Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
Results
Summary Discussed here are the results of a 1D and 3D forward with 1D and 2D inverse Controlled Source Electromagnetics (CSEM) modeling exercise over a know target offshore West Africa with the intent of demonstrating the technology, develop methodology for analysis and better understanding its limitations. The 3D dataset presented here is considered rich consisting of 11 transmitter (Tx) tow lines and 43 electric and magnetic seafloor receivers (Rx) arranged in a 3D grid pattern (see Figure 1) and acquired with a time-shared dual base frequency square waveform of 0.25 and 0.0625 Hz. Aspects of detectability, comparison to modeled results (1D, 2D and 3D), and resolution of known structure both horizontal and vertical, comparison with logged resistivities in a constrained and unconstrained sense are analyzed. For example โ detection of the target structure seems apparent in the unconstrained 2D inversions of the CSEM data. However, this inversion has reduced ability to image separate, vertically stacked pay and laterally non-extensive bodies in the unconstrained inversions and forward modeled data. Contrast will also be made between the 1D and 2D inversions to assess the limits in imaging the hydrocarbon filled structures, along with 3D forward calculated responses to be compared with the acquired data. Combined and joint inversion of the CSEM and Marine MagnetoTelluric (MMT) information will also be considered in order to better constrain the deeper resistivity section not sampled by well logs and to improve imaging of the 3D complexity of our known target. Introduction As many descriptions of the CSEM acquisition methodology have been presented in the past, the authors will spare the audience these details. For full description of the acquisition methodology, see Elligsrud et al. 2002 or Eidesmo et al. 2002. The survey acquisition design presented many challenges as seafloor Rx and Tx tow-lines were required to respect exclusion zones for all current and future drilling and seabed infrastructure associated with the development of the field. Drop accuracy and positioning of the seafloor Rxโs was required to be proven in a non-sensitive โgreen zoneโ before deployment in the sensitive โred zoneโ (see Figure 1) in proximity to development activities, with ROV recovery of Rx anchors planned in the case of conflict with seafloor equipment. Test were performed to demonstrate that CSEM acquisition would not interfere with drilling operations, in particular the electromagnetic Tx energy interference with the magnetic guidance used for drilling operations and acoustic positioning of the towed Tx and Rx locations not interfering with other seafloor acoustic positioning systems. A dual frequency, time-share Tx waveform of 0.25 and 0.0625 Hz was used to broaden the bandwidth and depth range of investigation with respect to the skin depth signal penetration limitations in a conductive environment. A majority of Rxโs were also equipped with vertical electrical sensors to measure the full vector compliment of the electric field. It may seem inappropriate to model such a complex dataset and target in 1D, however the ready availability of fast forward modeling and inverse code on a standard PC as well lends the 1D technique well to an initial pseudoquantitative interpretation.
Summary The complex pore structure of carbonate rocks controls their velocity-pressure dependence, resulting in a wide range of pore compressibility. This complexity calls for a proper prediction of carbonate reservoir properties due to changes in the stress during drilling and production. A comprehensive laboratory study on 247 carbonate samples has been previously undertaken to understand how mineral composition, pore type, and rock-fluid chemical interactions control seismic wave propagation. This study described the roles of pore type and mineral composition on the large scatter observed around the main velocityporosity trend, and the suitability of Gassmann''s theory to quantify fluid effects on velocity of carbonate rocks. During those studies a scatter in the pore stiffness versus porosity plot for the low porosity plugs was observed. The goal of this study is to understand the main cause of such a scatter and to start a systematic study of the control of the pore type on the sensitivity of elastic wave velocity to pressure changes. After analyzing the velocity and pore stiffness versus porosity crossplots, obtained by previous bench-top measurements, we selected and measured the elastic properties of 12 dry carbonate plugs characterized by similar porosities but different pore stiffness values, at variable confining pressure up to 20 MPa (small steps increments of 2.5 MPa) and zero pore pressure. This paper focuses on understanding the main factors affecting the pore stiffness scatter of low porosity carbonate plugs and their sensitivity to pressure effects. Results show an inverse relation between velocity sensitivity to pressure and pore stiffness: the sensitivity is high for low pore stiffness plugs and low for those with high pore stiffness. The thin sections of the analyzed core plugs show an increase in the amount of compliant pores for more sensitive plugs, while the less sensitive plugs are dominated by stiff pores. Introduction A detailed knowledge of the sensitivity of velocity to pressure and its relationship to pore type and microstructure is fundamental to describing the elastic behavior of carbonate rocks. Change in stress during drilling and production may lead to changes in the reservoir properties, primarily due to pore pressure changes. The effect of pore pressure on fluids is opposite to the effect of pore pressure on the rock frame: an increase in pore pressure causes a decrease in the rock velocity by softening the elastic rock frame, but the stiffening the pore fluids may counteract this behavior (Mavko and Mukerji, 1995; Mese, 2005; Avseth et al., 2005). Previous experimental and theoretical rock physics studies have focused mostly on the sensitivity of clastic rocks to pressure effects (Birch, 1960; Nur, 1971; Mavko and Mukerji, 1995; Avseth et al., 2005; Zinszner and Pellerin, 2007). This dependence is mainly due to the presence of mechanical microdefects, such as microcracks and grain contacts: the differential pressure (difference between confining pressure and pore pressure) acts to close the microcracks and to stiffen the grain contacts (Mavko and Mukerji, 1995; Avseth et al., 2005; Zinszner and Pellerin, 2007). However, limestones often show less sensitivity of velocity to differential pressure than in sandstones.
