An innovative method for the 4D monitor of storage in CCS (Carbon

dioxide Capture and Storage) and oil and gas reservoirs and aquifers

Junzo Kasahara1,3,4 and Kayoko Tsuruga2,4

1 NTT-data-CCS, Co. Ltd.,

2 JGI, Inc., 3 Department of Earth Sciences, Faculty of Science, Shizuoka University

4 University of Marine Science and Technology,

ABSTRACT

This paper is to describe an innovative seismological technology for the 4D survey to monitor of change in

storage in CCS (Carbon dioxide Capture and Storage) and oil and gas reservoirs and aquifer. We present two

fundamental technologies: an integrated interpretation method for WARRP (Wide-angle Refraction Reflection

Profiling) seismic data to determine precise underground structure and the seismic ACROSS (Accurately

Controlled and Routinely Operated Signal System) source to perform continuous monitor f characteristic change

of seismic waves.

The integrated set of 4D analysis tools gives ray-paths and travel-times and generating synthetic waveforms

under given velocity structure model. By this method, we can choose the most appropriate locations for sources

and receivers to watch characteristic change of CO2 storage and reservoirs.

We describe a non-destructive seismic source of well controlled seismic ACROSS source, which can be used to

continuously watch a target. This method, whose potential has been demonstrated in very large-scale surveys, is

nevertheless flexible enough to meet the specific requirements of seismic exploration and monitoring of CO2

storage and oil, gas and aqueous fields.

Keywords: continuous monitoring, 4D seismic survey, CCS, aquifer, oil-and gas reservoir, seismic velocity,

underground structure

1. Introduction

The CCS (Carbon dioxide Capture and Storage) and peak oil are two major topics with linkage to global

climate change (e.g., IPCC, 2005, 2007) and the energy crisis. In recent, the very rapid decrease of ice sheets in

the Arctic Ocean has been observed (Figure 1). Although some scientists think the Earth’s global temperature is

greatly controlled by generation of clouds which can be affected by amount of variation of cosmic ray radiation to

the Earth. However, on 17 April 2009, the U.S. Environmental Protection Agency (EPA) formally declared

Carbon Dioxide (CO2 ) as one of pollutants that endanger public health and welfare (Tura et al., 2009).

Figure 1: Sea ice concentration in September, sea ice motion and sea-level pressure (previous November-April) for 2004 and 2008

(Shimada and Kamoshida, 2008).

In view point of global climate change, the injection of super critical (>31 °C and >7.4 Ma) CO2 into the ground

can be one of solutions (Xue and Nakao, 2008). Many countries have tried to capture and storage the CO2. For

example, in Norway one million tons/year CO2 was injected into the ground in Sleipner. In Snohvit, 0.7 million

tons/year CO2 has been injected into the ground since April 2008. There are other successful CO2 injections

regions in Weyburn in Canada and In Salah in Algeria (IPCC, 2005). In Japan, a pilot experiment of CO2 injection

was carried out in Nagaoka between 2003 and 2005 (Xue and Matsuoka, 2008). In Nagaoka, the injection of CO2

was monitored by electrical resistivity measurements in wells and seismic cross-hole tomography by three wells.

Although the amount of injection was as small as 10,400 tons, the CCS project has been continued as Japanese

national project, and it is now one of the most important national projects in Japan. In order to minimize the

environmental destruction by CCS, it is important to do 4D study of the storage zone.

Not only the CO2, but methane (CH4 ) also strongly affects to the global climate change. Figures 2 and 3 show

variation of atmospheric methane during the last 21 kyr and the last 400 years, respectively. Some scientists

suggest the strong correlation between the abrupt collapse of methane hydrate and the PPTM (Paleocene Eocene

Thermal Maximum) at 55 Ma (Dickens et al., 1995, Kaiho et al., 1996). Green house effect caused by CH4 is

much higher than CO2. The global warming of the Earth may accelerate the discharge of CH4 from gas hydrated

buried in the ground.

Figure 2: Variability of mean atmospheric CH4 concentration

during the last 21 kyr recorded in the Greenland (GRISP) ice

core (Dallenbach and Yayanos, 2000).

Figure 3: Variation of atmospheric methane concentration

over the last 400 years (Houghton, 1997).

In view point of energy, in order to reduce the peak oil problem, the estimation of remaining capacity of oil and

gas reservoirs is one of the most cost effective methods to solve this problem. In addition to the pumping of oil

and gas, injection of CO2 into the oil and gas reservoir by the EOR method also may cause change the 3D shape

of reservoir. Water pumping from aquifers also raises similar changes as the case of oil and gas. Due to the water

pumping, the thickness and shape may be changed.

In order to monitor temporal changes in target zones, which are CO2 storage and reservoir, we use seismic waves.

To obtain correct estimation of size and physical properties of the target zone by repeated 3D (4D) seismic

surveys have been applied to this problem. However, the 4D seismic reflection method is expensive and difficult

to do frequently during a decade-long time frame. It is also necessary to keep the same source- receiver pairs and

processing for the comparison of plural period data. In this key note speech, the authors present an innovative

seismic method to carry out the 4D monitor of target zones.

