active faults, causing observed hot springs. Our simulations suggest groundwater recharge rates of 0.5 ≤ uR ≤ 1 m/yr and a mean background heat flow of Hb ≈ 0.080–0.134 W/m2 for the investigated region. 5 Jellinek, M.A., M. Manga, and M.O. Saar Did melting glaciers cause volcanic eruptions in eastern California? Probing the mechanics of dike formation, Journal of Geophysical Research, 109/B9, B09206, 2004. Abstract[1] A comparison of time series of basaltic and silicic eruptions in eastern California over the last 400 kyr with the contemporaneous global record of glaciation suggests that this volcanism is influenced by the growth and retreat of glaciers occurring over periods of about 40 kyr. Statistically significant cross correlations between changes in eruption frequency and the first derivative of the glacial time series imply that the temporal pattern of volcanism is influenced by the rate of change in ice volume. Moreover, calculated time lags for the effects of glacial unloading on silicic and basaltic volcanism are distinct and are 3.2 ± 4.2 kyr and 11.2 ± 2.3 kyr, respectively. A theoretical model is developed to investigate whether the increases in eruption frequency following periods of glacial unloading are a response ultimately controlled by the dynamics of dike formation. Applying results from the time series analysis leads, in turn, to estimates for the critical magma chamber overpressure required for eruption as well as constraints on the effective viscosity of the wall rocks governing dike propagation. 4 Saar, M.O., and M. Manga Seismicity induced by seasonal groundwater recharge at Mt. Hood, Oregon, Earth and Planetary Science Letters, 214, pp. 605-618, 2003. AbstractGroundwater recharge at Mt. Hood, Oregon, is dominated by spring snow melt which provides a natural large- amplitude and narrow-width pore-fluid pressure signal. Time delays between this seasonal groundwater recharge and seismicity triggered by groundwater recharge can thus be used to estimate large-scale hydraulic diffusivities and the state of stress in the crust. We approximate seasonal variations in groundwater recharge with discharge in runoff- dominated streams at high elevations. We interpolate the time series ofnumber ofearthquakes, N , seismic moment, M o , and stream discharge, Q , and determine cross-correlation coefficients at equivalent frequency bands between Q and both N and M o . We find statistically significant correlation coefficients at a mean time lag of about 151 days. This time lag and a mean earthquake depth of about 4.5 km are used in the solution to the pressure diffusion equation, under periodic (1 year) boundary conditions, to estimate a hydraulic diffusivity of U W 10 3 1 m 2 /s, a hydraulic conductivity ofabout K h W 10 3 7 m/s, and a permeability ofabout k W 10 3 15 m 2 . Periodic boundary conditions also allow us to determine a critical pore-fluid pressure fraction, P P / P 0 W 0.1, ofthe applied near-surface pore-fluid pressure perturbation, P 0 W 0.1 MPa, that has to be reached at the mean earthquake depth to cause hydroseismicity. The low magnitude of P P W 0.01 MPa is consistent with other studies that propose 0.01 9 P P 9 0.1 MPa and suggests that the state of stress in the crust near Mt. Hood could be near critical for failure. Therefore, we conclude that, while earthquakes occur throughout the year at Mt. Hood, elevated seismicity levels along pre-existing faults south of Mt. Hood during summer months are hydrologically induced by a reduction in effective stress 3 Saar, M.O., and M. Manga Continuum percolation for randomly oriented soft-core prisms, Physical Review E, 65/056131, 2002. AbstractWe study continuum percolation of three-dimensional randomly oriented soft-core polyhedra (prisms). The prisms are biaxial or triaxial and range in aspect ratio over six orders of magnitude. Results for prisms are compared with studies for ellipsoids, rods, ellipses, and polygons and differences are explained using the concept of the average excluded volume, ⟨vex⟩. For large-shape anisotropies we find close agreement between prisms and most of the above-mentioned shapes for the critical total average excluded volume, nc⟨vex⟩, where nc is the critical number density of objects at the percolation threshold. In the extreme oblate and prolate limits simulations yield nc⟨vex⟩≈2.3 and nc⟨vex⟩≈1.3, respectively. Cubes exhibit the lowest-shape anisotropy of prisms minimizing the importance of randomness in orientation. As a result, the maximum prism value, nc⟨vex⟩≈2.79, is reached for cubes, a value close to nc⟨vex⟩=2.8 for the most equant shape, a sphere. Similarly, cubes yield a maximum critical object volume fraction of φc=0.22. φc decreases for more prolate and oblate prisms and reaches a linear relationship with respect to aspect ratio for aspect ratios greater than about 50. Curves of φc as a function of aspect ratio for prisms and ellipsoids are offset at low-shape anisotropies but converge in the extreme oblate and prolate limits. The offset appears to be a function of the ratio of the normalized average excluded volume for ellipsoids over that for prisms, R=⟨¯vex⟩e/⟨¯vex⟩p. This ratio is at its minimum of R=0.758 for spheres and cubes, where φc(sphere)=0.2896 may be related to φc(cube)=0.22 by φc(cube)=1−[1−φc(sphere)]R=0.23. With respect to biaxial prisms, triaxial prisms show increased normalized average excluded volumes, ⟨¯vex⟩, due to increased shape anisotropies, resulting in reduced values of φc. We confirm that Bc=nc⟨vex⟩=2Cc applies to prisms, where Bc and Cc are the average number of bonds per object and average number of connections per object, respectively. 2 Saar, M.O., M. Manga, K.V. Cashman, and S. Fremouw Numerical models of the onset of yield strength in crystal-melt suspensions, Earth and Planetary Science Letters, 187, pp. 367-379, 2001. AbstractThe formation of a continuous crystal network in magmas and lavas can provide finite yield strength, τy, and can thus cause a change from Newtonian to Bingham rheology. The rheology of crystal–melt suspensions affects geological processes, such as ascent of magma through volcanic conduits, flow of lava across the Earth’s surface, melt extraction from crystal mushes under compression, convection in magmatic bodies, and shear wave propagation through partial melting zones. Here, three-dimensional numerical models are used to investigate the onset of ‘static’ yield strength in a zero-shear environment. Crystals are positioned randomly in space and can be approximated as convex polyhedra of any shape, size and orientation. We determine the critical crystal volume fraction, φc, at which a crystal network first forms. The value of φc is a function of object shape and orientation distribution, and decreases with increasing randomness in object orientation and increasing shape anisotropy. For example, while parallel-aligned convex objects yield φc=0.29, randomly oriented cubes exhibit a maximum φc of 0.22. Approximations of plagioclase crystals as randomly oriented elongated and flattened prisms (tablets) with aspect ratios between 1:4:16 and 1:1:2 yield 0.08<φc<0.20, respectively. The dependence of φc on particle orientation implies that the flow regime and resulting particle ordering may affect the onset of yield strength. φc in zero-shear environments is a lower bound for φc. Finally the average total excluded volume is used, within its limitation of being a ‘quasi-invariant’, to develop a scaling relation between τy and φ for suspensions of different particle shapes. 1 Saar, M.O., and M. Manga Permeability-porosity relationship in vesicular basalts., Geophysical Research Letters, 26/1, pp. 111-114, 1999. AbstractThe permeability κ and porosity ϕ of vesicular basalts are measured. The relationship between κ and ϕ reflects the formation and emplacement of the basalts and can be related to the crystal and vesicle microstructure obtained by image analysis. Standard theoretical models relating κ and ϕ that work well for granular materials are unsuccessful for vesicular rocks due to the fundamental difference in pore structure. Specifically, κ in vesicular rocks is governed by apertures between bubbles. The difference between calculated and measured κ reflects the small size of these apertures with aperture radii typically O(10) times smaller than the mean bubble radii.

### PROCEEDINGS REFEREED

 10 Myre, J.M., E. Frahm, D.J. Lilja, and M.O. Saar Solving Large Dense Least-Squares Problems: Preconditioning to Take Conjugate Gradient From Bad in Theory, to Good in Practice, IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), pp. 987-995, 2018. AbstractSince its inception by Gauss, the least-squares problem has frequently arisen in science, mathematics, and engineering. Iterative methods, such as Conjugate Gradient Normal Residual (CGNR), have been popular for solving sparse least-squares problems, but have historically been regarded as undesirable for dense applications due to poor convergence. We contend that this traditional “common knowledge” should be reexamined. Preconditioned CGNR, and perhaps other iterative methods, should be considered alongside standard methods when addressing large dense least-squares problems. In this paper we present TNT, a dynamite method for solving large dense least-squares problems. TNT implements a Cholesky preconditioner for the CGNR fast iterative method. The Cholesky factorization provides a preconditioner that, in the absence of round-off error, would yield convergence in a single iteration. Through this preconditioner and good parallel scaling, TNT provides improved performance over traditional least-squares solvers allowing for accelerated investigations of scientific and engineering problems. We compare a parallel implementations of TNT to parallel implementations of other conventional methods, including the normal equations and the QR method. For the small systems tested (15000 × 15000 or smaller), it is shown that TNT is capable of producing smaller solution errors and executing up to 16× faster than the other tested methods. We then apply TNT to a representative rock magnetism inversion problem where it yields the best solution accuracy and execution time of all tested methods. 9 Garapati, N., B.M. Adams, J.M. Bielicki, P. Schaedle, J.B. Randolph, T.H. Kuehn, and M.O. Saar A Hybrid Geothermal Energy Conversion Technology – A Potential Solution for Production of Electricity from Shallow Geothermal Resources, Energy Procedia, 114, pp. 7107-7117, 2017. AbstractGeothermal energy has been successfully employed in Switzerland for more than a century for direct use but presently there is no electricity being produced from geothermal sources. After the nuclear power plant catastrophe in Fukushima, Japan, the Swiss Federal Assembly decided to gradually phase out the Swiss nuclear energy program. Deep geothermal energy is a potential resource for clean and nearly CO2-free electricity production that can supplant nuclear power in Switzerland and worldwide. Deep geothermal resources often require enhancement of the permeability of hot-dry rock at significant depths (4-6 km), which can induce seismicity. The geothermal power projects in the Cities of Basel and St. Gallen, Switzerland, were suspended due to earthquakes that occurred during hydraulic stimulation and drilling, respectively. Here we present an alternative unconventional geothermal energy utilization approach that uses shallower, lower-temperature, naturally permeable regions, that drastically reduce drilling costs and induced seismicity. This approach uses geothermal heat to supplement a secondary energy source. Thus this hybrid approach may enable utilization of geothermal energy in many regions in Switzerland and elsewhere, that otherwise could not be used for geothermal electricity generation. In this work, we determine the net power output, energy conversion efficiencies, and economics of these hybrid power plants, where the geothermal power plant is actually a CO2-based plant. Parameters varied include geothermal reservoir depth (2.5-4.5 km) and turbine inlet temperature (100-220 °C) after auxiliary heating. We find that hybrid power plants outperform two individual, i.e., stand-alone geothermal and waste-heat power plants, where moderate geothermal energy is available. Furthermore, such hybrid power plants are more economical than separate power plants. 8 Rossi, E., M. Kant, F. Amann, M.O. Saar, and P. Rudolf von Rohr The effects of flame-heating on rock strength: Towards a new drilling technology, Proceedings ARMA 2017, 2017. AbstractThe applicability of a combined thermo-mechanical drilling technique is investigated. The working principle of this method is based on the implementation of a heat source as a mean to either provoke thermal spallation on the surface or to weaken the rock material, when spallation is not possible. Thermal spallation drilling has already been proven to work in hard crystalline rocks, however, several difficulties hamper its application for deep resource exploitation. In order to prove the effectiveness of a combined thermo-mechanical drilling method, the forces required to export the treated sandstone material with a polycrystalline diamond compact (PDC) cutter are analyzed. The main differences between oven and flame treatments are studied by comparing the resulting strength after heat-treating the samples up to temperatures of $$650\, ^{\circ}C$$ and for heating rates ranging from $$0.17 \,^{\circ}C/s$$ to $$20 ^{\circ}C/s$$. For moderate temperatures ($$300-450 \,^{\circ}C$$) the unconfined compressive strength after flame treatments monotonously decreased, opposed to the hardening behavior observed after oven treatments. Thermally induced intra-granular cracking and oxidation patterns served as an estimation of the treated depth due to the flame heat treatment. Therefore, conclusions on preferred operating conditions of the drilling system are drawn based on the experimental results. 7 Vogler, D., R.R. Settgast, C.S. Sherman, V.S. Gischig, R. Jalali, J.A. Doetsch, B. Valley, K.F. Evans, F. Amann, and M.O. Saar Modeling the Hydraulic Fracture Stimulation performed for Reservoir Permeability Enhancement at the Grimsel Test Site, Switzerland, Proceedings of the 42nd Workshop on Geothermal Reservoir Engineering Stanford University, 2017. AbstractIn-situ hydraulic fracturing has been performed on the decameter scale in the Deep Underground rock Laboratory (DUG Lab) at the Grimsel Test Site (GTS) in Switzerland in order to measure the minimum principal stress magnitude and orientation. Conducted tests were performed in a number of boreholes, with 3–4 packer intervals in each borehole subjected to repeated injection. During each test, fluid injection pressure, injection flow rate and microseismic events were recorded amongst others. Fully coupled 3D simulations have been performed with the LLNL’s GEOS simulation framework. The methods applied in the simulation of the experiments address physical processes such as rock deformation/stress, LEFM fracture mechanics, fluid flow in the fracture and matrix, and the generation of micro-seismic events. This allows to estimate the distance of fracture penetration during the injection phase and correlate the simulated injection pressure with experimental data during injection, as well as post shut-in. Additionally, the extent of the fracture resulting from simulations of fracture propagation and microseismic events are compared with the spatial distribution of the microseismic events recorded in the experiment. 6 Garapati, N., J. Randolph, S. Finsterle, and M.O. Saar Simulating Reinjection of Produced Fluids Into the Reservoir, Proceedings of 41st Workshop on Geothermal Reservoir Engineering, 2016. AbstractABSTRACT In order to maintain reservoir pressure and stability and to reduce reservoir s ubsidence, reinjection of produced fluids into the reservoir is common practice . Furthermore, studies by Karvounis and Jenny (2012 ; 2014), Buscheck et al. (2015), and Saar et al. (2015) found that preheating the working fluid in shallow reservoirs and then injecting the fluid into a deep reservoir can increase the reservoir life span, the heat extraction efficiency, and the economic gains of a geothermal power plant . We have modif ied the TOUGH2 simulator to enable the reinjection of produced fluids with the same chemical composition as the produced fluid and with either a prescribed or the production temperature . T he latter capability is useful, for example, for simulating injecti on of produced fluid into another (e .g., deeper) reservoir without energy extraction. Each component of the fluid mixture , produced from the production well , is reinjected into the reservoir as an individual source term. In the current study, we investigate a CO 2 – based geothermal system and focus on the effects of reinjecting small amounts of brine that are produced along with the CO 2 . Brine has a significantly smaller mobility (inverse kinematic viscosity) than supercritical CO 2 at a given temperature and thus accumulates near the injection well. Such brine accumulation reduces the relative permeability for the CO 2 phase, which in turn increases the pore – fluid pressure around the injection well and reduces the well in j ectivity index. For this reason, and as injection of two fluid phases is pr oblematic, we recommend removal of any brine from the produced fluid before the cooled CO 2 is reinjected into the reservoir. We also study the performance of a multi – level geothermal system (Karvounis and Jenny, 2012; 2014; Saar et al., 2015) by injection of preheated brine from a shallow reservoir (1.5 – 3 km) into a deep reservoir (5 km). We f i nd that preheating brine at the shallow reservoir extends the lifespan of the deep, hot reservoir, thereby increasing the total power production. 5 Garapati, N., J.B. Randolph, J.L. Valencia Jr., and M.O. Saar Design of CO2-Plume Geothermal (CPG) subsurface system for various geologic parameters, Proceedings of the Fifth International Conference on Coupled Thermo-Hydro-Mechanical-Chemical (THMC) Processes in Geosystems: Petroleum and Geothermal Reservoir Geomechanics and Energy Resource Extraction, 2015. AbstractRecent geotechnical research shows that geothermal heat can be efficiently mined by circulating carbon dioxide through naturally permeable rock formations — a method called CO2 Plume Geothermal — the same geologic reservoirs that are suitable for deep saline aquifer CO2 sequestration or enhanced oil recovery. This paper describes the effect of thermal drawdown on reservoir pressure buildup during sequestration operations, revealing that geothermal heat mining can decrease overpressurization by 10% or more. Geothermal Energy Production at Geologic CO2 Sequestration sites: Impact of Thermal Drawdown on Reservoir Pressure (PDF Download Available). Available from: https://www.researchgate.net/publication/273193986_Geothermal_Energy_Production_at_Geologic_CO2_Sequestration_sites_Impact_of_Thermal_Drawdown_on_Reservoir_Pressure [accessed Jun 12, 2017]. 