- Home
- Conferences
- Conference Proceedings
- Conferences
76th EAGE Conference and Exhibition 2014
- Conference date: June 16-19, 2014
- Location: Amsterdam, Netherlands
- Published: 16 June 2014
201 - 250 of 1028 results
-
-
Recognizing Geothermal Resources as Unconventionals by the New Geothermal Play Type Catalog
By I. MoeckSummaryOne of the key elements in assessment, characterisation and development of resources is the play type which has an almost mythical status among the oil and gas community – the successful play is thing of which legends are made, and playmakers are regarded as heroes of exploration and industry. While play concepts are clearly defined for hydrocarbon resources considering the source rock and petroleum charge system, the reservoir-seal formation and the trap type, worldwide applicable play concepts for deep geothermal resources are still lacking although industrial energy production from geothermal resources is as old as petroleum production. To address this deficit, a new geothermal play type catalog is developed based on geologic criteria. This geothermal play concept follows the basic approach of general resources assessment to group exploration targets according geologically controlled tiers, connected by similar geodynamic, sedimentary, magmatic and/or tectonic evolution. Worldwide case studies are demonstrated and exemplify the application of this new geothermal play concept. With this new catalog of geothermal plays it will be ultimately possible to transfer lessons learned not only among specific catalog types but also technology from geothermal plays to unconventional hydrocarbon plays and vice versa.
-
-
-
Some Geological and Geophysical Aspects of Geothermal Exploration in the Netherlands
Authors A.A. Van De Weerd, L. Van Leeuwen and A.G. VondrakSummaryThe first geothermal doublet was realized in 2007 in the West Netherlands Basin in Cretaceous sandstones at a depth of approximately 1650m. Water at approximately 600C is pumped up through the production well. Heat is extracted and the cooled water is pumped back into the same formation using a second well. The distance between producer and injector in the aquifer is approximately 1200–1500m. The extracted thermal energy is used for greenhouse heating. This success was followed by eight doublets and plans for more. The driving force is demand for heat by greenhouse complexes combined with availability of subsurface data (well logs and seismic) from oil and gas exploration. Geothermal exploration is very similar to exploration for hydrocarbons. The aim of geothermal exploration is to locate permeable and thick sandstone formations that can sustain production and injection rates in the order of 150–250m3/hour. Using simple reservoir models, the cool temperature-front of robust prospects is expected to reach the producer after 25–30 years. The geology of the areas and basins with realized and planned geothermal reservoirs and the experiences with realized geothermal wells will be discussed.
-
-
-
Geological History of the Southern Margin of the West Netherlands Basin - The Geothermal Energy Perspective
Authors Y.W.L. A’Campo, J.H.P. De Bresser and A.A. Van De WeerdSummaryThe Dutch Westland area, situated in the onshore part of the West Netherlands Basin, is of interest for geothermal energy exploration because good geothermal reservoirs are present in the Late Jurassic to Early Cretaceous deposits, and a high demand for hot water exists from the numerous greenhouses in the area. This study aimed to provide a better regional understanding of the Late Jurassic to Early Cretaceous tectonic history of the West Netherlands Basin, in particular regarding reservoir distribution and the possible influence of early tectonic structures on the deposition of reservoir sands. In addition, detailed knowledge of the fault pattern is important for the placement of geothermal wells. Detailed seismic mapping combined with stratigraphic cross-sections enabled a study of the distribution and development of geothermal reservoirs within the Alblasserdam, Delft Sandstone and the Berkel Sandstone Members. By flattening seismic cross-sections at several mapped horizons, the pre-inversion geometry of the basin was reconstructed. In addition, seismic attribute analyses have significantly improved the mapping of the faults and may illustrate depositional patterns within the Berkel Sandstone Member. The revealed tectonic history and reservoir distribution aid to accurately plan and execute geothermal wells.
-
-
-
The Permian Rotliegend Reservoir Architecture of the Dutch Koekoekspolder Geothermal Doublet
Authors H.F. Mijnlieff, M.R. Bloemsma, M.E. Donselaar, S. Henares, A.E. Redjosentono, J.G. Veldkamp and G.J. WeltjeSummaryThe Dutch Koekoekspolder geothermal doublet was drilled in 2011 targeting the Permian Rotliegend reservoir. The encountered reservoir properties were less favorable than expected pre-drill. Post-drill integrated evaluation of vintage data and the new data from the geothermal wells resulted in a refined depositional model of the area. The Rotliegend sediments comprise predominantly horizontally, bimodally laminated fine to coarse grained red sandstones. These sediments were deposited on an aeolian sandflat. On this sandflat in the relative low-lying areas, coinciding with palaeo-lows of the pre-Permian palaeotopgraphy gypscretes were formed. These gypscretes were periodically exhumed and (partly) eroded providing gypsum grains which were subsequently incorporated in the predominant quartz rich sand sediments on the sandflat forming the typical bimodally laminated Aeolian deposits. The gypscretes are now present as pervasively anhydrite cemented horizons haphazardly vertically distributed.
-
-
-
Summary of the EU Project GEISER on Induced Seismicity in Geothermal Engineering
Authors D.F. Bruhn, K. Águstsson, A. Zang, X. Rachez, S. Wiemer, J.D. van Wees, P. Calcagno and E. HuengesSummaryGEISER was a European project on understanding and mitigation of induced seismicty in geothermal operations. The project involved several European research institutions as well as industry and was funded by the European Commission within FP7.
GEISER addressed a better understanding of the key parameters that control induced seismicity in response to an injection. Data from several events of induced seismicity were collected and analysed. Mechanical models were developed to understand the processes leading to induced seismicity and were combined with probabilistic seismic hazard assessment approaches to propose a new, physics based probabilistic forewarning system. This system requires the determination of a maximum acceptable seismic magnitude and its accepted probability of occurrence. The reliability of the dynamic model is based on the availability of rock physics and seismic data, with models updated from real-time monitoring. In addition to this new approach, a number of recommendations and guidelines for licensing authorities, developers and operators of geothermal projects are proposed.
