https://dx.doi.org/10.4314/ijs.v21i1.

1
Ife Journal of Science vol. 21, no. 1 (2019) 001

ON THE EFFECTIVENESS OF USING QUANTITATIVE AVO ANALYSIS IN

FLUID AND LITHOLOGY DISCRIMINATION IN AN OFFSHORE NIGER
DELTA FIELD, NIGERIA
Adeoti, L.*, Ikoro, C. O., Adesanya, O. Y., Ayuk, M. A., Oyeniran, T. A and Allo, O. J.
Department of Geosciences, University of Lagos, Akoka, Lagos, Nigeria. *Corresponding Author, Email:
[email protected] (Tel.: +2348034739175)
(Received: 24th July, 2018; Accepted: 1st December, 2018)

ABSTRACT

Quantitative Amplitude Variation with Offset (AVO) analysis of “Jay” Field, offshore Niger Delta, was carried
out with a view to properly discriminating fluid and lithology using near, mid and far offset seismic and well data.
Seismic and well data were interpreted and analyzed. Synthetic seismogram was generated using density (r) and
sonic logs. AVO modeling, seismic AVO attribute analysis and AVO inversion were carried out and the results
from well log interpretation using 70-API gamma-ray cut-off, neutron-density over lay and resistivity logs
revealed that the field consists of intercalation of sand and shale with typical deltaic depositional environment
log signatures. Four identified sand reservoirs (a, b, c and d) with high resistivity values and negative separation in
the neutron-density overlay suggested that the field was hydrocarbon bearing probably containing gas or
condensate. Two sand reservoirs showed good rock physics results, 'Sand a' at 11,632 ft TVD with 18% porosity
(ϕ), 0.25 water saturation (Sw), decreasing ratio of compressional wave velocity to shear wave velocity (Vp/VS)
and Poisson's ratio () relative to the background shale signified AVO response typical of a hydrocarbon bearing
sand. 'Sand e' at 5,925 ft TVD, with 30% ϕ, Sw of 1, no change in Vp /VS and relative to the background shale
implied that an AVO response was unlikely. Gradient analysis result for the synthetic seismic at the top and base
of the two sands agreed with Rutherford's classification scheme for class IV AVO for 'Sand a' and no AVO
response for 'Sand e'. AVO attribute analysis and impedance inversion of the seismic volumes confirmed AVO
result for the two sands. The study established that AVO technique could be effectively used for fluid and
lithology discrimination in the “Jay” Field, Niger Delta.

Keywords: Amplitude Variation with Offset (AVO), Seismic and Well data, Rock Physics, Reservoir Sands

INTRODUCTION multinational companies have been awarded some

Several authors have shown that the Niger Delta deep offshore blocks and even ultra-deep
Basin has spectacularly maintained a thick concessions.
sedimentary apron and salient Petroleum
geological features favorable for petroleum Seismic reflection method has been effective in
generation, expulsion and trapping from the such challenging areas in detecting structures
Onshore through the Continental shelf and to the capable of trapping hydrocarbon but among the
deep water terrains (Whiteman, 1982). From the various seismic technique for hydrocarbon
very beginning of oil exploration in Nigeria in detection and monitoring in the subsurface, the
1937, till early 1993, virtually all exploration and Amplitude Variation with Offset (AVO) analysis
production activities were restricted to land and appears to be quite promising with pore fluid
swamps. Where prospecting ventured offshore, it identification. Only AVO analysis, which requires
was in areas not greater than 200 m water depth. special handling of the data, can distinguish
But then in 1993, the Federal Government opened lithology changes from fluid changes (Castagna,
a new frontier in oil and gas exploration, heralding 1993; Chiburis et al., 1993; Adekanle and
the bright prospects of a promising future, by Enikanselu, 2013; Adeoti et al., 2017). Almost all
allocating some offshore blocks in water depths major companies use AVO routinely as a tool to
reaching 2500 m. These deep water depths and "de-risk" exploration targets and to better define
greater depths than 2500 m will undoubtedly the extent and the composition of existing
impact positively the country's production and hydrocarbon reservoirs (Omudu and Ebeniro,
reserve blueprint. Though these deep-water 2005; Castagna and Chopra, 2007; Adeoti et al.,
operations are technically challenging and 2014; Schlumberger, 2014).
massively capital intensive, experienced
002 Adeoti et al.: On the Effectiveness of Using Quantitative AVO Analysis in Fluid and Lithology

