LBNL-44876

Monitoring Underground Gas Storage in a FracturedReservoir Using Time Lapse VSP

Thomas M. Daley,’ Mark A. FeighnerYandErnestL. Majer’

lEarthSciences DivisionErnestOrlando Lawrence Berkeley NationalLaboratory

Universityof CaliforniaBerkeley, Czilifornia94720

2Departmentof Mathematicsand SciencesSolano Community College

Suisun,California 94585

March 2000

ThisworkwassupportedbytheAssistantSecretaryforFossilEnergy,FederalEnergyTechnologyCenter(FETC),ProcessingwasperformedattheCenterforComputationalSeismology,whichissupportedbytheDirector,Officeof Science,Officeof BasicEnergySciences,Divisionof EngineeringandGeosciences,of theU.S.Departmentof EnergyunderContractNo.DE-AC03-76SFOO098.

-. --.—- -,....,., -,.&, .,.. . . ., . . . . . , . .,,,.,%,. ,., .-~--, . .. . . . . . . . .<,.s.., ... . . . .. . ..- ... , .,., , . . . . . . . .. . . . . . . ,, .------- . . . . . .

DISCLAIMER

This repo~ was.prepared as an account of work sponsoredby an agency .ofthe United States Government. Neitherthe United States Government nor any agency thereof, norany of their employees, make any warranty, express orimplied, or assumes any legal liability or responsibility forthe accuracy, completeness, or usefulness of anyinformation, apparatus, product, or process disclosed, orrepresents that its use would not infringe privately ownedrights. Reference herein to any specific commercialproduct, process, or service by trade name, trademark,manufacturer, or otherwise does not necessarily constituteor imply its endorsement, recommendation, or favoring bythe United States Government or any agency thereof. Theviews and opinions of authors expressed herein do notnecessarily state or reflect those of the United StatesGovernment or any agency thereof.

DISCLAIMER

Portions of this document may be illegible

in electronic image products. Images are

produced from the best available original

document.

2

Abstract. This paper reports on time lapse VSP study in a naturally fractured

reservoir used for underground gas storage. Four 9-component VSPS were acquired ilong

with a walkaway and then repeated after the reservoh properties changed. The initial

survey was conducted with the reservoir mostly gas saturated and the follow-up survey

was conducted with the reservoir mostly water saturated. The near-offset 9-c VSP was

investigated for indications of shear-wave splitting. No clear evidence of S-wave splitting

or other anisotropic wave propagation could be attributed to the reservoir horizon. Ray

trace modeling was performed using the VSP velocity results which showed that very

little horizontal propagation was within the reservoir for the ofEsetVSP sites. Because of

the ray-path limitations, the offset VSP site data could not constrain time-lapse changes

to the reservoir horizon. The P-wave walkaway did show a small time-lapse change

within the reservoir horizon. This time-lapse change in velocity was interpreted in terms

of a crack based equivalent media model to give a trade-off curve between crack density

and saturation. A porous media approach was used to estimate porosity which shows

time-lapse changes. Other analysis of the VSP including reflectivity and frequency vs.

time analysis showed no clear time-lapse changes (plots in Appendix D). Surface seismic

data near the VSP well was analyzed and modeled and we conclude that AVO analysis

is not applicable to this site.

3

1. Introduction

The Northern Indiana Public Service Company (NIPSCO) operates naturally

fractured reservoirs for seasonal storage ofnatural gas. The gas is injected during

summer and withdrawn during winter. As part of DOE sponsored research in ilactured

gas production, Lawrence Berkeley National Laboratory (LBNL) designed a VSP

(Vertical Seismic Profile) experiment to aid delineation of NIPSCO’S dolomitic Trenton

Formation reservoir, and to study the seismic effects of variable gas pressures. The

reservoir is in NIPSCO’S Royal Center field in Northern Indiana.

The effects of fracturing on seismic wave propagation have been widely studied

(O’Connell and Budiansky, 1974 Crampin, et al., 1986, Schoenberg and Sayers, 1995,

Pyrak-Nolte, et al., 1990) with one of the more significant effects being shear-wave

splitting (also called shear-wave birefringence). Nine-component VSP’S (3 component

sources and 3 component receivers) provide an excellent method of measuring shear-wave

splitting (Majer, et al., 1988, Daley, et al., 1988, Daley and McEvilly, 1990, Winterstein

and Meadows, 1991).

2. Background

The Royal Center field has limited geophysical characterization, with few well logs

and only field wide monitoring of gas injection/withdrawal volumes. However, the

annual displacement of water by gas within the natural fractures of the reservoir made

this field a good candidate for time-lapse monitoring. There has been no hydrofiacing

or other intentionally induced fracturing of the reservoir.

The Trenton formation is a paleozoic Ordovician dolomite which is part of the mostly

shale and limestone stratigraphy of the Royal Center field (Figure 1). It is believed

by the field operators that the top section of the Trenton dolomite is unfractured and

forms a cap for the reservoir. The thickness of the cap section (as well as its fracture

, I

4

content) was assumed variable and was not well defined by the field operators. Only

the distribution of wells indicated a potential dominant fracture orientation of NW-SE

along the axis of an proposed anticline (see Figure El in Appendix E). The dip within

the Royal Center Field is only 1% - 2%.

As part of a fractured gas research program run by DOE’s Federal Energy

Technology Center (FETC), we designed a VSP survey which would take advantage of

the reservoir’s time-lapse changes to delineate spatial variations in reservoir properties

as well as providing basic information for other seismic imaging proposed at Royal

Center. Among the poorly defined properties of the reservoir are thickness of fracture

interval, extent of fracturing above the reservoir (possible leakage), spatial distribution

of fractures, and dominant orientation of reservoir fractures. We hoped to use multiple

offset 9-component VSP’S to monitor shear-wave splitting (and other seismic properties)

at a productive storage well (implying multiple connected fractures) and at a poor

storage well (implying no fracturing of poorly connected fractures). From available well

sites, wells 157 (good storage) and 46 (no storage) were selected. The two major goals of

this project, shown in schematic form in Figure 2, were to measure the reservoir fracture

orientation from S-wave splitting and to estimate the iiacture density distribution

from spatial distribution of time-lapse changes in seismic travel times. A proposed

optimal survey design included source locations as shown in Figure 3, however field site

restrictions (including land and road access limitations) led to a less optimal distribution

of source location as described in the data acquisition section below (Figure 4).

There were two phases of the time-lapse VSP data acquisition at the NIPSCO

Royal Center field site. These two VSP surveys acquired data from essentially identical

acquisition geometry (source/receiver locations) under distinctly different reservoir

conditions. During the initial survey in December 1996 the reservoir gas pressure was

near its maximum of about 400 psi; during the second survey in May 1997 the reservoir

gas pressure was reduced to about 250 psi. Since the natural water pressure within the

A.. . :.

5

Trenton formation is about 310 psi, the reservoir was mostly gas saturated in 12/96 and

mostly water saturated in 5/97.

3. Data Acquisition

The initial data acquisition for the NIPSCO VSP experiment took place during

Dec 1996. The sensors were 3-component wall-locking geophones in a 5 level string with

8 foot spacing between recording depths. Downhole digitization was used to enable

multi-level recording on a 7-conductor wireline and to reduce ambient electrical noise in

the data. Because the 157 well was pressurized,a lubricator assembly long enough to

contain the entire string was assembled on site and used for the duration of the survey

(Figure 5). The source were vibroseis trucks, one Fayling P-wave truck and 2 Mertz

S-wave trucks. The vibroseis sources (both P- and S-wave) used a 12 to 99 Hz sweep,

12 s long with a 3 second listen time. The sample rate was 1 ms. Each VSP data

set has over 600 Mbytes of data, with about 10,000 seismograms. The S-wave trucks

were positioned at each source site such that they generated an in-line and a cross-line

polarized shear-wave (relative to the line connecting the source location and the well).

Source site access was a problem because we could not obtain county permission to use

the paved roads for source points and we could not use farm land at the well 46 site.

Farm land surrounding the 157 well was made available by the land owners, however the

soil proved too unconsolidated to support the 40,000+ pound vibroseis trucks. We were

left to locate our source sites on dirt roads, which are in a strict North-South/East-West

grid. The available roads determined the source locations used in the survey. The

following VSP data was therefore acquired at Royal Center field in Dec. 1996 (where

S1 is an S-wave polarized parallel to the line connecting source and well and S2 is an

S-wave source polarized tangential to the line connecting source and well):

Well 46

. Run 1: 970 ft. to 578 ft. at 8ft. intervals, P, S1 and S2 sources at zero offset (well

, I

6

pad)

Well 157

●Runl : 1088 ft. to296 ft. at8 ft. sensor intervals for P, Sland S2 sourcesat well

pad (site 1), and P2 source at site 2.

. Run 2 : 1088 ft. to 696 ft. at 8 ft. sensor intervals for P source at site 7, and S1 and

S2 sources at site 2.

● Run 3 : 1088 ft. to 696 ft. at 8 ft. sensor intervals for P, S1 and S2 sources at site 3.

. Run 4 : 1088 ft. to 696 ft. at 8 ft. sensor intervals for P, S1 and S2 sources at site 4.

. Run 5: 1016 ft. to 984 ft. at 8 ft. sensor intervals for P and S2 sources on East

walkaway at 50 ft. source intervals horn about 200 ft. to 1550 ft.

. Run 6: 1088 ft. to 696 ft. at 8 ft. sensor intervals for P source at site 6 (2740 ft.

offset).

Figure 4 shows a schematic location map of the source sites. The second phase

VSP survey (May 1997) was identical except that

1) Well 46 was not used (because the lack of access permission away from the well

pad eliminated our ability to perform spatial imaging and therefore compromised the

survey),

2) Run 1 data (site 1) at well 157 was acquired up to 96 feet at 8 ft intervals to give

better shallow velocity control,

3) Run 6 data at site 6 was acquired with both P and S-Wave sources because of

better off-road access conditions.

3.1. Data Processing

The data processing was mainly performed using the FOCUS-3D seismic processing

package produced by COGNI-SEIS (versions 4.0 and 4.1). The coordinate rotation aud

particle motion analysis was done using LBNL software.

The processing flow for each data set is as follows:

-—’?.. ., . .<. . . . —-.—— .—. .- -.

7

1) Convert data horn field formatted data tapes.

2) Edit and stack uncorrelated traces (in May 1997, stacking was done in the field).

3) Correlate traces and sort by source type.

4) Use P-wave arrival to calculate 3 component geophone rotation angles.

5) Use the calculated rotation angles to rotate each source type data set into

vertical, horizontal in-line, and horizontal cross-line geophone orientations.

6) For zero offset, use S-wave arrivals to measure s-wave splitting and calculate

anisotropy axis of symmetry to estimate fracture orientation.

7) Pick arrival times, calculate P and S-wave velocities and Poisson’s ratio.

The phase 1 well 46 data set had problems because of background noise and

rotation analysis error from near vertical P-wave propagation (in-field restrictions

discussed above prohibited using an offset P-wave source location). Without reliable

orientation information, the shear-wave data could not be analyzed. The well 46 data

set has not been analyzed beyond step 5, and was not repeated in phase 2 because of

these problems.

The phase 1 well 157 data sets have fair to good data quality, limited by noise

bursts in the well which required hand editing. These noise bursts were probably caused

by gas release into the well. The phase 2 data sets appeared to have more noise bursts,

probably because the reservoir was now mostly water saturated, allowing more gas

bubbling. After performing extensive hand edits in the phase-1 processing, effort was

spent on developing automated noise burst editing routines in the phase 2 data sets

(Figure 6). By editing the noise bursts in the uncorrelated recordings, the impact on the

correlated data was minimized. A 40 Hz, 60 dB/octave low pass filter was applied to

the data before the zero-offset time picking. The rotation angle analysis was done using

P-wave particle motion eigenvalue analysis (Daley, et al., 1988). The resulting edited,

correlated, rotated seismograms are shown in Appendix A.

The calculation of anisotropy axis of symmetry was done with a 4-component

, I

8

orientation analysis (Alford, 1986 Thomsen, 1988). However, close inspection of field

data has shown a source amplitude variation between the two S-wave vibroseis trucks

which invalidates the assumptions of this analysis. Inspection of vibroseis baseplate

accelerometers found a 25$Z0amplitude ddference between the two shear-wave vibroseis.

However, correction for this difference still leaves a peak amplitude difference of a factor

of 4 between the data from each shear-wave truck. Additionally, there is noticeable

difference between the two shear-wave wavelets, despite wavelet analysis which was

performed to chose a consistent vibroseis reference sweep for both s-wave sources (see

Figures 7a-b). If the two shear-wave sources were not generating identical wavelets

into the subsurface, this is another violation of the underlying assumptions of the

4-component anisotropy axis-of-symmetry orientation analysis. Therefor we are not

using the results horn this analysis of the axis of symmetry of anisotropy. The lack of

significant S-wave splitting for the zero offset at well 157 (which is unaffected by the

amplitude variations) indicates that the ~component orientation analysis would have

been inconclusive even if the S-wave amplitude variation was not present in the data

because S-wave splitting is the attribute used for 4-component analysis. The s-wave

splitting analysis, like the velocity analysis utilized travel time information obt tied

from the rotated data. ‘Travel time picking was tested using cross-correlation of fist

arrival wavelets, and hand picking of peaks or zero

for all methods. The results shown in Figures 8a-d

maximum of the correlated wavelet.

crossings, with results in agreement

and Appendix B are for picks of the

4.

4.1.

Results

Site 1- Vertical Propagation Results

Initial results from the Dec 1996 VSP indicated

effects and strong reflections (an example is shown in

minimal shear-wave splitting

Figure 7a-d). Analysis of velocity

—---.-7- TYr--— ——. . ----

9

structure obtained from the zero offset VSP showed largevelocity variations in the Royal

Center Field ranging from 9000 ft/s to nearly 20,000 ft/s in the Trenton Dolomite. A

large velocity inversion was measured related to the low velocity Eden shale formation

which overlies the high velocity Trenton dolomite. This velocity inversion has significant

impact on the offset VSP raypaths and on our ability to measure properties within the

reservoir which will be discussed below. With the 9-component VSP, P- and S-wave

velocities and their ratio and Poisson’s ratio can be calculated. These data plots are in

Appendix B.

The May 1997 survey successfully repeated the Dec 1996 survey. The two time-lapse

data sets have been analyzed for spatial or temporal variations in P- and S-wave travel

time, shear-wave polarization, and shear-wave splitting. The phase 1 and phase 2

zero-offset data analysis both indicated essentially isotropic propagation. When we

compared the measured travel times between phase 1 and phase 2, we found no change

in S-wave or P-wave time within the 3-4 ms scatter of the data points and allowing for

a static time shift which is attributable to near-surface conditions (Figure 8a-d). This

means that the measured velocity from the zero offset VSP data set was not affected

by the change in reservoir gas saturation. A comparison of shear-wave particle motion

(Figure 9) indicates the polarizations stayed predominantly aligned with the sources

(isotropic propagation), with some indication splitting which can not be attributed to

the reservoir zone.