-
-
-
Minimizing Induced Seismicity in Synthetic Geothermal Reservoirs by Fracture Control in Hydraulic Stimulation
Authors A. Zang, J.S. Yoon, O. Stephansson, O. Heidbach and G. ZimmermannSummaryHydraulic stimulation of rock mass at depth is an essential component to enhance permeability in energy technologies like hydrocarbon, shale gas and geothermal reservoirs. An unwanted by-product of stimulation, however, is the generation of induced seismic events with larger magnitudes. We use a discrete element model of naturally fractured rock with pore fluid flow algorithm in order to analyze two scenarios of high-pressure fluid injection at depth and associated induced seismicity. Our result for a crystalline geothermal reservoir stimulated in a three-stage fracturing process suggests that cyclic treatment is a safer alternative to conventional hydraulic fracturing as both the total number of induced events and their magnitudes are lowered. The ratio of pumped-in hydraulic energy to released seismic energy is in agreement with field data from shale gas and enhanced geothermal systems. This work is motivated by laboratory confined indenter tests on granite cores where a narrow damage zone is found in cyclic fracture control while a wider fracture damage pattern is observed in conventional, displacement-controlled experiments.
-
-
-
ThermoGIS - Unlocking the Geothermal Potential of The Netherlands
Authors M. Pluymaekers, J.D. van Wees and H. VeldkampSummaryA resource and performance assessment methodology has been developed to designate prospective high permeable clastic aquifers and to assess the amount of potential geothermal energy in the Netherlands. It builds from the wealth of deep subsurface data from oil and gas exploration and production which is publicly and digitally available. The performance assessment resulted in e.g. potential maps giving an overview of the expected doublet power which can be retrieved on an economic basis and a potential indicator map ( figure 3 ), which provides insight in subsurface suitability for specific applications. These maps can be interactively accessed in the web-based portal ThermoGIS (www.thermogis.nl). This application complements existing subsurface information systems available in the Netherlands and supports the geothermal community in assessing the feasibility of a geothermal system at any location and provides a non-expert insight in the geothermal energy potential on a national and regional scale.
-
-
-
Optimization of Geothermal-Well-Doublet Placement
Authors R.C.A. Smit, H. Salimi and K.H.A.A. WolfSummaryCurrently, geothermal-energy production is marginally economical. Through optimization of the well positions, the profitability of a project can be improved.
This paper studies optimization of the well positions such that the net present value of a project is maximized in a 2D geothermal reservoir for the selected heterogeneity structures. For this purpose, an automated gradient-based optimization method is used. The reservoir simulations are performed using the Finite-Element Method in the program COMSOL Multiphysics. The major features of the simulation results are discussed in detail.
It is shown that the effect of heterogeneities on the thermal-retardation factor is small. Furthermore, from an economical standpoint, it makes little sense to assume a doublet lifetime of more than 30 years. A higher heat-recovery factor does not necessarily mean that a doublet is more profitable. There exists an optimum well spacing for doublets such that additional gain of later breakthrough (when placing the production well further away from the injector) is negated by the loss in pressure support of the injection well.
Heat production from aquifers can be optimized through the usage of different doublet layouts. It is found that a checkers-board well arrangement is more effective than a tram-rail well arrangement.
-
-
-
Overpressure Prediction in Shales
Authors T. Han, M. Pervukhina and M.B. ClennellSummaryAccurate overpressure detection and prediction is essential to the scientific understanding of the history and safe engineering exploitation of a sedimentary basin. We predict the velocity of shales using the Clay-Plus-Silt (CPS) model from logging measured parameters, the modeled velocity is then used as normal velocity trend to detect and predict overpressure using Eaton equation. The results show that the CPS model predicts accurate velocity in normal-pressure shales, and the modeled shale velocity can be used to detect and predict overpressure more accurately than the established normal velocity trend in overpressured wells.
-
-
-
Cretaceous Mudrock Compaction Trends Across Haltenbanken, Offshore mid-Norway
Authors A.M.P. Cicchino, C. Sargent, N.R. Goulty and A.M. RamdhanSummaryThe Cretaceous mudrocks at Haltenbanken are buried 2–4 km below seafloor. Density porosity ranges over a factor of two at the same depths in different wells, with higher porosities in the west. Although the Cretaceous mudrocks are overpressured, pore pressure is laterally invariant across Haltenbanken; hence porosity differences cannot be due to differences in effective stress. Differences remain after plotting porosity against temperature. Porosity differences are somewhat reduced by adjusting depths for recent exhumation and burial. We suggest the following sequence of events: the Cretaceous mudrocks were normally compacted and hydrostatically pressured in the early Pliocene, because Tertiary sedimentation rates were low; Pliocene exhumation and deposition of Naust Formation glacial sediments occurred rapidly, without porosity loss because low-permeability overburden inhibited pore water expulsion; enhanced temperatures accelerated clay diagenesis, and lithification occurred with a laterally invariant pore-pressure profile because horizontal permeability is much higher than vertical permeability. If this interpretation is correct, the Cretaceous mudrocks at Haltenbanken exhibit chemical undercompaction, where pore water expulsion was inhibited during clay diagenesis and lithification. This finding illustrates the risk in predicting pore pressure in exploration wells using data from offset wells, and suggests a method for estimating Pliocene exhumation history on the Norwegian margin.
-
-
-
A Comparison of Seismicity-Based Permeability Characterization Methods
Authors M. Hosseini, C.W. Neuhaus, A. Rahimi Zeynal, M. Mueller and M.P. ThorntonSummarySeismicity-based permeability characterization uses the distribution of microseismic events associated with subsurface fluid injection (e.g. hydraulic fracturing) to estimate reservoir bulk permeability at the reservoir scale. While existing methodologies mostly originate from the linear pore pressure diffusion solution for a low frequency range of the slow wave propagation equation, the novel methodology presented in this paper is based on coupling the radial pressure diffusion equation with a Mohr-Coulomb shear failure criterion taking into account geomechanical data. Evaluating the spatio-temporal dynamics of induced microseismicity therefore allows us to calculate the permeability as a function of the pressure differential that is needed for rock failure and the subsequent occurrence of microseismicity.
-
-
-
Efficient Prediction of Permeability Based on 3d Images of Core-samples
Authors S. Linden, H. Hagen and A. WiegmannSummaryWe introduce a novel approach to compute the permeability on very large CT images of core samples. The LIR-tree is used for spatial partitioning of the geometry. The images are coarsened in areas where the velocity and pressure do not vary much while keeping the original resolution near the solid surfaces. In addition, solid regions do not occupy computational memory. Variables are arranged in a way such that each cell is able to satisfy the Stokes-equations independently from its neighbor cells. Pressure and velocity are discretized on staggered grids but instead of using one velocity variable on the cell faces we introduce two velocity variables. They discretize the two one-sided limits at the center of the cell surface. The discretization of the momentum and mass conservation equations yields one small linear system per cell. This structure allows to use the block Gauß-Seidel-algorithm as iterative solver.
