Numerical Modeling of the Seismic Wave Field

Source
Numerical Modeling of the Seismic Wave Field

Source
Numerical Modeling of the Seismic Wave Field

Source
 To obtain a representative image of the subsurface.
 Improve the signal to noise ratio: e.g. by measurement of several channels
and stacking of the data (white noise is suppressed).
 Present the reflections on the record sections with the greatest possible
resolution and clarity and the proper geometrical relationship to each other
by adapting the waveform of the signals.
 Isolate the wanted signals (isolate reflections from multiples and surface
waves).
 Obtain information about the subsurface (velocities, reflectivity etc.).
 Obtain a realistic image by geometrical correction.
 Conversion from travel time into depth and correction from dips and
diffractions.
Seismic Imaging

Data acquisititie

Data processing

Interpretation
The overall seismic data workflow
1. Natural
 Earthquakes
 Volcanic Explosions
 Magma Flow
2. Artificial
 Sledge hammer, weight drop, shotgun (shallow work)
 Buffalo Gun, dynamite (oil exploration)
 Dinoseis/Vibroseis (oil exploration, crustal studies)
 Airgun array at sea (oil exploration, crustal studies)
Dinosaur
Dinoseis
Dinoseis

In a gas gun, a mixture of

propane and oxygen is
contained in a truck-mounted
bell, with a movable
diaphragm in contact with the
ground
Vibroseis
 The vibroseis method pounds or
vibrates the earth to create shock
waves which are transmitted to a
recording truck.
 Four or five trucks usually work
close together in a line.
 A 2 component mass that has
the unique ability to generate both
Pressure wave output, inline shear
wave output, and crossline shear
wave output.
Vibroseis

Vibrating plate mounted beneath a truck

generates a long oscillatory wavetrain.
Plate placed on the ground, and the
truck is raised so most of its weight is on
the plate.
Electronic control system generates a
low amplitude sinusoidal signal that
varies in frequency, from 12 to 80 Hz for
example, over several seconds.
Signal drives hydraulics of the vibrator
causing the plate to vibrate the ground
with the same signal.
Vibroseis

Vibroseis is less damaging to

ground and nearby structures,
because signal is relatively low
amplitude. It has been used in major
cities.
A source, such as a vibrator
unit, dynamite shot, or an air
gun, generates acoustic or
elastic vibrations that travel
into the Earth, pass through
strata with different seismic
responses and filtering
effects, and return to the
surface to be recorded as
seismic data.
 The generation and recording of seismic data. Acquisition
involves many different receiver configurations, including laying
geophones or seismometers on the surface of the Earth or seafloor,
towing hydrophones behind a marine seismic vessel, suspending
hydrophones vertically in the sea or placing geophones in a wellbore
(as in a vertical seismic profile) to record the seismic signal.
 • Changes in rock type or fluid
content reflect the sound
waves towards the surface
• Receivers (hydrophones/
geophones) record the
returning seismic signal.
• Sound waves reflected by
different boundaries arrive at
different times.

Seismic horizons represent

changes in density and allow
the subsurface geology to
be interpreted
 Marine seismic data can be
acquired using single streamers
(second vessel) or multiple
streamers (first vessel) of
various lengths or ocean bottom
cable.
3D seismic data of various
footprints, or areas covered by
an array of multiple towed
streamers, are also routinely
acquired (four vessels on the
right).
This marine seismic vessel is towing multiple streamers that contain
hydrophones. Air guns emit a signal like a sound wave that travels
through the water into the Earth, passes through strata with different
seismic responses and filtering effects, and returns to the
hydrophones to be recorded as seismic data.
Coil shooting
Marine seismic vessels
are typically about 75 m
[246 ft] long and travel
about 5 knots [9.3 km/hr
or 5.75 statute miles/hr]
while towing arrays of air
guns and streamers
containing hydrophones a
few meters below the
surface of the water.
The tail buoy helps the crew locate the end of the streamers.
The air guns are activated periodically, such as every 25 m (about
10 seconds), and the resulting sound wave travels into the Earth, is
reflected back by the underlying rock layers to hydrophones on the
streamer and then relayed to the recording vessel.
Marine seismic acquisition
can be performed by a
source vessel and a
recording vessel with
streamers or an ocean-
bottom cable or, as shown
here, seabed geophones.
Energy from the source
vessel in the form of P-
waves travels through the
Earth and is reflected and
converted into P-waves and
S-waves recorded by the
receiver groups and relayed
to the recording vessel.
 Optimal acquisition varies according to local
conditions and involves employing the appropriate
source (both type and intensity), optimal
configuration of receivers, and orientation of
receiver lines with respect to geological features.
This ensures that the highest signal-to-noise ratio
can be recorded, resolution is appropriate, and
extraneous effects such as air waves, ground roll,
multiples and diffractions can be minimized or
distinguished, and removed through processing.
Schematic block diagram of a multichannel seismic recording system.
Channel refers to electrical input to recording system, i.e., carries data
from a receiver, such as from a group of geophones, to a recorder.
Might be from a single geophone as in engineering work, or a group of
9 geophones, common in oil exploration.
• In oil exploration work,
recording systems can
record up to 8000 channels.
• Simultaneous recording of
500 to 2000 channels is
common during 3D seismic
acquisition, and 120 to 240
channels during onshore 2D
seismic acquisition.
A common seismic display
that shows trace amplitude
versus time as an
oscillating line about a null
point.
Graphs of intensity of sound as received by the recorders
The seismic trace we record is
the result of the earth's
reflectivity (a composite of a
number of wavelets generated at
many subsurface interfaces in
the ground) convolved with the
source wavelet (and it's ghosts),
multiples, the recording system
and some noise.
 Velocity function
 Fold
 Point of intersection with other
profiles
 Location of borehole loggings
 Topography
 Shotnumber (additive to CMP), etc.
 What kind of measurement was carried out?
On land/sea using dynamite/vibrator?

