Datapages, Inc.Print this page

Density Analysis to Predict Carbonate Reservoirs in Volcanic Clastic Environment, Implications for Field Development

Ary Wahyu Wibowo1; Astri Pujianto1; Wisnu Hindadari1; Rusalida Raguwanti2; Yuda Faisal Yushendri2 (1) Exploration, PT.Pertamina EP, Jakarta, Indonesia (2) Geophysics, Upstream Technology Center (UTC), Pertamina (Persero), Jakarta, Indonesia.

INTRODUCTION
The Field located in the Ciputat Sub-basin, a north-west half graben within North West Java Basin. Syn-Rift in this sub-basin divided into two phases. Syn-Rift Phase-1 began with the deposition of the Volcanic Clastic and Carbonate sediment in several location. The sediments have a role as basement for the formation above. In the Syn-Rift Phase-2, fluvio-deltaic clastic sediment deposited unconformable above Syn-Rift Phase-1 sediment. Carbonate sediment of Syn-Rift Phase-1 are primary reservoir in the Field and is known as Carbonate Basement.
Carbonate Basement is the build up that has limited distribution. The northern part of Carbonate overlapping into Igneous Rock and in the other part onlapping with the Volcanic Rock. Drilling data indicate the hydrocarbon discovered in the Carbonate controlled by matrix porosity in the upper part and fracture in the lower part. Other basement reservoir, Igneous and Volcanic Rock does not produce hydrocarbon although there are indications during drilling.
The boundary between Carbonate Basement with Igneous or Volcanic Rock can be detected by applying cross plot refers to rock sensitivity. Cross plot between Density vs P-Impedance indicate the Igneous Rock have distinct density value than  Carbonate Basement, while the Volcanic Rock have similar density value with Carbonate Basement. P-Impedance (AI) analysis is generated to discriminate Volcanic Rock with Carbonate Basement in the eastern part of the Field. Density attribute is applied to distinguish Carbonate Basement with Igneous Rock.
An understanding of Carbonate Basement facies and distribution will assist in confirming hydrocarbon reserves and development of the Field, in order to determine infill or step-out well, and IOR-EOR strategy.


TECTONIC AND SEDIMENTARY SETTING
Regionally, North West Java Basin is part of the southern edge of the Sunda or Sundaland Microplate. Noble (1997) describes the configuration of the basement in the North West Java Basin which have formed a series of Horst and Half Graben with the north-south direction (Figure 1). The half graben, such as Ciputat, Pasibungur, and Cipunegara formed in the early phase of rifting at the Eocene age (Sribudiyani, et al, 2003).

 

 

Figure 1: Basement configuration map shows the Highs and Lows in the North West Java Basin (after Ron Noble, Anditya, 2000)

 

 

 

 

Generally, North West Java Basin began by rifting in the Middle to Late Eocene, formed half graben. Graben sedimentation intiated by Lacustrine-Fluvio sediment, including source rock deposition. Transgression occurs in Middle Oligocene to Middle Miocene and the fluvio-deltaic sediments of Syn-Rift were deposited. Carbonate sediment were deposited in Post-Rift phase. The sealing rocks deposited on it during maximum transgression. Late Miocene until Pliocene period is the formation of  the compression structure (Satyana, 2005). Sediments in the North West Java Basin comes from the north and east of the Sunda Microplate (Atkinson et al, 1993). The Stratigraphy of North West Java Basin is almost similar to the South Sumatra Basin (Schlumberger, 1986, Figure 2).
 