- Geology > Rock Type > Sedimentary Rock > Carbonate Rock (1.00)
- Geology > Geological Subdiscipline > Geomechanics (1.00)
SUMMARY Inversion of 3D time-domain electromagnetic data for high conductivity contrasts is a challenging endeavor. Finding a good model update is difficult for small sensitivities associated with resistive backgrounds. Convergence to acceptable solutions can be catalyzed with techniques of overly-conductive initialization and referencing and overcooling. Convergence can be further accelerated via a steepest-descent-type scaling of the model update. INTRODUCTION Although apparently simple, a conductive plate in a resistive background represents many of the difficult challenges that we face in the 3D modeling and inversion of time-domain electromagnetic (TEM) data. In the forward modeling context, we contend with large contrasts in material properties, high absolute conductivities that result in small diffusion distances, and low conductivities that result in wave solutions, boundary condition issues and equation instability (Newman and Alumbaugh, 1997; Haber et al., 2007). Oldenborger and Oldenburg (2007) report on successful modeling of conductive plate scenarios for contrasts up to 10:10 S=m. Inversions of 3D TEM data have been successful in environments with moderate conductivity contrasts (Napier, 2007). As conductivity increases, our need for cheap forward modeling within the inversion requires geometric discretization (Oldenborger and Oldenburg, 2007), corrective sources (Haber et al., 2007), matrix factorization (Oldenburg et al., 2007) or iterative methods (Newman and Commer, 2005). However, as the conductivity contrast increases, finding an inverse solution becomes problematic. Experience with synthetic models and field data suggests that we can recover highly conductive targets in moderately conductive backgrounds, but that we cannot recover highly conductive targets in resistive backgrounds. We present a set of synthetic experiments involving a conductive plate in a series of increasingly resistive halfspaces. The techniques of overlyconductive initialization and referencing (OCIR) and overcooling are used to catalyze convergence of the inverse problem at high contrasts. Although results are specific to a particular inversion algorithm, insight regarding enhanced sensitivity, and choice of initial and reference models is generally applicable. SYNTHETIC MODEL Our model is the 100S=m plate in a halfspace illustrated in Figure 1. The transmitter is a square loop with a step-off current discretized for a duration of 1s with 7 steps per decade from 10-10 s. Synthetic data are collected over a two-dimensional array of surface receivers. The data consist of 91 stations with measurements of Hx and Hz at 9 log-spaced time channels from 10- 10 s. The forward modeling is performed on a 116x84x84 mesh with 12:5m cell widths over the core of the mesh containing the transmitter, target and receivers. Data are simulated for variable background conductivity from 10-10 S=m. Inversions are performed using a 66x50x50 mesh with a core cell thickness of 25m. Although no random noise is added to the data, there is correlated error inherent to modeling and inverting on different meshes. INVERSION RESULTS We present the inversion results starting with the most conductive background and moving toward freespace. Inversions are performed using the least-squares, smoothness constrained algorithm of Haber et al. (2007) that employs a cooling schedule for regularization.