 What kind of spread was used?

Single ended/ split spread
Your real voyage in

consists not only in seeking

new knowledge
but also in having
 Alteration of seismic data to suppress noise, enhance signal
and migrate seismic events to the appropriate location in space.

 To improve the signal to noise ratio

 Isolation of the wanted signals (Reflections isolated from
multiples and surface waves)
 To obtain a higher resolution by adapting the waveform of the
signals
 To obtain a realistic image by geometrical correction
 To obtain information about the subsurface
Severe Noise Level Medium Noise Level Mild Noise Level
No Noise
Poor S/N Ratio Medium S/N Ratio Mild S/N Ratio

High quality spatial sampling and / or high quality target illumination

Mid quality spatial sampling and / or mid quality target illumination

Poor quality spatial sampling and / or Poor quality target illumination

Processing steps typically include:
1- analysis of velocities and frequencies,
2- static corrections,
3- deconvolution,
4- normal moveout,
5- dip moveout,
6- stacking, and
7- migration (which can be performed before or after stacking).
Source Receivers Raw data
Q t t t t t t

Subsurface geology Seismic section

 energy source signal There are many reflectors
receiver
on a seismic section. Major
changes in properties usually
Bed 1
produce strong, continuous
lower velocity reflectors as shown by the
higher velocity
arrow.
Bed 2

A seismic reflector is a boundary

between beds with different
properties. There may be a change
of lithology or fluid fill from Bed 1
to Bed 2. These property changes
cause some sound waves to be
reflected towards the surface.
CDPs are typically marked at intervals along the top of
seismic lines and they are regularly spaced to form a
horizontal scale. Here, 80 CDPs represent about 1 kilometre
(km).

53
Gaps in land seismic
data are due to
omissions where data
could not be acquired

For example, it is not

always possible to
transmit the signal
above pipes, in
sensitive areas and
above buildings

Signals from
farther away will
provide information
for deeper horizons
Two way time (TWT) is
recorded on the vertical axis of
the seismic line in fractions of 0.0 seconds or
a second. Sometimes it is more
convenient to express time as sea level
milliseconds.

TWT is the time required for 0.5 seconds or

the seismic wave to travel from
500 milliseconds
the source to some point below
the surface and back up to the
receiver.

1.0 seconds or
1000 milliseconds
The offset between the source and other
receivers induces a delay in the arrival
time of a reflection from a horizontal
surface at depth.

With increasing
seismic velocity
(increasing depth) the
hyperbolas become
flatter.
Ignorants are often greatly
tempted to read geology more
or less directly from the seismic
data.
 Data Processing Flow Chart
Field Tape Scaling
Mute
Transcription to Stack
Static correction
Interval Format Migration
NMO
Geometry Time variant filter
Remove bad trace Sort to
CDP Domain Time variant scale
Geometric spreading
compensation Multiple Attenuation
DISPLAY
Exponential gain Deconvolution

F-K Velocity Filter

Basic scheme of the seismic data processing
Measurements
Loading of the data/ conversion Navigation
Demultiplexing
Pre-Processing
Geometry
Editing/ Muting/ Stacking
Amplitude correction
Frequency filter
Deconvolution
CMP-Sorting Prestack
Velocity analysis

