INTEG, GEOPIC, ONGC, Dehradun,

Email Id: [email protected]

10
th
Biennial International Conference & Exposition

P408

SEISMIC VELOCITY MODEL BUILDING: AN AID FOR BETTER
UNDERSTANDING OF SUBSURFACE- A CASE STUDY FROM
CAMBAY BASIN, INDIA

Ajeet Kr.Pandey*, Ravendra Kumar, Manish Shukla, Anil Negi, A. K. Tandon

Summary

A seismic velocity model is necessary to map depth and thickness of subsurface layers interpreted from seismic reflection
images. Depth conversion is a way to remove the structural ambiguity inherent in time and verify structure. A good approach
to depth conversion, especially in a complex geological environment, is first to perform a depth migration with a velocity
model optimized for structural imaging, second to render the resulting laterally positioned depth image to time using the initial
velocities, and finally to convert the depth-migrated seismic data (now in the time domain) to true depth using a true vertical
velocity model. Velocity modeling is a step forward beyond direct conversion because velocity information adds two features
to the conversion to depth. First, the velocity model can be evaluated numerically, visually, and intuitively for reasonableness
(i.e., tested independently of its ability to predict depth, thus increasing its reliability), something that cannot be done with a
global time-depth correlation. Second, velocity modeling enables the use of velocity information from both seismic and wells,
providing a much broader data set for critical review and quality control.

Key words: TD relationship, raw depth grid, velocity transformations, slope relationship, mis-tie analysis, error correction.

Introduction

Present study deals with velocity modeling of Kalol field
with the help of VelPAK module of Kingdom Suite by
average velocity method derived by stacking and interval
velocities of the area. More particularly it describes an
integrated methodology of velocity modeling by using TD
relationship and time surfaces to have a better understating
of sub-surface which is most useful for further G & G
activities.

Kalol Field:

Kalol field, located 20 km NW of Ahmedabad is on
production since 1968 and produces oil and gas from nine
main pay zones The field has an aerial extent of 300 sq km.
The field is broad and is flanked by two important intra-
basinal depressions, namely Wamaj Low in the West and
Nardipur low towards east. The field has a prominent high
oriented NNW-SSE located towards the central and
western portions which appears to be a resultant of
transpressional tectonism. As per the figures of 1.4.2012
the field has In Place (PD) reserves of 149.86 MMt of oil
and 7989 MMm
3
of free gas with an Ultimate (PDD)
component of 22.55 MMt oil and 6465.5 MMm
3
free gas
component. Pay Zone II, III and IV are both oil and gas
bearing while Pay zones V, VI+VII, IX+X, XI and XII are
oil bearing.

Tectonically, Cambay basin is divided by a series of NE-
SW trending faults into five prominent tectonic blocks /
sub-basins which are bounded by transverse basement
faults, namely Sanchor - Patan, Mehsana - Ahmedabad,
Tarapur -Cambay, Bharuch - Jambusar and Narmada
Tapti blocks from north to south. The area of the current
study falls in the Ahmedabad sub block. The NE-SW
trending transverse faults are basement faults and were
reactivated in the Middle Eocene-Oligocene and Late
Miocene ages. Strike slip along the faults resulted in
segmentation of the basin, localized uplift and erosion and
development of the main structural hydrocarbon traps
(Biswas, 1987; Zutshi et al., 1997). The Ahmedabad
tectonic block has an important doubly plunging anticlinal





2
trend, namely, Kalol -Ahmedabad - Wasna trend. It is over
this trend that main Kalol field is located.


Figure.1: Index map showing study area


Figure.2: Base map showing wells taken for the study

Basic Theory:

There are mainly two methods namely Dix equation
method & Smooth function method for converting RMS
velocity function to an interval velocity function. In
present study Dix conversion is used. A brief of this
method is given below:

Dix Equation Method

This is the old standby recognized by most geoscientists
and used to compute the interval velocities posted along
with stacking velocities across the top of a stacked section.
With Dix Equation conversion, a single interval velocity is
found between any two knees within the RMS velocity
function using the equation:


Interval velocity functions are converted to RMS functions
with the equation:



The interval velocity function consists of a series of blocks,
with each block having a constant velocity, giving the Dix
Equation Method several interferences:

For each single point in the RMS velocity function, there
must be two points in the interval velocity function to
constrain the function to a uniform velocity over an
interval. Converting from RMS to interval velocity and
back again with Dix Equation does not yield exactly the
same function as started with because of the additional
points added. The result is close, but not exact.

With Dix Equation, blocked interval velocity function
can have large velocity discontinuities between blocks.
During migration or time-to-depth conversion, adjacent
seismic data samples might be processed with radically
different interval velocities introducing undesirable
artifacts into the output section.