at each developmental stage (Doust and

The inability of the seismic data to properly Omatsola, 1990). There are three major
discriminate the fluid and lithology types in “Jay” lithostratigraphic units recognized in the Niger
Field informed quantitative AVO analysis, fluid Delta: Akata, Agbada and Benin Formations
substitution modeling and rock physics analysis. (Short and Stauble, 1967) (Fig. 2). The Akata
These methods will integrate well and seismic data Formation is a shale unit recognised as the major
to provide detailed information even beyond the source of oil and gas. The Agbada Formation
drilled region. consists of sands and shales units, while the Benin
Formation is composed mainly of sands. These
GEOLOGY OF THE STUDY AREA lithostratigraphic units form one of the largest
The study area 'Jay' Field is located in the shallow regressive deltas in the world with an area of some
offshore, South-Western Niger Delta, about 58 2 500,000 km (Kulke, 1995), a sediment volume of
kilometers (31 Nautical miles) off the coast of 3 about 500,000 km and a sediment thickness of
Warri Delta state, Nigeria (Figure 1). The area more than 10 km in the basin depocenter (Kaplan
covers about 29200 sq. feet (2.71 sq. kilometers). et al., 1994). The Niger Delta Province contains
The study area lies within the shelf area of the only one identified petroleum system. This system
Niger Delta. The Niger Delta is a prograding is referred to here as the Tertiary Niger Delta
depositional complex within the Cenozoic (Akata–Agbada) Petroleum System. The Tertiary
Formation of Southern Nigeria. It covers an area Niger Delta is a sedimentary structure formed as a
of about 75,000 square kilometers. It extends complex regressive off-lap sequence of clastic
from the Calabar Flank and the Abakaliki Trough sediments ranging in thickness from 9,000 –
in Eastern Nigeria to the Benin Flank in the west 12,000 m (Etu-Efeotor, 1998). Starting from
and it opens to the Atlantic Ocean in the southern different depocentres, the Niger Delta Basin has
territory. The delta protrudes into the Gulf of coalesced to form a single united system since
Guinea as an extension from the Benue Trough Miocene era. Due to the history or relative
and Anambra Basin Provinces (Evamy et al., unbroken progradation throughout the Tertiary
1978). The Niger Delta Basin is situated in the period, these three depositional lithofacies are
Gulf of Guinea and extends throughout the readily identified despite local facies variations, as
Niger Delta oil and gas province. From the three regional and diachronons formations
Eocene to the present (Fig. 2), the delta has ranging from Eocene to Recent age (Short and
prograded southwestward, resulting in depobelts Stauble, 1967).
that represent the most active portion of the delta

Figure 1: Location of 'Jay' Field, Offshore Niger Delta (Google Map)

 Adeoti et al.: On the Effectiveness of Using Quantitative AVO Analysis in Fluid and Lithology 003

Figure 2: Stratigraphic Column Showing Figure 3: Base Map showing the seismic coverage and well location in
Formations of the Niger Delta (Modified “Jay” field.
from Doust and Omatsola, 1990).
MATERIALS AND METHODS ('Sand e') were picked for the AVO analysis.
Seismic and well logs data used for this research
were provided by Chevron Nigeria Limited Synthetic Seismogram
through the Department of Petroleum Resources The synthetic seismogram was generated by
(DPR). The seismic data include: near, mid and far convolving the corrected sonic log and r log to get
stacked volumes. The base map of the study area is the impedance. The impedance was then used to
presented in Figure 3. The well logs comprise generate a reflectivity stick which was then
gamma ray, resistivity, P-wave sonic (Vp), density convolved with a wavelet to get the synthetics. A
(r), caliper, water saturation (Sw) and volume of wavelet extracted from seismic was used to
shale (Vsh). However, only one well was made generate the synthetics for this study.
available and used for this study, Jay 01 well was
drilled at Latitude (5.753941N) and Longitude Rock Physics Analysis
(4.684741E). The Rock Physics analysis was carried out using
cross-plots. The acoustic impedance (Zp) versus
Well Log Interpretation Poisson ratio (s) was plotted and checked against
The well data were loaded and quality checked four parameters water saturation (Sw), resistivity,
(QC) for any inconsistencies, editing and ϕ, and volume of shale (Vsh). The interpretation
normalization. Well data consisting of gamma ray of the expected gas effect that plot outside the
log, resistivity log, neutron log, density log and background trend was carried out according to
sonic logs were employed for identifying Bacon et al. (2003).
anomalous zones, lithology discrimination, fluid
content, porosity (ϕ), Sh and depositional
AVO Modeling and Seismic AVO
environment in “Jay” Field. New logs were
Attributes
generated using empirical and theoretical
The far, mid and near offset gathers were stacked
transforms. Well logs were compared with seismic
to check for AVO effects. The AVO modelling was
data and various cross plots of seismic attributes
done through the creation of synthetic
with log data were done. Two sand reservoirs:
seismogram from logs after fluid substitution.
hydrocarbon bearing ('Sand a') and water bearing
Since post-stack seismic was used, attributes such
004 Adeoti et al.: On the Effectiveness of Using Quantitative AVO Analysis in Fluid and Lithology

