ORIGINAL PAPER

Active and Passive Seismic as an Indicator of LargeEquipment Interactions with the Oil Sand

A. D. Sharifabadi • T. G. Joseph • D. R. Schmitt

Received: 4 January 2010 / Accepted: 16 May 2010 / Published online: 30 May 2010

� Springer Science+Business Media B.V. 2010

Abstract The strain softening of oil sand in the

underfoot of ultra class mobile mining equipment, due

to the loading action of large mobile mining equipment

such as trucks and shovels, yields a highly unstable

condition for the operation of this ultra-class equipment.

Soft ground conditions in oil sand, due to the low

stiffness of the material a condition especially present in

the summer, can cause high rack, pitch, and roll in

trucks, leading to fatigue failure in structural compo-

nents. For shovels, poor ground stability can cause twists

in car bodies and undercarriages, resulting in major

damages. Track and shovel frame failures due to this

instability result in high maintenance costs. The authors

carried out a geophysical study of the oil sand in order to

evaluate the ground conditions under large mobile

mining equipment. A geophysical investigation per-

formed in summer 2001 encountered 6–8 m of thick soft

material, commensurate with very low velocities,

caused by the loosening of the surface material by

heavy mining machinery and excavation; and a transi-

tion zone of up to 25–26 m depth approaching the in situ

oil sand below. The depth of the oil sand zones can be

calculated by using the refraction analysis technique.

Spectral Analysis of Surface Waves was used to

estimate the ground stiffness. A new technique is

proposed to evaluate the changing ground stiffness

during the use of ultra-class mobile mining equipment.

Keywords Oil sand � Surface wave �Large mobile mining equipment � Ground stiffness

1 Introduction

Trucks and shovels dynamically load the ground during

the normal course of operations. The degree of impact

on the ground depends very much on its initial condition.

Soft ground, such as clay or oil sand, exhibits visco-

elastic–plastic behaviour, which on immediate unload-

ing, results in considerable permanent deformation. The

deformation is somewhat alleviated after a lengthy

period of relaxation. Joseph et al. (2003) reported that

after only a few operation cycles, trucks and shovels

operating on soft ground become less stable.

The oil sand deposits of northern Alberta are mined

using ultra-class scale ([325 tonne capacity trucks

and [46 m3 capacity shovels) bulk handling equip-

ment in an open-pit setting. Oil sand has specific

and unusual properties that allow it to behave like

A. D. Sharifabadi (&)

Geotechnic, AMEC Earth and Environmental, Calgary,

AB, Canada

e-mail: dehmoobed@hotmail.com

T. G. Joseph

School of Mining Engineering, University of Alberta,

Edmonton, AB, Canada

e-mail: tim.joseph@ualberta.ca

D. R. Schmitt

Department of Physics, University of Alberta, Edmonton,

AB, Canada

e-mail: doug@phys.ualberta.ca

123

Geotech Geol Eng (2010) 28:727–743

DOI 10.1007/s10706-010-9335-3

sandstone in winter and weak clay in summer as the

temperature varies from -30 to ?30�C, respectively.

Standard refraction analysis techniques were used

to demarcate the oil sand zones with different

stiffness values (see Fig. 1). Lower velocities near

the surface indicated the higher degree of disturbance

caused by operating machineries; creating a loose or

mobile zone. Higher velocities at deeper layers were

commensurate with a more compact and undisturbed

nature of at-depth material, described as ‘‘transition’’

and deeper still ‘‘in situ’’ zones. It has been reported

that after only a few cycles, trucks and shovels

operating on progressively softening ground become

less stable (Joseph et al. 2003; Welz and Schmitt

2002).

The objective of this study was to investigate oil

sand behaviour in terms of ground stiffness or shear

modulus due to the operational loading by mining

equipment. Another important objective was to

develop a better understanding of the in situ proper-

ties of oil sands. Geophysical methods incorporating

passive seismic data analysis were used to understand

the interaction between the ground and heavy mining

equipment. The mining equipment was passively

recorded as the ground motion source and correlated

with ground reactions to dynamic loading from large

mobile mining equipment.

