Environmental and Exploration Geophysics II

Frequency and Bandwidth: their

relationship to Seismic Resolution
tom.h.wilson
tom.wilson@mail.wvu.edu

Department of Geology and Geography

West Virginia University
Morgantown, WV

Tom Wilson, Department of Geology and Geography

How can we build this wavelet?

Tom Wilson, Department of Geology and Geography

1
The range of frequencies
present in the wavelet
controls its ability to resolve
the top and bottom of a layer
of given thickness.
The wavelet or transient
mechanical disturbance
generated by the source can
be thought of as a Composite Wavelet
(sum of cosines)

superposition or summation 1.2

of sinusoids with varying

1
0.8

Amplitude
0.6
0.4

frequency and amplitude.

0.2
0
-0.2
-0.4
-0.06 -0.04 -0.02 0 0.02 0.04 0.06
Time (seconds)

Tom Wilson, Department of Geology and Geography

Hilterman, 1985

The examples below illustrate the effect of increasing

the frequency range or bandwidth of the wavelet.

O. Ilmaz, 1987
Tom Wilson, Department of Geology and Geography

2
 The following simple example helps illustrate
the concept of an amplitude spectrum. Below is
a signal consisting of two sinusoids.

Tom Wilson, Department of Geology and Geography

Each sinusoid is associated with a specific

frequency. There are two frequency
components. The 32 sample per cycle component
has a frequency of 4 and the 8 samples per
cycle component has a frequency of 16. The
amplitude of the 32 sample/cycle component is
twice that of the 8 sample/cycle component.

The frequency spectrum (above) of the

“signal” at the top of the previous slide is an
equivalent representation of the signal.
Tom Wilson, Department of Geology and Geography

3
 Frequency domain

Time domain

O. Ilmaz, 1987
Tom Wilson, Department of Geology and Geography

Amplitude spectrum

Phase spectrum

Individual
frequency
components

Time-domain wavelets
Zero Phase Minimum Phase
Tom Wilson, Department of Geology and Geography Hilterman, 1985

4
 Extracting information about
wavelet frequency content from
an isolated reflection event.

The dominant period (τc) of

the response corresponds to
the time from one peak to
the next or from one trough
to the next. The reciprocal
of this dominant period is a
measure of the dominant
frequency (fc) of the signal
or wavelet spectrum.
The reciprocal of the half-width of
the response-envelop (τb) provides
an estimate of the bandwidth (fb)
of the signal spectrum.

Tom Wilson, Department of Geology and Geography

Hilterman, 1985

The dominant
frequency and
bandwidth measured
from the time-domain
representation of the
signal wavelet can be
used to provide a
sketch of the wavelet
spectrum.
Just as importantly
these measures can
be related directly to
the resolution
properties of the
seismic wavelet.

Hilterman, 1985
Tom Wilson, Department of Geology and Geography

5
 Let’s come back to this issue in a minute, but first let’s pull some
ideas together to develop a basic understanding of how the seismic
signal arises in terms of reflection coefficients and wavelets.

Shape of up-
going wave is
reversed

In space the leading

positive cycle changes
to a leading negative
cycle

Tom Wilson, Department of Geology and Geography

Exxon in-house
course notes

Shape of up-
going wave is
unchanged

Exxon in-house
course notes

Tom Wilson, Department of Geology and Geography

6
 positive

negative

Tom Wilson, Department of Geology and Geography

Exxon in-house
course notes

As the two layers move closer and closer together we get to a point
where the second cycle in the wavelet reflected from the top of the
layer overlaps with the arrival of the lead cycle in the wavelet
reflected from the base of the layer. This occurs at two-way time
equal to 1/2 the dominant period of the wavelet.

Tom Wilson, Department of Geology and Geography

Exxon in-house
course notes

7
 At this point there is maximum constructive
interference between the reflections from the top and
bottom of the layer. The composite reflection event (at
right above) attains peak amplitude.

Tom Wilson, Department of Geology and Geography Exxon in-house

course notes

The peak period of the wavelet can be determined using

peak-to-trough times which can be thought of as
corresponding to one half the dominant period of the wavelet.
Multiply those times by two to get the dominant period.

Tom Wilson, Department of Geology and Geography

8
 Reflection
Coefficients
trough

Side
peak
lobe

Maximum constructive interference illustrated for the

zero phase wavelet. The peak-to-trough time equals
τc/2, which also equals delay time between consecutive
reflection events
Tom Wilson, Department of Geology and Geography

Once the separation in

time drops to less than
half the dominant period
of the wavelet
destructive interference
in the reflections from
the top and bottom of
the layer will occur.
However, as the layer
continues to thin, the
dominant period of the
composite reflection
event does not drop
below 1/τc. However,
the amplitude of the
composite continues to
drop. But not the period.
Tom Wilson, Department of Geology and Geography
Exxon in-house
course notes

9
 Maximum
Constructive
Interference

Seismic Wavelet

trough

Side
peak
lobe

Two-way interval The peak-to-

time separating trough time
reflection
equals τc/2.
coefficients is τc/2

Tom Wilson, Department of Geology and Geography

These amplitude relationships are summarized below in the

model seismic response of a thinning layer similar to that
which you will generate in lab today.

