Experimental Study on Monitoring CO2 Sequestration by Conjoint

Article pubs.acs.org/est

Experimental Study on Monitoring CO2 Sequestration by Conjoint Analysis of the P‑Wave Velocity and Amplitude Hao Chen,*,† Shenglai Yang,† Kangning Huan,‡ Fangfang Li,† Wei Huang,† Aiai Zheng,† and Xing Zhang† †

Key Laboratory of Petroleum Engineering of MOE, China University of Petroleum, Beijing 102249, People’s Republic of China Special Ultrasonic Equipment Factory of Guangling, Yangzhou, Jiangsu 225002, People’s Republic of China

‡

ABSTRACT: CO2 sequestration has been considered to be one of the most straightforward carbon management strategies for industrial CO2 emission. Monitoring of the CO2 injection process is one of the best ways to make sure the safety storage but is also a major challenge in CO2 geological sequestration. Previous field and laboratory researches have shown that seismic methods are among the most promising monitoring methods because of the obvious reduction in P-wave velocities caused by CO2 injection. However, as CO2 injection continues, the P-wave velocity becomes increasingly insensitive according to the pilot projects when CO2 saturation is higher than 20−40%. Therefore, the conventional seismic method needs improvement or replacement to solve its limitations. In this study, P-wave velocity and amplitude responses to supercritical CO2 injection in brine-saturated core samples from Jilin oilfield were tested using core displacement and an ultrasonic detection integrated system. Results showed that neither the Pwave velocity nor amplitude could simply be used to monitor the CO2 injection process because of the insensitive or nonmonotonous response. Consequently, a new index was established by synthetically considering these two parameters to invert and monitor the CO2 process, which can be thought of as a newer and more effective assessment criterion for the seismic method.

1. INTRODUCTION In recent years, greenhouse gases such as CO2 have been increasing in the atmosphere, which has caused some concerns about climate change.1 CO2 sequestration in geological formations is considered to be the most effective method for fossil-fuel-derived CO2. Field tests have suggested that an ideal site for CO2 sequestration would be located in a geologically simple setting, with permeable reservoir rock, such as porous sandstone.2 Generally, to make sure CO2 is safely stored in the subsurface, the best way is to monitor the process of CO2 injection,3 which will make it possible to open a door for the control and modification of CO2 storage or flooding. Until now, seismic monitoring is one of the most promising methods in CO2 sequestration. Comparatively, it is much more economical, and the processing and acquisition of the data are fairly more conventional. In addition, the CO2 injection process and the flow of reservoir fluids are not affected by seismic monitoring because of its very small strains in reservoir rocks. It has gotten very good results in most CO2 injection projects.4,5 Variations of the velocity and amplitude caused by CO2 injection are two key parameters for the effectiveness of seismic monitoring. However, the seismic velocity is used much more widely because it can be easily detected with high accuracy, while the amplitude changes are usually difficult to measure either in the laboratory or in the field. Generally, the wave © 2013 American Chemical Society

velocity is closely related to the mineralogical composition, porosity of the formation, stratum pressure, and fluid types and content. The compressibility and density of formation would be changed because of the displacement of formation water along with CO2 injection, followed by changes of the propagation characteristics of the seismic waves.6,7 So far, significant changes of the P-wave velocities caused by fluid injection into porous media and the underlying mechanisms have been studied extensively.8,9 However, with the continuous injection of CO2, estimation of the volume of CO2 quantitatively is becoming increasingly important both economically and for safety. In addition, because of the implementation of emission trade in the near future, quantitative evaluation of the injected CO2 will be extremely essential.3 Furthermore, every geological formation has its capacity to contain a certain amount of CO2, and detecting and quantifying the behavior and distribution of CO2 is beneficial for mapping and locating CO2 zones, tracking movements of CO2 fronts, and monitoring the process of CO2 sequestration. SSuccess in such exercises will allow the possibility for the field engineers to control or improve CO2 storage.10 Received: Revised: Accepted: Published: 10071