SUMMARY For a controlled-source electromagnetic (CSEM) survey in a shallow-water environment, the presence of the sea surface significantly hinders the interpretation of the measured data. The electromagnetic (EM) wavefields are disturbed by the sea surface. The removal of the sea-surface-related wave phenomena from the data is an important step in order to robustly interpret the collected data. We propose a processing method by which the sea-surface-related multiples would be removed, while a priori knowledge of the EM source wavelet becomes superfluous. The governing equations are obtained from an appropriate application of the EM reciprocity theorem that relates, on one side, the EM fields in the actual measurement configuration including the sea surface and on the other side, the EM fields in a desired source configuration and in the absence of the sea surface, where the water layer is extended to infinity. INTRODUCTION In a controlled-source electromagnetic (CSEM) survey (Eidesmo et al., 2002), it is necessary to interpret the measurements in such a way that a prediction of the presence of hydrocarbons in the sedimentary layers can be made. However, in a shallow-water environment, the presence of the sea surface hinders this interpretation. Electromagnetic (EM) wavefields are partly reflected and partly transmitted by the sea surface. This means that the source signal is contaminated by its socalled source ghost signal, and that the received signals are contaminated by the so-called receiver ghost. Further, the receiver ghost can be considered as secondary source signals that are transmitted in the earth. Hence, removal of all these seasurface- related electromagnetic wavefields from the data is a prerequisite step before actual interpretation of the data can take place (Loseth and Amundsen, 2007). In this paper, we show that an appropriate use of the electromagnetic reciprocity theorem leads to the mathematical equations for a consistent removal of sea-surface-related wavefields. The actual EM wavefield is denoted as {E(x), H(x)} in the frequency domain (with frequency parameter s = j?). The position in space is denoted as x = {x1, x2, x3}. The coordinates x1 and x2 denote the horizontal directions, while x3 denotes the vertical direction pointing into the earth. The wavefield is generated by an EM source in the sea at xS = {x1, x2, Xs 3}.On the sea bottom (not necessarily a horizontal plane), we measure the EM wavefield reflected by the earth geology at the sea bottom (Figure 1). We assume that there exists a horizontal plane at x3 = xobs 3 between the source level and the sea bottom. It is assumed that the sea water is homogeneous with complex permittivity e and real constant permeability m. In this domain, outside the source domains, the EM fields satisfy Maxwell''s equations. For very low frequencies, we may neglect the displacement currents. Then, we may write e = s s, where s denotes the electrical conductivity, which is assumed to be frequency independent within the frequency band of operation. A first step in the processing of CSEM data is the decomposition of the data into constituents that represent upgoing and downgoing wavefields.
- Africa > South Africa > Western Cape Province > Indian Ocean (0.66)
- North America > United States (0.15)
- Geophysics > Seismic Surveying > Seismic Processing (1.00)
- Geophysics > Electromagnetic Surveying (1.00)
Summary Relating seismic wave velocity and maturity in organic-rich shale is still a fundamental issue in the study of organicrich rocks. We found that the inception of the maturity peak -expressed in terms of vitrinite reflectance- (Ro % = 0.65) separates the pressure-dependent anisotropic behavior of organich-rich rocks in two domains: for maturities less than 0.65 (from immature to peak mature rocks), rocks exhibit a low pressure-sensitivity (velocity and anisotropy) but increasing magnitude of anisotropy with increasing Ro%; for maturities greater than 0.65, rocks show an higher sensitivity of velocity to pressure and decreasing magnitude of anisotropy. To start understanding the role of maturation processes and how the spatial arrangement of kerogen could control the elastic properties of these rocks, we complemented traditional rock-physics measurements with the analysis of images obtained using confocal laser scanning microscopy. This latter represents a suitable technique to image organic matter yielding relevant inputs for rock-physics computational analysis. Introduction Organic-rich shales are intrinsically heterogeneous and complex rocks, comprising an inorganic framework in which organic matter may be dispersed in different amounts. Such heterogeneity and the associated production challenges make organic-rich shales questionable in commercial viability, mostly because of cost. In-situ conversion technologies are still emerging and it is still unclear how they will change the technological state of the art. However, two key issues to address include the efficient use of heat and the identification of the most promising source among possible targets. Geochemical characterization of the kerogen dispersed in organic-rich shales has been extensively used to quantitatively discriminate the type and maturity