4 Garapati, N., J.B. Randolph, and M.O. Saar Superheating Low-Temperature Geothermal Resources to Boost Electricity Production, Proceedings of the 40th Workshop on Geothermal Reservoir Engineering 2015, 2, pp. 1210-1221, 2015. AbstractLow-temperature geothermal resources (<150°C) are typically more effective for direct use, i.e., district heating, than for electricity production. District or industrial heating, however, requires that the heat resource is close to residential or industrial demands in order to be efficient and thus economic. However, if a low-temperature geothermal resource is combined with an additional or secondary energy source that is ideally renewable, such as solar, biomass, biogas, or waste heat, but could be non-renewable, such as natural gas, the thermodynamic quality of the energy source increases, potentially enabling usage of the combined energy sources for electricity generation. Such a hybrid geothermal power plant therefore offers thermodynamic advantages, often increasing the overall efficiency of the combined system above that of the additive power output from two stand-alone, separate plants (one using geothermal energy alone and the other using the secondary energy source alone) for a wide range of operating conditions. Previously, fossil superheated and solar superheated hybrid power plants have been considered for brine/water based geothermal systems, especially for enhanced geothermal systems. These previous studies found, that the cost of electricity production can typically be reduced when a hybrid plant is operated, compared to operating individual plants. At the same time, using currently-available high-temperature energy conversion technologies reduces the time and cost required for developing other less-established energy conversion technologies. Adams et al. (2014) found that CO 2 as a subsurface working fluid produces more net power than when brine systems are employed at low to moderate reservoir depths, temperatures, and permeabilities. Therefore in this work, we compare the performance of hybrid geothermal power plants that use brine or, importantly, CO 2 (which constitutes the new research component) as the subsurface working fluid, irrespective of the secondary energy source used for superheating, over a range of parameters. These parameters include geothermal reservoir depth and superheated fluid temperature before passing through the energy conversion system. The hybrid power plant is modeled using two software packages: 1) TOUGH2 (Pruess, 2004), which is employed for the subsurface modeling of geothermal heat and fluid extraction as well as for fluid reinjection into the reservoir, and 2) Engineering Equation Solver (EES), which is used to simulate well bore fluid flow and surface power plant performance. We find here that for geothermal systems combined with a secondary energy source (i.e., a hybrid system), the maximum power production for a given set of reservoir parameters is highly dependent on the configuration of the power system. The net electricity production from a hybrid system is larger than that from the individual plants combined for all scenarios considered for brine systems and for low-grade secondary energy resources for CO 2 based geothermal systems. Superheating of Low-Temperature Geothermal Working Fluids to Boost Electricity Production: Comparison between Water and CO2 Systems (PDF Download Available). Available from: https://www.researchgate.net/publication/271702360_Superheating_of_Low-Temperature_Geothermal_Working_Fluids_to_Boost_Electricity_Production_Comparison_between_Water_and_CO2_Systems [accessed Jun 12, 2017]. 3 Saar, M.O., Th. Buscheck, P. Jenny, N. Garapati, J.B. Randolph, D. Karvounis, M. Chen, Y. Sun, and J.M. Bielicki Numerical Study of Multi-Fluid and Multi-Level Geothermal System Performance, Proceedings World Geothermal Congress 2015, 2015. AbstractWe introduce the idea of combining multi-fluid and multi-level geothermal systems with two reservoirs at depths of 3 and 5 km. In the base case, for comparison, the two reservoirs are operated independently, each as a multi-fluid (brine and carbon dioxide) reservoir that uses a number of horizontal, concentric injection and production well rings. When the shallow and the deep reservoirs are operated in an integrated fashion, in the shallow reservoir, power is produced only from the carbon dioxide (CO 2), while the brine is geothermally preheated in the shallow multi-fluid reservoir, produced, and then reinjected at the deeper reservoir’s brine injectors. The integrated reservoir scenarios are further subdivided into two cases: In one scenario, both brine (preheated in the shallow reservoir) and CO 2 (from the surface) are injected separately into the deeper reservoir’s appropriate injectors and both fluids are produced from their respective deep reservoir producers to generate electricity. In the other scenario, only preheated brine is injected into, and produced from, the deep reservoir for electric power generation. We find that integrated, vertically stacked, multi-fluid geothermal systems can result in improved system efficiency when power plant lifespans exceed ~30 years. In addition, preheating of brine before deep injection reduces brine overpressurization in the deep reservoir, reducing the risk of fluid-induced seismicity. Furthermore, CO2-Plume Geothermal (CPG) power plants in general, and the multi-fluid, multi-level geothermal system described here in particular, assign a value to CO2, which in turn may partially or fully offset the high costs of carbon capture at fossil-energy power plants and of CO2 injection, thereby facilitating economically feasible carbon capture and storage (CCS) operations that render fossil-energy power plants green. From a geothermal power plant perspective, the system results in a CO2 sequestering geothermal power plant with a negative carbon footprint. Finally, energy return on well costs and operational flexibility can be greater for integrated geothermal reservoirs, providing additional options for bulk and thermal energy storage, compared to equivalent, but separately operated reservoirs. System economics can be enhanced by revenues related to efficient delivery of large-scale bulk energy storage and ancillary services products (frequency regulation, load following, and spinning reserve), which are essential for electric grid integration of intermittently available renewable energy sources, such as wind and solar. These capabilities serve to stabilize the electric grid and promote development of all renewable energies, beyond geothermal energy. Numerical Study of Multi-Fluid and Multi-Level Geothermal System Performance (PDF Download Available). Available from: https://www.researchgate.net/publication/274138343_Numerical_Study_of_Multi-Fluid_and_Multi-Level_Geothermal_System_Performance [accessed Jun 12, 2017]. 2 Buscheck, T.A., J.M. Bielicki, M. Chen, Y. Sun, Y. Hao, T.A. Edmunds, J.B. Randolph, and M.O. Saar Multi-Fluid Sedimentary Geothermal Energy Systems for Dispatchable Renewable Electricity, Proceedings to the World Geothermal Congress, 2015. AbstractSedimentary geothermal resources typically have lower temperatures and energy densities than hydrothermal resources, but they often have higher permeability and larger areal extents. Consequently, spacing between injection and production wells is likely to be wider in sedimentary resources, which can result in more fluid pressure loss, increa sing the parasitic cost of powering the working fluid recirculation system, compared to hydrothermal systems . For hydrostatic geothermal resources , extracting heat requires that brine be lifted up production wells, such as with submersible pumps, which can consume a large portion of the electricity generated by the power plant. CO 2 is being considered as an alternative working fluid (also termed a supplemental fluid) because its advantageous thermophysical properties reduce this parasitic cost, and because of the synergistic benefit of geologic CO 2 sequestration (GCS). We expand on this idea by: (1) adding the option for multiple supplemental fluids (N 2 as well as CO 2 ) and injecting these fluids to create overpressured reservoir conditions , (2) utiliz ing up to three working fluids: brine, CO 2 , and N 2 for heat extraction, (3) using a well pattern designed to store supplemental fluid and pressure , and (4) time – shifting the parasitic load associated with fluid recirculation to provide ancillary services ( frequen cy regulat ion , load fo llowing , and spinning reserve) and bulk energy storage (BES) . Our approach uses concentric rings of horizontal wells to create a hydraulic divide to store supplemental fluid and pressure, much like a hydroelectric dam. While, as with any geothermal system, electricity production can be run as a base – load power source, p roduction wells can alternatively b e controlled like a spillway to supply power when demand is greatest. For conventional geothermal power, the parasitic power load for fluid recirculation is synchronous with gross power output. In contrast, our approach time – shift s much of this parasitic load, which is dominated by the power required to pressurize and inject brine . Th us, most of the parasitic load can be scheduled durin g minimum power demand or when, due to its inherent var iability, there is a surplus of renewable energy on the grid . Energy storage is almost 100 percent efficient because it is achieved by time – shifting the parasitic load. Consequently, net power can near ly equal gross power during peak demand so that geothermal energy can be used as a form of high – efficiency BES at large scales . A further benefit of our approach is that production rates (per well) can exceed the capacity of submersible pumps and thereby t ake advantage of the productivity of horizontal wells and better leverage we ll costs — which often constitute a major portion of capital costs . Our vision is a n efficient, dispatchable , renewable electricity system approach that facilitates deep market penet ration of all renewable energy sources: wind, solar, and geothermal, whi le utilizing and permanently storing CO 2 in a commercially viable manner 1 Bailey, P., J. Myre, S.C.D. Walsh, D.J. Lilja, and M.O. Saar Accelerating Lattice Boltzmann Fluid Flow Simulations Using Graphics Processors, IEEE, pp. 550-557, 2009. AbstractLattice Boltzmann Methods (LBM) are used for the computational simulation of Newtonian fluid dynamics. LBM-based simulations are readily parallelizable; they have been implemented on general-purpose processors, field-programmable gate arrays (FPGAs), and graphics processing units (GPUs). Of the three methods, the GPU implementations achieved the highest simulation performance per chip. With memory bandwidth of up to 141 GB/s and a theoretical maximum floating point performance of over 600 GFLOPS, CUDA-ready GPUs from NVIDIA provide an attractive platform for a wide range of scientific simulations, including LBM. This paper improves upon prior single-precision GPU LBM results for the D3Q19 model by increasing GPU multiprocessor occupancy, resulting in an increase in maximum performance by 20%, and by introducing a space-efficient storage method which reduces GPU RAM requirements by 50% at a slight detriment to performance. Both GPU implementations are over 28 times faster than a single-precision quad-core CPU version utilizing OpenMP.

### THESES

 2 Saar, M.O. Geological Fluid Mechanics Models at Various Scales, Dissertation University of California, Berkeley, 153 pp., 2003. AbstractIn this dissertation, I employ concepts from fluid mechanics to quantitatively investigate geological processes in hydrogeology and volcanology. These research topics are addressed by utilizing numerical and analytical models but also by conducting field and lab work. Percolation theory is of interest to a wide range of physical sciences and thus warrants research in itself. Therefore, I developed a computer code to study percolation thresholds of soft-core polyhedra. Results from this research are applied to study the onset of yield strength in crystal-melt suspensions such as magmas. Implications of yield strength development in suspensions, marking the transition from Newtonian to Bingham fluid, include the pahoehoe-‘a’a transition and the occurrence of effusive versus explosive eruptions. I also study interactions between volcanic processes and groundwater as well as between groundwater and seismicity (hydroseismicity). In the former case, I develop numerical and analytical models of coupled groundwater and heat transfer. Here, perturbations from a linear temperature-depth profile are used to determine groundwater flow patterns and rates. For the hydroseismicity project I investigate if seasonal elevated levels of seismicity at Mt. Hood, Oregon, are triggered by groundwater recharge. Both hydroseismicity and hydrothermal springs occur on the southern flanks of Mt. Hood. This suggests that both phenomena are related while also providing a connection between the research projects involving groundwater, heat flow, and seismicity. Indeed, seismicity may be necessary to keep faults from clogging thus allowing for sustained activity of hydrothermal springs. Finally, I present research on hydrologically induced volcanism, where a process similar to the one suggested for hydroseismicity is invoked. Here, melting of glaciers, or draining of lakes, during interglacial periods reduce the confining pressure in the subsurface which may promote dike formation and result in increased rates of volcanism. In general, problems discussed in this dissertation involve interactions among processes that are traditionally associated with separate research disciplines. However, numerous problems in the geosciences require a multidisciplinary approach, as demonstrated here. In addition, employing several analytical and numerical methods, such as signal processing, inverse theory, computer modeling, and percolation theory, allows me to study such diverse processes in a quantitative way. 1 Saar, M.O. The Relationship Between Permeability, Porosity, and Microstucture in Vesicular Basalts, MSc Thesis University of Oregon, 101 pp., 1998. AbstractThis thesis presents measurements of permeability, $$k$$, porosity, $$\Phi$$, and microstructural parameters of vesicular basalts. Measurements are compared with theoretical models. A percolation theory and a Kozeny-Carman model are used to interpret the measurements and to investigate relationships between porosity, microstructure, and permeability. Typical permeabilities for vesicular basalts are in the range of $$10^{-14}$$ < $$k$$ < $$10^{-10} m^2$$. Best permeability estimates, following power laws predicted by percolation theory, are obtained when samples are used that show 'impeded aperture widening' due to rapid cooling and no bubble collapse (scoria and some flow samples). However, slowly cooled diktytaxitic samples contain elongated, 'collapsed' bubbles. Measurements indicate that the vesicle pathway network remains connected and preserves high permeabilities. Image-analysis techniques are unsuccessful if used for Kozeny-Carman equation parameter determination for vesicular materials, probably because the average interbubble aperture size that determines $$k$$ is not resolved with such a technique.

show/hide list of publications

### REFEREED PUBLICATIONS IN JOURNALS

 66 Ogland-Hand, J.D., J.M. Bielicki, Y. Wang, B.M. Adams, T.A. Buscheck, and M.O. Saar The value of bulk energy storage for reducing CO2 emissions and water requirements from regional electricity systems., Energy Conversion and Management, 181, pp. 674-685, 2019. AbstractThe implementation of bulk energy storage (BES) technologies can help to achieve higher penetration and utilization of variable renewable energy technologies (e.g., wind and solar), but it can also alter the dispatch order in regional electricity systems in other ways. These changes to the dispatch order affect the total amount of carbon dioxide (CO2) that is emitted to the atmosphere and the amount of total water that is required by the electricity generating facilities. In a case study of the Electricity Reliability Council of Texas system, we separately investigated the value that three BES technologies (CO2- Geothermal Bulk Energy Storage, Compressed Air Energy Storage, Pumped Hydro Energy Storage) could have for reducing system-wide CO2 emissions and water requirements. In addition to increasing the utilization of wind power capacity, the dispatch of BES also led to an increase in the utilization of natural gas power capacity and of coal power capacity, and a decrease in the utilization of nuclear power capacity, depending on the character of the net load, the CO2 price, the water price, and the BES technology. These changes to the dispatch order provided positive value (e.g., increase in natural gas generally reduced CO2 emissions; decrease in nuclear utilization always decreased water requirements) or negative value (e.g., increase in coal generally increased CO2 emissions; increase in natural gas sometimes increased water requirements) to the regional electricity system. We also found that these values to the system can be greater than the cost of operating the BES facility. At present, there are mechanisms to compensate BES facilities for ancillary grid services, and our results suggest that similar mechanisms could be enacted to compensate BES facilities for their contribution to the environmental sustainability of the system. 65 Dambly, M.L.T., M. Nejati, D. Vogler, and M.O. Saar On the direct measurement of shear moduli in transversely isotropic rocks using the uniaxial compression test, International Journal of Rock Mechanics and Mining Sciences (IJRMMS), 113, pp. 220-240, 2019. AbstractThis paper introduces a methodology for the direct determination of the shear moduli in transversely isotropic rocks, using a single test, where a cylindrical specimen is subjected to uniaxial compression. A method is also developed to determine the orientation of the isotropy plane as well as the dynamic elastic constants using ultrasonic measurements on a single cylindrical specimen. Explicit formulae are developed to calculate the shear moduli from strain gauge measurements at different polar angles. The calculation of shear moduli from these formulae requires no knowledge about Young’s moduli or Poisson’s ratios and depends only on the orientation of the isotropy plane. Several strain gauge setups are designed to obtain the shear moduli from different numbers and arrangements of strain gauges. We demonstrate, that the shear moduli can be determined accurately and efficiently with only three strain gauge measurements. The orientation of the isotropy plane is measured with different methods, including ultrasonic measurements. The results show, that the isotropy plane of the tested granitic samples slightly deviates from the foliation plane. However, the foliation plane can still determine the orientation of the isotropy plane with a good approximation. 64 Myre, J.M., I. Lascu, E.A. Lima, J.M. Feinberg, M.O. Saar, and B.P. Weiss Using TNT-NN to Unlock the Fast Full Spatial Inversion of Large Magnetic Microscopy Datasets, Earth Planets and Space, 2019. AbstractModern magnetic microscopy (MM) provides high-resolution, ultra-high sensitivity moment magnetometry, with the ability to measure at spatial resolutions better than 10^−4 m and to detect magnetic moments weaker than 10^−15 Am^2 . These characteristics make modern MM devices capable of particularly high resolution analysis of the magnetic properties of materials, but generate extremely large data sets. Many studies utilizing MM attempt to solve an inverse problem to determine the magnitude of the magnetic moments that produce the measured component of the magnetic field. Fast Fourier techniques in the frequency domain and non-negative least-squares (NNLS) methods in the spatial domain are the two most frequently used methods to solve this inverse problem. Although extremely fast, Fourier techniques can produce solutions that violate the non-negativity of moments constraint. Inversions in the spatial domain do not violate non-negativity constraints, but the execution times of standard NNLS solvers (the Lawson and Hanson method and Matlab’s lsqlin) prohibit spatial domain inversions from operating at the full spatial resolution of an MM. In this paper we present the applicability of the TNT-NN algorithm, a newly developed NNLS active set method, as a means to directly address the NNLS routine hindering existing spatial domain inversion methods. The TNT-NN algorithm enhances the performance of spatial domain inversions by accelerating the core NNLS routine. Using a conventional computing system, we show that the TNT-NN algorithm produces solutions with residuals comparable to conventional methods while reducing execution time of spatial domain inversions from months to hours or less. Using isothermal remanent magnetization measurements of multiple synthetic and natural samples, we show that the capabilities of the TNT-NN algorithm allow scans with sizes that made them previously inaccesible to NNLS techniques to be inverted. Ultimately, the TNT- NN algorithm enables spatial domain inversions of MM data on an accelerated timescale that renders spatial domain analyses for modern MM studies practical. In particular, this new technique enables MM experiments that would have required an impractical amount of inversion time such as high-resolution stepwise magnetization and demagnetization and 3-dimensional inversions. 63 Kant, M.A., E. Rossi, J. Duss, F. Amman, M.O. Saar, and P. Rudolf von Rohr Demonstration of thermal borehole enlargement to facilitate controlled reservoir engineering for deep geothermal, oil or gas systems, Applied Energy, 212, pp. 1501-1509, 2018. AbstractThe creation of deep reservoirs for geothermal energy or oil and gas extraction is impeded by insu cient stimulation. Direction and extension of the created fractures are complex to control and, therefore, large stimulated and interconnected fracture networks are di cult to create. This poses an inherent risk of un- economic reservoirs, due to insu cient heat-sweep surfaces or hydraulic shortcuts. Therefore, we present a new technique, which locally increases the cross section of a borehole by utilizing a thermal spallation process on the sidewalls of the borehole. By controlled and local enlargement of the well bore diameter, initial fracture sources are created, potentially reducing the injection pressure during stimulation, initiating fracture growth, optimizing fracture propagation and increasing the number of accessible preexisting frac- tures. Consequently, local thermal borehole enlargement reduces project failure risks by providing better control on subsequent stimulation processes. In order to show the applicability of the suggested technique, we conducted a shallow field test in an underground rock laboratory. Two types of borehole enlargements were created in a 14.5 m deep borehole, confirming that the technology is applicable, with implications for improving the productivity of geothermal, oil and gas reservoirs. 