Seismic ACROSS source Receiver array

Target oil/gas reservoir

Production well

Injection well

Refracted ray

Reflected ray

Seismic ACROSS source Receiver array

Target oil/gas reservoir

Production well

Injection well

Refracted ray

Reflected ray

Figure 4: Simple explanation of monitor of characteristic change of CO2 storage zones or reservoirs by the 4D method using the

ACROSS seismic source and a receiver array. In this method, we use refracted waves through the reservoir and reflected waves

above and below the target zone.

The seismic waves penetrate in the ground and refract through and/or reflect at a target zone (Figure 4). There

are several different kinds of seismic waves: compressional (P) wave, shear (S) wave and surface waves. If the

CO2, oil, gas or water contents in a target zone decrease or increase, P-wave velocity (Vp) and/or S-wave velocity

(Vs ) in the zone may change. For example, CO2 injection causes Vp decrease more than 10 %. In Nagaoka, Japan,

28% of Vp decrease was observed (Xue and Matsuoka, 2008). Vp of water is 1.5 km/s and Vp of reservoir rocks

such as limestone, dolomite, sandstone and shale are more than 3.5 km/s. By this very large difference of Vp,

aquifer can be very strong P-wave reflectors, and thickness change of aquifer may cause reflection change from

top and bottom of aquifer. Refracted seismic wave through aquifer cause seismic shadow zone which is the place

of no or small arrivals of seismic waves, and the width of seismic shadow zone may vary by changes of the place

and thickness of aquifers. Change of oil and gas contents also cause change of seismic wave characteristics.

We propose two innovative technologies. The first one is determination of precise underground structure using

WARRP (Wide-angle Refraction Reflection Profiling) seismic method. The second one is to perform continuous

monitoring of temporal change of characteristics of seismic waves by use of an instrument called ACROSS

(Accurately Controlled and Routinely Operated Signal System) .

2. Outline of the method

Even if we find any change of travel times and/or waveforms of seismic waves between sources and receivers,

we cannot determine the place of change without precise knowledge of seismic wave paths. The integrated

interpretation method determines Vp and/or Vs structure surrounding the target zone (Kasahara et al., 2008). In the

seismic structural study, we can use conventional seismic source such as airguns in offshore and VIBROSEIS in

onshore. Using such velocity structure, we can determine the most appropriate location for seismic sources and

seismometers (or hydrophones).

In addition to the knowledge of seismic ray paths, we use very stable seismic source to enable year’s long

continuous operation because any change of seismic source signatures may cause false temporal change. By use

of seismic waves, the ACROSS developed in Japan was designed to carry out continuous watch of physical

property changes in any target zones in the ground. One of seismic ACROSSs has been continuously operated

for six years. It has transmitted 10-20Hz chirp seismic waves from Toki city, in Japan and have been observed by

Hi-net (High sensitivity seismograph Network) by NEID (National Research Institute for Earth Science and

Disaster Prevention). The frequency bands can be extended to 50 Hz.

3. Determination of the P and S-wave structure of ground surrounding target zone

Figure 5: Concepts of OBS - hydrophone streamer-airgun WARRP as example. Left: Vp profile with depth. Right: Shooting at the

ocean surface and receivers (OBSs) at the ocean bottom. Similar assemblage could be for the onshore exploration.

In order to evaluate time-lapse changes of underground structures, it is necessary to know the present-day

seismic structures for the source-the target zone-receivers. The concept of exploration method is shown in Figure

5. In the example, we use the case in the offshore.

3.1. Offshore refraction and wide-angle reflection survey

In the offshore, a combination of refraction and wide-angle reflection surveying is often called an OBS (Ocean

Bottom Seismometer) (e.g., Kasahara et al, 1997) - air gun survey. The case study is the continental shelf survey

around Japan carried out between 2004 and 2008 using OBSs and airguns. The total length of survey and total

OBSs used are 25,000 km and 5,000 OBSs, respectively. During the survey, two independent airgun shootings,

one for seismic reflection and one for seismic refraction, were carried out. A large volume airgaun with 8,040 in3

are used as seismic source, fired about every 90 seconds to obtain records spaced 200 m apart. The shot time is

calibrated with a GPS clock. The shot positions are determined by DGPS. Common OBS spacing is 5 km, and the

length of each survey line is 500 to 1,200 km for a continental shelf survey in Japan (Kaneda et al., 2005;

Nishizawa et al., 2005a,b ).

OBS is an instrument to monitor seismic waves at the ocean bottom (Kasahara et al., 1995, 1997). It contains

three components of 2 or 4.5 Hz seismometers and a hydrophone. Seismic waves are digitized by 24 bit A/D

converters and stored in hard disk. The sampling rate is typically 200 Hz for the above survey. The absolute time

obtained by an internal clock is used for data analysis. OBSs are deployed by free-fall mode from a support vessel

and are retrieved from the ocean bottom using an electrical corrosion release mechanism by an acoustic release

command. Effective maximum water depth of OBS is 6,000 m. OBS positions are determined by distance

measurements using an acoustic transponder system with GPS positioning. The accuracy of the calculated location

is approximately 10 m (Oshida et al., 2008). Water depth is also determined within ~10m by combination of the

Seabeam depth data and acoustic transponder measurements.

