uu.seUppsala University Publications
Change search
Link to record
Permanent link

Direct link
BETA
Niemi, Auli
Publications (10 of 82) Show all publications
Dessirier, B., Tsang, C.-F. & Niemi, A. (2018). A new scripting library for modeling flow and transport in fractured rock with channel networks. Computers & Geosciences, 111, 181-189
Open this publication in new window or tab >>A new scripting library for modeling flow and transport in fractured rock with channel networks
2018 (English)In: Computers & Geosciences, ISSN 0098-3004, E-ISSN 1873-7803, Vol. 111, p. 181-189Article in journal (Refereed) Published
Abstract [en]

Deep crystalline bedrock formations are targeted to host spent nuclear fuel owing to their overall low permeability. They are however highly heterogeneous and only a few preferential paths pertaining to a small set of dominant rock fractures usually carry most of the flow or mass fluxes, a behavior known as channeling that needs to be accounted for in the performance assessment of repositories. Channel network models have been developed and used to investigate the effect of channeling. They are usually simpler than discrete fracture networks based on rock fracture mappings and rely on idealized full or sparsely populated lattices of channels. This study reexamines the minimal requirements to describe a channel network in terms of groundwater flow and solute transport, leading to an extended description suitable for unstructured arbitrary networks of channels. An implementation of this formalism in a Python scripting library is presented and released along with this article. A new algebraic multigrid preconditioner delivers a significant speedup in the flow solution step compared to previous channel network codes. 3D visualization is readily available for verification and interpretation of the results by exporting the results to an open and free dedicated software. The new code is applied to three example cases to verify its results on full uncorrelated lattices of channels, sparsely populated percolation lattices and to exemplify the use of unstructured networks to accommodate knowledge on local rock fractures.

Keywords
channel network, groundwater flow, solute transport, graph theory
National Category
Oceanography, Hydrology and Water Resources
Research subject
Hydrology
Identifiers
urn:nbn:se:uu:diva-321572 (URN)10.1016/j.cageo.2017.11.013 (DOI)000423005900018 ()
Funder
Swedish Radiation Safety Authority, SSM2016-763
Available from: 2017-05-08 Created: 2017-05-08 Last updated: 2018-03-06Bibliographically approved
Dessirier, B., Tsang, C.-F. & Niemi, A. (2018). A new scripting library for modeling flow and transport in fractured rock with channel networks. Computers & Geosciences, 111, 181-189
Open this publication in new window or tab >>A new scripting library for modeling flow and transport in fractured rock with channel networks
2018 (English)In: Computers & Geosciences, ISSN 0098-3004, E-ISSN 1873-7803, Vol. 111, p. 181-189Article in journal (Refereed) Published
Abstract [en]

Deep crystalline bedrock formations are targeted to host spent nuclear fuel owing to their overall low permeability. They are however highly heterogeneous and only a few preferential paths pertaining to a small set of dominant rock fractures usually carry most of the flow or mass fluxes, a behavior known as channeling that needs to be accounted for in the performance assessment of repositories. Channel network models have been developed and used to investigate the effect of channeling. They are usually simpler than discrete fracture networks based on rock fracture mappings and rely on idealized full or sparsely populated lattices of channels. This study reexamines the fundamental parameter structure required to describe a channel network in terms of groundwater flow and solute transport, leading to an extended description suitable for unstructured arbitrary networks of channels. An implementation of this formalism in a Python scripting library is presented and released along with this article. A new algebraic multigrid preconditioner delivers a significant speedup in the flow solution step compared to previous channel network codes. 3D visualization is readily available for verification and interpretation of the results by exporting the results to an open and free dedicated software. The new code is applied to three example cases to verify its results on full uncorrelated lattices of channels, sparsely populated percolation lattices and to exemplify the use of unstructured networks to accommodate knowledge on local rock fractures.