NMO-Correction
Stacking
Migration
Poststack
Post-processing

Printing/storage
Printing/storage
Raw data Processed
(migrated) data
Almost every seismic  UKOOA P1/90 navigation
file
processing system will
 SEG-Y seismic trace format
have it's own format for
 SHELL-V5 velocity format
seismic trace data,
 ESSO-V2 (formerly used
velocity files, horizon but now not favoured)
files etc.
SEG-D:
multiplexed time sequential form
SEG-2:
SEG-Y: demultiplexed trace sequential form

Digital storage:
• IBM Real (4-Byte floating point, standard)
• IEEE (4-Byte floating point)
• 4-Byte Integer
Multiplexer
 Older systems or systems with many separate channels
do not have for each separate channel a separate AD
converter or enough writing capacity to save all data from
one shot.
 To solve this problem all the values at the separate
channels are sampled for each time sample, after which
all values for the next time sample are sampled and
recorded. The data are not ordered for each channel
(channel 1, channel 2, channel 3, etc.), but for each time
sample (Timesample 1 - all channels, Timesample 2 - all
channels, etc.).
 For the processing all channels must be sorted out which
is called: Demultiplexen.
Demultiplexer
SEG-Y Importing:
1. Select SEG-Y file or files to import,

2. Define header locations,

3. Extract a geometry database from the SEG-Y headers,

4. Preview the geometry (source and detector attributes),

5. Construct a database.
1. Determine Source and
receiver position for
measured data
2. Calculate CMP position
3. Specify a certain BIN in
which the measured trace
belongs
Trace: The seismic data recorded for one channel.
A trace is a recording of the Earth's response to seismic
energy passing from the source, through subsurface
layers, and back to the receiver.
A seismic trace represents the response of the elastic
wavefield to velocity and density contrasts across
interfaces of layers of rock or sediments as energy travels
from a source through the subsurface to a receiver or
receiver array.
A geometrical arrangement of
seismic sources: a source array, with each individual
source being activated in some fixed sequence in time
OR
receivers: a geophone or hydrophone array that is
recorded by one channel.

Nests can contain numerous closely spaced geophones.

Schematic diagram of air gun geometry

An array is a geometrical
arrangement of seismic
sources. This schematic
diagram shows an air-
gun array towed several
hundred meters behind a
seismic vessel to provide
an energy source for the
acquisition of marine
seismic data.
A wave moves outward from
source in expanding spherical
wave fronts.
By conservation of energy,
the energy in a unit area of the
growing wave front decreases
as r-2, where r is the radius of
the sphere or distance from
the source.
 After wavefront has travelled distance r,
energy is distributed over spherical
shell of area 4πr2.
 So energy per unit area of spherical
wavefront is E/4πr2.
 Amplitude is proportional to square root
of energy, so amplitude of seismic wave
decays as 1/r where r is distance from
source.
 Wavefront only spherical if subsurface
seismic velocity is constant, i.e. ray is
straight.
 In general, spreading of seismic
wavefront spreading is not spherical
and term geometric spreading should
be used, but "spherical spreading" also
found.
 The change in the
amplitude of an electrical
signal from the original
input to the amplified
output.
• Automatic Gain Control (AGC): is the commonest (and often most dangerous) scaling type used. A
sliding window of fixed length is used to compute the average amplitude within the window. This
average is compared to a reference level and the gain computed for a point in the window. The window
then slides down one sample and the next gain correction is computed. The process continues until the
whole trace has been gained.
• ROBUST AGC: is sometimes used in which the minimum and maximum samples within a window are
excluded before computing the average. In this method amplitude extremes are preserved.
• REVERSIBLE AGC: This type of AGC is sometimes applied before multi-channel noise suppression
such as multiple removal. The principal is to equalise the data but store the AGC scalars. The multi-
channel process is then applied and the AGC scalars are then removed. Under certain circumstances this
can be effective, but must be treated with caution when AVO analysis is required.
• EQUALIZATION: usually refers to a whole trace AGC in which the average amplitude of each trace is
set to a fixed value.
• POWER: time to a power usually between 1.8 and 2.4.
• PROGRAMMED GAIN CONTROL: usually the amplitude decay curve (amplitude versus time) of a
trace or group of traces is smoothed and the resulting function gained such that the decay is minimal.
Larger gain values are generally applied at later times to compensate for energy decay.
• EXPONENTIAL: specified in dB per second.
 The exponential function is one of the
most important functions in mathematics. It
is written as exp(x) or ex, where e equals
approximately 2.71828183 and is the base of
the natural logarithm.
In mathematics, exponential growth (or
geometric growth) occurs when the growth
rate of a function is always proportional to
the function's current size. Such growth is
said to follow an exponential law. This
implies that for any exponentially growing
quantity, the larger the quantity gets, the
faster it grows.
But it also implies that the relationship
between the size of the dependent variable
and its rate of growth is governed by a strict
law, of the simplest kind: direct proportion.
 To remove the contribution of selected seismic traces
in a stack to minimize air waves, ground roll and other
early-arriving noise.
Low-frequency traces and long-offset traces are
typical targets for muting.
zero-offset data
Seismic data whose
source and receiver
share a common
location.

