Abstract The development of innovative techniques and concepts, and the emergence of new plays in carbonate rocks are creating a resurgence of oil and gas discoveries worldwide. The maturity of a basin and the application of exploration concepts have a fundamental influence on exploration strategies. Exploration success often occurs in underexplored basins by applying existing established geological concepts. This approach is commonly undertaken when new basins ‘open up’ owing to previous political upheavals. The strategy of using new techniques in a proven mature area is particularly appropriate when dealing with unconventional resources (heavy oil, bitumen, stranded gas), while the application of new play concepts (such as lacustrine carbonates) to new areas (i.e. ultra-deep South Atlantic basins) epitomizes frontier exploration. Many low-matrix-porosity hydrocarbon reservoirs are productive because permeability is controlled by fractures and faults. Understanding basic fracture properties is critical in reducing geological risk and therefore reducing well costs and increasing well recovery. The advent of resource plays in carbonate rocks, and the long-standing recognition of naturally fractured carbonate reservoirs means that new fracture and fault analysis and prediction techniques and concepts are essential. A key area of progress has been integration of stratigraphic, structural, geomechanical and diagenetic analysis to populate reservoir models accurately. Dramatic increases in computing and digital imaging capabilities are being harnessed to improve spatial analysis and spatial statistics in reservoirs and ultimately improve 3D geocellular models.

Abstract Carbonate oil reservoirs are sometimes regarded with apprehension in the petroleum industry since it can be difficult to predict the quality of, and ensure high recovery factors from, this rock family. Particular problems are the complex and heterogeneous nature of porosity in carbonate rocks, often leading to large ranges in permeability for any given porosity, and the organization of carbonate successions most commonly as vertically heterogeneous, but laterally persistent, layers. Important issues that arise time and again in carbonate reservoir description include (a) predicting reservoir quality at inter-well scales and in uncored wells, (b) recognizing problematic high-permeability layers, (c) determining the permeability component to allocate to fractures and connected vug systems, and (d) populating reservoir models with representative physical parameters. Because porosity in carbonate rocks generally presents as diverse and heterogeneous, conventional core plugs are seldom representative of large rock volumes and significant issues remain in terms of the scale-compatibility of the various datasets for measured physical parameters that are used in carbonate reservoir description. Many of the world's largest carbonate reservoirs were discovered and developed shortly after the Second World War and are now showing signs of maturity, expressed variously as poor pressure support, water or gas breakthrough and stranded resources. The proportion of the world's ‘conventional’ petroleum that is reservoired in carbonate rocks is commonly estimated at around 50–60% and many large carbonate reservoirs are likely to have a production lifetime beyond 50 years. It is no coincidence then that the petroleum industry has been the primary source of funding of and promotion of research into carbonate rocks and depositional systems, often with impacts extending well beyond oil and gas exploitation.

Abstract The relatively simplistic facies models for lacustrine carbonates do not currently incorporate either the diversity of microbialite carbonate development or the influence of volcanic-related processes found in rift settings. The basic nature of the carbonate factories in these systems, whether microbial, macrophytic, skeletal or abiogenic, is not resolved. Lacustrine microbialites can develop in shallow lakes as concentrations of microbialite mounds covering many hundreds of square kilometres, or as bathymetrically controlled facies belts, but in many rift settings vent-related thermal and non-thermal carbonates (travertines and tufas) are a major component. Subaqueous vent-related carbonates, with evidence of microbial activity, can produce seismic-scale carbonate build-ups in deeper lakes or apparently more stratiform accumulations in shallow lakes. In lakes with only volcanic catchments, Mg and silica activity, coupled with high carbonate alkalinity and microbial influences, can potentially generate a complex set of mineral–microbe interactions and products, creating a unique set of challenges for predicting and understanding reservoirs in such settings.