of the kerogenconstituting macerals (Teichmรผller, 1986). However, we still lack fundamental relationships relating the maturity of the organic fraction and the rock-physics properties used to support seismic exploration. Such relationships are a key driver to establish baseline properties of these rocks, and hence to enable advanced control strategies such as the ability to remotely map rock-property changes induced by temperature and/or saturation of generated oil and gas throughout the reservoir. Kerogen maturity has been often expressed in terms of Hydrogen Index (HI)--and by proxy, kerogen type--when relating the maturity of organic-rich rocks to rock-physics properties (Vernik and Nur, 1992; Prasad and Nur, 2001). However, Peters et al. (2004) showed that HI mostly expresses the hydrocarbon generative potential (HGP) and the quality of a source rock (i.e. oil prone vs. gas prone), which are both functions of the kerogen-constituting maceral types. As a consequence, the use of HI alone can be ambiguous, since low HI values are not unequivocally related to maturity but rather to the probable origin of maceral types within each maceral group. Sample preparation, as well as measuring maceral composition and elastic properties can be difficult and time-consuming, which makes it difficult to build large databases and accurately model data. In this paper, we present a relationship between maturity, expressed in terms of vitrinite reflectance (Peters et al., 2004), and the anisotropic Thomsen''s parameter. Then we complement traditional rock-physics measurements with the analysis of images obtained using confocal laser scanning microscopy.
- Geology > Rock Type > Sedimentary Rock > Organic-Rich Rock > Coal (1.00)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Mudrock > Shale (1.00)
- Geology > Geological Subdiscipline > Geomechanics (1.00)
- Geology > Geological Subdiscipline > Geochemistry (1.00)
Summary In most clastic reservoirs experiencing pressure depletion due to production, the hydraulically connected sands in the reservoir naturally compact to some degree. As a consequence, the much lower permeability reservoir shales may experience mechanical tension. The effective seismic response of the reservoir interval is thus a mix of both hardening and softening reservoir components. This phenomenon alters the predicted overall stress sensitivity from that anticipated for a homogeneous, fully connected reservoir interval. The time period over which this effect might be observed is influenced by the rate at which the shales reach pressure equilibrium with the surrounding sands. This work indicates that sub-seismic shale layers of approximately 1m thickness take less than 12 months to equilibrate, whilst thicker shale layers of 8m can take over 10 years. It is concluded that the mechanical and dynamic response of sub-seismic reservoir shale must be considered when quantitatively assessing the 4D seismic signature from frequently shot time-lapse surveys with a periodicity of 6 to 12 months, but also perhaps, for conventional 4D seismic surveys shot over 5 to 10 years. These conclusions are strongly affected by the permeability of the shale layers, the stress state, and are also a function of net to gross and depositional environment. Introduction A quantitative understanding of the variation in the elastic properties of reservoir rocks with changes in stress is an essential element of the petroelastic model that links saturation and pressure changes to the corresponding 4D seismic signatures. At present, the use of laboratory measurements to calibrate this link for a feasibility study or qualitative 4D signature assessment of pressure changes appear adequate for most practical purposes. However, the exact magnitude of the in-situ stress sensitivity is still largely uncertain, and cannot be relied upon for precise determination of pressure changes (Eiken and Tondel 2005). This point is highlighted by the results of Floricich et al. (2006) and Stephen and MacBeth (2006), who combine seismic observations and engineering data to conclude that the reservoir''s stress sensitivity as observed by the seismic data is less than that anticipated. Uncertainty is also reported in other seismic datasets such as those of Fletcher (2004), who reveals an unexplained difference in the sign of the amplitudes and time-shifts associated with depletion. A possible explanation for such seismic observations is offered by a number of well-documented factors that may act to reduce or enhance the stress sensitivity (Nunez and MacBeth 2006). The list includes: inaccuracies in specifying the "dry" rock frame properties due to sample preparation; internal damage due to stress unloading and cutting; the bias of the plug sampling scheme; the effective stress coefficient; the true triaxial stress state of the reservoir; creep; and dispersion effects. An additional item for consideration is the existence of heterogeneities below the scale at which the stress sensitivity laws are being applied. Thus, dynamic interpretation from 4D seismic currently ignores the relatively thin (1 to 10m) shales which are smaller than the size of a typical sand body (20 to 30m) and the seismic wavelength (50m to 100m).