April 21, 2013 July 4, 2013 August 5, 2013 August 5, 2013 dx.doi.org/10.1021/es4017409 | Environ. Sci. Technol. 2013, 47, 10071−10077

Environmental Science & Technology

Article

Figure 1. Schematic of experimental setup.

from 6.6 to 7.26 cm and a constant diameter equal to 2.5 cm were cored from the same oilfield and then characterized and tested in the Key Laboratory of Petroleum Engineering of MOE. Six investigated core samples were cleaned prior their use, by using various solvents (toluene, benzene, and chloroform) according to the petroleum and natural gas industry standards (SY-T 5336-1996). Their characteristics including their porosities, permabilities, and pore volumes are shown in Table 2. 2.3. Experimental Procedure. The desired temperature was set, and the two transfer cylinders filled with brine and CO2 were allowed to equilibrate for 24 h in the air-bath oven. A clean, dried core was first weighed and then placed into the core holder. Step 1. To avoid the introduction of impurities, couplants like glycerol and lubricating grease were not smeared. However, the waveforms are perfect, which indicates the importance of the integrity of the transducer and port. The time delay of the ultrasonic system is 3.5 μs under the conditions of 250 V emission voltage, 1 time compression ratio, 200 ns pulse width, and 35 dB attenuation. Step 2. After application of 5 MPa overburden pressure, dinitrogen was injected at a high rate for 2 h to remove any remaining toluene in the core. The core was then vacuumed and saturated with brine to measure the porosity and to establish the initial condition of 100% brine saturation. Next, the core was flushed with 20 pore volumes of brine at 0.05 mL/ min to restore any wettability alteration that may have occurred by miscible cleaning. Step 3. Absolute permeability was measured by brine injection. In all cases, including absolute and relative permeability measurements, the confining/axial pressure was set at 5 MPa. Injection was then stopped, and the core assembly was given 2 h to allow the pressure gradient along the core to settle out and reach equilibrium. Then waveforms were measured, and ultrasonic parameters like velocity and amplitude were obtained. In the meantime, the permeability of the core samples was also tested. The confining/axial

In this study, on the basis of integration of ultrasonic sensors and a holder plug with end-face polishing, transducers were directly acting on porous media and the problems of conventional ultrasonic methods were greatly improved. In addition, transmitting, receiving, and signal-processing software was programmed and the test accuracy was increased. On this basis, this study focuses on dynamic changes of the P-wave velocity and amplitude during supercritical CO2 injection into six sandstone core samples from Jilin oilfield; the underlying changes of CO2 saturation were related to both of the P-wave velocity and amplitude during CO2 injection when the pore pressure is constant. With integrated consideration of the relationship of these two parameters and CO2 saturation, a new index is established to monitor variation of CO2 saturation during CO2 flooding or sequestration. This study has resulted in an improved understanding of the response of the ultrasonic parameters to CO2 injection.

2. EXPERIMENTAL SECTION 2.1. Experimental Apparatus. Figure 1 shows the schematic of the experimental apparatus used in this study. This setup consists mainly of the following devices: a pair of ultrasonic transducers ( f = 500 kHz), a data acquisition system used for real-time data acquisition and transfer, an ultrasonic transmit−receive unit, three 100DX syringe pumps, a core holder, three pressure transducers, two high-pressure stainless steel cylinders, an air-bath oven, and two six-way valves. 2.2. Materials. The brine sample was collected from Jilin oilfield. Details of the chemical properties are listed in Table 1. Sandstone cores having cylindrical shape with lengths ranging Table 1. Chemical Properties of the Formation Water Sample property

value (mg/L)

property

value (mg/L)

NaCl NaSO4 NaHCO3

7110 1550 2830

CaCl2 MgCl2

140 110

10072

dx.doi.org/10.1021/es4017409 | Environ. Sci. Technol. 2013, 47, 10071−10077


Article

Table 2. Properties of the Core Used for Ultrasonic Measurements no.