2 Site Characterization, Equipment Selection

and Source

Alberta’s oil sand deposits are the biggest oil sand

reserve in the world and represent a major source of

oil. Canada’s crude bitumen exists entirely within the

province of Alberta and is found in sand and

carbonate sedimentary formations in three regions:

the Athabasca, Cold Lake, and Peace River oil sand

areas.

Syncrude Canada Limited is the world’s largest

producer of synthetic crude oil from oil sands and the

largest single source producer in Canada. Syncrude’s

operations are located just outside Fort McMurray in

the Athabasca Oil Sands.

Summer and winter refraction data measuring the

seismic properties of in-situoil sand was collected

from North Mine; near Mildred Lake on lease 17, and

the Aurora mine about 35 km to the north. The three

bitumen grades present at the Syncrude mines, 6, 8,

and 13.5%; were considered as part of this experi-

mental seismic refraction survey.

The summer experiment gathered seismic surfaces

wave using 14 Hz geophones at 1 m spacing. This

geophones spacing was chosen to avoid spatial aliasing

of wave lengths as small as 2 m (Nyquist wavenumber

k = 2p/kmin = p). This small wavelength contains

V1 = 380 m/s h1 = 4 m

V2 = 790 m/s h2 = 8 m

V3 = 1500 m/s

Mobile or soft zone

Approaching in situ zone

Transition zone

Fig. 1 Ground condition and corresponding velocity profile (Sharif-Abadi 2006)

728 Geotech Geol Eng (2010) 28:727–743

123

information about the shear wave velocity of the first 1

to 2/3 m. The sampling was collected by a 240-channel

geometrics geode seismic recorder at a sampling rate of

0.250 ms, providing a Nyquest frequency of 2,000 Hz

for refraction data. The passive seismic configuration

used was 4 ms sampling was conducted for a 65 s

record in order to obtain a complete picture of a truck

passing through a linear array of seventy-two 14 Hz

geophone sensors. The refraction source was an

accelerated weight drop for passive seismic and used

for both summer and winter data collection.

Passive sources are any sources that cause ground

motion over which there is little or no control, such as

road traffic, wind causing movement of objects, or

people walking. In this study the motion of mining

equipment such as trucks or shovels was the passive

source for ground motion.

3 Summer Refraction Data Acquisition

and Analysis

Seventy-two vertical 14 Hz geophones were placed

along a single line at 1 m spacing and, at an ambient

temperature of 28�C in an active mining area. The

source used was an accelerated weight drop and the

sampling rate was 0.250 ms. Data was recorded using

a Geometrics system and subsequently processed

using the Vista-Windows Seismic Processing soft-

ware. The passive record sensed high levels of noise

due to the heavy equipment operation. This noise was

filtered to remove the background influence of

motors, pumps, and tire treads from the gross loading

action of the equipment.

The data acquired showed high attenuation com-

mensurate with the expected greater fluidic nature of

the oil sand material at warmer temperatures. An

analysis of the refraction data showed that the

attenuation generally increased with the bitumen

content (increasing % grade by volume). Figures 2, 3,

and 4 illustrate examples of the seismic offset time

shot for operations on 6, 8, and 13.5% oil sand and

the corresponding refraction analysis. Figure 5 shows

the AGC (automatic gain control) plot of the unscaled

plot of the 13.5% oil sand with high level attenuation.

Table 1 shows the overall results of the refraction

data analysis by percentage bitumen content.

Forward and reverse refraction methods were used

to estimate the zones thickness in both directions. The

method employs refraction of wave arrivals from

shots offset in opposing directions targeting a given

receiver, having left the refracting surface from

effectively the same lateral location. The refractor

velocity and depth below the receiver location can be

determined for any receiver that records forward and

reverse refracted arrivals.

During refraction field tests, refraction data was

shot to a line of geophones from both ends and the

first arrival p-waves from both directions was

identified. Figure 6 illustrates how the forward and

reverse method allowing identification of arrival

times. It is obvious that the transition zone is

evident from the first 10 m of geophone response,

where the material properties vary considerably. The

truck travel route, Fig. 7, likely caused the variance

for the last 7 m of geophones (65 through 72),

where the actual route diverged from the geophone

line.