Stacking seismic data acquired with

separated sources and receivers gives the
data the appearance of zero-offset data.
zero-offset reflection

the raypath for a zero-offset reflection from a dipping reflector

and the resultant apparent dip.
 Often called statics, a bulk shift of a seismic trace in time
during seismic processing.
A common static correction is the weathering correction,
which compensates for a layer of low seismic velocity material
near the surface of the Earth.
Other corrections compensate for differences in topography
and differences in the elevations of sources and receivers.
1. Subdivision of time shift for source and
Receiver
2. All traces with equal source are corrected for
the time shift of the specific source
3. All traces with equal receiver are corrected for
the time shift of the specific receiver
4. The statics correction is the sum of the
corrections for appropriate source and receiver
 The difference in the arrival times or traveltimes of a
reflected wave measured by receivers at two different offset
locations.
 Normal moveout (NMO) is moveout caused by the
separation between a source and a receiver in the case of
a flat reflector.
 Dip moveout (DMO) occurs as an effect in addition to NMO
when reflectors dip.
 Problems that require static corrections can also produce
moveout.
 The procedure in seismic processing that compensates for
the effects of the separation between seismic sources and
receivers.
 The procedure in
seismic processing
that compensates for
the effects of the
separation between
seismic sources and
receivers in the case
of a horizontal
reflector.
 The effect of the separation between receiver and
source on the arrival time of a reflection that does not
dip, abbreviated NMO.
 A reflection typically arrives first at the receiver
nearest the source.
 The offset between the source and other receivers
induces a delay in the arrival time of a reflection
from a horizontal surface at depth.
 A plot of arrival times versus offset has a hyperbolic
shape.
 With increasing seismic velocity (increasing depth)
the hyperbolas become flatter.
 The traces from different
source-receiver pairs that share
a midpoint, such as receiver 6
(R6), can be corrected during
seismic processing to remove
the effects of different source-
receiver offsets, called normal
moveout or NMO.
After NMO corrections, the
traces can be stacked to
improve the signal-to-noise
ratio.
 The difference in the arrival times or traveltimes of a
reflected wave, measured by receivers at two different
offset locations, that is produced when reflectors dip.
Seismic processing compensates for DMO.
The procedure in seismic processing that compensates
for the effects of a dipping reflector. DMO processing
was developed in the early 1980s.
 In multichannel seismic acquisition, the
point on the surface halfway between the
source and receiver that is shared by
numerous source-receiver pairs. Such
redundancy among source-receiver pairs
enhances the quality of seismic data when
the data are stacked.

The common midpoint is vertically above the

common depth point, or common reflection point.
Common midpoint is not the same as common
depth point, but the terms are often incorrectly used
as synonyms.
 The traces from different
source-receiver pairs that share
a midpoint, such as receiver 6
(R6), can be corrected during
seismic processing to remove
the effects of different source-
receiver offsets, called normal
moveout or NMO.
After NMO corrections, the
traces can be stacked to
improve the signal-to-noise
ratio.
 The common midpoint is the
halfway point between source
and receiver at the surface.
The set of traces recorded
from different source-receiver
pairs that have the same
common midpoint (CMP) is
called a CMP gather.
Moveout corrections and
stacking result in redundancy
of the data that improves the
signal-to-noise ratio.
 In multichannel seismic acquisition where beds do not
dip, the common reflecting point at depth on a reflector,
or the halfway point when a wave travels from a source
to a reflector to a receiver. In the case of flat layers, the
common depth point is vertically below the common
midpoint.
In the case of dipping beds, there is no common depth
point shared by multiple sources and receivers, so dip
moveout processing is necessary to reduce smearing, or
inappropriate mixing, of the data.
The common depth point is the
halfway point in the travel of a wave
from a source to a flat-lying reflector to
a receiver.
 In multichannel seismic acquisition, the common
midpoint on a reflector, or the halfway point when a
wave travels from a source to a reflector to a receiver
that is shared by numerous locations if the reflector is
flat-lying.
Like common depth point, this term is commonly
misused, because in the case of dipping layers, common
reflection points do not exist.
 The loss of energy or amplitude of waves as
they pass through media.
Seismic waves lose energy through absorption,
reflection and refraction at interfaces, mode
conversion and spherical divergence, or spreading
of the wave.
In case of Formation Evaluation