Our hypothesis for the apparent inability of the zero-offset survey to detect

either fracture-induced anisotropy or velocity changes caused by variable gas

pressure/saturation is that the zone of fracturing associated with the ‘llenton Formation

reservoir around well 157 is too thin to be seen by the relatively long wavelengths of our

VSP survey. With dominant frequency of 40 Hz and fast velocities of about 10,000 ft/s

for S-waves, our S wavelength was about 250 ft The relatively thin reservoir (probably

about 30 to 50 ft) was not “visible” to the waves propagating vertically through the

,

10

Trenton Formation. Additionally, the shallow velocity structure and available well depth

appears to limit the ability of offset VSP geometry to sample a wide range of incident

angles for the Trenton reservoir, thereby limiting the use of amplitude vs offset analysis.

However, there are small, 1 to 3 ms, time changes observed at depth (after allowing

for the larger time changes due to the near surface material changes). The two most

compelling examples are the P-wave walkaway data and the shear-wave data from sites

3 and 4.

Analysis focused on interpreting these small, 2-3 ms travel time changes seen at

depth as time-lapse velocity changes confined to the reservoir (presumably due to

varying gas pressure and saturation). Before analysis of the walkaway and offset data

could begin, we needed to understand the propagation paths in the subsurface; this need

led to development of ray traced models for the VSP.

4.2. Modeling

The effects of the subsurface velocities on wave propagation can be estimated by

seismic modeling. We performed ray trace modeling, based on velocities measured by

the zero-offset VSP in phase 1 (in Appendix B), to investigate the ability of the offset

source locations to “see” more of the reservoir than the zero offset. Our initial modeling

used the following velocity structure:

Layer Depth (ft) P Velocity (ft/s) S Velocity (ft/s)

0-300 6,100 2720

300-500 15,000 9200

550-670 13,000 6200

670-930 9,500 4800

930-1100 18,000 9500

,- ----, -. -.- r.,? . . . ,. -.7””,-- . .. . . ... . . . . . ... . . .. . . ...- -- -—— ..— -—- --- . . -

11

Unfortunately, the results from this model indicated that the velocity structure at Royal

Center significantly limits the propagation within the reservoir. More detailed modeling

with subdivisions of each layer (Figure 10a-b) shows that only a small percentage

of the offset VSP raypath is within the Trenton formation, and less is within the

reservoir (using a high velocity, unikactured cap at the top of the Trenton overlying a

lower velocity fractured reservoir). The highly variable velocity layers (9000 to 18000

ft/s P-wave velocities) with a strong velocity inversion constrains most of the wave

propagation to shallow high velocity layers. The high velocity Silurian limestone bedrock

above the slow velocity Maysville formation caused most of the energy to be propagated

in the bedrock formation. Only a small percentage of the ray paths are in the Trenton

formation reservoir zone (below approximately 950 to 1000 feet). We believe this effect

is the limit on the ability of far offset VSP sites to detect spatial changes in reservoir

properties.

4.3. Time Lapse

Given

anisotropy

detect and

that the

Analysis

reservoir fractures were not directly detected with shear-wave

measurements, we looked to time lapse changes in reservoir properties to

estimate the extent of fractures. We believe that the displacement of gas

by water within the fractures should cause changes in seismic velocity which can be

modeled via theoretical relationships between fractures and seismic propagation such as

equivalent media velocity variations or fracture stiffness variations. In both cases the

presence of gas as a fracture filling should create lower velocities than the presence of

water as a fracture filling material. Lower velocities will lead to increased travel times as

measured by a VSP survey. As described above, the Dec 1996 (phase 1) survey had gas

saturation in the reservoir, while the May 1997 (phase 2) survey had water saturated

reservoir conditions with an associated decrease in reservoir pressure. Our basic tool for

VSP time lapse analysis is travel time differencing. We expect little change in travel

, I

12

time above the reservoir (except for the near surface which has velocity changes caused

by seasonal water saturation changes) and an increase in travel time difference at and

below the reservoir. For each source site we have three independent measures of time

lapse change in travel time; the P-wave, the in-line S-wave (S1), and the cross-line

S-wave (S2). Appendix C has the plots of travel time change for these three sources

for all source points. Additionally, the walkaway survey provides well sampled spatial

measurement of the time lapse changes along one azimuth. As mentioned above, the “

time

were

4.4.

lapse data sets with the most

the P-wave walkaway and the

compelling evidence for seismic

site 3 and 4 S-wave data.

Time-Lapse P-wave Walkaway Data

velocity variations

The P-wave walkaway data was analyzed by comparing the time lapse changes

at the shallowest sensor (984 ft.) with the changes at the deepest sensor (1016 ft.).

Without knowing the actual reservoir interval (production is over a large perforated

casing interval), these sensors straddle the “best guess” of the reservoir available from

field engineers ( 984 to 1016 feet). We hypothesize that the deeper sensor will record

a larger time lapse change due to velocity changes within the reservoir induced by

changing from gas to water saturation. Figure 11 shows there is a consistent 0.1 to 0.5

ms difference in the time lapse delay between the two walkaway sensor depths. The

mean value for the 27 walkaway source locations is 0.2 ms with a standard deviation

of 0.1 ms. The consistency of the time-lapse d.iflerence gives us confidence that this

observation is caused by physical property changes within the reservoir zone. We are

thus using a volumetric average over the reservoir region probed by the walkaway, rather

than trying to delineate spatial variability in the reservoir, as was our original plan.

Knowing the time-lapse change in P-wave propagation time, we need to determine

the actual propagation distance in the reservoir to determine the change in seismic

velocity. The propagation distance is modeled in via raytracing described previously to

.. . .. /.., . . . . . . , . . . —..

determine the P-wave propagation

survey. The propagation distante

distance (raylength)

is dependent on the

13

in the reservoir for the walkaway

depth chosen for the interface

between Trenton cap rock (high velocity, presumably unfractured) and Trenton reservoir

(lower velocity, fractured). Figure 10 shows raypaths in the reservoir horizon. The

reservoir ray lengths were determined by differencing the lengths for top and bottom

sensors. We found values for the difference ray lengths in the reservoir ranging between

20 feet (near offsets) and 60 feet (far offsets). When the 0.2 ms time-lapse travel-time

variation is attributed to these propagation distances, we calculate 5% to 15’?ZOvelocity

changes in the reservoir relative to the velocity measured in the near offset survey at site

1. Additionally, inspection of the interval velocities from site 1 finds the interval centered

near the reservoir (1016’) has a 10% velocity change, so we have confidence in using

10% as our estimate of velocity change in the Trenton reservoir due to changing gas

saturation. With an estimate of time-lapse reservoir velocity change, we can estimate

the material property variation necessary to cause the velocity change.

5. Estimates of Fracture Density and Saturation in the

Reservoir

Without being able to resolve the anisotropic properties of the reservoir, we decided

to estimate reservoir properties using an equivalent media description of fractures

(O’Connell and Budiansky, 1974). This model is appropriate for our situation of

long wavelengths (about 100 m) compared to fracture size (presumably 0.1 to 1 m)

and isotropic wave propagation. In this model, the effect of thin randomly oriented

ellipsoidal cracks with variable saturation on material properties is calculated as though

there is a media having equivalent properties to the fractured solid. The crack density,

q is defined by e = (21V/7r)(A2/P), where N = cracks per unit volume, A is the area of

a crack, and P is the perimeter. This relationship for a partially saturated cracked rock

,

is given by45 (v – U) (2 -Z)

‘= fi(l – v2) [(1 –f)(l+3v)(2– ~) – 2(1 –2v)]

where ~ = the crack density,

v = Poisson’s ratio for the untracked matrix,

P = Poisson’s ratio for the cracked rock, and

~ = saturated fraction of cracks.

14

(1)

In the NIPSCO data, we used

Vp = 15,700 ft/s and Vs = 9600 ft/s for gas saturated conditions, and

Vp = 17,200 and Vs = 9,700 ft/s for water saturated conditions.

These velocities come fkom the 72 ft interval velocity centered at 1016 ft. Poisson’s ratio

for an untracked dolomite (v= 0.33) was taken from Carmichael, 1982.

For our analysis of the NIPSCO data, this analysis leads to a trade off between the

crack density and the crack saturation (the ilaction of cracks water saturated). Figure

12 shows the trade off curve of crack density and saturation for the Trenton reservoir

calculated for the velocity variations seen in the walkaway survey. These two curves

represent the changes in reservoir properties at well 157. For any given crack density,

the difference between the two curves represents the percentage of fractures whose water

is displaced by gas. For a reservoir with higher fracture density, fewer cracks need to

be gas saturated to give the velocity changes observed in the VSP. If other information

about crack density becomes available (for instance from core measurements) the curves

in Figure 12 can be used. To provide a more physical interpretation of crack density, we

can assume a circular fracture of 1 m2 area. We then get the curves shown in Figure 13

which has fractures per unit volume instead of dimensionless crack density.

,

.+ .. ——. ---- .=-. .

15

5.1. Equivalent Porosity Estimate

Because there is not strong evidence of fracture induced wave propagation effects

within the reservoir for Site 1 data (near vertical propagation), we believed it would

be instructive to use a matrix porosity model of the time-lapse changes observed in

the walkaway VSP. We used an approximation to Gassmann’s relations (Gassmann,

1951) proposed by Mavko, et al., 1998 for data with P-wave velocity (VP). They use the

modulus M! = pVP. For a system with two fluids saturating a matrix, we use:

IWO= matrix rock modulus,

A4tll = fluid 1 modulus,

~fzz = fluid 2 modulus,

~satl = modulus of rock saturated with fluid 1,

~~a~z= modulus of rock saturated with fluid 2.

For the case of two fluids displacing the same porosity, such as the gas displacing water

in the Trenton reservoir, we can state the porosity @ as follows:

Mf 11 M 12

~ = (Mi-Mm) – (Mcj%) (2)atl sat2

(MO-M.=,,) – (MO-M.=,2)

We use water for fluid 1 (p = 1.0, VP = 1531 m/s), and methane gas for fluid 2 (p =

0.34, Vp = 430 m/s). Evaluation of this equation gives a porosity of 0.32.

It is interesting to note that evaluation of the crack density plots in Figure 12 for

a differential saturated fraction of 3270 gives a crack density of c = 0.27. However, this

should not be considered a true estimate fracture density, because Gassmann’s relations

and the O ‘Cormel and Budiansky relationships have different conceptual models of rocks

and can not in theory be combined. It would be much preferable to maintain a crack

based model of the reservoir and use other information to estimate the fracture density

or the partial saturation change.

, I

16

6. Offset Site Data - Possible Anisotropy

The site 3 and 4 S-wave data were analyzed by comparing the S1 and S2 source

polarizations at each site. Since sites 3 and 4 are essentially 90 degrees apart in azimuth,

the S1 and S2 sources have opposite azimuthal polarization at each site. That is, the S1

source at site 3 has the same azimuthal polarization as the S2 source at site 4, while

the S2 source at site 3 has the same azimuthal polarization as the S1 source at site 4.

We therefore expect reversed relative time lapse change when comparing S1 and S2

for sites 3 and 4, if the reservoir has an aligned fracture set which is not on a North-

South or East-West azimuth. Figure 14 shows that we do observe such reversed time

lapse changes, and both sites have approximately the same magnitude of change. For

site 3, S2 (North/South polarized) data has about 1.5 ms more time lapse change.

For site 4, S1 (North/South polarized) has about 1.5 ms more time lapse change.

Since the North/South polarized source shows the largest change at both sites, we can

propose a conceptual model of a reservoir with vertical fractures having a predornimmtly

East/West azimuth. This is because shear waves polarized normal to the fracture strike

azimuth will be slowed compared to shear waves polarized parallel to the ikacture

azimuth. Unfortunately, the data do not have a spatially constrained time-lapse change.

We can not resolve a change in the reservoir zone ( 975 to 1025 ft.). The increasing

time-lapse change seen below 850 ft. may be related to vertical fracturing extending

above the reservoir, however this is not consistent with the vertical propagating waves

at site 1.

7. Other Analysis

There is remtig analysis beyond the scope of this report which holds promise for

fracture and gas detection using reflection and mode-conversion properties. In the phase

1 data we observed a strong reflection from the Trenton formation and a strong mode

. . .. . .. ~,, —, ,,, -.,J~,, ,,.a..,..,,,,..,,<,,.............>.,,,,,.,,,.,,..,..t ....IA.. ,. h.,. .-. . .,. ,. . .. .,: .:’.., . .,.. .4! “-,—-,-” -

17

conversion (Figure 7). The mode conversion seems associated with the Maysville and

Eden formation contact, while the S-wave reflection seems associated with the Trenton

formation and may be affected by reservoir gas saturation. Initial investigation of

zero-offset (site 1) P-wave reflectivity, using F-K separation of upgoing and downgoing

energy, did not yield any clear time-lapse changes (see plots in Appendix D). Frequency

vs time analysis of the walkaway P-wave also yielded no clear time-lapse changes (plots

in Appendix D).

8. Surface Seismic Analysis

As part of the FETC fractured gas research program, LBNL was asked to study

2-D surface seismic data which had been interpreted using Trenton formation reflection

amplitude anomalies as representative of gas saturation. We also thought it would be

instructive to compare two crossing surface lines which intersect near the location of

well 157. Figure E-1 shows the location of lines RC-1 and RC-5 along with many of the

wells in the Royal Center field. The VSP well, S:157-T, is at the intersection of these

two lines. Figure E-2 shows CDP gathers from these two lines in the vicinity of well

157. The Trenton formation reflection (top of Trenton) is at about 0.17 s time. Figure

E-3 shows the CDP stacked sections from lines RC-1 and RC-5 with the VSP stacked

upgoing reflections inserted in the middle. Again the Trenton reflection is at about 0.17

s. We see that the reflection wavelet for the VSP is quite different with earlier energy.

The two surface lines are similar with RC-1 having reduced side lobes (red) compared to

the main peak (blue). The bottom of Trenton reflection ( 0.2 s) is quite similar on both

surface lines and the VSP, indicating little effect from the reservoir which presumably

varies laterally away from the VSP well. Our reprocessing of these two seismic lines did

not show the strong lateral variation in the Trenton

interpreted as lateral variation in fracturing.

The possibility of using AVO techniques in the

reflection wavelet which had been

surface seismic was investigated,

18

however the velocity structure

be obtained. Figure E4shows

Trenton reflection, and we can

is a strong limit on the actual incident angles which can

modeling of the surface seismic cdp gathers for the top of

see that there is very limited angular coverage. Therefor

we have concluded that the Trenton is not a good candidate for AVO studies.

9. Summary and Conclusions

Underground storage of natural gas with seasonal injection and withdrawal

provides an excellent opportunity to study time-lapse changes in reservoir properties. In

particular, the NIPSCO site is attractive because water displaces gas within the reservoir,

thus maximizing the changes in material properties and elastic wave propagation. We

designed and acquired a time-lapse VSP study aimed at characterizing the reservoir

properties using the spatial and temporal variations in seismic wave propagation.

Unfortunately, the study was initially compromised by surface access restrictions

which limited the scope of data acquisition, and by questions about the amplitude

equivalence of two shear-wave sources. Nonetheless, multiple 9-C VSPS were collected

while the reservoir was gas saturated and repeated with the reservoir water saturated.

