Flow Unit Model Research Articles

Reservoir hydrocarbon potential of the Miocene Agbada Formation using RMS amplitude map and hydraulic flow unit model, Hatch Field offshore, Niger Delta Basin, Nigeria

The Miocene Agbada Formation is the hydrocarbon-producing unit in the Niger Delta region because it has excellent stratigraphic and structural traps. Understanding the reservoirs' hydraulic flow characteristics in the Agbada Formation is necessary. The flow of fluids in a reservoir is controlled by the effective porosity of the reservoir that can be spatially correlated with root mean square (RMS) amplitude and HFUs.The study aims to produce an RMS map and HFUs model for reservoir units D and E in the Hatch Field Offshore Niger Delta Basin and determine the reservoirs' hydrocarbon potential using the reservoir seismic data, core analysis, and well-log data. The seismic attribute of RMS amplitude was applied for both reservoir units D and E in the Hatch field. The calculated FZI was categorised using a conditional statement with two HFUs. The hydraulic flow unit model for both reservoir units was built utilising a simulation grid design of 157x100x5 for reservoir unit D and 204x120x5 for reservoir unit E. The sequential indicator simulation (SISIM) algorithm was applied to simulate the HFUs model for reservoir units D and E. The RMS amplitude maps show bright spots that are possible hydrocarbon prospects in P1, P2, and P3 for reservoir unit D. However, reservoir unit E has bright spots that are also likely hydrocarbon prospects here named P4 and P5. The brighter amplitudes observed in prospects P4 and P5 signify significant amounts of hydrocarbons in reservoir unit E. The results of the hydraulic flow units’ model is represented by the spatial variability of flow units within the reservoir. The coefficient of determination between well-log data and the model results range from 0.9258 to 0.9655 and 0.8586 to 0.9868 for reservoir units D and E, respectively. A litho-saturation cross plot comparing lithofacies and HFUs confirms that HFU 1 comprises sandstone while HFU 2 is composed of marine shale. The reservoirs' spatial distribution of hydraulic flow units shows that unit D has poor hydraulic flow characteristics and hydrocarbon potential due to the high proportion of marine shale facies in HFU 2 compared to HFU 1. Reservoir unit E predominantly has a higher proportion of sandstone in HFU 1 compared to HFU 2, making it a better reservoir with good hydrocarbon potential than reservoir unit D.

Effects of annealing time on nanoscale structural heterogeneity and magnetic properties of Fe<sub>80</sub>Si<sub>9</sub>B<sub>10</sub>Cu<sub>1</sub> amorphous alloy

The evolution of nanoscale structural heterogeneity and its effect on magnetic properties of Fe<sub>80</sub>Si<sub>9</sub>B<sub>10</sub>Cu<sub>1</sub> amorphous alloy during structural relaxation after being annealed for different times are investigated in this work. The nanoscale structural heterogeneity is found to degenerate gradually with relaxation by using the small-angle X-ray scattering and atomic force microscope. Combined with Mössbauer spectroscopy analysis results, the enhanced comprehensive soft magnetic properties of the relaxed alloys can be attributed to the degeneration of nanoscale structural heterogeneity. From the flow unit model, the volume fraction of flow units decreases with relaxation proceeding, and some of the flow units annihilate and transform into the ideal elastic matrix. On the one hand, the relaxed sample with greater packing density has stronger magnetic exchange interaction and higher saturation magnetic flux intensity. On the other hand, the number density of quasi-dislocation dipoles decreases with the annihilation of flow units in the relaxation process, leading the pinning effect of the domain wall to be weakened. Consequently, the magnetic anisotropy decreases after relaxation, which results in the reduction of coercivity. In this work, the structural mechanism of the evolution of magnetic properties in the relaxation process of Fe<sub>80</sub>Si<sub>9</sub>B<sub>10</sub>Cu<sub>1</sub> amorphous alloy is investigated from the perspective of structural heterogeneity, which is helpful in establishing the correlation between the structure and magnetic properties of Fe-based amorphous alloys.