Abstract The Upper Devonian Grosmont platform in Alberta, Canada, is the world's largest heavy oil reservoir hosted in carbonates, with 400–500 bbbl of initial oil in place at an average depth of about 250–400 m. Past pilot activity between 1975 and 1987, mainly with ‘huff and puff’ technology, had variable results and the reservoir was deemed uneconomical at the time. More advanced thermal recovery technologies, such as steam assisted gravity drive and electrical in-situ retorting, much higher world market prices for oil and certain political pressures have led to a flurry of activity in the Grosmont since 2006. To be successful, all these schemes require a solid geological, geophysical and petrophysical reservoir evaluation. Our study, which aims to aid in thermal recovery of this reservoir, shows that the current reservoir characteristics were created by a succession of five major factors and/or processes: sedimentary stratigraphy, dolomitization, polyphase and polygenetic fracturing, polyphase and polygenetic karstification, and biodegradation. Most of the present porosity and permeability is due to fracturing and karstification. The sedimentary stratigraphy of the Grosmont reservoir consists of six stacked carbonate units interbedded with marls and some evaporites. The latter two originally acted as aquitards during diagenesis but are breached or missing in parts of the area today. Dolomitization by density-driven reflux was the first pervasive diagenetic process. Most dolostones are fine-crystalline and tight, however, and the only notable porosity caused by and/or related to dolomitization is scattered molds and vugs. A dense fracture network was created in three or four phases. Most fractures probably originated from collapse following subsurface salt dissolution and/or from Laramide tectonics far to the west, whereby pulsed crustal loading in the fold-and-thrust belt created a dynamic forebulge in the Grosmont region via multiple pulses of basin-wide crustal flexing, each followed by relaxation. The fracture network probably was reactivated and/or expanded by glacial loading and post-glacial isostatic rebound in the Pleistocene and Holocene, respectively. The region experienced three or four prolonged periods of epigene (top-down) karstification, although there is tangible evidence for only two of them in the Grosmont platform. The first of these episodes was a ‘warm epigene karstification’ during the Jurassic–Cretaceous, and the second was/is a ‘cold epigene karstification’ that started sometime in the Cenozoic and is continuing to this day. In addition, there is circumstantial evidence for hypogene (bottom-up) ‘karstification’ (=dissolution) throughout much of the geological history of the Grosmont since the Late Devonian, with two possible maxima around the time of hydrocarbon emplacement, that is Early–Middle Cretaceous and Early Cenozoic, respectively. Epigene karstification was accompanied and/or followed by extensive biodegradation. Present bitumen in-situ viscosities are >1 million cP and API gravities range from 5 to 9°. Taken together, these conditions pose considerable challenges for any type of thermal recovery scheme.

Abstract Hydrothermal dolomite (HTD) bodies are known as high-quality hydrocarbon reservoirs; however few studies focus on the geometry and distribution of reservoir characteristics. Across the platform-to-basin transition of the Ramales Platform, fault-controlled HTD bodies are present. Three kinds of bodies can be distinguished based on their morphology, that is, elongated HTD corridors, a massive HTD body (Pozalagua body) and an HTD-cemented breccia body. The differences in size and shape of the HTD bodies can be attributed to differences in local structural setting. For the Pozalagua body, an additional sedimentological control is invoked to explain the difference in HTD geometry. A (geo)-statistical investigation of the reservoir characteristics in the Pozalagua body revealed that the HTD types (defined based on their texture) show spatial clustering controlled by the orientation of faults, joints and the platform edge. Porosity and permeability values are distributed in clusters of high and low values; however, they are not significantly different for the three HTD types. Two dolomitization phases (i.e. ferroan and non-ferroan) can be observed in all HTD bodies. In general, the HTDs resulting from the second non-ferroan dolomitization phase have lower porosity values. No difference in permeability is found for the ferroan and non-ferroan dolomites.