Figure 2: Regional Stratigraphy of West Indonesian Basin. Schlumberger (1986)

 

In Ciputat Sub-Basin, Syn-Rift is divided into two phases. Syn-Rift Phase-1 have begun with the deposition of the volcanic rock that equivalent to the Jatibarang Formation and carbonate sediment deposited on it in several location. At the end of this phase, there are erosional surface due to hiatus before the Syn-Rift Phase-2 sediment deposited. The Syn-Rift Phase-1 sediment have a role as basement for the formation above.
Syn-Rift Phase-2 was began in transgression period, the fluvio-deltaic sediment of Talangakar Formation (TAF) is deposited unconformable on the Carbonate and Volcanic sediment of Syn-Rift Phase-1, followed by deposition of the Upper Talangakar sediment. Carbonate of the Baturaja Formation were deposited conformable on it as Post-Rift sediment and covered by shale of the Cibulakan Formation at maximum transgression.
Carbonate sediment of the Syn-Rift Phase-1 are primary reservoir and is known as Carbonate Basement. Furthermore, along with the compressional structure, the Carbonate of the Parigi Formation is deposited alternating with sand-shale of the Cisubuh formation.
The evolution of the Sub-basin can be illustrated as shown in Figure 3.

Figure 3: Illustration of Pre-Rift to Post-Rift Evolution in Ciputat Sub-basin

 

STRATIGRAPHY
Most of the well in the basin of the Ciputat Sub-Basin have penetrated sediment under Talang Akar Formation (TAF), which is composed by Carbonate, Volcanic, and Igneous Rock. They have a role as Basement for the deposition of sediment on it.
Stratigraphically, the three rocks under Talang Akar Formation can be distinguished into Pre-Rift sediment, composed of igneous rock (#R-01 well) and Syn-Rift Phase-1 sediment, composed of volcanic rock (#X-6 and #X-7 well) and carbonate (#X-1 and #X-5 well). Syn-Rift Phase-2 sediment consisting of the Fluvio-Deltaic Sediment of the Talang Akar Formation is deposited unconformable on the Syn-Rift Phase-1, followed by the deposition of Post-Rift sediment on it.

 

 

 

Figure 4: West-East Seismic Cross Setion, The Carbonate have limited distribution and onlapping into Volcanic Rock.

 

 

 

Pre-Rift
According to well data that have penetrated the Pre-Rift sediment in Ciputat Sub-basin, the Pre-Rift sediment composed of granite, monzonite, diorite and schist.
Syn-Rift Phase-1

  • Carbonate

Syn-rift Phase-1 sediment composed of volcanic rocks, shales and carbonate. Carbonate encountered in #X-1 and #X-5 well. Petrographic analysis has been carried out on some sample of cutting, SWC, and cores.
Carbonates are generally crystalline or crystallization, but sometimes it still shows the initial texture
In some places, the carbonate have been deformed and generate fractures, although a few of core sample indicates a healing of fractures, however it has proven to be the primary reservoir in the Field of Ciputat Sub-basin.

  • Shale

Dark gray to black and hard. These rocks have been strongly deformed and formed fractures that filled with calcite mineral.

  • Volcanic Rock

Volcanic Rock as basement encountered in #X-6 and #X-7 well. Based on petrographic analysis, Volcanic composed of pyroclastic have been altered into clays mineral. Fractures that occur in these rocks have been filled by calcite mineral.
Syn-Rift Phase-2

  • Talangakar Formation (TAF).

TAF consists of sandstone, shale and coal. Sandstone is primary reservoir and proven as hydrocarbon reservoir. Based on petrographic analysis, sandstones can be classified as quartz wacke, and subarkose sublitharenite that deposited in fluvio-deltaic environment.