3.2. Data processing and interpretation of data

Field data with pre-processing

Final Vp depth Model

Field seismic data

Synthetic waveformsOK

NO

Integrated analysis for the refracted-wide angle reflected seismic waves (offshore & onshore)

Seismic reflection section and prior geophysical & geological knowledge

P-S converted waves

Deep & shallow reflected waves

Geophone location

Interpretation of later phases

Gravity modeling

Topography

Fitnesscheck

Travel-time inversion

Interactive forward modeling

Travel-time fitting with observed seismic records

Stripped out from shallow part

Consistency check of Vp depth model and 2D-seismic

Vp

dept

h M

odel

Topography PSDMInitial model based on field data and

2D-seismic section

Synthetics and travel-time calculation

Field data with pre-processing

Final Vp depth Model

Field seismic data

Synthetic waveformsOK

NO

Integrated analysis for the refracted-wide angle reflected seismic waves (offshore & onshore)

Seismic reflection section and prior geophysical & geological knowledge

P-S converted waves

Deep & shallow reflected waves

Geophone location

Interpretation of later phases

Gravity modeling

Topography

Fitnesscheck

Travel-time inversion

Interactive forward modeling

Travel-time fitting with observed seismic records

Stripped out from shallow part

Consistency check of Vp depth model and 2D-seismic

Vp

dept

h M

odel

Travel-time inversion

Interactive forward modeling

Travel-time fitting with observed seismic records

Stripped out from shallow part

Consistency check of Vp depth model and 2D-seismic

Vp

dept

h M

odel

Topography PSDMInitial model based on field data and

2D-seismic section

Synthetics and travel-time calculation

Figure 6: Integrated analysis for the WARRP (Wide-angle Refraction Reflection Profiling) seismic method (Kasahara et al., 2008).

We build Vp (and/or Vs) model which satify refraction and reflection data observed by OBSs and seismic reflection section obtained

by ordinary 2D seismic reflection survey. We also confirm whether or not the Vp (and/or Vs) structure satisfies gravity data.

The data processing and interpretation flow was described in Tsuruga et al. (2008) and Kasahara et al. (2008).

Figure 6 shows the scheme for innovative interpretation method for WARRP data. Usually the Vp structure in the

ground is determined by travel-time data of refracted wave and reflected waves and the travel-time inversion

method (or seismic tomography) has been widely used. However, the process to obtain Vp structure is not straight

forward because it is highly non-linear. To avoid this non-linearity, we interpret many seismic wave characteristics

in reflected waves and later refracted arrivals to obtain a reasonable initial model. In addition to travel-time data,

we interpret amplitudes of seismic waves by comparison of observed and synthetic waveforms. W also use S

waves to obtain Vs structure which defines Poisson’s ratio of the ground. The Poisson’s ratio can give information

on contents of liquid materials.

Processing and Interpretation

Receiver-gather seismic records which comprise seismic traces with offset distance for airgun shots are produced

for each OBS station. The SEG formatted records are 60 seconds long for each shot. The seismic record section is

produced by the “Pasteup” software made by Fujie (Fujie et al., 2009). A screen shot of Pasteup is shown in

Figure 7. The receiver-gather seismic records for each OBS are displayed. The selection of appropriate

expression of waveforms as seen in Figure 8 makes easier identification of the onsets and late arrival.

Figure 7: Interactive tools for the interpretation (Fujie et al., 2008). OBS gather seismic records with every 200 m distance trace are

displayed in the center. In this example, OBS is #3. The offset distance in horizontal axis is -30km to +70 km. The vertical axis is

travel times. Other panels are parameter windows for display.

A combination of forward modeling and travel time tomography methods (Fujie et al., 2000, 2003; Korenaga et

al., 2000) are used repeatedly. The modeling tool (Fujie et al.,2008) produces ray paths and travel times (Figure 9).

In the fastest travel time calculation by Fujie et al. (2000, 2003), and Korenaga et al. (2000), the later phase is not

calculated. We modify the travel-time calculation to compute later arrivals (Kubota et al., 2009). The time delay in

the sediments strongly influences to the final seismic structure in the deep ocean conditions and cause difficultyto

determine Vp in the soft sediments because nearby structures are masked by strong water wave arrivals. To avoid

the strong influence due to sediments, we use TWT (Two Way Time) of sediments obtained by ordinary seismic

reflection records.

Figure 8: Various expression of observed seismic records (Fujie et al., 2008).

Figure 9: Modeling of Vp structure. Screen (Fujie et al.,2008). Top: layer model with ray paths. Colors are Vp values. Bottom:

Calculated travel times and observed travel times with error bounds.