62 Ahkami, M., T. Roesgen, M.O. Saar, and X.-Z. Kong High-resolution temporo-ensemble PIV to resolve pore-scale flow in 3D-printed fractured porous media, Transport in Porous Media, pp. 1-17, 2018. AbstractFractures are conduits that can enable fast advective transfer of (fluid, solute, reactant, particle, etc.) mass and energy. Such fast transfer can significantly affect pore-scale physico-chemical processes, which can in turn affect macroscopic mass and energy transport characteristics. Here, flooding experiments are conducted in a well-characterized fractured porous medium, manufactured by 3D printing. Given steady-state flow conditions, the micro-structure of the two-dimensional (2D) pore fluid flow field is delineated to resolve fluid velocities on the order of a sub-millimeter per second. We demonstrate the capabilities of a new temporo-ensemble Particle Image Velocimetry (PIV) method by maximizing its spatial resolution, employing in-line illumination. This method is advantageous as it is capable of minimizing the number of pixels, required for velocity determinations, down to one pixel, thereby enabling resolving high spatial resolutions of velocity vectors in a large field of view (FOV). While the main goal of this study is to introduce a novel experimental and velocimetry framework, this new method is then applied to specifically improve the understanding of fluid flow through fractured porous media. Histograms of measured velocities indicate log-normal and Gaussian-type distributions of longitudinal and lateral velocities in fractures, respectively. The magnitudes of fluid velocities in fractures and the flow interactions between fractures and matrices are shown to be influenced by the permeability of the background matrix and the orientation of the fractures. 61 Kong, X.-Z., C. Deuber, A. Kittilä, M. Somogyvari, G. Mikutis, P. Bayer, W.J. Stark, and M.O. Saar Tomographic reservoir imaging with DNA-labeled silica nanotracers: The first field validation, Environmental Science &Technology, 52/23, pp. 13681-13689, 2018. AbstractThis study presents the first field validation of using DNA-labeled silica nanoparticles as tracers to image subsurface reservoirs by travel time based tomography. During a field campaign in Switzerland, we performed short-pulse tracer tests under a forced hydraulic head gradient to conduct a multisource−multireceiver tracer test and tomographic inversion, determining the two-dimensional hydraulic conductivity field between two vertical wells. Together with three traditional solute dye tracers, we injected spherical silica nanotracers, encoded with synthetic DNA molecules, which are protected by a silica layer against damage due to chemicals, microorganisms, and enzymes. Temporal moment analyses of the recorded tracer concentration breakthrough curves (BTCs) indicate higher mass recovery, less mean residence time, and smaller dispersion of the DNA-labeled nanotracers, compared to solute dye tracers. Importantly, travel time based tomography, using nanotracer BTCs, yields a satisfactory hydraulic conductivity tomogram, validated by the dye tracer results and previous field investigations. These advantages of DNA-labeled nanotracers, in comparison to traditional solute dye tracers, make them well-suited for tomographic reservoir characterizations in fields such as hydrogeology, petroleum engineering, and geothermal energy, particularly with respect to resolving preferential flow paths or the heterogeneity of contact surfaces or by enabling source zone characterizations of dense nonaqueous phase liquids. 60 Samrock, F., A.V. Grayver, H. Eysteinsson, and M.O. Saar Magnetotelluric image of transcrustal magmatic system beneath the Tulu Moye geothermal prospect in the Ethiopian Rift, Geophysical Research Letters, 2018. AbstractContinental rifting is initiated by a dynamic interplay between tectonic stretching and mantle upwelling. Decompression melting assists continental break-up through lithospheric weakening and enforces upflow of melt to the Earth’s surface. However, the details about melt transport through the brittle crust and storage under narrow rift-aligned magmatic segments remain largely unclear. Here we present a crustal scale electrical conductivity model for a magmatic segment in the Ethiopian Rift, derived from 3-D phase tensor inversion of magnetotelluric data. Our subsurface model shows that melt migrates along pre-existing weak structures and is stored in different concentrations on two major interconnected levels, facilitating the formation of a convective hydrothermal system. The obtained model of a transcrustal magmatic system offers new insights into rifting mechanisms, evolution of magma ascent, and prospective geothermal reservoirs. 59 Amann, F., V. Gischig, K. Evans, et al., A. Kittilä, S. Wiemer, M.O. Saar, S. Löw, Th. Driesner, H. Maurer, and D. Giardini The seismo-hydro-mechanical behaviour during deep geothermal reservoir stimulations: open questions tackled in a decameter-scale in-situ stimulation experiment, Solid Earth, pp. 115-137, 2018. AbstractIn this contribution, we present a review of scientific research results that address seismo-hydromechanically coupled processes relevant for the development of a sustainable heat exchanger in low-permeability crystalline rock and introduce the design of the In situ Stimulation and Circulation (ISC) experiment at the Grimsel Test Site dedicated to studying such processes under controlled conditions. The review shows that research on reservoir stimulation for deep geothermal energy exploitation has been largely based on laboratory observations, large-scale projects and numerical models. Observations of full-scale reservoir stimulations have yielded important results. However, the limited access to the reservoir and limitations in the control on the experimental conditions during deep reservoir stimulations is insufficient to resolve the details of the hydromechanical processes that would enhance process understanding in a way that aids future stimulation design. Small-scale laboratory experiments provide fundamental insights into various processes relevant for enhanced geothermal energy, but suffer from (1) difficulties and uncertainties in upscaling the results to the field scale and (2) relatively homogeneous material and stress conditions that lead to an oversimplistic fracture flow and/or hydraulic fracture propagation behavior that is not representative of a heterogeneous reservoir. Thus, there is a need for intermediate-scale hydraulic stimulation experiments with high experimental control that bridge the various scales and for which access to the target rock mass with a comprehensive monitoring system is possible. The ISC experiment is designed to address open research questions in a naturally fractured and faulted crystalline rock mass at the Grimsel Test Site (Switzerland). Two hydraulic injection phases were executed to enhance the permeability of the rock mass. During the injection phases the rock mass deformation across fractures and within intact rock, the pore pressure distribution and propagation, and the microseismic response were monitored at a high spatial and temporal resolution. 58 Rossi, E., M.A. Kant, C. Madonna, M.O. Saar, and Ph. Rudolf von Rohr The Effects of High Heating Rate and High Temperature on the Rock Strength: Feasibility Study of a Thermally Assisted Drilling Method, Rock Mechanics and Rock Engineering, 51/9, pp. 2957-2964 , 2018. AbstractIn this paper, the feasibility of a thermally assisted drilling method is investigated. The working principle of this method is based on the weakening effect of a flame-jet to enhance the drilling performance of conventional, mechanical drilling. To investigate its effectiveness, we study rock weakening after rapid, localized flame-jet heating of Rorschach sandstone and Central Aare granite. We perform experiments on rock strength after flame treatments in comparison to oven heating, for temperatures up to 650 $$^{\circ}$$C and heating rates from 0.17 to 20 $$^{\circ}$$C/s. The material hardening, commonly observed at moderate temperatures after oven treatments, can be suppressed by flame heating the material at high heating rates. Our study highlights the influence of the heating rate on the mechanism of thermal microcracking. High heating rate, flame treatments appear to mostly induce cracks at the grain boundaries, opposed to slow oven treatments, where also a considerable number of intragranular cracks are found. Herewith, we postulate that at low heating rates, thermal expansion stresses cause the observed thermal cracking. In contrast, at higher heating rates, thermal cracking is dominated by the stress concentrations caused by high thermal gradients. 57 Hobé, A., D. Vogler, M.P. Seybold, A. Ebigbo, R.R. Settgast, and M.O. Saar Estimating Fluid Flow Rates through Fracture Networks using Combinatorial Optimization, Advances in Water Resources, 122, pp. 85-97, 2018. AbstractTo enable fast uncertainty quantification of fluid flow in a discrete fracture network (DFN), we present two approaches to quickly compute fluid flow in DFNs using combinatorial optimization algorithms. Specifically, the presented Hanan Shortest Path Maxflow (HSPM) and Intersection Shortest Path Maxflow (ISPM) methods translate DFN geometries and properties to a graph on which a max flow algorithm computes a combinatorial flow, from which an overall fluid flow rate is estimated using a shortest path decomposition of this flow. The two approaches are assessed by comparing their predictions with results from explicit numerical simulations of simple test cases as well as stochastic DFN realizations covering a range of fracture densities. Both methods have a high accuracy and very low computational cost, which can facilitate much-needed in-depth analyses of the propagation of uncertainty in fracture and fracture-network properties to fluid flow rates. 56 Mikutis, G., C.A. Deuber, L. Schmid, A. Kittilä, N. Lobsiger, M. Puddu, D.O. Asgeirsson, R.N. Grass, M.O. Saar, and W.J. Stark Silica-encapsulated DNA-based tracers for aquifer characterization, Environmental Science & Technology, 52, pp. 12142-12152, 2018. AbstractEnvironmental tracing is a direct way to characterize aquifers, evaluate the solute transfer parameter in underground reservoirs, and track contamination. By performing multitracer tests, and translating the tracer breakthrough times into tomographic maps, key parameters such as a reservoir’s effective porosity and permeability field may be obtained. DNA, with its modular design, allows the generation of a virtually unlimited number of distinguishable tracers. To overcome the insufficient DNA stability due to microbial activity, heat, and chemical stress, we present a method to encapsulated DNA into silica with control over the particle size. The reliability of DNA quantification is improved by the sample preservation with NaN3 and particle redispersion strategies. In both sand column and unconsolidated aquifer experiments, DNA-based particle tracers exhibited slightly earlier and sharper breakthrough than the traditional solute tracer uranine. The reason behind this observation is the size exclusion effect, whereby larger tracer particles are excluded from small pores, and are therefore transported with higher average velocity, which is pore size-dependent. Identical surface properties, and thus flow behavior, makes the new material an attractive tracer to characterize sandy groundwater reservoirs or to track multiple sources of contaminants with high spatial resolution. 55 Leal, A.M.M., D.A. Kulik, W.R. Smith, and M.O. Saar An overview of computational methods for chemical equilibrium and kinetic calculations for geochemical and reactive transport modeling, Pure and Applied Chemistry, 89/5, pp. 597-643, 2017. AbstractWe present an overview of novel numerical methods for chemical equilibrium and kinetic calculations for complex non-ideal multiphase systems. The methods we present for equilibrium calculations are based either on Gibbs energy minimization (GEM) calculations or on solving the system of extended law of mass-action (xLMA) equations. In both methods, no a posteriori phase stability tests, and thus no tentative addition or removal of phases during or at the end of the calculations, are necessary. All potentially stable phases are considered from the beginning of the calculation, and stability indices are immediately available at the end of the computation to determine which phases are actually stable at equilibrium. Both GEM and xLMA equilibrium methods are tailored for computationally demanding applications that require many rapid local equilibrium calculations, such as reactive transport modeling. The numerical method for chemical kinetic calculations we present supports both closed and open systems, and it considers a partial equilibrium simplification for fast reactions. The method employs an implicit integration scheme that improves stability and speed when solving the often stiff differential equations in kinetic calculations. As such, it requires compositional derivatives of the reaction rates to assemble the Jacobian matrix of the resultant implicit algebraic equations that are solved at every time step. We present a detailed procedure to calculate these derivatives, and we show how the partial equilibrium assumption affects their computation. These numerical methods have been implemented in Reaktoro (reaktoro.org), an open-source software for modeling chemically reactive systems. We finish with a discussion on the comparison of these methods with others in the literature. 54 Luhmann, A.J., B.M. Tutolo, B.C. Bagley, D.F.R. Mildner, W.E. Seyfried Jr., and M.O. Saar Permeability, porosity, and mineral surface area changes in basalt cores induced by reactive transport of CO2-rich brine, Water Resources Research, 53, pp. 1-20, 2017. AbstractFour reactive flow-through laboratory experiments (two each at 0.1 mL/min and 0.01 mL/min flow rates) at 150°C and 150 bar (15 MPa) are conducted on intact basalt cores to assess changes in porosity, permeability, and surface area caused by CO2-rich fluid-rock interaction. Permeability decreases slightly during the lower flow rate experiments and increases during the higher flow rate experiments. At the higher flow rate, core permeability increases by more than one order of magnitude in one experiment and less than a factor of two in the other due to differences in preexisting flow path structure. X-ray computed tomography (XRCT) scans of pre- and post-experiment cores identify both mineral dissolution and secondary mineralization, with a net decrease in XRCT porosity of ∼0.7%–0.8% for the larger pores in all four cores. (Ultra) small-angle neutron scattering ((U)SANS) data sets indicate an increase in both (U)SANS porosity and specific surface area (SSA) over the ∼1 nm to 10 µm scale range in post-experiment basalt samples, with differences due to flow rate and reaction time. Net porosity increases from summing porosity changes from XRCT and (U)SANS analyses are consistent with core mass decreases. (U)SANS data suggest an overall preservation of the pore structure with no change in mineral surface roughness from reaction, and the pore structure is unique in comparison to previously published basalt analyses. Together, these data sets illustrate changes in physical parameters that arise due to fluid-basalt interaction in relatively low pH environments with elevated CO2 concentration, with significant implications for flow, transport, and reaction through geologic formations. 53 Luhmann, A.J., B.M. Tutolo, C. Tan, B.M. Moskowitz, M.O. Saar, and W.E. Seyfried, Jr. Whole rock basalt alteration from CO2-rich brine during flow-through experiments at 150°C and 150 bar, Chemical Geology, 453, pp. 92-110, 2017. AbstractFour flow-through experiments at 150 °C were conducted on intact cores of basalt to assess alteration and mass transfer during reaction with CO2-rich fluid. Two experiments used a flow rate of 0.1 ml/min, and two used a flow rate of 0.01 ml/min. Permeability increased for both experiments at the higher flow rate, but decreased for the lower flow rate experiments. The experimental fluid (initial pH of 3.3) became enriched in Si, Mg, and Fe upon passing through the cores, primarily from olivine and titanomagnetite dissolution and possibly pyroxene dissolution. Secondary minerals enriched in Al and Si were present on post-experimental cores, and an Fe2O3-rich phase was identified on the downstream ends of the cores from the experiments at the lower flow rate. While we could not specifically identify if siderite (FeCO3) was present in the post-experimental basalt cores, siderite was generally saturated or supersaturated in outlet fluid samples, suggesting a thermodynamic drive for Fe carbonation from basalt-H2O-CO2 reaction. Reaction path models that employ dissolution kinetics of olivine, labradorite, and enstatite also suggest siderite formation at low pH. Furthermore, fluid-rock interaction caused a relatively high mobility of the alkali metals; up to 29% and 99% of the K and Cs present in the core, respectively, were preferentially dissolved from the cores, likely due to fractional crystallization effects that made alkali metals highly accessible. Together, these datasets illustrate changes in chemical parameters that arise due to fluid-basalt interaction in relatively low pH environments with elevated CO2. 52 Myre, J.M., E. Frahm, D.J. Lilja, and M.O. Saar TNT-NN: A Fast Active Set Method for Solving Large Non-Negative Least Squares Problems, Procedia Computer Science, 108C, pp. 755-764, 2017. AbstractIn 1974 Lawson and Hanson produced a seminal active set strategy to solve least-squares prob- lems with non-negativity constraints that remains popular today. In this paper we present TNT-NN, a new active set method for solving non-negative least squares (NNLS) problems. TNT-NN uses a different strategy not only for the construction of the active set but also for the solution of the unconstrained least squares sub-problem. This results in dramatically improved performance over traditional active set NNLS solvers, including the Lawson and Hanson NNLS algorithm and the Fast NNLS (FNNLS) algorithm, allowing for computational investigations of new types of scientific and engineering problems. For the small systems tested (5000 × 5000 or smaller), it is shown that TNT-NN is up to 95× faster than FNNLS. Recent studies in rock magnetism have revealed a need for fast NNLS algorithms to address large problems (on the order of 105 × 105 or larger). We apply the TNT- NN algorithm to a representative rock magnetism inversion problem where it is 60× faster than FNNLS. We also show that TNT-NN is capable of solving large (45000 × 45000) problems more than 150× faster than FNNLS. These large test problems were previously considered to be unsolvable, due to the excessive execution time required by traditional methods. 51 Walsh, S.D.C., N. Garapati, A.M.M. Leal, and M.O. Saar Calculating thermophysical fluid properties during geothermal energy production with NESS and Reaktoro, Geothermics, 70, pp. 146-154, 2017. AbstractWe investigate how subsurface fluids of different compositions affect the electricity generation of geothermal power plants. First, we outline a numerical model capable of accounting for the thermophysical properties of geothermal fluids of arbitrary composition within simulations of geothermal power production. The behavior of brines with varying compositions from geothermal sites around the globe are then examined using the model. The effect of each brine on an idealized binary geothermal power plant is simulated, and their performances compared by calculating the amount of heat exchanged from the fluid to the plant’s secondary cycle. Our simulations combine (1) a newly developed Non-linear Equation System Solver (NESS), for simulating individual geothermal power plant components, (2) the advanced geochemical speciation solver, Reaktoro, used for calculation of thermodynamic fluid properties, and (3) compositional models for the calculation of fluid-dynamical properties (e.g., viscosity as a function of temperature and brine composition). The accuracy of the model is verified by comparing its predictions with experimental data from single-salt, binary-salt, and multiple-salt solutions. The geothermal power plant simulations show that the brines considered in this study can be divided into three main categories: (1) those of largely meteoric origin with low salinity for which the effect of salt concentration is negligible; (2) moderate-depth brines with high concentrations of Na+ and K+ ions, whose performance is well approximated by pure NaCl solutions of equivalent salinity; and (3) deeper, high-salinity brines that require a more detailed consideration of their composition for accurate simulation of plant operations. 