Guy Desaubliaux

Engie · DEG

The Bonneville Basin is a continental lacustrine system accommodating extensive microbial carbonate deposits corresponding to two distinct phases: the deep Lake Bonneville (30,000 to 11,500 ¹⁴C BP) and the shallow Great Salt Lake (since 11,500 ¹⁴C BP). A characterization of these microbial deposits and their associated sediments provides insights into their spatio‐temporal distribution patterns. The Bonneville phase preferentially displays vertical distribution of the microbial deposits resulting from high‐amplitude lake level variations. Due to the basin physiography, the microbial deposits were restricted to a narrow shoreline belt following Bonneville lake level variations. Carbonate production was more efficient during intervals of relative lake level stability as recorded by the formation of successive terraces. In contrast, the Great Salt Lake microbial deposits showed a great lateral distribution, linked to the modern flat bottom configuration. A low vertical distribution of the microbial deposits was the result of the shallow water depth combined with a low amplitude of lake level fluctuations. These younger microbial deposits display a higher diversity of fabrics and sizes. They are distributed along an extensive ‘shore to lake’ transect on a flat platform in relation to local and progressive accommodation space changes. Microbial deposits are temporally discontinuous throughout the lake history showing longer hiatuses during the Bonneville phase. The main parameters controlling the rate of carbonate production are related to the interaction between physical (kinetics of the mineral precipitation, lake water temperature and runoff), chemical (Ca²⁺, Mg²⁺ and HCO3‐ concentrations, Mg/Ca ratio, dilution and depletion) and/or biological (trophic) factors. The contrast in evolution of Lake Bonneville and Great Salt Lake microbial deposits during the lacustrine history leads to discussions on major chemical and climatic changes during this interval as well as the role of physiography. Furthermore, it provides novel insights into the composition, structure and formation of microbialite‐rich carbonate deposits under freshwater and hypersaline conditions. This article is protected by copyright. All rights reserved.

To bridge the gap between the marginal coarse-grained delta and the basin axis depocenter in high-sediment supply setting, we propose to study outcrops from the Ilias Gilbert-type delta to pro-delta along the southern margin of the Gulf of Corinth in Greece. An entire sedimentological profile is documented from source to sink based on standard field observations enhanced by 3D photogrammetric models (UAV acquisition). We primarily focus on the facies distributions, facies associations, internal architectures and the sandbody morphologies. The Gilbert-type delta bottomset is classically described as produced by decantation (Gilbert, 1885) and limited mass transport deposit in the toe of the delta (Postma and Roep, 1985). However, this field based study reveals four bottomset-dynamics under specific processes integrated within a stratigraphic frame. Each bottomset dynamic has specific facies, facies association and architecture. One sandy-rich bottomset has been identified with a slope apron geometry produced by high-density turbidity during the transgressive stage. During the normal regression, the bottomset records a complete erosional by-pass stage, i.e. the sediments are totally exported throughout the pro-delta. Downstream of the Gilbert-type delta bottomsets (i.e. > 500 m water depth), the sedimentary system is dominated by conglomeratic to sandy rich channels with an axis of 60° from the delta axis and located along slope. The channels are limited by both external and internal margin with specific facies and architectures only on the basinward-side. Channels are commonly characterized by low sinuosity geometries with a mean slope of 2°. The external margin could locally include sediment waves as well as Mass Transport Deposits. The facies, architecture of the margin and direction of the channel argue for a bottom-currents driven system feeding by the gravity driven sedimentary flows from the delta. The conglomeratic channel outcrops along more than 3 km, constituting a new coarse-grained end-member in the deep water depositional system. This study provides the key elements to recognize the facies and architectures of the different bottomset typologies in relation with the associated pro-deltaic and mixed turbiditic-contouritic system. These sedimentary and stratigraphic models improve the prediction for the sand and conglomerate distribution and their connectivity within the various parts of the delta toe induces by a high-sediment discharge.