We compare our method to three other solvers on a large complex Berea sandstone dataset. The method is 3–6 times faster, scales well for up to 32 processors and has very low memory requirements. An additional benefit of this work is that the range of permeabilities from the benchmark can be narrowed down significantly.
-
-
-
A New Fault-Enhancement Attribute Based on First Order Directional Derivatives of Complex Trace
Authors P.C.P. Pampanelli, G.M. Faustino, P.M.C. Silva, A. Kolisnyk and M. GattassSummaryIn this work we present a new fault-enhancement attribute based on image processing techniques for edge detection. The proposed method is simple to implement and has a strong mathematical background. Since faults can be viewed as discontinuities along horizons, in order to enhance such discontinuities we apply a first-order directional derivative. Here, we use the vector perpendicular to the instantaneous phase gradient as orientation vector. It is important to mention that this orientation vector is calculated analytically. In addition, we use the directional derivative of the imaginary components of the complex trace to improve the continuity of our fault enhancement attribute. In order to assess the proposed method, we use the volume of the Netherlands offshore F3 block downloaded from the Opendtect website and compare the obtained results with variance attribute. We conclude that our method is simple, sufficiently accurate, and detects faults without enhancing acquisition footprint. Furthermore, the results show that the instantaneous phase gradient provides consistent dip estimation.
-
-
-
Structural Oriented Signal Decomposition Using Adaptive Frequency Distributions
Authors A.A. Aqrawi and A.A. AqrawiSummarySignal decomposition is often used for delineating turbidite plays, as well as channels and deltas. Post-stack seismic is comprised of a range of frequencies stacked together to create a coherent and concise image. Using signal decomposition, one can decompose the seismic data, singling out desired frequencies, and dulling out the rest. This makes capturing changes of phase, and hence channels, easier. There have been two major features lacking in these attributes, namely an even distribution of frequency occurrences, and loss of information along dipping stratigraphic structures. We propose a new technique that uses an adaptive algorithm, in addition to a local three dimensional trace extraction that filter perpendicular to the stratigraphy. The adaptive algorithm caters to the very nature of seismic segmenting the frequency distribution accordingly. This preserves an even distribution of the sum of frequency occurrences, and hence reduces over-and-under sampling considerably. This results in an improved delineation of the both minor and major channels, a higher imaging resolution and contrast, and a reduction in geological noise. The fact that the trace is filtered perpendicular to the local stratigraphy, this further enhances the resolution, and gives a more accurate result.
-
-
-
Application of GLCM-based Seismic Attributes for Anisotropy Detection
Authors C. Eichkitz, J. Amtmann and M.G. SchreilechnerSummaryThe grey level co-occurrence matrix is a measure of the texture of an image. It describes how often different combinations of pixel brightness values occur in an image. Based on this, several textural attributes can be calculated. We apply a workflow for full 3D GLCM calculation to synthetic data and a real data example from the Vienna Basin. The aim of this work is to test the GLCM attributes on their applicability for anisotropy detection. For this purpose we calculate the GLCM attributes in single space directions and compare the results of these calculations to each other.
-
-
-
Seismic Facies Classification Using Visualization-induced Self-organizing Mapping
More LessSummaryIn multi-seismic-attribute analysis area, facies classification provides an effective way for lithology discrimination and reservoir characterization. SOM method has been widely used in seismic attributes analysis. But the application of SOM method is usually performed in an empirical manner, such as clusters’ number selection and classification procedure, which introduce artificial bias and unstable result. In this paper, we introduce a visualization-induced SOM method. With distance preservation on mapping grid, seismic multidimensional data is able to be visualized on 2-D plane in a very natural and accurate way, and automatic and stable facies classification could also be implemented. We use a simple classification example to illustrate the advantage and verify its effectiveness on a field data.
-
-
-
Automatic Geobody Detection Using Multi-class Sparse Representation
Authors G. Larrazabal, C. Ramirez and G. GonzalezSummarySalt segmentation and characterization can be viewed as a classification problem where two groups or classes (SALT and SEDIMENT) are to be assigned to each seismic data element at a given coordinate. In a previous work we propose to construct a classification machinery based on the theory of sparse representation in order to carry out salt characterization in an automatic fashion. Nonetheless, the complexity of the seismic data as well as the limitations of the migration algorithms makes this dual distinction a difficult challenge. In order to overcome this obstacle, we propose to extend our original work from the characterization of two classes to a multiclass segmentation framework. In this manner, we allow the sparse representation method to extend the number of possible outcomes when classifying a given test sample thus reducing the number of misclassifications.
-
-
-
Detailed Geological Site Characterisation of Jemgum Salt Cavern Gas Storage Facility (Lower Saxony Basin, Germany)
Authors B. Otto and C. HanitzschSummaryA detailed geological site characterisation has been conducted at the new Jemgum Salt Cavern Gas Storage Facility in Northern Germany. Specific geomechanical requirements as well as complex Zechstein internal stratification had to be taken into account. 3D seismic processing including data regularisation and anisotropic controlled beam prestack depth migration provided an accurate image in depth. In order to accelerate modeling of the complex salt structure consisting of two salt canopies a new seismic interpretation workflow based on fault segments instead of horizons has been applied. This approach overcomes severe multi-z-value limitations inherent to grid based seismic interpretation software. Key advantages are an improved visibility, an increased picking efficiency and a higher degree of freedom while shaping the salt model. At all stages of modeling appropriate seismic attributes have been successfully incorporated. The 3D structural model provided a key element for all subsequent subsurface planning as well as drilling operations. Finally a blind well at the steep flank of the salt dome indicates a small lateral uncertainty of the salt edge.
-
-
-
Geologically Validated Stratigraphic Earth Models from Seismic Dat
Authors A. Laake, H. Olsen, B. Frizzell and I. SchmidtSummaryLarge-scale glacial deposits present challenges to seismic earth model building because of their rapidly changing velocities in incised tunnel valleys with glacial- and water-laid fill. We present an approach for detailed geological interpretation of seismic horizons as boundaries for stratigraphic and lithological units using the structurally sharpened continuous colour (SRGB) technique. In the first step, colour-textured time slices reveal geological features contained in the seismic data cube. In the second step, colour-textured horizons are correlated with stratigraphic and depositional information for geological interpretation. The framework of interpreted horizons and seismically derived facies can then be used as a framework for velocity earth modelling. This approach can be used with any type of 3D geophysical data, in both time and depth domains.