Separating β relaxation from α relaxation in fragile metallic glasses based on ultrafast flash differential scanning calorimetry

Despite numerous studies on the glass transition and the related relaxation dynamics, the physical mechanism of activation of multiscale relaxation events under various external stimuli in amorphous materials is still unclear. In this study, by combining traditional differential scanning calorimetry (DSC) and Flash DSC with heating rates spanning over five orders of magnitude, the thermodynamic responses have been systematically studied for several fragile and strong metallic glasses (MGs). A common endothermic event, before the glass transition, is detected when the heating rate increases above a critical value for fragile MGs. This endothermic event is verified to represent the activation of secondary $\ensuremath{\beta}$ relaxation (Johari-Goldstein relaxation), which is commonly found in amorphous materials. For fragile MGs, with an increase in fragility, the critical heating rate to separate $\ensuremath{\beta}$ relaxation from $\ensuremath{\alpha}$ relaxation decreases. In contrast, the $\ensuremath{\beta}$ relaxation does not appear within the current experimental heating rate limit for strong glass systems. Finally, based on the potential energy landscape model and the flow unit model of the heterogeneous structure for MGs, a pathway is proposed for the fragile and strong MGs to understand the physical mechanism for the separation of $\ensuremath{\beta}$ relaxation from $\ensuremath{\alpha}$ relaxation via ultrafast heating. This study clearly demonstrates that the Flash DSC with a wide heating rate range is an effective tool to study the relaxation dynamics in amorphous materials.

Developing a workflow to select representative reservoir models combining distance-based clustering and data assimilation for decision making process

Different types of static and dynamic uncertainties can play critical roles to select an efficient and reliable production strategy in the oil fields, yielding a large number of models to analyze. The definition of such a production strategy requires flow simulation for each model. Since the flow simulation process for this large model ensemble requires a great CPU effort, selecting a subset of representative models (RMs) is often used for accelerating the analysis. However, providing an applicable and efficient method to identify the RMs that statistically reflect the same dynamic and static properties as the full set of models is still challenging. In this study, we develop a new workflow to select RMs based on the integration of distance based clustering and data assimilation aimed at the field development. The workflow begins with the generation of geostatistical models using Latin hypercube (LH) sampling and then reducing them through distance-based clustering according to their static features. Therefore, there is no need to apply numerical simulation for scenario reduction. Next, discrete Latin hypercube with geostatistical realizations (DLHG) method is performed to combine other types of uncertainty with the reduced geostatistical models and build the simulation models. Eventually, data assimilation (DA) is applied to select the subset of models that match the past reservoir performance and production data. In addition, the uncertainty range are evaluated for the defined well and field objective functions using the cumulative distribution function (CDF) and non-parametric Kolmogorov-Smirnov (K-S) test. The methodology is successfully applied to a synthetic benchmark case named UNISIM-II-D considering the flow unit modeling. The results show that the workflow can effectively and suitably identify the RMs considering different uncertainty types, reservoir heterogeneity and many field and well objective functions. The selected RMs are used for making production forecasts and development planning support.

Flow Unit Model of Channel Sand Body and Its Effect on Remnant Oil Distribution: A Case Study of PI Formation in the Eastern Transition Zone of Daqing Oilfield

To analyze the effect of various flow units in a channel sand body on remnant oil, we established a connection between various flow unit types and the remnant oil distribution. Using stratigraphic correlation and the characterization of sedimentary microfacies, we describe a single sand body, point bar, and narrow channel located at the injection-production well pattern of well B2-60-FB271 in the Eastern transition zone of the Daqing Placanticline. Architecture models of the point bar and narrow channel are also established using a series of parameters from different measurement methods. Four types of flow units (strong-current limiting, medium-current limiting, weak-current limiting, and none-current limiting) in the point bar sand body were identified, whereas one type, unshielded unit, was identified in the narrow channel. Geological parameters, such as porosity, permeability, and pore-throat radius (r50), were optimized to quantitatively characterize these various flow units. Samples were obtained from well B2-60-FB271 and analyzed by the freeze-fluorescence thin section technique. According to the displacement degree, the microscopic remnant oil was divided into three types: (1) free-state remnant oil, (2) semi-free-state remnant oil, and (3) bound-state remnant oil. We found that the strong-current limiting flow unit in the point bar is the enrichment area of free-state microscopic remnant oil and that the medium-current limiting and weak-current limiting flow units also have relatively high free microscopic remnant oil. These constitute the remaining oil enrichment areas in the study area.

Classification and application of flow units in carbonate reservoirs

In the carbonate oil field offshore of Sumatra block in Sunda basin, RQI (Rock Quality Index) and FZI (Flow Zone Index) have proved to be effective method of dividing the flow units. The method utilizes the porosity and permeability of the core rock samples to divide the flow units of the K oil field carbonate reservoirs. The field consists of six flow units, and they represent different flow characteristics. The results show that each flow unit has a specific relationship of porosity and permeability and a specific capillary pressure curve (J function); this feature is related to the rock pore structure and pore size. In the application, the flow unit model is built by a stochastic modeling method, and the permeability model and the oil saturation model are built under the control of the flow unit model.