- North America > United States (0.30)
- Europe > United Kingdom (0.29)
- Geophysics > Time-Lapse Surveying > Time-Lapse Seismic Surveying (1.00)
- Geophysics > Seismic Surveying (1.00)
- North America > United States > Gulf of Mexico > Central GOM > East Gulf Coast Tertiary Basin > Green Canyon > Block 205 > Genesis Field > Neb Formation (0.99)
- Europe > United Kingdom > Atlantic Margin > West of Shetland > Faroe-Shetland Basin > Judd Basin > Block 204/25 > Greater Schiehallion Field > Schiehallion Field (0.99)
- Europe > United Kingdom > Atlantic Margin > West of Shetland > Faroe-Shetland Basin > Judd Basin > Block 204/20 > Greater Schiehallion Field > Schiehallion Field (0.99)
- (4 more...)
- Reservoir Description and Dynamics > Reservoir Characterization > Seismic processing and interpretation (1.00)
- Reservoir Description and Dynamics > Reservoir Characterization > Reservoir geomechanics (1.00)
- Reservoir Description and Dynamics > Reservoir Characterization > Four-dimensional and four-component seismic (1.00)
Introduction Summary Analytical and numerical solutions of time domain electromagnetic responses of earth model with axis symmetry have been derived and model responses have been calculated with a pulse source stimulating inside metal casing. Anomalous pattern of TEM responses varying with casing parameters (thickness, conductivity and magnetic permeability) and sensitivity to resistivity changes of formation outside of casing were analyzed. Model results have shown that through casing TEM sounding is feasible and has advantages of multi-components, multi-parameters, deep and untouched probing. Electrical and electromagnetic well logging methods have been the most direct and reliable methods for quantitative evaluation of oil and gas reservoirs. Studies on through casing resistivity logging (TCRL) have been initialized since 1980s by several well logging service companies for purpose of residual oil evaluation of reservoir in producing field. Baker Atlas and Schlumberger have, early or late, developed TCRL apparatus and carried out technical services world wide with promising application results. Some difficulties still remain and need to be solved properly in data processing and interpretation due to the casing effect. For instance, cementing zone correction, correction of casing effects, resolution of thin beds, limitations for evaluating residuals by using the attenuation index, and the sensitivity to saturated water bearing layers, etc.. Some of these problems caused by environmental factors, while others are caused by the imperfect of processing and interpretation methods, and the limitation of the observation system itself is the key factor. The limitation of the observation method is stated as: this system delivers DC current through casing wall and then diffusing to the formation. Formation resistivity probing at various depth can be achieved only by varying the electrode spacing since the DC fields have no sounding capability. Moreover, the down-hole current supply is limited to be <6A by the logging cable. The time domain electromagnetic method (TEM) has advantages of high resolution, high data quality, high efficiency of cost and productivity among numerous electromagnetic methods. It specially suitable for structure mapping in exploration stage and dynamic monitoring of reservoir for production stage due to its high sensitivity to resistive layer. The TEM method which is mature on theory and has been widely used for prospecting on surface is going to be employed for down-hole through casing resitivity imaging with high resolution and large penetration depth. The key factor for bringing this idea to practice is to develop down-hole tool for large power pulse source and highly sensitive receivers. To emit large power pulse with low frequency in cased well can break through the shielding of the casing. To observe data in time domain in the cased well, or between wells, or on surface, or combine of well and surface observations can gave better vertical resolution and larger probing distances than conventional electrical logging. Dynamic reservoir monitoring can be easily achieved if time lapse observations are carried out. Axis Symmetry Earth Model with Casing As shown in Figure 1, a two dimensional model in cylindrical coordinates (r,z) with axis symmetry in ? is set up to simulate a pulse source in cased well.