core

diameter (cm)

length (cm)

dry weight (g)

pore volume (cm3)

porosity (%)

absolute permeability (mD)

1 2 3 4 5 6

H13-2 H13-6 H-29 H-30 H-26 H-44

2.46 2.46 2.52 2.54 2.52 2.51

6.96 6.70 7.26 6.91 6.60 6.91

59.05 62.56 76.89 72.68 73.76 75.10

10.37 7.23 6.98 7.02 4.87 5.26

31.30 22.70 19.30 20.10 14.80 15.43

340.70 144.60 64.79 35.82 5.59 4.99

SCO2 1 − SCO2 1 = + Kf K CO2 Kb

pressure was increased to 10, 15, 20, 25, and 30 MPa after each ultrasonic measurement and permeability test. Step 4. The confining/axial pressure was maintained by an Isco pump set at 30 MPa, and then brine was injected into the core to increase the pore pressure to 5, 15, 20, and 25 MPa with a constant pressure mode. The time of equilibrium was 2 h before each ultrasonic test. Step 5. The backpressure was kept constant at 25 MPa, pure CO2 was injected into the brine-saturated core samples with a steady flow rate of 0.2 mL/min, the production brine was collected, and the ultrasonic parameters were measured during CO2 injection.

where SCO2 is the saturation of CO2 in the pore spaces and Kb and KCO2 are the bulk modulus of brine and CO2, respectively. In addition, the bulk density of the rock ρsat is the weighted average of the fluid density ρf occupying the pore space and the matrix density ρm of the rock, which can be calculated using eq 5: ρsat = (1 − ϕ)ρm + ϕρf

ρf = (1 − SCO2)ρb + SCO2ρCO

2

LC ≈

(1)

In this study, the pore fluids are brine and CO2, respectively. Because the shear modulus of a rock is expected to be unaffected by fluid saturation at relatively low frequencies, the shear modulus of the wet rock, Gsat, is simply equal to that of dry rock, Gd, as follows: (2)

As is shown in eq 3, the bulk modulus of the saturated porous rocks is related to the dry rock modulusKd, solid matrix Km, fluid bulk modulus Kf, and porosity ϕ.

K f = SCO2 eK CO2 + (1 − SCO2)e Kb

(1 − Kd /K m) ϕ Kf

+

1−ϕ Km

−

Kd Km2

(7)

(8)

where e is the exponential value; it is closely related to the distribution of the fluids in the rock samples. Consequently, so far, the measured Vp is more accurate considering the complex factors of direct numerical simulation. Amplitude is another important ultrasonic parameter that involves a lot of information on the core sample. It is closely related to the attenuation of ultrasonic. The bigger the attenuation, the lower the amplitude is. According to Biot

2

K sat = Kd +

κK fl /fη

where κ is the permeability, Kfl is the viscosity, and f and η are the seismic frequency and bulk modulus of the most viscous fluid phase. Saturations that are heterogeneous over scales larger than ∼LC have wave-induced pore-pressure gradients that cannot equilibrate. We refer to this state as patchy saturation. In fact, for low seismic frequencies, velocities corresponding to homogeneous and patchy saturations represent approximate lower and upper bounds for given saturations and dry rock properties.14,15 For this reason, in 1995, Brie et al.16 proposed a modified equation with an empirically based fluid-mixing law, as follows:

4

Gsat = Gd

(6)

where ρb and ρCO2 are the bulk densities of brine and CO2, respectively. Therefore, through a combination of the calculated saturated bulk modulus and the dry shear modulus of the rock obtained from experimental tests, Vp could be modeled theoretically. However, a problem with mixed-fluid phases is that the velocities depend not only on the saturations but also on the spatial distributions of the phases within the rock. The Gassmann−Wood model is applicable only if the CO2 and brine phases are mixed uniformly at a very small scale, so the different wave-induced increments of the pore pressure in each phase have time to diffuse and equilibrate during a seismic period. The critical spatial scale is shown in eq 7:

K sat + 3 Gsat ρsat

(5)

The bulk density ρf of the brine−CO2 mixture is given by

3. RESULTS AND DISCUSSION The P-wave properties of the fluid-saturated natural cores mainly depend on factors such as the density, viscosity, coefficient of compressibility, porosity, and permeability, which are all functions of the types and saturations of fluids saturated in the porous media. Consequently, the change of the P-wave properties of the rock−fluid system is the coupling response of the rock and its internal fluids to the changes of external conditions. In other words, by detection of the P-wave parameters of the rock−fluid system under specified conditions, the saturation of fluids in porous media could be reversed theoretically. On the basis of the concept of elasticity of porous media, in 1951, Gassmann developed a widely used equation to predict velocity changes in partially saturated rocks. As eq 1 shows, the P-wave velocity (Vp) in a homogeneous isotropic material can be calculated from the bulk modulus Ksat, shear modulus Gsat, and bulk density ρsat.11 Vp =

(4)

(3)

According to Wood’s fluid-mixing law, the bulk modulus of the brine−CO2 mixture inside the pore spaces can be obtained as follows:12,13 10073



Article

Figure 2. Relationship of the P-wave velocity, amplitude, and permeability versus the confining pressure of natural core H13-6 (60 °C).

theory,17,18 the attenuation of ultrasonic for fluid-saturated porous media mainly depends on the frequency of the wave and viscosity of the fluids in porous media. For the brine-saturated rocks undergoing CO2 injection, the existence of gas would cause great attenuation. First of all, bubbles decrease the bulk modulus of the fluids and the flow could be much easier even in the low-pressure gradient, which causes lots of energy losses at the interface. In addition, the distribution, extrusion, and movement of the bubbles also cause attenuation. For low CO2 saturation, relaxation is the dominate factor and the response of the fluid droplet saturated in the pores to compression and expansion of the cracks leads to attenuation. Meanwhile, the heat formed by repeated compression and expansion of the gas in the pores enters into the brine-saturated rocks. Consequently, the amplitude of the P-wave first decreases and then increases with CO2 injection. 3.1. P-Wave Velocity, Amplitude, and Permeability versus Pressure. Figure 2 presents the results of applying confining and axial pressures for brine-saturated natural cores H13-6 under 60 °C. It tends to increase both the Vp and amplitude and decrease the permeability exponentially. Obviously, as the pressure increases, more and more pennyshaped pores and thin cracks are closed and particles including grains and cement in the rocks contacted much tighter, which led to the logarithmic decline of the core permeability. Analysis indicates that both the density and bulk modulus increased because of the compaction effect, but the degree of the bulk modulus is comparatively more significant. Therefore, Vp increases as the confining pressure increases. In contrast, the pore pressure shows the opposite effect on Vp and amplitude when the confining and axial pressures are constant at 30 MPa, as shown in Figure 3. It is considered that the increase of the pore pressure not only keeps the pores and cracks open and counteracts some of the confining pressure effect but also fills the pores and increases the overall density. Therefore, higher pore pressures cause larger decreases in Vp. The decreases perfectly conform to the obvious quadratic regression relationship. It is evident that the experimental results basically coincide with theoretical analysis.

Figure 3. Relationship of Vp, amplitude, and permeability versus the pore pressure of natural core H13-6 (60 °C).