Flow units as dynamic defects in metallic glassy materials.

In a crystalline material, structural defects such as dislocations or twins are well defined and largely determine the mechanical and other properties of the material. For metallic glass (MG) with unique properties in the absence of a long-range lattice, intensive efforts have focused on the search for similar ‘defects’. The primary objective has been the elucidation of the flow mechanism of MGs. However, their atomistic mechanism of mechanical deformation and atomic flow response to stress, temperature, and failure, have proven to be challenging. In this paper, we briefly review the state-of-the-art studies on the dynamic defects in metallic glasses from the perspective of flow units. The characteristics, activation and evolution processes of flow units as well as their correlation with mechanical properties, including plasticity, strength, fracture, and dynamic relaxation, are introduced. We show that flow units that are similar to structural defects such as dislocations are crucial in the optimization and design of metallic glassy materials via the thermal, mechanical and high-pressure tailoring of these units. In this report, the relevant issues and open questions with regard to the flow unit model are also introduced and discussed.

Hydraulic flow units' estimation from seismic data using artificial intelligence systems, an example from a gas reservoir in the Persian Gulf

In recent years, considering the reservoir pressure drop in productive wells, designing the optimal well trajectory for production and injection in enhanced oil recovery (EOR) plans requires to determine hydraulic flow unit (HFU) in the reservoir. HFUs can also be used for petrophysical zonation of reservoirs as well as permeability predictions in uncored intervals or zones with low quality core data of wells. In the present study we have tried to integrate 3D seismic data with well data in order to find a quantitative relationship between flow zone index (FZI) and seismic attributes through employing linear regression and artificial intelligence models. Using this approach, FZI can be predicted using the information which is propagated in the whole field and achievable in the early stage of field development and consequently a suitable HFUs model may be represented. To this end, the suitable attributes for FZI estimation were selected by stepwise linear regression from extracted acoustic impedance (AI) and sample based attributes. Afterward, three optimal intelligent systems including probabilistic neural network (PNN), multi-layer feed forward network (MLFN), and radial basis function networks (RBFN) were employed. The obtained results reveal that PNN is the most accurate estimator compared to MLFN, RBFN, and multi-attribute regression methods. In the final stage, PNN was applied to develop 3D hydraulic flow unit model for the reservoir section of the investigated carbonate gas field located in Persian Gulf.

In-situ synchrotron X-ray diffraction study of dual-step strain variation in laser shock peened metallic glasses

Atomic-structure evolution is significant in understanding the deformation mechanism of metallic glasses. Here, we firstly find a dual-step atomic strain variation in laser-shock-peened (LSPed) metallic glasses during compression tests by using in-situ synchrotron X-ray diffraction. Under low compressive load, LSP-deformed zone's atomic-structure shows low Young's Modulus (E); with load increase, atomic-structure are re-hardened, showing high E. An atomic deformation mechanism is proposed by using flow unit model, that LSP could induce interconnected flow units and homogenize the atomic-structure. These interconnected flow units are metastable and start to annihilate during compressive loading, causing the dual-step atomic strain variation.

Flow unit model in metallic glasses

Metallic glass is a promising metallic material with many unique properties, and also considered as a model system to study the mysteries of amorphous materials. Recently, many experimental and simulation results supported the existence of “flow unit” in metallic glass. In this paper, we review the background, the theoretical and experimental evidences of flow unit model. Flow units are considered as those loosely packed regions embedded inside the elastic matrix and behave like viscous liquid. Compared with the matrix, flow unit regions have low modulus and strength, low viscosity, high atomic mobility and stand in the saddle points on energy landscape. Therefore, flow units can be treated as dynamical defects in metallic glass. The feature, activation and evolution process of flow unit region in metallic glass as well as their correlation with property in metallic glass are also reviewed. Through dynamical mechaincal methods like dynamical mechanical spectra and stress relaxation, flow unit region and its properties can be distinguished and studied. A three-parameter physical model is proposed to describe the mechnical behaivors of flow units. The activations and evolutions of flow unit under different temperature and strain conditions are studied. A three-stage evolution process is found and the relation with mechanical performance and relaxation behavior is established. The characteristics of flow units are also related to various properties of metallic glass, like plasticity, strength, fracture and boson peaks. By using the thermal, mechanical and high pressure aging procedues, the properties of metallic glass can be manipulated as desired through adjusting the density of flow units. We show that the flow unit model not only helps to understand the mechanism behind many long-standing issues like deformation, glass transition dynamic relaxations, and the connection between strucutre and properties and performance of metallic glasses, but also is crucial for tuning and designing the properties of metallic glasses.