50 Leal, A.M.M., D. Kulik, G. Kosakowski, and M.O. Saar Computational methods for reactive transport modeling: An extended law of mass-action, xLMA, method for multiphase equilibrium calculations, Advances in Water Resources, 96, pp. 405-422, 2016. AbstractWe present a numerical method for multiphase chemical equilibrium calculations based on a Gibbs energy minimization approach. The method can accurately and efficiently determine the stable phase assemblage at equilibrium independently of the type of phases and species that constitute the chemical system. We have successfully applied our chemical equilibrium algorithm in reactive transport simulations to demonstrate its effective use in computationally intensive applications. We used FEniCS to solve the governing partial differential equations of mass transport in porous media using finite element methods in unstructured meshes. Our equilibrium calculations were benchmarked with GEMS3K, the numerical kernel of the geochemical package GEMS. This allowed us to compare our results with a well-established Gibbs energy minimization algorithm, as well as their performance on every mesh node, at every time step of the transport simulation. The benchmark shows that our novel chemical equilibrium algorithm is accurate, robust, and efficient for reactive transport applications, and it is an improvement over the Gibbs energy minimization algorithm used in GEMS3K. The proposed chemical equilibrium method has been implemented in Reaktoro, a unified framework for modeling chemically reactive systems, which is now used as an alternative numerical kernel of GEMS. 49 Tutolo, B.M., D.F. Mildner, C.V. Gagnon, M.O. Saar, and W.E. Seyfried Nanoscale constraints on porosity generation and fluid flow during serpentinization, Geology, 44/2, pp. 103-106, 2016. AbstractField samples of olivine-rich rocks are nearly always serpentinized—commonly to completion—but, paradoxically, their intrinsic porosity and permeability are diminishingly low. Serpentinization reactions occur through a coupled process of fluid infiltration, volumetric expansion, and reaction-driven fracturing. Pores and reactive surface area generated during this process are the primary pathways for fluid infiltration into and reaction with serpentinizing rocks, but the size and distribution of these pores and surface area have not yet been described. Here, we utilize neutron scattering techniques to present the first measurements of the evolution of pore size and specific surface area distribution in partially serpentinized rocks. Samples were obtained from the ca. 2 Ma Atlantis Massif oceanic core complex located off-axis of the Mid-Atlantic Ridge and an olivine-rich outcrop of the ca. 1.1 Ga Duluth Complex of the North American Mid-Continent Rift. Our measurements and analyses demonstrate that serpentine and accessory phases form with their own, inherent porosity, which accommodates the bulk of diffusive fluid flow during serpentinization and thereby permits continued serpentinization after voluminous serpentine minerals fill reaction-generated porosity. 48 Leal, A.M.M., D.A. Kulik, and M.O. Saar Enabling Gibbs energy minimization algorithms to use equilibrium constants of reactions in multiphase equilibrium calculations, Chemical Geology, 437, pp. 170-181, 2016. AbstractThe geochemical literature provides numerous thermodynamic databases compiled from equilibrium constants of reactions. These databases are typically used in speciation calculations based on the law of mass action (LMA) approach. Unfortunately, such LMA databases cannot be directly used in equilibrium speciation methods based on the Gibbs energy minimization (GEM) approach because of their lack of standard chemical potentials of species. Therefore, we present in this work a simple conversion approach that calculates apparent standard chemical potentials of species from equilibrium constants of reactions. We assess the consistency and accuracy of the use of apparent standard chemical potentials in GEM algorithms by benchmarking equilibrium speciation calculations using GEM and LMA methods with the same LMA database. In all cases, we use PHREEQC to perform the LMA calculations, and we use its LMA databases to calculate the equilibrium constants of reactions. GEM calculations are performed using a Gibbs energy minimization method of Reaktoro — a unified open-source framework for numerical modeling of chemically reactive systems. By comparing the GEM and LMA results, we show that the use of apparent standard chemical potentials in GEM methods produces consistent and accurate equilibrium speciation results, thus validating our new, practical conversion technique that enables GEM algorithms to take advantage of many existing LMA databases, consequently extending and diversifying their range of applicability. 47 Buscheck, T.A., J.M. Bielicki, T.A. Edmunds, Y. Hao, Y. Sun, J.B. Randolph, and M.O. Saar Multifluid geo-energy systems: Using geologic CO2 storage for geothermal energy production and grid-scale energy storage in sedimentary basins, Geosphere, 12/3, pp. 1-19, 2016. AbstractWe present an approach that uses the huge fluid and thermal storage capac ity of the subsurface, together with geologic carbon dioxide (CO 2 ) storage, to harvest, store, and dispatch energy from subsurface (geothermal) and surface (solar, nuclear, fossil) thermal resources, as well as excess energy on electric grids. Captured CO 2 is injected into saline aquifers to store pres – sure, generate artesian flow of brine, and provide a supplemental working fluid for efficient heat extraction and power conversion. Concentric rings of injection and production wells create a hydraulic mound to store pressure, CO 2 , and thermal energy. This energy storage can take excess power from the grid and excess and/or waste thermal energy and dispatch that energy when it is demanded, and thus enable higher penetration of variable renewable en – ergy technologies (e.g., wind and solar). CO 2 stored in the subsurface func – tions as a cushion gas to provide enormous pressure storage capacity and displace large quantities of brine, some of which can be treated for a variety of beneficial uses. Geo thermal power and energy-storage applications may generate enough revenues to compensate for CO 2 capture costs. While our ap – proach can use nitrogen (N 2 ), in addition to CO 2 , as a supplemental fluid, and store thermal energy, this study focuses on using CO 2 for geothermal energy production and grid-scale energy storage. We conduct a techno-economic assess ment to determine the levelized cost of electricity using this approach to generate geothermal power. We present a reservoir pressure management strategy that diverts a small portion of the produced brine for beneficial con – sumptive use to reduce the pumping cost of fluid recirculation, while reducing the risk of seismicity, caprock fracture, and CO 2 leakage. 46 Tutolo, B.M., X.-Z. Kong, W.E. Seyfried Jr., and M.O. Saar High performance reactive transport simulations examining the effects of thermal, hydraulic, and chemical (THC) gradients on fluid injectivity at carbonate CCUS reservoir scales, International Journal of Greenhouse Gas Control, 39, pp. 285-301, 2015. AbstractCarbonate minerals and CO2 are both considerably more soluble at low temperatures than they are at elevated temperatures. This inverse solubility has led a number of researchers to hypothesize that injecting low-temperature (i.e., less than the background reservoir temperature) CO2 into deep, saline reservoirs for CO2 Capture, Utilization, and Storage (CCUS) will dissolve CO2 and carbonate minerals near the injection well and subsequently exsolve and re-precipitate these phases as the fluids flow into the geothermally warm portion of the reservoir. In this study, we utilize high performance computing to examine the coupled effects of cool CO2 injection and background hydraulic head gradients on reservoir-scale mineral volume changes. We employ the fully coupled reactive transport simulator PFLOTRAN with calculations distributed over up to 800 processors to test 21 scenarios designed to represent a range of reservoir depths, hydraulic head gradients, and CO2 injection rates and temperatures. In the default simulations, 50 °C CO2 is injected at a rate of 50 kg/s into a 200 bar, 100 °C calcite or dolomite reservoir. By comparing these simulations with others run at varying conditions, we show that the effect of cool CO2 injection on reservoir-scale mineral volume changes tends to be relatively minor. We conclude that the low heat capacity of CO2 effectively prevents low-temperature CO2 injection from decreasing the temperature across large portions of the simulated carbonate reservoirs. This small thermal perturbation, combined with the low relative permeability of brine within the supercritical CO2 plume, yields limited dissolution and precipitation effects directly attributable to cool CO2 injection. Finally, we calculate that relatively high water-to-rock ratios, which may occur over much longer CCUS reservoir lifetimes or in materials with sufficiently high brine relative permeability within the supercritical CO2 plume, would be required to substantially affect injectivity through thermally-induced mineral dissolution and precipitation. Importantly, this study shows the utility of reservoir scale-reactive transport simulators for testing hypotheses and placing laboratory-scale observations into a CCUS reservoir-scale context. 45 Tutolo, B.M., A.J. Luhmann, X.-Z. Kong, M.O. Saar, and W.E. Seyfried Jr. CO2 sequestration in feldspar-rich sandstone: Coupled evolution of fluid chemistry, mineral reaction rates, and hydrogeochemical properties, Geochimica Et Cosmochimica Acta, 160, pp. 132-154, 2015. AbstractTo investigate CO2 Capture, Utilization, and Storage (CCUS) in sandstones, we performed three 150 °C flow-through experiments on K-feldspar-rich cores from the Eau Claire formation. By characterizing fluid and solid samples from these experiments using a suite of analytical techniques, we explored the coupled evolution of fluid chemistry, mineral reaction rates, and hydrogeochemical properties during CO2 sequestration in feldspar-rich sandstone. Overall, our results confirm predictions that the heightened acidity resulting from supercritical CO2 injection into feldspar-rich sandstone will dissolve primary feldspars and precipitate secondary aluminum minerals. A core through which CO2-rich deionized water was recycled for 52 days decreased in bulk permeability, exhibited generally low porosity associated with high surface area in post-experiment core sub-samples, and produced an Al hydroxide secondary mineral, such as boehmite. However, two samples subjected to ?3 day single-pass experiments run with CO2-rich, 0.94 mol/kg NaCl brines decreased in bulk permeability, showed generally elevated porosity associated with elevated surface area in post-experiment core sub-samples, and produced a phase with kaolinite-like stoichiometry. CO2-induced metal mobilization during the experiments was relatively minor and likely related to Ca mineral dissolution. Based on the relatively rapid approach to equilibrium, the relatively slow near-equilibrium reaction rates, and the minor magnitudes of permeability changes in these experiments, we conclude that CCUS systems with projected lifetimes of several decades are geochemically feasible in the feldspar-rich sandstone end-member examined here. Additionally, the observation that K-feldspar dissolution rates calculated from our whole-rock experiments are in good agreement with literature parameterizations suggests that the latter can be utilized to model CCUS in K-feldspar-rich sandstone. Finally, by performing a number of reactive transport modeling experiments to explore processes occurring during the flow-through experiments, we have found that the overall progress of feldspar hydrolysis is negligibly affected by quartz dissolution, but significantly impacted by the rates of secondary mineral precipitation and their effect on feldspar saturation state. The observations produced here are critical to the development of models of CCUS operations, yet more work, particularly in the quantification of coupled dissolution and precipitation processes, will be required in order to produce models that can accurately predict the behavior of these systems. 44 Garapati, N., J.B. Randolph, and M.O. Saar Brine displacement by CO2, energy extraction rates, and lifespan of a CO2-limited CO2-Plume Geothermal (CPG) system with a horizontal production well, Geothermics, 55, pp. 182-194, 2015. AbstractSeveral studies suggest that CO2-based geothermal energy systems may be operated economically when added to ongoing geologic CO2 sequestration. Alternatively, we demonstrate here that CO2-Plume Geothermal (CPG) systems may be operated long-term with a finite amount of CO2. We analyze the performance of such CO2-limited CPG systems as a function of various geologic and operational parameters. We find that the amount of CO2 required increases with reservoir depth, permeability, and well spacing and decreases with larger geothermal gradients. Furthermore, the onset of reservoir heat depletion decreases for increasing geothermal gradients and for both particularly shallow and deep reservoirs. 43 Tutolo, B.M., A.T. Schaen, M.O. Saar, and W.E. Seyfried Jr. Implications of the redissociation phenomenon for mineral-buffered fluids and aqueous species transport at elevated temperatures and pressures, Applied Geochemistry, 55, pp. 119-127, 2015. AbstractAqueous species equilibrium constants and activity models form the foundation of the complex speciation codes used to model the geochemistry of geothermal energy production, extremophilic ecosystems, ore deposition, and a variety of other processes. Researchers have shown that a simple three species model (i.e., Na+, Cl?, and NaCl(aq)) can accurately describe conductivity measurements of concentrated NaCl and KCl solutions at elevated temperatures and pressures (Sharygin et al., 2002). In this model, activity coefficients of the charged species (e.g., Na+, K+, Cl?) become sufficiently low that the complexes must redisocciate with increasing salt concentration in order to meet equilibrium constant constraints. Redissociation decreases the proportion of the elements bound up as neutral complexes, and thereby increases the true ionic strength of the solution. In this contribution, we explore the consequences of the redissociation phenomenon in albite–paragonite–quartz (APQ) buffered systems. We focus on the implications of the redissociation phenomenon for mineral solubilities, particularly the observation that, at certain temperatures and pressures, calculated activities of charged ions in solution remain practically constant even as element concentrations increase from <1 molal to 4.5 molal. Finally, we note that redissociation has a similar effect on pH, and therefore aqueous speciation, in APQ-hosted systems. The calculations and discussion presented here are not limited to APQ-hosted systems, but additionally apply to many others in which the dominant cations and anions can form neutral complexes. 42 Adams, B.M., T.H. Kuehn, J.M. Bielicki, J.B. Randolph, and M.O. Saar A comparison of electric power output of CO2 Plume Geothermal (CPG) and brine geothermal systems for varying reservoir conditions, Applied Energy, 140, pp. 365-377, 2015. AbstractIn contrast to conventional hydrothermal systems or enhanced geothermal systems, CO2 Plume Geothermal (CPG) systems generate electricity by using CO2 that has been geothermally heated due to sequestration in a sedimentary basin. Four CPG and two brine-based geothermal systems are modeled to estimate their power production for sedimentary basin reservoir depths between 1 and 5km, geothermal temperature gradients from 20 to 50°Ckm-1, reservoir permeabilities from 1×10-15 to 1×10-12m2 and well casing inner diameters from 0.14m to 0.41m. Results show that CPG direct-type systems produce more electricity than brine-based geothermal systems at depths between 2 and 3km, and at permeabilities between 10-14 and 10-13m2, often by a factor of two. This better performance of CPG is due to the low kinematic viscosity of CO2, relative to brine at those depths, and the strong thermosiphon effect generated by CO2. When CO2 is used instead of R245fa as the secondary working fluid in an organic Rankine cycle (ORC), the power production of both the CPG and the brine-reservoir system increases substantially; for example, by 22% and 20% for subsurface brine and CO2 systems, respectively, with a 35°Ckm-1 thermal gradient, 0.27m production and 0.41m injection well diameters, and 5×10-14m2 reservoir permeability. 41 Luhmann, A.J., M. Covington, J. Myre, M. Perne, S.W. Jones, C.E. Alexander Jr., and M.O. Saar Thermal damping and retardation in karst conduits, Hydrology and Earth System Sciences, 19/1, pp. 137-157, 2015. AbstractWater temperature is a non-conservative tracer in the environment. Variations in recharge temperature are damped and retarded as water moves through an aquifer due to heat exchange between water and rock. However, within karst aquifers, seasonal and short-term fluctuations in recharge temperature are often transmitted over long distances before they are fully damped. Using analytical solutions and numerical simulations, we develop relationships that describe the effect of flow path properties, flow-through time, recharge characteristics, and water and rock physical properties on the damping and retardation of thermal peaks/troughs in karst conduits. Using these relationships, one can estimate the thermal retardation and damping that would occur under given conditions with a given conduit geometry. Ultimately, these relationships can be used with thermal damping and retardation field data to estimate parameters such as conduit diameter. We also examine sets of numerical simulations where we relax some of the assumptions used to develop these relationships, testing the effects of variable diameter, variable velocity, open channels, and recharge shape on thermal damping and retardation to provide some constraints on uncertainty. Finally, we discuss a multitracer experiment that provides some field confirmation of our relationships. High temporal resolution water temperature data are required to obtain sufficient constraints on the magnitude and timing of thermal peaks and troughs in order to take full advantage of water temperature as a tracer. 40 Tutolo, B.M., A.J. Luhmann, X.-Z. Kong, M.O. Saar, and W.E. Seyfried Jr. Experimental observation of permeability changes in dolomite at CO2 sequestration conditions, Environmental Science and Technology, 48/4, pp. 2445-2452, 2014. AbstractInjection of cool CO2 into geothermally warm carbonate reservoirs for storage or geothermal energy production may lower near-well temperature and lead to mass transfer along flow paths leading away from the well. To investigate this process, a dolomite core was subjected to a 650 h, high pressure, CO2 saturated, flow-through experiment. Permeability increased from 10–15.9 to 10–15.2 m2 over the initial 216 h at 21 °C, decreased to 10–16.2 m2 over 289 h at 50 °C, largely due to thermally driven CO2 exsolution, and reached a final value of 10–16.4 m2 after 145 h at 100 °C due to continued exsolution and the onset of dolomite precipitation. Theoretical calculations show that CO2 exsolution results in a maximum pore space CO2 saturation of 0.5, and steady state relative permeabilities of CO2 and water on the order of 0.0065 and 0.1, respectively. Post-experiment imagery reveals matrix dissolution at low temperatures, and subsequent filling-in of flow passages at elevated temperature. Geochemical calculations indicate that reservoir fluids subjected to a thermal gradient may exsolve and precipitate up to 200 cm3 CO2 and 1.5 cm3 dolomite per kg of water, respectively, resulting in substantial porosity and permeability redistribution. 39 Luhmann, A.J., X.