During the modeling, the initial model for the travel time tomography is extremely important because travel

time-tomography is strongly non-linear. The ray paths are strongly dependent on a model structure. In order to

avoid the initial model dependency, global searches or a trial-and-error approach are required. First, an initial

model for forward modeling and tomography is produced including water column and sedimentary layers. In

forward modeling, theoretical travel times are calculated and superposed on observed waveforms in the Pasteup

screen (Figures 7 and 10). The Vp depth working model and the 2D seismic reflection records in time section are

compared during the interpretation (Figure 11). The final Vp structure well satisfies the seismic reflection records

as seen in the example. After extensive study by forward modeling, the seismic structure is transferred to

tomographic analysis (Figure 6).

In the final step, synthetic seismograms are calculated by a finite difference algorithm (FDM) (Larsen, 2000). If

the synthetic structure is close to the true structure, most arrivals, including amplitude and arrival times, are well

simulated (e.g., Kasahara et al., 2008).

Figure 10: Example of OBS gather seismic records (top) and ray paths in the Vp layer structure model (Kasahara et al., 2008).

Vertical axes of top and bottom are travel times with move-out velocity of 8.0km/s and depth in km, respectively. WW: water wave;

Pg: refracted P in the crust; Pn: refracted P in the mantle; and PmP: reflected P at the Moho. Horizontal axis for all are offset distance

from OBS. 0-offset is the position of OBS. The offset distance is the one between the OBS and each shot position. Shot spacing is

200 m for this example. Amplitude of each seismic trace is shown by rainbow color. Calculated travel-times are superimposed on the

seismic record.

Figure 11: (Top) Layer structure model with depth at

subduction zone. Philippine Sea plate subducts beneath Japan

Arc from the right to left. Vertical axis: depth. Horizontal axis:

distance from the end of the line. (Bottom) seismic reflection

records superposed by crustal velocity structure model. Vertical

axis: TWT. Horizontal axis: distance from the end of line. Note

that major reflectors are consistent to strong reflectors.

When seismic waves travel through the ground, they refract, reflect and cause P to S or S to P mode conversions

at the sharp or discontinuous boundaries of P-wave velocity (Vp) and/or S-wave velocity (Vs) (Figure 12)

(Tsuruga and Kasahara, 2010b). In the conventional seismic surveys, reflected P-waves are mostly used to obtain

seismic reflection records. Refracted waves, S-waves and surface waves are ignored because they are considered

as noise. During the interpretation, converted phases between P and S waves are evaluated.

Figure 12: Mode conversions between P and S at the boundary of sedimentary layer and hard rock basements.

Example of data interpretation

A huge WARRP dataset using 2D seismic reflection surveys and OBS-air gun was obtained by the Japan Coast

Guard during the continental shelf survey between 2004 and 2008 (e.g., Kaneda et al., 2005; Nishizawa et al.,

2005a,b). The results have been submitted from Japanese Government to the United Nation in 2008 for the

extension of continental shelf. Figure 13 is an example of interpretation (Tsuruga et al., 2008). Using Vp

structure shown in the Figure 13c, synthetic seismic record was obtained, and it can satisfy most of characteristics

identified in the observed seismic record section shown in Figure 13a.

Figure 13: An example of (a) observed data, (b) synthetic seismic record section and (c) Vp structure in depth (Tsuruga et al, 2008).

Vertical axes of (a) -(b) and (c) are travel-times with move-out velocity of 8.0 km/s and depth in km, respectively. Horizontal axes for

all are offset distance from the OBS. Pg: refracted P in the crust, Pn: refracted P in the mantle, Sg: refracted S in the crust, and PmP:

reflected P at the Moho.

4. Continuous monitoring of seismic waves through the reservoir

The ACROSS is a method using either seismic or electromagnetic waves transmitted into the ground

(Kumazawa et al., 2000). One of the seismic ACROSS sources is located in Toki City (Table 1). It is called the

Tono seismic source (35˚ 23´ N and 137˚ 12´ 55˝E, 265 meter-above-sea-level). Transmission has been

continuous since 2002.

3.5 x 3.5 x 2.3 m

2.5 x 2.5 x

1.3 m

10 x 10 x

1.3 m

6 x 3.5 x

2.3 m

6 x 3.5 x

2.3 m

Size of

Coupler

SandstoneSandstone

& Mudstone

Unconso-

lidated

Mudstone

GraniteMudstoneLithology

L: 7.8 Hz

S: 16 Hz

25 Hz

at 10tonf x 2[1] 50 Hz

[1] 25 Hz

[2] 35 Hz

[1] 25 Hz

[2] 35 Hz

Angular

Frequency

At 20tonf

Variable

L: 82 kgm

S: 20 kgm

Variable

( Fixed）

[1] 4.1kgm

Fix

[1] 1.6 kgm

Fix

[1] 8.0 kgm

[2] 3.9 kgm

Fix

[1] 8.0 kgm

[2] 3.9 kgm

Eccentric

Moment

Mori

（MRI/JMA）

Toyohashi

(Nagoya Univ）

Horonobe

(JAEA)

Awaji

(Nagoya/Tokyo

/Kyoto Univs.)