Heterogeneity: the soul of metallic glasses

Owing to the superior mechanical and physical properties, metallic glasses (MGs) have attracted tremendous attention as promising candidates for structural and functional applications. Unfortunately, the ability to form uncontrollable glasses, the poor stability and the unpredicted catastrophic failure stemming from the disordered structure, as the Achilles' heel of MGs, severely restrict their large-scale applications. A number of phenomenological models, such as free volume model, shear transformation zone (STZ) model, flow unit model, etc., have been proposed, intending to relate microstructures to properties of MGs. However, few sophisticated structure-property relationships are established due to a poor understanding of the microstructure of MGs. Recently, heterogeneity is commonly believed to be intrinsic to MGs, and it can be used to establish the structure-property relationship of MGs. In this paper, we review the recent progress of MGs from the angle of heterogeneity, including the static heterogeneities and dynamic heterogeneities. The perspectives of the scientific problems and the challenges of metallic glass researches are also discussed briefly.

Flow units perspective on elastic recovery under sharp contact loading in metallic glasses

The obscure nature of glass physics has led to develop various correlations between different parameters and properties of metallic glasses. Despite these correlations, the clear picture of plastic deformation is still lacking. We have measured elastic recovery in metallic glasses by indentation, and found the elastic recovery correlate with different properties and parameters of metallic glasses. All these observations can be quite well explained with flow unit model which could provide clearer picture on the plastic deformations and nature of the metallic glasses.

A study on remaining oil distribution in a carbonate oil reservoir based on reservoir flow units

Abstract Reservoir Γ north, which belongs to Zhanzhol, Kazakhstan, a fracture-pore typed carbonate oilfield, has a variety of reservoir types. According to the combination pattern of different voids and the relationship between porosity and permeability, carbonate reservoirs are classified into four types which are fracture-cavity-pore typed (referred to as composite typed here after), fracture-pore typed, pore typed and fracture typed, and the identification of which by well logging data is realized. There is a little difference in the producing degree within the same reservoir type for composite typed and fracture typed reservoirs. However, there are large differences in the producing degree for fracture-pore typed and pore typed reservoirs. In order to appraise the producing degree of reservoirs more exactly, key parameters affecting the producing degree of fracture-pore typed and pore typed reservoirs are selected separately, which, for the fracture-pore typed are reservoir quality index ( RQI ), pore throat radius at mercury saturation of 50% ( R 50 ), total permeability ( K f + K m ), and the ratio of fracture permeability to matrix permeability ( K f / K m ), and for the pore typed are RQI, R 50 , and matrix permeability ( K m ). With the integrated consideration of matrix and fracture, a flow unit classification method for dual medium reservoir is established, and taking these parameters as clustering variants, four types of reservoirs are further divided into six kinds of reservoir flow units by the means of neural network clustering technology. Based on reservoir flow unit modeling, a numerical simulation model is built with each type of reservoir flow units adopting its corresponding relative permeability curve, based on which, the distribution regularity of remaining oil of reservoir Γ north is studied and demonstrated. Eight new wells are designed according to the research result, and put into production in 2013 with the initial daily oil production rate being 2.3 times that of nearby old wells.

A CONSTRUÇÃO DE MODELOS REALÍSTICOS DE RESERVATÓRIOS SILICICLÁSTICOS ATRAVÉS DA INTEGRAÇÃO DE ESTUDOS DAS HETEROGENEIDADES DEPOSICIONAIS E DIAGENÉTICAS EM UMA ABORDAGEM DE UNIDADE DE FLUXO