- Reservoir Description and Dynamics > Reservoir Fluid Dynamics > Flow in porous media (0.56)
- Reservoir Description and Dynamics > Formation Evaluation & Management > Open hole/cased hole log analysis (0.55)
- Well Completion > Completion Monitoring Systems/Intelligent Wells > Downhole tools and equipment (0.54)
Summary We analyzed the sensitivity of multicomponent (MC) seismic (P-P, P-SV, and SH-SH) reflection coefficients (RCs) and traveltimes to water saturation and pore pressure by taking the appropriate partial derivatives. Applying this approach to a poorly consolidated sandstone reservoir partially saturated with light oil and brine, demonstrates that P-P traveltimes have the largest sensitivity to water saturation, but the least sensitivity to pore pressure. In contrast, SH-SH traveltimes have the least sensitivity to water saturation, and the most sensitivity to pressure. P-SV traveltimes have intermediate sensitivities to pressure and saturation, but are more affected by saturation than pressure. By analyzing the sensitivity of MC seismic RCs to water saturation at the reservoir top, i.e., shale over oil saturated sandstone, and at the oil-water contact (OWC), i.e., oil-saturated sandstone over brine-saturated sandstone, we found that the absolute value of amplitudes at all angles is greatest for P-P, smallest for P-SV, and intermediate for SH-SH. In addition, the absolute values of AVO (amplitude variation vs. offset) gradients at the reservoir top and OWC can be organized in descending order as (P-SV, SH-SH, PP) and (P-P, P-SV, SH-SH), respectively. For the sensitivity of both interfaces to pressure (the reservoir top and OWC), angle-dependent relations are extracted. Introduction Seismic reservoir monitoring of a waterflooded reservoir is a challenging problem. Detecting small changes in seismic traveltimes and RCs due to changes in dynamic reservoir properties, i.e., saturation and pressure, is the key to success. Seismic RCs respond to interface properties, i.e., contrasts in P and S-wave velocities and in density between adjacent layers. On the other hand, seismic traveltimes respond to the reservoir interval properties. These seismic attributes are complicated functions of elastic parameters. Elastic parameters themselves are also related to the dynamic reservoir properties using a set of complex equations defined as the rock physics model, e.g., Gassmann (1951) or Hertz-Mindlin (Mindlin, 1949) theories. Consequently, making any judgment about the sensitivity of the MC seismic attributes with respect to saturation and pressure is difficult and dependent on the rock physics model employed. In this paper, we consider the travel times and amplitudes for pre-stack MC seismic data and propose a quantitative approach to obtain their sensitivity to saturation and pressure at different offset ranges. Sensitivity Analysis of MC Seismic Traveltimes In general, MC seismic traveltimes can be expressed as functions of P and S-wave velocities, interval thickness, and incident angle. P and S-wave velocities can then be related to saturation and pressure within the reservoir using rock physics equations. The following four steps summarize the calculation of the traveltime sensitivities: First, the partial derivatives of traveltimes with respect to P and S-wave velocities for a reservoir interval can be calculated either numerically using ray tracing, an Eikonal solver, or analytically assuming straight rays. Second, take the partial derivatives of the P and S-wave velocities relative to bulk and shear moduli, and density of the reservoir layer. Third, take the partial derivatives of the bulk and shear moduli, and density relative to saturation using Gassmannโs theory and with respect to pressure using a stress-sensitive rock physics model.
- North America > Canada > Saskatchewan (0.24)
- North America > United States > Texas (0.16)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Sandstone (1.00)
- Geology > Geological Subdiscipline > Geomechanics (1.00)
ABSTRACT SUMMARY Effective and reliable reservoir monitoring is critically important for optimizing oil/gas production and ensuring safe geological carbon sequestration. It requires an optimal seismic sensor deployment that uses a minimum number of sensors to record the most significant information resulting from reservoir property changes. The monitoring survey has typically been designed using conventional seismic-wavefield illumination analyses. However, seismic-wavefield illumination cannot alone yield an optimum seismic acquisition for effectively monitoring reservoir property changes. We introduce a new approach for designing optimal seismic monitoring surveys by analyzing sensitivities of elastic wavefields with respect to reservoir property changes. The method is based on differentiating the elastic-wave equation with respect to geophysical parameters. The resulting equations are solved using a finitedifference scheme. Numerical studies demonstrate that seismic survey designs based on elastic-wave sensitivity analysis can be very different than those based on elastic-wavefield illuminations under the same optimal criteria. Sensitivity analyses can also be used to investigate whether a VSP or a surface seismic survey is more effective for monitoring P- or S-wave speed changes in a target region. INTRODUCTION Seismic-wave illumination analyses have been used for seismic survey designs (Spitzer et al., 