The Great Salt Lake is a modern hypersaline lake, in which an extended modern and ancient microbial sedimen- tary system has developed. Detailed mapping based on aerial images and field observations can be used to identify non- random distribution patterns of microbial deposits, such as paleoshorelines associated with extensive polygons or fault- parallel alignments. Although it has been inferred that cli- matic changes controlling the lake level fluctuations explain the distribution of paleoshorelines and polygons, straight mi- crobial deposit alignments may underline a normal fault sys- tem parallel to the Wasatch Front. This study is based on ob- servations over a decimetre to kilometre spatial range, result- ing in an integrated conceptual model for the controls on the distribution of the microbial deposits. The morphology, size and distribution of these deposits result mainly from environ- mental changes (i.e. seasonal to long-term water level fluctu- ations, particular geomorphological heritage, fault-induced processes, groundwater seepage) and have the potential to bring further insights into the reconstruction of paleoenvi- ronments and paleoclimatic changes through time. New ra- diocarbon ages obtained on each microbial macrofabric de- scribed in this study improve the chronological framework and question the lake level variations that are commonly as- sumed.

The deep siliciclastic reservoir (>3500 m) of the Upper Cape Hay Formation of the Bonaparte Basin (Petrel gas field, Petrel sub-basin, Permian) exhibits wide heterogeneity in porosity (2–26%) and permeability (0.001–2500 mD). To investigate this variability, 42 samples were taken from five wells drilled through this formation. Six facies were identified from core descriptions and microscopic study of the sandstones. These facies are typical of a tide-dominated estuary, and include (1) mud flat, (2) sand flat, (3) top of tidal sand bar, (4) middle of tidal sand bar, (5) bottom of tidal sand bar, and (6) outer estuary facies. The paragenetic sequence comprises the emplacement of early aggregates of ferrous clay mineral precursors, mechanical compaction, recrystallization of those ferrous clay mineral precursors to Fe-rich chlorite and crystallization of Fe-rich chlorite forming coatings around detrital grains, chemical compaction, development of quartz overgrowth, feldspar alteration, crystallization of dickite and illite-rich illite/smectite (I-S) mixed layers, and ferrous calcite cementation. The middle and top of the tidal bars generally exhibit the highest porosity (Φ > 10%) and permeability values (k > 1 mD). Feldspar alteration released silica and aluminium into the reservoir promoting the development of dickite and illite-rich I/S mixed layers, which tended to destroy porosity and permeability, as calcite cements and quartz overgrowths. Diagenetic chlorite coatings around detrital grains are restricted to the sand bar facies deposited at the end of the last third-order transgressive systems tract of the Cape Hay Formation. The formation and conservation of ferrous clay precursors seems to be possible in an estuarine environment where seawater and fresh water are mixed and tidal sand bars are formed. These ferrous clay precursors recrystallized to Fe-rich chlorite coating after mechanical compaction. These coatings inhibited quartz cementation and prove to be the key parameter behind good reservoir qualities.

Due to the multiple controlling factors involved, it is a challenging task to identify and quantify the processes influencing the distribution and heterogeneity of marine organic-rich rocks. To improve our understanding of these deposits, we model their burial history and stratigraphic evolution as well as processes linked to marine organic matter history throughout the Lower Jurassic in of the Northern Lusitanian Basin (Western Iberian Margin). This 15 Ma-long interval is modelled using 100-kyr time steps to simulate lithologies and organic matter heterogeneity as layers with a thickness of 2 to 5 m, depending on the sedimentation rate in the basin. The model is calibrated by well and outcrop data which provide structural and biostratigraphic constraints, as well as information on the depositional facies and geochemistry of the sediments. The results show that the presence of organic-rich intervals is linked to first-order variations in the basin geometry and sedimentation rates. Without considering any variation of primary productivity or oxygen content in surface sea-waters, the parameters of basin geometry and sedimentation rate are sufficient to predict the main characteristics of source rocks, i.e. their occurrence, thickness and mineralogy at the basin-scale. However, to fit the measured organic carbon contents, we need to take account of other parameters such as variations of primary productivity or changes in dissolved oxygen concentration. This article is protected by copyright. All rights reserved.

