Seismic Data Processing

Recording Media
Reflection Seismic
SEISMIC DATA PROCESSIING
A set of logical operations on the
input data aimed at reducing the
unwanted components and
gathering the wanted
components.
SEISMIC DATA PROCESSIING
1.Demultiplexing / Format Conversion
2.Trace header generation <…observers’ data
3.Spherical divergence correction
4.Deconvolution before stack
5.Band pass filter
6.Trace normalization
7.Velocity Analysis
8. Normal Move Out Correction
9. CMP Stack
SEISMIC DATA PROCESSIING
10. Residual statics estimation & application
11. Dip Move Out Correction
12. Velocity analysis
13. DMO stack
14. Random noise attenuation
15. Decon after stack
16. Time Variant Filter
17. Migration
18. Scaling
 Demultiplexing

 Required if the Seismic Data is recorded

in
 multiplexed format
 Conversion of scan sequential mode to
trace sequential mode.
 Essentially a Matrix transposition (rows to
columns and vice versa)
 Multiplexing
of data

Multiplexed Data are written in Sample order

Demultiplexing of
data
Demultiplexing of data
 Trace Headers Generation

Generation of addresses to the traces

 Geographical positioning
 Facilitates for the unique identification
 Sorting with respect to a common group
(common shot, common receiver,
common midpoint & common offset)
Trace Headers Generation
 Different Types of Gathers

Common Receiver gather Common Mid Point gather

 Spherical Divergence Correction
 Seismic Amplitude decays as a function
of time due to spherical spreading and
inelastic attenuation.
 Compensation is done using a gain
function that is inverse of the decay curve.
 Objective is to see that nearly same
amount of energy is reaching at every
layer of the subsurface.
Amplitude Decay
 Normal Moveout (NMO)

• NMO correction is ‘Dynamic’; it is a function of time

(To), source to receiver offset, and Velocity.
 Pre-Stack Analysis
 In the case of marine data, pre-stack analysis includes selection
of velocity analysis locations, identifying records and traces that
require editing, and determining deconvolution parameters.
 In case of land data processing, especially for noisy data,
prestack analysis is also done to analyze frequency content of
the data and to choose parameters for processes that enhance
signal-to noise ratio.
 Band-pass filter, velocity filter, and deconvolution tests are run
for this purpose. Band-pass filtering discriminates between
signal and noise on the basis of frequency.
 Velocity filters discriminate between signal and noise on the
basis of apparent velocity.
 Pre-Stack Analysis
 Decon can attenuate undesirable events such as short period multiples and
enhance the vertical resolution, by collapsing wavelets.
 Marine data normally have less random and coherent noise than land data,
but they suffer from a higher degree of multiple reflections in many offshore
areas.
 A common midpoint sort is generated for conducting many processing steps,
such as applying elevation statics, velocity analysis, residual statics, and
stack to name few.
 Figure shows the flow chart of pre-stack analysis.
 Distances and angle changes on the seismic line must be taken in
consideration to obtain the correct distance from the source to receiver for
each trace.
 Field statics (that account for trace to trace elevations differences) are
usually applied before NMO to derive velocity analysis from seismic data.
Pre-Stack Analysis
Pre-Stack Analysis
 Static Corrections
• The elevation differences among the traces of a
CMP gather cause delays.
• The Low Velocity Layer (LVL) near the surface
also introduces delays in the observed travel times.
• The data has to be corrected to a reference surface
(Datum) removing these differences.
• These corrections are static; they don’t change
with time; hence the name ‘Static correction’.
 Static Corrections
In order to obtain a seismic section that accurately shows the
subsurface structure, the datum plane at a known elevation above
mean sea level and below the base of the variable-velocity weathered
layer
The value of the total statics (ΔT) depends on the following factors:
1) The perpendicular distance from the source to the datum plane.
2) Surface topography; that is, the perpendicular distance from the
geophone to the datum plane.
3) Velocity variations in the surface layer along the seismic line.
4) Irregularities in thickness of the near-surface layer.
In computing ΔT it is usually assumed that the reflection ray path in the
vicinity of the surface is vertical.
The total correction is:
ΔT = Δts+Δtr
Where Δts = The source correction, in ms & Δtr = The receiver
correction, in ms
Static Corrections
 Static Corrections
• Figure shows common shot gathers from a seismic land line,
where statics (due to near-surface formation irregularities)
caused the departure from hyperbolic travel times on the
gathers at the right side of the display.
 Deconvolution
• A seismic trace can modeled as the convolution of the input
signature with the reflectivity function of the earth impulse
response, including source signature, recording filter,
surface reflections, and geophone response.
• It is also has primary reflections (reflectivity series),
multiples, and all types of noise.
• If decon were completely successful in compressing the
wavelet components and attenuating multiples it would
leave only the reflectivity of the earth on the seismic trace.
• In so doing, vertical resolution is increased and earth
impulse response or reflectivity is approximately recovered.
 Deconvolution
 Deconvolution is a process that improves the vertical resolution of
seismic data by compressing the basic wavelet, which also increases
bandwidth of the wavelet.
 In addition to compressing or shortening reflection wavelets
deconvolution can also be used to attenuate ghosts, instrument
effects, reverberations and multiple reflections.
 The earth is composed of layers of rocks with different lithologies
and physical properties.
 In seismic exploration, their densities and the velocities at which the
seismic waves propagate through them define rock layers.
 The product of density and velocity is called acoustic impedance.
 It is the impedance contrast between layers that causes the reflections
that are recorded along a surface profile
 Decon is normally applied before stack (DBS). But it is sometimes is
applied after stack (DAS)
Deconvolution
 Mute
 This is the process of excluding parts of the traces that contain
only noise or more noise than signal. The two types of mute
used are front-end mute and surgical mute.
 Front-End Mute
 In modern seismic work, the far geophone groups are quite
distant from the energy source.
 On the traces from these receivers, refractions may cross and
interfere with reflection information from shallow reflectors.
However, the nearer traces are not so affected.
 When the data are stacked, the far traces are muted (zeroed)
down to a time at which reflections are free of refractions.
 The mute schedule is a set of time, trace pairs that define the
end of the muting.
 Mute changes the relative contribution of the components of
the stack as a function of record time.
Front-End Mute
 In the early part of the record, the long
offset may be muted from the stack
because the first arrivals are disturbed by
refraction arrivals, or because of the
change in their frequency content after
applying normal moveout
 The transition where the long offsets
begin to contribute may be either gradual
or abrupt.
 However, an abrupt change may
introduce frequencies that will distort the
design of the deconvolution operator.
Surgical Mute
 Muting may be over a certain
time interval to keep ground
roll, airwave, or noise patterns
out of the stack.
 This is especially applicable if
the noise patterns are in the
same frequency range as the
desired signal.
 Convolution to filter out the
noise may also attenuate the
desired signal.
 Figure illustrates the surgical
mute approach.
 Velocity Analysis
• The word velocity seldom
appears alone in seismic
literature. Instead it will
occur in combinations such
as instantaneous velocity,
interval velocity, average
velocity, RMS velocity,
NMO velocity, stacking
velocity, migration
velocity, apparent velocity,
etc.
• Figure can be used to
illustrate some of these
 Velocity Analysis
Instantaneous Velocity
The velocity indicated by V(xa, za) is the velocity that would be
measured at a point a distance xa from the left of the Figure and at a
depth za is an example of instantaneous velocity.