-Z. Kong, B.M. Tutolo, N. Garapati, B.C. Bagley, M.O. Saar, and W.E. Seyfried Jr. Experimental dissolution of dolomite by CO2-charged brine at 100oC and 150 bar: Evolution of porosity, permeability, and reactive surface area, Chemical Geology, 380, pp. 145-160, 2014. AbstractHydrothermal flow experiments of single-pass injection of CO2-charged brine were conducted on nine dolomite cores to examine fluid–rock reactions in dolomite reservoirs under geologic carbon sequestration conditions. Post-experimental X-ray computed tomography (XRCT) analysis illustrates a range of dissolution patterns, and significant increases in core bulk permeability were measured as the dolomite dissolved. Outflow fluids were below dolomite saturation, and cation concentrations decreased with time due to reductions in reactive surface area with reaction progress. To determine changes in reactive surface area, we employ a power-law relationship between reactive surface area and porosity (Luquot and Gouze, 2009). The exponent in this relationship is interpreted to be a geometrical parameter that controls the degree of surface area change per change in core porosity. Combined with XRCT reconstructions of dissolution patterns, we demonstrate that this exponent is inversely related to both the flow path diameter and tortuosity of the dissolution channel. Even though XRCT reconstructions illustrate dissolution at selected regions within each core, relatively high Ba and Mn recoveries in fluid samples suggest that dissolution occurred along the core’s entire length and width. Analysis of porosity–permeability data indicates an increase in the rate of permeability enhancement per increase in porosity with reaction progress as dissolution channels lengthen along the core. Finally, we incorporate the surface area–porosity model of Luquot and Gouze (2009) with our experimentally fit parameters into TOUGHREACT to simulate experimental observations. 38 Adams, B.M., T.H. Kuehn, J.M. Bielicki, J.B. Randolph, and M.O. Saar On the importance of the thermosiphon effect in CPG (CO2-Plume geothermal) power systems, Energy, 69, pp. 409-418, 2014. AbstractCPG (CO2 Plume Geothermal) energy systems use CO2 to extract thermal energy from naturally permeable geologic formations at depth. CO2 has advantages over brine: high mobility, low solubility of amorphous silica, and higher density sensitivity to temperature. The density of CO2 changes substantially between geothermal reservoir and surface plant, resulting in a buoyancy-driven convective current – a thermosiphon – that reduces or eliminates pumping requirements. We estimated and compared the strength of this thermosiphon for CO2 and for 20 weight percent NaCl brine for reservoir depths up to 5 km and geothermal gradients of 20, 35, and 50 °C/km. We found that through the reservoir, CO2 has a pressure drop approximately 3–12 times less than brine at the same mass flowrate, making the CO2 thermosiphon sufficient to produce power using reservoirs as shallow as 0.5 km. At 2.5 km depth with a 35 °C/km gradient – the approximate western U.S. continental mean – the CO2 thermosiphon converted approximately 10% of the energy extracted from the reservoir to fluid circulation, compared to less than 1% with brine, where additional mechanical pumping is necessary. We found CO2 is a particularly advantageous working fluid at depths between 0.5 km and 3 km. 37 Tutolo, B.M., X.-Z. Kong, W.E. Seyfried Jr., and M.O. Saar Internal consistency in aqueous geochemical data revisited: Applications to the aluminum system, Geochimica et Cosmochimica Acta, 133, pp. 216-234, 2014. AbstractInternal consistency of thermodynamic data has long been considered vital for confident calculations of aqueous geochemical processes. However, an internally consistent mineral thermodynamic data set is not necessarily consistent with calculations of aqueous species thermodynamic properties due, potentially, to improper or inconsistent constraints used in the derivation process. In this study, we attempt to accommodate the need for a mineral thermodynamic data set that is internally consistent with respect to aqueous species thermodynamic properties by adapting the least squares optimization methods of Powell and Holland (1985). This adapted method allows for both the derivation of mineral thermodynamic properties from fluid chemistry measurements of solutions in equilibrium with mineral assemblages, as well as estimates of the uncertainty on the derived results. Using a large number of phase equilibria, solubility, and calorimetric measurements, we have developed a thermodynamic data set of 12 key aluminum-bearing mineral phases. These data are derived to be consistent with Na+ and K+ speciation data presented by Shock and Helgeson (1988), H4SiO4(aq) data presented by Stefánsson (2001), and the Al speciation data set presented by Tagirov and Schott (2001). Many of the constraining phase equilibrium measurements are exactly the same as those used to develop other thermodynamic data, yet our derived values tend to be quite different than some of the others’ due to our choices of reference data. The differing values of mineral thermodynamic properties have implications for calculations of Al mineral solubilities; specifically, kaolinite solubilities calculated with the developed data set are as much as 6.75 times lower and 73% greater than those calculated with Helgeson et al. (1978) and Holland and Powell (2011) data, respectively. Where possible, calculations and experimental data are compared at low T, and the disagreement between the two sources reiterates the common assertion that low-T measurements of phase equilibria and mineral solubilities in the aluminum system rarely represent equilibrium between water and well-crystallized, aluminum-bearing minerals. As an ancillary benefit of the derived data, we show that it may be combined with high precision measurements of aqueous complex association constants to derive neutral species activity coefficients in supercritical fluids. Although this contribution is specific to the aluminum system, the methods and concepts developed here can help to improve the calculation of water–rock interactions in a broad range of earth systems. 36 Garapati, N., J.B. Randolph, J.L. Valencia Jr., and M.O. Saar CO2-Plume Geothermal (CPG) Heat Extraction in Multi-layered Geologic Reservoirs, Energy Procedia, 63, pp. 7631-7643, 2014. AbstractCO2-Plume Geothermal (CPG) technology involves injecting CO2 into natural, highly permeable geologic units to extract energy. The subsurface CO2 absorbs heat from the reservoir, buoyantly rises to the surface, and drives a power generation system. The CO2 is then cooled and reinjected underground. Here, we analyze the effects of multi-layered geologic reservoirs on CPG system performance by examining the CO2 mass fraction in the produced fluid, pore-fluid pressure buildup during operation, and heat energy extraction rates. The produced CO2 mass fraction depends on the stratigraphic positions of highly permeable layers which also affect the pore-fluid pressure drop across the reservoir. 35 Buscheck, T.A., J.M. Bielicki, M. Chen, Y. Sun, Y. Hao, T.A. Edmunds, M.O. Saar, and J.B. Randolph Integrating CO2 Storage with Geothermal Resources for Dispatchable Renewable Electricity, Energy procedia, 63, pp. 7619-7630, 2014. AbstractWe present an approach that uses the huge fluid and thermal storage capacity of the subsurface, together with geologic CO2 storage, to harvest, store, and dispatch energy from subsurface (geothermal) and surface (solar, nuclear, fossil) thermal resources, as well as energy from electrical grids. Captured CO2 is injected into saline aquifers to store pressure, generate artesian flow of brine, and provide an additional working fluid for efficient heat extraction and power conversion. Concentric rings of injection and production wells are used to create a hydraulic divide to store pressure, CO2, and thermal energy. Such storage can take excess power from the grid and excess/waste thermal energy, and dispatch that energy when it is demanded, enabling increased penetration of variable renewables. Stored CO2 functions as a cushion gas to provide enormous pressure-storage capacity and displaces large quantities of brine, which can be desalinated and/or treated for a variety of beneficial uses. Geothermal power and energy-storage applications may generate enough revenues to justify CO2 capture costs. 34 Randolph, J.B., M.O. Saar, and J.M. Bielicki Geothermal energy production at geologic CO2 sequestration sites: Impact of thermal drawdown on reservoir pressure, Energy Procedia, 37, pp. 6625-6635, 2013. AbstractRecent geotechnical research shows that geothermal heat can be efficiently mined by circulating carbon dioxide through naturally permeable rock formations — a method called CO2 Plume Geothermal — the same geologic reservoirs that are suitable for deep saline aquifer CO2 sequestration or enhanced oil recovery. This paper describes the effect of thermal drawdown on reservoir pressure buildup during sequestration operations, revealing that geothermal heat mining can decrease overpressurization by 10% or more. 33 Gottardi, R., P.-H. Kao, M.O. Saar, and Ch. Teyssier Effects of permeability fields on fluid, heat, and oxygen isotope transport in extensional detachment systems, Geochemistry, Geophysics, Geosystems, 14/5, pp. 1493-1522, 2013. Abstract[1] Field studies of Cordilleran metamorphic core complexes indicate that meteoric fluids permeated the upper crust down to the detachment shear zone and interacted with highly deformed and recrystallized (mylonitic) rocks. The presence of fluids in the brittle/ductile transition zone is recorded in the oxygen and hydrogen stable isotope compositions of the mylonites and may play an important role in the thermomechanical evolution of the detachment shear zone. Geochemical data show that fluid flow in the brittle upper crust is primarily controlled by the large-scale fault-zone architecture. We conduct continuum-scale (i.e., large-scale, partial bounce-back) lattice-Boltzman fluid, heat, and oxygen isotope transport simulations of an idealized cross section of a metamorphic core complex. The simulations investigate the effects of crust and fault permeability fields as well as buoyancy-driven flow on two-way coupled fluid and heat transfer and resultant exchange of oxygen isotopes between meteoric fluid and rock. Results show that fluid migration to middle to lower crustal levels is fault controlled and depends primarily on the permeability contrast between the fault zone and the crustal rocks. High fault/crust permeability ratios lead to channelized flow in the fault and shear zones, while lower ratios allow leakage of the fluids from the fault into the crust. Buoyancy affects mainly flow patterns (more upward directed) and, to a lesser extent, temperature distributions (disturbance of the geothermal field by ~25°C). Channelized fluid flow in the shear zone leads to strong vertical and horizontal thermal gradients, comparable to field observations. The oxygen isotope results show δ18O depletion concentrated along the fault and shear zones, similar to field data. 32 Walsh, S.D.C., and M.O. Saar Developing extensible lattice-Boltzmann simulators for general-purpose graphics-processing units, Communications in Computational Physics, 13/3, pp. 867-879, 2013. AbstractLattice-Boltzmann methods are versatile numerical modeling techniques capable of reproducing a wide variety of fluid-mechanical behavior. These methods are well suited to parallel implementation, particularly on the single-instruction multiple data (SIMD) parallel processing environments found in computer graphics processing units (GPUs). Although recent programming tools dramatically improve the ease with which GPUbased applications can be written, the programming environment still lacks the flexibility available to more traditional CPU programs. In particular, it may be difficult to develop modular and extensible programs that require variable on-device functionality with current GPU architectures. This paper describes a process of automatic code generation that overcomes these difficulties for lattice-Boltzmann simulations. It details the development of GPU-based modules for an extensible lattice-Boltzmann simulation package – LBHydra. The performance of the automatically generated code is compared to equivalent purposewritten codes for both single-phase,multiphase, andmulticomponent flows. The flexibility of the new method is demonstrated by simulating a rising, dissolving droplet moving through a porous medium with user generated lattice-Boltzmann models and subroutines. 31 Kong, X.-Z., and M.O. Saar DBCreate: A SUPCRT92-based program for producing EQ3/6, TOUGHREACT, and GWB thermodynamic databases at user-defined T and P, Computers and Geosciences, 51, pp. 415-417, 2013. AbstractSUPCRT92 is a widely used software package for calculating the standard thermodynamic properties of minerals, gases, aqueous species, and reactions. However, it is labor-intensive and error-prone to use it directly to produce databases for geochemical modeling programs such as EQ3/6, the Geochemist’s Workbench, and TOUGHREACT. DBCreate is a SUPCRT92-based software program written in FORTRAN90/95 and was developed in order to produce the required databases for these programs in a rapid and convenient way. This paper describes the overall structure of the program and provides detailed usage instructions. 30 Kong, X.-Z., and M.O. Saar Numerical study of the effects of permeability heterogeneity on density-driven convective mixing during CO2 dissolution storage, Int. J. Greenhouse Gas Control, 19, pp. 160-173, 2013. AbstractPermanence and security of carbon dioxide (CO2) in geologic formations requires dissolution of CO2 into brine, which slightly increases the brine density. Previous studies have shown that this small increase in brine density induces convective currents, which greatly enhances the mixing efficiency and thus CO2 storage capacity and rate in the brine. Density-driven convection, in turn, is known to be largely dominated by permeability heterogeneity. This study explores the relationship between the process of density-driven convection and the permeability heterogeneity of an aquifer during CO2 dissolution storage, using high-resolution numerical simulations. While the porosity is kept constant, the heterogeneity of the aquifer is introduced through a spatially varying permeability field, characterized by the Dykstra-Parsons coefficient and the correlation length. Depending on the concentration profile of dissolved CO2, we classify the convective finger patterns as dispersive, preferential, and unbiased fingering. Our results indicate that the transition between unbiased and both preferential and dispersive fingering is mainly governed by the Dykstra-Parsons coefficient, whereas the transition between preferential and dispersive fingering is controlled by the permeability correlation length. Furthermore, we find that the CO2 dissolution flux at the top boundary will reach a time-independent steady state. Although this flux strongly correlates with permeability distribution, it generally increases with the permeability heterogeneity when the correlation length is less than the system size. 29 Luhmann, A.J., X.-Z. Kong, B.M. Tutolo, K. Ding, M.O. Saar, and W.E. Seyfried Jr. Permeability reduction produced by grain reorganization and accumulation of exsolved CO2 during geologic carbon sequestration: A new CO2 trapping mechanism, Environmental Science and Technlogy, 47/1, pp. 242-251, 2013. AbstractCarbon sequestration experiments were conducted on uncemented sediment and lithified rock from the Eau Claire Formation, which consisted primarily of K-feldspar and quartz. Cores were heated to accentuate reactivity between fluid and mineral grains and to force CO2 exsolution. Measured permeability of one sediment core ultimately reduced by 4 orders of magnitude as it was incrementally heated from 21 to 150 °C. Water-rock interaction produced some alteration, yielding sub-?m clay precipitation on K-feldspar grains in the core’s upstream end. Experimental results also revealed abundant newly formed pore space in regions of the core, and in some cases pores that were several times larger than the average grain size of the sediment. These large pores likely formed from elevated localized pressure caused by rapid CO2 exsolution within the core and/or an accumulating CO2 phase capable of pushing out surrounding sediment. CO2 filled the pores and blocked flow pathways. Comparison with a similar experiment using a solid arkose core indicates that CO2 accumulation and grain reorganization mainly contributed to permeability reduction during the heated sediment core experiment. This suggests that CO2 injection into sediments may store more CO2 and cause additional permeability reduction than is possible in lithified rock due to grain reorganization. 28 Covington, M., A.F. Banwell, J. Gulley, and M.O. Saar Quantifying the effects of glacier conduit geometry and recharge on proglacial hydrograph form, Journal of Hydrology, 414-415, pp. 59-71, 2012. AbstractThe configuration of glacier hydrological systems is often inferred from proxy data, such as hydrographs, that are collected in proglacial streams. Seasonal changes in the peakedness of hydrographs are thought to reflect changes in the configuration of the subglacial drainage system. However, the amount of information that proglacial hydrographs contain about drainage system configurations depends critically on the degree to which the drainage systems modify recharge hydrographs. If the drainage system does not modify recharge hydrographs, then proglacial hydrographs primarily reflect the recharge conditions produced by supraglacial inputs. Here, we develop a theoretical framework to determine the circumstances under which glacier drainage systems can modify recharge hydrographs and the circumstances under which recharge pulses pass through glaciers unchanged. We address the capability of single conduits, simple arborescent conduit networks, and linked cavity systems to modify diurnal recharge pulses. Simulations of discharge through large sets of such systems demonstrate that, unless large reservoirs or significant constrictions are present, the discharge hydrographs of simple glacial conduit systems are nearly identical to their recharge hydrographs. Conduit systems tend not to modify hydrographs because the changes in storage within englacial and subglacial conduit networks on short time scales are typically small compared to their ability to transmit water. This finding suggests that proglacial hydrographs reflect a variety of factors, including surface melt rate, surface water transfer, and subglacial water transfer. In many cases the influence of suglacial processes may be relatively minor. As a result, the evolution of proglacial hydrographs cannot be used unambiguously to infer changes in the structure or efficiency of englacial or subglacial hydrological systems, without accurate knowledge of the nature of the recharge hydrograph driving the flow. 27 Covington, M.D., A.J. Luhmann, C. Wicks, and M.O. Saar Process length scales and longitudinal damping in karst conduits, Journal of Geophysical Research – Earth Surface, 117, F01025, 2012. Abstract[1] Simple mathematical models often allow an intuitive grasp of the function of physical systems. We develop a mathematical framework to investigate reactive or dissipative transport processes within karst conduits. Specifically, we note that for processes that occur within a characteristic timescale, advection along the conduit produces a characteristic process length scale. We calculate characteristic length scales for the propagation of thermal and electrical conductivity signals along karst conduits. These process lengths provide a quantitative connection between karst conduit geometry and the signals observed at a karst spring. We show that water input from the porous/fractured matrix is also characterized by a length scale and derive an approximation that accounts for the influence of matrix flow on the transmission of signals through the aquifer. The single conduit model is then extended to account for conduits with changing geometries and conduit flow networks, demonstrating how these concepts can be applied in more realistic conduit geometries. We introduce a recharge density function, ϕR, which determines the capability of an aquifer to damp a given signal, and cast previous explanations of spring variability within this framework. Process lengths are a general feature of karst conduits and surface streams, and we conclude with a discussion of other potential applications of this conceptual and mathematical framework. 