Interpretation of 3D Seismic and Subsurface data from the Buffalo ...

Interpretation of 3D Seismic and Subsurface data from the
Buffalo Valley Field, Chavez county, New Mexico
Sandeep Pyakurel, Thomas H. Wilson, Jaime Toro, and Shahab Mohaghegh
Department of Geology and Geography, 419 White Hall, West Virginia University, Morgantown, WV 26506-6300
Abstract
Structure and isopach maps derived from formation top picks
reveal the presence of differential subsidence across the
Buffalo Valley field during the Pennsylvanian and Permian
periods. The Buffalo Valley field is located along the Northwest
Shelf of the Delaware Basin. 3D seismic data reveal that the
field is bounded by two normal faults, both of which drops to
the southeast. The fault bounded block rotates slightly to the
southeast during deposition of the reservoir intervals.
Production from the Morrow Formation is concentrated within
this fault block. The Morrow production trend in this area is
roughly north-south at an angle to the northeast trending
Northwest Shelf in this area.
3D seismic data volumes consisting of P-wave and PS-wave
data are interpreted and compared in this study. VSP data
(zero offset an some 3D VSP coverage ) are incorporated into
the seismic interpretation. The tie between surface seismic
ands well-log derived synthetics is generally poor due to
Fresnel zone effects in these heterogeneous intervals.
Correlation of formation top depths to arrival times in the zerooffset
corridor stack are used to identify reflection events in the
surface seismic data. In this study we examine the relationship
of reflection amplitude and other seismic attributes associated
with reflections from the Atoka-Morrow interval to 5- year
cumulative production data from the field.
Background
The study area is located within the Buffalo Valley field in the
southern part of the Chavez County, New Mexico (Figure 1).
The area lies on the Northwest Shelf of the Delaware Basin.
Production in the Buffalo Valley field is largely from a channel
sand at the base of the Morrow Formation. Reservoir depth in
the field is approximately 9000 feet subsurface. The Morrow
Formation consists mostly of interbedded sand, shale and thin
limestone deposits. The reservoir quality sandstone lies in the
lower part of the Morrow formation and is interpreted to be an
inter-distributary channel fill deposit. The base of the Morrow
coincides with the Mississippian unconformity. Channels at the
base of the Morrow scour into the Mississippian Barnett shale,
which overlies the Chester Formation. The top of the Morrow is
overlain by Atoka Formation which is comprised mostly of sand
shale and carbonate.
Figure 1: Location map.
Seismic data from the
Buffalo Valley area were
provided by WesternGeco
Synthetic Tie
The tie between reflection events and
formation tops was established at the
Read and Stevens, Inc., Lula #3 well (API
No. 30-005-60481). Data available for the
Lula #3 include a near offset VSP and
near offset corridor stack (Figure 3).
Arrival times of reflection events in the
surface seismic data were approximately
44 milliseconds early relative to those in
the VSP corridor stack. Weathering
corrections and surface consistent static
corrections are the likely cause of this
difference.
0.600
0.700
0.800
0.900
1.000
1.100
Trobough 1-A
Lula3
Bogle State Com 1
6035 ft 4625 ft
- Line 1034 + Corridor & Synth -
SP: 1.0 20.0 40.0 60.0 80.0 100.0 120.0 140.0
A
Wolf
Cis
Can
St
At
1C
2C
Miss
WDev
0.600
0.700
0.800
0.900
1.000
1.100
Figure 3: Surface
Seismic, VSP Corridor
Stack & synthetics
Fresnel zone effects likely prevent a
good match of synthetic to corridor stack
and VSP corridor stack to surface
seismic. Some of this heterogeneity can
be observed in the 3D VSP data around
the Lula #3 well (Right). The Fresnel
zone radius at lower Pennsylvanian
Morrow depths is approximately 1500
feet using an RMS velocity of 16400 feet
per second derived from the well velocity
survey, and dominant frequency of
approximately 62.5 Hz. Given the 3D
0.800
0.900
1.000
1.100
1.200
SP:
Lula3
- Corridor Stack & Upgoing VSP -
75.0 100.0 125.0 1.0 25.0 50.0 150.0 154.0
Wolf
Cis
Can
St
At
1C
2C
Miss
WDev
Figure 2: Zero Offset
VSP & Corridor Stack
Surface seismic data in the vicinity
of the Lula #3 well (Figure 3) along
with the near offset corridor stack and
bandpass filtered synthetic seismic
traces compiled from the Lula #3
sonic illustrate the seismic tie used to
interpret reflection events. The east
west inline 1034 is split at the Lula #3
well. The corridor stack along with a
series of bandpass filtered synthetics
have been inserted into the gap at
the well location for comparison to
surrounding seismic data. Formation
top correlations to reflection events
are illustrated. The section is quite
heterogeneous at local scales, and
Lula3
Line:
Trace:
1.000
1.100
2069.0
10340.0
St
At
1C
2C
Miss
Figure 3: 3D VSP
VSP response, it is not surprising to see disagreement in the
details of the seismic response observed between the various
data sets and the synthetic seismogram. Based on the above
comparisons we place the base of the Morrow (top of the
Chester) at 1.074 sec in the surface seismic. The reflection from
the top of the Atoka arrives at approximately 1.041 sec.