4 Determination of Ground Stiffness by Using

a Geophysical Technique with Summer Data

The velocity of oil sand is needed in the inversion

method to estimate the ground stiffness profile. An

inversion method was used on the summer data to

estimate the ground stiffness profile. This approach

may be used for ground and equipment performance

modeling during different seasons of the year. The

relative ground modulus changes during mining

activity can potentially be determined. The prelimin-

ary study of passive data revealed that the ground

modulus drops after a few truck runs (as explained in

the passive seismic data analysis).

Characterizing the near subsurface using geophys-

ical methods has been of great interest in recent years

since these methods are cheaper and faster than

conventional drilling and borehole logging. Being

able to determine a modulus-depth profile without the

aid of boreholes via surface wave geophysics is seen

as being highly advantageous. The most powerful

tools for evaluating the subsurface are spectral

analysis of surface waves (SASW) via non-intrusive

Rayleigh waves (Matthews et al. 1996, Haegeman

and Van-Impe 1999).

Knowing the properties of soil layers and having a

profile are the key factors in analyzing an overlying

dynamic loading source. SASW is currently used

Geotech Geol Eng (2010) 28:727–743 729

123

primarily for the evaluation of subsurface wave

velocity profiles.

The results obtained from this method represent of

the average properties of a relatively large mass of

soil. This method can be very cost-effective for

investigations (Haegeman and Van-Impe 1999 and

Ganji et al. 1998). An alternative technique for

determining ground stiffness is continuous surface

wave (CSW) analysis Continuous Surface Wave

analysis relies on the propagation properties of

vertically polarized seismic surface waves, where

the penetration depth by a surface wave is dependent

on the wavelength and frequency (Moxhay et al.

2001).

Shear wave velocity is a factor in identifying the

shear strength of a given formation. Rayleigh wave

dispersion has been used as a method for evaluating

the shear modulus of near-surface materials. Using

Rayleigh waves to obtain the shear wave velocity has

two steps; (a) finding the dispersion relationship for

the Rayleigh wave, and (b) applying the inverse

procedure to convert the dispersion curve to the shear

wave velocity versus the depth (Beaty and Schmitt

2003).

4.1 Determine Stiffness with Rayleigh Waves

One technique uses the surface dispersion curve to

determine the near-surface stiffness. A tool that has

increasingly been used to evaluate the shear modulus

of near-surface materials is Rayleigh surface-wave

dispersion (Stokoe and Nazarian 1985; Rix et al.

Fig. 2 Shot gather (time-

offset) in the 6% bitumen

oil sand and p-wave

velocities

730 Geotech Geol Eng (2010) 28:727–743

123

2001; Louie 2001). Usually the dispersion curves for

Rayleigh waves are used to examine the variability of

near-surface properties up to 15 m deep Rayleigh

waves travel along the earth-air interface and usually

contain more energy than body waves and are able to

provide substantial information on shear wave veloc-

ity, identifying structures of the near-surface (Beaty

2000; Ewing et al. 1957).

The dispersion nature of Rayleigh waves and the

velocity depend on its frequency. The Rayleigh wave

phase velocities depend primarily on the shear-wave

velocity structure of the near-surface materials.

Lower frequencies, or longer wavelengths, have deep

penetration, which is appropriate for determining the

depth stiffness (these phenomena contain information

about the deep layers). On the other hand, high

frequencies or lower wavelength Rayleigh waves are

appropriate for determining the upper layer stiffness.

A rule of thumb is that the depth of the relation of

Rayleigh waves is one-half of the wavelength.

Dispersion curves show the velocity of the wave at

each wavelength or frequency. These can be inverted

to obtain the shear wave velocity profiles of an area.