26 Alexander, S.C., and M.O. Saar Improved characterization of small u for Jacob pumping test analysis methods, Ground Water, 50/2, pp. 256-265, 2012. AbstractNumerous refinements have been proposed to traditional pumping test analyses, yet many hydrogeologists continue to use the Jacob method due to its simplicity. Recent research favors hydraulic tomography and inverse numerical modeling of pumping test data. However, at sites with few wells, or relatively short screens, the data requirements of these methods may be impractical within physical and fiscal constraints. Alternatively, an improved understanding of the assumptions and limitations of Theis and, due to their widespread usage, Jacob analyses, leads to improved interpretations in data-poor environments. A fundamental requirement of Jacob is a “small” value of u = f(r2/t), with radial distance, r, and pumping time, t. However, selection of a too stringent (i.e., too low) maximum permissible u-value, umax, results in rejection of usable data from wells beyond a maximum radius, rmax. Conversely, data from small radii, less than rmin, where turbulent- and vertical-flow components arise, can result in acceptance of inappropriate data. Usage of drawdown data from wells too close to the pumping well, and exclusion of data from wells deemed too far, can cause unrealistic aquifer transmissivity, permeability, and storativity determinations. Here, data from an extensive well field in a glacial-outwash aquifer in north-central Minnesota, USA, are used to develop a new estimate for umax. Traditionally quoted values for umax range from 0.01 to 0.05. Our proposed value for Jacob distance-drawdown analyses is significantly higher with umax up to 0.2, resulting in larger allowable rmax-values and a higher likelihood of inclusion of additional wells in such pumping test analyses. 25 Covington, M.D., A.J. Luhmann, F. Gabrovsek, M.O. Saar, I. Willis, and C.M. Wicks Mechanisms of heat exchange between water and rock in karst conduits, Water Resources Research, 47, W10514/10, 2011. Abstract[1] Previous studies, motivated by understanding water quality, have explored the mechanisms for heat transport and heat exchange in surface streams. In karst aquifers, temperature signals play an additional important role since they carry information about internal aquifer structures. Models for heat transport in karst conduits have previously been developed; however, these models make different, sometimes contradictory, assumptions. Additionally, previous models of heat transport in karst conduits have not been validated using field data from conduits with known geometries. Here we use analytical solutions of heat transfer to examine the relative importance of heat exchange mechanisms and the validity of the assumptions made by previous models. The relative importance of convection, conduction, and radiation is a function of time. Using a characteristic timescale, we show that models neglecting rock conduction produce spurious results in realistic cases. In contrast to the behavior of surface streams, where conduction is often negligible, conduction through the rock surrounding a conduit determines heat flux at timescales of weeks and longer. In open channel conduits, radiative heat flux can be significant. In contrast, convective heat exchange through the conduit air is often negligible. Using the rules derived from our analytical analysis, we develop a numerical model for heat transport in a karst conduit. Our model compares favorably to thermal responses observed in two different karst settings: a cave stream fed via autogenic recharge during a snowmelt event, and an allogenically recharged cave stream that experiences continuous temperature fluctuations on many timescales. 24 Randolph, J.B., and M.O. Saar Impact of reservoir permeability on the choice of subsurface geothermal heat exchange fluid: CO2 versus water and native brine, Geothermal Resources Council (GRC) Transactions, 35, pp. 521-526, 2011. AbstractAbstract Geothermal systems utilizing carbon dioxide (CO 2) as the subsurface heat exchange fluid in naturally porous, permeable geologic formations have been shown to provide improved geothermal heat energy extraction, even at low resource temperatures. Such systems, termed CO 2 Plume Geothermal (CPG) systems, have the potential to permit expansion of geothermal energy utilization while supporting rapid implementation through the use of existing technologies. Here, we explore CPG heat extraction as a function of reservoir permeability and in comparison to water and brine geothermal heat extraction. We show that for reservoir permeabilities below 2 x 10 -14 m 2 , CO 2 -based geothermal provides better electric power production efficiency than both water-and brine-based systems. Impact of reservoir permeability on the choice of subsurface geothermal heat exchange fluid: CO2 versus water and native brine (PDF Download Available). Available from: https://www.researchgate.net/publication/266220008_Impact_of_reservoir_permeability_on_the_choice_of_subsurface_geothermal_heat_exchange_fluid_CO2_versus_water_and_native_brine [accessed Jun 7, 2017]. 23 Randolph, J.B., and M.O. Saar Combining geothermal energy capture with geologic carbon dioxide sequestration, Geophysical Research Letters, 38, L10401, 2011. Abstract[1] Geothermal energy offers clean, renewable, reliable electric power with no need for grid-scale energy storage, yet its use has been constrained to the few locations worldwide with naturally high geothermal heat resources and groundwater availability. We present a novel approach with the potential to permit expansion of geothermal energy utilization: heat extraction from naturally porous, permeable formations with CO2 as the injected subsurface working fluid. Fluid-mechanical simulations reveal that the significantly higher mobility of CO2, compared to water, at the temperature/pressure conditions of interest makes CO2 an attractive heat exchange fluid. We show numerically that, compared to conventional water-based and engineered geothermal systems, the proposed approach provides up to factors of 2.9 and 5.0, respectively, higher geothermal heat energy extraction rates. Consequently, more regions worldwide could be economically used for geothermal electricity production. Furthermore, as the injected CO2 is eventually geologically sequestered, such power plants would have negative carbon footprints. 22 Davis, M.A., S.D.C. Walsh, and M.O. Saar Statistically reconstructing continuous isotropic and anisotropic two-phase media while preserving macroscopic material properties, Physical Review E, 83, 026706, 2011. AbstractWe propose a method to generate statistically similar reconstructions of two-phase media. As with previous work, we initially characterize the microstructure of the material using two-point correlation functions (a subset of spatial correlation functions) and then generate numerical reconstructions using a simulated annealing method that preserves the geometric relationships of the material’s phase of interest. However, in contrast to earlier contributions that consider reconstructions composed of discrete arrays of pixels or voxels alone, we generate reconstructions based on assemblies of continuous, three-dimensional, interpenetrating objects. The result is a continuum description of the material microstructure (as opposed to a discretized or pixelated description), capable of efficiently representing large disparities in scale. Different reconstruction methods are considered based on distinct combinations of two-point correlation functions of varying degrees of complexity. The quality of the reconstruction methods are evaluated by comparing the total pore fraction, specific surface area of the percolating cluster, pore fraction of the percolating cluster, tortuosity, and permeability of the reconstructions to those of a set of reference assemblies. Elsewhere it has been proposed that two-phase media could be statistically reproduced with only two spatial correlation functions: the two-point probability function (the probability that two points lie within the same phase) and the lineal path function (the probability that a line between two points lies entirely within the same phase). We find that methods employing the two-point probability function and lineal path function are improved if the percolating cluster volume is also considered in the reconstruction. However, to reproduce more complicated geometric assemblies, we find it necessary to employ the two-point probability, two-point cluster, and lineal path function in addition to the percolating cluster volume to produce a generally accurate statistical reconstruction. 21 Randolph, J.B., and M.O. Saar Coupling carbon dioxide sequestration with geothermal energy capture in naturally permeable, porous geologic formations: Implications for CO2 sequestration, Energy Procedia, 4, pp. 2206-2213, 2011. AbstractCarbon dioxide (CO2) sequestration in deep saline aquifers and exhausted oil and natural gas fields has been widely considered as a means for reducing CO2 emissions to the atmosphere as a counter-measure to global warming. However, rather than treating CO2 merely as a waste fluid in need of permanent disposal, we propose that it could also be used as a working fluid in geothermal energy capture, as its thermodynamic and fluid mechanical properties suggest it transfers geothermal heat more efficiently than water. Energy production and sales in conjunction with sequestration would improve the economic viability of CO2 sequestration, a critical challenge for large-scale implementation of the technology. In addition, using CO2 as the working fluid in geothermal power systems may permit utilization of lower temperature geologic formations than those that are currently deemed economically viable, leading to more widespread utilization of geothermal energy. Here, we present the results of early-stage calculations demonstrating the geothermal energy capture potential of CO2-based geothermal systems and implications of such energy capture for the economic viability of geologic CO2 sequestration. 20 Saar, M.O. Review: Geothermal heat as a tracer of large-scale groundwater flow and as a means to determine permeability fields, special theme issue on Environmental Tracers and Groundwater Flow, editor-invited peer-reviewed contribution, Hydrogeology Journal, 19, pp. 31-52, 2011. AbstractA review of coupled groundwater and heat transfer theory is followed by an introduction to geothermal measurement techniques. Thereafter, temperature-depth profiles (geotherms) and heat discharge at springs to infer hydraulic parameters and processes are discussed. Several studies included in this review state that minimum permeabilities of approximately 5 × 10−17 < kmin <10−15 m2 are required to observe advective heat transfer and resultant geotherm perturbations. Permeabilities below kmin tend to cause heat-conduction-dominated systems, precluding inversion of temperature fields for groundwater flow patterns and constraint of permeabilities other than being 19 Myre, J., S.D.C. Walsh, D.J. Lilja, and M.O. Saar Performance analysis of single-phase multiphase, and multicomponent lattice-Boltzmann fluid flow simulations on GPU clusters, Concurrency and Computation: Practice and Experience, 23, pp. 332-350, 2010. AbstractAbstract The lattice-Boltzmann method is well suited for implementation in single-instruction multiple-data (SIMD) environments provided by general purpose graphics processing units (GPGPUs). This paper discusses the integration of these GPGPU programs with OpenMP to create lattice-Boltzmann applications for multi-GPU clusters. In addition to the standard single-phase single-component lattice-Boltzmann method, the performances of more complex multiphase, multicomponent models are also examined. The contributions of various GPU lattice-Boltzmann parameters to the performance are examined and quantified with a statistical model of the performance using Analysis of Variance (ANOVA). By examining single- and multi-GPU lattice-Boltzmann simulations with ANOVA, we show that all the lattice-Boltzmann simulations primarily depend on effects corresponding to simulation geometry and decomposition, and not on the architectural aspects of GPU. Additionally, using ANOVA we confirm that the metrics of Efficiency and Utilization are not suitable for memory-bandwidth-dependent codes. Copyright © 2010 John Wiley & Sons, Ltd. 18 Dasgupta, S., M.O. Saar, R.L. Edwards, C.-C. Shen, H. Cheng, and C.E. Alexander Jr. Three thousand years of extreme rainfall events recorded in stalagmites from Spring Valley Caverns, Minnesota, Earth and Planetary Science Letters, 300, pp. 46-54, 2010. AbstractAnnual layer analysis in two stalagmites collected from Spring Valley Caverns, southeastern Minnesota, reveals hydrological response of the cave to extreme rainfall events in the Midwest, USA. Cave-flooding events are identified within the two samples by the presence of detrital layers composed of clay sized particles. Comparison with instrumental records of precipitation demonstrates a strong correlation between these cave-flood events and extreme rainfall observed in the Upper Mississippi Valley. A simple model is developed to assess the nature of rainfall capable of flooding the cave. The model is first calibrated to the last 50-yr (1950–1998 A.D.) instrumental record of daily precipitation data for the town of Spring Valley and verified with the first 50 yr of record from 1900 to 1949 A.D. Frequency analysis shows that these extreme flood events have increased from the last half of the nineteenth century. Comparison with other paleohydrological records shows increased occurrence of extreme rain events during periods of higher moisture availability. Our study implies that increased moisture availability in the Midwestern region, due to rise in temperature from global warming could lead to an increase in the occurrence of extreme rainfall events. 17 Walsh, S.D.C., and M.O. Saar Interpolated lattice-Boltzmann boundary conditions for surface reaction kinetics, Physical Review E, 82, 066703, 2010. AbstractThis paper describes a method for implementing surface reaction kinetics in lattice Boltzmann simulations. The interpolated boundary conditions are capable of simulating surface reactions and dissolution at both stationary and moving solid-fluid and fluid-fluid interfaces. Results obtained with the boundary conditions are compared to analytical solutions for first-order and constant-flux kinetic surface reactions in a one-dimensional half space, as well as to the analytical solution for evaporation from the surface of a cylinder. Excellent agreement between analytical and simulated results is obtained for a wide range of diffusivities, lattice velocities, and surface reaction rates. The boundary model’s ability to represent dissolution in binary fluid mixtures is demonstrated by modeling diffusion from a rising bubble and dissolution of a droplet near a flat plate. 16 Randolph, J.B., and M.O. Saar Coupling geothermal energy capture with carbon dioxide sequestration in naturally permeable, porous geologic formations: A comparison with enhanced geothermal systems, Geothermal Resources Council (GRC) Transactions, 34, pp. 433-438, 2010. AbstractGeothermal energy offers clean, consistent, reliable electric power with no need for grid-scale energy storage, unlike wind and solar renewable power alternatives. However, geothermal energy is often underrepresented in renewable energy discussions and has considerable room for growth. New technology and methods will be critical for future investment, and rapid implementation of new techniques will be critical in ensuring geothermal energy plays a significant role in the future energy landscape world – wide. Here, we discuss a novel approach with the potential to permit expansion of geothermal energy utilization while supporting rapid implementation through the use of exist – ing technologies: geothermal heat use in naturally porous, permeable geologic formations with carbon dioxide as the working heat exchange fluid. 15 Walsh, S.D.C., and M.O. Saar Macroscale lattice-Boltzmann methods for low-Peclet-number solute and heat transport in heterogeneous porous media, Water Resources Research, 46, W07517, 2010. Abstract[1] This paper introduces new methods for simulating subsurface solute and heat transport in heterogeneous media using large-scale lattice-Boltzmann models capable of representing both macroscopically averaged porous media and open channel flows. Previous examples of macroscopically averaged lattice-Boltzmann models for solute and heat transport are only applicable to homogeneous media. Here, we extend these models to properly account for heterogeneous pore-space distributions. For simplicity, in the majority of this paper we assume low Peclet number flows with an isotropic dispersion tensor. Nevertheless, this approach may also be extended to include anisotropic-dispersion by using multiple relaxation time lattice-Boltzmann methods. We describe two methods for introducing heterogeneity into macroscopically averaged lattice-Boltzmann models. The first model delivers the desired behavior by introducing an additional time-derivative term to the collision rule; the second model by separately weighting symmetric and anti-symmetric components of the fluid packet densities. Chapman-Enskog expansions are conducted on the governing equations of the two models, demonstrating that the correct constitutive behavior is obtained in both cases. In addition, methods for improving model stability at low porosities are also discussed: (1) an implicit formulation of the model; and (2) a local transformation that normalizes the lattice-Boltzmann model by the local porosity. The model performances are evaluated through comparisons of simulated results with analytical solutions for one- and two-dimensional flows, and by comparing model predictions to finite element simulations of advection isotropic-dispersion in heterogeneous porous media. We conclude by presenting an example application, demonstrating the ability of the new models to couple with simulations of reactive flow and changing flow geometry: a simulation of groundwater flow through a carbonate system. 14 Covington, M., C.M. Wicks, and M.O. Saar A dimensionless number describing the effects of recharge and geometry on discharge from simple karst aquifers, Water Resources Research, 45, W11410, 2009. Abstract[1] The responses of karstic aquifers to storms are often used to obtain information about aquifer geometry. In general, spring hydrographs are a function of both system geometry and recharge. However, the majority of prior work on storm pulses through karst has not studied the effect of recharge on spring hydrographs. To examine the relative importance of geometry and recharge, we break karstic aquifers into elements according to the manner of their response to transient flow and demonstrate that each element has a characteristic response timescale. These fundamental elements are full pipes, open channels, reservoir/constrictions, and the porous matrix. Taking the ratio of the element timescale with the recharge timescale produces a dimensionless number, γ, that is used to characterize aquifer response to a storm event. Using sets of simulations run with randomly selected element parameters, we demonstrate that each element type has a critical value of γ below which the shape of the spring hydrograph is dominated by the shape of the recharge hydrograph and above which the spring hydrograph is significantly modified by the system geometry. This allows separation of particular element/storm pairs into recharge-dominated and geometry-dominated regimes. While most real karstic aquifers are complex combinations of these elements, we draw examples from several karst systems that can be represented by single elements. These examples demonstrate that for real karstic aquifers full pipe and open channel elements are generally in the recharge-dominated regime, whereas reservoir/constriction elements can fall in either the recharge- or geometry-dominated regimes. 13 Walsh, S.D.C., M.O. Saar, P. Bailey, and D.J. Lilja Accelerating geoscience and engineering system simulations on graphics hardware, Computers and Geosciences, 35/12, pp. 2353-2364, 2009. AbstractMany complex natural systems studied in the geosciences are characterized by simple local-scale interactions that result in complex emergent behavior. Simulations of these systems, often implemented in parallel using standard central processing unit (CPU) clusters, may be better suited to parallel processing environments with large numbers of simple processors. Such an environment is found in graphics processing units (GPUs) on graphics cards. This paper discusses GPU implementations of three example applications from computational fluid dynamics, seismic wave propagation, and rock magnetism. These candidate applications involve important numerical modeling techniques, widely employed in physical system simulations, that are themselves examples of distinct computing classes identified as fundamental to scientific and engineering computing. The presented numerical methods (and respective computing classes they belong to) are: (1) a lattice-Boltzmann code for geofluid dynamics (structured grid class); (2) a spectral-finite-element code for seismic wave propagation simulations (sparse linear algebra class); and (3) a least-squares minimization code for interpreting magnetic force microscopy data (dense linear algebra class). Significant performance increases (between 10×× and 30×× in most cases) are seen in all three applications, demonstrating the power of GPU implementations for these types of simulations and, more generally, their associated computing classes. 