Keywords
Channel network, Groundwater flow, Solute transport, Graph theory
National Category
Earth and Related Environmental Sciences
Identifiers
urn:nbn:se:uu:diva-343655 (URN)10.1016/j.cageo.2017.11.013 (DOI)000423005900018 ()
Available from: 2018-05-09 Created: 2018-05-09 Last updated: 2018-05-09Bibliographically approved
Hedayati, M., Wigston, A., Wolf, J. L., Rebscher, D. & Niemi, A. (2018). Impacts of SO2 gas impurity within a CO2 stream on reservoir rock of a CCS pilot site: Experimental and modelling approach. International Journal of Greenhouse Gas Control, 70, 32-44
Open this publication in new window or tab >>Impacts of SO2 gas impurity within a CO2 stream on reservoir rock of a CCS pilot site: Experimental and modelling approach
Show others...
2018 (English)In: International Journal of Greenhouse Gas Control, ISSN 1750-5836, E-ISSN 1878-0148, Vol. 70, p. 32-44Article in journal (Refereed) Published
Abstract [en]

In order to evaluate chemical impacts of SO2 impurity on reservoir rock during CO2 capture and storage in deep saline aquifers, several batch reactor experiments were performed on laboratory scale using core rock samples from the pilot CO2 injection site in Heletz. In this experiment, the samples were exposed to pure N-2(g), pure CO2(g), and CO2(g) with an impurity of 1.5% SO2(g) under reservoir conditions for pressure and temperature (14.5 MPa, 60 degrees C). Based on the set-up and the obtained experimental results, a batch chemical model was established using the numerical simulation program TOUGHREACT V3.0-OMP. Comparing laboratory and simulation data provides a better understanding of the rock-brine-gas interactions. In addition, it offers an evaluation of the capability of the model to predict chemical interactions in the target injection reservoir during exposure to pure and impure CO2. The best match between the geochemical model and experimental data was achieved when the reactive surface area of minerals in the model was adjusted in order to calibrate the kinetic rates of minerals. The simulations indicated that SO2(g) tends to dissolve rather quickly and oxidizes under a kinetic control. Hence, it has a stronger effect on the acidity of the brine than pure CO2(g) and as a result, increased mineral dissolution and caused the precipitation of sulfate and sulfide minerals. Ankerite, dolomite, and siderite, the most abundant carbonates in the sandstone rock sample, are subject to stronger dissolution in the presence of SO2 gas. The performed simulations confirmed a slower dissolution rate for ankerite and siderite than for dolomite. The model reproduced the precipitation of pyrite and anhydrite as observed in the laboratory. The dissolution of dolomite observed in the batch reaction test with pure N-2 is assumed to be due to slight contamination with oxygen and modelling supported this. The inclusion of SO2 increased the porosity over that of the pure CO2 case, and is thus considered to increase the permeability and injectivity of the reservoir as well. Exposure to SO2 also increased the concentration of trace elements. The calibrated kinetic parameters determined in this study will be used to model the injection and long-term behavior of CO2 at the Heletz field site, and may be used for similar geologic reservoirs.

National Category
Geochemistry
Identifiers
urn:nbn:se:uu:diva-340872 (URN)10.1016/j.ijggc.2018.01.003 (DOI)000428773100004 ()
Funder
EU, FP7, Seventh Framework Programme, 309102, 309067
Available from: 2018-02-05 Created: 2018-02-05 Last updated: 2018-06-01Bibliographically approved
Tsang, C.-F., Figueiredo, B. & Niemi, A. (2018). Importance of stress effects on inputs to fracture network models used for subsurface flow and transport studies. International Journal of Rock Mechanics And Mining Sciences, 101, 13-17
Open this publication in new window or tab >>Importance of stress effects on inputs to fracture network models used for subsurface flow and transport studies
2018 (English)In: International Journal of Rock Mechanics And Mining Sciences, ISSN 1365-1609, E-ISSN 1873-4545, Vol. 101, p. 13-17Article in journal (Refereed) Published
National Category
Geosciences, Multidisciplinary
Identifiers
urn:nbn:se:uu:diva-334138 (URN)10.1016/j.ijrmms.2017.11.012 (DOI)000418717200002 ()
Funder
Swedish Radiation Safety Authority, SSM2016-763The Geological Survey of Sweden (SGU), 1724
Available from: 2017-11-21 Created: 2017-11-21 Last updated: 2018-01-18Bibliographically approved
Wolf, J. L., Niemi, A., Bensabat, J., May, F., Ruetters, H. & Rebscher, D. (2017). 2D reactive transport simulations of CO2 streams containing impurities in a saline aquifer, Heletz, Israel. In: Dixon, T Laloui, L Twinning, S (Ed.), 13Th International Conference on Greenhouse Gas Control Technologies, Ghgt-13: . Paper presented at 13th International Conference on Greenhouse Gas Control Technologies (GHGT), NOV 14-18, 2016, Lausanne, SWITZERLAND (pp. 3537-3546). Elsevier
Open this publication in new window or tab >>2D reactive transport simulations of CO2 streams containing impurities in a saline aquifer, Heletz, Israel
Show others...
2017 (English)In: 13Th International Conference on Greenhouse Gas Control Technologies, Ghgt-13 / [ed] Dixon, T Laloui, L Twinning, S, Elsevier, 2017, p. 3537-3546Conference paper, Published paper (Refereed)
Abstract [en]