One method to get dispersion curves from Rayleigh

waves is to carry out Tau–p (s–p) transformation on

the data, followed by a one-dimensional Fourier

transformation along the s direction. s–p transforms

essentially carry the sum of the amplitudes along a

line in the offset-time domain (seismic data) with

intercept time s and slope p called the ‘‘slowness’’

(slowness is the inverse of velocity). This sum will

map onto a point (s–p). This is a simple, well-known

linear wave field transformation that takes an input

data set in the time-offset domain and transforms it

Fig. 3 Shot gather (time-

offset) in the 8% bitumen

oil sand and p-wave

velocities

Geotech Geol Eng (2010) 28:727–743 731

123

into a new data set in the intercept time—slowness

domain. In this case, the input data set, p, is the

observed seismogram wavefield. In S, the trans-

formed wavefield, p is the horizontal slowness and sis the time intercept.

Sðs;PÞ ¼Zþ1

�1

Pðsþ px; xÞdx ð1Þ

A simple way to look at the transform is to think of

each point in the s–p plane being the sum of all the

points in the t–x plane lying along a straight line with

a slope of p and a time-axis intercept of s. The

seismogram is decomposed into plane wave elements

(McMechan et al. 1982; McMechan and Yedlin 1981;

Louie 2001).

After the dispersion curve has been attained, it

must be inverted to a shear wave velocity profile.

Figure 8 shows the sequence of obtaining the

dispersion curve for the seismic data (Beaty 2000,

Matthews et al. 2000).

For inversion, the p-wave velocity and the thick-

ness of the layer are needed.

Obtaining the phase velocity dispersion curve

from Rayleigh waves requires two steps:

1. Isolate the Rayleigh wave from the other arrivals

on the seismograph (this step can be performed

by windowing the surface data in either the

offset-time or frequency-wave number domain

(f–k)).

2. Extract the dispersion curve. Different methods

can be used:

Fig. 4 Shot gather (time-

offset) in the 13.5%

bitumen oil sand and

p-wave velocities (high

level attenuation)

732 Geotech Geol Eng (2010) 28:727–743

123

Fig. 5 AGC scaled plot for

the 13.5% bitumen oil sand

Table 1 P-wave velocity

and thickness of summer

operating condition oil sand

Oil sand depth descriptor Oil sand grade (%) P-wave (m/s) Thickness (m)

First zone (loose) 6 1,200 2.25

Second zone (in situ) 1,500 Continuous

First zone (loose) 8 410 9

Second zone (transition) 1,100 28

Third zone (in situ) 1,950 Continuous

First zone (loose) 13.5 380 6.7

Second zone (transition) 790 26

Third zone (in situ) 1,500 Continuous

0

0.02

0.04

0.06

0.08

0.1

0.12

0 10 20 30 40 50 60 70 80

Offset - Geophones (m)

Tim

e (S

ec)

y = -0.002x + 0.1507R2 = 0.9974

y = -0.0011x + 0.0977R2 = 0.9993

y = 0.0021xR2 = 0.995

y = 0.0011x + 0.0136R2 = 0.9996

Reverse Forward

Fig. 6 Forward and reverse

method for summer

refraction data (p-wave)

Geotech Geol Eng (2010) 28:727–743 733

123

• Exploit the phase information of the Fourier

transformation of the surface wave (used in

earthquake seismology).

• Use dispersion information to locate the f–k

spectrum or frequency-slowness representing

the data (used for an array of multiple

geophones):

• Perform a s–p transformation on the data

followed by one-dimensional Fourier

transformation along s direction.

• Pick the peaks associated with the surface

wave energy in the f–k domain.

4.2 Dispersion Curve for Summer Data

Determination of the s–p transform of the surface

wave dataset is the first step in the method described

in McMechan and Yedlin (1981) for obtaining

dispersion curves from a multichannel seismogram.

Both s–p and f–p transformations were carried out on

the summer data (see Fig. 9).

In the t-offset domain followed by the s–p

transform, the f–p display and a reconstruction of

the data and the data reconstructed from the s–p have

preserved the information contained in the data well.

The difference between the original data set and

reconstruction will show if the original data set is

aliased or if the initial value for s and p was not

correct. Darker areas on the f–p map indicated a

concentration of energy related directly to the

dispersion curves.