Figure 4 shows Vp changes caused by the confining and pore pressures for core samples H13-2, H13-6, and H-44. It is clearly seen that the higher the porosity, the smaller the variations in the velocity. The reason is that higher-porosity rocks usually contain fewer thin cracks but more round pores and cavities. Consequently, during the confining pressure increases, the resistance of the brine in the pore to the pore closures in the saturated rock does not contribute significantly to the velocity variation because the round pores or cavities in the rock are already hard to deform. Comparatively, velocity changes caused by pore pressure seem much less distinctive for brine-saturated rock samples. 3.2. P-Wave Velocity and Amplitude versus CO2 Saturation. Figure 5 presents the change of Vp, cumulative brine recovery (CBR), and velocity reduction during CO2 injection. A distinctive trend was observed for the velocity curve, with a sharp fall followed by a more gradual decline. Because the pore pressure remains constant during CO2 injection, the displacement of brine in the pores of the core sample was analyzed to be the predominant reason. Velocity 10074



Article

contrast between CO2 and brine. Actually, both CO2 injection and pore-pressure buildup could change the ultrasonic velocity, as we discussed above. Therefore, there was a backpressure regulator connected to the outlet valve to keep the pore pressure constant and avoid pressure buildup during CO2 injection. In addition, the general trend for the velocity change is the bigger the porosity, the larger the initial Vp and Vp reduction. However, there was no clear correlation between the magnitude of the initial Vp and the total reduction of Vp for these core samples. Figure 7 shows the Vp and amplitude changes obtained during CO2 injection. It is clear that Vp decreased significantly

Figure 4. Vp changes caused by the confining and pore pressures for core samples with different porosities.

Figure 7. Change of the Vp and amplitude with CO2 injection.

in low CO2 saturation and became less sensitive when CO2 saturation was higher than 35%. The results are consistent with both laboratory- and field-scale measurements of previous research conclusions, while the amplitude declines quickly, similar to low CO2 saturation, but then increases sharply as CO2 injection continues. It is considered that, for brine-saturated cores, there was no obvious movement for the brine because of complete compressive deformation to the whole pores, so the attenuation was small and the amplitude was high. With the injection of CO2, more and more pores were occupied by CO2 and the attenuation caused by brine migration increased. Meanwhile, the migration changed the wave energy into heat energy irreversibly. As CO2 injection continued, the dominant migration caused greater fractional loss; therefore, the amplitude dropped very low. After that, more and more brine in the pores was replaced by much lower viscous CO2; attenuation was mainly caused by a sliding fraction between the particles, which led to much lower attenuation. So, the amplitude increased after reaching the lowest point. 3.3. Prediction of CO2 Saturation. Obvious variations were obtained in the measured Vp and amplitude of the natural core samples, and these considerable changes could largely be attributed to CO2 injection. However, the amount of CO2 injection could not be predicted accurately and quantitatively simply depended on the Vp and amplitude because the changes of these two parameters are nonlinear or nonmonotonous during CO2 flooding or sequestration. Xue et al. conducted experiments with porous sandstones and found that there is a clear and monotonously increasing relationship between CO2 saturation and the resistivity values. The following field-scale measurements showed that the use of resistivity is a useful way

Figure 5. Vp, CBR, and velocity reduction with time of core sample H26 (60 °C).

reduction matches well with the changing curve of CBR. Both mainly have three stages: monotonous increasing, drastic increasing, and gradual increasing. Apparently, over 60% of the velocity reduction occurred within the first and second stages (about 20 min) of CO2 injection. Figure 6 shows the relationship of the natural core porosity with the initial Vp and Vp reduction before and after CO2 injection. These decreases were caused by the compressibility

Figure 6. Relationship of the natural core porosity with the initial Vp and Vp reduction. 10075