In order to evaluate the chemical impacts of CO2 stream impurities on reservoir rocks, 2D reactive transport simulations using the code TOUGHREACT V3.0 were performed. The underlying reservoir properties are based on in-situ data from the CO2 injection test site Heletz, Israel. Two different CO2 compositions (mole fractions 99 % CO2 + 1 % SO2 and 98.8 % CO2 + 1 % SO2 + 0.2 % NO2, respectively) were chosen to represent oxidising impurities. Different modelling approaches, namely trace gas transport (TGT) and additional brine injection (ABI), were applied to investigate the influence of these modelling approaches on qualitative and quantitative simulation results. The simulations using either approach show an accumulation of SO2 and NO2 close to the injection well due to the preferential dissolution of these acidic impurities compared to CO2. Both modelling approaches indicate the same general chemical impact and related mineral reactions. Within the affected rock volume a distinct ankerite to anhydrite conversion occurs, which slightly enhances porosity. While the same qualitative conclusions independently from the chosen modelling approach were obtained, the quantitative magnitude of mineral conversion and the spatial extent of impurity affected rock material depend on the chosen modelling approach and thus need further investigation with respect to e.g. validation by field test data. (C) 2017 The Authors. Published by Elsevier Ltd.

Place, publisher, year, edition, pages
Elsevier, 2017
Series
Energy Procedia, ISSN 1876-6102 ; 114
Keywords
Impurities, reactive transport, SO2, NO2, TOUGHREACT V3.0, CCS
National Category
Geosciences, Multidisciplinary
Identifiers
urn:nbn:se:uu:diva-349328 (URN)10.1016/j.egypro.2017.03.1483 (DOI)000419147303068 ()
Conference
13th International Conference on Greenhouse Gas Control Technologies (GHGT), NOV 14-18, 2016, Lausanne, SWITZERLAND
Funder
EU, FP7, Seventh Framework Programme, 309102
Available from: 2018-04-26 Created: 2018-04-26 Last updated: 2018-04-26Bibliographically approved
Tian, L., Wilkinson, R., Yang, Z., Power, H., Fagerlund, F. & Niemi, A. (2017). Gaussian Process Emulators for Quantifying Uncertainty in CO2 Spreading Predictions in Heterogeneous Media. Computers & Geosciences
Open this publication in new window or tab >>Gaussian Process Emulators for Quantifying Uncertainty in CO2 Spreading Predictions in Heterogeneous Media
Show others...
2017 (English)In: Computers & Geosciences, ISSN 0098-3004, E-ISSN 1873-7803Article in journal (Refereed) Published
Abstract [en]

We explore the use of Gaussian process emulators (GPE) in the numerical simulation of CO2 injection into a deep heterogeneous aquifer. The model domain is a two-dimensional, log-normally distributed stochastic permeability field. We first estimate the cumulative distribution functions (CDFs) of the CO2 breakthrough time and the total CO2 mass using a computationally expensive Monte Carlo (MC) simulation. We then show that we can accurately reproduce these CDF estimates with a GPE, using only a small fraction of the computational cost required by traditional MC simulation. In order to build a GPE that can predict the simulator output from a permeability field consisting of 1000s of values, we use a truncated Karhunen-Loève (K-L) expansion of the permeability field, which enables the application of the Bayesian functional regression approach. We perform a cross-validation exercise to give an insight of the optimization of the experiment design for selected scenarios: we find that it is sufficient to use 100s values for the size of the training set and that it is adequate to use as few as 15 K-L components. Our work demonstrates that GPE with truncated K-L expansion can be effectively applied to uncertainty analysis associated with modeling of multiphase flow and transport processes in heterogeneous media.