Toki

(JAEA)

3.5 x 3.5 x 2.3 m

2.5 x 2.5 x

1.3 m

10 x 10 x

1.3 m

6 x 3.5 x

2.3 m

6 x 3.5 x

2.3 m

Size of

Coupler

SandstoneSandstone

& Mudstone

Unconso-

lidated

Mudstone

GraniteMudstoneLithology

L: 7.8 Hz

S: 16 Hz

25 Hz

at 10tonf x 2[1] 50 Hz

[1] 25 Hz

[2] 35 Hz

[1] 25 Hz

[2] 35 Hz

Angular

Frequency

At 20tonf

Variable

L: 82 kgm

S: 20 kgm

Variable

( Fixed）

[1] 4.1kgm

Fix

[1] 1.6 kgm

Fix

[1] 8.0 kgm

[2] 3.9 kgm

Fix

[1] 8.0 kgm

[2] 3.9 kgm

Eccentric

Moment

Mori

（MRI/JMA）

Toyohashi

(Nagoya Univ）

Horonobe

(JAEA)

Awaji

(Nagoya/Tokyo

/Kyoto Univs.)

Toki

(JAEA)

Table 1: Currently operated ACROSS in Japan.

In this section, we briefly explain what is the seismic ACROSS and show an example of ACROSS measurements

near Lake Hamana in Central Japan.

4.1 Principle of ACROSS observations

The detection of subtle changes in the characteristics of seismic waves requires on stable and precise

signal transmission and observation. The use of powerful seismic sources generated by chemical explosions may

damage the ground conditions near the source. If such damage near the source occurs, it is difficult to separate the

effect of change due to the source characteristics from ones due to the propagation path from the source to the

receivers.

50 cm50 cm50 cm

Figure 14: Seismic ACROSS in the Tono Geoscience Center, JAEA.

The ACROSS has been developed by the group of Dr. Kumazawa in Tono Geoscience Center of JAEA

Japan. The seismic source for the ACROSS system is small enough not to damage the ground surface. In the

ACROSS system, the transmitted seismic waves are accurately controlled, in phase and frequency, using a GPS

clock. The ACROSS is an eccentric mass-rotation-type signal generator, which is able to precisely control the

excitation of harmonic waves (Figure 15). The combination of clockwise and counterclockwise signal rotation

provides the linear vibration (Kunitomo and Kumazawa, 2004). The seismometers to receive ACROSS signals

were calibrated to obtain amplitude, phase and cross coupling characteristics (Tsuruga et al., 2010a).

minimum frequency spacing 1/Ts = 5 mHz

signal spacing 1/Tf = 20 mHz

Frequency domain datasignal

channelnoise

channel

minimum frequency spacing 1/Ts = 5 mHz

signal spacing 1/Tf = 20 mHz

Frequency domain datasignal

channelnoise

channel

stacking unit period Ts = 200 s

FM period Tf = 50 s

Time domain data

stacking unit period Ts = 200 s

FM period Tf = 50 s

Time domain data

~500 spectral lines for signal

~2000 noise level data

~500 spectral lines for signal

~2000 noise level data

Figure 15: Source in time domain, frequency domain and

spectra. Bottom left: time domain transmitting signal. Top

right: spectral domain line spectra of the ACROSS source. Line

spectra have at every 0.02 Hz separation. Right: amplitude

and phase spectra in frequency domain.

Figure 16: One month stacked spectrum observed at 61 km in

the SE of the Tono ACROSS source. Top: observed

displacement spectrum D(ω) in m; Middle: source

spectrum F(ω in N). Bottom: transfer function H(ω)

obtained by D(ω)/F(ω) in m/N (Tsuruga et al., 2005).

The transmission with the frequency sweep from f1 (Hz) to f2 (Hz) during Ts (second) time-window can generate

a series of spectral peaks from f1 to f2 with every 1/Ts Hz spacing (Figure 15). The division of stacked data by the

source characteristics gives the transfer function. Figure 16 shows an example of source and receiving spectra

obtained by one month stacking at 61 km distance from the ACROSS Tono source. The division of receiver

spectrum by source spectrum gives the transfer function between the source and the receiver (Earth response).

The observed signals are stacked in time and/or frequency domains. The stacking can greatly improve the

signal-to-noise ratio (S/N) of each spectral peak because the frequency and phase are precisely controlled (Figure

17). To obtain the transfer function, we performed the following procedures: (1) weighted-stacking of observed

data in time and/or frequency domains (e.g., Nagao et al., 2010) , (2) estimated the spectral data for each

transmitted frequency with observation error estimated as the root-mean-square of nearby noise level, and finally

(3) derive a transfer function, H={Hij}, defined by a 9-component second-order tensor obtained by dividing the

3-component observed displacements, Di, by the 3-directional signal force, Fj, where i and j represent directions

(e.g., radial, transverse and vertical directions). The transfer function can also be represented as a time series by

inverse Fourier transform. Hasada et al. (2010) proposes the “Sompi” meth for the interpretation of ACROSS

data.

estimated error

estimated noise

estimated noise level

D()

F()

H()

Figure 17: Effects of stacking in S/N increase with duration(left) and waveforms (right). Seismic station is 58 km from the Toki

ACROSS seismic source (Kasahara et al., 2010). 4-8 days have satisfactory S/N as seen in waveforms.