1.000
1.100
0.800
0.900
1.000
1.100
1.200
Figure 5: Type log for study
area (Well 30005060321)
Top of the
Atoka
Top of the
Morrow
Morrow Basal
Channel
Top of the
Barnet Shale
Top of the
Chester
Seismic time structure
Time structure on the Chester and
Atoka reflection events (figures 6 and
7, respectively) reveals that the field is
bounded along its western and eastern
margins by two faults across which the
strata drop to the east. Seismic profiles
from the 3D survey (Figure 8) reveal
that these faults are nearly vertical.
The isochore map for the
Atoka to Chester reflection
events (Figure 9) suggests
some movement along
these faults occurred during
deposition of this lower
Pennsylvanian interval.
Log Response
Formation top picks for most of the
wells were based on the gamma
ray and density log response. The
top of the Morrow (Figure 5) was
generally difficult to pick but is
often associated with a drop in
density associated with the
predominantly clastic Morrow
deposits. The basal Morrow
channel is a low γ ray, high
porosity zone that overlies the high
Top of the Morrow
γ ray Barnett shale (Figure 5)
Basal Morrow
Channel
Figure 6: Time structure
on top of the Chester Fm.
Figure 7: Structure contour
map: top of the Atoka Fm.
Figure 8: Seismic profile
passing through the Lula #3
well.
Can – top of the Canyon
Formation;
St – top of the Strawn
Formation;
At – top of the Atoka
Formation;
1C – top of the Mississippian
Chester Formation.
Figure 9: Atoka to Chester
Isochore Map.

Structure on the top of the Chester
and Atoka formations is shown in
Figures 10 and 11. In general the
structure follows the northeast trend
of the Northwestern Shelf; however,
north trending structural indentations
interrupt the regional trend within the
field. Comparison of the time and
structure maps suggests that the
north trending faults observed in the
3D seismic data are probably
associated with the structural
indentation portrayed in the well
derived structure maps.
Figure 11: Structure on the
top of the Atoka Formation
Structure & Isopach Maps
Figure 10: Structure on the
top of the Atoka Formation
Figure 12: Isopach map: top of
the Atoka to top of the Chester
The isochore map of the Atoka to Chester top (Atoka through
Morrow) reflection times (Figure 9) shows thickening along the
westernmost fault as well as across the southern extent of the fault
block in the survey area. The southern part of the block appears to
have rotated south-southwest during deposition, and at the same
time there appears to have been local subsidence across the
western fault that diminished to the north along its length. Reduced
travel time differences observed in the Atoka-Morrow isochore to
the north, along the trend of the western fault, show some similarity
to the decreased thickness of the interval observed in the isopach
map (Figure 12).
The Atoko-Morrow isopach map (Figure 12) reveals that in the
northern part of the 3D survey area the Atoka-Morrow interval is
over 100 feet thick, and increases to over 200 feet thick in the
south. The well derived maps provide a much coarser view of the
subsurface, with well spacing of about 2000 feet on average.
However a pronounced trend of thickened Atoka-Morrow is clearly
revealed along the trend of the seismic interpreted fault block.
