C H A P T E R

2
Understanding Seismic
Wave Propagation

O U T L I N E

Types of Seismic Waves 18 Understanding Convolution 27

Body Waves 18
Cross-Correlation 29
Wave Parameters 20
Autocorrelation 30
Seismic Reflection 21
Time Domain 30
Law of Reflection 21
Law of Refraction 22 Frequency Domain 30
Critical Angle 23 Fourier Transform 31
Reflection Coefficient 24 Inverse Fourier Transform 31
Transmission Coefficient 26

The oil industry uses seismic reflection measurements to gain knowledge about the geo-
logical structures in the subsurface in other to locate oil and or gas reservoirs. This method
works by sending an acoustic or pressure wave into the earth, which get reflected back
when it meets a geological boundary where there is contrast in rock properties. An example
of geological boundary is a boundary between sand and shale layers. This wave is then
reflected back to the surface after a few seconds and it is then recorded by a receiver.
The time that it takes for the wave to return to the surface tells us how far the geological
boundary is located (Figure 2.1).
Seismic survey is based on the theory of elasticity and therefore tries to deduce elastic prop-
erties of rocks, such as acoustic impedance, Poisson’s ratio and Vp/Vs ratio, by measuring
their response to elastic disturbances called seismic (elastic) waves.

Seismic Data Analysis Techniques in Hydrocarbon Exploration 17 # 2014 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/B978-0-12-420023-4.00002-2
18 2. UNDERSTANDING SEISMIC WAVE PROPAGATION

Source Receiver

Geological boundary
0.2 s
Sand 0.2 s n1 r1

n2 r 2

Shale

FIGURE 2.1 Conceptualized reflected seismic ray.

TYPES OF SEISMIC WAVES

Two main types of waves are considered in seismic exploration: body wave and surface
wave. Surface waves move along the surface of the earth and they are of lower frequency than
body waves.

Body Waves
Body waves propagate through the earth’s inner layers, and they are further classified as
P waves and S waves. These waves are of higher frequency than surface waves.
P Waves
P waves are formed when energy is applied exactly at right angles to a medium. Particle
motion under the influence of the wave is then in the direction of propagation of the wave. As
a result of the particle motion, the rock particles are alternatively compressed and rarefracted
or pulled apart as the waves propagate. P waves are also known as compressional waves,
because of the pushing and pulling they do. They are the fastest kind of seismic waves.

Rarefraction Compression
Particle motion

P wave

Direction of travel of the wave

S wave

Particle motion

FIGURE 2.2 Conceptualized P- and S-wave propagation. Source: USGS.

I. BASIC SEDIMENTOLOGY AND SEISMIC DATA ACQUISITION

 TYPES OF SEISMIC WAVES 19
In seismic processing, one generally assumes that we are dealing with P waves (Figure 2.2)
only, travelling through isotropic and homogeneous material.
If the elastic moduli (K) and the bulk density (r) of the rocks are known, P-wave velocity
can be calculated using the following equation:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
K þ 4=3m
Vp ¼
r
CHARACTERISTICS OF P WAVES
• It is a sound wave or pressure wave.
• It is always the first arriving seismic wave.
• Its velocity is the fastest of all seismic waves.
• Its particle motion is in the direction of propagation of the wave.
• It can propagate in solids or fluids (oil, gas and water).
S Waves
S wave, also called shear waves, is the second type of body wave. S wave is formed when
energy is applied in a direction parallel to the surface of a medium. S wave does not propagate
through fluids. Shear wave exploration cannot be done with conventional marine seismic
acquisition technology. Ocean bottom nodes (OBN), in which receivers are mounted on
the seabed (Figure 2.3), can be used to measure S waves.

Remotely operated vehicle (ROV) Node on the seabed

FIGURE 2.3 Trilobit 4C ocean bottom nodes deployed on the seabed using ROV (remotely operated
vehicle) system. Source: WorldOil Online Magazine, vol. 232(9), September 2011.