4.3 Inversion

One technique employed by Beaty (2000) uses the

surface dispersion curve to determine the near-

surface stiffness. To obtain a reasonable S-wave

profile, good estimates of the values for P-wave and

density profiles should be sufficient. The forward

modeling technique was used to obtained theoretical

dispersion curves in an elastic, layered medium and

followed by the outline inversion technique to obtain

the velocity profile from the measured dispersion

curve. Beaty (2000) used a matrix propagator method

to obtain theoretical dispersion curves for Rayleigh

waves in a vertically varying medium consisting of a

set of n homogeneous layers overlying a homoge-

neous half-space. To invert the dispersion curve data

for the S-wave velocity profile, the forward

model was incorporated into a simulated annealing

algorithm.

After the dispersion curve has been attained, it

must be inverted to a shear wave velocity profile.

Even though both summer and winter data was

aliased, the summer data showed less alias from the

s–p graph. Therefore, an inversion was carried out on

the summer refraction data using a simulated anneal-

ing algorithm (Beaty 2000).

The p-wave and S-wave velocities are related to

the elastic constants through the relations

Vp ¼

ffiffiffiffiffiffiffiffiffiffiffiffiffiffikþ 2l

q

s

Fig. 7 Trucks route

Detection of motion on the ground surface

Dispersion curve: Phase velocity of Rayleigh wave versus frequency

Variation of shear wave velocities with depth

Small strain stiffness profile (G0 versus depth)

Processing

Inversion

G0=ρ.Vs2

Acquisition

Fig. 8 Sequence of obtaining shear wave velocities

734 Geotech Geol Eng (2010) 28:727–743

123

Vs ¼ffiffiffilq

r

where q, density; l, shear modulus; k, Lame’s first

parameter. The input for the model is P-wave velocity

profile, density, layer thickness.

Table 2 illustrates the results of the inversion and

calculation of elastic modules based on the shear

modulus. The elastic modulus results are very similar

to the previous triaxial test results presented by

Joseph et al. (2003).

5 Summer Passive Data

Ninety-six geophones, 72 in one line and 24 in a

perpendicular decay line were set-up for passive and

refraction data acquisition (see Fig. 10). The distance

Slowness (s/m)

Freq

uenc

y (H

z)

f - p Transform

0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02

0

20

40

60

80

100

120

140

t - offset

20 40 60

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

τ - p

0.01 0.02

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

t - offset (reconstructed)

20 40 60

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

Fig. 9 Summer data

f–p and s–p graphs

Geotech Geol Eng (2010) 28:727–743 735

123

between each geophone was set at 1 m, with trucks

passing within 1–2 m of the line. Three site locations

of varying bitumen content and corresponding initial

softness were evaluated.

The sum of the amplitudes in line A of 72

geophones for trucks passing was calculated for each

location. Figure 11 shows that the sum of the

amplitudes increased as the truck velocity increased.

The same phenomenon as in the winter data was

seen; the sum of the amplitudes increased at the

beginning and then suddenly decreased after a

number of cycles (see Fig. 12).

6 Winter Refraction Data

In the case of winter data acquisition, where the

ambient temperature was -38�C, the top layer of oil

sand was frozen to a depth of several metres, yielding

higher velocities compared to those beneath. No

reliable P-wave refraction was discernable from this

data, and the spacing of 3 parallel lines of 24

geophones at 3-m intervals (Fig. 14) was not enough

to attain refraction from the stiff ground. Spatial

aliasing was seen in all data sets and was probably

caused by insufficient distance between geophones

and the thin frozen layer of oil sand at the surface.

Noise was a problem due to the presence of active

heavy machinery. The high noise levels were present

even in the geophones furthest from the shot point.

The surface waves were very strong in this frequency

Fig. 10 Summer test set-up

Table 2 Presents the inversion results

Changing oil

sand property

Oil sand

grade (%)

P-wave velocity

(m/s)

Thickness

(m)

S-wave velocity

(m/s)

Shear modulus

(MPa)

Elastic modulus

(MPa)

First zone (mobile/soft) 13.5 380 6.7 137 37 94

Second zone (transition) 790 26 236 120 301

Third zone (in situ) 1,500 Infinity 248 136 339

y = 1E+08x - 2E+08

R2 = 0.9072

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

7 7.5 8 8.5 9 9.5

Velocity (m/s)

Am

plit

ud

e *1

0E9

(Vo

lt)

Fig. 11 Sum of the amplitude versus truck velocity for an

unloaded truck

736 Geotech Geol Eng (2010) 28:727–743

123

range, as shown in Fig. 13. However, the direct

waves gave a range of velocities for the frozen layer

of 1,500–1,700 m/s. Figure 13 illustrates the direct

wave of a winter data set.