Article

promising tools for monitoring the process of CO 2 sequestration in aquifers. According to the McElroy data, high-resolution cross-well tomography can resolve 1% velocity variation.19 Unlike the nonlinear or nonmonotonous change of the conventional seismic method, our experimental results show that the VAI that we established is appreciably changed linearly from 6.6% to 18.5% for natural core samples from Jilin oilfield with a porosity of 14.8−31.3% after gas breakthough of CO2 injection. This suggests that the CO2 injection process may be detected and the movement of injected CO2 in the aquifer may be monitored by either cross-well or highresolution surface seismic methods. In addition, the linear decrease in VAT caused by CO2 injection makes it possible to use seismic methods in mapping CO2 zones, tracking CO2 front movements, and monitoring the CO2 sequestration process or CO2 flooding process, which are widely promoted worldwide. Moreover, success in such exercises will allow the possibility for engineers to control or modify the sequestration or flooding process accordingly. 3.5. Existing Problems. Generally, the degree of changes in Vp depends mainly on the crack concentration, porosity, pore structure and geometry, mineral composition of the rock, pore fluid properties, and interactions between the rock frame and pore fluid.6 The amplitude is much more difficult to calculate, or even measure accurately, either in the laboratory or in the field compared with Vp because of the influence of many factors such as scattering, geometrical spreading, source and receiver coupling, and unknown source functions.20 However, the difficulties that normally affect the estimation of absolute attenuation can be minimized if we attempt to estimate the relative attenuation under time-varying reservoir conditions. In addition, it should be noted that CO2 injection may cause pore-pressure buildup in the actual operation of injected CO2 into the aquifers, which is an unavoidable problem especially in reservoirs with permeability of less than 20 mD.21 The reason is that the CO2 injection pressure is much higher compared to the initial pore pressure of the formation brine. This suggests that a field seismic survey would obtain combined results of CO2 saturation variation and pore-pressure buildup. However, Xue et al. found that the effect of the pore-pressure increase was much smaller than that caused by CO2 saturation.2 S-wave velocity and electromagnetic measurements could be used to separate these two effects.10,22 Our future work will focus on this topic.

to overcome the limitations of the seismic method in CO2 sequestration.12 Our study focused on the ultrasonic data that we obtained in the experiments. By no-dimension analysis of both Vp and amplitude, a new parameter considering both Vp and amplitude (VAI) was introduced to represent the change of CO2 saturation during CO2 injection based on LINEST function approximation, as shown in eq 9:

VAI =

Vpc Vi

−m

A pc Ai

(9)

where Vpc and Apc are the Vp (m/s) and amplitude (V) at different CO2 saturations, Vi and Ai are the initial Vp (m/s) and amplitude (V) for the brine-saturated rocks, and m is the slope of the VAI, which can be obtained by linear regression. In order to compare the VAI of various core samples, we process the VAI by no-dimension analysis (VAID) also, as shown in Figure 8. There is a clear and monotonous downtrend

Figure 8. Change of VAID with CO2 injection for core sample H-26.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: +86-135-0123-4602. Figure 9. VAID of natural cores with CO2 injection.

Notes

The authors declare no competing financial interest.

■

between CO2 saturation and VAID. Figure 9 gives the relationship of VAI with CO2 saturation for six natural core samples. The fitting results are satisfying, which indicates that the use of conjoint analysis of Vp and amplitude may be a useful method to overcome the limitations of the conventional seismic method in CO2 monitoring, which usually simply rely on Vp or amplitude. In addition, the general trend is the bigger the porosity, the more evident the VAI changes. 3.4. Promising Applications of VAI in CO2 Monitoring. Seismic methods are widely agreed to be one of the most

ACKNOWLEDGMENTS

This study was supported by National Basic Research Program of China (Grant 2011CB707304) and Natural Science Foundation of China (Grant 50874114). Supported by Science Foundation of China University of Petroleum, Beijing (No. 2462013YXBS003). The authors thank Jilin oilfield for its encouragement and assistance throughout the project and preparation of this paper. 10076