Further smooth function is used which gives conceptual
best fitted curve for velocity. As it is seen from the
Figure.3 that original RMS function stays out as it is not
suited with the conceptual velocity curve. It has to be
modeled to get the optimum velocity in the range of
velocities calculated at the well point by changing the Dix
conversion factor or percentage. The comparison of these
two conversion method are given in Figure.3.

Finally, for the extraction of average velocity from interval
velocity in time the following equation is used:


Where Vavg & Vint represent average and interval velocities.






3

Figure.3: Representation of RMS to Interval Velocity
Conversion Options


Figure.4: Representation of actual velocity vs. model velocity


Figure.5: The paradox of velocity modeling: (a) when one
minimizes the error at the tie point (direct conversion) the model
can become unrepresentative of the true function, but (b) when
one honors the true function the error at the tie point increases
[Courtesy: Edward L. Etris et al, 2001, CSEG)

The Figure.3, 4, & 5 represents the relationship of depth
and velocity as derived by velocity modeling. The
difference between modeled and actual velocities can be
observed from the curve. The main aim of any velocity
modeling is to reduce the gap between modeling and actual
curves. (Figure.4)

In Figure 5(a & b) it is demonstrated that limitation of any
velocity modeling if velocity model is forced to produced
zero depth error at well point it loses its controls other than
the well point and in second case if it will be tried to make
the velocity model which ties over all surface then the non-
depth zero error will be produced. The velocity model
build by second case will be superior if the non-zero depth
error is minimum. One of the major causes of error in depth
predictions is misties in time between seismic horizons and
the corresponding geologic well pick in time. Direct
methods hide these errors by forcing the wells to tie, thus
altering the velocity provided independently by the well
and creating a new back-calculated velocity to ensure a
correct tie. This means that the translation function is now
no longer simply a model of the true velocity in the ground
rather it is a composite correction factor.

Methodology:

In order to create depth maps at desired stratigraphic levels
Velocity Modeling was undertaken using the available
stacking velocity data. The primary protocol followed was
to convert Time-Mapped horizon into Depth ensuring that
the resultant map did not violate the depth at which
stratigraphic level was being encountered in a well.
A data bank of 239 wells was used, spread evenly across
the seismic volume, to ensure that sufficient control points
were available for the resultant depth grid to bring out the
horizon spread within volume. These control point were
used as guides for establishing a slope relationship
between two known points at well bore and taking time
grid trend.
Data utilized:

Well data (239 vertical wells)

Well Markers

Mapped and Gridded Time Seismic Horizons

Seismic Velocity Cube (Stacking Velocity)

TD / Sonic Log (77 verticals wells)

The desired data, namely, seismic survey, well, well picks,
TD or Sonic logs, Velocity volume, mapped horizons / grid
were imported into the VelPak module for velocity
modeing. It is to be ensured that the time range of the





4
imported grid in VelPak matches with those in actual
interpretation. Later, Velocity Module is used for
displaying the staking velocities with respect to activated
grid. After this the staking velocities are converted to
interval velocities using the Dix Conversion. The resultant
Interval Velocity volume is used to extract average
velocity corresponding to active time grid in X,Y and Z
format where Z is the average velocity while X, Y are the
co-ordinates of Inline and Cross Line cross overs in the
active time grid. Validation of these average velocities at
well positions is shown in Fig.7.

These X, Y and Z values are gridded using available
gridding algorithms (Collocated co-kriging or global
method), during this procedure the corresponding Time
Grid is used as a Parent Grid which provides control trend
for the velocity gridding (Fig.13). This grid is further
validated at well point to have a better control for depth
conversion. Example of this validation is given at well
location A in Fig.7. Later the depth conversion is done
using Simple Average Velocity method where above
gridded average velocities (Fig. 13) are used. This output
is used to produce Raw Depth Grid.

It is important to define the stratigraphic levels at which
the depth needs to be created. These layered definitions are
propagated at each of the chosen well to ensure uniformity
in establishing a time depth relationship (Fig. 8).


Figure.6: Process flow of velocity modeling considered for the
study

Prior to advancing in depth creation it is imperative to
establish a Tie between well picks / formation top and
Raw Depth Grid which will quantify the mistie or the
error value at each well position (Fig. 8& 9). These error
values are further grided to generate error grid. This error
grid, linked to a particular stratigraphic level (Parent
Grid) is used to create Final Depth Grid by compensating
these error factor/ values in the Raw Depth Grid extracted
earlier (Fig.10).The Final Depth Grid, thus generated, is
exported to use for map generation purpose.

It is also advisable to export, independently, the error and
average velocity grid so that these too can be mapped for
the purpose of quality check.


Figure.7: Validation of gridded average velocity values
(extracted vs. actual)


Figure.8: Defining layers for calibration and tie processes


Figure.9: Calibration process with respect to K-III top





5

Figure.9: Tie process with respect to K-III top


Figure.10: Error grid generation and analysis


Figure.11: Time & Depth structure map at K-IX Top

Results & Discussions:

This is a case study and gives the practical approach for
velocity modeling which can be further utilized for time to
depth conversion.