7 Passive Data Acquisition

Winter data was acquired using a 3-line array of

geophones and the moving trucks as a source of

vibration. The purpose of the test was to (a) evaluate

the interaction between the ground and mobile

mining equipment under firm ground conditions

and, (b) to show that seismic equipment and survey

techniques could monitor interactions. Figure 14

shows the winter test set-up (an advance over the

original summer test set-up used in Fig. 7) using 3

parallel lines, each having 24 vertical 14 Hz geo-

phones placed with an instrument spacing of 3 m,

giving a test array of 72 m length, similar to the

0

2

4

6

8

10

12

14

16

18

0 10 20 30 40 50 60 70 80

Geophone #

Am

plit

ud

e *1

0E6

(Vo

lt)

Last Truck Passing

First Truck Passing

Fig. 12 Changing sum of

amplitudes in each

geophone

Fig. 13 Winter surface and

refraction data

Geotech Geol Eng (2010) 28:727–743 737

123

original summer configuration. The geophone acqui-

sition system was set at a 250 Hz acquisition rate.

The only source of vibration was set as a controlled

single truck passing through the array, so that a few

seconds before the truck reached the array, the

geophones would be manually activated to start

recording data. As expected, the middle line B had

greater amplitude compared to that of lines A and C

(see Figs. 15, 16) for a single run. Figure 16

illustrates the maximum amplitude of each geophone

for different runs. The results also showed that the

rear of the truck put more energy into the ground due

to the truck’s approximately 2/3:1/3 rear to front axle

load distribution.

7.1 Winter Passive Data Analyses

Figure 17 shows plot of winter passive data; the truck

velocity is determined by the slope of the inclined

line. Figure 18 illustrates the velocity of the truck for

the entire test (trucks passing 50 times through the

line). As was expected, the passive data shows that a

truck traveling at a slow velocity has a higher band of

amplitude than that of a faster truck yielding a

reduced band (Ws [ Wf) (see Fig. 19).

The geophones’ amplitude from units in the same

column was calculated, (e.g., 1, 25 and 49) giving an

accurate correlation between the velocity of the truck,

and the sum of the amplitudes. Figure 20 shows the

Fig. 14 Test set-up for

passive seismic winter data

acquisition

Fig. 15 Passive data shot acquired for a single run of truck

738 Geotech Geol Eng (2010) 28:727–743

123

correlation between truck velocity and the sum of the

amplitudes.

From Fig. 20 it can be concluded that trucks with

higher velocity transfer more amplitude or energy

into the ground. The time between two consecutive

runs is termed the relaxation time. The relaxation

time for 50 runs performed during the test was not

constant. Since oil sand in summer behaves like a

viscous elastic material, one important parameter

related to the degree of permanent ground deforma-

tion relative to the next truck passing over the same

ground is the relaxation time. The tests yields four

levels of time gaps between runs (see Fig. 21),

although the expected effect on ground performance

was not thought to be significant for the stiffer ground

in winter.

The extent of ground deformation can be estimated

by normalizing the sum of the amplitude across the

0

2

4

6

8

10

12

1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69

Geophone (#)

Am

plit

ud

e (1

04 )

1 2 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

Line B (Geophone # 25 to 48)

Line A (Geophone #1 to 24)

Line C (Geophone # 49 to 72)

