Land Acquisition for land seismic operations

1

1. Land Seismic Acquisition
Parameters – Part 1

2. Swath or Patch Acquisition
Land data is acquired in swaths or patches.
Land swaths or patches have minimally at least one receiver line that is common
between two adjacent patches so refractive statics can be calculated in
processing providing surface consistency.
Size of the patches are determined by the equipment on the crew and the survey
design.

3. Wide azimuth data
Wide azimuth data improves the illumination of reservoirs beneath complex
structures and provides images that can clearly qualify reservoir
compartmentalization or permeability, This in turn will affect the reservoir
simulation that is run to determine the amount of hydrocarbons recoverable, and
it will affect the economics of the field and possibly the facilities used offshore.
With patch acquisition there are an increasing number of traces as offset
increases. This can help with multiple attenuation by emphasizing those ranges
of traces which exhibit the largest separation between primary and multiple
signal (Bouska, 1998).

4. COV Tiles
Each COV tile couples the
azimuth and offset binning
naturally. COV Tiles also handle
irregularity better than
conventional offset gathering in
many aspects.
1-Fold
The reciprocal azimuth is added
so that across the tile there is
only 1-fold, and it is more
spatially continuous with
discontinuities between the
tiles.
COV tiles are the ideal data
domain for Kirchhoff migrations.
Many design the seismic
acquisition to properly populate
the COV tiles.

5. Parameterization of seismic data
acquisition (Parameters can be used to buy
spec data)
Migration Aperture
Does the area of the
survey cover enough of
the area of interest to
properly image the
structures
Size of the smallest
structure to image
Depth of shallowest
zone of interest
1.5 x the depth of the
deepest zone of
interest
Thickness of zone of
interest
Size of survey Source and receiver
station spacing
Source and receiver
line spacing
Size of recording swath Maximum Frequency
Number of shots and
receivers and time to
acquire
Cost
Stop
Inline Fold Crossline Fold Source type
3D Fold at different
levels
Recording Parameters
Not affordable
Affordable
Based on Hardage
(2010)
Structural imaging

6. Test Line
Acquire 2D
test line
Check different
sweeps of data
Brute stack using
elevation statics
Though we mention high density, the survey should be designed so that we have
just enough to meet the requirements for interpreting the data yet be
economical.
Analyze the
data looking at
frequency,
penetration,
etc.
Seismic acquisition tends to be a compromise to acquire the data economically
but acquiring the data to achieve our goals.
Not about quantity but about quality that is why the test line is important.
Or, check shot
points &
charge size

7. In the past seismic data was acquired to obtain a better full stack to do
interpretation but there are issues with the full stack understanding the
structure.
Past goal of seismic acquisition – full
stack
Higher order move out need to flatten to
see AVO
Multiple
Velocity too fast because presence
of hydrocarbons slows down the
velocity.

Frequency decreases
Natural gas
30 Degrees
Mute tends to be 30 degrees missing far offsets
The fold is important for SNR.

8. Currently, the goal is angle stacks which enable us to calculate rock properties. Gather
conditioning is used to produce flatter gathers resulting in higher resolution.
Present goal of seismic acquisition –
partial stacks
Flatter gathers after 3
rd
order moveout
Residual velocities creating flatter gathers
Natural gas
45
Multiple attenuated
with radon
Nears MidsFars
3015
Intercept Gradient Curvature
Vp Vs Density Rock Properties for the geomodel
VVAz and AVAz (azimuthal data)
Fold for the angle stacks is important
instead of fold for the entire survey.

9. Rock Properties From Seismic
Inversion Products
Seismic Rock Property Equation
P-Impedance (Acoustic Impedance) I
p=AI=r*a
S-Impedance I
s=SI=r*b
Vp/Vs a/b
Poisson Ratio PR(J)=(0.5a
2
)-b
2
/(a
2
-b
2
)
Poisson Ratio PR(J)=0.5*(IP
2
-2IS
2
)/(IP
2
-IS
2
)
Poisson Ratio PR(J)=0.5*lr/(IP
2
-IS
2
)
Lambda Rho lr=(AI
2
-2SI
2
)=(I
p
2
-2I
s
2
)
Mu Rho mr=SI
2
=I
s
2
Near Impedance NI=(AI
2-AI
1)/(AI
2+AI
1)
Elastic Impedance (q)
Young’s Modulus E=rb
2((3a
2-4b
2)/(a
2-b
2))
Young’s Modulus E=2I
S
2
(1+ PR(J))/r
Young’s Modulus Density
(lithology and brittleness)
Er=mr(3I
P
2
-4I
S
2
)/(I
P
2
-I
S
2
)
Closure stress scalar PR(J)/ (1- PR(J))
Bulk Modulus (k) k=I
pa-4/3I
sb)sin41()sin8()tan1(
222
qqq
r
kk
sp VVEI
--+
=

10. Rock properties can come from a geostatistical, multi-linear regression or neural network
analysis to find the optimum transform which relates the seismic attributes to any well log
data such as Vp, Vs, Density, Vp/Vs, Porosity etc.
Seismic
Attribute
Seismic
Attribute
Seismic
Attribute
Seismic
Attribute
Neural Net
Well Logs
Optimal Well
Property
from
Combination
of Seismic
Attributes
Transform
Neural Net using Seismic Attributes
Neural Net volumes have higher resolution than seismic data allowing for thinner beds to
be seen.