-
-
-
Mapping of Sand Injectites from Colour-processed Multimeasurement Seismic Data
By A. LaakeSummaryIn a PSDM data set from the Central North Sea we found sand injectite structures of up to 1 km in diameter and 300 m in height. Their shape and thickness varies with the location within a submarine delta. The dewatering of sediments as well as the pressure from overlaying sediments deposited by the delta determines the shape and thickness of the injectites. Sand injectites can be delineated and extracted from seismic prestack depth migration (PSDM) data using the structural attribute from the structurally sharpened continuous-colour red green blue (SRGB) process. Multimeasurement towed streamer acquisition and isometrical sampling enabled the detection of complex sand injectites. Combining seismic sections and horizons textured with seismic attributes in a chair diagram rendering provides a useful approach for geological interpretation in 3D. Texturing seismic horizons with SRGB colours enables the interpreter recognizing morphological patterns that can reveal clues for the interpretation of depositional and erosional environments. Correct mapping of sand injectites can be used for mapping drilling hazards where the sands are still not consolidated. Consolidated sand injectites may be important reservoirs as well as pathways for fluid migration across seals in sandstone reservoirs with frequent shale interbeds.
-
-
-
Velocity Estimates of Injected Carbon Dioxide in 4D Seismic Monitoring by Non-linear Elastic Wavefield Inversion
By A. SakaiSummaryGeologic sequestration by carbon dioxide injection into saline aquifers and others is one of the promising options to mitigate global climate changes. Elastic parameters are fundamental in estimating rock properties, fluid saturation and pore pressure in carbon dioxide injection and 4D seismic survey efficiently provides spatial and temporal changes of elastic parameters as well as baseline parameters.
4D seismic survey data, collaboratively acquired by Japex in 2003 and 2005 on the area of onshore RITE/METI carbon dioxide injection test site in Japan, were analysed incorporated with repeated time-lapse logging data matched in scale by the current method. In this study, 1) newly built the most probable full-scale baseline and monitor elastic wave velocity model of the 4D seismic survey in the depth domain using time scale adjusted and calibrated 4D AI cube in the time domain and 2) examined the feasibility of nonlinear elastic wavefield inversion of synthetic 4D seismic data assuming regular 3D geometry, though the original 3D was covered by irregular geometry, with common smoothest quasi-linear initial macro-velocity model for baseline and monitor surveys. Results of non-linear elastic wavefield inversion (FWI) with frequency-cascade scheme successfully showed estimates of 4D elastic wave velocity changes at excellent quality and accuracy.
-
-
-
Preparatory Work for the Seismic Monitoring of CO2 Storage at a Prospective Site in the St. Lawrence Lowlands, Canada
Authors L. Perozzi, B. Giroux, R. Kofman and D. SchmittSummaryThe success of the geological storage of CO2 depends on the capability to monitor movements of the injected fluid into the subsurface. For understanding the effects of CO2 as a pore fluid on the overall rock seismic response, a series of ultrasonic measurements on two samples of saline aquifer sandstones of the Potsdam Group have been made showing clear P-wave variations with CO2 varying phase state.
Significant amplitude variation was only observed in the Covey Hill sample. The laboratory measurements and well log data were used to calibrate a numerical model that was used to perform poro-viscoelastic forward modeling of time-lapse walkaway VSP surveys. The results show that supercritical CO2 injected in the reservoir, results in a clear seismic signature. Finally, the modeling results also indicate that the possibility to rely on AVO analysis to monitor the CO2 plume is compromised by a refracted wave that appear early in the wave field.
-
-
-
Experimental Evidence of Calcite Dissolution and Induced Precipitation during supercritical CO2 Residence
Authors L. Pimienta, L. Esteban, J. Sarout, K. Liu, C. Delle Piane and B. ClennellSummaryPrior to injecting CO2 in water-saturated carbonate reservoirs, one needs to investigate the effect of the residence of supercritical CO2 (SCCO2) on the rock integrity and overall physical properties.
In this study, a Savonnieres limestone is characterised in terms of its physical properties, pore chemistry and textural features prior and after 2 or 4 hours SCCO2 residence under in situ stress/temperature conditions. More precisely, elastic waves (Vp and Vs) at ultrasonic frequencies, electrical resistivity (Rt), helium porosity-permeability and pore chemistry are measured before and after SCCO2 aging. In addition, X-ray CT monitoring is carried out during the different steps.
While water chemistry highlights an enhanced calcite dissolution related to the duration of SCCO2 residence, a change in the physical properties is observed between the two residence steps. It is shown from the physical properties that (i) the rock building minerals were dissolved after 2 hours; and (ii) the rock overall integrity increases after 4 hours, highlighting a possible re-precipitation phenomenon.
-
-
-
Hydraulic Forces in Geological Storage of CO2
By K.U. WeyerSummaryTraditionally the ‘storage’ of CO2 in the deep subsurface has been seen as (1) a hydrostatic process whereby migration of sequestered CO2 is driven by upwards and downwards directed vertical buoyancy forces, and (2) that the buoyancy driven movement is upwardly constrained by aquitards (caprocks). Thus, a seemingly perfect storage concept was created for both storage in deep aquifers and under the caprocks of depleted hydrocarbon reservoirs which has subsequently been adopted by industrial, governmental, and academic agencies.
Weyer (2013) showed that this concept is based on the purely mathematical treatment of velocity potentials [energy/unit volume] within engineering hydraulics and the supporting assumption that water is incompressible. In fact water is compressible as is supercritical CO2. Bear (1972 , p. 159–161) and de Marsily (1986 , p. 52) established that the force potential treatment of Hubbert (1940) is the proper way to deal with compressible fluids. This was confirmed by Scheidegger (1974 , p. 79) who, having dealt with both approaches, stated unequivocally “It is thus a force potential and not a velocity potential which governs flow through porous media” (emphasis added). This paper deals with hydraulic forces derived from force potentials [energy/unit mass] and indicates the differences between hydrostatic and hydrodynamic conditions.
