Reservoir geomodelling is an effective tool in exploration, development or even during production for studying geological stages of any field lifecycle. Improvements in the algorithmic logic of subsurface parameters is the key to achieving greater accuracy in delineating potential prospects and/or fields.
Keeping this in mind, Dimension Strata has developed an advanced 3D geomodelling system in-house, using its own 3D geomodelling software, OrbStrata. The system was put into test and has successfully performed in several prominent oil/gas fields; the latest being a gas field in Brunei Darussalam, in which the advanced 3D geomodel achieved a staggering zoomed-in capability of down to 502m zone per grid cell without loss of resolution of the cell grids, thereby enabling identification of previously undetected/unexplored potential zones of interest.
Use of Facies Paleo-Trend Modelling
The study was deliberately performed with a very limited dataset to test with only a single 3D seismic depth cube and one geological report, and the geomodelling algorithm was accurately able to delineate probable structural features and prospective depositional geobodies within the field. Two-phase approaches were used for the study; construction of a structure having lithological characteristics, followed by Facies Paleo-Trend Modelling (FPTM) of the modelled geobodies to simulate the interpreted layers within the 3D model.
In the context of this study, a novel stochastic method was used along with a proprietary algorithm that provided a considerably improved structural framework as well as the 3D facies framework of the field.
Overview of the study
The study combined volume-based modelling (VBM) with Dimension Strata’s 3D seismic interpretation and reservoir geomodelling system, in which amplitude anomaly-based seismic interpretation is used and amplitude anomaly factor (ANF) is employed in determining different amplitude classes for specific horizons and/or faults interpretation. Proportional horizons were used for iso-stratigraphic property during automatic calculation for stacking interpreting 3D horizons.
Grid construction, structural and facies paleo-trend gridding
Initially, the potential reservoir zones were identified using the integrated seismic analysis (ISA) within the horizon modelling phase. This incorporation ultimately placed the zone boundaries of all modelled layers of the study field.
Although the data provided for the study was limited only to a seismic cube, incorporation of present-day deposition pattern within the block and previously studied depositional settings were used as inputs for the FPTM trend to be developed. This study was completely based on geometrical inputs that control particular body shapes (thickness/width, etc.). The facies modelling workflow for the studied reservoir followed three robust steps based on the abovementioned technical steps.
During the initial stage, depositional-bodies were generated using FPTM algorithm; then the boundaries of the interpreted geobodies were defined; finally, the internal geometry and heterogeneity of the facies were developed to generate the final FPTM of the study reservoirs.
Three horizons were interpreted for the study using OrbStrata’s advanced geomodelling algorithm. An empirical geodatabase incorporating seismic amplitude anomaly along with amplitude normalised factor (ANF) was used to determine the target horizons for the study. As the study was based exclusively on seismic data (in depth), the determination of ANF helped delineating probable lithofacies classes within the studied field. For the study, amplitude class was developed based on the zone of interest and exclusively implicates the classification within the zone boundaries.
Structurally the reservoir zone interpreted in between the top and bottom reservoir surfaces is devoid of any faults. It is a classical four-way dip closure structure with an average orientation of the structure being N25E and S220W. The structure is thick at the central part of the structural model and thin on the sides of the structure. The average azimuth of the structure is N40°E.
Facies paleo-depositional modelling
The subsurface structure, its horizons and the seismic facies interpreted all lead in the same direction in re-evaluating the paleo-depositional settings of the field. From the seismic facies, it is observed that the facies probability model (P50 to P90 sands), from bottom to the top horizon indicates two major sediment supply directions (one from the Northwest and the other from Northeast, as evident from the facies geomodel) in the modelled grid.
Analysing the seismic facies carefully, it was concluded that the facies tend to get dirtier (mixed sand to shale in nature) from the bottom layers gradually towards the top layers of the model. This also signifies that the facies altogether might have retrograded (transgression) to the younger successions (within the studied interval). The facies geomodel also depicts a probable paleoshoreline shift parallel to the initial NNE-SSW orientation down further South towards the present-day coastline of Brunei.
The algorithm used a combination of sequential indicator simulation (SIS) for flexibility in defining different variables within the model, truncated Gaussian simulation (TGS without any preferential trending) along with DSSB’s facies paleo-trend modelling (FPTM) algorithm to develop 3D seismic-based facies putting more weight of the seismic amplitude anomaly for determining logical facies distribution spatially. It is also to be noted that the study used all available modelling parameters only with the help of seismic data (in depth domain). Taking into consideration the current limitations set for the study, the output geomodel(s) that it has produced are of considerable accuracy, as are the subsurface settings within the target zone of the study field.
Determining three-dimensional geometries of the interpreted facies was a key component of this study. As the facies and facies classes were directly associated with the quality of the prospective reservoir zone, facies geometries were carefully interpreted.
According to the facies associations and literature review from previous studies performed on this area, the study concluded that the prevalent depositional setting of the area is offshore deep-sea fan sedimentation. It was also evident from the seismic facies modelled within the facies model. From the topmost surface of the target reservoir zone, it was evident (considering the amount of data provided for the study) that different sand facies distributed in a fan shape along the Southwest and Southeast portion of the model.
The produced model could explicitly describe the crucial changes in geometry of the facies, which potentially is of great importance for future drilling campaigns within the modelled zone. Fan-shaped deposits were present on the Layer-5, especially on the Southwest and Central Northeast part of the model.
The facies model has been tested with the available gas-water contact (GWC) of studied zone, hence it paves the way for higher probability of development for particular facies proportions above the GWC level (-3,050m SSTVD).
Based on the execution of all the major interpretational modules of DSSB’s algorithm, depositional facies are genetically cleaner and are above the GWC. The spatial distribution of the potential seismically derived sand facies is interpreted. The overall interpretation of the different probabilities of sand occurrence within the model states that P50 (probability of occurrence) and P60 sands have a higher percentage of occurrence within the middle layers (k=7 to k=13) whereas the probability of occurrence of P70 and P80 sands is from k=1 to k=13 layers of the model. P90 sands are assumed to be found within k=9 to k-19 layers within the model.
This advanced geomodelling technique equipped with the facies paleo-trend modelling is tested on several prominent hydrocarbon fields/prospects in Asia Pacific, showing immense potential in resolving complex depositional facies interpretations with substantially improved outcomes.
Dr AKM Eahsanul Haque is technical director at Dimension Strata