as far-minus-near, (far-minus-near) multiplied by AVO Inversion

near and (far-minus-near) multiplied by far Model-based and Sparse-spike acoustic
attributes were investigated; placed side by side for impedance (AI) inversion algorithms were applied
comparison for AVO effect. Cross-plot of near to the integrated well and seismic data to identify
against far-near, an approximation of intercept impedance contrasts, possible fluid saturated
versus gradient plot according to Castagna and zones and lateral extent of the reservoirs even
Swan 1997 was carried out. The cross-plot beyond the drilled zones in “Jay” Field. The
interpretation was done using Castagna et al., 1998 inversion was carried out using well logs (r and
AVO classification (Figures 4 and 5). sonic), interpreted horizons, extracted wavelet and
seismic volumes (near, mid and far). The QC of
the inversion was done using the cross-plot of the
well Zp against the inverted Zp

Figure 4: AVO Intercept (A) and Gradient (B) Cross- Figure 5: AVO Classification, Angle against Reflection
plot (Castagna and Swan, 1997). Coefficient (Castagna et al., 1998)

RESULTS AND DISCUSSION presence of gas. A shallower 'Sand e' having low
gamma ray and low resistivity response with the
Well Log corresponding neutron-density log tracking on
Figure 6 shows the well log interpretation with each other suggests water bearing sand reservoir.
delineation of the sand and shale units. Four These are sands (a and e) selected for the AVO
hydrocarbon bearing sands (a, b, c and d) were studies; 'Sand a' at 11,623 ft (3,536 m) TVD with ϕ
identified using a 70-API gamma ray cut-off in the (18%) and Sw (0.25) (sixth track), 'Sand e' at 5,925 ft
second track, regions where the gamma ray (1,806 m) TVD with ϕ (30%) and Sw (1).
reading falls below the cut-off point is classified as
sand (displayed in yellow) while greater than 70- Synthetic Seismogram
API is classified as shale (displayed in ash color). The wavelet generated from the well (elastic wave
In addition, gamma reading revealed blocky option) produces a synthetic seismogram with
serrated signature. High resistivity response on the correlation coefficient of 0.5 (Figure 7). The
fifth track is associated with the four sand units synthetic seismogram was used for well-to-seismic
while the neutron-density overlay shows a tie for the study area. The wavelet shows
negative separation for the four sands suggesting amplitude with peak of about 1.
Adeoti et al.: On the Effectiveness of Using Quantitative AVO Analysis in Fluid and Lithology 005

Figure 6: Identified hydrocarbon bearing Sands (a, b, c, d and e).

Figure 7: Synthetic Seismogram using Wavelet from “Jay 01”.

006 Adeoti et al.: On the Effectiveness of Using Quantitative AVO Analysis in Fluid and Lithology

Rock Physics Analysis resistivity) logs respectively.

Figure 8 shows the well correlation with the
calculated rock physics parameters. The rock Bacon et al. (2003) used as a model for the
physics result for the hydrocarbon bearing sands interpretation is inserted in Figure 9a. The analysis
show low Vp/Vs, low s and low μ-λ; as against the further revealed that these distinct plots coincide
background shale. Rock physics result for the with the depth range of the hydrocarbon bearing
water bearing sand show no change in the sands picked from log with good ϕ (14% - 28%),
parameters suggesting that an AVO response is Sw (< 0.5), VSh (0.1 - 0.3) and resistivity of over 100
unlikely. Cross-plot analyses as shown in Figures 9 Ωm. The water bearing sand falls within the
(a-d) revealed sands that are distinctively away background trend. The result shows the AVO
from the background shale and are brine saturated effect at the gas reservoirs.
lithology when colored by (Sw, ϕ, VSh and

Figure 8: Rock Physics parameters for Sands (a, b, c, d and e).

Figure 9a: Cross-plot of s against Zpcolor coded with S. w

Adeoti et al.: On the Effectiveness of Using Quantitative AVO Analysis in Fluid and Lithology 007

Figure 9b: Cross-plot of s against Zpcolor coded with ϕ.

Figure 9c: Cross-plot of s against Zpcolor coded with V Sh.

Figure 9d: Cross-plot of s against Zp color coded with Resistivity.