Nesse artigo, é apresentada uma abordagem para caracterização e desenvolvimento de reservatórios baseada na análise de heterogeneidades deposicionais, suas hierarquias e como essas heterogeneidades devem ser focadas em um modelo de unidade de fluxo. O termo unidade de fluxo genética como utilizado nesse artigo representa um elemento que tem uma escala definida, correspondendo a um subsistema de deposição (associação de elementos arquiteturais), parte de uma unidade estratigráfica de quarta ordem. Essas unidades são utilizadas para prover controles geológicos consistentes para a análise da variabilidade da qualidade do reservatório, em escala de campo de petróleo, especialmente em projetos que estejam em estágios de melhoria avançada de recuperação de fluido. Essa necessidade de dependência de escala para a definição das unidades de fluxo contrasta com as definições anteriores. A capacidade preditiva de um modelo baseado em unidades de fluxo genéticas reside no reconhecimento dessas unidades em perfis elétricos, considerando um arcabouço de estratigrafia de alta resolução. O impacto dos processos diagenéticos nas variações entre porosidade e permeabilidade também é discutido nesse artigo, com ênfase na integração dessa análise ao processo de modelagem de reservatórios. Em alguns casos, o efeito da diagênese ou de outros processos pós-deposicionais (fraturamento, por exemplo) pode comprometer a aplicabilidade da abordagem por unidades de fluxo genéticas como é proposto aqui.

Flow Unit Modeling in Complex Reservoirs: ABSTRACT

Flow Unit Modeling in Complex Reservoirs: ABSTRACT

Depositional Controls on Reservoir Properties in a Braid-Delta Sandstone, Tirrawarra Oil Field, South Australia

The Tirrawarra Sandstone contains 146 million bbl of oil in Tirrawarra field in the Cooper basin of South Australia. We used core, well logs, and petrophysical data to construct a depositional-facies-based flow-unit model of the reservoir, which describes rock properties and hydrocarbon saturations in three dimensions. Using the model to calculate volumes and residency of original and remaining oil in place, we identified an additional 36 million bbl of oil in place and improved understanding of past production patterns. The Tirrawarra Sandstone reservoir was deposited in a Carboniferous-Permian proglacial intracratonic setting and is composed of lacustrine and fluvial facies assemblages. The stratigraphic framework of these nonmarine facies is defined by distinctive stacking patterns and erosional unconformities. Mudstone dominated zones that are analogous to marine maximum flooding surfaces bound the reservoir. At its base a progradational lacustrine-delta system, composed of lenticular mud-clast-rich sandstones enclosed in mudstone, is truncated by an unconformity. Sandstones in these lower deltaic facies lost most of their porosity by mechanical compaction of ductile grains. Above the unconformity a braid-delta system, dominated by aggradational bed-load channel-fill sandstones, forms the core of he reservoir. Sediment reworking by channel migration and locally by shore-zone processes created quartz-rich, multilateral sandstones, which retained the highest porosity and permeability of all the reservoir facies and contained most of the original oil in place. The braid delta is erosionally truncated and overlain by a more proximal braid-plain system. Braided-channel sandstones, however, are overlain by lenticular meandering-channel sandstones, which in turn grade upward into widespread mudstones and coals. Thus, this uppermost part of the reservoir displays a retrogradational stacking pattern and upward-decreasing reservoir quality. On the basis of the systematic arrangement of facies within this stratigraphic framework, as well as facies-related differences in composition and texture, we identified reservoir flow units that have characteristic petrophysical properties. Our results demonstrate that depositional variables are the primary controls on reservoir quality and productivity in the Tirrawarra Sandstone.

Flow Unit Modeling of a Heterogeneous Shallow Shelf Carbonate Reservoir (Clear Fork/Glorieta), North Robertson Unit, Gaines County, Texas: ABSTRACT

Flow Unit Modeling of a Heterogeneous Shallow Shelf Carbonate Reservoir (Clear Fork/Glorieta), North Robertson Unit, Gaines County, Texas: ABSTRACT

Stochastic Modeling of Lower Delta and Coastal Plain Reservoir in the Eocene-Misoa Formation, Ceuta Field, Venezuela: ABSTRACT

Stochastic modeling technique minimizes uncertainties in the values used for reservoir parameters and this reduces the risk in reservoir management practices. This technique was successfully employed in modeling south Ceuta field in Venezuela which will be subjected to secondary recovery due to significant pressure decline. The Lower Misoa reservoir is divided into five sedimentological units; the upper three units am characterized by distributary channels, mouth bars and crevasse plains, whereas the lower units we deposited in coastal plain bars and tidal channels. The deterministic data was used to build a stochastic model with HERESIM simulator which employs a Gaussian Truncated Technique. Eight stochastic realizations were generated and the geometry and the quality of each sedimentary unit was defined for every realization. A quantitative flow unit model was developed in a grid of 125 x 60 x 280 uniform blocks, the dimensions of each block being 180 m square and .3048 m in height. The two uppermost units exhibit better quality rocks. Eighty percent of this interval contains facies with good lateral continuity, ranging from 1-3 km within the coastal bar deposits, and from 1-2 km within the tidal channels. The quality of the deposits degrades towards southwest and West. Permeabilitymore » ranges from 100 to 480 md in the best units and from 31 to 250 md in the lower units. Reservoir porosity varies from 10-14%. An interface between HERESIM and the ECLIPSE reservoir simulator was developed to establish a dynamic fluid flow simulation. Four out of the eight realizations produced excellent match for the pressure-production history without any need for data manipulation. The technique therefore, could save approximately 60% of history match time.« less