S-wave velocity can be calculated using the following equation:

rffiffiffi
m
Vs ¼
r

I. BASIC SEDIMENTOLOGY AND SEISMIC DATA ACQUISITION

20 2. UNDERSTANDING SEISMIC WAVE PROPAGATION

P-wave velocity is always faster than S-wave velocity. The ratio of P-wave velocity to
S-wave velocity is greater than or equal to the square root of 2
Vp pffiffiffi
 2
Vs

CHARACTERISTICS OF S WAVES
• The particle motion under the influence of the wave is in the direction of the energy applied
and is then perpendicular to the propagation of the wave.
• S waves can only propagate in solid.
• There is no compression and expansion as a result of the particle motion, and thus no
pressure disturbance.
• One can detect S waves through the measurement of the particle motion.

WAVE PARAMETERS

Figure 2.4 illustrates some important wave parameters that are applicable in
seismic exploration.

Wavelength Peak
l

Amplitude

Time zero

Trough

FIGURE 2.4 Sinusoidal waveform (seismic trace).

A sinusoidal waveform is described by the mathematical expression below:

’ ðt; xÞ ¼ A cos ð2pft þ fÞ
The above equation described a seismic trace.
The parameters in the above equation are defined below:

• The amplitude is the maximum displacement from time zero.

• Period (t) is the time, in seconds, required to complete 1 waveform.

I. BASIC SEDIMENTOLOGY AND SEISMIC DATA ACQUISITION

 SEISMIC REFLECTION 21
• Frequency ( f ) is the number of times a waveform repeats per second. It is the reciprocal of
period. Frequency is measured in Hertz (Hz), where 1 Hz equals 1 waveform/s.
1
f ¼
T
For example, a waveform with duration of 30 ms has a frequency of 33.3 Hz. Note that 1 s
equal to 1000 ms:
• A wavelength (l) is the distance over which one waveform is completed. It is measured in
metre or feet:
velocity ðvÞ
wavelength ðlÞ ¼
frequency ð f Þ
• Wave number is the reciprocal of wave length:
wave number ¼ 1/wavelength.
• Phase indicates the initial position of the waveform. Phase is an angular quantity that is
measured in radians or degrees.
There is a very important relationship between frequency ( f ), wavelength (l) and velocity
(v) for all waves. This is given by
v ¼ fl
where v is seismic velocity, f is frequency of the propagating wave and l is the wavelength.

SEISMIC REFLECTION

Law of Reflection
The law of reflection states that the angle of incidence equals the angle of reflection. It also
states that the incident ray, the reflected ray and the normal at the point of incident all lie in
the same plane (Figure 2.5).

Receiver
Shot point

Incidence angle Reflected angle

n1 r1

n2 r2 Reflection point

FIGURE 2.5 Conceptualized law of reflection.

For a horizontal reflector, the reflection point is half way from source point to receiver. As a
rule of thumb in the industry, the maximum offset (the offset is the distance between the

I. BASIC SEDIMENTOLOGY AND SEISMIC DATA ACQUISITION

22 2. UNDERSTANDING SEISMIC WAVE PROPAGATION

source and the receiver) is equal to the depth of the reflector. For straight rays the angle of
incidence is about 28 . But rays in the real earth are not straight. The angle of incidence
for real earth is approximately 35 (Figure 2.6).

Receiver
Source

Depth

35°
Incidence angle Reflected angle

FIGURE 2.6 Conceptualized curve ray path in the earth.

For a dipping reflector, the reflection point does not lie at the midpoint between the source
and the receiver but up-dip of the source receiver pair (Figure 2.7).

Source Receiver

Reflected ray
Incident ray

Sand
Dipping reflector

Shale

FIGURE 2.7 Reflection point on a dipping reflector.

Note that the incident ray generates not only the reflected ray but also the refracted ray at
an angle given by Snell’s law.
Law of Refraction
The law of refraction is known as Snell’s law. It states that the ratio of the sine of the angle of
incident to the sine of the angle of refraction is a constant for a given pair of strata

I. BASIC SEDIMENTOLOGY AND SEISMIC DATA ACQUISITION

 CRITICAL ANGLE 23
sin yi
¼n
sin yr
where n is a constant, yi is the angle of incidence and yr is the refracted angle.