The Great Salt Lake, Utah, USA is a shallow, hypersaline, intracontinental lake hosting extensive microbial deposits. At a large spatial scale, the distribution of these deposits is driven by environmental and geodynamical factors (i.e. water-level fluctuations and a fault-related framework). A detailed mapping of the Buffalo Point area, in the north-western part of Antelope Island, indicates the presence of an anomalous concentration of microbial deposits dated ca. 5.8 ka BP and distributed along a lobe-shaped geometry. This uncommon microbial deposit geometry results from an extensive colonization of a conglomerate substrate exhibiting an accumulation of m-sized rounded Cambrian quartzite boulders. We suggest that this conglomerate substrate provides a stable nucleation point that promotes the development and preservation of the lobe-shaped microbial deposits. Microbial deposits may also have protected the conglomerate substrate from erosional processes and thereby increased the preservation potential of the lobe-shaped structure. Based on the characteristics of the conglomerate (e.g. grain size, texture) and its location (i.e. 200 m beyond the average shoreline), this lobe-shaped structure likely results from subaqueous debris or a hyperconcentrated density flow that transports sedimentary material from the Buffalo Point slopes downward to the shore. We estimate the age of the conglomerate deposition to be between 21 and 12 ka BP. The initiation of the flow may have been triggered by various mechanisms, but the existence of a major active normal fault in the vicinity of these deposits suggests that an earthquake could have destabilized the accumulated sediments and resulted in conglomerate emplacement. The catastrophic 15 ka BP Bonneville Flood, which led to a drop in the lake level (approximately 110 m), may also provide an explanation for the initiation of the flow.

The Great Salt Lake is a modern hypersaline system in which an extended modern and ancient microbial sedimentary system has developed. Detailed mapping based on aerial images and field observations can be used to identify non-random distribution patterns of microbial deposits, such as paleoshorelines associated with extensive polygons or fault-parallel alignments. Although it has been inferred that climatic changes controlling the lake level fluctuations explain the distribution of paleoshorelines and polygons, straight microbial deposit alignments may underline a normal fault system parallel to the Wasatch Front. This study is based on observations over a dm to km spatial range, resulting in an integrated conceptual model for the controls on the distribution of the microbial deposits. The morphology, size and distribution of these deposits result mainly from environmental changes (i.e. seasonal to long-term water level fluctuations, particular geomorphological heritage, fault-induced processes, groundwater seepage) and have the potential to bring further insights into the reconstruction of paleoenvironments and paleoclimatic changes through time. New radiocarbon ages obtained on each microbial macrofabrics described in this study improve the chronological framework and question the lake level variations that are commonly assumed.

Active, carbonate-mineralizing microbial mats flourish in a tropical, highly evaporative, marine-fed lagoonal network to the south of Cayo Coco Island (Cuba). Hypersaline conditions support the development of a complex sedimentary microbial ecosystem with diverse morphologies, a variable intensity of mineralization and a potential for preservation. In this study, the role of intrinsic (i.e. microbial) and extrinsic (i.e. physicochemical) controls on microbial mat development, mineralization and preservation was investigated. The network consists of lagoons, forming in the interdune depressions of a Pleistocene aeolian substratum; they developed due to a progressive increase in sea-level since the Holocene. The hydrological budget in the Cayo Coco lagoonal network changes from west to east, increasing the salinity. This change progressively excludes grazers and increases the saturation index of carbonate minerals, favouring the development and mineralization of microbial mats in the easternmost lagoons. Detailed mapping of the easternmost lagoon shows four zones with different flooding regimes. The microbial activity in the mats was recorded using light–dark shifts in conjunction with microelectrode O2 and HS- profiles. High rates of O2 production and consumption, in addition to substantial amounts of exopolymeric substances, are indicative of a potentially strong intrinsic control on mineralization. Seasonal, climate-driven water fluctuations are key for mat development, mineralization, morphology and distribution. Microbial mats show no mineralization in the permanently submersed zone, and moderate mineralization in zones with alternating immersion and exposure. It is suggested that mineralization is also driven by water-level fluctuations and evaporation. Mineralized mats are laminated and consist of alternating trapping and binding of grains and microbially-induced magnesium calcite and dolomite precipitation. The macrofabrics of the mats evolve from early colonizing Flat mats to complex Cerebroid or Terrace structures. The macrofabrics are influenced by the hydrodynamic regime: wind-driven waves inducing relief terraces in windward areas and flat morphologies on the leeward side of the lagoon. Other external drivers include: (i) storm events that either promote (for example, by bioclasts covering) or prevent (for example, by causing erosion) microbial mat preservation; and (ii) subsurface degassing, through mangrove roots and desiccation cracks covered by Flat mats (i.e. forming Hemispheroids and Cerebroidal structures). These findings provide in-depth insights into understanding fossil microbialite morphologies that formed in lagoonal settings. This article is protected by copyright. All rights reserved.