The velocity structure of the RoyaI Center site included a strong velocity inversion

which greatly limited the volume of the reservoir which could be probed with the VSP

method. The near offset VSP (site 1) was analyzed for shear-wave splitting (under the

assumption of verticaI fracturing with a dominant orientation), but no definitive result

could be obtained. This is partially due to problems with shear source amplitudes, but

the travel time analysis also found no splitting within the reservoir horizon (and within

the resolution of the data). We believe the reservoir is too thin for the wavelengths

obtainable with the VSP method in this fast velocity material (dolomite). The far offset

source sites did show some shear wave splitting, and a time-lapse change in velocity

‘was observed in the P-wave walkaway. The P-wave walkaway data could be constrained

to time-lapse change in velocity within the reservoir horizon. This result was analyzed

. . ... —-....,-

19

in terms of fracture density and partial crack saturation using the long wavelength,

equivalent media approach. ‘Tradeoff curves between crack density and saturation were

generated for the gas saturated and water saturated conditions. The difference between

these two curves represents the percentage of the fracture space utilized for gas storage

for a given fracture density.

Other studies included variation in near-offset P-wave reflections (no changes

observed) and frequency vs time analysis of the walkaway data (no clear time-lapse

change). We also conducted an analysis of surface seismic data acquired near the VSP

well. We found no difference in the Trenton reservoir between north-south and east-west

reflection lines. We also used our ray tracing model based on VSP velocities to study the

CMP reflection gathers. We found that, like the offset VSP surveys, the CMP gathers

had very little variation in angle of incidence of rays. This means the AVO analysis

would not be useful for the Trenton reservoir. Again like the VSP surveys, the shallow

velocity inversion with large contrasts is preventing lateral sampling of the Trenton at

any reflection point.

In conclusion, the Trenton reservoir has good potential for assessing the effects

of varying gas and water saturation. However, surface seismic sources will be limited

in their ability to image the effects because of the velocity structure and the relative

thinness of the reservoir. We believe that borehole source methods, such as crosswell

or singlewell imaging, provide the best opportunity to study the seismic response by

avoiding the shallow velocity inversions and providing higher frequency data.

20

10. Acknowledgements

This work was supported by the Assistant Secretary for Fossil Energy, Federal

Energy Technology Center (FETC). Processing was performed at the Center for

Computational Seismology, which is supported by the Director, Office of Science,

Office of Basic Energy Sciences, Division of Engineering and Geosciences, of the U.S.

Department of Energy under Contract No. DE-AC03-76SFOO098. We would like to

thank FETC project manager Thomas Mroz for his support, and Jim Crispman of

NIPSCO for his help with field information and access. We would like to thank the

LBNL field crews for both data acquisition sessions, especially Don Lippert, Cecil

HofEpaiur,Ken Williams, Jim Galvin and John Beck.

,.

21

Figure 1. Geologic column for the Royal Center field. The approximate depths of major

units at the VSP well 157 are: Glacial Till 0-150 ft., Silurian limestone and shale 150-675

ft., Eden shale 675-930 ft., Trenton dolomite 930-1100 ft.

Figure 2.

Figure 3.

Figure 4.

Schematic objectives of NIPSCO VSP.

Original design concept for NIPSCO VSP.

Seismic source locations for NIPSCO VSP data acquisition. Field conditions

and local access laws prevented locations as in original concept (Figure 2).

Figure 5. Sixty ft. lubricator pipe used to introduce multi-level sensor

pressurizedVSP well.

/ I

string in the

Figure 6. Uncorrelated vibroseis data showing effects of automatic noise burst editor

before (left) and after (right) edits.

Figure 7a. Seismograms from orthogonal S-wave experiment at site 1. The left side

data set is in-line horizontal oriented sensors recording an in-line oriented S-wave source

over the depth range 296 ft. to 1088 ft. The right side data set is cross-line horizontal

oriented sensors recording a cross-line oriented S-wave source over the same depth range.

These data sets are used for estimating shear wave anisotropy which can be controlled

by subsurface fracturing.

Figure 7b. Four orthogonal S-wave recordings for depths 936 to 1088 ft (the approximate

reservoir zone). Top-left is in-line source and in-line sensor (XX), top-right is in-line

source and cross-line sensor (XY), bottom-left is cross-line source and in-line sensor

(YX), bottom-right is cross-line source and cross-line sensor (yY). The seismograms are

normalized to the maximum of each source.

Figure 8a. Time lapse

(gas saturated reservoir)

change in

minus the

P-wave travel time for site 1. Times are 1996 data

1997 times (water saturated reservoir).

22

Figure 8b. S-wave splitting in terms of travel time difference (S1 (in-line, east-

west) source minus S2 (cross-line, north-south) source for site 1 in 1996 (gas saturated

reservoir).

Figure 8c. S-wave splitting in terms of travel time difference (S1 (in-line, east-west)

source minus S2 (cross-line, north-south) source for site 1 in 1997 (water saturated

reservoir).

Figure 8d. Time lapse change in S-wave travel time for site 1. Times are 1996 data,

(gas saturated reservoir) minus the 1997 times (water saturated reservoir) for S1 (in-line,

east-west) source and S2 (cross-line, north-south) source.

Figure 9. Comparison of S-wave hodograms (particle motion) for in-line (east-west)

source polarization (left) and cross-line (north-south) source polarization (right) for three

depths within or below the reservoir. The S-wave polarization is dominantly isotropic

with some ellipticity and a moderate east-west component on the cross-line (north-south)

source data (right).

Figure lOa. Ray tracing for P-wave walkaway survey. Note that very little propagation

occurs in the Trenton formation. Velocities are from the 1997 site 1 VSP modified to

better match the walkaway travel times.

——— - ..-.—, -1

23

Figure 10b. Close-up view of raypaths within the reservoir zone of the Trenton. A

conceptual model of an unfractured Trenton cap with minimal fracturing above a highly

fractured Trenton reservoir was applied to estimate the raylengths within the reservoir.

These raylengths were used in the fracture density and saturation estimates for the

reservoir. The five sensor locations are the walkaway sensor depths.

Figure 11. Time lapse changes in P-wave travel time for top and bottom walkaway

sensors. The sensors were not moved between each station recording. Times were picked

on the unrotated vertical component. The station spacing was 50 ft. The difference

between these two curves provides the estimate of time-lapse velocity changes between

gas filled and water filled reservoir conditions using the raylengths estimated raytracing

(Figure 10). The average difference for all stations between the 1016 and 984 ft sensors

is 0.2 ms with a standard deviation of 0.1 ms.

Figure 12. Trade off curves for dimensionless crack density c and saturation @ for the

two VSP surveys. The difference between the curves for a given ~ represents the partial

saturation used for gas storage.

Figure 13. Trade off curves for volumetric crack density and saturation ~ for the two

VSP surveys using a 1 m2 circular fracture. The difference between the curves for a given

crack volume represents the partial saturation used for gas storage.

Figure 14. Time-lapse changes in S-wave travel time between 1996 (gas saturated

reservoir) and 1997 (water saturated reservoir) VSP surveys. The north-south polarized

S sources have larger time changes than the east-west polarized sources.

Appendix A: VSP Data Plots

-.—.- ,. -. . . .. ... .... . . ........ . . .. .... .. ,. —.—. .. . . -,,..

25

Figure Al. Cross correlation oftwo Peltongenerated sweeps showing 85ms delay due

to varying trigger time shift in 1996 data set.

Figure A2. Cross correlation of two data traces with a 100 ms time shift added to show

full wavelet. 85 ms delay from figure Al has been removed.

Figure A3a - A3c. Seismograms from Well 46, 1996. The three

components horizontal in-line (left), horizontal cross-line (center),

The traces have been rotated (using P-wave particle motion) and

panels are geophone

and vertical (right).

are scaled using the

maximum of all three panels. A3a is P-wave source, A3b is S-wave in-line source, A3c is

S-wave cross-line source.

Figure A4a - A4d. Seismograms from Well

are geophone components horizontal in-line

157, site 1, 1996 survey. The three panels

(left), horizontal cross-line (center), and

vertical (right). The traces have been rotated (using P-wave particle motion) and are

scaled using the maximum of each trace. A3a is P-wave source at site 1, A3b is S-wave

in-line source, A3c is S-wave cross-line. A4d is P-wave source at site 2 (used for rotation

analysis of site 1 data).

Figure A5a - A5c. Seismograms from Well 157, site 2, 1996 survey. The three panels are

geophone components horizontal in-line

(right). The traces have been rotated

using the mtium of each trace. 5a -

cross-line, respectively.

(left), horizontal cross-line (center), and vertical

(using P-wave particle motion) and are scaled

A5c are the P-wave, S-wave in-line, and S-wave

Figure A6. Seismograms from Well 157, site 3, 1996 survey. The three panels are

geophone components horizontal in-line (left), horizontal cross-line (center), and vertical

(right). The traces have been rotated (using P-wave particle motion). Left panels are

S-wave cross-line source, right panels are S-wave cross-line (each scaled to maximum of

three components).

26

Figure A7a - A7c. Seismograms from Well 157, site 4,1996 survey. The three panels are

geophone components horizontal in-line (left), horizontal cross-line (center), and verticaJ

(right). The traces have been rotated (using P-wave particle motion) and are scaled

using the maximum of each trace. A7a - A7c are the P-wave, S-wave in-line, and S-wave

cross-line, respectively.

Figure A8a - A8f. Seismograms from Well 157, walkaway, 1996 survey. The 5 panels

are geophone depths 984, 992, 1000, 1008 and 1016 ft. left to right. The 27 traces per

panel are the 27 offset locations (50 ft. spacint). The traces are scaled to the maximum

of each section except where noted. A8a-A8c are P-wave source, geophone components

vertical (trace scaling), unrotated horizontal 1 (trace scaling), and unrotated horizontzd

2. A8d-A8f are S-wave cross-line source, vertical, unrotated horizontal 1, and u.nrotated

horizontal 2.

Figure A9. Seismograms fkom Well 157, site 6 P-wave source, 1996 survey. The three

panels are horizontal in-line (left), horizontal cross-line (center) and vertical (right). The

traces are normalized to their maximum.

..-. .—,.-_— .,,. ., ..,., \_. .... . . .,,, .,. .......... . ... .-+. .— - ..—

27

Appendix B: Travel Time, Velocity and Rotation Angle Tables

and Plots

28

Figure B1. Rotation angles for 1996 site 1.

Figure B2. Travel times for 1996 site 1 without low pass filter.

Figure B3. Travel times for 1996 site 1 with low pass filter.

Figure B4. Average velocities for 1996 site 1.

Figure B5. Interval velocities for 1996 site 1, with an 80 interval.

Figure B6. Travel times for 1996 site 3. The columns are depth (ft.), P time, S1

(in-line) time and S2 (cross-line) time.

Figure B7. Travel times for 1996 site 4. The columns are depth (ft.), P time, S1

(in-line) time and S2 (cross-line) time.

Figure B8. Travel times for 1996 site 6. The columns are depth (ft.), P time.

Figure B9. Plot of P-wave interval velocities for 1996 site 1 using a 40 ft. interval.

Figure B1O. Plot of S-wave (in-line source) interval velocities for 1996 site 1 using a 40

ft. interval.

Figure B1l. Plot of S-wave (cross-line source) interval velocities for 1996 site 1 using a

40 ft. interval.

—r——,.—. ,....,.......... ... ........,.,’,,,,, >,, %. ..-, - ,. . . . . . -,. > , . . 4>>. ... ?.. .. ... . -,,-+c.~ ..,.,.J.J*SI.3,., ,., .- . .,.,~.. -- . . . ..’ ! . ----—- ‘- “’ -

29

Appendix C: Time Lapse and Travel Time Difference Plots

30

Figure Cla. Time-lapse change in P-wave travel time at site 1.

Figure Clb. Time-lapse change in S-wave (in-line source) travel time at site 1.

Figure Clc. Time-lapse change in S-wave (cross-line source) travel time at site 1.

Figure Cld. Travel time difference between S1 (in-line, E-W) and S2 (cross-line, N-S)

at site 1 for 1996 and 1997 surveys.

Figure C2a. Time-lapse change in P-wave travel time at site 2.

Figure C2b. Time-lapse change in S-wave travel time at site 2 for S1 (in-line, EW)

and S2 (cross-line, N-S) for 1996 minus 1997.

Figure C2C. Travel time difference between S1 (in-line, EW) and S2 (cross-line, N-S)

at site 2 for 1996 and 1997 surveys.

Figure C3a. Time-lapse change in P-wave travel time at site 3.

.. . . . .. .--.-, ,... . . ... ... .... .---,-TJZ ., . . . . .. . . .. .. ..,. ..l, . . ,. . . ,, . .,., ,, , . —.. .,

31

Figure C3b. Time-lapse change in travel time at site 3 for S1 (in-line, E-W) and S2

(cross-line, N-S).

Figure C3C. Travel time difference between S2 (cross-line) and Sl(in-line) for 1996 and

1997 surveys.

Figure C3d. Time-lapse travel time difference for site 3, S2 minus S1 and 1996 minus

1997.

Figure C4a. Time-lapse change in P-wave travel time at site 4.

Figure C4b. Time-lapse change in travel time at site 4 for S1 (in-line, E-W) and S2

(cross-line, N-S).

Figure C4C. Travel time difference between S2 (cross-line) and Sl(in-line) for 1996 and

1997 surveys at site 4.

Figure C4d. Time-lapse travel time difference for site 4, S2 minus S1 and 1996 minus

1997.

Figure C5a. Time lapse P-wave time change for walkaway survey for two sensors (984

and 1016’).

Figure C5b. Time lapse and spatial change for P-wave walkaway.

Figure C5C. Time lapse S-wave (cross-line source) time change for walkaway survey for

two sensors (984 and 1016’).

Figure C5d. Time lapse and spatial change for S-wave (cross-line source) time change

for walkaway survey for two sensors (984 and 1016’).

Figure C6. Time lapse P-wave time change for site 6 (1996 - 1997).

32

Appendix D: Other Studies and Analysis

33

Figure Dia. Particle motion for comparison for 1997 site 1 S-waves. S1 (in-line, E-W)

source (left side) and S2 (cross-line, N-S) source (right side). Level number is sequential

from shallowest (96 ft.).

Figure Dlb. Particle motion for comparison for 1997 site 1 S-waves. S1 (in-line, E-W)

source (left side) and S2 (cross-line, N-S) source (right side). Level number is sequential

from shallowest (96 ft.).

Figure D2a. Upgoing data fkom 1996 site 1 P-wave source after F-K filter.

Figure D2b. Upgoing data from 1997 site 1 P-wave source after F-K filter.

Figure D3a. Frequency vs time (top) and spectra (bottom) plot for average of two

farthest 1996 walkaway P-wave sites for sensor at 984 ft.

Figure D3b. Frequency vs time (top) and spectra (bottom) plot for average of two

farthest 1996 walkaway P-wave sites for sensor at 1016 ft.

Figure D3c. Frequency vs time (top) and spectra (bottom) plot for average of two

farthest 1997 walkaway P-wave sites for sensor at 984 ft.

Figure D3b. Frequency vs time (top) and spectra (bottom) plot for average of two

farthest 1997 walkaway P-wave sites for sensor at 1016 ft.

34

Appendix E - Surface Seismic Analysis

-.. .-. — . . .. .,. , . .. . .. ..... . ..... ... . . .. .. . —-- . —.....!.. ,,. ,

35

Figure El. Location map of some surface seismic lines and wells in the Royal Center

field. The surface lines analyzed are RC-1 and R.C-5 which intersect near VSP well 157.

Figure E-2. Example CDP gathers from surface line RC-5 (left) and RC-1 (right) near

the intersection of the lines. The reflector at about 0.17 s is the top of the Trenton

reservoir formation.