FIGURE 2.8 Conceptualized Snell’s law. Shot point

Receiver

Incidence angle Reflected angle

n1 r1

n2 r2
Refracted angle
Refracted ray

CRITICAL ANGLE
As the angle of incidence increases in Figure 2.8, a point is finally reached where the
refracted ray does not emerge at the second layer but lie along the interface. This particular
angle of incidence at which the angle of refraction is 90 and the refracted ray lies along the
interface is known as the critical angle. At and beyond the critical angle, there is no transmit-
ted ray and therefore a very high reflected ray will be recorded (Figure 2.9).

FIGURE 2.9 Conceptualized critical angle. Critical angle

qc
n 1 r1

n2 r2 Refracted ray
q 2 = 90°

Therefore,
sin yi Vp1
¼
sin 90 Vp2
But, sin 90 ¼ 1.
At critical angle,
Vp1
sin ycritical ¼
Vp2

I. BASIC SEDIMENTOLOGY AND SEISMIC DATA ACQUISITION

24 2. UNDERSTANDING SEISMIC WAVE PROPAGATION

A critical refracted wave travels along the interface between layers and is refracted back
into the upper layer at the critical angle. The waves refracted back into the upper layer are
called head waves or first-break refractions because at certain distances from a source, they
are the first arriving energy. Recorded first-break refraction is shown in Figure 2.10.
Note that these first-break refractions can give us important information about the shallow
velocities on land seismic data.
Note also that seismic data are acquired in such a way that reflections from horizons of
interest are in the pre-critical region, even at the farthest offset in the data.

FIGURE 2.10 First-break refraction data.

In reality, part of the seismic energy arriving at an interface is transmitted and refracted,
and another part of the energy is reflected at that same interface. Given that there are many
reflectors in the subsurface, there will be many paths from source to receiver, each of them
with a different travel time. The proportion of energy reflected depends on the material
properties of the two bounding layers and on the angle of incidence.

REFLECTION COEFFICIENT

In each layer in Figure 2.11, we have the density, r, and velocity, n. The product of velocity
and density is a material property of the layers and is known as acoustic impedance. The
acoustic impedance is an important property of a rock layer. This is because it determines

I. BASIC SEDIMENTOLOGY AND SEISMIC DATA ACQUISITION

 REFLECTION COEFFICIENT 25
the reflection response of the earth. In the earth, reflections occur at the interfaces between
layers and the reflection amplitudes depend mostly on the difference of acoustic impedance
from layer to layer. The term AI unit is used for the unit of acoustic impedance.

Source Receiver Signal from receiver

Sand

Time
Incidence amplitude
Reflected amplitude

n1 r1
Reflection from
n2 r2 bottom of layer 1

Shale Reflection from

bottom of layer 2

FIGURE 2.11 Reflection coefficient.

Reflected amplitude is the difference between the two impedances of each layer divided by
their sum. The ratio of the reflected amplitude to the incidence amplitude is called the reflec-
tion coefficient and is given by the difference in acoustic impedance divided by their sum

Z2  Z1
reflection coefficient ¼
Z2 þ Z1

where Z1 and Z2 are acoustic impedance of layers 1 and 2, respectively.

Calculations of reflection coefficients should give results between 1 and þ1.
The reflection coefficient can still be expressed as

V2 r2  V1 r1
RC ¼
V2 r2 þ V1 r1

The sum in the denominator of the reflection coefficient equation does not have very much
effect. It is the difference in the numerator that really dominates the reflection coefficient.
For example, if we use the reflection coefficient equation and calculated that the reflection
coefficient is 0.2, then this implies that a wave with amplitude of 20% of the original ampli-
tude of the wave reaching a reflecting interface is returned towards the surface. As the energy
is proportional to the square of the amplitude, 4% of the energy is reflected, and that the
remaining energy passes through the interface (Figure 2.12).