12 Walsh, S.D.C., H. Burwinkle, and M.O. Saar A new partial-bounceback lattice-Boltzmann method for fluid flow through heterogeneous media, Computers and Geosciences, 35/6, pp. 1186-1193, 2009. AbstractPartial-bounceback lattice-Boltzmann methods employ a probabilistic meso-scale model that varies individual lattice node properties to reflect a material’s local permeability. These types of models have great potential in a range of geofluid, and other science and engineering, simulations of complex fluid flow. However, there are several different possible approaches for formulating partial-bounceback algorithms. This paper introduces a new partial-bounceback algorithm and compares it to two pre-existing partial-bounceback models. Unlike the two other partial-bounceback methods, the new approach conserves mass in heterogeneous media and shows improvements in simulating buoyancy-driven flow as well as diffusive processes. Further, the new model is better-suited for parallel processing implementations, resulting in faster simulations. Finally, we derive an analytical expression for calculating the permeability in all three models; a critical component for accurately matching simulation parameters to physical permeabilities. 11 Walsh, S.D.C., and M.O. Saar Magma yield stress and permeability: Insights from multiphase percolation theory, Journal of Volcanology and Geothermal Research, 177, pp. 1011-1019, 2008. AbstractMagmas often contain multiple interacting phases of embedded solid and gas inclusions. Multiphase percolation theory provides a means of modeling assemblies of these different classes of magmatic inclusions in a simple, yet powerful way. Like its single phase counterpart, multiphase percolation theory describes the connectivity of discrete inclusion assemblies as a function of phase topology. In addition, multiphase percolation employs basic laws to distinguish separate classes of objects and is characterized by its dependency on the order in which the different phases appear. This paper examines two applications of multiphase percolation theory: the fi rst considers how the presence of bubble inclusions in fl uences yield stress onset and growth in a magma’s crystal network; the second examines the effect of bi-modal bubble-size distributions on magma permeability. We fi nd that the presence of bubbles induces crystal clustering, thereby 1) reducing the percolation threshold, or critical crystal volume fraction, φ c , at which the crystals form a space-spanning network providing a minimum yield stress, and 2) resulting in a larger yield stress for a given crystal volume fraction above φ c . This increase in the yield stress of the crystal network may also occur when crystal clusters areformed due toprocesses otherthanbubble formation, such as heterogeneouscrystallization, synneusis, and heterogeneity due to deformation or fl ow. Further, we fi nd that bimodal bubble size distributions can signi fi cantly affect the permeability of the system beyond the percolation threshold. This study thus demonstrates that larger-scale structures and topologies, as well as the order in which different phases appear, can have signi fi cant effects on macroscopic properties in multiphase materials. 10 Walsh, S.D.C., and M.O. Saar Numerical Models of Stiffness and Yield Stress Growth in Crystal-Melt Suspensions, Earth and Planetary Science Letters, 267/1-2, pp. 32-44, 2008. AbstractMagmas and other suspensions that develop sample-spanning crystal networks undergo a change in rheology from Newtonian to Bingham flow due to the onset of a finite yield stress in the crystal network. Although percolation theory provides a prediction of the crystal volume fraction at which this transition occurs, the manner in which yield stress grows with increasing crystal number densities is less-well understood. This paper discusses a simple numerical approach that models yield stress in magmatic crystalline assemblies. In this approach, the crystal network is represented by an assembly of soft-core interpenetrating cuboid (rectangular prism) particles, whose mechanical properties are simulated in a network model. The model is used to investigate the influence of particle shape and alignment anisotropy on the yield stress of crystal networks with particle volume fractions above the percolation threshold. In keeping with previous studies, the simulation predicts a local minimum in the onset of yield stress for assemblies of cubic particles, compared to those with more anisotropic shapes. The new model also predicts the growth of yield stress above (and close to) the percolation threshold. The predictions of the model are compared with results obtained from a critical path analysis. Good agreement is found between a characteristic stiffness obtained from critical path analysis, the growth in assembly stiffness predicted by the model (both of which have approximately cubic power-law exponents) and, to a lesser extent, the growth in yield stress (with a power-law exponent of 3.5). The effect of preferred particle alignment on yield stress is also investigated and found to obey similar power-law behavior. 9 Edwards, R.A., B. Rodriguez-Brito, L. Wegley, M. Haynes, M. Breitbart, D.M. Petersen, M.O. Saar, S.C. Alexander, E.C. Alexander Jr., and F. Rohwer Using pyrosequencing to shed light on deep mine microbial ecology, BMC Genomics, 2006. Abstract Background Contrasting biological, chemical and hydrogeological analyses highlights the fundamental processes that shape different environments. Generating and interpreting the biological sequence data was a costly and time-consuming process in defining an environment. Here we have used pyrosequencing, a rapid and relatively inexpensive sequencing technology, to generate environmental genome sequences from two sites in the Soudan Mine, Minnesota, USA. These sites were adjacent to each other, but differed significantly in chemistry and hydrogeology. Results Comparisons of the microbes and the subsystems identified in the two samples highlighted important differences in metabolic potential in each environment. The microbes were performing distinct biochemistry on the available substrates, and subsystems such as carbon utilization, iron acquisition mechanisms, nitrogen assimilation, and respiratory pathways separated the two communities. Although the correlation between much of the microbial metabolism occurring and the geochemical conditions from which the samples were isolated could be explained, the reason for the presence of many pathways in these environments remains to be determined. Despite being physically close, these two communities were markedly different from each other. In addition, the communities were also completely different from other microbial communities sequenced to date. Conclusion We anticipate that pyrosequencing will be widely used to sequence environmental samples because of the speed, cost, and technical advantages. Furthermore, subsystem comparisons rapidly identify the important metabolisms employed by the microbes in different environments. 8 Christiansen, L.B., S. Hurwitz, M.O. Saar, S.E. Ingebritsen, and P.A. Hsieh Seasonal seismicity at western United States volcanic centers, Earth and Planetary Science Letters, 240, pp. 307-321, 2005. AbstractWe examine 20-yr data sets of seismic activity from 10 volcanic areas in the western United States for annual periodic signals (seasonality), focusing on large calderas (Long Valley caldera and Yellowstone) and stratovolcanoes (Cascade Range). We apply several statistical methods to test for seasonality in the seismic catalogs. In 4 of the 10 regions, statistically significant seasonal modulation of seismicity (> 90% probability) occurs, such that there is an increase in the monthly seismicity during a given portion of the year. In five regions, seasonal seismicity is significant in the upper 3 km of the crust. Peak seismicity occurs in the summer and autumn in Mt. St. Helens, Hebgen Lake/Madison Valley, Yellowstone Lake, and Mammoth Mountain. In the eastern south moat of Long Valley caldera (LVC) peak seismicity occurs in the winter and spring. We quantify the possible external forcing mechanisms that could modulate seasonal seismicity. Both snow unloading and groundwater recharge can generate large stress changes of > 5 kPa at seismogenic depths and may thus contribute to seasonality. 7 Saar, M.O., M.C. Castro, C.M. Hall, M. Manga, and T.P. Rose Quantifying magmatic, crustal, and atmospheric helium contributions to volcanic aquifers using all stable noble gases: Implications for magmatism and groundwater flow, Geochemistry Geophysics Geosystems, 6/3, 2005. Abstract[1] We measure all stable noble gases (He, Ne, Ar, Kr, Xe) in spring waters in the Oregon Cascades volcanic arc and in eastern Oregon, USA. We show that in order to estimate magmatic helium (He) contributions it is critical to simultaneously consider He isotopic ratios, He concentrations, and mixing of He components. Our component mixing analysis requires consideration of all measured noble gases but no other elements and is particularly insightful when strong dilution by air-saturated water has occurred. In addition, this approach can allow distinction between crustal and magmatic He components and facilitates their identification in deep groundwaters that have been diluted by near-surface water. Using this approach, we show that some cold springs on the eastern flanks of the Oregon Cascades exhibit He isotopic ratios that indicate significant magmatic He contributions comparable to those observed in thermal springs on the western flanks. Furthermore, while these magmatic He contributions are largest in deep groundwaters near the Cascades crest, greater magmatic excess He fractions than may be inferred from He isotopic ratios alone are present in all (deep) groundwaters including those at larger distances (>70 km) from the volcanic arc. We also suggest that excess He and heat discharge without dilution by air-saturated water may be restricted to spring discharge along faults. 6 Saar, M.O., and M. Manga Depth dependence of permeability in the Oregon Cascades inferred from hydrogeologic, thermal, seismic, and magmatic modeling constraints, Journal of Geophysical Research, 109/B4, B04204, 2004. Abstract[1] We investigate the decrease in permeability, k, with depth, z, in the Oregon Cascades employing four different methods. Each method provides insight into the average permeability applicable to a different depth scale. Spring discharge models are used to infer shallow (z < 0.1 km) horizontal permeabilities. Coupled heat and groundwater flow simulations provide horizontal and vertical k for z < 1 km. Statistical investigations of the occurrences of earthquakes that are probably triggered by seasonal groundwater recharge yield vertical k for z < 5 km. Finally, considerations of magma intrusion rates and water devolatilization provide estimates of vertical k for z < 15 km. For depths >0.8 km, our results agree with the power law relationship, k = 10−14 m2 (z/1 km)−3.2, suggested by Manning and Ingebritsen [1999] for continental crust in general. However, for shallower depths (typically z ≤ 0.8 km and up to z ≤ 2) we propose an exponential relationship, k = 5 × 10−13 m2 exp (−z/0.25 km), that both fits data better (at least for the Cascades and seemingly for continental crust in general) and allows for a finite near-surface permeability and no singularity at zero depth. In addition, the suggested functions yield a smooth transition at z = 0.8 km, where their permeabilities and their gradients are similar. Permeabilities inferred from the hydroseismicity model at Mount Hood are about one order of magnitude larger than expected from the above power law. However, higher permeabilities in this region may be consistent with advective heat transfer along active faults, causing observed hot springs. Our simulations suggest groundwater recharge rates of 0.5 ≤ uR ≤ 1 m/yr and a mean background heat flow of Hb ≈ 0.080–0.134 W/m2 for the investigated region. 5 Jellinek, M.A., M. Manga, and M.O. Saar Did melting glaciers cause volcanic eruptions in eastern California? Probing the mechanics of dike formation, Journal of Geophysical Research, 109/B9, B09206, 2004. Abstract[1] A comparison of time series of basaltic and silicic eruptions in eastern California over the last 400 kyr with the contemporaneous global record of glaciation suggests that this volcanism is influenced by the growth and retreat of glaciers occurring over periods of about 40 kyr. Statistically significant cross correlations between changes in eruption frequency and the first derivative of the glacial time series imply that the temporal pattern of volcanism is influenced by the rate of change in ice volume. Moreover, calculated time lags for the effects of glacial unloading on silicic and basaltic volcanism are distinct and are 3.2 ± 4.2 kyr and 11.2 ± 2.3 kyr, respectively. A theoretical model is developed to investigate whether the increases in eruption frequency following periods of glacial unloading are a response ultimately controlled by the dynamics of dike formation. Applying results from the time series analysis leads, in turn, to estimates for the critical magma chamber overpressure required for eruption as well as constraints on the effective viscosity of the wall rocks governing dike propagation. 4 Saar, M.O., and M. Manga Seismicity induced by seasonal groundwater recharge at Mt. Hood, Oregon, Earth and Planetary Science Letters, 214, pp. 605-618, 2003. AbstractGroundwater recharge at Mt. Hood, Oregon, is dominated by spring snow melt which provides a natural large- amplitude and narrow-width pore-fluid pressure signal. Time delays between this seasonal groundwater recharge and seismicity triggered by groundwater recharge can thus be used to estimate large-scale hydraulic diffusivities and the state of stress in the crust. We approximate seasonal variations in groundwater recharge with discharge in runoff- dominated streams at high elevations. We interpolate the time series ofnumber ofearthquakes, N , seismic moment, M o , and stream discharge, Q , and determine cross-correlation coefficients at equivalent frequency bands between Q and both N and M o . We find statistically significant correlation coefficients at a mean time lag of about 151 days. This time lag and a mean earthquake depth of about 4.5 km are used in the solution to the pressure diffusion equation, under periodic (1 year) boundary conditions, to estimate a hydraulic diffusivity of U W 10 3 1 m 2 /s, a hydraulic conductivity ofabout K h W 10 3 7 m/s, and a permeability ofabout k W 10 3 15 m 2 . Periodic boundary conditions also allow us to determine a critical pore-fluid pressure fraction, P P / P 0 W 0.1, ofthe applied near-surface pore-fluid pressure perturbation, P 0 W 0.1 MPa, that has to be reached at the mean earthquake depth to cause hydroseismicity. The low magnitude of P P W 0.01 MPa is consistent with other studies that propose 0.01 9 P P 9 0.1 MPa and suggests that the state of stress in the crust near Mt. Hood could be near critical for failure. Therefore, we conclude that, while earthquakes occur throughout the year at Mt. Hood, elevated seismicity levels along pre-existing faults south of Mt. Hood during summer months are hydrologically induced by a reduction in effective stress 3 Saar, M.O., and M. Manga Continuum percolation for randomly oriented soft-core prisms, Physical Review E, 65/056131, 2002. AbstractWe study continuum percolation of three-dimensional randomly oriented soft-core polyhedra (prisms). The prisms are biaxial or triaxial and range in aspect ratio over six orders of magnitude. Results for prisms are compared with studies for ellipsoids, rods, ellipses, and polygons and differences are explained using the concept of the average excluded volume, ⟨vex⟩. For large-shape anisotropies we find close agreement between prisms and most of the above-mentioned shapes for the critical total average excluded volume, nc⟨vex⟩, where nc is the critical number density of objects at the percolation threshold. In the extreme oblate and prolate limits simulations yield nc⟨vex⟩≈2.3 and nc⟨vex⟩≈1.3, respectively. Cubes exhibit the lowest-shape anisotropy of prisms minimizing the importance of randomness in orientation. As a result, the maximum prism value, nc⟨vex⟩≈2.79, is reached for cubes, a value close to nc⟨vex⟩=2.8 for the most equant shape, a sphere. Similarly, cubes yield a maximum critical object volume fraction of φc=0.22. φc decreases for more prolate and oblate prisms and reaches a linear relationship with respect to aspect ratio for aspect ratios greater than about 50. Curves of φc as a function of aspect ratio for prisms and ellipsoids are offset at low-shape anisotropies but converge in the extreme oblate and prolate limits. The offset appears to be a function of the ratio of the normalized average excluded volume for ellipsoids over that for prisms, R=⟨¯vex⟩e/⟨¯vex⟩p. This ratio is at its minimum of R=0.758 for spheres and cubes, where φc(sphere)=0.2896 may be related to φc(cube)=0.22 by φc(cube)=1−[1−φc(sphere)]R=0.23. With respect to biaxial prisms, triaxial prisms show increased normalized average excluded volumes, ⟨¯vex⟩, due to increased shape anisotropies, resulting in reduced values of φc. We confirm that Bc=nc⟨vex⟩=2Cc applies to prisms, where Bc and Cc are the average number of bonds per object and average number of connections per object, respectively. 2 Saar, M.O., M. Manga, K.V. Cashman, and S. Fremouw Numerical models of the onset of yield strength in crystal-melt suspensions, Earth and Planetary Science Letters, 187, pp. 367-379, 2001. AbstractThe formation of a continuous crystal network in magmas and lavas can provide finite yield strength, τy, and can thus cause a change from Newtonian to Bingham rheology. The rheology of crystal–melt suspensions affects geological processes, such as ascent of magma through volcanic conduits, flow of lava across the Earth’s surface, melt extraction from crystal mushes under compression, convection in magmatic bodies, and shear wave propagation through partial melting zones. Here, three-dimensional numerical models are used to investigate the onset of ‘static’ yield strength in a zero-shear environment. Crystals are positioned randomly in space and can be approximated as convex polyhedra of any shape, size and orientation. We determine the critical crystal volume fraction, φc, at which a crystal network first forms. The value of φc is a function of object shape and orientation distribution, and decreases with increasing randomness in object orientation and increasing shape anisotropy. For example, while parallel-aligned convex objects yield φc=0.29, randomly oriented cubes exhibit a maximum φc of 0.22. Approximations of plagioclase crystals as randomly oriented elongated and flattened prisms (tablets) with aspect ratios between 1:4:16 and 1:1:2 yield 0.08<φc<0.20, respectively. The dependence of φc on particle orientation implies that the flow regime and resulting particle ordering may affect the onset of yield strength. φc in zero-shear environments is a lower bound for φc. Finally the average total excluded volume is used, within its limitation of being a ‘quasi-invariant’, to develop a scaling relation between τy and φ for suspensions of different particle shapes. 1 Saar, M.O., and M. Manga Permeability-porosity relationship in vesicular basalts., Geophysical Research Letters, 26/1, pp. 111-114, 1999. AbstractThe permeability κ and porosity ϕ of vesicular basalts are measured. The relationship between κ and ϕ reflects the formation and emplacement of the basalts and can be related to the crystal and vesicle microstructure obtained by image analysis. Standard theoretical models relating κ and ϕ that work well for granular materials are unsuccessful for vesicular rocks due to the fundamental difference in pore structure. Specifically, κ in vesicular rocks is governed by apertures between bubbles. The difference between calculated and measured κ reflects the small size of these apertures with aperture radii typically O(10) times smaller than the mean bubble radii.