Figure E-3. CDP stacked sections for surface lines RC-1 (left, CDPS 576- 596) and

RD-5 (right, CDPS 618- 638) with the well 157 site 1 VSP P-wave upgoing stack displayed

in the middle (labeled as CDP 1). The top of the Trenton reservoir formation is at about

0.18 S.

Figure E-4. Ray trace modeling of surface seismic CDP gather at Royal Center field

using the velocity model derived from the well 157 VSP. The lack of variation in reflection

angle of incidence at the Trenton reservoir formation (about 970 ft.) shows the limitations

on any AVO analysis.

36

References

Alford, R.M., 1986. Shear data inthepresence ofaztiuthd tisotropy, 56th A.nn.,... Internat. Mtg., Sot. Explor. Geophys., Expanded Abstracts.

Carmichael, R.S., Handbook of PhysicaJ Properties of Roch, Volme II, CRC Press,

hlC., 1982.

Crampin, S., Bush, I., Naville, C and Taylor, D., 1986. Estimatfig the ~ternal StrUCtI-Ue

of reservoirs with shear-wave VSPS, Leading Edge, 5, 35-39.

Daley, T. M., T.V. McEvilly, and E.L. Majer, 1988. Analysis of P and S Wave Vertical

Seismic Profile Data From the Salton Sea Scientific Drilling Project, J. Geophys.

Res., .R 93, Bll, p13025-13036.

Daley, T.M. and T.V. McEviUy, Shear wave anisotropy in the Parkfield Varian Well

VSP, Bull. Seisrn. Sot. Am., 80, 857-869, 1990.

Gassmann, F., 1951. Uber die Elastizitat poroser Medien. Vier. der Natur. GesWchaj3

in Zurich., 96, pi-23.

Jech, J., and Psencik, I., 1992. Kinematic inversion for qP- and qS-waves in homogeneous

hexagonally symmetric structures, Geophys. J. In-t., 108, 604-612.

Liu, E., Crampin,S. and Queen, J.H. and Rizer, W. D., 1993, Velocity and attenuation

anisotropy caused by micro cracks and macrofractures from multi-azimuthal

reverse VSPS, Canadian Journal of Exploration Geophysicists, 29, 177-188.

Majer, E. L., McEvilly T. V., Eastwood F., and Myer L., 1988, Fracture detection

using P-wave and S-wave vertical seismic profiling at the Geysers, Geophysics,

53, 76-84.

Mavko, G., Mukerji, T., Dvorkin, J., 1998, The rock physics handbook Tools for seismic

analysis in porous media, Cambridge University Press.

4

-.—: ,,, ,,, ,,, ,,~i .,. , “.,,. +...! ; .. .. ,.... -.’! . . ..>. <->,..-:.;, -, . . ?. -. . . ...-+.... .. . F ... ..4s., .,?. f.:.>? ,, -..< .!’,,- -. . . . . ,;H7 > .-.,. . ... .—

:%

37

O'Connell., R.J., and Budiansky, B., Seismic Velocities in Dryand Saturated Cracked

Solids, Journal of Geophysical Research, 79, n35, p5412-5426.

Pyrak-Nolte, L.J., Myer, L.R., and Cook, N.G.W, 1990. Anisotropy in Seismic Velocities

and Amplitudes from Multiple Parallel Fractures, Journal of Geophysical

Research, 95, nB7, pl1345-11358.

Schoenberg, M., and Sayers, C.M., Seismic Anisotropy of fractured rock, Geophysics,

60, nl, p204-211.

Thomsen, L., 1988. Reflection seismology over azimuthally anisotropic media,

Geophysics, 53, 304-313.

Winterstein, D.F., and Meadows, M.A., 1991. Shear-wave polarizations and subsur face

stress directions at Lost Hills field, Geophysics, 56, 1331-1348.

This manuscript was prepared with the AGU 1#~ macros v3.O.

STRATIGRAPHIC - COLUMN.

ROYAL CENTER FIELD

: GROUP OR$ FORMATIONm :

Y-%1 MISSISSIMWA

W=== I WAl,DROM

I

I MAYSVtLtX

I TRCMTOM

z~g l-- .—~1Yi BLACK

$

Rlvcfi

I

COL-‘CG CURVE UMN

fCAMMA RAW I

-+----+

I-Hlc

1NESSLITHOLOGY

bo’-w MFT-wl OnAvu.

2ff-mri 00L0WTXUMC2TO% CK.RTY.

12d-w WAL2J UMCITOMC.

--lIs-w OoLowmc uu221ti

m-m’ LIUESTOWI SHAE.

W-111$ Llutxlornc.

TI+ Llh12iTCWa ~.

\ t-3’ WALE.

sd-12d LIMIX70MFA W4AL21 ~

2W-2 1 SHAM.

~Isd-l?a’ 00 L0MIIC

l--tI,Td-ttd] L,b12~, 0W2UM w., .

&E====uJ+2ft OoLwlm SAxOY aitmY.

Figure 1. Geologic column for the Royal Center field. The approximate depths of major

units at the VSP well 157 are: Glacial Till 0-150 ft., Silurian limestone and shale 150-675

ft., Eden shale 675-930 ft., Trenton dolomite 930-1100 ft.

----

. .

NIPSCO VSP Experiment Objectives

Objective 1: Determine fracture orientation from s-wave splitting.

output

S-wave

Objective 2: Estimate fracture location from winter/summer

changes in P & S wave attributes

I FractureZone

,

Figure 2. Schematic objectives of NIPSCO VSP.

.

NIPSCO VSP DESIGN

Concept - Equal Offset Distance, Multiple AzimuthsAcquire atAcquire at

●

Productive and Non-Productive WellsMaximum and Minimum Gas Pressures

●e

●

-1500’

● = 9-Component VSP Source Location

Figure 3. Original design concept for NIPSCO VSP.

NIPSCO VSP Source Locations

I

O P, S1, S2at Well Pad (only for 12/96 survey)

T— Well 460

-5000’

w

Site4

0 P, S1, S2

1250’ I~ P, S1, S2 at Well Pad P & S1 every 50’

0 Site 1 1— Well 157@’

T 300’ ~ ~

o 1500’ )() Site3

640’ P, s-1, S2 P, S1 , S2

c)P Site2

2750’

w ~ p (and S1 in 5/97 survey)

Site6

Figure 4. Seismic source locations for NIPSCO VSP data acquisition. Field conditions

and local access laws prevented locations as in original concept (Figure 2). I

, I

Figure 5. Sixty ft. lubricator pipe used to introduce multi-level sensor string in the

pressurizedVSP well.

. .... . . ........ — ... . ....:. .. .-.’..! ! <... -. . .?..

of noise bursts in uncorrelated data(left) and after automatic editing (right)

L7ncorrelated1s1source14

— .——... —

AfterREMSPK14FFID

Cn?.11

7

7

7d

8

8

8

8

8

8

8

8

8\Time,(see]

8—.—

9 FFIDCml

1,

1.

7.

8.

8.

8.

8.

8.

8.

8.

8.

8.pnJ

8.=—.. _____

9

.—. —_

3 9 s

.—. ——.— ——— n

.-

70-

80-

90-

oo-

10-.

20-.

30-.

40-.

50-.

60-.

70--

80--

90--=

I

—

,

Figu .re 6. Uncorrelated vibroseis data showing effects of automatic noise burst editor

before (left) and after (right) edits

157-1 S1 source Rot. Hi-sRSTATRCN 1

157-1 S2 Source Rot. H2“m.1...mn#n,.-)

Rum - :.2;6 3$8 440 S;2 S14 t@ 728 800 8?2 944 1016 10BE-..M,ZL.N <

P.UEV 296 368 440 S12 S44 6S6 72S 800 872 944 1016 108B

-EiR!ionl-

1- 1#.— .. . ,---- —.. . -.

Figure 7a. Seismograms from orthogonal S-wave experimental site 1. The left side,.

data set is in-line horizontal oriented sensors recording an in-line oriented S-wave source.

over the depth range 296 ft. to 1088 ft. The right side data set is cross-line horizontalI

1 oriented sensors recording a cross-line oriented S-wave source over the same depth range.

These data sets are used for estimating shear wave anisotropy which can be controlled,,,

by subsurface fracturing.

. ...

4.,.

Exit File ~lsplay ~nalysls FJck ~elrt,. ,

RsTA’rRcN 1Rcsmr

0.00

0.10

0<20

0.30

T&?

RSTATRCN 1RCS3!AT 936 952

0.00

O.lo------------------

TJi!3

968 904 1

I..>-..[

. . .

. .

. . .

1 10-48 10

[

.. ..-.

.- ...,

.. .. .

1S1 ROT-S.

1080

1S2 ROT-S2

016 lC” II

[;

1

.. . .. .

.. . .. .

... . .. .

1064 10BO

i

. . .. . ..

. . .. . ..

. . .. . ..

0.00

----------------.0.10

-----------------0.20

.---------------- .0.30

104 1064 10800.00

-.........-...------------------.0.10

---------------..0.20

}

--------------.--.0,30

liEmmm -

LMouse Button 1: Select section.

Mouse Button 2“ Deselect section.

Mouse Button 3: Display section POPUP menu.

.-— — !.. —

1%ctlm M.trltutos Pmd

Quick Draw Control:

‘“”t &s-lin9 SectIon~l

Display Attributes: scaling:1

Display Typo~ .—

JAuto Deblas] rEnsem

Q!wJ+X!wd

Gain (db)F

I AGr Window lim-,“.-!.!... l-’.4.-..!......— — .——

Figure 7b. l?ourorthogonal S-wave recordings for depths 936 to 1088 ft (the approximate

reservoir zone). Top-left is in-line source and in-line sensor (XX), top-right is in-line

source and cross-line

(YX), bottom-right is

nor___zed . . -he I-.

sensor (XY), bottom-left is cross-line source and in-line sensor

cross-line source and cross-line sensor (YY). The seismograms are

mul.. .- eac flrce

/

‘)

NIPSCO VSPSite 1 (100’)

P Time Difference (96-97)

o

200

400:

1000

‘)1200

>

-.—.. . -

0 2 4 6 8 10

Time (ins)

Figure 8a. Time lapse change in P-wave travel time forsitel. Times are1996 data

(gas saturated reservoir) minus the 1997 times (water saturated reservoir).

o

200

400

$$ 600

Q:

800

1000

1200

NIPSCO VSP

Site 1 1996 S-Wave Splitting

S1 (SH, E-W) - S2 (SH, N-S)

I ---B--S1 - S2[

I I I 1 I I I 1 I I I 1 I 1 I 1 I 1 I (

01234567 89

S1 - S2 (ins)

,

Figure 8b. S-wave splitting in terms of travel time difference (S1 (in-line, east-

west) source minus S2 (cross-line, north-south) source for site 1 in 1996 (gas saturated

reservoir).

o

200

400

zw

800

1000

1200

0

May 1997 Site 1

S–Wave Time Difference

40 Hz LP Filtered

.

1 1 t I

2 4 6 8 10

S1-S2 (ins)

Figure 8c. $wavesplittingi ntermsoft ravel time difference(sl (in-line, east-west)

source minus S2 (cross-line, north-south) source for site I in 1997 (water saturated

reservoir).

o

200

NIPSCO VSPSite 1 (100’)

S Time-Lapse Difference (96-97)

4

+ S1 (SH, E-W)+ S2 (SH, N-S)

400

600 –

800 –

1000

-5 -3 -1 1 3 5

Time (ins)

Figure 8d. Time lapse changein S-wave travel time for site 1. Times are 1996 data

(gas saturated reservoir) minus the 1997 times (water saturated reservoir) for S1 (in-line,

east-west) source and S2 (cross-line, north-south) source.

,

NIPSCO VSP Site 11997

S1 - SH (E-W) S1 - SH (N-S), , , .;,.;, ,,,,- , ,;.,;,,$,,,

‘Iti”gmktck%kiiwl

1....,.’~

<

I:..:,,,,**,

,;~.p:>“,

LEvElJ2~

a

LWL120

7--%s’z I

d7

Figure 9. Comparison of S-wave hodograms (particle motion) for in-line (east-west)

source polarization (left) and cross-line (north-south) source polarization (right) for three

depths within or below the reservoir. The S-wave polarization is dominantly isotropic

with some ellipticity and a moderate east-west component on the cross-line (north-south)

l-l$

source data (right).—..- —. . .. .... .. ... . .. . .—. -

0.

200

D 400

h

( 600.

f

) 800-

iI looo -

I

1182-

2 4 :1

0 0 ; :; 4

0 0 0 00

00 0

I...f oI I o 0I 1 I I

,1 = .,...-... s-, ,, -,. .,,,- .,, .. >.,

Y Mays-aF’ -..’: .--” .. ’’-:”:-, ‘.: . . . . - .- ~~ . . —

. . . .. ,., . . . . .. ,,, .,.< ,. .,= ,.,:,...... f.,>...: -,..,=,- ..., .; ,.,:: ~ J.. ., ‘ , -r.:,,.-4,, . .{. .-- r . . . .

,., ...

. . . . . . . . . ,.,: -’.--” . ~, -..’.... .,

r .l’- “-“. “- .“ -: .,: “,, .’,+. . . - ; .’.=. .:” :;:, ,. ,:, .- ,,(; , . . .,/-., ,, ,.

. . . .,: ..,- -,:’- ‘ :., .

,:. -. ..~. .;t c.. ,’,. . . . . ,4 .,, .,,:. ,“., ..,,.::.”,7:,-,(’ .,, ,, :-,” .-’ ,,\ .,,,,- ,.:,,. .,. .,,,, .-: ,. .,: ;-:.--,:s .>:,,. “., . . . . . .~.

,,,,. ‘,.,..

,? --- -.. . . , ,’:: ,’ ::.,. . ,. ..>: ‘~,’’.+;;,, .:,;’. ,., ;T@nmiv-b ‘“,.:,..,,.-’ .“. :,.., .-. , :’” ,.. ”

L- ,. . . ,<, ,: .:;”--- , . .-. .,.-’;.:- ~,.,:, ,,. ,...f, . .,...:-;,v. , ,: +“ ,,’.- .- ,,,;.,:>. .,~,.:. “.,,,,...,.,.,.,,-........<,:,., ~..,,,,,:.-., ,.w -.>., .. :.,,,.,,.; -,::.,:,- ,:- ..’ ,,-/..,,” -:, ,.,:...- :, .!.. ... . . -,. .,.: ,., -

,., .,,. ---:’ ,,., ,,:’

.. . ,! ..’. .;- ,;., ,-, . .. .,: -,J,.’’--,,:, ,., ..,, .-~”.,: ,;. .s..,~, ~,. ~.,.:. ,.-: .”’ ,.. - .: ..,.,..,,. ..,, ,,, ., ~.,(,.-,,. ,,,, ,s,:... .“,. :, ,. ..,- ., .”.,

=6:.: ,,-... .’..: :,’

.-, .-.: T :. y.<,.. . .. . . , .,L, ,.. . . ,,,. ,,. . . . . . . . . ‘,(,’,.+-;..” ,. . :.3’.”,:’.-, ;: -. . ...,,, :’”-,. ‘ ;?,

. .- !, : ,“ .,, . .. , -.-, ,.< ,. J,. .- . . . . . . . . ,., ;.. -., .-. ,, . -. ~. ,,:. ,:., , ;, . .,,

file: zero–5.97B, date: Wed Dec,,. ,,

1 15:57:45 1999, user: tomd, host: Ccs

Figure lOa. Ray tracing for P-wavewalkaway survey. Note that very little propagation

occurs in the Trenton formation. Velocities are from the 1997 site 1 VSP modified to

better match the walkaway travel times.