Keywords
CO2, Bayesian, Permeability, KL expansion, Monte Carlo, Cumulative distribution function, Uncertainty analysis
National Category
Geosciences, Multidisciplinary Computer Sciences
Identifiers
urn:nbn:se:uu:diva-298748 (URN)10.1016/j.cageo.2017.04.006 (DOI)000404697000011 ()
Funder
EU, FP7, Seventh Framework Programme, 227286EU, FP7, Seventh Framework Programme, 282900EU, FP7, Seventh Framework Programme, 309067
Available from: 2016-07-06 Created: 2016-07-06 Last updated: 2018-02-06Bibliographically approved
Basirat, F., Yang, Z. & Niemi, A. (2017). Pore-scale modeling of wettability effects on CO2–brine displacement during geological storage. Advances in Water Resources, 109, 181-195
Open this publication in new window or tab >>Pore-scale modeling of wettability effects on CO2–brine displacement during geological storage
2017 (English)In: Advances in Water Resources, ISSN 0309-1708, E-ISSN 1872-9657, Vol. 109, p. 181-195Article in journal (Refereed) Published
Abstract [en]

Wetting properties of reservoir rocks and caprocks can vary significantly, and they strongly influence geological storage of carbon dioxide in deep saline aquifers, during which CO2 is supposed to displace the resident brine and to become permanently trapped. Fundamental understanding of the effect of wettability on CO2-brine displacement is thus important for improving storage efficiency and security. In this study, we investigate the influence of wetting properties on two-phase flow of CO2 and brine at the pore scale. A numerical model based on the phase field method is implemented to simulate the two-phase flow of CO2-brine in a realistic pore geometry. Our focus is to study the pore-scale fluid-fluid displacement mechanisms under different wetting conditions and to quantify the effect of wettability on macroscopic parameters such as residual brine saturation, capillary pressure, relative permeability, and specific interfacial area. Our simulation results confirm that both the trapped wetting phase saturation and the normalized interfacial area increase with decreasing contact angle. However, the wetting condition does not appear to influence the CO2 breakthrough time and saturation. We also show that the macroscopic capillary pressures based on the pressure difference between inlet and outlet can differ significantly from the phase averaging capillary pressures for all contact angles when the capillary number is high ( log Ca > -5). This indicates that the inlet-outlet pressure difference may not be a good measure of the continuum-scale capillary pressure. In addition, the results show that the relative permeability of CO2 can be significantly lower in strongly water-wet conditions than in the intermediate-wet conditions.

Place, publisher, year, edition, pages
Elsevier, 2017
National Category
Oceanography, Hydrology and Water Resources
Research subject
Hydrology
Identifiers
urn:nbn:se:uu:diva-315304 (URN)10.1016/j.advwatres.2017.09.004 (DOI)000416037100014 ()
Funder
EU, FP7, Seventh Framework Programme, 309067Swedish Research Council, 637-2014-445Swedish Energy Agency, 43526-1
Available from: 2017-02-14 Created: 2017-02-14 Last updated: 2018-02-23Bibliographically approved
Figueiredo, B., Tsang, C.-F., Rutqvist, J. & Niemi, A. (2017). Study of hydraulic fracturing processes in shale formations with complex geological settings. Journal of Petroleum Science and Engineering, 152, 361-374
Open this publication in new window or tab >>Study of hydraulic fracturing processes in shale formations with complex geological settings
2017 (English)In: Journal of Petroleum Science and Engineering, ISSN 0920-4105, E-ISSN 1873-4715, Vol. 152, p. 361-374Article in journal (Refereed) Published
Abstract [en]