  • Upper Talangakar Formation

Composed by sandstone, shale, limestone, and coal. In this formation, hydrocarbon discovered in lower zone. While at the top of this formation is not found hydrocarbon prospects.
Post-Rift
Post-rift phase began with the deposition of Limestone of Baturaja Formation and covered by Upper Cibulakan Formation, composed of shale, limestone and siltstone.
GEOPHYSICAL ANALYSIS AND INTERPRETATION
Geophysical methods that applied in this study include the re-interpretation of 3D seismic data, analysis and modeling of rock physics, seismic inversion modeling, and multi-attribute analysis.
Re-interpretation of 3D seismic data have to be built after the re-processing of seismic with azimutal stack method produces a different result and better than the previous processing. Analysis and rock physic modeling is required to determine the parameters to distinguish Carbonate and other Basement (Igneous and Volcanic Rock). Based on the sensitivity analysis of the rock physic parameters versus lithology and fluid type can be extracted the best rock physic parameters. By applying seismic inversion modeling and seismic multi-attribute analysis, some seismic parameters will be extracted and then be applied to assist in building a geological model of the reservoir.
The data used for seismic re-interpretation are 3D seismic data that acquired in 2007, and then re-processed using the azimuthal method by UTC in 2010. Sesimic interpretation indicates  that the Carbonate onlapping into Volcanic Rock, it is contributed as stratigraphic trap. 
Rock physics analysis is conducted by applying seismic data and well data. First stage of rock physic analysis is to generate a cross-plot of some attributes in certain geological intervals, seismic resolution analysis by calculating tuning thickness, log prediction, and determine parameters of seismic inversion.
Cross-plot analysis was conducted to the seismic parameters without S-wave analysis due to angle of the PSTM seismic gather insufficient for AVO analysis or simultaneous inversion (gather angle only 200), so the cross-plot analysis are preferred to discriminate reservoir lithology with the surrounding lithology. Attributes that used as input in the cross-plot analysis are Density, Gamma Ray (GR), P-wave and P-impedance (AI). In Figure 5 can be seen there is no obvious data distribution pattern for lithology. That is, the rock parameter between reservoir with the surrounding lithology is almost similar.
Attributes are extracted from Gamma Ray (GR) cannot discriminate lithology, but Density attribute has the opportunity to be able to distinguish the lithology by cross-plot analysis due to has different regression trend. By applying P-impedance of seismic inversion, Carbonate in several well is difficult to distinguish with the Igneous Rock (Figure 6).

 

 

Figure 5: Cross-plot of some attributes to determine the sensitivity of each attribute toward the lithology.

 

 

 

 

Figure 6: Character of R-1 (Igneous Rock) and X-5 (Carbonate) which shows the similarity value of P-impedance, so that the AI seismic inversion is difficult to distinguish them, but from the density log value ​​provide a significant difference.

 

 

 

In Figure 5, the Density attribute has the opportunity to discriminate Carbonate against Igneous Rock. Cross-plot between P-impedance with Density have been built to know the sensitivity of Density attribute between Carbonate and other Basement rock. Obviously, Density attribute can be applied to distinguish Carbonate with the Igneous Rock as shown in Figure 7, it is recommended to generate the Density volume combined with AI Inversion volume to map the Carbonate distribution.
Inversion modeling conducted to determine the properties and lateral porosity distribution of rock by applying an inverse modeling integrated with Rock Physics analysis, Swetness Attribute, and Multi Attribute Analysis (Neural Network).

 

Figure 7: Lithologic trend for Carbonate and Igneous Rocks in AI and Density cross-plot of all wells showed sensitivity of density attribute to distinguish lithology.

 

 