1998; Curtis, 2004). However, it is not an rigorous approach for designs of source/receiver distributions for reservoir monitoring. Optimal sensor distribution is critical for effective and reliable seismic monitoring of oil/gas reservoirs and geological carbon sequestration (Wells et al., 2006). Ideally, sensors should be placed at locations where they can record the most significant changes in timelapse seismic signals caused by geophysical property changes due to movement of fluid/gas/CO2 flows within a target region to be monitored. Therefore, sensitivities of seismic wavefields to geophysical property changes need to be investigated. In addition, one CO2 primary leakage path during geological carbon sequestration is through fault zones. Therefore, a sensor network must be optimally deployed to effectively detect changes caused by possible leakage through those faults. We develop an elastic full-wave finite-difference method for sensitivity analysis of elastic waves with respect to changes in reservoir properties such as P- or S-wave speeds. We derive the sensitivity equations from analytical differentiations of the elastic-wave equations. The sensitivity equations are coupled with the wave equations in a way that elastic waves arriving in a target reservoir behave as a secondary source to sensitivity fields. Using elastic-wave sensitivity components, we define an optimal monitoring criteria based on seismicenergy changes caused by reservoir property changes. Numerical demonstrations with a single layer model show that the wavefield illuminations are not capable of leading any survey design for monitoring of the geophysical property changes occurred within in homogeneous domain. Synthetic analyses with the elastic Marmousi model indicate that optimal survey designs obtained using elastic-wavefield illuminations are quite different than those obtained using sensitivity illumination analyses. Selection of the receiver distribution for detecting P-wave speed changes in the target can also be different from that for detecting S-wave speed changes. In addition, after the seismic survey design is determined, the sensitivity analysis can still be utilized to assess the relationship between the geophysical property changes and the receiver locations.
- Geophysics > Seismic Surveying > Surface Seismic Acquisition (1.00)
- Geophysics > Seismic Surveying > Seismic Processing (0.91)
Summary Geophones are most commonly used in seismic exploration on land, seabed and in boreholes. The design of these commonly used geophones has remained basically the same for many years. The conceptual design of traditional geophones limits their performance in terms of overall sensitivity, noise floor and harmonic distortion. Calibration of traditional geophones has always been an issue, especially the lack of calibration of tolerances associated with the moving mass. Conventionally, the amount of moving mass has to be input to a geophone tester with the design specification being the only source of information. The Geophone Accelerometer (GAC) sensor was first developed in 1990 for borehole seismic applications as an omni-tiltable device that can be used in deviated wells without the need for a gimbal mechanism (Obuchi and Fujinawa., 1991). The improved performance of GAC sensors in terms of calibrated response and vector fidelity has opened up many advanced applications. In this paper we trace the development of the GAC and highlight its performance in comparison to traditional geophones. Introduction A geophone consists of a moving coil suspended in a magnetic flux by means of a pair of springs as shown in Fig 1. When seismic energy arrives at the geophone, the energy moves the housing of the geophone and the moving coil tends to stay in place. Referring to Fig 2, the relative motion between the housing and coil generates electric output, eg. The output of a geophone is terminated by a shunt resister R to damp the motion of the moving mass at optimal condition (typically at D=0.7). Original GAC design The Geophone Accelerometer (GAC) sensor was first developed in 1990 for borehole seismic applications. The GAC sensor is an omni-tiltable device which can be used in deviated wells without the need for a gimbal mechanism. Since in borehole seismic the downhole noise environment is very small, a key feature of GAC technology was the redesign of the magnetic circuit to improve the overall sensitivity. The GAC sensor is designed to be used in an over-damping configuration by using an imaginary short circuit of an operational-amplifier, so that the electric output is proportional to acceleration, and the response is essentially flat over a large frequency band. As the over damping reduces the movement of the moving coil, the result is that the total harmonic distortion of the sensor becomes very small. The signal output from a geophone is proportional to acceleration at frequencies near the natural frequency. If the total damping is large, it is possible to enlarge the frequency range that is proportional to acceleration by increasing the total damping factor, D. To achieve large total damping, it is necessary to have a large magnetic flux density with small moving mass. Fig 3 shows the structure of the GAC that provides large magnetic flux density. The GAC sensor was designed to be used in an over damping configuration by using an imaginary short circuit of an operational-amplifier as shown in figure 3.