Hydraulic fracturing has been applied to extract gas from shale-gas reservoirs. Complicated geological settings,such as spatial variability of the rock mass properties, local heterogeneities, complex in situ stress field, and preexistingbedding planes and faults, could make hydraulic fracturing a challenging task. In order to effectivelyand economically recover gas from such reservoirs, it is crucial to explore how hydraulic fracturing performs insuch complex geological settings. For this purpose, numerical modelling plays an important role because suchconditions cannot be reproduced by laboratory experiments. This paper focuses on the analysis of the influenceof confining formations and pre-existing bedding planes and faults on the hydraulically-induced propagation ofa vertical fracture, which will be called injection fracture, in a shale-gas reservoir. An elastic-brittle model basedon material property degradation was implemented in a 2D finite-difference scheme and used for rock elementssubjected to tension and shear failure. A base case is considered, in which the ratio SR between the magnitudesof the horizontal and vertical stresses, the permeability kc of the confining formations, the elastic modulus Epand initial permeability kp of the bedding plane and the initial fault permeability kF are fixed at reasonablevalues. In addition, the influence of multiple bedding planes, is investigated. Changes in pore pressure andpermeability due to high pressure injection lasting 2 h were analysed. Results show that in our case during theinjection period the fracture reaches the confining formations and if the permeability of those layers issignificantly larger than that of the shale, the pore pressure at the extended fracture tip decreases and fracturepropagation becomes slower. After shut-in, the pore pressure decreases more and the fracture does notpropagate any more. For bedding planes oriented perpendicular to the maximum principal stress direction andwith the same elastic properties as the shale formation, results were found not to be influenced by theirpresence. In such a scenario, the impact of multiple bedding planes on fracture propagation is negligible. On theother hand, a bedding plane softer than the surrounding shale formation leads to a fracture propagationasymmetrical vertically with respect to the centre of the injection fracture with a more limited upward fracturepropagation. A pre-existing fault leads to a decrease in fracture propagation because of fault reactivation withshear failure. This results in a smaller increase in injection fracture permeability and a slight higher injectionpressure than that observed without the fault. Overall, results of a sensitivity analysis show that fracturepropagation is influenced by the stress ratio SR, the permeability kc of the confining formations and the initialpermeability kp of the bedding plane more than the other major parameters.

Keywords
Shale-gas, Hydraulic fracturing stimulation, Fracture propagation, Elastic-brittle model, Bedding plane, Fault reactivation
National Category
Geosciences, Multidisciplinary
Identifiers
urn:nbn:se:uu:diva-321462 (URN)10.1016/j.petrol.2017.03.011 (DOI)000401042000030 ()
Funder
EU, Horizon 2020, 640979The Geological Survey of Sweden (SGU), 1724
Available from: 2017-05-05 Created: 2017-05-05 Last updated: 2017-06-16Bibliographically approved
Joodaki, S., Yang, Z. & Niemi, A. (2017). The effect of designing parameter of WAG injection on enhancement of CO2trapping in heterogeneous formations: A numerical study. Greenhouse Gases: Science and Technology, 7(6), 1008-1019
Open this publication in new window or tab >>The effect of designing parameter of WAG injection on enhancement of CO2trapping in heterogeneous formations: A numerical study
2017 (English)In: Greenhouse Gases: Science and Technology, E-ISSN 2152-3878, Vol. 7, no 6, p. 1008-1019Article in journal (Refereed) Published
Abstract [en]

Dissolution and residual trapping of CO2 injected in saline aquifers can be influenced by injection strategies applied. In this study, we focus on the water-alternating-gas (WAG) strategy and investigate the importance of parameters needed to design an effective WAG injection sequence, including (i) CO2 and water injection rates, (ii) WAG ratio, and (iii) number of cycles. Using TOUGH2-ECO2N, we perform 3D numerical simulations of sequences of CO2 and water injection into a heterogeneous formation. Hysteresis in relative permeability and capillary pressure functions is considered based on the Land trapping model. Results show that to design a WAG injection in a high permeable formation, the WAG ratio and number of injection cycles are more important parameters than the CO2 and water injection rates. Increasing the total amount of water injection (i.e., decreasing the WAG ratio for given total amount of injected CO2) improves the CO2 dissolution and residual trapping. It is also shown that increasing the number of injection cycles has a negative effect on both residual and dissolution trapping as measured at the end of the injection sequence, because both the free-phase and the dissolved CO2 plumes in the one-cycle injection scenario reach farther distances and occupy larger reservoir volumes than in the multi-cycle injection. This result means that while water injection following the CO2 injection improves trapping in comparison with the CO2-only injection strategy, the WAG scheme with multiple cycles should not be chosen to enhance trapping for the scenario considered in this study. 