,

1-898=

925-

950-

;D

:; 975-

+h

:( looo- i

f;e

“t 1025-‘)

: lo50-

: lo75-

“ 1o99-

Distance (feet)

25 50 75 100 150 2(I I

6%’~+: ““”.,:..,. ’...+,.,.’

, 1 I I 1 I I

. . ...-.: :’ ---- .. .

-4..:,.-,,.,’ -. . . . . .

.,, . . . . ...-.’. .- ..:..,! . . ...”.. .. . .. . . . . . . . . .. .. . . . .

:-, ,,. . ‘:!.:.,.. ~,, ,. .,. ,:...:., .. -,.,,,-,., . . .“:; ,.... , ..,, .- .,, .,.. ,. -..”.,, ,. ...- ,. :... ;:. .: .,,:, . .-.,., . . ..;jf~-.:i’;’. --”’‘.:,’.~-:.’” ~. :;: .: ”’,., ~ ‘~---, ., “-~” :’

. . . . . . ... - . . ;,..,: ., =-., -,. , . ..!

.,. ,. :’...” /—

,.

,, .,,, ,.... .,- ,,, -:, :.:.~.~,.: :,,, .-.:-...“’ ,.:. ‘,’,: ,,,,:.,,~,$, : ,;, . .: : , ,. ---- .,-,.-,.,.. . ,,j ., ..- . . .,

file: zero-5 .97B, date: Tue Nov 30 11:33:23 1999, user: tomd, host: CCS

Figure 10b. Close-up view of raypaths within the reservoir zone of the Trenton. A

conceptual model of an unfractured Trenton cap with minimal fracturing above a highly

fractured Trenton reservoir was applied to estimate the raylengths within the reservoir.

These raylengths were used in the fracture density and saturation estimates for the

reservoir. The five sensor locations are the walkaway sensor depths.

.—

NIPSCO VSP WalkawayP-Wave Time Difference 96–97984 ft and 1016 ft Sensors

●✌✎

.;

EEEIl1 I 1 I 1 I i I t I i I I I I I 1

0369 12 15 18 21 24 27

Station

,

Figure 11. Time lapse changes in P-wave travel time for top and bottom walkaway

sensors. The sensors were not moved between each station recording. Times were picked

on the unrotated vertical component. The station spacing was 50 ft. The difference

between these two curves provides the estimate of time-lapse velocity changes between

gas filled and water filled reservoir conditions using the raylengths estimated raytracing

(Figure 10). The average difference for all stations between the 1016 and 984 ft sensors

is 0.2 ms with a standard deviation of 0.1 ms.

)

1.0

0.8

0.2

0.0

NIPSCO Time–Lapse VSPTrenton ReservoirFracture Saturation vs Density

--- . Mostly Water ( 5/97)Mostlv Gas (12/96)

=4/

/I

I

I

I

I

I

I

I /

t

.

---~—., . . ..... . . .,, ..,,. . . .. ...>-.- .-a . . . . . . . . . . . . ..,:. .. .. . .. ... ,. ,-. . .,_ .. . . . . . . . .T—----- ,, .

0.0 0.5 1.0 1.5 2.0

e (dimensionless crack density)

Figure 12. Trade off curves for dimensionless crack density eandsaturation~ for the

two VSP surveys. The difference between the curves for agivenc represents the partial

saturation used for gas storage.

0 ;)

,.

NIPSCO Time-Lapse VSPTrenton ReservoirFracture Saturation vs Density

●

;.

1.0

0.8

0.2

) 0.0

0 2 4 6 8 10

1 m2 Cracks per m 3

—— --Mostly Water ( 5/97)— Mostly Gas (12/96)

/

(I

I

I

I

I

I

I

I

I

I

I

Figure 13. Trade off curves for volumetric crack density andsaturationd for the two

VSPsurveys using alm2 circular fracture. The differencebetweenthe curves foragiven

crack volume represents the partial saturation used for gas storage.

,

P

600

700

800

1100

1200

0

NIPSCO VSPSites 3(1500’ E) and 4(1250’ N)

S Time-Lapse Difference (96-97)

I

+3S1 (E-W, SV)x“

++3S2 (N-S, SH)‘--+--4S1 (N-S, SV)-%--4S2 (E-W, SH)

~-== - - - -

- -x- -2- ---x.

x-%- --- * EL

1 I I I 1 I 1 1 1

1 2 3 4

I

,.- ,~-.-- ~ ,. , ., .. ,, ,. ,, .,,- ,“ ............... ....... ... .... . ......::,<.,..&..<.,-. .-*. ,.. ,,~ .m, . . . .31/ .:<,. . ,, .:. ,. “-”~’ ---- :

5

Time (ins)

Figure 14. Time-lapse changes in S-wave travel time between 1996 (gas saturated

reservoir) and 1997 (water saturated reservoir) VSP surveys. The north-south polarized

Ssources have larger time changes than the east-west polarized sources.

Appendix A: VSl? Data Plots

>

.._.,. ........ . . . ..,,.. ....1.,. .. -,.., . . . . . . . .. . . . . . . . . . . ... ,.. . . . . . . .- ._. ,. . . . . . . . . . . . . . . . . . . .. .. .. . . . . .

,,, —- -

Exit— Parameter5 Display Help I

CECKMRELEV

CURLAT1

73$UI

o.L15------

Il.06------

07------

f38------

t19------

111------

0.11------

L_____ ----- ----- _____ _____ _____ _____ ---- ---- _\

-#. m

_____ _____ --- - -il

----- -- ---- ---- ---- ---- ____ ---- ____ ____ ___ -Cl

.---- ------------------______________----cl

_____ - ----- ----- ----- _____ _____ _____ ----- -II

____ ----- ____ --0.10

L_____--------------__________----_____--------0.11

Figure Al. Cross correlation of two Pelton generated sweeps showing 85ms delay due

to varying trigger time shift in1996 data set.

----

—. .- . .

SHO1’ 30 31

i---osos

1

---0.10

---0.s

---0.20

---0.2.5

---0.30

.---0.35

,---0.40

r---n4s

.-

0.06-------------------‘------------------------------0.06

--------------------.--...,08

o.12--------------- -------------------------------0.12

0.14--------------------------------------------------0.14,

Figure A2.

full wavelet.

Cross correlation of two

85 ms delay from figure

data traces with a 100 ms time shift added to show

Al has been removed.

%%. . 6.

Exit File Qlsplay &3alysls ~lck Mel

46 PI Shifted and RotatedCEmN 1 ? 3SsLsv sjs 642 706 770 834 898 962

T~~

r---------------010

LY,w--------------OHM

I---------------0.20I+{H-!---------------O.2S

‘K{ttHW1-Yl----------------030

- ---------------0.35

- ---------------0.40

. . . . . . .—. .0,45

I-4 1-

lEMMiaHfEaEmima Node VAWG J lScaling Global A

Zoom lUnzoodLabel RELEV Al

-36~adddd~~~~~-12Mouse Button 1: Select section,

Mouse Button 2: Deselect section,Format Horz JICursor Off_I[

Mouse Button 3 Display section POPUP menu, Colonnap I I Iw LRJ[

. . -.

Figure A3a - A3c. Seismograms from Wel146, 1996. The three

components horizontal in-line (left ), horizontal cross-line (center),

The traces have been rotated (using P-wave particle motion) and

——.. .

panels are geophone

and vertical (right).

are scaled using the

maximum of all three panels. A3a is P-wave source, A3b is S-wave in-line source, A3c is

S-wave cross-line source.

Exit File DkDkW Analvsis Pick Hdn I

CsiiulR2LEV

=

46 S1 Shiftedand Rotate

5;8 642 106 770 034 898 9625% 642 706 770 834 898 962

!d3578 sd2 706

---------------0.25

----------------0.30I

----------------0.35

t

.------’----------0.40

<,,......,.------------.--nd<

Mouse Button 1: Select section.

-36+lm I I I l~~~a&J-12

Mouse Button 2: Deselect section.F013Mt lion -1 {CU=SO= Off -1 [

Mouse Button 3 Display section popup menu. Colonnap I I Im LR AI

Figure A3b

.

. .

Exit File Qlsplay &nalysIs &lck Mel

Clmr 1RsLrv

l%=

46 S2 shiftedand Rotated3

20.00

r --------------- 0.0.s

Wm!---------------o.lo

1I---------------0.1s

}-

111)---------------0.20

---------------0.25

H.1111---------------0.30

1---------------0.35114---------------0.40

.45-

,,, 1,,, m Mode VAWGd lScaling Global -

Zooml”nzoc+abe, R,,,v 4[

-36~aJJJJ~~~~~-12

Mouse Button 1: Select section,

Mouse Button 2 Deselect section,Format Horz 41 Curs0r CIff_I[

Mouse Button 3 Display section popup menu. Colomap I I I IQ LR4j

Figure A3c

.

Exit File Qisplay A.nalysls Pick Men

157-1 PI SourceRSTATRCN1 2 3RZLEV

o

0

0

0

0

0

0

0

0

0

,,, $. IIimmm r--Mode VAWG 4 {Scaling Trace -I[

ZOoml”nzm+xd RELW 4[

-24<lfl Ill I {m~~~~+O

Mouse Button 1: Select section,

Mouse Button 2: Deselect section.Focmat Horz JICursoK Off_I [

Mouse Button 3: Display section popup menu. Colonnap la LR41

kigure A4a - A4d. Seismograms from Well 157, site 1, 1996 survey. The three panels

are geophone components horizontal in-line (left), horizontal cross-line (center), and

vertical (right). The traces have been rotated (using P-wave particle motion) and are

scaled using the maximum of each trace. A3a

in-line source, A3c is S-wave cross-line. A4d is

al,-.., <lc r.. -.. el ..).

is P-wave source at site 1, A3b is S-wave

P-wave source at site 2 (used for rotation

.

Exit File Qlsplay Analysis Mck ~elp I

157-1 S1 SourcerRCN 1 2 “3

Mouse Button 1: Select section.

Mouse Button 2 Deselect section.

Mouse Button 3 131splay section popup menu.

Figure A4b

Mode VAWG J [scaling Trace A

hcmlunzoo~hbel RELEV Al

-24dtl Jr JJM9MMd+oFormat I-lorz J{ Cursoc off-1[

Colomap I I Im LR AI

Exit File Qlsplay ~nalysls lick

157-1 S2 SourceRSTATRCN 1 2 3Ru.Ev

0.00 0,00

0.05 0.05

0,10 0,10

0.1s 0.15

0,20 0.20

0,25 0.25

0.30 0,30

0.3s 0.35

0.40 0.40

0,4s 0.4sr,I* u,

liiMEm9 Mode VAWG _I I scaling Trace ~

ZOcm Iunzoo~Label RELEV J I

-24<ln lrJJ~~~m~+OMouse Button l: Select section.

Mouse Button 2 Deselect section.Format Horz _I ICursor Off J I

Mouse Button 3: Dkplay section popup menu. Colomlap 1 I ❑ LR41

1

Figure A4c

.

I Exit File Qisplay ~nalysls ~Ick ~elp

RSTAP.!X.N

=

157-1El, P2 SourceRotatedTRCN 1 2

696 768 240 912 904 10S6 696 766 840 912 984 10S63696 760 B40 912 964 10S6

Im9iEm !fode VAWG A [scaling l=

Zoontl Unzoo~Labol RELEV J

Mouse Button 1: Select section.

-24~~ I lrJ~~~~~+O

Mouse Button 2: Deselect section.Format Horz JICucsor offal{

Mouse Button 3 Display section popup menu, Colormap I I I Iml LR A

Figure A4d

I

Exit File DlsDlaY Analysis Pick ~elp I—.. —

RSTATRCN 1IULEV

157-2 SourceP2 Rotated2

124 888 952 101(

,,

1000

[

000

--0,0s

--0.10

>

---0,2st

I)I

---0,30

---0,3s

---0,40

_o,4s--,.

I* !-

Bimilm Mode VAWG 4 15Caling Trace - I

a+ ’abe’ ~-24~~~JJJ~~M~~+o

Mouse Button 1: Select section.

Mouse Button 2 Deselect section.‘ronnat I-lorz ~[cursor off-[

Mouse Button 3 Display section Dopup menu. Colomap la LRA[

Figure A5-a - A5c. Seismograms from Well 157, site ~~1996 survey.The three panels are-—....- —

geophone components horizontal in-line (left), horizontal cross-line (center), and vertical

(right). The traces have been rotated (using P-wave particle motion) and are scaled

using the maximum of each trace. 5a - A5c are the P-wave, S-wave in-line, and S-wave

cr--- line, -fi-sect “’alv..

—

IiI

Exit File J2kmlay .Analysls Pick1

Hell

RSTATRCN

0.00:

0.0s-.

0.10-.

O.ls-.

157-2 Source S1 Rotated2

0.20

0. 2.s

0.30

0, 3s

0.40

0,4s

36 10S6 696 736 776 B16 8S6 896 936 976 1016 10S6

[

0,00

--0.0s

--0.10

\--0.1s

)

~ --0.20

5

P

---0,2s>

1

--0,30

)

--0.3s

--0,40

!3_o.4s

14 1/

Ilimm, Mode VAWG J Iscaling Trace J

.... llhxmo~.aiml _

Mouse Button 1: Select section.

-24~n I I lr~~~~&J+O

Mouse Button 2 Deselect section,Format Horz JICursor Off J[

Mouse Button 3 Display section popup menu. Colormap I I I la LR41

Figure A5b

●

Exit File DkDlaY AnalYsis Pick HelrI I

RSTATRCN 1SEQNO

0.

0

0,

0.

0.

0.

0.

0

0.

0.

rd

I’(*157-2 s~”~~~ 92 Rotated2 31 10 19 28 31 46 1 10 19 28

—o 00

--0,05

--0.10

--0.15

--0.20

--025

--0.30

--0.35

--0.40

~o,4s

=

Mode VAWG ~ [Scaling SectionJ I

Zoom lUnzooALabel SEONO -1l—— I

-2441nn I I lfi~fi~~+O

Mouse Button 1: Select section.

Mouse Button 2Deselectsectlon.Format Horz _I ICUrSOr off~]

Mouse Button 3 Display section popup menu. ColorMap 1 I I Iti LRJI

‘.

Figure A5c

I

Exit File Qisplay ~nalysls ~lck ~el,

157-3 E2. S2 SourceRotated-S 157-3E2. S1 SourceI@tatedRSTAP.UzV

,TRCN1 2 3

0.00 0.00

0.0s o.0s

0.10 0,10

O.ls 0.1s

0.20 0.20

02s 0.2s

o 30 0,30

0.3s 0.35

0,40 0.40

n 4s ilA,5

~TRCN1 2 3

0.00 0 00

0.0s 0.0s

0.10 0.10

0.ls 0,3s

0.20 0.20

0.2s 0.2s

0.30 0 30

0.3s o.3s

0.40 0.40

n 4< n 4s

N I 1./(

Imiilm Mode VAWG #jscallng m

zoonlunzoo~Lah31 REI.EV JI

Mouse Button 1: Select section.