Fig. 16 Max amplitude for each geophone for 50 runs of truck

Fig. 17 Truck velocity

from passive data

Geotech Geol Eng (2010) 28:727–743 739

123

width (from the summing up procedure) by the sum

of total amplitude for the same run. The more

amplitude or energy transferred into the ground, the

lower the ground stiffness. Less energy corresponds

to harder ground conditions. A study compiled from

each of the 4 relaxation groups shown in Fig. 21,

proved that after a number of runs for each group, the

total energy first increased, corresponding to the

progressively softening ground. However, with addi-

tional runs, the energy suddenly leveled off to a

constant value. This lead to a tentative conclusion

that after several runs the ground reached a stable

condition and higher stiffness in winter condition (see

Fig. 22). Unlike results of energy increases being

parallel for all groups, this condition was certainly

more predominant in later runs of the test (group 4)

than in group 2. Figure 23 is normalization of Fig. 23

with the maximum amplitude in each column of

geophones (percentage of the maximum amplitude

0

2

4

6

8

10

12

1 6 11 16 21 26 31 36 41 46

# Run

Vel

oci

ty(m

/s)

Fig. 18 Simple truck velocity for winter passive data

Fig. 19 Slow truck

velocity versus fast truck

velocity

740 Geotech Geol Eng (2010) 28:727–743

123

value) and different lines representing progressive

relative ground stiffness changes by run and in turn

grossly representing parameters that may affect the

degree of softening: truck velocity, ground profile,

and relaxation time.

To evaluate onboard the truck’s onboard informa-

tion system’s (VIMS) ability to predict poor ground

conditions, the seismic-data-acquisition system was

set at 250 Hz (1/250 Hz = 4 Ms). However, this

setting was mismatched with the OEM on-board data-

acquisition system (VIMS) at 1 Hz, so making a

correlation between the two sets was difficult.

y = 2E+07x + 1E+08R2 = 0.882

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10.5 11.5

Velocity (m/s)

Am

plit

ud

*10

E8

(Vo

lt)

Sum 1 Sum 2 Sum 23 Sum 24Sum 3

Total Sum

1 24

25 49

50 72

Fig. 20 Sums of

amplitudes for each array

position

0:00:00

0:14:24

0:28:48

0:43:12

0:57:36

1:12:00

1:26:24

1:40:48

0 10 20 30 40 50

# Run

Tim

e (M

in)

Group 2 Group 3 Group 4Group 1

h:min:s

Fig. 21 Relaxation time between runs

Group Two (4-20)

0

5

10

15

20

25

21161161

Geophones

Am

plit

ud

e *1

0E6

(Vo

lt)

Group 2Group 3

Fig. 22 Sum of geophones

columns amplitude versus

relaxation group

Geotech Geol Eng (2010) 28:727–743 741

123

Studying the trends of the rack, pitch, and roll

(Joseph and Welz 2003) from the predominant truck

motion did however provided some correlation. The

difference in energy was determined via the sum of

the truck’s front-to-back energy (amplitudes) using

the seismic data. Figure 24 shows that the rack

yielded a reasonable shape correlation, indicating an

influence on the ground.

Figure 25 provides a closer look at the passive

data. Here, the front-to-back motion is evident from

the curves caused by front and back of truck tire due

to the undulated ground and poor ground condition.

However, devising an algorithm to separate the front

and back motion or to distinguish the path of the

curve apex was difficult for several reasons. The time

of the front-to-back motion depends on the velocity,

which may be extracted from the passive data, but

only as an average value. Also, the start time is

manually activated, so that distinguishing the exact

time of passing adjacent to any given geophone was

made very difficult in post processing.

8 Conclusion

The collected data reveals that the near surface

material loosen by heavy mining machinery and

excavation and that it has lower velocity. The deeper

material is more compact and undisturbed due to a

change in Bitumen content of the oil sand.

In summer, the p wave velocity of loose oil sand

material is between 300 and 800 m/s where the p

wave velocity of in situ oil sand is between 1,500 and

2,000 m/s. The shear modulus of the soften oil sand

(13.5%) due to the mobile mining equipment activ-

ities varies from 30 to 140 MPa. The elastic modulus

of the 13.5% oil sand was shown to be predictable

from passive seismic analysis at approximately

90 MPa for lose material and 400 MPa for in situ

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

12611161

Geophones

Per

cen

t (%

)

1 2 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

Sum 1 Sum 2 Sum 23 Sum 24Sum 3

1 24

25 49

50 72

Fig. 23 Comparing the amplitude normalization for different runs

-8

-6

-4

-2

0

2

4

6

1.0 3.0 5.0 7.0 9.0 11.0

Time (Sec)

Rac

k ca

lcu

late

d f

rom

sei

smic

am

plit

ud

e (*

10E

6)

-20

-15

-10

-5

0

5

10

15

20

Rac

k ca

lcu

late

d f

rom

VIM

S d

ata

(kP

a)

Seismic Truck

Fig. 24 Comparing the seismic data and VIMS data

742 Geotech Geol Eng (2010) 28:727–743

123

oil sand. In winter, the p wave velocity of the frozen

oil sand layer is close to in situ oil sand at

approximately 2,000 m/s.