-
-
-
Interfacial Interactions of CO2-Water-Bentheimer Sandstone System - Dissolution and Contact Angle
Authors N. Shojai Kaveh, E.S.J. Rudolph, W.R. Rossen and K.H.A.A. WolfSummaryCO2 storage in deep saline aquifers and depleted gas reservoirs is largely controlled by interfacial phenomena among fluid phases and rock pore spaces. Reduction of the interfacial tension may result in the mobilization of connate water (capillary trapping). In addition, dissolving CO2 in formation water is one of the most important mechanisms to (permanently) store CO2 (solubility trapping). However, dissolution is generally negatively correlated with salinity, and dissolution of CO2 lowers the IFT of CO2/water systems. Therefore, understanding of the wettability and dissolution behavior is of importance for displacement behavior and storage capacity. In this paper, wetting properties of Bentheimer sandstone rock are investigated with an unsaturated aqueous phase at various pressures. In addition dissolution behavior of CO2 bubbles in aqueous phase are determined. The Experiments show that the change in contact angle is influenced by a number of mechanisms such as dissolution and bubble size variation, rather than by the wetting properties of the surface alone. During the experiments, two dissolution regimes were observed. Analysis of the data showed that this behavior cannot be caused by wettability alterations of the surface from strongly to less water-wet, but by bubble volume reduction due to CO2 dissolution in the aqueous phase.
-
-
-
Reservoir-condition Pore-scale Imaging - Contact Angle, Wettability, Dynamics and Trapping
Authors M.G. Andrew, B. Bijeljic and M.J. BluntSummaryFirstly capillary trapping is examined in a range of five different rock types, including both carbonates and sandstones. Rocks are imaged both after drainage and imbibition, and in all cases between 65–70% of the CO2 in place after drainage was trapped. Trapped cluster size distributions are compared to rock connectivity as determined using pore network modelling. Better connected pore-spaces tend to have more large clusters relative to small clusters, and visa-versa. This is important as small clusters are more difficult to remobilise by viscous and gravitational forces. They also present a relatively larger surface area for reaction and mineralization.
Secondarily wettability is analysed by measuring contact angle manually. In order to do this the contact line was found in 3D and the data set resampled onto planes perpendicular to the contact line at a particular point. Contact angles ranging from 35–55o were found, indicating that the super-critical CO2-brine-carbonate system is weakly water wet. The range in contact angles is interpreted as the result of contact angle hysteresis associated with surface heterogeneity.
Finally the first images of CO2 drainage at reservoir conditions are also presented, imaged at Diamond Light Source, represented an unprecented depth of information about pore-scale flow processes.
-
-
-
Seismic Response of CO2 Saturated Red Wildmoor Sandstone under Varying Temperatures and Pressures
Authors J. Naseryan Moghadam, N. Haque Mondol, H. Hellevang, Ø. Johnsen and P. AagaardSummaryUnderground geological sequestration of CO2 is considered as one of the most promising solutions for overcoming the global warming problem. Time lapse seismic monitoring enables us to monitor the CO2 plume migration and is considered as an integral component of a geological CO2 sequestration project because the seismic behavior of the rock is a function of both mineralogical composition and also the physical properties of the pore fluids. CO2 can be presented as gaseous, liquid and supercritical states at the uppermost kilometer of the sedimentary basin while in CO2 sequestration operation the supercritical and liquid states of CO2 are preferred due to the higher sweep efficiency. In this study, both compressional (Vp) and shear (Vs) wave velocities of a CO2 saturated Red Wildmoor sandstone sample under different temperature and pressure conditions are measured in a uniaxial hydrostatic cell equipped with acoustic transmitting and receiving transducers. The observed velocities illustrate that by introducing CO2 into the dry core, the resultant wave velocities decrease up until the critical condition in which by further increasing the CO2 pressure the Vs remains unchanged while the Vp gradually increases up to the maximum pressured applied in this study. Also it was observed that at lower temperature, the gas to liquid CO2 phase transition occurs with a sudden decrease in the amounts of the observed velocities while at higher temperature the velocity reduction is smoother possibly due to reaching the supercritical state. The observed velocities are in good agreement with Gassmann predicted velocities as well as literature data.
-
-
-
Numerical Simulation of Convective Mixing During Carbon Dioxide Sequestration in Saline Aquifers
Authors M. Pasdar, B. Rostami and S.M. Seyyedi Nasooh AbadSummaryThe growing concern about global warming has increased interests in geological storage of CO2. Investigating the risk of leakage of CO2 from injection sites (saline aquifers) is of great importance, because free CO2 has always a chance to leak out through pathways such as faults, fractures etc. Predicting the final fate of injected CO2 requires an understanding of the basics mechanisms involved in storage process. In this study, convective mixing, one of the main mechanisms during CO2 sequestration has been investigated numerically. Scaling analysis of convection parameters enables us to evaluate suitable injection sites without the need for costly simulations.
-
-
-
Comparison of Objective Functionals in Seismic Full Waveform Inversion
Authors C.E. Jimenez Tejero, D. Dagnino, V. Sallares and C. RaneroSummaryThe FWI method is a powerful tool that allows one to obtain high-resolution information from the subsurface. However, the method is highly non-linear as in the convergence to the solution it might get trapped in local-minima. Among other techniques, it becomes crucial a suitable choice of the objective function. We have selected five objective functions to perform a comparative study under a common 2D-acoustic FWI scheme: the L2-nom, cross-correlation travel time (CCTT), non-integration-method (NIM), envelope and phase objective functions. We test with a 2D-canonical model the susceptibility of the functions to the initial model perturbations. To complete de study with a more realistic synthetic example we test the functions with the Marmousi model. The L2-norm and phase objective functions give the highest resolution images and the CCTT, NIM and envelope objective functions lead to smooth models. However in realistic initial conditions, L2 and phase misfits fail in recovering the velocity model in contrast to the CCTT, NIM and envelope functions that maintain a more consistent behavior.
-
-
-
Robust Full Waveform Inversion with Normalized Plane-wave Data
More LessSummaryIn order to reduce the computational burden of Full waveform inversion (FWI), we have developed FWI algorithm using plane-wave encoding. Generally, the number of plane-wave gathers is much fewer than that of common shot gathers. Therefore, we can minimize the computational cost to handle massive 3D data sets. In addition, by adopting normalized wavefield excluding source effect, the source wavelet estimation does not need in our inversion process. We validated our algorithm through numerical tests.