In addition to these characteristics, this system has the benefit of more than six years field experience with

continuous transmissions in the Tono and Awaji regions in Japan. Considering the currently available technologies,

the ACROSS is the only system able to satisfy the necessary conditions for time-lapse study in CO2 storage, oil

and gas reservoirs and/or aquifer.

4.2 Feasibility Study in Tokai Region, Japan

Although the following example is not direct application to the 4D study in CO2 storage, oil-gas and

aquifer, the shallow target is just a scale-down measurement of the following case. In such case, a stacking

duration could be very short. The temporal change may be very easily observed by stacking of short dataset.

J4

J5T6

ACROSS source(Toki city, Gifu)

Explosive survey

Hamana Lake

J4

J5T6

ACROSS source(Toki city, Gifu)

Explosive survey

Hamana Lake

J4

J5T6

ACROSS source(Toki city, Gifu)

Explosive survey

Hamana Lake

Figure 18: Location map of seismic stations for the

ACROSS experiment from Nov. 2004 to Sept. 2005. and

small represent the locations of ACROSS sources at Toki

City, Gifu, and twenty-two temporal seismic stations,

respectively. and denote seismic stations of the Hi-net

seismic network developed by NIED and the permanent

seismic network of Nagoya University, respectively. The

locations of the six chemical explosions and 328 receivers of

the seismic experiment in Central Japan in 2001 (Iidaka et al.,

2003) are represented by and thick solid lines, respectively.

in the shaded area show the epicenters of deep

low-frequency tremors. and show newly built ACROSS

source locations installed by Nagoya Univ. and by Metrological

Research Institute, respectively. Circle centered at the ACROSS

source shows every 25 km distance.

Figure 19: P-wave velocity (Vp) structure used for evaluation

of travel-times and waveforms.

Figure 20: Ray paths for different wave types (Kasahara et al.,

2010).

By re-interpret observed seismic data, we obtained another Vp model which is similar to Iidaka et al., (2003)

(Figure 19), calculated ray paths (Figure 20) , travel times and synthetic seismograms (Figure 21).

Figure 21: Comparison of transfer functions obtained by the ACROSS Tokai experiment (green and yellow wiggles) with synthetic

seismograms at offset distances of 50 to 80 km (after Tsuruga et al., 2005) correspond to the theoretical travel times of the PmP

and PxP phases.

By comparison of synthetic seismograms and observed seismic records, we can identify a very

strong impulsive arrival around 56 km offset distance from the ACROSS source. This impulsive

arrival is interpreted as reflection phases (PI2P) from 15km shown in Figure 20. Onset are also

explained by calculated travel-times.

5. Discussion

We show one of example obtained by ACROSS transmission. The Horai seismic station belongs to

Hi-Net and it is installed at the depth of 200 m in borehole. Location is at 57 km from the Toki souce

and near T6 station in Figures 18 and 19. Figure 22 shows nine month records of S arrival part of

Hrr component (radial transmission and radial receiver) (Yoshida et al., 2004). It can be noticed

that onset of S waves around 10.0 seconds does not show significant change, but amplitudes of later

arrivals around 10.7 seconds show large during spring-summer time. Although it is difficult to

interpreter the cause of temporal variation due to lack of precise Vp and Vs structure, we can say

the continuous and long period observation can give the evidence to time lapse.

Figure 22: Temporal change of waveforms during nine months at Horai seismic station (200m deep)(Yoshida et al., 2004). Bottom: a

part of one-month averaged seismic traces around S arrivals Top: superposition of all traces. Wavefroms around 10.7 seconds show

large temporal change. Onset time of S-arrival does not show significant change during 9-month periods.

6. Summary and Conclusions

The 4D monitor of storage in CCS and oil and gas reservoirs and aquifer is one of the most important

technologies to watch the condition of CO2 storage and reduce the operational costs in oil field and aquifer. This

paper described an innovative seismological technology for the 4D survey. We presented two fundamental

technologies. The first one is determination of precise underground structure using WARRP seismic method. The

second technology is to perform continuous monitor of characteristic change of seismic waves refracting through

the underground or reflecting at the surface and/or the bottom of a target zone by use of an instrument called

ACROSS.

The integrated set of 4D analysis tools analyze ray-paths and travel-times and generating synthetic waveforms

under given a P- and S- wave velocity structure model. Beside ordinary reflection waves, the method does also

support refraction as well as wide-angle reflection waves, a unique capability which broadens the range of

imaging and imaging possibilities. The comparison of the synthetic waveforms with observed records from an

array of geophones enables to precisely discriminate the phases and reflectors involved. By this method, we obtain

very accurate velocity structure and ray paths to choose the most appropriate locations for sources and receivers to

watch characteristic change of CO2 storage and reservoirs.