Microbialites of the Great Salt Lake (GSL) were mapped in 1938 but remained largely understudied since then. High resolution aerial imageries coupled to detailed field mapping of facies indicate a non-random distribution of microbialites along the GSL shoreline. Here, we discuss the main factors involved in the formation and preservation of microbialites since the lake reached its modern configuration (ca. 10 Ka). Microbialites and microbial mats were investigated on the western margin of Antelope Island at 0-3m waterdepth. Our results suggest an interplay of tectonic (active faults, groundwater seepage) and climatic (long term to seasonal lake level fluctuations) processes in the control of the distribution, size and morphologies of the microbial deposits. Our observations have direct implications for the prediction of microbialite distribution in sedimentary systems. Figure 1: Antelope field site. Repartition of microbialites in the western side of Antelope Island, influenced by the effect of faults, desiccation cracks and paleo-shorelines. 361

Evidence of several major unconformities in the Lower Palaeozoic succession in Iran lead to question the role of tectonic/eustatism/climate in terms of their formation. The studied Palaeozoic succession in the Kuh-e Surmeh Anticline is characterized by the preservation of two thin Ordovician and Lower Permian Formations separated by a large hiatus encompassing the Upper Ordovician up to the lowermost Permian. The Ordovician sequences were deposited in shallow shoreface to lower offshore environments and the Lower Permian corresponds to a wave-dominated estuarine system evolving to a delta system. These mainly clastic successions represent good reservoirs in the regional Palaeozoic petroleum system of the Iranian plate separated by a major unconformity. The local erosion of the Zakeen Fm., observed in the neighbouring areas close to the Kuh-e Surmeh anticline, can be used to specify the role of the different controlling factors. We show that the influence of regional tectonics (Hercynian Orogeny/extensional deformation) and climate (Hirnantian and Carboniferous glaciations) related to the late Ordovician and Late Carboniferous/Early Permian succession cannot be ruled out from the possible candidates participating in erosion, but local diapir doming seems to better explain part of this local intense erosion.

The late Messinian mixed carbonate-siliciclastic platforms of the Sorbas Basin, known as the Terminal Carbonate Complex, record significant changes in carbonate production and geometry. Their facies and stratigraphic architecture result from complex interactions between base-level fluctuations, evaporite deformation/dissolution and detrital inputs. A 3D quantitative approach (with DIONISOS software) is used to explore the basin-scale platform architecture and to quantify the carbonate production of the Terminal Carbonate Complex. The modelling strategy consists in integrating detailed 2D field-based transects and modern carbonate system parameters (e.g. carbonate production rates, bathymetric and hydrodynamic ranges of production). This approach limits user impact and so provides more objective output results. Tests are carried out on carbonate production rates, subsidence and evaporite deformation/dissolution. Numerical modelling provides accurate predictions of geometries, facies distributions and depositional sequence thicknesses, validated by field data. Comparative statistical testing of the field transects and of the various model outputs are used to discern the relative contribution of the parameters tested to the evolution of basin filling. The 3D visualisation and quantification of the main carbonate producers (ooids and microbialites) are discussed in terms of changes in base-level and detrital supply. This study demonstrates that base-level fluctuations have the greatest impact on the carbonate budget. Evaporite deformation/dissolution affects the type and amount of carbonate production, inducing a transition from an ooid- to microbialite-dominated system and also has a major effect on stratigraphic architecture by inducing the migration of depocentres. The numerical modelling results obtained using modern carbonate system parameters could also be applied to subsurface ooid-microbialite reservoirs, and the Terminal Carbonate Complex is a good analogue for such systems.