-
-
-
Spectral Implementation of Full Waveform Inversion Based on Reflections
Authors Z. Wu and T. AlkhalifahSummaryUsing the reflection imaging process as a source to model reflections for full waveform inversion (FWI), referred to as reflection FWI (RFWI), allows us to update the background component of the model, and avoid using the relatively costly migration velocity analysis (MVA), which usually relies on extended images. However, RFWI requires a good image to represent the current reflectivity, as well as, some effort to obtain good smooth gradients. We develop a spectral implementation of RFWI where the wavefield extrapolations and gradient evaluation are performed in the wavenumber domain, obtaining clean dispersion free and fast extrapolations. The gradient, in this case, yields three terms, two of which provide us with each side of the rabbit ear kernel, and the third, often ignored, provides a normalization of the reflectivity within the kernel, which can be used to obtain a reflectivity free background update. Since the image is imperfect (it is an adjoint, not an inverse), an optimization process for the third term scaling is implemented to achieve the smoothest gradient update. A rare application of RFWI on the reflectivity infested Marmousi model shows some of the potential of the approach.
-
-
-
Hybrid Ray-born and Finite-difference Full Waveform Inversion
Authors H.C. Tengesdal, A.N. Minakov and H. KeersSummaryIn this contribution we present a hybrid method for full seismic waveform inversion incorporating the ray-Born and finite-difference methods. We also compute synthetic seismograms using both Methods for several random seismic velocity models and perform a comparison of the pressure waveforms. The main motivation in this contribution is to show that the ray-Born modeling method, which is less computationally expensive, is accurate enough to provide an efficient alternative to purely numerical methods.
-
-
-
Influence of Inaccurate Wavelet Amplitude on Frequency-domain Full Waveform Inversion
More LessSummaryIn this paper, we study quantitatively the influence of seismic wavelet amplitude error on different scale anomaly inversion in frequency-domain full waveform inversion (FWI). We design 40 anomaly models with different size and 8 perturbation seismic wavelets, and evaluate quantitatively reconstructed model error by calculating the optimal similarity coefficient and model fitting error. Tests show that when perturbation of seismic wavelet amplitude is larger than 30%, FWI obtains completely incorrect and unacceptable inverted results even if initial model is just the true model. In addition, the utilization of transmission data mitigates influence of inaccurate seismic wavelet amplitude on FWI partly. Finally, the effects of wavelet amplitude error on the shape and value of reconstructed models are different, complicated, in particular when the scale of anomaly is lower than half of wavelength. Therefore, inaccuracy of seismic wavelet amplitude is a key factor in frequency-domain FWI and should be considered in practice.
-
-
-
Efficient Regularization Technique for Full Waveform Inversion Using Selective Smoothing Coefficients
More LessSummaryA smoothing filter has a good effect on regularizing functions harmonizing adjacent values. For this reason, we use regularization when we perform full waveform inversion. However, regularization with the same smoothing coefficients blurs boundaries between layers and important structures so the result becomes unclear. In order to compensate for the blurring problem, we designed a selective smoothing filter that adopts different smoothing coefficients as the inversion process runs. At first, we establish effective maximum wavenumbers regarding to all smoothing coefficients. Second, we determine wavenumber coverage effectively with a few of frequencies. Then, the coverage and frequencies decide varying smoothing coefficients for the inversion. In order to verify the accuracy of our regularization, we perform full waveform inversion with selected smoothing coefficients on the SEG/EAGE salt model. The inverted result shows for the correct salt velocity and clear structures including sand lens, faults, and layer boundaries. It is observed that our scheme is both simple and powerful for stabilizing and sharpening inverted image.
-
-
-
Comparison of Stochastic Optimization Methods on Two Analytic Objective Functions and on a 1D Elastic FWI
Authors A. Sajeva, M. Aleardi, A. Mazzotti, E. Stucchi and B. GaluzziSummaryWe compare the performance of three different stochastic optimization methods on two analytic objective functions varying the number of parameters, and on a 1D elastic full waveform inversion (FWI) problem. The three methods that we consider are the Adaptive Simulated Annealing (ASA), the Genetic Algorithm (GA), and the Neighbourhood Algorithm (NA) which are frequently used in seismic inversion.
The application of these algorithms on the two analytic functions is aimed at evaluating the rate of convergence for different model space dimensions. The first function consists in a convex surface, and the second one is a multi-minima objective function which also permits to verify the ability of each method to escape from entrapment in local minima. Our study shows that among the three optimization methods GA displays the better scaling with the number of parameters. The ASA method is often the most efficient in case of low dimensional model spaces, whereas NA seems to perform less efficiently than the other two and to be more prone to get trapped in local minima.
Tests of 1D elastic FWI on synthetic data, inverting for density, P and S-wave velocity for a total of 21 unknowns confirm the conclusions drawn from the previous examples.
-
-
-
1D Elastic FWI and Uncertainty Estimation by Means of a Hybrid Genetic Algorithm-Gibbs Sampler Approach
Authors M. Aleardi and A. MazzottiSummaryConversely from gradient-based deterministic approaches, stochastic optimization methods, like Genetic Algorithms (GA), search for the global minimum of the misfit function within a given parameter range not requiring any calculation of gradients of error surfaces. But, more importantly, they collect a series of models and associated likelihoods that can be used to estimate the posterior probability density distribution (PPD). However, it is demonstrated that GA do not honour the principle of the importance sampling. Therefore, a biased estimation of the PPD is produced if it is directly computed from the collected models and their associated likelihoods. On the other hand applying Markov Chain Monte Carlo methods (MCMC), such as Metropolis-Hastings and Gibbs sampler, can provide accurate PPD but at considerable computational cost. In this work we use a hybrid method, which combines the speed of GA, to find an optimal solution, with the accuracy of a subsequent Gibbs sampling (GS), to obtain a reliable estimation of the PPD. We apply this method on a 1D elastic Full Waveform Inversion (FWI) on synthetic and actual data. It turns out that GA, if appropriately implemented, yield PPD estimations that, although with an underestimated variance, can be very close to the correct ones.