008 Adeoti et al.: On the Effectiveness of Using Quantitative AVO Analysis in Fluid and Lithology

AVO Analysis response for hydrocarbon bearing reservoirs. For

Gradient analyses of the seismic wavelet derived 'Sand a' (Fig. 10b), the base plots are along the
synthetic for the top and base of 'Sand a' and 'Sand background brine saturated trend while the top of
e' (Figs. 10 a and b) revealed AVO curve and the sand plots fall on the second quadrant which is
intercept-gradient plot. A positive amplitude (Fig. distinct from the background trend and it is
10a) that is neither rising nor falling is observed for therefore qualified as a class IV AVO based on the
top and base of 'Sand e', which is not a valid AVO Rutherford and Williams (1989) classification.

Figure 10a: Gradient Analysis for 'Sand a' according to Rutherford and Williams (1989).

Figure 10b: Gradient Analysis for 'Sand e' according to Rutherford and Williams (1989).
 Adeoti et al.: On the Effectiveness of Using Quantitative AVO Analysis in Fluid and Lithology 009

Seismic Attributes Cross-plots trend and therefore is classified as class IV AVO

Figures 11 (a and b) are the seismic attributes response (after Castagna et al., 1998; Adeoti et al.,
cross-plots which revealed that the 'Sand a' is class 2017). Figure 11b, the cross-plot of Gradient
IV AVO). The top of sands on the seismic volume against intercept for 'Sand e' does not show any
is the input for the cross-plot analysis. The plot significant deviation from background trend
towards the second quadrant as captured by the compared to Figure 11a.
ash box (Fig. 11a) deviates from the background

Figure 11a: Cross-plot of Gradient against Intercept for Figure 11b: Cross-plot of Gradient against Intercept for
'Sand a'. 'Sand e'.
Comparison of Seismic Attributes near) and (far-minus-near) multiplied by far are
Figure 12 shows the comparison between the good indicators of gas sands as revealed by their
original seismic (near, mid and far) and generated amplitude events. The seismic attributes also
seismic attributes (far-minus-near and (far-minus- differentiated gas bearing reservoir sands (green
near) *far). The amplitude event decreases from with orange events on seismic) from shale-sand
the near offset to the mid offset and much more (sharp blue with often yellow in-between)
for the far offset as in captured white circles 'Sand interfaces with bright amplitude.
a' region (Fig. 12). Seismic attributes (far-minus-

Figure 12: Comparison of the Seismic Attributes (far- near and (far- near*far) with Original Seismic (near, mid, far)
Volumes.
010 Adeoti et al.: On the Effectiveness of Using Quantitative AVO Analysis in Fluid and Lithology

AVO Inversion shale while the blue with bright amplitude

The model-based impedance inversion (Figure corresponds to fluid effect suggesting a
13a) serves as the input for the other inversions. hydrocarbon bearing sand. The insert of gamma
The model-based impedance inversion and resistivity logs confirm the inversion result for
corresponds to the interpreted horizons used as lithology discrimination with a near perfect
input for the regions Zp contrast. Linear Sparse correlation. The correlation in Figures 13 (c and d)
Spike impedance inversion is displayed in Figure of the original information from the well log with
13b which shows events of low impedance as seen the inverted data has error of 1196.61 for the Zp;
from the drilled reservoir region and continuing correlation of 0.97 for the synthetics and error of
beyond it, separated by faults. Impedance contrast 0.23, a good match between the horizons already
showing in different color codes corresponds to picked from seismic and well log.
different lithologies. The purple corresponds to

Figure 13a: Model-base Inverted Zp.

Figure 13b: Inverted Zwith

p inserted Gamma ray and resistivity logs.
 Adeoti et al.: On the Effectiveness of Using Quantitative AVO Analysis in Fluid and Lithology 011

Figure 13c: Cross-plot of well Zagainst

p Inverted Z p.

Figure 13d: Correlation of well Zand

p inverted Z p.