I. BASIC SEDIMENTOLOGY AND SEISMIC DATA ACQUISITION

26 2. UNDERSTANDING SEISMIC WAVE PROPAGATION

RC = 0.2
Reflected amplitude

4%
Incidence amplitude

96%

Refracted amplitude

FIGURE 2.12 Reflected amplitude.

Note that reflection coefficient equation is an approximation. It only holds true for rays at
right angles to the interface.
Note also that ray path models can be used to determine the actual amplitude of the reflec-
tions returned from any layer.

TRANSMISSION COEFFICIENT

Source–receiver pair at normal incidence

Sand

Incidence amplitude

n1 r1 Reflected amplitude

n2 r2

Transmitted amplitude

Shale

FIGURE 2.13 Transmission coefficient.

Transmission coefficient (Figure 2.13) describes the amplitude of the transmitted wave
and is given by the expression
2Z1
transmitted amplitude ¼
Z2 þ Z1
If the reflection coefficient is R, then a little mathematics gives the transmission coefficient
as 1 þ R, so that for a wave going from a softer rock (sand) to harder rock (shale), the trans-
mitted amplitude is greater than the incident amplitude. This is because the transmitted
amplitude is measured in a different material from the incidence and reflected amplitudes.

I. BASIC SEDIMENTOLOGY AND SEISMIC DATA ACQUISITION

 UNDERSTANDING CONVOLUTION 27
For a wave going downwards the transmission coefficient is 1 þ R and for a wave going
upwards the transmission coefficient is 1  R (Figure 2.14).

Subsurface

Sandstone (softer rock)

Layer 1 n1 r1

Shale (harder rock) n2 r2

Layer 2

a
Subsurface

Sandstone (softer rock)

Layer 1 n1 r1

n2 r2
Shale (harder rock)

Layer 2

b
FIGURE 2.14 (a) The arrow shows a down-going wave. Transmission coefficient is 1 þ R. (b) The arrow shows a
up-going wave. Transmission coefficient is 1  R.

Note: The two-way transmission coefficient is expressed as

two-way transmission coefficient ¼ ð1 þ RÞð1  RÞ ¼ 1  R2

UNDERSTANDING CONVOLUTION

From well log data (velocity and density log), we can construct a lithologic variation such
as changes from sand to shale (Figure 2.15).

FIGURE 2.15 Conceptualized stratigraphic layer of Geologic section

the earth.
Depth

Shale
n1 r1
Sand
Shale

Sand

Shale n2 r2
Sand

Product of the velocity and density log gives the acoustic impedance (AI) log or profile as
shown in Figure 2.16.

I. BASIC SEDIMENTOLOGY AND SEISMIC DATA ACQUISITION

28 2. UNDERSTANDING SEISMIC WAVE PROPAGATION

Density log (g/cc) Velocity log (m/s) Impedance log (AI unit)

FIGURE 2.16 Density and velocity logs (derived from sonic log). The product of density and velocity logs gives
the acoustic impedance log.

Recall that, seismic reflection occurs when there is contrast in acoustic impedance across a
layer boundary. From the acoustic impedance log, the geoscientist can calculate the reflection
coefficient for each reflecting interface in the subsurface.
These reflection coefficients formed the reflectivity series which is either displayed as time
or depth.
The earth reflectivity is a set of impulse response with amplitudes proportional to the
reflection coefficient of each reflecting horizon and time of occurrence equal to the two-
way reflection time (Figure 2.17).

Amplitude
R1
R2
Time

R3

R4

R5

R6

FIGURE 2.17 Earth reflectivity series from each reflecting interface.