The present study brings new insights on the small-scale characteristics of relatively distal turbidite deposits (classically defined as depositional lobes or sheets) from the deep-water clastic slope system of the Annot Sandstones (Grès d'Annot), in the Trois Evêchés area. This “seismic scale” outcrop has been studied in detail to characterize sedimentary geometries and facies associations within and between sheet-like sandstone-bodies. In terms of facies, their internal architecture is dominated by numerous erosional features, variable scale cross-bedding and dewatering structures. Facies range from gravelly sandstones to fine-grained sandstones and their spatial distribution inside sandstone-bodies is provided, allowing us to discuss the spatial changes in flow behaviour. A detailed dataset, made of very high-resolution correlations in thick to thin-bedded sheet-like sandstone-bodies, enables to identify five main types of architectural elements, composed of both channelised and unchannelised depositional bodies. Gravelly-filled channel-elements, laterally stacked channel-elements (interpreted as being related to sinuous channel migrations) and wing-like channel-elements constitute the channelised sheet systems whereas tabular sandstones (sheets sensu stricto), prograding and/or dome-shaped units constitute layered sheet systems. Sinuous channels are represented by the association of erosive-based, laterally stacked bedsets and very coarse-grained, low-angle cross-bedded facies. These latter facies are interpreted as the result of flow deconfinement and overbanks above channel margins. The stratigraphic relationships between these architectural elements permit establishing a conceptual model of the distribution and organisation of the components. This distribution scheme is based on the longitudinal differentiation of gravity processes, which are directly linked with architectural element types. This model is composed of proximal confined units where channelization processes, amalgamation and bypass dominate, to more distal and less confined units where compensation, constructive structures and high deposition rates control sedimentary architectures. The Trois Evêchés sub-basin may be considered as a good example of a channelised lobe to unchannelised lobe transitional environment, which explains the lateral and vertical high variability during its infill. The widespread apparition of residual facies, such as cross-stratified sandstones as well as basal lags, suggests significant bypass. The appearance of both channelised and unchannelised units is consistent with this transitional environment. The provided detailed dataset in this paper displays the associated wide range of architectural elements in these settings, and highlights the significant heterogeneities of sand-rich confined turbidite sheet systems. These heterogeneities are under the seismic resolution and may influence fluid migration in sand-rich deep-water reservoirs.

This paper presents the sedimentary analysis of an exceptional Ordovician glacial tunnel valley in the eastern part of the Anti-Atlas. The valley infill comprises two major glacial erosion surfaces (striated pavements) each overlain by a fining-upward glacial unit. These units are composed of five distinct facies associations, recording the evolution from subglacial to proglacial environments, and an additional sixth facies association, overtopping the tunnel valley infill, and associated with post-glacial environments. The tunnel valley infill also records a transitional environment between the subglacial and proglacial settings, which is compared with the Antarctic ice-sheet margin. These three environments are defined by the position of the grounding line and the coupling line. The new proposed depositional model also differs from usual Ordovician depositional models in which the main tunnel valley infill is interpreted as essentially proglacial outwash deposits, in a range of glaciomarine to glaciofluvial environments. Overall, a substantial part of the valley infill (~ 50% of volume) was deposited in a subglacial setting. The sedimentary bodies could form potentially thick and laterally extended, although these were limited by the shape and extent of the subglacial accommodation space. Finally, the sedimentary record, when compared with regional analogues, also provides information for the palaeogeographic reconstruction of the Ordovician ice-sheet in this region.