Future efforts will include conversion from timer to depth based on
available sonic and well control from the field
Production Data
Five year cumulative production data from the Buffalo Valley field
are shown in figures 13 and 14. Production data from two wells
(highlighted in yellow) in the southwest corner of the area were
omitted since production from these wells was much higher (60471
and 25725 Bbl ) than that in surrounding wells making it difficult to
portray the production pattern on a linear contour scale when these
two wells were included. The average 5 year cumulative production
obtained from 57 wells in the field is 4842.07 Bbl
Figure 13: Oil production (units
in barrels of oil)
Seismic Attributes
Figure 15: Average instantaneous
frequency observed in a 0.014 sec
window centered on the Morrow
Figure 14: Gross sand
thickness within the
Morrow Interval
Figure 16: Average instantaneous
envelope observed in a 0.014 sec
window centered on the Morrow.
Preliminary attribute assessment suggests some relationship
between spatial variations of instantaneous frequency (Figure 15) and
envelope (Figure 16) to production distribution. The average value of
the attribute was extracted from a 14 ms window centered roughly in
the Morrow reflection event. A distinct channel-like pattern is not
observed in the attribute maps. The productive Morrow sands are
generally less than 30 feet thick and are below the resolution limit of
the 3D seismic data; however, areas of low instantaneous frequency
coincide roughly with highs in the 5-year cumulative production within
the survey area.
P to S converted wave data
The 3D seismic data set also includes a converted P to S wave
volume. Side-by-side examples of the P-wave and PS-wave data
are shown below (figures 17 and 18 respectively). A pronounced
zone of reduced amplitude is observed in the PS-wave data
volume that is not observed in the P-wave section. This zone of
low amplitude and reflection coherence may be associated with a
narrow fracture zone (potentially gas filled).
Figure 18: Display
of P to S-wave
data along profile
shown in Figure
17.
Figure 17: P-Wave
seismic profile
through the Lula #3
VSP well.
Conclusion
A tie between synthetic and surface seismic proved difficult to
establish. Comparison of well log derived synthetics to the VSP
near-offset corridor stack was essential to identifying two-way
travel times to formation tops and corresponding reflection events
observed in the 3D surface seismic. The response of the Atoka-
Morrow intervals observed in the 3D VSP confirms observations
of the well log response in the field that the Atoka-Morrow interval
is heterogeneous at interwell and seismic (Fresnel Zone) scales.
The log response of the Morrow, is difficult to correlate from well
to well whereas the top of the Chester Formation proved to be a
more useful marker horizon.
The 3D seismic data from the field reveal that the play is
bounded by nearly vertical faults that step down to the east. The
faults are oriented roughly north-south at an angle to the NE
trending shelf in this area. The eastern fault terminates to the
north while offsets along the western fault decrease to the south.
There appears to have been some movement along the faults
during deposition of the Atoka-Morrow interval. Observable fault
offsets are located near the center of the field on the western fault
and near the southern end of the field along the eastern fault.
Production from the Pennsylvanian Morrow Formation along
the Northwest Shelf of the Permian Basin is generally confined to
channel sands at the base of the Morrow. These sands are
generally identifiable by low γ ray and high neutron porosity log
response. Log derived structure and isopach maps reveal the
influence of the faults observed in the seismic. Structure maps
clearly reveal that the field sits on an interruption in the regional
trend of the Northwest Shelf. The Atoka-Morrow isopach reveals
thickening of the Atoka-Morrow interval along the trend of the
fault block. Atoka-Morrow thickening increases from north to
south suggesting that the fault block may have rotated to the
south during deposition of the Atoka-Morrow interval, consistent
with observations made in the 3D seismic data volume.
Analysis of the P to S converted wave data suggests it will help
identify additional faults or fracture zones within the field. Nearly
vertical low amplitude zones appear in the P to S-wave data that
could not otherwise be easily identified in the P-wave volume.
The analysis of seismic attributes and their correlation to
production is in its early stages. Possible relationships between
spatial variability in seismic attribute response and production
from the Morrow channel will be examined in more
comprehensive detail as the study continues.
References
Oil and gas fields of southeastern New Mexico, Roswell Geological
Society, 1999 & 1977 Supplements.
Acknowledgements
This research was funded through DOE/NETL contract DE-FC26-
03NT41629. Our thanks to WesternGeco for providing the 3D
seismic data from the area. The comments of Dave Oldham were
greatly appreciated. Much of the well data were obtained from the
New Mexico Energy, Minerals, and Natural Resources Department
Oil Conservation Division at http://ocdimage.emnrd.state.nm.us/.
Production data used in the study were obtained from the New
Mexico Tech “Go-Tech” site at http://octane.nmt.edu/data/ ongard/.