Abstract This paper represents the second part of an integrated study that is focussed on the development and distribution of reservoir bodies and properties in epeiric carbonate systems. It is based on outcrop analogue data from Triassic ‘layer-cake’ carbonates in the South German Basin, which were deposited along an epicontinental, very gently inclined carbonate ramp. The reservoir facies consists of skeletal and oolitic carbonate grainstones (Φ max 23%, K max 700 mD), which are organized in a pronounced hierarchy of stratigraphic cycles. Based on outcrops, cores, gamma ray (GR) logs and thin sections, a high-resolution, geocellular 3D facies model was generated, which covers the area of a Middle East giant gas field (25×36 km). The spatial distribution of reservoir properties was systematically investigated on different scales. The lateral distribution of reservoir properties remains in the same order of magnitude for hundreds of metres, within in the same stratigraphic position. However, on a kilometre scale, facies bodies, diagenetic trends and thus reservoir properties show gradual lateral changes. Vertically, in contrast, properties change commonly on a decimetre scale and are largely controlled by stratigraphic cycles. Petrophysical modelling enhanced the understanding of key factors and processes controlling both reservoir quality and quantity.

Abstract Outcrop studies integrated with subsurface data of core, cuttings and different well logs were used to investigate reservoir characteristics of the Middle–Late Eocene Pila Spi Formation at Taq Taq oil field of Kurdistan Region of northeastern Iraq. Reservoir studies include petrographic investigations, microfacies analyses and petrophysical evaluation in an attempt to provide an insight regarding the reservoir potential. The Pila Spi Formation is subdivided into four distinctive lithofacies (P1–P4), characterized by dolostones, dolomitic limestones and limestones. Several types of dolomite were recognized ranging from early diagenetic fenestral fine crystalline to late diagenetic coarse crystalline dolomite, which had positively influenced the reservoir characteristics by enhancing inter-crystalline, intra-skeletal and micro-vug porosity, especially for lithofacies units P2 and P3. Reservoir porosity is heterogeneous in distribution and ranges from 5 to 20%. Using porosity cut-off values of 8.2%, and water saturation cut-off values of 24%, six porosity units were identified from top to bottom (PU1–PU6). The best unit is PU2 (15 m thick), which is characterized by medium crystalline dolomite mosaic with average effective porosity of 21.5%. Permeability ranges between 0.1 and 1 md. Flow unit differentiation is discussed in term of porosity–permeability cross-plot and reservoir pore-throat classification (R35). Results indicate that most of the Pila Spi reservoir is of micro-port (matrix) flow type. However, reservoir quality enhancement is attributed to fracturing.

Abstract The Cretaceous Apulia Platform, exposed in the Murge area (southern Italy), suffered intense (palaeo)karstification. This study focuses on the controlling factors of karstification with emphasis on fracturing. Mechanical stratigraphy was used to calculate the fracture density within different sedimentary sequences. Several mechanical units were defined and a characteristic relationship was found between unit thicknesses and fracture density, that is, fracture density increases if layer thickness decreases. In some of the quarries studied, sedimentary cycles are clearly present that are also reflected in the fracture density logs. The degree of karstification within a mechanical unit is proportional to the mean fracture spacing. Based on fracture orientation data extracted from LIDAR scans, different orientation clusters were observed between fractures that are karstified and fractures that are not karstified, post-dating karstification. The clusters of karstified fractures are related to the compression of the southern Apennines. The fractures became dissolution enlarged during the Pleistocene uplift caused by bulging of the Apulia Platform. This main karstification phase occurred prior to Late-Pleistocene deposition and before the formation of orthogonal fracture sets.