### PROCEEDINGS REFEREED

 10 Myre, J.M., E. Frahm, D.J. Lilja, and M.O. Saar Solving Large Dense Least-Squares Problems: Preconditioning to Take Conjugate Gradient From Bad in Theory, to Good in Practice, IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), pp. 987-995, 2018. AbstractSince its inception by Gauss, the least-squares problem has frequently arisen in science, mathematics, and engineering. Iterative methods, such as Conjugate Gradient Normal Residual (CGNR), have been popular for solving sparse least-squares problems, but have historically been regarded as undesirable for dense applications due to poor convergence. We contend that this traditional “common knowledge” should be reexamined. Preconditioned CGNR, and perhaps other iterative methods, should be considered alongside standard methods when addressing large dense least-squares problems. In this paper we present TNT, a dynamite method for solving large dense least-squares problems. TNT implements a Cholesky preconditioner for the CGNR fast iterative method. The Cholesky factorization provides a preconditioner that, in the absence of round-off error, would yield convergence in a single iteration. Through this preconditioner and good parallel scaling, TNT provides improved performance over traditional least-squares solvers allowing for accelerated investigations of scientific and engineering problems. We compare a parallel implementations of TNT to parallel implementations of other conventional methods, including the normal equations and the QR method. For the small systems tested (15000 × 15000 or smaller), it is shown that TNT is capable of producing smaller solution errors and executing up to 16× faster than the other tested methods. We then apply TNT to a representative rock magnetism inversion problem where it yields the best solution accuracy and execution time of all tested methods. 9 Garapati, N., B.M. Adams, J.M. Bielicki, P. Schaedle, J.B. Randolph, T.H. Kuehn, and M.O. Saar A Hybrid Geothermal Energy Conversion Technology – A Potential Solution for Production of Electricity from Shallow Geothermal Resources, Energy Procedia, 114, pp. 7107-7117, 2017. AbstractGeothermal energy has been successfully employed in Switzerland for more than a century for direct use but presently there is no electricity being produced from geothermal sources. After the nuclear power plant catastrophe in Fukushima, Japan, the Swiss Federal Assembly decided to gradually phase out the Swiss nuclear energy program. Deep geothermal energy is a potential resource for clean and nearly CO2-free electricity production that can supplant nuclear power in Switzerland and worldwide. Deep geothermal resources often require enhancement of the permeability of hot-dry rock at significant depths (4-6 km), which can induce seismicity. The geothermal power projects in the Cities of Basel and St. Gallen, Switzerland, were suspended due to earthquakes that occurred during hydraulic stimulation and drilling, respectively. Here we present an alternative unconventional geothermal energy utilization approach that uses shallower, lower-temperature, naturally permeable regions, that drastically reduce drilling costs and induced seismicity. This approach uses geothermal heat to supplement a secondary energy source. Thus this hybrid approach may enable utilization of geothermal energy in many regions in Switzerland and elsewhere, that otherwise could not be used for geothermal electricity generation. In this work, we determine the net power output, energy conversion efficiencies, and economics of these hybrid power plants, where the geothermal power plant is actually a CO2-based plant. Parameters varied include geothermal reservoir depth (2.5-4.5 km) and turbine inlet temperature (100-220 °C) after auxiliary heating. We find that hybrid power plants outperform two individual, i.e., stand-alone geothermal and waste-heat power plants, where moderate geothermal energy is available. Furthermore, such hybrid power plants are more economical than separate power plants. 8 Rossi, E., M. Kant, F. Amann, M.O. Saar, and P. Rudolf von Rohr The effects of flame-heating on rock strength: Towards a new drilling technology, Proceedings ARMA 2017, 2017. AbstractThe applicability of a combined thermo-mechanical drilling technique is investigated. The working principle of this method is based on the implementation of a heat source as a mean to either provoke thermal spallation on the surface or to weaken the rock material, when spallation is not possible. Thermal spallation drilling has already been proven to work in hard crystalline rocks, however, several difficulties hamper its application for deep resource exploitation. In order to prove the effectiveness of a combined thermo-mechanical drilling method, the forces required to export the treated sandstone material with a polycrystalline diamond compact (PDC) cutter are analyzed. The main differences between oven and flame treatments are studied by comparing the resulting strength after heat-treating the samples up to temperatures of $$650\, ^{\circ}C$$ and for heating rates ranging from $$0.17 \,^{\circ}C/s$$ to $$20 ^{\circ}C/s$$. For moderate temperatures ($$300-450 \,^{\circ}C$$) the unconfined compressive strength after flame treatments monotonously decreased, opposed to the hardening behavior observed after oven treatments. Thermally induced intra-granular cracking and oxidation patterns served as an estimation of the treated depth due to the flame heat treatment. Therefore, conclusions on preferred operating conditions of the drilling system are drawn based on the experimental results. 7 Vogler, D., R.R. Settgast, C.S. Sherman, V.S. Gischig, R. Jalali, J.A. Doetsch, B. Valley, K.F. Evans, F. Amann, and M.O. Saar Modeling the Hydraulic Fracture Stimulation performed for Reservoir Permeability Enhancement at the Grimsel Test Site, Switzerland, Proceedings of the 42nd Workshop on Geothermal Reservoir Engineering Stanford University, 2017. AbstractIn-situ hydraulic fracturing has been performed on the decameter scale in the Deep Underground rock Laboratory (DUG Lab) at the Grimsel Test Site (GTS) in Switzerland in order to measure the minimum principal stress magnitude and orientation. Conducted tests were performed in a number of boreholes, with 3–4 packer intervals in each borehole subjected to repeated injection. During each test, fluid injection pressure, injection flow rate and microseismic events were recorded amongst others. Fully coupled 3D simulations have been performed with the LLNL’s GEOS simulation framework. The methods applied in the simulation of the experiments address physical processes such as rock deformation/stress, LEFM fracture mechanics, fluid flow in the fracture and matrix, and the generation of micro-seismic events. This allows to estimate the distance of fracture penetration during the injection phase and correlate the simulated injection pressure with experimental data during injection, as well as post shut-in. Additionally, the extent of the fracture resulting from simulations of fracture propagation and microseismic events are compared with the spatial distribution of the microseismic events recorded in the experiment. 6 Garapati, N., J. Randolph, S. Finsterle, and M.O. Saar Simulating Reinjection of Produced Fluids Into the Reservoir, Proceedings of 41st Workshop on Geothermal Reservoir Engineering, 2016. AbstractABSTRACT In order to maintain reservoir pressure and stability and to reduce reservoir s ubsidence, reinjection of produced fluids into the reservoir is common practice . Furthermore, studies by Karvounis and Jenny (2012 ; 2014), Buscheck et al. (2015), and Saar et al. (2015) found that preheating the working fluid in shallow reservoirs and then injecting the fluid into a deep reservoir can increase the reservoir life span, the heat extraction efficiency, and the economic gains of a geothermal power plant . We have modif ied the TOUGH2 simulator to enable the reinjection of produced fluids with the same chemical composition as the produced fluid and with either a prescribed or the production temperature . T he latter capability is useful, for example, for simulating injecti on of produced fluid into another (e .g., deeper) reservoir without energy extraction. Each component of the fluid mixture , produced from the production well , is reinjected into the reservoir as an individual source term. In the current study, we investigate a CO 2 – based geothermal system and focus on the effects of reinjecting small amounts of brine that are produced along with the CO 2 . Brine has a significantly smaller mobility (inverse kinematic viscosity) than supercritical CO 2 at a given temperature and thus accumulates near the injection well. Such brine accumulation reduces the relative permeability for the CO 2 phase, which in turn increases the pore – fluid pressure around the injection well and reduces the well in j ectivity index. For this reason, and as injection of two fluid phases is pr oblematic, we recommend removal of any brine from the produced fluid before the cooled CO 2 is reinjected into the reservoir. We also study the performance of a multi – level geothermal system (Karvounis and Jenny, 2012; 2014; Saar et al., 2015) by injection of preheated brine from a shallow reservoir (1.5 – 3 km) into a deep reservoir (5 km). We f i nd that preheating brine at the shallow reservoir extends the lifespan of the deep, hot reservoir, thereby increasing the total power production. 5 Garapati, N., J.B. Randolph, J.L. Valencia Jr., and M.O. Saar Design of CO2-Plume Geothermal (CPG) subsurface system for various geologic parameters, Proceedings of the Fifth International Conference on Coupled Thermo-Hydro-Mechanical-Chemical (THMC) Processes in Geosystems: Petroleum and Geothermal Reservoir Geomechanics and Energy Resource Extraction, 2015. AbstractRecent geotechnical research shows that geothermal heat can be efficiently mined by circulating carbon dioxide through naturally permeable rock formations — a method called CO2 Plume Geothermal — the same geologic reservoirs that are suitable for deep saline aquifer CO2 sequestration or enhanced oil recovery. This paper describes the effect of thermal drawdown on reservoir pressure buildup during sequestration operations, revealing that geothermal heat mining can decrease overpressurization by 10% or more. Geothermal Energy Production at Geologic CO2 Sequestration sites: Impact of Thermal Drawdown on Reservoir Pressure (PDF Download Available). Available from: https://www.researchgate.net/publication/273193986_Geothermal_Energy_Production_at_Geologic_CO2_Sequestration_sites_Impact_of_Thermal_Drawdown_on_Reservoir_Pressure [accessed Jun 12, 2017]. 4 Garapati, N., J.B. Randolph, and M.O. Saar Superheating Low-Temperature Geothermal Resources to Boost Electricity Production, Proceedings of the 40th Workshop on Geothermal Reservoir Engineering 2015, 2, pp. 1210-1221, 2015. AbstractLow-temperature geothermal resources (<150°C) are typically more effective for direct use, i.e., district heating, than for electricity production. District or industrial heating, however, requires that the heat resource is close to residential or industrial demands in order to be efficient and thus economic. However, if a low-temperature geothermal resource is combined with an additional or secondary energy source that is ideally renewable, such as solar, biomass, biogas, or waste heat, but could be non-renewable, such as natural gas, the thermodynamic quality of the energy source increases, potentially enabling usage of the combined energy sources for electricity generation. Such a hybrid geothermal power plant therefore offers thermodynamic advantages, often increasing the overall efficiency of the combined system above that of the additive power output from two stand-alone, separate plants (one using geothermal energy alone and the other using the secondary energy source alone) for a wide range of operating conditions. Previously, fossil superheated and solar superheated hybrid power plants have been considered for brine/water based geothermal systems, especially for enhanced geothermal systems. These previous studies found, that the cost of electricity production can typically be reduced when a hybrid plant is operated, compared to operating individual plants. At the same time, using currently-available high-temperature energy conversion technologies reduces the time and cost required for developing other less-established energy conversion technologies. Adams et al. (2014) found that CO 2 as a subsurface working fluid produces more net power than when brine systems are employed at low to moderate reservoir depths, temperatures, and permeabilities. Therefore in this work, we compare the performance of hybrid geothermal power plants that use brine or, importantly, CO 2 (which constitutes the new research component) as the subsurface working fluid, irrespective of the secondary energy source used for superheating, over a range of parameters. These parameters include geothermal reservoir depth and superheated fluid temperature before passing through the energy conversion system. The hybrid power plant is modeled using two software packages: 1) TOUGH2 (Pruess, 2004), which is employed for the subsurface modeling of geothermal heat and fluid extraction as well as for fluid reinjection into the reservoir, and 2) Engineering Equation Solver (EES), which is used to simulate well bore fluid flow and surface power plant performance. We find here that for geothermal systems combined with a secondary energy source (i.e., a hybrid system), the maximum power production for a given set of reservoir parameters is highly dependent on the configuration of the power system. The net electricity production from a hybrid system is larger than that from the individual plants combined for all scenarios considered for brine systems and for low-grade secondary energy resources for CO 2 based geothermal systems. Superheating of Low-Temperature Geothermal Working Fluids to Boost Electricity Production: Comparison between Water and CO2 Systems (PDF Download Available). Available from: https://www.researchgate.net/publication/271702360_Superheating_of_Low-Temperature_Geothermal_Working_Fluids_to_Boost_Electricity_Production_Comparison_between_Water_and_CO2_Systems [accessed Jun 12, 2017]. 3 Saar, M.O., Th. Buscheck, P. Jenny, N. Garapati, J.B. Randolph, D. Karvounis, M. Chen, Y. Sun, and J.M. Bielicki Numerical Study of Multi-Fluid and Multi-Level Geothermal System Performance, Proceedings World Geothermal Congress 2015, 2015. AbstractWe introduce the idea of combining multi-fluid and multi-level geothermal systems with two reservoirs at depths of 3 and 5 km. In the base case, for comparison, the two reservoirs are operated independently, each as a multi-fluid (brine and carbon dioxide) reservoir that uses a number of horizontal, concentric injection and production well rings. When the shallow and the deep reservoirs are operated in an integrated fashion, in the shallow reservoir, power is produced only from the carbon dioxide (CO 2), while the brine is geothermally preheated in the shallow multi-fluid reservoir, produced, and then reinjected at the deeper reservoir’s brine injectors. The integrated reservoir scenarios are further subdivided into two cases: In one scenario, both brine (preheated in the shallow reservoir) and CO 2 (from the surface) are injected separately into the deeper reservoir’s appropriate injectors and both fluids are produced from their respective deep reservoir producers to generate electricity. In the other scenario, only preheated brine is injected into, and produced from, the deep reservoir for electric power generation. We find that integrated, vertically stacked, multi-fluid geothermal systems can result in improved system efficiency when power plant lifespans exceed ~30 years. In addition, preheating of brine before deep injection reduces brine overpressurization in the deep reservoir, reducing the risk of fluid-induced seismicity. Furthermore, CO2-Plume Geothermal (CPG) power plants in general, and the multi-fluid, multi-level geothermal system described here in particular, assign a value to CO2, which in turn may partially or fully offset the high costs of carbon capture at fossil-energy power plants and of CO2 injection, thereby facilitating economically feasible carbon capture and storage (CCS) operations that render fossil-energy power plants green. From a geothermal power plant perspective, the system results in a CO2 sequestering geothermal power plant with a negative carbon footprint. Finally, energy return on well costs and operational flexibility can be greater for integrated geothermal reservoirs, providing additional options for bulk and thermal energy storage, compared to equivalent, but separately operated reservoirs. System economics can be enhanced by revenues related to efficient delivery of large-scale bulk energy storage and ancillary services products (frequency regulation, load following, and spinning reserve), which are essential for electric grid integration of intermittently available renewable energy sources, such as wind and solar. These capabilities serve to stabilize the electric grid and promote development of all renewable energies, beyond geothermal energy. Numerical Study of Multi-Fluid and Multi-Level Geothermal System Performance (PDF Download Available). Available from: https://www.researchgate.net/publication/274138343_Numerical_Study_of_Multi-Fluid_and_Multi-Level_Geothermal_System_Performance [accessed Jun 12, 2017]. 2 Buscheck, T.A., J.M. Bielicki, M. Chen, Y. Sun, Y. Hao, T.A. Edmunds, J.B. Randolph, and M.O. Saar Multi-Fluid Sedimentary Geothermal Energy Systems for Dispatchable Renewable Electricity, Proceedings to the World Geothermal Congress, 2015. AbstractSedimentary geothermal resources typically have lower temperatures and energy densities than hydrothermal resources, but they often have higher permeability and larger areal extents. Consequently, spacing between injection and production wells is likely to be wider in sedimentary resources, which can result in more fluid pressure loss, increa sing the parasitic cost of powering the working fluid recirculation system, compared to hydrothermal systems . For hydrostatic geothermal resources , extracting heat requires that brine be lifted up production wells, such as with submersible pumps, which can consume a large portion of the electricity generated by the power plant. CO 2 is being considered as an alternative working fluid (also termed a supplemental fluid) because its advantageous thermophysical properties reduce this parasitic cost, and because of the synergistic benefit of geologic CO 2 sequestration (GCS). We expand on this idea by: (1) adding the option for multiple supplemental fluids (N 2 as well as CO 2 ) and injecting these fluids to create overpressured reservoir conditions , (2) utiliz ing up to three working fluids: brine, CO 2 , and N 2 for heat extraction, (3) using a well pattern designed to store supplemental fluid and pressure , and (4) time – shifting the parasitic load associated with fluid recirculation to provide ancillary services ( frequen cy regulat ion , load fo llowing , and spinning reserve) and bulk energy storage (BES) . Our approach uses concentric rings of horizontal wells to create a hydraulic divide to store supplemental fluid and pressure, much like a hydroelectric dam. While, as with any geothermal system, electricity production can be run as a base – load power source, p roduction wells can alternatively b e controlled like a spillway to supply power when demand is greatest. For conventional geothermal power, the parasitic power load for fluid recirculation is synchronous with gross power output. In contrast, our approach time – shift s much of this parasitic load, which is dominated by the power required to pressurize and inject brine . Th us, most of the parasitic load can be scheduled durin g minimum power demand or when, due to its inherent var iability, there is a surplus of renewable energy on the grid . Energy storage is almost 100 percent efficient because it is achieved by time – shifting the parasitic load. Consequently, net power can near ly equal gross power during peak demand so that geothermal energy can be used as a form of high – efficiency BES at large scales . A further benefit of our approach is that production rates (per well) can exceed the capacity of submersible pumps and thereby t ake advantage of the productivity of horizontal wells and better leverage we ll costs — which often constitute a major portion of capital costs . Our vision is a n efficient, dispatchable , renewable electricity system approach that facilitates deep market penet ration of all renewable energy sources: wind, solar, and geothermal, whi le utilizing and permanently storing CO 2 in a commercially viable manner 1 Bailey, P., J. Myre, S.C.D. Walsh, D.J. Lilja, and M.O. Saar Accelerating Lattice Boltzmann Fluid Flow Simulations Using Graphics Processors, IEEE, pp. 550-557, 2009. AbstractLattice Boltzmann Methods (LBM) are used for the computational simulation of Newtonian fluid dynamics. LBM-based simulations are readily parallelizable; they have been implemented on general-purpose processors, field-programmable gate arrays (FPGAs), and graphics processing units (GPUs). Of the three methods, the GPU implementations achieved the highest simulation performance per chip. With memory bandwidth of up to 141 GB/s and a theoretical maximum floating point performance of over 600 GFLOPS, CUDA-ready GPUs from NVIDIA provide an attractive platform for a wide range of scientific simulations, including LBM. This paper improves upon prior single-precision GPU LBM results for the D3Q19 model by increasing GPU multiprocessor occupancy, resulting in an increase in maximum performance by 20%, and by introducing a space-efficient storage method which reduces GPU RAM requirements by 50% at a slight detriment to performance. Both GPU implementations are over 28 times faster than a single-precision quad-core CPU version utilizing OpenMP.

### THESES

 2 Saar, M.O. Geological Fluid Mechanics Models at Various Scales, Dissertation University of California, Berkeley, 153 pp., 2003. AbstractIn this dissertation, I employ concepts from fluid mechanics to quantitatively investigate geological processes in hydrogeology and volcanology. These research topics are addressed by utilizing numerical and analytical models but also by conducting field and lab work. Percolation theory is of interest to a wide range of physical sciences and thus warrants research in itself. Therefore, I developed a computer code to study percolation thresholds of soft-core polyhedra. Results from this research are applied to study the onset of yield strength in crystal-melt suspensions such as magmas. Implications of yield strength development in suspensions, marking the transition from Newtonian to Bingham fluid, include the pahoehoe-‘a’a transition and the occurrence of effusive versus explosive eruptions. I also study interactions between volcanic processes and groundwater as well as between groundwater and seismicity (hydroseismicity). In the former case, I develop numerical and analytical models of coupled groundwater and heat transfer. Here, perturbations from a linear temperature-depth profile are used to determine groundwater flow patterns and rates. For the hydroseismicity project I investigate if seasonal elevated levels of seismicity at Mt. Hood, Oregon, are triggered by groundwater recharge. Both hydroseismicity and hydrothermal springs occur on the southern flanks of Mt. Hood. This suggests that both phenomena are related while also providing a connection between the research projects involving groundwater, heat flow, and seismicity. Indeed, seismicity may be necessary to keep faults from clogging thus allowing for sustained activity of hydrothermal springs. Finally, I present research on hydrologically induced volcanism, where a process similar to the one suggested for hydroseismicity is invoked. Here, melting of glaciers, or draining of lakes, during interglacial periods reduce the confining pressure in the subsurface which may promote dike formation and result in increased rates of volcanism. In general, problems discussed in this dissertation involve interactions among processes that are traditionally associated with separate research disciplines. However, numerous problems in the geosciences require a multidisciplinary approach, as demonstrated here. In addition, employing several analytical and numerical methods, such as signal processing, inverse theory, computer modeling, and percolation theory, allows me to study such diverse processes in a quantitative way. 1 Saar, M.O. The Relationship Between Permeability, Porosity, and Microstucture in Vesicular Basalts, MSc Thesis University of Oregon, 101 pp., 1998. AbstractThis thesis presents measurements of permeability, $$k$$, porosity, $$\Phi$$, and microstructural parameters of vesicular basalts. Measurements are compared with theoretical models. A percolation theory and a Kozeny-Carman model are used to interpret the measurements and to investigate relationships between porosity, microstructure, and permeability. Typical permeabilities for vesicular basalts are in the range of $$10^{-14}$$ < $$k$$ < $$10^{-10} m^2$$. Best permeability estimates, following power laws predicted by percolation theory, are obtained when samples are used that show 'impeded aperture widening' due to rapid cooling and no bubble collapse (scoria and some flow samples). However, slowly cooled diktytaxitic samples contain elongated, 'collapsed' bubbles. Measurements indicate that the vesicle pathway network remains connected and preserves high permeabilities. Image-analysis techniques are unsuccessful if used for Kozeny-Carman equation parameter determination for vesicular materials, probably because the average interbubble aperture size that determines $$k$$ is not resolved with such a technique.