The reduction in amplitude of an electromagnetic

wave passing through the formation, usually measured in
decibels/meter, dB/m.
The term is used in particular with reference to the
propagation resistivity log and the electromagnetic
propagation log.
 A step in seismic signal processing to recover high
frequencies, attenuate multiples, equalize amplitudes,
produce a zero-phase wavelet or for other purposes that
generally affect the waveshape.
Deconvolution, or inverse filtering, can improve
seismic data that were adversely affected by filtering, or
convolution that occurs naturally as seismic energy is
filtered by the Earth.
Deconvolution can also be performed on other types
of data, such as gravity, magnetic or well log data.
zero-phase  Pertaining to seismic data
whose wavelet is
symmetrical about zero
time.
 Deconvolution during
seismic processing can
convert data of mixed phase
to zero-phase data, but is
not always successful.
 Zero-phase data tend to
provide sharper definition
and less distortion between
stratigraphic features in the
subsurface, such as sand
and shale layers.
Phase types
Phase types
(i) Minimum phase: The minimum-phase signal, shown in (a), is described as a front-loaded
signal. This means that the energy in the signal is concentrated in the front of the pulse.
The signal is not symmetrical. The phase of this signal will vary for each frequency
component of the signal. Note that the convolution between two minimum-phase signals is
always minimum phase. However, convolution between one minimum-phase and one
nonminimum-phase signal does not produce a minimum phase signal.
For a group of signals with the same amplitude spectra, the minimum-phase signal will
have the smallest phase shift at all frequencies, cause the least time delay, have the most
front-loaded energy distribution, and have the largest time-zero sample value.
Phase types
(ii) Mixed phase: The mixed-phase signal, shown in (b), is described as a signal with its energy
concentrated in the center of the pulse. It can be divided into minimum-phase and
maximum-phase signals. The signal is usually not symmetrical. The phase of this signal will
vary for each frequency component of the signal.

(iii) Maximum phase: The maximum-phase signal, shown in Figure (c), is described as an end-
loaded signal. This means that the energy in the signal is concentrated toward the end of
the pulse. The signal is not symmetrical. The phase of this signal will vary for each
frequency component of the signal. The characteristics of the maximum-phase signal are
the opposite of the minimum-phase signal.
Phase types
(iv) Zero phase: The zero-phase signal, shown in (d), is symmetrical and centered on zero time.
The zero-phase signal has the shortest duration and largest peak amplitude of any signal
with the same amplitude spectrum. These characteristics make it the most desirable of all
the signals because of its resolution capability. The phase of the zero-phase signal is zero
for all frequency components contained within the signal.
 The use of frequency (abbreviated as f) and
wavenumber (k, the reciprocal of
wavelength) as the reference framework,
obtained by using the Fourier transform over
time and space.
 Frequency is the measurement of the number of times that a
repeated event occurs per unit of time. It is also defined as the rate
of change of phase of a sinusoidal waveform.
To calculate the frequency of an event, the number of
occurrences of the event within a fixed time interval are counted,
and then divided by the length of the time interval.
An alternative method to calculate frequency is to measure the
time between two consecutive occurrences of the event (the period)
and then compute the frequency f as the reciprocal of this time:

where T is the period.

 Frequency has an inverse relationship to the
concept of wavelength.
The frequency f is equal to the speed v of the
wave divided by the wavelength λ (lambda) of
the wave:
 The distance between
analogous points in a
wave train, measured
perpendicular to the
wavefront.
In seismic data, the
wavelength is the seismic
velocity divided by
frequency.
 The angular wavenumber or circular wavenumber, k, often
misleadingly abbreviated as "wavenumber", is defined as

where λ is the wavelength in the medium, ν (Greek letter nu) is

the frequency, vp is the phase velocity of wave, ω is the angular
frequency, E is the energy, ħ is the reduced Planck constant, and c is
the speed of light in vacuum.
The wavenumber is the scalar of the wave vector.