Abstract A correlation is demonstrated between the presence of crack-seal texture and power-law kinematic aperture-size (width) distributions among opening-mode fractures in rocks of dominantly carbonate mineralogy. Crack-seal opening increments (opening-displacement increment sizes or ‘gaps’) within individual fractures follow narrow normal or log-normal size distributions, suggesting that fracture widening accumulates in characteristic (usually micrometre-scale) size increments. The scale invariance in overall fracture width distributions present in some fracture sets most likely arises from grouping of these increments (localization) to form larger fractures (millimetre- to centimetre-scale widths). Such localization could be a consequence of the tendency for larger, less cemented fractures to break preferentially during subsequent deformation. Cement accumulation patterns thus provide a mechanism for positive feedback whereby large-fracture growth exceeds small-fracture growth. Using characteristically sized growth increments, a fracture growth model accurately simulates fracture arrays having power-law fracture-width distributions. Model parameters can be altered to produce characteristic-width fracture size distributions. The results have implications for how fracture porosity and permeability evolve in carbonate reservoirs.

Abstract In Asón Valley hydrothermal dolomite area (Basque–Cantabrian Basin, northern Spain), an overlapping stepover area between two major basement faults, the Cabuerniga and Ruahermosa transtensional faults, was the location for different scales and types of extensive fractures. This fracture mesh affected the Albian Ranero limestone and was formed in a dilational jog, a regional fluid throughflow area. It acted as pathway for overpressured fluids that controlled the dolomite mineralization in the Ranero massif area. The study of synchronous structural features, fluid flow channelling and dolomitization processes indicated their tectonic control. The fluid circulated and concentrated preferentially in more fractured areas with increased permeability, such as extensional chimneys, creating dolomite bodies. Repeated extensive tectonic activity enhanced fracture porosity, promoting overpressured fluid migration and cyclical dolomitization events. The studied fracture pattern suggests the presence of sinistral transtension during the formation of extensional joint-sets that channellized the mineralizing fluid-flow. A gradation in structural features and dolomite facies and textures is visible from the main Pozalagua fault dolomite body to peripheral dolomite bodies as the Breccia body and Ranero megajoints. These differences could reflect a proximal–distal trend from the main fluid-flow area along the Pozalagua fault to the Ranero megajoints.

Abstract The Late Carboniferous strata exposed in the Sacramento Mountains in Southern New Mexico, USA, have long been considered classic exposures documenting reciprocal high-frequency mixed carbonate–siliciclastic cyclicity and shelf-edge algal-mound growth. The growth style and internal architecture of these phylloid algae mounds depend on their position on the shelf and are controlled by potential accommodation space, depth of the photic zone and hydrodynamic energy. The combination of these parameters results in a laterally variable amount of reworked phylloid algae debris and in-situ mound core facies along the depositional profile. This variable architecture can be observed on the outcrop and results in a complex distribution of these two lithofacies in three dimensions that is challenging to reproduce in a 3D geocellular model. Two geostatistical estimation algorithms are used to stochastically model carbonate buildups: surface-based and multipoint statistics (MPS)-based. The surface-based model uses two-point statistics and is built by first recreating the overall geometry of the mound and then reproducing the internal architecture using indicator Gaussian simulation, but requires strong secondary trend data to reproduce the correct facies architecture. The MPS model successfully recreates both the geometry and internal architecture of the mound, but requires a complicated training image and complex multigrid simulation that would be hard to implement in subsurface. This comparison demonstrates that modelling carbonate buildup geometry and internal architecture is not trivial and requires complex workflow with secondary trends. These secondary trends require a significant amount of prior knowledge that is easily extracted from outcrop observations, but would be difficult to assess in subsurface data.

Abstract Although karstic networks may have a major impact on fluid flow in reservoir characterization, they exhibit great intrinsic heterogeneity that makes their characterization very complex. This work proposes an integrated workflow to study and stochastically simulate karstic networks. This approach is based on the study of outcropping caves. Topological and geometrical parameters are automatically extracted from cave surveys. The extracted geometrical parameters are used to determine the input parameters of the stochastic simulations. These simulations utilize a structure-based, pixel-based and geostatistical approach. To check for consistency, a procedure is proposed to compare the topological and geometrical parameters of observed and simulated karst networks using multivariate analyses. The proposed integrated workflow has been successfully applied to a real case study involving karsts in Jurassic limestone from the south of France. The obtained karstic networks reproduce observed topological and geometrical parameters even when the employed simulation approach relies solely on geometrical parameters.