Figure.12: Time & Depth structure map at K-III Top

The depth map of K-IX level has been created using well
tie with 184 wells of Kalol and Wadu fields (Fig.11).The
position of prominent high and lows remains intact but
there are places where the axis of the lows appears to have
slightly changed. The axis of low, located towards the
north eastern portion of the area, changes from E-W to
NNW-SSE. Within the large flat area occurring towards
central east portion of the map, few lows appear which
were not so clear in the time domain map. The high and
low justify the mapped faults. At K-IX level, there is a
good match (around 97%) between the Formation tops and
the depths values read from the depth map at well
positions. Thus the overall slope picture created on depth
matches with that observed in time.


Figure.13: Average velocity map at K-IX & K-III level

The depth map at K-III level has been created using 185
wells of Kalol and Wadu field which were falling within
the area (Fig.11). The time and depth map appear to be
quite similar and most of the geomorphic features
identified in time remain preserved in depth also. The axial
trends of features have maintained their azimuthal
relationship in time and depth. However at local levels
there are small pockets in which the lows have widened





6
while the high points have converged. These changes are
localized and subtle. The high and low justify the mapped
faults. Thus at K-III level, there is a good match (around
98%) between the Formation tops and the depths values
read from the depth map at well positions.

Conclusions:

For proper reservoir development it is important to
demonstrate subtle stratigraphic surficial undulations
which help in accurate demonstration of sub-seismic faults
and fault sealing efficiencies. The depth pictures obtained
by average velocity modeling better demonstrate the sub
surface undulations at stratigraphic levels. The depth
matching at each well location (having K-III and K-IX
tops) was performed for validation of this method which
comes around 98% for K-III and 97% for K-IX levels.
Moreover it is inferred from the study that the proper
conditioning of sonic logs, TD relationship, velocity
volumes & selection of right methodologies are utmost
important to build a right picture of sub surface through
seismic velocity modeling. Validation away from well
location was also done to know the depth pattern between
different well locations. It has been seen from the time and
depth maps at these levels that the preservation of major
structures (deeper and shallower event) are retained.
Sometimes it is very tough to have the same picture of
structure in time and depth due to inaccurate velocity
volume and improper velocity modeling methodologies.
These methods generally insist the depth surfaces to tie
with the well tops due to which distortion of overall depth
picture was introduced. While making velocity model of
any area one should choose the methodology as per data
availability, e.g. if sonic logs are not conditioned then
better to use TD relationship of the wells.

Acknowledgement:

We wish to express the sincere gratitude to Director (E) for
giving an opportunity to do this project at GEOPIC. We
express our heartfelt thanks to HOI-GEOPIC for his
excellent guidance and valuable suggestions during the
execution of this project.

We extend our sincere thanks to Asset Manager,
Ahemadabad Asset & his team for time to time discussion
and support.

We are thankful to the members of the interpretation
group, for their kind cooperation and valuable guidance.
N.B. The views expressed here are those of the authors
only and do not reflect the views of ONGC.

References:

Edward L. Etris, Nick J. Crabtree, Jan Dewar: True Depth
Conversion: More Than A Pretty Picture, CSEG Recorder,
November, 2001

Samuel H. Bickel, Veritas Geophysical Services, Houston:
Depth mapping of time horizons, SEG Expanded Abstract,
1998

Cameron M., Fomel S., and Sethian, J. : Time-to-depth
conversion and seismic velocity estimation using time-
migration velocity, Geophysics, 73, no. 5, VE205-VE210,
(2008)

MELVAh D. CARTER: Depth conversion using
normalized interval velocities, GEOPHYSICS: THE
LEADING EDGE OF EXPLORATION JANUARY, 1989
Kaufman, H., 1953, Velocity functions in seismic
prospecting, Geophysics, 18, 289-297.

Reilly, M., 1993, Integration of well and seismic data for
3D velocity model building, First Break, 11, no. 6, 247-
260.

Ross, W. S., 1994, The velocity-depth ambiguity in
seismic traveltime data, Geophysics, 50, no. 5, 830-843.

Schultz, P., 1999, The Seismic Velocity Model as an
Interpretation Asset, 1999 Distinguished Instructor Short
Course, Distinguished Instructor Series, No. 2, SEG.

Surdam, R.C., 1997, A New Paradigm for Gas Exploration
in Anomalously Pressured Tight Gas Sands in the Rocky
Mountain Laramide Basins, in R.C. Surdam, ed., Seals,
traps and the petroleum system: AAPG Memoir 67, 283-
298.

Tieman, H.J., 1994, Investigating the velocity-depth
ambiguity of reflection traveltimes, Geophysics, 59, no.
11, 1763-1773.