■

Article

REFERENCES

(1) Herzog, H. J. What future for carbon capture and sequestration? Environ. Sci. Technol. 2001, 35 (7), 148−153. (2) Xue, Z. Q.; Ohsumi, T. Seismic wave monitoring of CO2 migration in water-saturated porous sandstone. Explor. Geophys. 2004, 35 (1), 25−32. (3) Nakatsuka, Y.; Xue, Z. Q.; Yamada, Y.; Matsuoka, T. Experimental study on monitoring and quantifying of injected CO2 from resistivity measurement in saline aquifer storage. Energy Procedia 2009, 1 (1), 2211−2218. (4) Arts, R.; Eiken, O.; Chadwick, A.; Zweigel, P.; van der Meer, L.; Zinszner, B. Monitoring of CO2 injected at sleipner using time lapse seismic data. Energy. 2004, 29 (9−10), 1383−1392. (5) Ivandic, M.; Yang, C.; Lüth, S.; Cosma, C.; Juhlin, C. Time-lapse analysis of sparse 3D seismic data from the CO2 storage pilot site at Ketzin, Germany. J. Appl. Geophys. 2012, 84, 14−28. (6) Wang, Z. J.; Nur, Amos, M. Effects of CO2 flooding on wave velocities in rocks with hydrocarbons. SPE Reserv. Eng. 1989, 4 (4), 429−436. (7) Xue, Z. Q.; Ohsumi, T.; Koide, H. Laboratory measurements of seismic wave velocity by CO2 injection in two porous sandstones. GHGT6 2002, 359−364. (8) Mavko, G.; Mukerji, T. Bounds on low-frequency seismic velocities in partially saturated rocks. Geophysics 1997, 63 (3), 918− 924. (9) White, J. E. Computed seismic speeds and attenuation in rocks with partial gas saturation. Geophysics 1975, 40 (2), 224−232. (10) Wang, Z.; Cates, M.; Langan, R. Seismic monitoring of a CO2 flooding in a carbonate reservoir: A rock physics study. Geophysics 1998, 63 (5), 1604−1617. (11) Gassmann, F. Elastic waves through a packing of spheres. Geophysics 1951, 16 (4), 673−685. (12) Kim, J.; Xue, Z. Q.; Matsuoka, T. Experimental Study on CO2 Monitoring and Saturation with Combined P-wave Velocity and Resistivity. CPS/SPE international Oil & Gas Conference and Exhibition, Beijing, China, June 8−10, 2010. (13) Wood, A. B. A textbook of sound; G. Bell and Sons Ltd.: London, 1941. (14) Mavko, G.; Mukerji, T. Bounds on low-frequency seismic velocities in partially saturated rocks. Geophysics 1998, 63 (3), 918− 924. (15) Mosen, K.; Johnstad, S. E. Improved understanding of velocitysaturation relationship using 4Dcomputer-tompgraphy acoustic measurement. Geophys. Prospect. 2005, 53 (2), 173−181. (16) Brie, A.; F. Pampuri, F.; Marsala, A. F.; Meazza, O. Shear sonic interpretation in gas-bearing sands. SPE Annual Technical Conference and Exhibition, Dallas, TX, Oct 22−25, 1995. (17) Biot, M. A. Theory of propagation of elastic waves in a fluidsaturated porous solid. J. Acoust. Soc. Am. 1956, 28 (2), 168−178. (18) Biot, M. A. Mechanics of deformation and acoustic propagation in porous media. J. Appl. Phys. 1962, 33 (4), 1482−1498. (19) Harris, J. M.; Nolen-Hoeksema, R. C.; Langan, R. T.; Van Schaack, M.; Lazaratos, S. K.; Rector, J. W. High-resolution crosswell imaging of a west Texas carbonate reservoir: Part I-Project summary and interpretation. Geophysics 1995, 60 (3), 667−681. (20) Harris, J. M.; Yin F.; Quan Y. L. Enhanced oil recovery monitoring using P-wave attenuation. SEG Annual Meeting, Denver, CO, Nov 10−15, 1996. (21) Van der Meer, L. Investigations regarding the storage of carbon dioxide in aquifers in the Netherlands. Energy Convers. Manage. 1992, 33 (5−8), 611−618. (22) Hoversten, G.; Gritto, R.; Daley, T.; Majer, E.; Myer, L. Crosswell seismic and electromagnetic monitoring of CO2 sequestration. GHGT6 2002, 371−376.

10077


Experimental Study on Monitoring CO2 Sequestration by Conjoint

Recommend Documents