The present paper investigates the outcrop-scale architecture of submarine gravity flow deposits in both moderately to highly confined settings. A significant dataset was gathered on the sand-rich, confined, siliciclastic turbidite system of the Annot Sandstone Formation (SE France). We focused on thick sedimentary bodies formed by depletive gravity flows, in relatively distal settings (sand-rich lobes). The purpose was to investigate the architectural variability of their small scale components (beds and bedsets) and consequently to evaluate the associated heterogeneities (facies changes, grain size, sedimentary features, sediment thickness, etc.). Therefore, we provide a set of detailed descriptions on well-exposed outcrops, through very high-resolution correlation panels built thanks to multiple vertical logs. These turbidite system end-members have rarely been described in such details and have often been considered as laterally continuous and homogeneous. On the basis of the hierarchical subdivisions of turbidite lobes, results show a high complexity in internal architecture from lobe bed to lobe scale whereas large scale (scenery-seismic scale) is characterized by apparently homogeneous tabular and very extensive sandbodies (thus corresponding to the classical definition of “sheet-sands”). This complexity implies heterogeneities not only in terms of sediment thickness changes (abrupt pinch-outs and variable scale compensation features), but also in terms of facies and associated sedimentary features distributions (tractive structures, internal scours with associated by-pass, freezed mudclast-rich layers, etc.). This facies variability is controlled by the interplay between internal erosive bedforms (i.e. large scours) and constructive bedforms, such as dome-shaped or laterally stacked bedsets that are related to local topographic controls and compensation processes. Variability then appears both laterally and vertically. Besides, results also showed small scale evidences of interactions between turbidite flows and basin floor, only expressed at lobe bed to lobe element scale, especially on the basin sides where paleoslope is subject to erosion and remobilisation and where flows can undergo significant changes. Concerning both seismic and reservoir implications, although those deposits are characterized by a sheet-like geometry, this high internal variability necessarily implies heterogeneities in terms of petrophysical characteristics (porosity, permeability) that may have a significant impact on fluid circulation. These elements lead us to reconsider both sedimentary processes involved in their building and also reservoir models that can be established on field analogues.

The worldwide known Annot Sandstone Formation has been considered as a reference of confined siliciclastic turbidite system. This formation crops out in SE France and represents the Upper Eocene to Lower Oligocene gravitary infill of complex foreland basins, developed in front of the Alpine thrusts. This system can be assimilated as a sand-rich turbidite ramp, sourced by multiple fan deltas leading to topographically complex sub-basins. Highly bypassing channelized systems dominate in the most proximal and most confined areas. They distally evolve to relatively less confined areas, in which sedimentary bodies appear to be more continuous and homogenous on a regional scale. These last architectural elements, defined as sheet-sands or depositional lobes, have been the focus of this study in poorly documented areas. From an important dataset made of very high resolution outcrop correlations (gathered mainly in the Trois Evêchés and Lauzanier sub-basins), we have quantified the complex distribution of sedimentary facies and structures, grain-size and key surfaces in sand-rich sheets. This was done to understand their variability from depositional event to architectural element scales and to better characterize dimensions and characteristics of their components. Six main types of architectural elements were defined, composed of both channelized and unchannelized elements. Channelized units show a high variability in terms of facies, geometry and patterns of infill that are related to multiple erosional and depositional processes, which will be discussed. We notably relate some evidences of sinuous channels, represented by lateral accretion deposits in the channel complex axis and by low angle cross-bedded facies. We interpret this particular facies as the result of flow deconfinement and overbanks above channel margins. The stratigraphic analysis of elementary objects allows us to propose a genetic model and a spatial distribution model of sheet-sand architectural elements. The latter provides their longitudinal repartition, from proximal vertically stacked internal channels to distal compensating sheets. Those features have not been accurately described in relatively distal sand-rich turbidite deposits so far and this high internal variability necessarily implies heterogeneities both in terms of petrophysical characteristics and reservoir connection. Besides, it also implies reconsidering sedimentary processes involved in these environments. Key Words: Annot Sandstones, Confined Systems, Reservoir Heterogeneities, Turbidite Sheet-sands.

Petrophysical properties were measured on oolitic limestone from the Oolithe Blanche formation (middle Jurassic) in the Paris Basin. Eighteen oriented blocks were collected from three outcrops regarding of the three main facies, oolitic shoal facies, tide dominated facies and prograding oolitic facies. We investigated the relationship between both compressional wave and electrical conductivity with different petrophysical properties such as porosity (water porosity and mercury porosity), permeability and capillarity imbibition. These have led us to relate the variations of petrophysical properties to several microstructural parameters, among them the microporous structure is the most important. Concerning P wave velocities, the general trend observed is as expected a decrease of velocities as the porosity increases but with a significant fluctuation of velocities for a given value of porosity. We have used two distinct effective medium approximations to describe the velocity variations: the self-consistent (SC) approximation and the cemented contact theory (CCT) but no unique model can simply explain our velocity behaviours. The main parameter that controls the P wave velocities is the distribution of microporosity inside oolites: for samples with velocities higher than 4 km/s the microporosity is mainly located in the rim of the ooids while for samples with velocity lower than 4 km/s the microporosity is homogeneous across the ooids. Acoustical fabrics, which are controlled by the facies, indicate that sedimentary textural components such as the amount of cement and arrangement of elements within the oolitic limestones (bioclasts, pellets and ooids) and their degree of connectivity may have some influences on the acoustic velocities.