Average velocity
The velocities indicated as V1,V2, V3, etc. are interval velocities.
They are the average velocity through an interval of depth or record time
and equal the thickness of the depth interval divided by vertical time
through the interval.
Average velocity to a particular depth is simply the depth divided by the
time it takes a seismic wave to propagate vertically to that depth.
Since seismic wave propagation times are usually measured as two-way
times, the average velocity to, say, za, in Fig. is (2za/Ta), where Ta is
the two-way time to depth za.
Average velocity is required to convert time to depth.
 Velocity Analysis
Root Mean Square or RMS velocity at a particular record time, Tn,
is calculated as follows:
1. Determine what interval times sum to the value Tn
2. Square the corresponding interval velocities
3. Multiply the squared interval velocities by their interval times
4. Sum the products obtained in step 3
5. Divide the sum obtained in step 4 by Tn
6. Take the square root of the value resulting from step 5, This is the
RMS velocity at time Tn

If all reflectors are flat or nearly flat, RMS velocity is the same as
NMO velocity.

NMO velocity is the velocity used to correct for NMO. If NMO

corrections are correct, and no other factors are involved, all
primary reflections on CMP gather records occur at the same
time on all traces.
 Velocity Analysis
 Stacking velocity is the velocity that gives the optimum common midpoint
(CMP) stack output when it is used for NMO corrections.
 It may be the same as NMO velocity but if there is significant dip on
reflectors, it probably will not be the same.
 Migration velocity is the velocity that optimizes the output of a migration
algorithm, i.e. – moves the reflected energy to the correct times and places.
 Apparent velocity is determined by dividing a horizontal distance by the
time a seismic signal appears to propagate across it.
 For source-generated noise, apparent velocity and propagation velocity are
equal but for reflections, it is much faster than NMO or stacking velocity.
 Apparent velocity is important in designing velocity or F-K filters.
 Stacking velocity, migration velocity and average velocity are the most
important to seismic data processing.