There is a correlation between the loading action

of a truck and the information collected by the trucks

onboard information system measurements of rack,

pitch, roll, and bounce. The ground softening due to

mobile mining equipment can be correlated to

seismic reaction.

Acknowledgments This study was financially supported by

James Progithin International Ltd and the University of

Alberta. The authors are grateful to the significant help of

Dr. Kristen Beaty and Marek Welz to this research.

References

Beaty KS (2000, Fall) Determination of near surface variability

using Raleigh waves. MSc Thesis, Department of Physics,

University of Alberta, Edmonton, 208 pp

Beaty KS, Schmitt DR (2003) Repeatability of multimode

Raleigh-wave dispersion studies. Geophysics 68(3):782–

790

Ewing WM, Jardetzky WS, Press F (1957) Elastic waves in

layered media. McGraw-Hill, New York

Ganji V, Gucunski N, Nazarian S (1998) Automated inversion

procedure for spectral analysis of surface waves. J Geo-

tech Geoenviron Eng 124(8):757–770

Haegeman W, Van-Impe WF (1999) Characterization of dis-

posal sites from surface wave measurements. J Environ

Eng Geophys 4(1):27–33

Joseph TG, Welz M (2003) Mechanical action and geophysical

reaction: equipment—oil sand interactions. In: Proceed-

ings of CIM coal and soft rock workshop, Saskatoon,

October 3–5, 10 pp

Joseph TG, Sharif-Abadi AD, Shi N (2003) A broken material

approach to modelling oil sand under dynamic load. In:

CD Proceedings of computer applications in the minerals

industries (CAMI), Calgary, September 8–10, 12 pp

Louie NJ (2001) Faster, better: shear-wave velocity to 100

metres depth from refraction micrometre arrays. Bull

Seismol Soc Am 91(2):347–364

Matthews MC, Hope VS, Clayton CRI (1996) The use of

surface waves in the determination of ground stiffness

profiles. In: Proceeding of the institution of civil engi-

neering—geotechnical engineering, vol 119, 2, pp 84–95

Matthews MC, Clayton CRI, Own Y (2000) The use of field

geophysical techniques to determine geotechnical stiff-

ness parameters. In: Proceeding of the institution of civil

engineering—geotechnical engineering, vol 143, 1, pp

31–42

McMechan GA, Yedlin MJ (1981) Analysis of dispersive

waves by wave-field transformation. Geophysics 46:869–

874

McMechan GA, Clayton RW, Mooney WD (1982) Application

of wave field continuation to the inversion of refraction

data. J Geophys Res 87(B2):927–935

Moxhay AL, Tinsley RD, Sutton JA (2001) Monitoring of soil

stiffness during ground improvement using seismic sur-

face waves. Ground Eng Mag 34(1):34–37

Rix GJ, Lai CG, Foti S (2001) Simultaneous measurement of

surface wave dispersion and attenuation curves. Geotech

Test J 24:350–358

Sharif-Abadi AD (2006) Cyclic performance of soft ground.

PhD thesis, University of Alberta, 267 pp

Stokoe KH, Nazarian S (1985) Use of Rayleigh waves in liq-

uefaction studies. In Woods RD (ed) Measurement and

use of shear wave velocity for evaluating dynamic soil

properties: Am Soc Civ Eng, pp 1–17

Welz M, Schmitt DR (2002) Passive seismic as an indicator of

larger equipment interaction with oil sand. In: 104th CIM

AGM, Vancouver April 27–May 1

Fig. 25 Closer view of the

passive data

Geotech Geol Eng (2010) 28:727–743 743

123