We described ACROSS which is a non-destructive seismic source of well controlled and low noise seismic

source and can be used to continuously watch a target. This method, whose potential has been demonstrated in

very large-scale surveys, is nevertheless flexible enough to meet the specific requirements of seismic exploration

and monitoring of CO2 storage and oil, gas and aqueous fields.

Acknowledgements

The authors give their great thanks to former colleagues in Japan Continental Shelf Survey Co.

Ltd., Kawasaki Geoengineering Co. Ltd, and JGI Inc., and Drs. M. Kumazawa, T. Kunitomo, Y.

Hasada, T. Nakajima and H. Nagao in the former Tono Geoscience Center.

References

Dallenbach A.T., and Yayanos, A. The distribution of methane in the upper waters of the Southern California

Bight, Journal of Geophysical Research, 97, 11269-11285, 2000.

Dickens., G.R., O’Neil, J. R., Rea, D.K., and Owen, R. M., Dissociation of oceanic methane hydrate as a cause of

the carbon-isotope excursion at the end of Paleocene, Paleooceanography, 10, 965-971, 1995.

Fujie, G., Kasahara, J., Sato, T. and. Mochizuki, K.,Traveltime and raypath computation: A new method in a

heterogeneous media, Geophysical Exploration (BUTSURI -TANSA), 53, 1-11, 2000.

Fujie,G., Mochizuki, K. and Kasahara, J., New travel time and raypath computation methods for reflection and

ps-conveted waves using a regular-grid. G Geophysical Exploration (BUTSURI TANSA), 56, 357-368, 2003.

Fujie, G., Kasahara, J., Murase, K., Mochizuki, K. and Kaneda, Y., Interactive analysis tools for the wide-angle

seismic data for crustal structure study (Technical Report), Geophysical Exploration (BUTSURI TANSA), 61,

26-33, 2008.

Hasada,Y., Kumzawa, M., Tsuruga, K., Kunitomo, T. and Kasahara, J., Automatic travel time determination from

a frequency domain transfer function: The Sompi event analysis, in “Active Geophysical Monitoring”, ed. by J.

Kasahara et al., pp387-401, Elsevier, Amsterdam 2010.

Houghton, J., Global Warming, the Complete Briefing, 2nd ed., Cambridge University Press, 251pp, 1997.

Iidaka, T., Iwasaki, T.,Moriya, T., Kumakawa, L., Kurashimo, E., Kawamura, T., Tamazaki, F., Koike, K. and Aoki,

G.:Configuration of subducting Philippine sea plate and crustal structure in the central Japan region,

Geophys. Res. Lett., 30, 10.1029/20Gl016517, 2003.

Ikuta, R., Watanabe, T., Soma, T., Saiga, A., Yamaoka, K., Tsuruga, K., Kunitono, T., Hasada, Y., Kasahara, J.,

Satomura, M., Kumazawa, M. and Fujii, N., Monitoring of the state of inter-plate coupling using reflection

phase - Preliminary results using a short array, paper presented in Proc. of “2nd International Workshop,

Active geophysical monitoring of the earth lithosphere”, Novosibirsk, Russia, 2005.

IPCC, Special Report on Carbon Dioxide cCpture and Storage, Cambridge University Press, 2005.

IPCC, Assessment Report No. 4: Climate Changes 2007-The physical Science Basis-.Cambridge University Press,

996pp, 2007.

Kaiho,K., Arinobu, T., Ishiwatari, R., et al., Latest Paleocene benthic foraminiferal extinction and environmental

changes at Tawanui, New Zealand, Paleooceanography, 11, 447-465, 1996.

Kaneda, K., Nishizawa, A., Katagiri, Y. and Kasahara, J., Crustal structure model of the Ogasawara Plateau

colliding with the Philippine Sea Plate, 2005 Japan Earth and Planetary Sci. Joint Meeting abstract, J028-011,

2005.

Kasahara, J., Matubara, T., Sato, T., Koresawa, S. and Katao, H., Development of MOOBS/H (magneto-optical

ocean bottom seismometer with hydrophone)-1 (In Japanese), J. Marine Acoustics Soc., 22, 253-267, 1995.

Kasahara, J., Matsubara, T., Sato, T., and Mochizuki, K., Development of high performance digital OBS/H(ocean

bottom seismometer and hydrophone) MOOBS/H-24A (In Japanese), J. Marine Acoustics Soc., 24, 39-47,

1997.

Kasahara, J., Tsuruga, K., Kubota, R., Tanaka, T., Murase, M., Nishiyama, E., Tamura,Y., Nishizawa, A. and

Kaneda, K., Integrated method to study the Earth’s crustal structure using refraction/wide-angle reflection

method, Proceeding of JSEG 60th Anniversary Symposium, 387-392, 2008.

Kasahara, J., Tsuruga, K., Hasada, Y., Kunitomo, T., Ikuta, R., Watanabe, T., Yamaoka, K., Fujii, N., and

Kumazawa, M., Detection and continuous monitor of strong seismic reflectors at the subducting Philippine Sea

Plate using the ACROSS Methodology, in “Active Geophysical Monitoring”, ed. by J. Kasahara et al.,

pp423-436, Elsevier, Amsterdam 2010.