-
-
-
Velocity Model Building Based on the Buried Hill Surface and Its Application for the Carbonate Reservoir in Tarim
More LessSummaryThe carbonate reservoirs in Tarim mainly include limestone fracture-cave reservoir and dolomite remnant buried hill reservoirs. The insider of the buried hill has complex structure and abundant fractures, and at the same time igneous rock is widely spread, not only inside the buried hill but also in shallow layers. These factors lead to strong lateral velocity variation. Therefore the accurate velocity model building technology is demanded to solve the imaging of the buried hill and the identification for the igneous rock and the carbonate rock. In this paper, a new velocity model building method based on the buried hill surface has been put forward, which uses different modelling strategy to handle with the overburden and underlying parts of the buried hill surface. Specifically, we use a quantitative description technique for modelling of overburden igneous rock, and grid-based tomography for modelling of underlying formations. When we have applied this method into the real seismic data and use the multi-well constrained PSDM technology at the same time, the key issues, such as the imaging of the buried hill and the identification for the igneous rock as well as the carbonate rock, can be solved very well.
-
-
-
GPU Accelerated Anisotropic PSTM Imaging Success in a Fold and Thrust Belt, Onshore Trinidad
Authors T. Dhanpaul, S. Ragbir, M.Q. Guo, Y.M. He and Y. TianSummaryNowadays PSDM has the unbeatable imaging advantages with the increasing computing capability. However, in some tectonically complex area, comparing to possibility of mislead by sudden velocity variation of PSDM results, practical experiences frequently told PSTM yielding good results. With the reinforcement of GPU technology,anisotropic PSTM can achieve an accurate result and and it can be used as a good guidance for further PSDM study. Trinidad lies on the South American tectonic plate and falls within the Orinoco Fold Belt which is a prolific oil producer. The study area is located within the intensely deformed southern basin of Trinidad.Seismic data recorded is typically very poor. This paper focuses primarily on some of the processing challenges caused by the undulating topography, complex subsurface structures, the presence of high-velocity rocks near at surface within the project area as well as the impact of back-scattered energy and severe migration wavefields and how these challenges were mitigated against thereby improving the overall data quality and its interpretability. The final results showed better resolution,enhanced structural geometries and fault definition. As a result, it is envisaged that when interpretation is completed on this dataset in the very foreseeable future that enormous potential would be unlocked.
-
-
-
Non-tomographic Resolution of Small-scale Velocity Anomalies Offshore Netherlands
Authors S. Greenwood, J.K. Fruehn, J.K. Fruehn, R. Lengkeek, R. Gras, E. Verkuil and D. SekulicSummaryFeatures with a lateral extent much shorter than the recording cable length, because the recorded seismograms to exhibit highly irregular moveout behaviour in preSDM CRP gathers during the initial phases of velocity model update. If these features are only a few hundred metres wide, then it is unlikely that conventional ray-based tomography will be able to invert for them. In addition, if the autopicker being used to supply residual moveout behaviour to the tomographic inversion scheme is using a parametric picker, then the complex moveout behaviour will not be adequately described, and the tomographic inversion may not converge.
Here we show one such example of this problem, for a structure offshore Netherlands, and describe a piecemeal solution using the measured autopicking error distribution to guide a local vertical velocity update. This approach helps to resolve the shallow velocity anomaly, permitting conventional tomographic update for the deeper section to proceed successfully.
Further, this approach will be contrasted with that of a non-parametric (generalised moveout) picking approach wherein the assumption of 2nd or 4th order parametric moveout is relaxed, so as to capture the more complex moveout behaviour in the CRP gathers so as to enable the subsequent tomographic inversion to better resolve small-scale structure.
-
-
-
Reducing Depth Misties by Making Anisotropy Conformal with Velocity and Geology - Example from the NW Australian Shelf
Authors S. Birdus, L. Vincent, R. Djumeno, A. Artemov and L. LiSummaryWe propose and successfully apply on a real 3D seismic dataset from the North-West Australian shelf a new technique that uses well information to correlate anisotropy with velocity for localized lithology driven anomalies. We assume that localized variations in both velocity and anisotropy are caused by changes in the lithology (shale vs carbonate vs sandstone etc). This should result in some correlation between anisotropy anomalies and velocity anomalies. We use well information to establish such a correlation. Our technique produces geology conformal PSDM anisotropic velocity models and reduces depth misties.
-
-
-
A New Tool for Estimation of Gas Hydrate Saturation
Authors D. Bhowmick, D.K. Gupta, A. Malhotra and U. ShankarSummaryThis work introduces a new empirical p-wave velocity-porosity transform that holds well over a wide range of porosity values and takes the effect of clay into consideration. The transform is based on field observations of sonic velocity and porosity logs. This method fits the velocity-porosity relation between Wood’s model and time average equation and the values of associated parameters are derived by fitting it over NGHP-01-05 site of Krishna-Godavari basin, India and Mt. Elbert 01 well of Alaska North Slope, USA. We further prove its applicability by using it to derive the gas hydrate saturation at sites NGHP-01-03 and NGHP-01-07 of KG basin and the Mt. Elbert 01 well from Alaska region. The results obtained using proposed transform are found to be in agreement with the previous results. Thus, the proposed empirical relation holds well on low as well as high porosity sediments.
-
-
-
Numerical Modelling of Water Reinjection in Geothermal Structure
Authors S.G. Kostyanev, S. Kocsardi, Z. Nagy and P. SzolgaySummaryA numerical method is given here to model the processes of filtration and heat transfer in a three-dimensional geothermal medium. The main goal of the analysis was to find and proof the correlation between pressure and temperature in a geothermal structure.
The Cellular Neural/Nonlinear Networks paradigm is a natural framework to describe the behaviour of locally interconnected dynamical systems which have an array structure. Emulated digital implementation of a CNN-Universal machine can use space variant templates (weighted interconnections) and multi-layered structures on different array processing architectures.
We want to find the optimal computational architecture which is satisfying the functional requirements, using the minimal precision and nevertheless achieving maximum computing power. To meet these requirements we want to process the special-regions with the highest possible parallelism.
-
-
-
Detection of Seismic Impedance Interfaces Based on Singularity Analysis
More LessSummaryThe local singularity of signal often carries a great deal of important information. Unlike Fourier transform, the wavelet transform has good local property in time and frequency domains, and it is suitable for analyzing the non-stationary signal. A new wavelet basis named impedance basis is proposed on the basis of the properties of impedance data in the paper, which can indicate discontinuity of impedance data in time. The wavelet transform based on impedance basis can effectively extract the information about impedance interfaces. The analysis for Marmousi impedance model and real seismic impedance data shows that the presented approach is applicable to detection of impedance interfaces. In addition, it is also beneficial to identification and interpretation of thin-bed.