In this study, Inversion Modeling conducted in Crystalline Carbonate that similar with Basement. Seismic data used in the Inversion Modeling is the Pre-Stack Time Migration (PSTM) of 3D Seismic that processed by Azimuthal Processing method. Inversion Modeling have been generated by using 10 well data that have Density, Sonic, GR (optional), and checkshot/VSP data. Constraint Sparse Spike Inversion (CSSI) is one of the selected seismic inversion methods to acquire absolute acoustic impedance from seismic data using the low-frequency component that extracted from well data.
Inversion modeling as advanced data processing has been generated by employing the high lateral resolution seismic data and well data with good vertical resolution. The integration of the two will widen the seismic bandwidth. Inversion Modeling will turn the interface property obtained from seismic amplitude into layer property as final result. Layer property will assist seismic interpretation and to improve the output. Acoustic impedance model is generated by tying well data with seismic data.
Acoustic Impedance quality can be controlled through high value of signal to noise ratio which can be achieved 12 db and cross correlation value close to 1 in most area. Quality control of Inversion Modeling can be done by compairing acoustic impedance trace from seismic with acoustic impedance from log data. Overlay between both of them must be matcing.
Carbonate property that extracted from Inversion Modeling show good lithological differences, although it still needs to be integrated with multi-attribute analysis to enhance the level of confidence. Thin section of cutting data is very helpful in the lithology description.
Stratigrphic slicer mapping is applied to  gain Amplitude Impedance Map with detail interval until 2 ms in objective level. This will produce a number of micro-layer of the certain top horizon.
Anomali of Acoustic Impedance value within Carbonate shows different reservoir characters from several well. Thin section analysis from cutting data indicates variation of lithology character in Carbonate.
Cross-plot from several log properties as the Rock physics analysis output showed differences in lithological character that can be observed.
Multi-attribute seismic analysis is a method used to select the seismic attribute that has a good physical relationship (high weight) with rock physic parameters of reservoir to be predicted. Input of seismic attributes can be either internal attributes (such as amplitude, frequency, seismic phase, etc.) or external attributes derived from inversion modeling (example: pseudo density model extracted from the model based inversion). According to the rock physic analysis, density is an excellent property to discriminate lithology of reservoir.

 

Figure 8: Acoustic Impedance Cross Section (top) and Pseudo Density Cross Section (bottom).

 

 

 

 

 

 

Figure 9: Comparision between Acoustic Impedance (AI) Cross Section (top) and Pseudo Density Cross Section (bottom). In AI Cross Section, the anomaly between Carbonate (Well #X-5) and Igneous Rock (Well #R-1) is not significantly distinguished, but in Pseudo Density the reservoir character can be distinguished.

 

 

 

 

Figure 8 shows the cross section of the acoustic impedance and pseudo-density of 3D volume that extracted from inversion modeling. Furthermore, both of the 3D volume data will be applied as input data for external seismic attributes to predict the property volume in multi-attribute seismic analysis.
There are two methods that be used to generate the property volume: (1) Multi-Step-Wise Regression and (2) Neural Network (Neural Network method that used to predict the density volume is Probabilistic Neural Network/PNN). To predict the density volume, various seismic attributes, pseudo density and density log data will be used as input data. Prediction of density volume by this method is expected to increase the resolution and validation in each well.
Prediction of density distribution is generated by applying multi-attribute seismic method.  The process of multi-attribute analysis consists of two main phases: Training data and applying these attributes into seismic data.
To determine the attributes that will be applied in the prediction, the density log will be trained to some seismic attributes.  Prediction error value will decrease in line with the number of attributes used (more attributes are used, the prediction error value will be smaller).
Prediction of Density Volume using Probabilistic Neural Network method showing better results seen from the match between Predicted Density and Density Log with cross-correlation 0.9408, while prediction using Step-Wise Multiple Regression method has cross-correlation at 0.8067.
The output of the Probabilistic Neural Network indicates carbonate can be distinguished with igneous rocks, but it is difficult to distinguish from volcanic rock due to the density of the volcanic rock and carbonate are not much distinct (Figure 9 and 10)

 

Figure 10: Character shown by high Acoustic Impedance value in Well #R-1, corresponding with igneous rock minerals in thin section. Whereas Well #X-5 with lower acoustic impedance value ​​are carbonate with fossiliferous composition in thin section.

 

 

 

CARBONATE FACIES AND DISTRIBUTION
In Ciputat Sub-basin, Carbonate Basement is the build up that has limited distribution. The northern part of Carbonate onlapping to the Igneous Rock (#R-1 well) and in the other part is overlapping with the Volcanic Rock (#X-6 and #X-7 well).
The West-East correlation of seismic section that have been tied to well shows the Carbonate Basement facies changing into Volcanic Rock. The boundary of Carbonate is the onlap surface with Volcanic Rock (Figure 4).
Carbonate Basement distribution was initially considered monoclinic towards North, after development and delineation drilling can be predicted the Carbonate is build up with East-West distribution. In northern, the Carbonate Basement boundary can be determined from well data that show the distinction in Density Attribute with Igneous and Volcanic rock. Eastern and southern boundary indicated by differences in lithology and acoustic impedance attributes. (Figure 11).

