Ferreira, Joana (University of Lisbon, Faculty of Sciences, Department of Geology) | Azerêdo, Ana C. (University of Lisbon, Faculty of Sciences, Department of Geology and Instituto Dom Luiz) | Bizarro, Paulo (Partex Oil & Gas) | Ribeiro, Maria Teresa (Partex Oil & Gas) | Sousa, Ana (Partex Oil & Gas)
Characterizing highly heterogeneous carbonate reservoirs requires the integration and detailed analysis of petrophysics, facies, diagenesis, geometry, depositional environments and lateral and vertical variability. This is often challenging to conceptual models at oil-field scale, as this thorough analysis is hard to fully reproduce at reservoir-scale models. In order to improve skills and interpretations on both approaches, we addressed a case study from a Middle Jurassic outcrop of Portugal as an analogue for a carbonate reservoir.
The outcrop exhibits three barrier shoreface lithofacies: L1 - oolitic and bio-intraclastic grainstones (divided into: L1a- with planar stratification or unstructured; L1b - with diverse cross-stratification styles); L2 - coarser grained grainstones/rudstones; and L3- coral/algal biostromes. Outcrop analysis was combined with petrographic/diagenetic studies of rock samples. Regarding petrophysical properties, three methods were used to determine the porosity: thin-section impregnation with blue-dyed epoxy resin, rock-slab water saturation and, for a few samples, plug measurements in a Helium gas expansion porosimeter. The plugs were also used to acquire permeability values using a digital gas permeameter.
The results show that the outcrop is a tight reservoir, since most levels have low porosity (~3.5% average) and permeability (mostly <0.1md), though higher values (φ~10-15% and K~160md) occur locally. Most levels are classified as hybrid 1 or diagenetic reservoirs, according to
Normal industry workflows often do not fully consider geological data and conceptual models, and instead rely heavily on geostatistical propagation of well data. The results obtained indicate that there is an improvement in reservoir understanding with an integrated reservoir characterization and modelling process that accounts for actual depositional and diagenetic trends, as well as the distribution of the sedimentary bodies.
Carbonate reservoir characterization is often a complex task, due to the interplay between primary processes (e.g. depositional environments, facies changes) and secondary processes (e.g. burial, diagenesis, faulting and fracturing, cementation). In order to properly characterize and model such a reservoir, it is paramount to unravel the order by which such processes have affected the rock, leading to the present day petrophysical properties.
In the presented case study (onshore dolomitized carbonate reservoir in Central Asia), a multi-step approach was taken for its characterization and modelling. The characterization phase was focused in understanding the key processes and controls on porosity and permeability. From the core and log data, a detailed sedimentologic and diagenetic study was performed, to identify the depositional environments and facies, as well as the pore system geometry, and its impact on fluid flow. Furthermore, several trends on reservoir quality were identified, related to faults, and associated with depositional cyclicity.
From the above work, a reservoir model was built, to support field development planning and associated uncertainties. A structural and stratigraphic framework was built, and Flow Unit Types (FUT) were defined using seismic, cores, thin sections, logs and mercury injection capillary pressure data (MICP). Property modelling was carried out for porosity and permeability, honouring FUT, depositional and diagenetic trends. In particular, two trends were modelled: a fault-related trend, to introduce the impact of diagenetic leaching related to faults (observed in core data); and a cyclicity related trend, to introduce the impact of preferential fluid flow pathways that occur at or near cycle tops. The uncertainty in the reservoir property models was evaluated with different FUT, driven by depositional and diagenetic concepts.
The results indicate that a significant improvement in reservoir understanding can be achieved with the use of an integrated study and model workflow, focusing on the key control factors that affect the pore system and the distribution of permeability. In this way it was possible to recognize spatial trends and capture the relationship between petrophysical properties, pore architecture and sweep efficiency.
Reservoir characterization and modelling of highly heterogeneous carbonate reservoirs encompasses the interplay between petrophysical properties, facies, diagenesis, and their relationship with depositional environments. This case study describe a strongly dolomitized carbonate reservoir of Valanginian age onshore Kazakhstan, Central Asia. A reservoir model was built by using an integrated workflow with all the available data, namely seismic, cores, thin sections, logs and MICP. In order to build a robust subsurface model and reduce uncertainties, reservoir rock types (RRTs) were defined and modelled honouring depositional trends and diagenetic attributes.
Due to the complexity of the reservoir, the Winland R35 method, together with Lorenz plots and petrophysical groups, was used to derive the RRTs and to assign a porosity-permeability relationship for each RRT. The uncertainty in the reservoir property models was evaluated with different RRT connectivity scenarios, driven by depositional and diagenetic concepts.
With the integration of diagenetic trends in the model, it was possible to capture the heterogeneity of the reservoir and better understand the porosity and permeability distributions. This has led to development plan optimization through the definition of sweet spot areas and an improved STOIIP calculation.