11. What can be Derived From New
Seismic Data
The integrated team needs to define what they need from the new seismic
volume.
As can be seen rock properties can be derived from a multitude of seismic
attributes and these rock properties can be co-kriged with well data within the
geomodel to guide the interpolation between the wells.
Neural Net has higher resolution than seismic data allowing for thinner beds to
be seen.

12. Different Stages of Geomodeling
Can see with the use of machine learning the new seismic volume and
the data obtained from it can lead to understanding individual frac
stages which is the goal of our work leading to full field development.

13. It is important that we do modeling before we do seismic acquisition to
understand:
1) Seismic resolution
Acquiring seismic data need to look at
resolution and fluid and lithology
2) Possible types of AVO and understand the offsets that are required to
illuminate AVO
In most of our unconventional plays we are dealing with thin sands, and we want to
be able to illuminate sweet spots which are defined by lithology and fluid content.
By using proven techniques in conventional reservoirs, we can obtain sweeter
spots to place our horizontal wells.

14. Migration aperture and Fresnel Zone
Migration aperture is the distance needed to properly collapse the seismic
diffractions to image the geological structure properly.
To migrate the data properly the data needs to be across the whole migration
aperture.
Full Fold
0 40 40 0
Migration aperture is measured only in the full fold area.
Data outside of the migration aperture will not allow the structures to be
properly imaged due to the diffractions outside the migration aperture not being
collapsed properly or imaged properly.
Prospects outside of the migration aperture should never be drilled.
Migration Aperture

15. Fresnel Zone
Full Fold Area
Migration
Aperture
Apron around the survey
is low fold.
The full fold area is
where we have the full
fold.
The migration aperture
is where structures can
be properly collapsed.

16. How to Calculate the Fresnel Zone
Horizontal resolution before
migration: 255 m
Horizontal resolution before
migration: 1565 m
Horizontal resolution is dependent upon the depth, frequency and the velocity.
The deeper the event the lower the horizontal resolution.
????????????�?????? ���=
�??????� ??????�??????
2
�
0/����
Time Vel RMS Freq Fresnel Zone
1 1800 50 255
2 2000 40 447
3 3000 30 949
4 3500 20 1565

17. Vertical Resolution – Spectral
Decomposition
Vertical resolution and tuning are well known by most Geophysicists because of
spectral decomposition.
The basis of spectral decomposition is the stratigraphy resonates at wavelengths
which are dependent on the bedding thickness.
Subtle thickness variations and discontinuities can be imaged, and the thickness
of the formation can predict bedding quantitatively using vertical resolution
equals to 1/4l and l=a/frequency where l is the wavelength, a is the rock
velocity (interval velocity) and frequency.
With Spectral Decomposition the seismic is broken down into frequency bands
and allows for the quantification of amplitude variation with frequency, and
insight into the distribution of stratigraphic entities, faults and fractures, and/or
hydrocarbons.
Vertical resolution is dependent upon the frequency output of the source and the
processing of the seismic data.

18. Horizontal Resolution
The horizontal resolution is less understood by most Geophysicists.
Horizontal resolution tends to be more driven by the acquisition and processing
of the data.
Horizontal resolution before migration is equal to the migration aperture or
Fresnel zone.
After migration, the imaging aperture has shrunk to a small circle equal to 1/4l.
Spectral Decomposition can be used to bring out stratigraphy, subtle low
amplitude faults, discontinuities, heterogeneities within the data.

19. The lower frequencies help reduce side lobes in the wavelet.
Sidelobe energy
Thin beds
Wavelet has a spikier shape
and reduced sidelobes which
allows a less ambiguous
interpretation of close
reflectors.
Low frequency waves suffer less from attenuation and diffraction, which leads to
a deeper penetration.
Lower frequency seismic acquisition
(Broader Band)
Low frequency content allows a larger part of the low frequency model to be
derived directly from seismic when inverting the seismic data.

20. Why do we need low frequencies
Having low frequencies in the data:
1) Minimizes the amount of the low frequency model required for the
Deterministic Inversion.
2) If low frequencies are present the Relative Inversion approaches the
Deterministic Inversion.
3) Full Waveform Inversion (FWI) require low frequencies to help determine the
kinematically relevant, low-wavenumber components of the velocity model,
which are in turn are needed to avoid convergence of FWI to spurious local
minima (Li and Demanet, 2016).