-24~OJJrJ~~~M~+0

Mouse Button 2 Deselect section.Fozmat Horz dlCUCsOr IXT41

Mouse Button 3 Oisplay section POPUP menu. Colormap I I I la _LRJI.- .— --— -— .— .—. . . . —— .. .-

Figure A6. Seismograms from Well 157, site 3, 1996 survey. The three panels are

geophone components horizontal in-line (left), horizontal cross-line (center), and vertical

(right). Thetraces have been rotated (using P-wave particle motion). Left panels are

S-wave cross-line source, right panels are S-wave cross-line (each scaled to maximum ofi{ three components).I

Exit File Qisplay ~nalysls ~lck Mel

157-4PI SourceRotated

Mode VAWG 4 [SCalin9 ] Trace .

Zoom lUnzoom!Label RELEV -~~-24~ar I I l~~~g~+o

Moue the cursor over an object to display help here Format Horz JICursOr Off J [

Colormap I I I IQ LRJI. — — ..——— .- — .—

Figure A7a - A7c. Seismograms from Well 157, site 4, 1996 survey. The three panels are

geophone components horizontal in-line (left), horizontal cross-line (center), and vertical

(right). The traces have been rotated (using P-wave particle motion) and are scaled

using the maximum of each trace. A7a - A7c are the P-wave, S-wave in-line, and S-wave

c .lim ipec “ “y..

.

Exit File Qisplay &nalysls ~lck He

157-4 N2. S1 Source RotatedRSTATRCN 1 2

696 768 640 912 9ti 10S6 696 71 S40 912 9E4 10S6 > 76B 840 912 904 1056

Ifode Auto J [SCdiIMJ Trace J [

=UnzOo~Label RELEV Al~— 1

-z4~rJJJrJ~~~~~+oMouse Button 1: Select section,

Mouse Button 2: Deselect section.Format Horz J[CUK80K OfFJl

Mouse Button 3 Olsplay section POPUP menu. Colormap I I I Im LR AI

J

Figure A7b

I

.

Exit File @play Qnalysis ~lck Mel

RSTATRCNRELEV

000.

0 os-

0 lo-

0, l.s-

020-

0 2s.

0.30.

0,3s.

040.

157-4 N2, S2 source Rotated

: 768 04( 91 984 10S63

696 768 840 912 904 10S6912 904 l&S6 (

1.

t,, I m Node VAWG -I [Scaling I Trace .

‘ZoomlUnzoo~L.abel RELEV .

-2441n l~rd~~~~~+oMouse Button 1: SeleCt section.

MOUSe Button 2 DCSeleCt SeCtiOn.Format Horz -I{ CUrsor Off4{

Mouse Button 3 OkplaY SeCtiOn POPUP menu. Colormap I 9 LR41

Figure A7c

.,),

.

Exit File _DiSD]aY Analvsls Pick ., -,- 1

,,I

--

lima= Mode VAWG ~ Iscaling

—o 00

-----0.05

-----0.10

-----0.15

--.--0.20

----- 0.2s

-.---0.30

-----0.3s

-----040

_o.4s

1/

“race ~

-=J+Lab”’ JwY.!udMouse ELrtton l: Select section,

-36+lm In I [~~~~~-12

Mouse Button 2: Deselect section. Format Horz J JCUHOr Off A [

Mouse Button 3 Display section POPUP menu, Colormap ❑ I I Ilii LllJ[— ----- —. —-.

Figure A8a - A8f. SeismogramsfromWell157,walkaway,1996survey.The 5 panels

are geophone depths 984, 992, 1000, 1008 and 1016 ft.lefttoright.The27tracesper

panelarethe27offsetlocations(50ft.spacint).Thetracesarescaledtothemaximum

ofeachsectionexceptwherenoted.A8a-A8careP-wavesourcelgeophonecomponents

vertical(tracescaling),unrotatedhorizontal1(tracescaling),andunrotatedhorizontal

2.A8d-A8fareS-wavecross-linesource,vertical,unrotatedhorizontal1,andunrotated

horizontal2,

I

4 . . . . -----

Exit Fllc DlsDlav Analvsis Pick !ielp I

157 WalkawayP-wave Unrot HInm.m7 qliA qq7 1000 100II.—--. . . .ISPNU?I 17131925-i-713 192s-”i-713192s -

000

0 05

0,10

0.15

0.20

0.25

0,30

0,35

0.40

0.45

. . . 101625

0 00

----.-005

------010

------0.15

/

------020

-----0.25

------o 30

------0.35

------0.40

)

)

~------o 4s

-24~n In I IMM~fi~+oMouse Button 1: Select section.

Mouse Button 2 Deselect section.Format Horz 4 [ Cursor off 4 [

Mouse Button 3: Display SeCtiOn POPUP menu. Colonnap I I I 9 LR AI

FigureA8b

.

#..1“

Exit File Q15play f@ysls ~lck He

RELSV 984157 Walkaway P-wave Unrot H2

992 1000 100R Inlfi

I

]4 I I&//

ElilmEBm Uode VAWG 4[SCaling GlobalAl

Z00mlUnz00~Lah31 ESpNUM -II

-24<ljlm I I 1~~~~~+0

Mouse Button 1: Select section,

Mouse Button 2 Deselect section,Format Horz JICUKSOr Cff~l

Mouse Button 3: Display section popup menu, Colonnap I I la LR41

Figure A8c

Exit File Qisplay ~nalysls ~lck ~el~

RELEV !ESPNVM

ooo-

0,05---

0.10---

0,15---

0,20---

0 25---

0.30---

0.3s---

o,40---

o.45_

992. .

157 WalkawayS-WaveVertical1000 - 1008 10

\..

-...

-—

Mouse Button 1: Select section.

Mouse Button 2: Deselect section.

Mouse Button 3: DkD]aY section 130tmp menu.

Figure A8d

25—o 00

------o 0s

------0,10

------0.15

------020

------o2s

------0,30

------0.35

------0,40

_o.45

1/

Mode VAWG _slScaling I Global 4

Zc.omlUnzx.o~Label EsPNUIM 4

-24qDrJJJ~~~~PJ+o

Format Horz _l[CUE30r OSTJI

Colonnap I I I w LRJI

.

$- .,...,,

Exit File Qlsplay &witysls ~lck ~ell

RSLEV 98453P

2157 WalkawayS-wavetln

1000 Ifi..7 13 19 25

—o 00

----- 0.0s

-----0.10

-----0.1s

-----0.20

----- 0.2s

-----0.30

----- 0.3s

-----0.40

_o,45

1<l+

,(,

‘“ Emlm Mode VAWG AI SCaling Global_Il

Mouse Button 1: Select section.

lzooln[”slzoo@ssml ,SPN”M Al

-24~Dr I I 1~~~~~+0

Mouse Button 2: Deselect section.Format Horz AI Cursor CJff_I [

Mouse Button & Display section POPUP menu. Colosmap I I [m LR41

Figure A8e

I

—

3

.

--3

..’=+n+

S-4m.

u.m

.

!2

1 --A.

,,4

,

al$-l

2v-lE-4

Exit File Qlsplay &mlys15 ~lck He

RsTnRsLzv

—

1996 F-wave SouthOffsetRot,TRCN 1 2

696 744 792 340 888 936 904 32 10BO696 744 792 B411 BEE 936 984 10323

Iimlm Mode VAWG A [SCaling AGC A [

Zoom IUnzood Label RELEV -II

J

—— 1

-z45.1 CIJJJJHEHHd+oMouse Button 1: Select section.

Mouse Button 2: Deselect section,Format Horz d I Cursor off 4 [

Mouse Button 3 Display section popup menu. Colosmap I I I 19 LRJ[

-. .——

Figure A9. Seismograms from Well 157, site 6 P-wave source, 1996 survey. The three

panels are horizontal in-line (left), horizontal cross-line (center) and vertical (right). TheIIII.. traces are normalized to their maximum,,,

)’

Appendix B: Travel Time, Velocity and Rotation Angle Tables

and Plots

,,,

-:—-y.y-s~-f.,! .<,, .; ,, ,, -. ... y ~ , ..,G / .>.,~..t , .,fiq, . . . . . . . ~.-., ..J:. ! .*. r L-J .. ... fi. ... . ,. r,, ~.. , ..!,.., ,.,.W,-. . ! .,---- .

. .. . . .. . . . . .

mLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLNELLEVELLEVELLEVELLEVELLEVELLEVELLNELLEVELLEVELLEVELLEVELLEVELLEVELLEVEL

12

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

PHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHI

PHI

83.88483.26582.’77781.79877.59082.94279.65579.07276.97872.36574.67874.85374.19067.33967.29780.43974.92885.23074.42477.41065.71070.22486.53075.13266.01260.45959.88254.47761.15256.65569.46168.11172.98373.78558.39755.98356.72758.64260.14859.72064.87960.41961.33449.21831.99449.47840.93817.90943.67041.44849.57950.83525.64329.81341.11741.40642.23322.11543.64242.86845.91542.86435.38452.489

THETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETA‘THETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETA‘THETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHE$I’ATH!?!PA

169.292172.623172.499168.531173.822169.058165.905163.481157.250146.92228.51721.79727.29147.26651.254-67.956-67.035-67.000-65.187-73.702-81.124-82.805-77.368-79.120-81.626-111.368-105.702-114.478-100.904-98.785176.836172.297177.090177.216172.36966.21255.34766.52170.36872.16737.84734.87735.22540.34966.314

-63.847-66.519-45.047-68.992-74.853-73.171-75.725-63.492-65.981-75.999-80.176-84.044-85.868-81.016-84.038-87.946-91.760-92.476-86.485

angl

_ __ .—...

d;. atLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLEVELLNELLEVELLEVELLEVELLEVEL

6566676869707172737475767778798081828384858687888990919293949596979899100

—.— .-

PHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHIPHI

PHIPHIPHIPHIPHIPHIPHI

45.30850.61445.67428.82742.41538.16449.18947.22228.45835.52528.47634.92048.16241.03261.57649.92635.80036.14926.16744.04126.61356.26347.59361.03445.66434.21741.97546.84737.99446.06354.37751.93456.05847.65953.47854.463

Figure B1. Rotation angles for 1996 site 1.

..

THETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETATHETA

-88.702-84.830-90.606-97.617-91.058-91.034-65.012-68.429-57.748-67.851-70.822-56.124-61.198-46.299-62.192-68.968-73.234-76.490-69.704-74.052-74.397167.182168.232168.472168.012152.798-43.757-50.513-22.750-39.024-62.079-39.343-47.406-18.103-32.279-52.078

m

mlDepth (ft) P S1 S2 S2timl296 34.243530304 34.712791312 35.392941320 35.864471328 36.360809336 37.016701344 37.351009352 37.809921360 38.278141368 38.704559376 39.195580384 39.456600392 39.944859400 40.410782408 40.854019416 41.373211424 41.548771432 41.845779440 42.122330448 42.686779456 43.341370464 43.879440472 44.363621480 44.712940488 45.066570496 45.626499504 46.072121512 46.491501520 46.975842528 47.316231536 47.790260544 48.186111552 48.789230560 49.321819568 50.047230576 50.782471584 51.088551592 51.511452600 52.012821608 52.733559616 53.474548624 54.074749632 54.800690640 55.480610648 56.238220656 57.087490664 57.727089672 58.240028680 59.110809688 59.958099696 60.779610704 61.439129712 62.176399720 63.038029728 63.984081736 65.034142744 65.834061752 66.543823760 67.367828768 68.182159776 69.117889784 69.966713792 70.683823

119.006233119.705803120.115822120.431213121.735313123.708450124.654533124,415680124.889603124.667084124.384041124.641800125.799263126.613228127.923843127.856323128.622787129.050201129.138046130.044891132.435593134.140961135,843216136.861145136.758743138.775925138.662354140.742157140.611252142.172043141.417038143.242020143.410843143.943802144.539337143.913849145.690582145.559204146.821411149.521179149.434769151.787323151.526062153.354584156.234222156.694977159.355362160.740021161.590775162.757965165.748611168.165131168.978714170.644745172.808609174.288605178.160095179.368607181.359207182.256851184.711273185.725952188.229416

110.734299111.634079112.364594112.937149113.968613115.141823116.197197116.196777116.774208118.048241118.072067118.758469119.658127120.147186122.279739121.875092122.667236123.162331123.821877124.625664125.799057126.772858127.760223128.428177129.124176130.034393130.361145130.901505132.168198133.227264133.345718134.208099134.934921135.635345136.747864138.683762140.153366140.425079141.364487144.652786145.030945146.037262147.101776148.354507150.375275150.183350152.024979152.876190154.934799156.669952158.755936160.574753162.532288164.465271166.136551168.860397170.308121171.868851173.327957175.022888176.272934177.975876180.315216

100.201057100.079041101.962952103.455040103.086304103.500000104.500000105.500000106.000000105.963997107.500000109.620331109.931519110.778831112.024773112.842690113.492050114.551941114.853668115.953011116.389511117.427856118.302353119.095032119.770851121.601677121.986504123.662079123,223328124.297737123.529861124.521812125.759399126.578133127.339233129.977219131.189270131.531784132.232330134.817276133.778107136.527954137.510208139.115570140.475266141.444824142.572113144.022476145.709915147.547470149,406921151.119705152.968903154.906418156.806076158.697830160.734955162.177063163.758469165.501999166.800659168.823685170.627991

Figure B2.

157-1.times800 71.433563808 72.119133816 73.271698824 74.084908832 74.920418840 75.626907848 76.214813856 76.783684864 77.625107872 78.597267880 79.619743888 80.430779896 81.141747904 81.548927912 81.976883920 82.619438928 83.653877936 84.989647944 85.519402952 85.831291960 86.135567968 86.390182976 86.562363984 87.575020992 88.0024191000 88.3330231008 88.8637241016 89.3636471024 89.9018171032 90.3162081040 90.7131501048 91.1415411056 91.6492771064 91.9602431072 92.3505481080 92.7769091088 93.314301

190.076126190.598618193.050446194.171326196.026321198.643845199.087051199,225723200.785355202.344559202.973602204.184006206.464981207.796143208.949661209,547440211.561096212.976318215.752808216.725861216.947433216.510696219.220596221.052475220.601669221.692474220.904404220.815872221.637817223.783768223.362854222.818573224,958038225.545212227.763336227.457886226.911484

181.698883183.255035184.856918186.394379189.149445190.317581191.702072193.264847194.978699195.884476197.856216199.442673201.088577203.202164203.752411204.709106205.734756207.559860209.619675210.167404211.186493211.647995212.209579213.117416213.830414214.808517215.837204216.172653217.297150218.279297219.208893220.496109220.551468221.738800222.410065223.231033224.042007

171.987549173.470078175.196411177.246231179.044022180.444946181.883011183.896057185.191544186.570892188.166016189.691040190.886093192.238251193.768997195.435379196.578659198.690689199.332474200.044296200.809113201.301727201.277817201.704742203.355240203.552963204.951935206.287842207.293793207.699112208.801514209.551941210.218033211.239578211.674271212.900558213.590195

m;,’ ,,,, ,. , ,.+

., .,, .,,;,’, ,,

,, ,,,’ ;., , ,

Traveltimesfor1996site1 withoutlowpassfilter.-.

i+o.-lmor-m-mrlul Cooo)clm.no+c.rain el-op,-l mrqcor.atulrq ,+ CocqFldcONi-lm OLowm Inrlooeqelr.1-Jw ,TClm$-11.lwlwwqw Ulmrlm .+.-lwdwqc-umm mm+ c.Jinmcommqmwo qNmu-Id.-Ic4ut m-J qmmwmmtiu-11-m-l.-ll-+0I.m mrlN.-lmmtn-Jc.ari *w,+Lflm-smomr-l Calnwl.wultnL.4mmmtn-=Jmw0r4 t-wr4L.wcOwrmq.-I Nc.aqruImmmqqO mooCommt.wwmuldr- moulmeleh+~l- a,-ldmoar.lolnaq mtoa,-llod

,, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .mo.-itwm-=r-mww r-mmmodmromm -J.=Jinmtnwc-I-mLV mmmoo.+

CA wr-mt.t.l.r-I-PP c-bt.t.mmmamm WCOCOCOCOC9CJCOCOLV acocommm

a)

t!mw-#NOOawql.aO mw-#mOmwqllNo Oawuomc...vtnqme mwq!mom01-lr4m--JmmFw mmo,+wmrlemw Wpmaloorlcumq Wulwpmm

. mwcommm.nrncnoa Oammmmmmo-,o’lol molmmoooooo 000000.+.-4.-4s+!-!.-!.4.-!+.-! -1-1

I

i

I

I

DEPTH P_VSLOCITY SV.VSLOCITY SH_VELOCITY296:304.312.320.328.336.344.352.360.368.376.384.392.400.408.416.424.432.440.448.456,464.472.480.488.496.504.512.520.528.536.544.552.560.568.576.584.592.600.608.616.624.632.640.648.656.664.672.680.688.696.704.712.720.728.736.744.752.760.768.776.784.792.