The results indicate that a substantial improvement in reservoir understanding can be achieved with an integrated reservoir characterization and modelling process that accounts for depositional and diagenetic trends, especially in reducing subsurface uncertainty. Furthermore, it was possible to recognize spatial trends and capture the relationship between petrophysical properties, pore architecture and sweep efficiency. It is expected that the ultimate recovery will also improve.
The case study field is located onshore Kazakhstan, and comprises several oil bearing units. The principal reservoir corresponds to Aptian deltaic-marine sands, whereas this study addresses a secondary reservoir, which is the Valanginian carbonate. The producing structure is an E-W oriented anticline with a western downdip, where some faults are present.
The Carbonate reservoir was discovered as an upside in the mid-2000’s while drilling an exploration well. Encouraging flow tests from a 6 m interval have led to the kick-off of a detailed reservoir modelling exercise, in order to support a development plan. After that, a first pass static model was done with just a few wells. More recently, several appraisal wells were drilled to delineate the extent of the Carbonate reservoir.
The Valanginian Carbonate comprises fine grained limestone, dolomite and marl. This total interval is some 370-400 m thick (Figure 1). The oil bearing unit itself occurs in the uppermost part of the interval, and is mainly composed of skeletal dolopackstone, dolowackestone/dolopackstone, doloboundstones, with some intervals of dolomudstones. This oil bearing unit presents layer cake geometry, and is sealed by anhydrite.
This paper describes the work undertaken to build a 3D static model of a Lower Cretaceous Carbonate Reservoir located in Kazakhstan called X-Field. This reservoir has been pervasively dolomitized, and presents several challenges for development optimization. This model will be used to support further appraisal and development activities, in order to tackle key uncertainties, such as reservoir quality distribution.
All of the available data were quality controlled, analyzed and interpreted (including data from logs and cores), to produce porosity, permeability and RRT (reservoir rock type) models. These are believed to be representative of the reservoir's behavior and connectivity.
In order to identify the main flow zones and understand the reservoir's complexity, Reservoir Rock Typing (RRT) was performed on two cored wells by analyzing CCAL and SCAL data, including thin sections, MICP measurements, porosity and permeability. A comprehensive RRT methodology using Winland R35 method and poro-perm plot was followed, which resulted in defining five rock types. The outcome from the RRT study was confirmed by poro-perm plot, which showed the presence of five flow units.
The 3D model was built by using corner point grids (CPG), and contains a total of 2,380,050 cells. Several models of porosity and RRT were generated, representing "low??, "mid??, and "high?? case scenarios of reservoir quality distribution. Finally, permeability models were created for each scenario, conditioned to their respective Winland R35 porosity-permeability relationships per RRT.
Comparison between the different porosity (F), permeability (k), and RRT models and scenarios, will allow a better management of the reservoir uncertainties during the appraisal and development stages for this reservoir.
Carbonate reservoirs are commonly heterogeneous and their reservoir quality results from complex interactions between depositional facies and diagenetic processes. The Diagenetic Diagram is a powerful tool that helps in the characterization of the diagenetic processes that have affected the reservoir. From this knowledge, it is possible to significantly improve the understanding of the reservoir's pore system and permeability distributions, which are key factors for development optimization and production sustainability.
A multi-scale and multi-method study (petrography, blue-dye impregnation, selective staining and porosity determination) of Middle Jurassic carbonates from the Lusitanian Basin (Portugal) has been undertaken, to find the best systematic approach to these reservoirs. It has involved thorough diagenetic characterization of each lithotype (lithofacies, texture, porosity, qualitative permeability assessment and diagenetic evolution). The study area was selected based on its excellent and varied exposures of carbonate facies and availability of core.
Methodological and terminological challenges were faced during the study, especially dealing with data coming from several scales (macro, meso, and micro). In order to overcome these challenges, a diagenetic diagram was developed and applied to the selected rocks. It is a tool that allows the integration of data coming from outcrops, hand samples, cores, cuttings, thin sections, and laboratory experiments.
This is carried out in a dynamic, guided, systematic, and rigorous way, enabling the evaluation of the relationship between facies, diagenetic evolution and pore systems. The latter are characterized regarding size, geometry, distribution, and connectivity. This enables the identification and characterization of permeability heterogeneities in the rocks. It was concluded that the main porosity class (i.e. secondary) was created by diagenetic processes.
The proposed method has strong application potential for: detailed characterization and understanding of porosity and permeability in carbonate reservoirs, from a diagenetic evolution and fluid flow perspective (e.g. SCAL and pore system description); definition of diagenetic trends for modeling petrophysical properties and rock types. In this regard, the method is being applied to a Valanginian carbonate reservoir in Kazakhstan, and some preliminary results are presented in this paper. Refining this technique may be helpful for similar carbonate studies, enhancing the results of typical diagenetic studies by improving the characterization of reservoir properties at various scales, thus contributing to a more sustainable exploitation of hydrocarbon reservoirs.