21. Importance of near angles
Acoustic Impedance is inversion of the near angle stack or AVO Intercept.
AVO Intercept is the projected zero offset stack utilizing the gathers or the
partial stacks.
With the near angle stack and Intercept, the angle of incidence is low, so the
energy is going straight down and coming straight up as it is reflected off the
rock interfaces; therefore, the amplitude is proportional to the reflection
coefficient.
The near angle stack and the AVO Intercept tends to have higher frequencies
and the gathers tend to be flatter on the nears, which gives us a better stack.
Spectrally enhanced Acoustic Inversion with structural smoothing is a good
dataset to be used for coherency, curvature, etc.. It illuminates the fault edges
and allows us to see faults and discontinuities better.

22. Where is the fluid information on
gathers
Reflects lithology/porosity
because the Far Offsets tend
to represent P-wave and S-
wave data
The fluid effect is on the near offsets, and this is why they
are important in the seismic acquisition.

23. Importance of far offsets
Far angles illuminate the
AVO in the gather.
Class 2 AVO
Class 3 AVO
Class 2 AVO
It is the AVO in the seismic
data that drives the
success of the inversion.
Inversion reduces the
sidelobes with the low
frequency model and
removes the wavelet
effect increasing the
resolution.

24. Determination of the maximum far
offset – Picking refractors for statics
The far offset must be large enough so that the refractors can be properly picked
in order to calculate the best refraction statics. If the far offset is not large
enough then there may be issues in calculating the best refraction static
solution.
Reflector interferes
with refractor

25. Determination of the maximum far
offset – Depth ZOI
Maximum offset should be 1.5 times the depth of the ZOI.
Depth of ZOI
1.5 * Depth of ZOI
For Depth = Offset it equals
30 Degrees of incidence
angle which is the limitation
of the 2-term NMO equation
and the Straight Ray PSTM in
processing.
For 1.5 * Depth = Offset
incidence angle >30 Degrees
of angle requiring 3rd-term
NMO equation and the Curve
Ray PSTM in processing.
It will illuminate AVO in the
data better.

26. Determination of the maximum far offset –
mute pattern previous processing
Mute from previous processing of a 3D or 2D dataset in the area. This can tell you
what was used before.
Care needs to be taken to look at the mute pattern compared to the seismic
data.
Currently there are techniques to flatten the data out to farther offsets so this
mute pattern is just a guideline.

27. Determination of the maximum far
offset – multiple removal
Multiple
Area where
multiple can
be removed
Maximum offset
for multiple
removal

28. Determination of the maximum far
offset – AVO from well modeling
Biot-Gassmann substitution may be used to determine the far offset for seismic
data.
Class 3 AVO
Class 2 AVO
Class 2 AVO
Class 2 AVO
Class 3 AVO
Class 2 AVO

29. Determination of maximum far offset
The far offset should be whatever is greater from:
Offset required to pick refractors for calculation of refraction statics
1.5 times depth of ZOI
Mute pattern previous processing
Maximum far offset to cleanly remove multiples on the far offsets
Far offset required to illuminate any AVO seen in Biot-Gassmann modelling

30. Steps in designing a seismic survey
Using synthetic seismograms from well logs to determine the maximum frequency
to resolve the thickness of the target formation.
Fmax = (FDom*VInt)/4

Use well data and Biot-Gassmann substitution to determine the factors that will affect
the illumination of the hydrocarbons in the seismic data (AVO modeling). This can
also allow us to determine the maximum offset to improve the AVO response. A
general rule of thumb is the maximum offset in the zone of interest (ZOI) equals the
1.5 times the depth.
Use well data to create a wedge model to understand tuning.
Use older seismic data to understand frequency content, offset distribution using
previous mutes on the data, and understand tuning in the data.

31. Steps in designing a seismic survey
Use 2D data to determine the size of the structures, understand possible any AVO
present in the data, help determine the 3D acquisition parameters.
Determine the size of bins. Equal or square bins should be used to get the same
migration impulse response in the inline and crossline directions.
Determine the fold which are the number of traces in a gather. Fold reduces the
amount of noise in the data.

32. Seismic Modeling
Seismic acquisition modeling is required to determine:
1)Source type
2)Receiver type
3)Source spacing
4)Receiver spacing
5)Number of source shots
6)Number of receivers in a receiver array
7)Number of receiver channels in a receiver spread
8)Sampling rate
9)Record length
10)Predict cost
Seismic models are needed through the lifecycle of the data.

33. Conclusion
Seismic acquisition is a series of compromises that are made to be able to meet
budgetary needs set by the oil and gas company.
It is understanding these compromises and how it will affect the results of the
data. This is the purpose of these power points.
The goal is to balance seismic acquisition, processing, and interpretation, and to
do this we need to understand a little bit of each to achieve balanced results
that we want.
The integrated team within the oil and gas company needs to set the goals of
what will be achieved from the data to help make the necessary compromises.
Geophysics is difficult because it requires understanding so much, but our
knowledge only needs to be limited so that we can make decisions.
Goal of this series of power points is for you to use them as a reference and to
understand just enough to ask the right questions of the experts to be able to
make decisions.

34. Thank you!!!!