CONCLUSIONS negative reflection coefficient that decreases with

In this study, quantitative AVO analysis has been offset while 'Sand e' has positive reflection
carried out on “Jay” Field data for lithology and coefficient that neither increases or decreases with
fluid discrimination. Using gamma-ray cut-off of offset.
70, four hydrocarbon sands (a, b, c and d) and
water bearing 'Sand e' were identified. 'Sand a' at The seismic inversion results show 97%
11,623 ft depth and 'Sand e' at 5,925 ft depth were correlation coefficient between the inverted Zp
used for AVO analysis. The results from the Rock and well Zp which implies that seismic inversion is
physics cross-plots show that gas sands explicitly an effective tool in lithology and fluid prediction
plot away from the background shale and water in “Jay” Field offshore, Niger Delta. The study
saturated sands which indicates prospect in the has therefore shown that AVO could be a reliable
study area. tool for fluid and lithology discrimination in “Jay”
Field. However, shear velocity information should
Gradient analysis at the top and base of the two be acquired to increase the accuracy of the
selected sands (a and e) agreed with the procedure used in this study as against the
Rutherford and Williams (1989) classification empirically derived log used.
scheme for class IV AVO for 'Sand a' but no AVO
response for 'Sand e' which indicates that 'Sand a' ACKNOWLEDGEMENTS
is low impedance gas saturated Sands with The authors wish to thank the Department of
012 Adeoti et al.: On the Effectiveness of Using Quantitative AVO Analysis in Fluid and Lithology

Petroleum Resources and Chevron Nigeria Doust, H. and Omatsola, E. 1990. Niger Delta. In:
Limited for releasing the data for this study. Divergent and Passive Margin Basin
(Edwards P.A. and Santogrossi, P.A. eds.).
REFERENCES Tulsa, USA, 45:239-248.
Adekanle, A. and Enikanselu, P. A. 2013. Porosity E t u - E f e o t o r, J. O. 1 9 9 8 . S t r a t i g r a p hy,
Prediction from Seismic Inversion Sedimentolog y and Depositional
Properties over 'XLD' Field, Niger Delta, Environment of Reservoir
America Journal of Science and Sands of the Ivo Field, Niger Delta, Global
Industrial Research, 4(1), 31-35. Journal of Pure and Applied Sciences,
Adeoti, L., Adeleye, K.O., Itsemode, A. and Bello, 4(3): 64 –66.
M.A. 2014. Fluid prediction using AVO Evamy, B.D., Haremboure, J., Kamerling, P.,
analysis and forward modelling of deep Knaap, W.A., Molloy, F.A. and Rowlands,
reservoirs in Faith Field, Niger Delta, P.H. 1978. Hydrocarbon Habitat of
Nigeria, Arabian Journal of Geosciences, Ter tiar y Niger Delta, American
10:1-20. Association of Petroleum Geologists
Adeoti, L., Adesanya, O.Y, Oyedele, K. F, Bulletin, 62: 277- 298
Afinotan, P., Adekanle, A. 2017. Lithology Google Earth http://earth.google.com
and fluid prediction from simultaneous Kaplan, A., Lusser, C.V. and Norton, I.O. 1994.
seismic inversion over Sandfish field, Tectonic Map of the World. Panel 10
Niger Delta, Nigeria, Geosciences Journal, Tulsa. American Association of
10:1-15. Petroleum Geoscience Bulletin, 74:1-12
Bacon, M., Simm, R. and Redshaw, T. 2003. 3-D Kulke, H. 1995. Regional Petroleum Geology of
Seismic Interpretation, Cambridge the World. Part II: Africa, America,
University Press, Ltd, UK. 212. Australia and Antarctica: Berlin,
Castagna, J.P. 1993. Petrophysical Imaging using Gebrüder Borntraeger, 143-172.
AVO, The Leading Edge, 12(3): 172-178, Omudu, L. M. and Ebeniro, J. O. 2005. Cross-plot
342. of rock properties for fluid
Castagna, J.P. and Chopra, S. 2007. Introduction to discrimination, using well data in offshore
this special section - AVO, The Leading Niger Delta: Nigerian Journal of Physics,
Edge, 26(12): 1506-1507. 17:16-20.
Castagna, J.P. and Swan, H.W. 1997. Principles of Rutherford, S.R. and Williams, R.H. 1989.
AVO Cross-plotting, The Leading Edge, Amplitude-versus-offset Variations in
116(4): 337-342. Gas Sands, Geophysics, 54: 680-688
Castagna, J.P., Swan, H.W., and Foster, D. J. 1998. Schlumberg er Oilfield Glossar y 2014.
Framework for AVO gradient and www.glossary.oilfield.slb.com
intercept interpretation: Geophysics, Short, K.C. and Stauble, A. 1967. Outline Geology
63(3): 984-956. of Niger Delta, American Association of
Chiburis, E., Leaney, S., Skidmore, C., Franck, C. Petroleum Geologists, 51: 761 - 776.
and McHugo, S. 1993. Hydrocarbon Whiteman, A. 1982. Nigeria: Its Petroleum
Detection with AVO, Oilfield Review. Geology Resources and Potential,
5:1-3. Grantman and Trontman, London. 1: 394