9115.–9210.9248.9340.9423.9463.9584.9671.9754.9846.9920.10050.10122.10197.10277.10336.10479.10591.10707,10748.10766.10812.10871.10961.11049.11085.11148.11217.11268.11353.11405.11475.11494.11530.11520.11509.11594.11652.11692.11682.11667.11684.11673.11673.11656.11621.11630.11663.11625.11593.11566.11571.11561.11529.11483.11419.11401.11398.11377.11357.11318.11294.11292.

2819.2864.2913.2966.3006.3042.3081.3147.3197.3228.3293.3339.3379.3430.3433.3509.3549.3598.3642.3681.3709,3742.3775.3816.3856.3890.3940.3984.4005,4032.4088.4120.4156.4193.4216.4214.4226.4274.4302.4259.4302.4326.4349.4365.4359.4417.4416.4443.4435.4437.4428.4427.4423.4419.4422.4398.4407.4413.4422.4424.4438.4440.4426.

—2623.2671.

2725.

2781.

2814.

2832.

2872.

2939.

2990.

3057.

3126.

3182.

3214.

3255.

3282.

3345.

3385.

3434.

3492.

3528.

3523.

3537.

3550.

3581.

3641.

3645.

3704.

3705.

3765.

3779.

3854.

3860.

3910.

3951.

3989.

4061.

4066.

4124.

4142.

4120.

4175,

4162.

4222.

4223.

4196.

4234.

4213.

4226.

4253.

4271.

4241.

4228.

4254.

4259.

4252.

4261.

4213.

4229.

4226.

4249.

4235.

4255.

4240.

avgvel.dat800.808.816.824.832,840.848.856.864.872.880.888.896.904.912.920.928.936.944.952.960.968.976.984.992.1000.1008,1016.1024.1032.1040.1048.1056.1064.1072.1080.1088.

11285.11287.11218.11202.11183.11184.11202.11223.11203.11166.11122.11109,11110.11152.11190.11200.11156.11075.11099.11151.11204.11263.11333.11293.11328.11376.11398.11423.11443.11479.11517.11550.11573.11620.11657.11690.11708.

4436.4442.4447.4453.4430.4444.4454.4459.4460.4480.4476.4480.4483,4475.4502.4520.4536.4535.4528.4554.4570.4597.4623.4640.4662.4678.4693.4722.4734.4750.4766.4774.4809.4819.4840.4858.4876.

Figure B4. Average velocities for 1996 site 1.

.

4241.4271.4258.4274.4274.4258.4288.4325.4331.4337.4363.4376.4366.4376.4390.4416.4411.4419.4399.4416.4449.4494.4475.4474.4519.4533.4585.4623,4642,4633.4677.4724.4715.4738.4727.4768.4815.

m

98/10/1213:3253

DEPTH336.344.352.360.368.376.384.392.400.408.416.424.432.440.448.456.464.472,480.488.496.504.512.520.528.536.544.552.560.568.576.584.592.600.608.616.624,632.640.6d8.656.664.672.680.688.696.704.712.720.728.736.744.752.760.768.7’76.784.792.800.808.816.824.

P_VSLOCITY15488.16198.16911.16960.17187.17752.18448.19213.20196.19518.18768.17610.17644.18138.18540.18378.17294.16851.16141.16933.17632.18225.17746.17050.15786.15257.15689.15685.15639.14547.13870.13394.13125.12816.12753.12527.11901.11746.11138.10946.10828.10744.10730.10475.10222.9966.9772.9542.9598.9639.9508.9299.9323.9449.9753.9633.9619.9476.9613.9885.10360.10371.

Deth_sv SV_VSLOCITY Depth_sh

S30.

344.352.360.368.376.384.392.400.408.416.424.432.440.448.456.464.472.480.488.496.504.512.520.528.536.544.552.560.568.576.584.592.600.608.616.624.632.640.648.656.664.672.680.688.696.704.712.720.728.736.744.752.760.768.776.784.792.800.808.816.824.

LUQ2Z .

10786.10554.10694.9292.11486.11969.11132.11016.11817.’10070.9718.9623.9423.11411.9580.10167.10115.9386.9114.10394.10557.10946.10903.10313.9095.8037.8268.8566.6897.6747.6667.6485.6205.5794.6868.6655.6347.5826.6580.5763.5443.5129.4914.5024.4240.4333.4172.4309.4319.4526.4557.4460.4603.4635.4961.4934.4594.4673.4761.4674.4671.

--..”. . . 336.- 14262.344.352.360.368.376.384.392.400.408.416.424.432.440.448.456.464.472.480.488.496.504.512.520.528.536.544.552.560.568.576.584.592.600.608.616.624.632.640.648.656.664.672.680.688.696.704.712.720.728.736.744.752.760.768.776.784.792.800.808.816.824.

15568.13544.12472.12479.18645.19515.16731.18274.14453.9664.8199.7762.7615.8840.7158.7792.6696.6828.6464.8734.8624.10378.11095.10105.15310.11198.16346.12684.10723.9833.9230.9722.8387.6751.6180.5782.5206.5353.5974.4849.4831.4535.4578.4777.4502.4213.4253.4009.4065.4181.4516.4120.4082.4460.4230.4957.4765.4593.4718.5472.5274.

SS_VELOCITY

-.

832. 10038.840. 9705.848. 9560.856. 10098.864. 10648.872. 11265.880. 11369.888. 10688.896. 9690.904. 10073.912. 10994.920. 12207.928. 13348.936. 14676.944. 13203.952. 13205.960. 13928.968. 15276.976. 18196.984. 18163.992. 17749.1000. 17391.1008. 16757.1016. 15653.1024. 18158.1032. 18315.1040. 17921.1048. 17895.

—-. -— .

832.840.848.856.864.872.880.888.896.904.912.920.928.936.944.952.960.968.976.984.992.

1000.1008.1016.1024.1032.1040.1048.

..

5102.

4917.4909.4896.4729.5443.5524.5666.5562.5432.5568.5967.6517.7154.8024.7895.7879.7878.9241.

10368.9813.9924.8998.9545.9236.9282.9456.9707.

832.840.848.856.864.872.880.888.896.904.912.920.928.936.944.952.960.968.976.984.992.1000.1008.1016.1024.1032.1040.1048.

. .5628.6160.5849.5924.5833.6151.7291.6374.5783.5313.5530.5692.6453.6237.6002.6829.6552,8518.10152.13525.11279.12409.12622.13878.17724.11120.13813.13259.

Figure B5. Interval velocities for 1996 site 1, with an 80 interval.

696 121.190102704 121.942596712 122.021103720 122.365196728 123.443497736 125,073196744 125.326599752 127.288300760 126.844498768 127.102303776 128.488007784 128.913101792 130.580704800 130.206406808 131.142197816 132.052994824 132.644897832 133.968994840 133.964798848 134.072006856 134.792297864 135.108398872 136.671295880 136.162094888 137.374100896 136.760605904 137.298096912 137.631500920 138.342697928 138.426102936 139.609100944 140.429794952 139.533005960 139.665298968 140.105301976 138.373703984 139.913300992 141.1347961000 140.4996031008 139.4794011016 139.3798071024 139.6461941032 139.8621061040 140.0298001048 140.2877041056 140.1786041064 140.5610961072 140.4308011080 140.7505041088 141.123901

260.186615261.262604262.622498264.026215265.504211267,656586268.981812270.222809271.482788273.109314274.963104276.145905276.555115277.008789277.655396278.202698279,421112281.281189282.550903284.728302285.901215287.457489288.004303288.520111288.863403289.195709289,525085290.774200294.761993295.265503295.453705295.311615295.349701295.289612295.306702295.158112295.245300295.248505295.193115295.315399295.601501295.538910295.606293295.683990295.799286295.973511296.133698296.177002296.282898296.391510

250.693207252.228195252.487503255.160706255.687698256.663086258.250397259.698608260,596497261.828613263.054504264.336914265.503693266.779388267.948914268.823090269.994385271.263702272.569489273.634094274.943298275.866089277.098785278.080200279.253601280.033813281.380096282.698700285.381897286.153900285.633911285.668213285.933105286.174988286.410004286.709412287.206512287.450287287.684814288.115295289.019501289.968811290.248413290.547913290,385712291.311401291.346588291.812592292.083099292.456787

157-

—. ----- —---- .———

times

-—. ..-. ..

,.-.,,. ‘ m>,;,,.’ ‘>

Figure B6. Travel times for 1996 site 3. The columns are depth (ft.), P time, S1

(in-line) t,imeancl S2 (cross-line) time.

.

I 98/W1213:3240 I

696 105.702904704 106.180199712 106.839401720 107.529297728 108.329300736 109.363098744 109.931702752 110.614403760 111.382301768 111.928497776 112.629097784 112.872002792 113.719803800 114.433296808 115.031303816 115.736099824 116.174301832 116.665604840 117.240799848 118.028801856 118.958603864 119.775200872 120.203300880 120.717499888 121.223999896 121.876297904 122.350403912 122.818001920 123.334000928 124.367500936 125.856201944 125.965599952 126.087303960 126.259102968 126.354301976 126.196404984 126.568703992 126.6323011000 126.5901031008 126.8172001016 126.8302001024 126.6154021032 126.6202011040 126.5690001048 126.7906041056 127.3666001064 127.1856001072 127.4744031080 127.6896971088 128.060806

230.274399231.616302232.699402233.883301235.238495236.337708237.198105237.797501238.522903240.124802242.376099243.908005244.603897245.717300246.948807248.411499249.360703250.310806251.461899252.782898253.700699254.377197255.017197256.052795257.721588258.765991259.432404260.082214261.183289262.934113264.302612264.450592264.577087264.661591264.819397264.799805265.001312265.264313265.465698265.714508266.033295266.200104266.461609266.720703267.038788267.392395267.444092267.633514267.825806268.104401

157-4.times223.496002224.936096225.323303226.788406228.916794230.800598232.413101234.058807235.232803235.631302235.869202‘237.239197238.762695240.072800241.092697242.537994243.828293244.881607246.737900248.128799248.637207249.633408251.218597251.790604252.775497254.180496254.968506255.261902256.419800258.222900259.302795259.827698260.783600259.219788259.497986259.640594260.027191260.589203260.882202261.624298262.257507263.025208262.955597263.600006264.213013264.502899264.876709265.012207265.420898266.004913

. . . ————. -.— -—---- - . .. ...- . ..— .—.

Figure B7. Travel times for 1996 site 4. The columns are depth (ft.), P time, S1

(in-line) time and S2 (cross-line) time.

Depth96 P96696704712720728736744752760768776784792800808816824832840848856864872880888896904912920928936944952960968976984992100010081016102410321040104810561064107210801088

187.000000188.000000189.000000189.000000191!000000191.333328191.666656192.000000193.500000193.500000195.000000195.500000196.000000196.000000196.000000197.000000197.000000199.000000200.500000201.000000201.000000201.000000201.500000201.500000201.000000200.500000200.666656200.833328201.000000202.000000201.500000201.750000202.000000202.250000202.500000203.000000202.000000202.500000202.166672201.833344201.500000200.000000199.500000200.500000200.000000199.750000199.500000199.500000199.500000199.000000

——-

~,. , 157-(

.-..—— .—

times

—..... —

.. .

_.. — .

Figure B8. Travel times for 1996 site 6. The columns are depth (ft.), P time

I

.

May 199”/ Site 1Inverval VelocityAO ft Interval,...

I .1

200

400n

ii-

1000

.- ---- —..

o

1

++ ++

* ++ +

++++

+++++ ++ ++

++++*

++ ++

++++ +

* ++++

++*+

+

+

+

-1

II I 1 I 1 1 I 1 I I I 1 1 1 1 # 1 I 1

0 5000 10000 15000 20000

P=VELOCITY

Figure B9. Plot of P-wave interval velocities for 1996 site lusinga40 ft. interval.

“)..,

/

..* )

o

200

400

zw-

Jj600

Q:

800

1000

12.00

Site 1 May 1997S1 Interval Velocity40 Ft Interval

8

-,

?!2

1 I t I I I 1 1 I 1 1 I 1 I I I 1 I 1

0 5000 10000 15000 20000

Sl_VELOCITY (ft/s)

Figure B1O. Plot of S-wave (in-line source) interval velocities for 1996 site 1 using a 40

ft.interval.. .

)

.,

.)

o

200

400

600

800

1000

1200

0

Site 1 MayS2 Interval

1997Velocity

40 Ft Interval

*

,

, 4

I 1 I 1I

I 1 I 1I

1 1 8 1I

1 1 1 1

10000

s2_vELocITY (ft/s)

,

,

----- T-,—-T=;-.., .. . ,. , . . . . . ..... . .. . . . ,., . . . . ..s. ”.. -. ., ! -., ,. . .. - .. . :..% : 9 ... . . . . . .,. ~,.,. .. ...... ... “,,.... .,..-.

5000 15000 20000

Figure B1l. Plot of S-wave (cross-line source) interval velocities for 1996 site 1 using a

40 ft. interval.

,

Appendix C: Time Lapse and Travel Time IXfFerence Plots

>

. .--. -,r, , . . . . . . . . . . . . . . . . . . . ..- . . . , , .,, . . . . . . . . . . . . . . . . ..——. . - .-. -

NIPSCO VSPSite 1 (100’)

P Time Difference (96-97)

o

200

1000

.. , -e .. .- -—

1200

0 2 4 6 8 10

Time (ins)

Figure Cla. Time-lapse change in P-wave travel timeatsitel.

~ - —-.

.— -.

.*

— .