We present new insights for the characterization of the petroleum system evolution in North Africa based on a review of the stratigraphic architecture description of some selected North African Palaeozoic basins. During Palaeozoic time, the Gondwana platform was divided into sub-basins bounded by structural highs. Most of the highs were inherited from north-south and SW-NE Pan-African crustal faults which were reactivated during the Palaeozoic and later, in the Austrian and Alpine tectonic phases. We studied the stratigraphic architecture of the Palaeozoic succession around four main highs showing a clear tectonic activity during the Palaeozoic sedimentation. The Gargaff Arch, in Libya, is a major SW-NE broad anticline which slowly grew up during the Cambrian and Ordovician and stopped rising during the Silurian. The activity resumed during Late Silurian and early Devonian and during the Late Devonian. The Tihemboka High is a north-south anticline in between Libya and Algeria. The uplift started during the Cambro-Ordovician then stopped during most of the Silurian. The activity resumed during the Late Silurian and continued until the Lower Carboniferous. The Ahara High, separating the Illizi and Berkine basins in Algeria, has continuously grown during the Cambro-Ordovician, stopped rising during the Silurian, and grew again continuously during the Devonian. The Bled El-Mass High is a part of the Azzel-Matti Ridge separating the Ahnet and Reggane basins in Algeria. The high mostly rose during the Cambro-Ordovician then subsided relatively less quickly than the surrounding basins during the Silurian and Devonian. The uplift timing and chronology of each palaeohigh partly controlled the petroleum systems of the surrounding basins. Topographic lows favoured the occurrence of anoxic conditions and the preservation of Lower Silurian and Frasnian source rocks. Complex progressive unconformities developed around the palaeohighs form potential complex tectonostratigraphic traps. Finally, hydrocarbons could have been trapped around the highs during pre-Hercynian times, preserving reservoir porosity from early silicification. Mixed stratigraphic-structural plays could then be present today around the highs. © Petroleum Geology Conferences Ltd. Published by the Geological Society, London.