The Role of Diagenetic Studies in Flow-Unit Modeling: San Andres Formation, Yoakum County, Texas: ABSTRACT

The Permian San Andres Formation represents one of the most prolific hydrocarbon-producing intervals of the Permian basin. Dolostone lithofacies intercalated with thin evaporites accommodate highly compartmentalized reservoirs resulting from complex depositional and diagenetic histories. This compartmentalization often facilitates the use of these reservoirs in flow-unit studies. Perhaps more important than the relationship of productive intervals to depositional facies is the degree to which diagenetic processes have influenced reservoir properties. Detailed petrographic evaluation of the reservoir in question, though often overlooked, should be an integral part of flow-unit studies. Once a diagenetic sequence is established, the information may be incorporated in to the facies model to better understand how to subdivide the reservoir. Such an investigation has been conducted on the San Andres Formation in Reeves field of southeastern Yoakum County, Texas. Here, multistage diagenetic overprints are superimposed on depositional facies that vary in degree of lateral extent, thereby complicating the geometries of individual productive zones within the reservoir. Analysis of the reservoir reveals that Reeves San Andres sediments were subjected to dominant diagenetic processes, including dolomitization and sulfate implacement, both of which are major factors in porosity preservation, and a variety of minor processes that have had little effect onmore » reservoir quality. The recognition of diagenetic facies, and understanding of the processes that have created them, and identification of the implications of these processes on reservoir properties is a vital part of any flow-unit study.« less

Integrated Characterization of Carbonate Ramp Reservoirs Using Permian San Andres Formation Outcrop Analogs

The San Andres Formation (Permian, Guadalupian) of the Permian basin is representative of carbonate ramp reservoirs in that it has highly stratified character, complex facies and permeability structure, and generally low recovery efficiencies of 30% of original oil in place. The approach used here to describe carbonate ramp reservoirs such as the San Andres Formation produces detailed reservoir models based on integration of sequence stratigraphic analysis, petrophysical quantification through definition of rock fabric flow units, and fluid flow simulation. Synthesis of these subdisciplines clarifies which aspects of the geologic-petrophysical model are most significant in predicting reservoir performance and ultimately in understanding the location of remaining oil satur tion. The San Andres Formation crops out along the Algerita escarpment, a long, oblique-dip, continuous shelf-to-basin exposure in the central Guadalupe Mountains. These outcrops provide a unique opportunity to study lateral relationships in geologic and petrophysical structure analogous to those occurring between wells in subsurface reservoirs. On the basis of sequence stratigraphic analysis, three scales of cyclicity are recognized: depositional sequences, high-frequency sequences, and cycles. Examination of the cycles in two detailed window areas provides a practical scale for petrophysical quantification and fluid flow simulation. An understanding of cycle position within the high-frequency sequence framework also provides predictive information. Petrophysical analysis revealed six rock fabric groups dominated by intergranular, separate vug, or dense intercrystalline pore types. Comparison of these rock fabric groups with facies descriptions produced a rock fabric flow unit model that honors the geologic structure of the cycle and sequence framework. Permeability data were averaged within rock fabric flow units using a geometric mean approach based on fine-scale fluid flow modeling of deterministic and stochastically generated permeability fields. Two-dimensional black oil fluid flow models illustrate that (1) major differences in sweep efficiency and fluid flow performance are predicted when linear interwell interpolations are compared with actual interwell-scale geologic structure as determined by outcrop geologic and petrophysical mapping, (2) an understanding of static geologic/petrophysical conditions provides only a partial understanding of reservoir performance defined by the interaction of these static properties and dynamic properties of fluid flow interaction within the flow unit architecture, and (3) because of the orderly distribution of high- and low-permeability facies within cycle stacks of high-frequency sequences, this larger scale of geologic description can give a reasonable first-order approximation of fluid flow patterns and early breakthrough. End_Page 181------------------------------