I. BASIC SEDIMENTOLOGY AND SEISMIC DATA ACQUISITION

 CROSS-CORRELATION 29
In seismic acquisition, the source wavelet that is sent into the earth convolved with the
earth’s reflectivity series to produce the recorded seismic trace. This process used to derive
our seismic trace is called convolution.
Mathematically, convolution is:
wðtÞ  rðtÞ ¼ sðtÞ
In other words, convolution is an operation describing the interaction between a unit
impulse (source wavelet) and a system (earth reflectivity) to produce an output (seismic trace)
(Figure 2.18).
FIGURE 2.18 Conceptualized the convo- Reflectivity series
lutional model of the seismic trace. R1
Wavelet
R2

R3
=
R4

R5

R6

Note that the convolution of any time series (earth reflectivity) with a unit impulse (spike
produced from seismic energy) is simply the input time series (seismic trace).

CROSS-CORRELATION

Cross-correlation is a process for measuring the similarity of one time series (seismic trace)
to another time series (seismic trace). An example of cross-correlation is shown in Figure 2.19.
Seismic traces (time series)

FIGURE 2.19 Conceptualized cross-correlation.

I. BASIC SEDIMENTOLOGY AND SEISMIC DATA ACQUISITION

30 2. UNDERSTANDING SEISMIC WAVE PROPAGATION

AUTOCORRELATION
Autocorrelation is simply the cross-correlation of one time series with itself.

TIME DOMAIN

Convolution described time series and wavelets in terms of amplitude as a function of

time. This is time domain description of seismic trace. That is, in time domain, a seismic trace
is described as having certain amplitudes at certain time.

FREQUENCY DOMAIN

In frequency domain, seismic trace is described in terms of amplitudes and phases at

certain frequencies.
To describe seismic trace in the frequency domain requires both an amplitude spectrum
and a phase spectrum. The amplitude spectrum simply gives amplitude at each frequency.
The phase spectrum simply gives the phase at each frequency (Figure 2.20).

40

30
Amplitude

20

10

0 25 50 75 100
a Frequency (Hz)

180

90
Phase (°)

0
25 50 75 100

-90

-180
b
FIGURE 2.20 (a) Amplitude spectra. (b) Phase spectra.

Convolution of our seismic trace in the time domain is equivalent to multiplication of the
trace amplitude spectrum with the filter amplitude spectrum and the addition of the trace
phase spectrum with the filter phase spectrum in the frequency domain (Figure 2.21).

I. BASIC SEDIMENTOLOGY AND SEISMIC DATA ACQUISITION

 INVERSE FOURIER TRANSFORM 31
Input trace Trace amplitude spectrum Trace phase spectrum

X +

Filter amplitude spectrum Filter phase spectrum

= =
Output trace
Amplitude spectrum Phase spectrum

FIGURE 2.21 Conceptualized convolution in the frequency domain.

Correlation in time domain is equivalent to multiplication of the trace amplitude spectrum

with the filter amplitude spectrum and subtraction of the trace phase spectrum from the filter
phase spectrum in the frequency domain.
Notice that in convolution the phase spectrums are added together, while in correlation the
phase spectrums are subtracted one from another. This difference explains why the autocor-
relation of any trace is a zero phase.
Note also that autocorrelation of a seismic trace in the time domain is equivalent to mul-
tiplying the amplitude spectrum by itself and subtracting the phase spectrum from itself.

FOURIER TRANSFORM

The Fourier transform process allows the geoscientists to transform the seismic trace from
the time domain, which is a plot of amplitude versus time, to frequency domain, which is the
amplitude and phase spectra.

INVERSE FOURIER TRANSFORM

The geoscientists use inverse Fourier transform to transform the seismic trace from fre-
quency domain to time domain.

I. BASIC SEDIMENTOLOGY AND SEISMIC DATA ACQUISITION

32 2. UNDERSTANDING SEISMIC WAVE PROPAGATION

The seismic record consists of about 240 traces recorded every 25 m on the surface and
thousands of traces are in the total data volume. To transform this quantity of data into
the frequency domain, an algorithm has been developed that speeds up the process to allow
transformation from time domain to frequency domain and vice versa.
The algorithm to transform from time domain to the frequency domain is called the fast
Fourier transform (FFT).
The algorithm to transform from frequency domain to the time domain is the inverse fast
Fourier transform (IFFT).

I. BASIC SEDIMENTOLOGY AND SEISMIC DATA ACQUISITION