. ———_

- --.=-T-., T’-.<. K T-TzFzTb,: . ?. <..,.,,, ,. !,.. <,. ‘=?T$,...ff. f. ,,.,,, . .. .?.. . ... . . .. ,-.. ,-..,. , >~...~.,.-, . ...=--’ .?. $ -, *.:%.>..*.. r .hv2.A.. . . . . . ..------

‘)o

200

400

600

800

1000

1200

NIPSCO VSPSite 1 (100’)

S1 Time Difference (96-97)

1i I I I # I I I t

-5 -3 -1 1 3 5

Time (ins)

/

Figure Clb. Time-lapse change in S-wave (in-line source) travel time atsitel.

-

o

200

400

600

800

1000

1200

NIPSCO VSPSite 1 (100’)

S2 Time Difference (96-97)

=7-7

—

-5 -3 -1 1 3 5Time (ins)

Figure Clc. Time-lapse change in S-wave (cross-line source) travel time atsitel.

NIPSCO VSPSite 1 (100’)

S Time Difference ( S1-S2)

o

200

1000

1200 I I I I I 1 I I I I

10 12 14 16 18 20

Time (ins)

FigureCld. TraveltimedifferencebetweenS1 (in-line,E-W) andS2 (cross-line,N-S)

atsite1 for1996and 1997surveys.

/

0

200

400

600

800

1000

‘)1200

NIPSCO VSPSite 2 (640’ E)

P Time Difference

I I 1 1

5 7 9 11 13 15

Time (ins)

Figure C2a. Time-lapse change in P-wave travel time atsite2.

,-,–.,n - > , ,,-.T-m.-;- . . , ,.,,.. ,, .44 .-,...,.7,.., >.. . . . .,. ”..., .-.., ? > - , ,.,. ,., , .4-., s-.. . . -,,.,:, .,-. ,<,1 r. . 2... —...

,. .,,.

NIPSCO VSP

Site 2 (640’ E) Time Lapse

S–Wave Difference

200

400

600

800

1000

- . . . . . . - - - - - - -- -- - - - . . . . . . . . .

-.0

- - -- - . .

‘...-

------------ --_#-. . - - -

. . .●

------- -

- - -- . . . . . . . . . . . .- . -.---- - - . . . . . . >--

~------- ---....-

-------- - --- - - . - . . . . - - -

-1 1 3 5 7 9

Time (ins)

,

Figure C2b. Time-lapsechangeinS-wavetraveltimeatsite2 forS1 (in-line,E-W)

andS2 (cross-line,N-S)for1996minus1997.

,-

.),,0

200

400

600

800

1000

..

)1200

NIPSCO VSPSite 2 (640’ E)

S Time Difference (S2-S1)

PEl

-..------ --- . . . .

●✎✎✍✎,’,

“------ .:.----

------ -- . - -e

-..-.”

------ .+

.:.----

-...*-

10 12 14 16 18 20

Time (ins)

Figure C2C. Travel timedifference between Sl (in-line, EW)and S2(cross-line, N-S)

at site2 for 1996 and 1997 surveys.

,“

),

.,

~..

‘)o

200

400

JJw

600;

gn 800

1000

0

NIPSCO VSPSite 3 (1500’ E)

P Time Difference (96-97)

I I 1 1

2 4 6

Time (ins)

8

,

10

Figure C3a. Time-lapse change in P-wave travel time atsite3.

““),L- 0

200

400

600

800

1000

) 1200

NIPSCO VSPSite 3 (1500’)

S Time Difference (96-97)

a?-—_----=--y*----%-.------.-~=-.~ %==”

T-7 .. . ,,, . ...! . .. ., . . . . . . . . .- >,, ,,. ,. —m.. .. ,. .X,.* ..,. . . . . . . . <. . . . 7..<. .,. . . . . . . . . . .,-. —.. .— . . . . .

o 2 4 6 8 10

Time (ins)

FigureC3b. Time-lapse change intravel timeat site 3for Sl (in-line, EW) and S2

(cross-line, N-S).

NIPSCO VSPSite 3 (1500’ E)

S Time Difference (S2-S1)

200

i

400

-1

04

:800

1000

)1200

-14 -12 -lo -8 -6 -4

Time (ins)

,

Figure C3c. Travel time differencebetween S2 (cross-line) and Sl(in-line)for 1996 and

1997 surveys.

)9 “.

..

...

)

o

200

400

600

800

1000

1200

NIPSCO VSPSite 3 (1500’)

S Time Difference (96-97)

l-- S2-S1, 96-971

1 ! I 1 1 1 I I 1

-5 -3 -1 1 3 5

Time (ins)

Figure C3d. Time-lapse travel time difference for site 3, S2minus Sland 1996 minus

1997.

NIPSCO VSPSite 4 (1250’ N)

P Time Difference (96-97)“>

/

o

200

400

600

800

1000

1200

-.s$: I 1 I I I 1 I I

o 2 4 6 8 10

Time (ins)

,

Figure C4a. Time-lapse change in P-wave travel time atsite4.

o

200

400

600

800

1000

NIPSCO VSPSite 4 (1250’)

S Time Difference (96 - 97)

P=--+---*--+:.---a--- .=-

_= -._ __ -_-%= ---=.=__:2’--- - ------- =%

I 1 I 1 I I I 8 I 1

+

. . —. .— —.— .

0 2 4 6 8 10

Time (ins)

Figure C4b. Time-lapse change in travel time at site 4 for S1 (in-line, EW) and S2

(cross-line, N-S).

NIPSCO VSPSite 4 (1250’ N)

S Time Difference (S2-S1)

200

4001

800

iooo

1

------ . . . . . . . . .

-- - -... . . . - -- ..-+.:.:;:..- .-.*=-. . . . . .

. . . . . . . .-------

“-- . . . . . . . . . .-----

- -----

-lo -8 -6 -4 -2 0

Time (ins)

Figure C4c. Travel time difference between S2 (cross-line) and Sl(in-line)for 1996 and

1997 surveys at site 4.

/

o

200

400

600

800

1000

1200

NIPSCO VSPSite 4 (1250’)

S Time Difference (96-97)

1 1S2-S1, 96-97 [

1

I I I 1 1 I!

II

1

*

. ...- ,,, ,. -. ., . . .............. . .. . ... .. . . . . .,’ ..,. . . ,.—. - - - -.—

-5 -3 -1 1 3 5

Time (ins)

Figure C4d. Time-lapse travel time difference for site 4, S2minus Sl and1996 minus

1997.

5

4

;

#-1-1B

3

2

NIPSCO VSP WalkawayP–Wave Time Difference 96–97984 ft and 1016 ft Sensors

4 ,*.*,,

Mean Difference = 0.2Standard Deviation = 0.1

,,*

msms

I I 1 I 1 I I I I I 1 I I I I I t

0369 12 15 18 21 24 27

Station

#Figure C5a. Time lapse P-wave time change for walkaway survey for twosensors (984

and 1016’).

--.

).-

)

)

NIPSCO VSP walkawayP–wave Time Difference

Time Lapse (96-97) andSpatial (1016’- 984’)

0.5+

0.4

0.3

d

-0.3

-0.4 -

-0.5 [ I 8 I I 1 I I a I I I I I I I 1

0369 12 15 18 21 24 27

Station

Figure C5b. Time lapse and spatial change for P-wave walkaway.

..>.7,~.—. ,-. -~, .- ~,~ ~m,l ,-.-.~~,:.’ -,... .-— -- Y,. .-. .>.,. ‘l. <..,.: ;.<<.,., ., “ > :., ..,., . . . . . . . , ..:). . . . ,.1 . % . . .,. <,

—.--— -.— .

NIPSCO VSP WalkawayS-Wave Time Difference 96-97984 ft and 1016 ft Sensors

6

5

4

-2

0

-1

-2

-3 I ,1 I 1 I I 1 i 1 I I i I I I I I I

0369 12 15 18 21 24 27

Station

Figure C5c. Time lapse S-wave (cross-line source) time change for walkaway survey for

two sensors (984 and 1016’).

)

NIPSCO VSP WalkawayS-Wave Time Difference

Time Lapse (96 - 97) andSpatial (984’ - 1016’)

2.5

2.0

1.5

1.0

0.5

0.0

; -0.5

E -1.0

-1.52-$ -2.0

-2.5

-3.0

-3.5

-4.0

-4.5

-5.0 1I I I 1 1 I I I 1 I

0369 12 15

Station

--r-r-r18 21 24 27

Figure C5d. Time lapse and spatial change for S-wave (cross-line source) time change

for walkaway survey for two sensors (984 and 1016’).

..

.<)

-

NIPSCO VSPSite 6 P–Wave Time Difference

200i

400

6001

800

1000

1200

o 1 2 3 4 5 6

Time (ins)

Figure C6. Time lapse P-wave time change for site 6 (1996 - 1997).

3.’

Appendix D: Other Studies and Analysis

‘

,,,.~ ;,7- ,:.

—.. _

., ..,,, . . . ...?.“. ,..., ,., ,. ,,, , ,., .,>,.,.,!, -.: --=. -7Yz%7?.,. ~, ,.. .. . . . . . T- >..>,s ,,tl.> . ,, -- .+., , ..

——. . .. —.. —-_

v...

Figure Dia. Particle motion for comparison for 1997 site 1 S-waves. S1 (in-line, EW)

source (left side) and S2 (cross-line, N-S) source (right side). Level number is sequential

from shallowest (96 ft.)., .

)

,

‘ .)

‘)

Figure Dlb. Particle motion for comparison for 1997 site 1 S-waves. S1 (in-line, E-W)

source(leftside)andS2 (cross-line,N-S)source(rightside).Levelnumberissequential

fromshallowest(96ft.).

. ... .. . . .

‘IME = 120 TRi3CE = 71lick on the displa~ For Parameters Panel

p

96.up.agc

o-

200

.-.7.,7 T7T-. m. .f!.~ ..4.

—. . . . . .

II Figure D2a. Upgoing data from 1996 site lP-wave source after F-K filter.t . . .

,, ,~, ~< ..,’ ,, *4,. . . . . . . . .

‘, :. . ,,. ‘.;.,,’,;.:, ,. .,.~,.,’”, .. .

,. ... ,;,..,, ,

‘ ‘J.. ~ , ~ ,.,,. .,,, .. ;;,.- . . . . ...’.’..’ .. ...>,;..” ,.. -.:... -,~-\. .

,’, ~+ ..,. ,>’ :“_ ::._ ‘“,+..,,,”..,:.

..,.. ””., .,,. --”-””.,.-.-”.. ,, “ , . . . . . -

,.

. .,----- -A” ., . :“””,-, .. . .. , .,..- , -, . ,,, ,, .- ,,> .,.,”,.-.”,. ./.

.,

g..:: ,“,_.,”..”. . ...-..;=::;.:=:.;-.7.,... ..2..L.LLAL.UJ-‘ ‘‘ ‘‘,

‘$5_——.,. ... . . .——

“ItlE = 194 TRfiCE = 98;lick on the display for Parameters Panel

97.up.agc

w. _mAsuREo

DEPTH

o

~ Figure D2b. Upgoing data from 1997 site lP-wave source after F-K filter.

I d9 96984’

mv

I

225

222

219

216

213

210

2Q7

21J4

2XJ1

19E

195

192

1E!3

la6

la3

laO

177

174

171

254-

246

23EI

230

222

214

104 139

,1

~45

,a~

133

177

222

2B8

310

354

39B

442

174

/’‘%

/

II 35 711 1 D4 13E 174

FREQUENCY {Hz)4

Figure D3a. I?requencyvs time (top) and spectra (bottom) plot for average oftwo

farthest 1996 walkaway P-wave sites for sensor at 984 ft.

.

. ... -—-

dB 961016’227

224.

221

21B

215

212

209

206

203

200

197

lw-

191

180

185

1 ai!

179

176

173

25$

247

23$1

231

223

215

35 1 b4- 13B

.1

45

Do

134

173

223

267

311

355

40a

444

\

174

0 35 70 104- 13D 174

FREQUERIX [Hz)

Figure D3b. Frequency vs time (top) and spectra (bottom) plot for average of two

farthest 1996 walkaway P-wave sites for sensor at 1016 ft.

—

i

mu

227

224

221

21s

215

21!2

2D9

Zclf!

20$

200

107

1 B4

1!31

1 E&

1 H$

1 H2

17fi

17%

173

256

24’R

24’11

232

224

216

.3

~47

,91

.135

.Iml

,224

~26%

.312

357

,4Q1

~445

0 35 70 la4 139 -174-

, I I I II

v 35 70 1(I4 139 174-

FREQUENCY [Hz]

Figure D3c. Frequencyvs time (top) and spectra (bottom) plot for average of two

farthest 1997 walkaway P-wave sites for sensor at 984 ft.

—

229

220

223

220

217

2i4

211

M2(I8

205

Fl202

199

1m

193

190

187

184

181

178

175

257

249

241

2s 3

m-9 225

217

c1 35 70 104 1“39 i

2

45

91

1xi

i 80

224

281

313

357

402

446

74

0 35 70 1(34 Ii!l 174

FRECIUEHCY [Hz)

Figure D3d. Frequency vs time (top) and spectra (bottom) plot for average of two

farthest 1997 walkaway P-wave sites for sensor at 1016 ft.

.

Appendix E - Surface Seismic Analysis

,,>

. —. - ,-. . . .. . .. . .. ., . .,>.... ..s ., ...,.- 7 --- . . . . . . . . ., .,,.- . . ...’.-—- —- —.. — . . . . .

,.

1260000

1256000

1252000

1248000

124400C

124000(

123600(

123200(

122800(

122400

0-12

r

—64!

use

. . _.. _..

65 I , 1 t

I 1 ,I

1,

,: ,

w,-, t

m t ,b-l I 1 1

,1 1 1

I I 1 1 ! 1

@+ , 1

, , 1. I I 1

~

lo .—, t1

,1 t

-a

,1

, , ,.

z- 1 7

LIY : :I

I ; t —

1

%

! I ,t 1 t 1 I

U-J-=!

;1 18

1,

I, I

w1 t

m ,

amsm I I t 1 : t

GQI 1 r , ,

63 I ,

(9 I 1 1,

1,

t1 1

,1

,

-

w I

w I , 1, 1

w I I I I 1 I 1—

It , r —.

m f , ,

; Exit File Qisplay analysis ~ick.

H elr

~80 d i58~ After DSKRDCDP 576 588 592 596 ~ ~@ ~: 619OFFSET 138 139 193 179 1?2

623l;2~ 96

627 631 635

0.00 \137I 28 194 247 137

I 1 I 1 1 1 f I I 1 I [ ! I I I f I r I Iil’[I I [ I 1 I I Ill 1 I I I I J t t ) 1 I ) 1 1 I I 1 t I 1 0.00

0.10

H“’$+i+wi

0.20

0.30 ----*

o.40-

0.60-

?-

.--0.1

--0.20

--0.30

--0.40

--0.50

--0.60

.

Figure E-3. CDP stacked sections for surface lines RC-1 (left, CDPS 576- 596) and ~

RD-5 (right, CDPS 618- 638) with the well 157 site 1 VSP P-wave upgoing stack displayed

in the middle (labeled as CDP 1). The top of the Trenton reservoir formation is at about

0.18 S.

DE

Pth

(FPEt

1

0.

200 ~

400.

600.

800-

looo -

1200 -

1 2 3 4 5 50 0 0 0 50 0 0 : 0 0

0 0 0 0 0 0 0I I I 1 t I I I 1 1 I t I I I I 1 I I I I I I I I I t I I

file: maf-gascdp, date: Tue Jun 23 10:05:27 1998, user:

...-..-’ ,