 

Figure 11:  Pseudo Density Map (bottom) using Neural Network method provide a obvious picture to distinguish lithologic character in the Well #R-1 and Well #X-5 which is not visible in AI Map (note the circle).

 

 

 

 

CONCLUSIONS
Carbonate Basement in Ciputat Sub-Basin is the build up carbonate that has limited distribution. The northern part of Carbonate onlapping to the Igneous Rock and in the other part is overlapping with the Volcanic Rock. Drilling data indicates the hydrocarbon discovered in the Carbonate controlled by matrix porosity in the upper part and fractures in the lower part. Other basement reservoir, such as: igneous rock and volcanic rock does not produce hydrocarbon although there are indications during drilling.
Based on the cross-plot analysis, density attribute can be applied to discriminate Carbonate Basement lithology with other Basement, especially Igneous Rock in #R-1 well. Mapping of Carbonate lithology is recommended by using a combination of AI Inversion volume and Density Volume to distinguish Carbonate from Igneous and Volcanic Rock.
The boundary between Carbonate Basement with Igneous or Volcanic Rock can be detected by applying cross plot refers to rock sensitivity. Cross plot between Density vs P-Impedance indicate the Igneous Rock in #R-1 well have distinct density value than  Carbonate Basement in #X-5 well, while the Volcanic Rock have similar density value with Carbonate Basement.
P-Impedance (AI) analysis is generated to discriminate Volcanic Rock with Carbonate Basement in the eastern part, whereas the Density attribute is applied to distinguish Carbonate Basement with Igneous Rock.

REFERENCES
Chilingarian G.V, Mazzullo S.J, Rieke H.H., 1996., Carbonate Reservoir Characterization: a Geologic–Engineering Analysis., Part II., Elsevier.
Lucia FJ., 1995., Rock Fabric/Petrophysical Classification of Carbonate Pore Space for Reservoir Characterization., AAPG Bull 79., 9: 1275-1300.
Lucia FJ., 1983., Petrophysical Parameters Estimated from Visual Descriptions of Carbonate Rocks: a Field Classification of Carbonate Pore Space., J Pet Technol., March 1983:629-637.
Lucia FJ., 2007: Carbonate Reservoir Characterization: An Integrated Approach., Second Edition., Springer.
Mavko G, Mukerji T, Dvorkin J., 2009., The Rock Physics Handbook: Tools for Seismic Analysis of Porous Media., Second Edition., Cambridge University Press.
Mike Carlson., 2003., Practical Reservoir Simulation: Using, Assessing, and Developing Results., PennWell.
Pertamina BPPKA., 1996., Petroleum Geology of Indonesia Basins: Principles, Methods and Application, Volume III West Java Sea Basins., Pertamina BPPKA, Jakarta.
Pertamina and ITB., 2002., Regional Tectonic Study of Java and Hydrocarbon Potential of Central Java and West Java Basins., Pertamina and ITB.
Satyana, A.H., 2005., Oligo-Miocene Carbonate of Java, Indonesia:Tectono-Volcanic Setting and Petroleum Implications., Proceedings Indonesian Petroleum Association, the 30th Annual Convention.


Tucker, M.E, Wright, V.P., 1989., Carbonate Sedimentology., Blackwell Scientific Publications., Oxford.
Wayne M. AHR., 2008., Geology of Carbonate Reservoir: The Identification, Description, and Characterization of Hydrocarbon Reservoir in Carbonate Rocks., A John Wiley & Sons., Inc.

 

AAPG Search and Discovery Article #90166©2013 AAPG International Conference & Exhibition, Cartagena, Colombia, 8-11 September 2013