Korenaga, J., Holbrook, W. S., Kent, G. M., Kelement, P. B., Detrick, R. S., Larsen, C., Hopper, J. R. and

Dahl-Jensen, Crustal structure of the southeast Greenland margin from joint refraction and reflection seismic

tomography, Journal of Geophysical Research, 105, 21,591-21,614, 2000.

Kubota, R., Nishiyama, E., Murase, K., and Kasahara, Travel-time estimation of first arrivals and later phases

using the modified graph method for a crustal structure analysis, Geophysical Exploration (BUTSURI TANSA),

62, 105–113, 2009.

Kumazawa, M., Kunitomo, T., Yokoyama, Y., Nakajima, T. and Tsuruga, K., ACROSS: Theoretical and technical

developments and prospect to future applications, Technical report, Japan Nuclear Cycle Development

Institute, 9, 115-129, 2000.

Kunitomo, T. and Kumazawa, M., Active monitoring of the Earth‘s structure by the seismic

ACROSS-Transmitting and receiving technologies of the seismic ACROSS (In Japanese)-,Chikyu Monthly,

Active Monitoring, 47, 44-52, 2004.

Larsen, S., “E3D” 2D/3D elastic finite difference wave propagation code. UCRL.

(http://www.seismo.unr.edu/ftp/louie/class/455/e3d), 2000.

Nagao, H., Takahiro Nakajima T. , Kumazawa, M., and Kunitomo, T., Stacking strategy for acquisition of an

ACROSS transfer function, in “Active Geophysical Monitoring”, ed. by J. Kasahara et al., pp207-221,

Elsevier, Amsterdam, 2010.

Nishizawa, A., Kaneda, K., Katagiri, Y. and Kasahara, J., Crustal structure around the Oki-Daito Ridge in the

northern west Philippine Basin, 2005 Japan Earth and Planetary Sci. Joint Meeting abstract, J028-004, 2005a.

Nishizawa, A., Kaneda, K., Katagiri, Y. and Kasahara, J., Crustal structure across the southern Kyusyu-Palau

Ridge in the Philippine Sea, 2005 Japan Earth and Planetary Sci. Joint Meeting abstract, J028-006, 2005b

Oshida, A., Kubota, K., Nishiyama, E., Ando, J., Kasahara, J., Nishizawa, A. and Kaneda, K., New method to

determining OBS positions for the crustal structural study using airgun shootings and precise bathymetric data,

Geophysical Exploration (BUTSURI TANSA), 61, 15-25, 2008.

Shimada, K. and Kamoshida, T., Future catastrophic reduction of sea ice caused by activation of sea lce motion

and the upper ocean circulation in the Arctic Ocean Journal of Geography, 117, xix-xx, 2008.

Tura, A., Schueleke, J. and Soldo, J., Introduction to this special section: unconventional resources and CO2

monitoring, Leading Edge, 791, 2009.

Tsuruga, K., Hasada, Y., Kunitomo, T., Kasahara, J., Ikuta, R., Watanabe, T., Yamaoka, K., Fujii, N., Kumazawa,

M., Nagao, H., Nakajima, T., Saiga, A. and Satomura, M., Monitoring of the state of inter-plate

coupling using reflection phase -Preliminary results using a long array-, in Proc. of “2nd International

Workshop on Active Geophysical Monitoring of the Earth Lithosphere”, 23-27, Novosibirsk, Russia, 2005.

Tsuruga, K., Kasahara, J., Kubota, R., Nishiyama, E., Kamimura, A., Yoshihiro Naito, Y., Honda, H., Oikawa,

N., Tamura, Y., Nishizawa , A. and Kaneda, K., Evaluation and interpretation of the effect of heterogeneously

inserted thin/thick, high/low velocity layers in OBS/Air-gun crustal structure study, Geophysical Exploration

( BUTSURI TANSA), 61, 1-14, 2008.

Tsuruga, K. and Kasahara, J., Monitoring of the temporal changes of reservoirs by means of wide-angle

reflection and refraction of S-wave, in this symposium, 2010a.

Tsuruga, K., Kumazawa, M., Kunitomo, T., Shigeta, N. and Kasahara, J., Secondary calibration method of

seismometers by utilizing ACROSS signal, in “Active Geophysical Monitoring”, ed. by J. Kasahara et al.,

pp223-235, Elsevier, Amsterdam, 2010b.

Xue, Z. and Nakao, N., CO2 geological storage: International technology and policy trends, and research and

development challenges in Japan, Journal of Geography, 117, 722-733, 2008.

Xue, Z. and Matsuoka,T., Lessons from the first Japanese pilot project on saline aquifer CO2 storages, Journal of

Geography 117, 734-752, 2008.

Yoshida, Y., Ueno, H., Ishikawa, Y., Kunitomo, T. and Kumazawa, M. , The observation and analysis of

ACROSS signals by seismic network - Determination of the transfer function, in Proc. of “1st International

Workshop on Active Monitoring in the Solid Earth Geophysics” , 258-261, Mizunami, Japan, 2004.