Tom Wilson, Department of Geology and Geography

10
 The amplitude difference
- trough-to-peak remains
constant for two-way
travel times much
greater than half the
dominant period. Thickness =Vt/2

As the top and bottom of

the layers merge closer
and closer together, the
lead cycle in the
reflection from the base
of the layer overlaps with
the follow-cycle in the
reflection from the top
and the amplitude of the
composite reflection
event begins to increase.
Tom Wilson, Department of Geology and Geography

Layer thickness is simply Vt/2, where t is the two-way

interval transit time. Tuning occurs at two-way times equal
to one-half the dominant period (τc/2). If the interval
velocity of the layer in question is known, the dominant
period can be converted into the tuning thickness.

Tom Wilson, Department of Geology and Geography

11
Difference of arrival time
between the reflections from
the top and bottom of the
layer decreases abruptly at
about 8 milliseconds.
8 milliseconds represents the
two-way travel time through
the layer; it is also the time
at which tuning occurs and is
half the dominant period of
the seismic wavelet.
8 milliseconds is τc/2 and the
two way time through the
layer. Thus, τc/4 is the one-
way time through the layer.

Tom Wilson, Department of Geology and Geography

τc/4, the one-way time through the layer, equals 4

milliseconds. The interval velocity in the layer is 11,300 f/s.
Hence, the thickness of the layer at this point is ~45 feet.
This is the tuning thickness or minimum resolvable thickness
of the layer obtainable with the given seismic wavelet.
Tom Wilson, Department of Geology and Geography

12
 What is the amplitude spectrum of wavelet #5?

Broader spectra produce sharper, Ilmaz, 1987

shorter duration wavelets

Tom Wilson, Department of Geology and Geography

Spectral bandwidth, wavelet duration in the time domain

and resolution. τC is only one parameter that affects
resolution. τb is also an important parameter.

Greatest Bandwidth

Smallest Bandwidth

Tom Wilson, Department of Geology and Geography

Hilterman, 1985

13
 The Convolutional Model

Hilterman, 1985

Physical nature of the seismic response

Tom Wilson, Department of Geology and Geography

The output is a superposition of

reflections from all acoustic interfaces
Tom Wilson, Department of Geology and Geography
Exxon in-house
course notes

14
Tom Wilson, Department of Geology and Geography
Exxon in-house
course notes

Tom Wilson, Department of Geology and Geography

15
 One additional topic to consider in general is wavelet deconvolution
and how wavelet shape can affect geologic interpretations ….
Consider the following structural model
Subsurface structure - North Sea

Tom Wilson, Department of Geology and Geography

Neidel, 1991

Below is the synthetic seismic response computed for the

North Sea model.

Potential hydrocarbon trap?

Tom Wilson, Department of Geology and Geography

Neidel, 1991

16
 Consider the effect of wavelet shape on the geologic interpretation
of seismic response. In the case shown below, the primary reflection
from the base of the Jurassic shale crosses a side-lobe in the
wavelet reflected from the overlying basal Cretaceous interval.

Tom Wilson, Department of Geology and Geography

Neidel, 1991

Deconvolution is a filter operation which compresses and

simplifies the shape of the seismic wavelet. Deconvolution
improves seismic resolution and simplifies interpretation.

Tom Wilson, Department of Geology and Geography

17
 Neidel, 1991

North Sea Seismic display after deconvolution.

The geometrical interrelationships between
reflectors are clearly portrayed.
Tom Wilson, Department of Geology and Geography

Consider the following problem -

You are given the seismic wavelet shown below.

Using the estimation procedure discussed in class today measure the

appropriate feature on the above seismic wavelet and answer the following
questions:

What is the minimum resolvable thickness of a layer having an interval

velocity of 10,000fps? Show work on your handout

What is the phase of the wavelet? Why do you say that?

Tom Wilson, Department of Geology and Geography

18
 The minimum phase wavelet has its energy concentrated toward the front end of
the wavelet. The amplitude of the disturbance decays exponentially. This wavelet
is a causal wavelet and the location of the reflection coefficient is placed at
the wavelet onset, which can be difficult for the interpreter to pick.

Tom Wilson, Department of Geology and Geography

The zero phase wavelet is symmetrical. This wavelet is centered over the reflection
coefficient. The zero phase wavelet is produced through data processing and is not
generated naturally. It is non causal - half of the wavelet arrives before the
reflector appears in time. It is easy for an interpreter to pick reflection times using
the zero phase wavelet since highest amplitude occurs at the reflection boundary.

Tom Wilson, Department of Geology and Geography

19
 The zero-phase wavelet
is also considered to
have higher resolving
power. It is generally
more compact than the
equivalent minimum
phase wavelet and is,
overall, easier to
interpret.
The exploration data is
in a zero phase format.

Tom Wilson, Department of Geology and Geography

Hilterman, 1985

The default wavelet in Struct is the Ricker

wavelet. The Ricker wavelet is zero phase.

Tom Wilson, Department of Geology and Geography

Hilterman, 1985

20
 If you haven’t already, finish reading chapter 4.

Tom Wilson, Department of Geology and Geography

21