-
-
-
Time-frequency Transform Based NMO Correction with Auto-mute
Authors D.C. Adams and D. MarkusSummaryWe examine the use of a time-frequency transform based normal moveout (NMO) correction and its effects on normal moveout stretch distortion. Our analysis contributes to understanding the nature of normal moveout stretch, particularly in the time-frequency domain. Our normal moveout correction is calculated using either an S-transform or a TT-transform. In the S-transform method we make use of its time shift property. TT-transform allows us to shift local signals using an adaptive block move sum process. In either case normal moveout stretch produces a dimming effect on the traces that increases as the amount of stretch increases. With the S-transform method the dimming mechanism is related to frequency shifts relative to the transform kernel function. When the TT-transform method is used the dimming is observed to be a result of destructive interference between time-shifted localized signals. The result is a self-muted normal moveout corrected gather.
-
-
-
Three Dimensional Generalized Nonhyperbolic Moveout Approximation - Application on a 3D HTI Model
More LessSummaryReflection moveout approximation is widely used for seismic modeling, velocity analysis, time migration etc. However, most of existed approximations are developed for 2D case. We propose a three-dimensional generalized nonhyperbolic moveout approximation (GMA). It is called generalized approximation because it follows the 2D counterpart proposed by Fomel and Stovas. In our approximation, thirteen parameters are determined from the zero-offset reference ray and two non-zero offset reference rays. By testing the accuracy of the proposed approximation in a 3D HTI model, we find that the accuracy of our 3D GMA is very good.
-
-
-
Generalized Group- and Phase-domains Moveout Approximations for Converted-wave in a Horizontal and Homogeneous VTI Layer
More LessSummaryFor a horizontal and homogeneous transversely isotropic reflector with a vertical symmetry axis (VTI), we derive the generalized moveout approximations for P-SV converted-wave in group (t-x) domain and phase (τ-p) domain. They are called generalized approximation due to following the concept proposed by Fomel and Stovas. Both generalized moveout approximations include five parameters, separately; three parameters are determined from the zero-offset (horizontal slowness) ray and other two parameters are obtained from a non-zero offset (horizontal slowness) ray in group (phase) domain. The accuracy of our approximations is tested on a VTI model. Numerical results show that our group-domain approximation has good accuracy even for the large offset; our phase-domain approximation is equivalent to the exact solution. The potential applications of the proposed approximations include seismic modeling, traveltime inversion etc in a horizontal and homogeneous VTI layer.
-
-
-
Aspects of Residual Moveout after Downward Continuation
More LessSummaryThe case is considered that an approximate prestack depth migration with an inaccurate velocity model and a subsequent horizon analysis for residual moveout have been applied to a seismic survey. A Solution is developed in which the parameters required for the inversion are determined in the vicinity of the reflected events and in terms of travel times which have been previously determined from residual moveout analyses. The validity of the suggested algorithm is demonstrated for the inversion of synthetic data exhibiting dipping structures and laterally inhomogeneous velocities. For moderate offsets the approximations predict the residual moveout of migrated reflections and are related to the spreading functions for an exploding reflector in the true velocity model.
For the case of a vertically inhomogeneous medium expressions are obtained for the exact determination of residual moveout after downward continuation. The case is considered that for a reservoir structure the velocity is constant beyond a certain depth and residual moveout is to be preserved: an estimate is provided for the increase in offset as a function of the increase in thickness of the overburden.
-
-
-
Hydraulic Rock Typing Using Rock Friction Factor - Reynolds Number Type Curves
More LessSummaryAlthough many studies have been conducted to investigate the gas-flow behavior within porous rocks using the RFI-ReN type curves in the past, to our knowledge, application of the RFI-ReN type curves for hydraulically typing rock samples has not been previously reported. In this work, we present a new workflow for characterizing hydraulic heterogeneity and identifying rock types. A generalized analytical equation characterizing the relationship of RFI versus ReN is firstly derived by invoking the Forchheimer equation and the concepts of RFI and ReN. Next, four characteristic lengths are proposed to correlate a set of classic real gas-flow data published by Fancher et al. (1933) , Cornell (1952) and Tek (1957) . The most relevant characteristic length is determined through comparing their success in distinguishing the correlated type curves. This characteristic length is then used to construct a characteristic RFI-ReN type curve chart for the given set of rock samples. Rock samples with different hydraulic properties can be classified into groups based on locations of their type curves on the RFI-ReN type curve chart. Furthermore, we present that the spread of RFI-ReN type curves provides a graphic mean for characterizing the degree of heterogeneity of the hydraulic properties of tested rock samples.
-
-
-
Investigation on Log Responses of Bulk Density and Thermal Neutron in Coal Bed with Different Ranks
More LessSummaryIn this paper, two novel methods to correct the effect of ash on logs are firstly introduced. The corrected log responses of density in coal bed are in accordance with the changing rule of density of laboratory. The hydrogen index of coals with different ranks is simulated. And the simulated neutrons are well verified by logs. After obtaining the accurate variation of density and neutron along with coalification, the log response behaviors of density and neutron in coal bed with different ranks are analyzed and discussed in detail. The values of bulk density and neutron porosity for higher rank ash-free coals are also proposed, which make supplement for the petrophysical knowledge base. Finally, a cross plot of density versus neutron logs is proposed to identify coal ranks, and it has been successfully applied to various CBM fields of several basins in China.
-
-
-
Elastic Properties Characterization and Pore System Evaluation Using Kuster-Toksös and Gassmann Models
Authors L.C. Oliveira, I.A. Lima Neto, R.M. Misságia, M.A.R. Ceia, N.L. Archilha, L.T. Castro and F.R. SouzaSummaryCarbonates represent a significant portion of the Brazil’s deepwater oil production, whose importance has increased with recent discoveries in the post-salt and pre-salt oil deposits, justifying studies on the subject, the economic potential they represent. The study of the physical properties of rocks is important for reservoir characterization and monitoring. It correlates the seismic response of rock properties. Reservoirs carbonate rocks might be a complicated system pores, resulting in different compressibility that affect the velocities of seismic waves directly. The goal is to estimate elastic moduli considering interparticle porosity expected in limestones using Effective Elastic Media (EEM) theories as Kuster-Toksöz model (KT). This work consists in interpretive analysis of the elastic properties of ultrasonic data measured by triaxial rock deformation and physics system, installed in integrated reservoir modelling laboratory at LENEP/UENF in Macae, Rio de Janeiro, Brazil.
-