Abstract The stochastic stratigraphic well correlation method considers the stratigraphic correlation of well data as a set of possible models to sample and manage uncertainty in subsurface studies. This method was applied to the Malampaya buildup (a well documented offshore gas field located NW of the Palawan Island, Philippines), aged upper Eocene to lower Miocene. Previous studies highlight that rock petrophysical properties are mainly controlled by diagenesis. Correlation rules are thus developed in order to adapt the stochastic stratigraphic well correlation method to the study of diagenetic units. These rules are based on wireline log shape and diagenetic units types. Four stratigraphic correlation models are generated using the proposed correlation method: a deterministic one corresponding to the most probable model considering only well data and three stochastic ones. These correlation models are bound with geostatistical methods to build static reservoir models. Synthetic seismic profiles are computed from facies models conditioned to acoustic impedance models. It leads to comparable seismic amplitude images, highlighting the importance of considering several well correlation models for one given seismic survey. Stochastic stratigraphic correlations are shown to have a first-order impact on reservoir unit characterization, rock volumes and fluid flow response on the reservoir model.

Abstract This study presents a workflow for 3D modelling of carbonate reservoirs using multiple-point statistics (MPS) in the framework of a pre-existing model. It consists of the following steps: (1) applying a hierarchical classification scheme for carbonate geobodies; (2) based on this classification, retrieval of relevant data from Carbdb, a novel software to manage a database of analogue studies; (3) construction of training images based on the retrieved data from Carbdb; and (4) using the training images when building a 3D reservoir model with MPS. MPS makes use of training images to capture depositional patterns, which will then be reproduced during the stochastic simulations. Carbdb provides a library of quantitative data such as dimensions, geometries and the distribution pattern of geobody analogues necessary for building training images. The MPS workflow was applied to carbonate shoal bodies from a reservoir analogue, the Muschelkalk in SW Germany. Present-day shoal bodies from the Arabian Gulf were retrieved from Carbdb as possible modern analogues to generate training images. The realizations of this MPS approach are compared with a previously established 3D geocellular model that was built deterministically by interactive facies modelling. The MPS simulations produced geologically more realistic facies distributions with higher facies heterogeneity, similar to the depositional patterns observed in modern analogues.

Abstract This paper presents a method to stochastically simulate 3D karstic networks and more specifically branchwork pattern cave systems. Considering that they can be compared with 3D fluvial networks, the topological classification of Strahler and the corresponding ratios of Horton are used to define three morphometric parameters. These parameters are integrated in an algorithm that computes branches hierarchically to obtain a final network organized around the main observed inlet and outlet with a branching complexity controlled by the user. Each branch corresponds to a low-cost path between two points calculated with the 𝒜★ graph search algorithm. Speleogenetic information on inception horizons, palaeo-water tables and fractures is accounted for by adapted definitions of the searching functions of 𝒜★. The method is demonstrated on a 3D synthetic case with discrete fractures networks, inception horizons and a palaeo-water table. The simulated karsts have a realistic geometry and are geologically consistent.

Abstract Carbonate reservoirs contain an increasingly important percentage of the world's hydrocarbon reserves. This volume presents key recent advances in carbonate exploration and reservoir analysis. As well as a comprehensive overview of the trends in carbonate over the years, the volume focuses on four key areas: emerging plays and techniques – with special reference to lacustrine plays in syn-rift basins and development of super-giant heavy oil plays improved reservoir characterization – with examples from the Middle East and Europe and case studies of how outcrop analogues can provide key data for input to geological models impact of fractures and faults in carbonates –contributors highlight the need for integrated structural and diagenetic approaches in order to understand how fractures evolve as fluid-flow conduits advances in geomodelling of carbonate reservoirs –several papers discuss the application of new and innovative geomodelling and geostatistical techniques to carbonate reservoirs.