New Life for Old Data: Brownfield Reservoir Characterization and 3D
Geological Modeling from the West Baram Delta Province, Offshore
Sarawak, Malaysia*
By
1 1 2 1
Matthew Mulcahy , Tanwi Basu , A. Wahid Musbah , Gavin Douglas , Howard D.
2 2 2
Johnson , Kamarolzaman B. Yahya , and Mustafa Suleiman
Search and Discovery Article #20023 (2004)
1
Schlumberger DCS, Kuala Lumpur, Malaysia (tbasu@kuala-
2
lumpur.oilfield.slb.com) Petronas Carigali Sdn Bhd, Kuala Lumpur,
Malaysia (kamary@petronas.com.my)
Abstract
Reservoir characterization studies of the multiple, vertically-stacked Miocene
reservoirs in the Bokor and Betty fields (West Baram Delta, offshore Sarawak [Figure
1]) have created facies based 3D geological models to appraise remaining reserves and
determine strategies for economic redevelopment. This paper describes the reservoir
modeling workflow (Figure 2) and highlights lessons learned for similar
redevelopment projects in the future.
Since the discovery of these fields in the 1970’s, extensive subsurface data, including
core, well log, dipmeter, seismic and production data have been used to evaluate these
Late Miocene coastal/deltaic reservoirs. However, previously these data had only been
evaluated in traditional, 2D form and had never been subject to a modern 3D geological
modeling analysis. To do this, and to supplement the existing data, further in-depth
studies were performed, including (A) petrography (SEM and XRD analyses) to
determine framework grain and clay mineralogy and their distribution (pore-filling vs.
pore bridging as shown in Figure 3, (B) rock fabric/texture analysis from high-
resolution dipmeter data to assess fine-scale heterogeneities (Figure 4) and
(C) facies analysis of core and well logs to generate electrofacies models (Figure 5).
General Statement
Although core data were extremely limited in Betty and Bokor fields, they were
recognized to be critical to the reservoir modeling objectives. Consequently, a key
approach was the development of electrofacies models that integrated both the core and
wireline log data. In addition, it was necessary to review the conceptual and analog
depositional models in order to provide the optimum framework for understanding
facies and sand body distribution. This, together with analysis of well log patterns and
seismic amplitudes, enabled propagation of a genetically based, electrofacies
interpretation to each of the hundreds of reservoirs within every well within the two
fields (Figure 6). This was the precursor to distributing facies and rock properties data
in 3D geocellular models for each field; this has significantly improved reservoir
understanding and enhanced redevelopment decisions. In particular, the 3D geological
models provided the basis for history-matched full-field simulation models, which have
identified new and timely infill well targets and recompletion opportunities (Figure 7).
Finally, these studies have also provided additional insight into the depositional setting
and stratigraphic architecture of these multiple stacked (100s-1000s ft thick) wave- and
tide-influenced coastal/deltaic reservoirs. Additional details of the latter are summarized
below.
Figure 1. Location map showing location of Betty and Bokor fields, in the Baram Delta province,
offshore Sarawak, Malaysia. Area of more than 4 km sedimentary section is outlined. (After Ngah,
1999).
Figure 2. Workflow recommended for brownfield geomodeling, illustrating new life for old data.
Figure 3. Petrography through SEM and spot elemental analysis of Betty sandstones showing
angular pores within frame-working quartz grains (left) and pore-filling chlorites (right). XRD
analysis independently performed on the shales also resulted a clay percentage dominated by
chlorite and minor illite.
Figure 4. Rock Fabric Analysis via BorTex using the dipmeter logs showing least active
microresistive curves for massive, poorly stratified sandstone (left) in Betty-5. Highly active curves
are observed in clay-clast bearing storm-dominated-event beds (top-right) and intensely
bioturbated sandstones, both of which show high conductive heterogeneity as calculated from the
analysis (red shading on the heterogeneity track). This analysis captured the fine scale depositional
heterogeneity as conductive and resistive anomalies.
Figure 5. Electrofacies Analysis using neural network techniques via RockCell. The diagram
illustrates the raw input log data on the far left that is explicitly tagged from core description for
neural network training. In the middle are the facies probability and final facies estimation, which
can then be checked against original core description on the right.
Figure 6. Example of genetically-based electrofacies distribution for one reservoir unit in Bokor
Field. Seismic amplitudes and reservoir engineering data, together with core results, combine in
construction of the 3D geological model.
Figure 7. Hydrocarbon pore volume thickness map (HCPVo) derived from simulation
results on 3D geological model allow identification of infill targets and recompletion
opportunities
Betty Field
The Upper Cycle V (Upper Miocene) reservoirs in the Betty Field reflect repeated
progradation and retrogradation of the north-westward prograding West Baram Delta.
Generation of accommodation space was strongly influenced by episodic movement of
the bounding growth fault to the south. Facies and rock property characteristics show
that the reservoirs were highly wave reworked and redistributed alongshore to form
laterally continuous shoreface sand sheets with associated transitional to offshore
inner-neritic shelfal lithofacies. The reservoirs are highly heterogeneous vertically
(Figures 7 and 8), due to the frequent intercalation of inner-neritic shales poorer quality
distal lower shoreface and transition zone lithofacies, particularly in the lower part of
each coarsening upward parasequence.
Bokor Field
The Upper Cycle V/Lower Cycle VI (Upper Miocene) succession in the Bokor Field
displays extreme vertical stacking of approximately 130 separate reservoirs over a
6000-ft-thick hydrocarbon-bearing interval. The reservoirs (Figure 9) were deposited in
a more variable coastal/deltaic environment, which includes distributary channels,
mouth bars, tidal channels/estuaries, tidal flats and coastal barrier/shoreface sand
bodies. The Bokor reservoirs were deposited in a more axial deltaic setting, when the
coastal and inshore areas were influenced by both tidal and wave processes, similar to
the modern Niger Delta. The greater abundance of channel sand bodies results in a
higher degree of lateral heterogeneity, as is evident from seismic amplitude displays.
The latter confirm the gross paleogeographic setting of the Bokor Field and provide
independent data on channel size, shape and orientation.
Figure 8. A cross-section panel from the SSW (left) to ENE (right) going through the depositional
strike of L3 reservoirs in Betty field showing lateral continuity of most of the best reservoir quality
sandstone (in yellow) and inner-neritic shales (in gray). Vertical heterogeneity is also a function of
the distribution of poor quality sands (in orange). The tracks for each of the wellbores shows the
volume clay and the lithofacies distribution as estimated from RockCell neural network technique
in the left and the right, respectively.
Figure 9. A North (left) to South (right) cross-section panel from Bokor field,
illustrating the contrast between laterally heterogeneous lower coastal plain deposits (P)
and laterally continuous shoreline/shoreface (C) and shoreface/shelf (S) deposits.
Vertical heterogeneity is prevalent throughout, resulting in numerous stacked reservoirs.
Tracks for each wellbore show the volume of clay infilled, with electrofacies on the left
and porosity and permeability on the right.
Conclusions
These studies demonstrate how modern techniques and technologies can integrate
highly variable vintages of subsurface data into robust reservoir models capable of
improved reservoir understanding and more confident prediction of bypassed oil
location. They also highlight the additional value that can be gained from existing and
sometimes apparently neglected or underutilized data. Finally, we would like to
emphasize how modern technologies can be used to delivery timely results in these and
other brownfield reservoir characterization studies. The approach described here could
be applied to many other redevelopment projects.
Reference
Ngah, Khalid, 1999, Malaysia’s gas resources: Search and Discovery article
#10002
