Seismic data plays a crucial role in reservoir characterization. Quantitative seismic reservoir characterization workflow aims to extract reservoir properties from seismic data to identify sweet spots for exploration, appraisal and development drilling. This is all the more true if the data is to be used in 4D seismic surveys (Time-lapse seismic) for reservoir monitoring. 4D seismic data can help to capture the reservoir change, realize the reservoir dynamic characterization and facilitate an enhanced understanding for reservoir potential, so as to have the better business decisions and optimized field development plans. In this study, use a SAGD (Steam-Assisted Gravity Drainage) field to demonstrate how to extract reservoir dynamic change and evaluate the reservoir performance from 4D seismic data.

As a unique stratigraphic prospect of UAE, the carbonate Mishrif Formation in NN Field is composed of 15-25 m thick rudist grainstone that formed in a shoal environment. The effective reservoir is bounded by inter-shoal packstone-wackestone. Combination of porous reservoir and non-porous baffles indicate high heterogeneity caused by rapid changes in deposition. Current exploration and drilling proposal are precluded due to the ambiguous understanding on reservoir anisotropy and dim-identification from seismic due to the thin reservoir thickness. To mitigate the challenge from reservoir identification, Mishrif whole core was collected and the following analysis performed: thin section description, porosity and permeability (RCA), X-Ray diffraction (XRD), and mercury injection (MICP). The subsurface analysis of the Mishrif reservoir was augmented with litho-facies identification, sedimentary facies recognition, and diagenetic history. Paleogeography was integrated with sequence stratigraphy to predict possible reservoir distribution. Sequence stratigraphy focused on identifying the 4th order sequence interfaces such as first flooding surface (FFS), maximum flooding surface (MFS), and sequence boundaries (SQ). Subsequently, the paleogeomorphology of oil-bearing zone was conducted, and a method using two crucial sequence surfaces was optimized after comparing impression and residual thickness methods. Meanwhile, to quantitatively characterize this set of oil-bearing units, AVO and Pre-stack inversion was implemented to predict reservoir distribution and fluid habitat. The integrated study revealed that the Mishrif reservoir quality is controlled by original depositional facies and diagenetic processes. The rudist grainstone was shoal-related with deposition on a paleo-geographic high and originally high porosity and excellent pore-connectivity. The subsequent fresh water leaching and dissolution contributed to improvement of pore structure. In contrast, the inter-shoal limestone contains higher micrite deposited in slightly deeper water, due to lower porosity it resisted the weathering procedure. To overcome the challenge of thin reservoir thickness, selection of key surfaces which are used to construct the paleo-geographic configuration would be quite crucial. And the identification of such surfaces only from seismic would be of high uncertainty. Finally, the dual-interface method was adopted to delineate the paleogeomorphology of oil-bearing zone. This illustration of paleogeography displayed high similarity to the reservoir quantitatively derived from AVO and Pre-stack inversion, which improved reservoir prediction. This integrated method from core-based reservoir recognition, sequence-driven paleogeography, as well as quantitative AVO and Pre-stack inversion provides new insight to study heterogeneous carbonates and reduce uncertainty for thin reservoir prediction.

The sequence stratigraphic framework based on the combination of core, lithofacies and logging curve cycles, sometimes hardly reflect the lithology and sedimentary changes between wells, and is inconsistent with seismic data and production performance. Through the integrated research of core, well logging, seismic data and reservoir engineering, this paper proposed an update method. The new method can effectively solve the geological challenges in exploration and development and provide a reliable geological basis for efficient production of the oilfield. This method includes the following 3 steps, (1) identify the sequence stratigraphic boundary integrated the core and lithofacies analysis, and establish the well correlation sequence stratigraphic framework. (2) According to seismic and geological calibration, realize mutual constraint between wells and seismic and robust the sequence stratigraphic framework. (3) The sequence stratigraphic framework is optimized by using production dynamic data, which could grab the sequence stratigraphic framework more consistent with the deposition law. The isochronous sequence stratigraphic framework established by this method in B oilfield of the Middle East truly restores the structural characteristics of the progradational strata of the main production layer in B oilfield, and the sequence boundaries match well between drilling data and seismic data. Under the control of the isochronous sequence stratigraphic configuration, the ambiguous results of the previous division in sublayers according to the lithological isopach were updated, which solved the problems of diachronous oil layer and disordered oil-water relationship in this oilfield. This study also provides an effective isochronous sequence stratigraphic unit for reservoir prediction in exploration and development. Compared with the previous sequence stratigraphy research method in this area, the new method has two major advantages, (1) It complements the shortage of uncertain between wells and increases the accuracy for uncored interval. Furthermore, this method establishes a real isochronous sequence stratigraphic framework. 2) Combined with production dynamic data, the challenge of diachronous sublayers and confusion of oil-water relationship in the research results are avoided.

Faults often control the movement and aggregation of oil and gas. With the development of oil fields, the role of subtle faults is becoming more and more important. The accuracy of fault interpretation directly affects the direction of exploration and development. However, due to the limitation of the seismic resolution, it is hard to identify these faults according to routine methods such as coherence, variance, curvature, etc. To overcome such kind of challenge and better match the demand for fine fault identification, a method integrated deep learning and spectral decomposition was proposed.

With the development of exploration and development, thin reservoir prediction is becoming more and more important. However, due to the limit of seismic resolution, thin reservoir prediction has always been an important challenge in the Middle East. Thin reservoir prediction based on conventional geophysical techniques is not accurate enough to meet the requirements of development. In order to improve the accuracy of thin reservoir prediction, a new thin reservoir prediction technique is proposed. This technical workflow main includes 4 steps: (1) Sedimentary facies identification based on multidisciplinary analysis, (2) Sedimentary facies model and seismic forward modelling, (3) Seismic response characteristics analysis and seismic data conditioning under the guide of forward modelling, (4) Seismic meme inversion and thin reservoir prediction. The new drilled wells demonstrate the successful application of the techniques in the M field in the Middle East. The 5 layers of thin sandstones reservoir can be divided into two sets of high stand systems of sand and low stand systems of incised valley deposits. Geological model seismic forward modelling shows that the most sensitive seismic dominant frequency for effectively identifying the two groups of sandstone is 35Hz (Fig.1). Through the high resolution seismic processing, the main frequency of seismic data was optimized from 25Hz to 35Hz, which improves the recognition ability for the thin sand groups (Fig.2). Seismic facies analysis based on previous and new seismic data shows that different thin reservoir layer can be effectively identified by seismic facies. Under the constraints of seismic facies, the Seismic meme inversion can effectively predict the two sand groups (Fig.3). 51 km2 of thin reservoir favourable area was discovered and 16 wells were drilled with 91% success rate based on the new seismic inversion result in the southeast part of the oilfield. This technology can effectively integrate geological information and seismic conditioning techniques, and improve the accuracy of thin reservoir prediction results more reasonably, which can not only provide support for the exploration for thin reservoir but also efficient development. This technique is applicable not only to thin clastic reservoir but also to thin carbonate reservoir.