Keywords
Dissolution trapping, WAG injection, CO2 geological storage
National Category
Other Environmental Engineering
Identifiers
urn:nbn:se:uu:diva-329592 (URN)10.1002/ghg.1719 (DOI)000417761600005 ()
Funder
EU, FP7, Seventh Framework Programme, 309067
Available from: 2017-09-19 Created: 2017-09-19 Last updated: 2018-03-07Bibliographically approved
Figueiredo, B., Tsang, C.-F., Rutqvist, J. & Niemi, A. (2017). The effects of nearby fractures on hydraulically induced fracture propagation and permeability changes. Engineering Geology, 228, 197-213
Open this publication in new window or tab >>The effects of nearby fractures on hydraulically induced fracture propagation and permeability changes
2017 (English)In: Engineering Geology, ISSN 0013-7952, E-ISSN 1872-6917, Vol. 228, p. 197-213Article in journal (Refereed) Published
Abstract [en]

Fracture propagation caused by hydraulic fracturing operations can be significantly influenced by adjacent fractures. This paper presents a detailed coupled hydro-mechanical analysis to study the effects of nearby natural fractures on hydraulically induced fracture propagation and changes in fracture permeability. Two rock domains were considered in comparison: FD1, with one fracture, and FD2, with two adjacent parallel or non-parallel fractures. It was assumed that water injection occurred in a borehole that intersected the single fracture in FD1 and one of the two fractures in FD2. Simulations were made for a time period of 3 h with an injection period of 2 h followed by 1 h of shut-in. An elastic-brittle model based on the degradation of material properties was implemented in a 2D finite-difference scheme and used for elements of the intact rock subjected to tension and shear failure. The intact rock was considered to have a low but non-negligible permeability. A verification study against analytical solutions showed that the fracture propagation and stress concentrations due to differential boundary stresses could be accurately represented by our model. Next, a base case was considered, in which the stress ratio (SR) between the magnitudes of the maximum and minimum boundary stresses, the permeability k(R) of the intact rock and the initial permeability k(TF) of the tension failure regions were fixed. In FD2, the distance d(F) between the two natural fractures defined by the closest distance was also fixed. The results showed that in both rock domains, the fracture started to propagate when the pore pressure was approximately 85% of the magnitude of the minimum boundary stress. The propagation of a single fracture was significantly greater than the propagation of a double fracture system because, in the latter case, the pore pressure decreased when the two fractures connected. As a result, changes in permeability in FD2 were smaller than in FD1. At the end of injection, the maximum ratios between the final and initial permeability of the natural fractures were found to be approximately 3 and 2 for rock domains FD1 and FD2, respectively. For non-parallel fractures, the controlling factor for fracture propagation was the separation between the tips of the pressurised fracture and the neighbouring non-pressurised fracture. A sensitivity study was conducted to study the influence of the key parameters d(F), SR, k(R) and k(TF) on the simulation results. Fracture propagation showed more sensitivity to d(F) and SR than to the other parameters.

Place, publisher, year, edition, pages
ELSEVIER SCIENCE BV, 2017
Keywords
Hydraulic fracturing stimulation, Coupled hydro-mechanical effects, Fracture propagation and connectivity, Permeability changes, Elastic-brittle model
National Category
Earth and Related Environmental Sciences
Identifiers
urn:nbn:se:uu:diva-340135 (URN)10.1016/j.enggeo.2017.08.011 (DOI)000413282000017 ()
Funder
EU, European Research Council, 640979
Available from: 2018-01-29 Created: 2018-01-29 Last updated: 2018-01-29Bibliographically approved
Organisations

Search in DiVA

Show all publications