La présente étude s'intègre dans le cadre du projet de recherche SHPCO2 (Simulation Haute performance pour le stockage géologique de CO2), financé par l'ANR(1) et en partenariat avec l'IFP, ENSMSE(2), LAGA(3), INRIA(4). Il est dédié au développement d'une plate-forme informatique haute performance pour la simulation numérique du stockage géologique de CO2 avec la volonté de se confronter à des modèles géologiques « réalistes » (domaines faillés, géométries caractéristiques des bassins, variations latérales de faciès) et donc aux difficultés numériques de simulations des écoulements que ces contraintes impliquent. Dans ce cadre, l'unité GBS (service GEO du BRGM) est chargée de réaliser des modèles géologiques 3D maillés de l'aquifère du Dogger du bassin parisien, à partir du logiciel de modélisation 3D Petrel. Ces modèles doivent restituer d'une part, la géométrie 3D de la zone d'étude et d'autre part les propriétés pétrophysiques des roches (porosité, perméabilité) qui caractérisent cet aquifère. Application d'une méthodologie intégrée La zone d'étude est choisie en fonction du fort potentiel de stockage géologique du CO2 (maximum d'épaisseur de l'aquifère, incluant des faciès sédimentaires poreux) et de la qualité des données de puits disponibles. Cette zone est localisée au sud-est de Paris sur une superficie de 100 km x 100 km. Les modèles réalisés dans ce secteur bénéficient du programme de valorisation des données de subsurface du BRGM. Ils intègrent notamment des données de puits variés : •les logs fondamentaux issus de la base BEPH(5), •les logs diagraphiques digitalisés et les logs pétrophysiques calculés, •des descriptions de carottes (3 puits de référence sur le secteur étudié), •des contraintes biostratigraphiques (Garcia, 1993) •mais aussi des travaux d'interprétations géologiques réalisés à l'échelle du bassin de Paris: •des données sismiques (lignes retraitées de la non-exclusive du bassin parisien, BRGM), •un schéma structural simplifié de la zone étudiée (failles de Bray, Malnoue, Vittel, Belou, Saint-Martin de Bossenay, Valpuiseaux) construit à partir de plusieurs horizons pointés sur les profils sismiques de l'étude non-exclusive (BRGM) •des cartes paléogéographiques des séquences de dépôts du Bajocien Supérieur au Callovien Terminal (Gaumet, 1997) Notre méthodologie est basée sur l'intégration de ces informations pour réaliser un modèle géologiquement cohérent et réaliste en se basant sur les principes de la stratigraphie séquentielle. Depuis le toit de l'Aalénien jusqu'à l'Oxfordien Inférieur, dix lignes isochrones (« Maximum Flooding Surfaces ») sont corrélées sur 70 puits (corrélations modifiées d'après Gaumet F., 1997). Ces lignes sont ensuite interpolées en 3D, recalées selon une grille sismique de référence et confrontées au schéma de failles disponible. On obtient ainsi un modèle géométrique 3D dont les couches sédimentaires sont limitées par des surfaces isochrones. Le modèle de faciès est construit en superposant les cartes paléogéographiques sur les surfaces de même âge et conditionnées par un paramétrage en électrofaciès aux puits (modifiées d'après Gaumet F., 1997). Cette méthode permet de visualiser la répartition des variations latérales et verticales de faciès, nous renseignant ainsi sur la connectivité (ou non) des réservoirs. Cette information est essentielle à la simulation des écoulements dans la zone considérée. Le remplissage des mailles en propriétés pétrophysiques est simulé de manière stochastique dans le modèle géométrique 3D et permet de proposer un modèle statique du Dogger du bassin parisien. La variabilité de la porosité est simulée au sein de chaque faciès à partir de logs de porosité calculée, continue sur toute la profondeur étudiée (25 logs). La perméabilité est quant à elle déterminée à partir de lois Phi-K, recalculées pour chaque faciès selon des valeurs de couples porosité-perméabilité mesurées sur carottes et extraites des rapports de fin de sondages. Valorisation scientifiques et techniques Aujourd'hui, ces modèles permettent aux ingénieurs-réservoir du projet SHPCO2 de tester concrètement leur capacité de calcul et de simulation d'injection et d'écoulement du CO2. Plusieurs modèles volumiques, de tailles, de précisions géologiques variées qui intègrent des failles majeures sont réalisés : •un modèle de la zone d'injection (20 x 20 km) qui contient 2 millions de mailles (500 x 500 m), •2 modèles de 700 000 et 10 millions de mailles (500 x 500 m) sur la totalité de la zone d'étude (100 x 100 km). Ces modèles rentrent dans le cadre de la valorisation des travaux de recherche, de mise à jour et du développement des connaissances géologiques du bassin de Paris. Ils permettent de proposer une visualisation 3D de la répartition spatio-temporelle des faciès réservoirs (barrière oolithique et/ou bioclastique et shoals granulaires) et des niveaux potentiellement imperméables qu'ils renferment (faciès de plate-forme externe/bassin profond), tout en intégrant les zones de couverture au toit et au mur des réservoirs. Perspectives Ce type de modèle, réalisé ici dans le cadre d'une étude liée au stockage géologique de CO2, pourrait également être utile pour la compréhension géologique du bassin (paléogéographique et géodynamique) et constitue un outil précieux pour l'aide à la décision dans différents domaines appliqués tels que l'exploration pétrolière, la gestion des aquifères, l'évaluation du potentiel géothermique, aussi bien à l'échelle du réservoir qu'à l'échelle du bassin et pourquoi pas jusqu'aux affleurements. Enfin, pour répondre à des objectifs d'exploration, on pourrait envisager d'intégrer des paramètres traduisant les phénomènes de diagénèse et de fracturation. Ces paramètres pourraient s'intégrer dans le cadre d'une étude géostatistique plus poussée, afin de traduire plus précisément la variabilité spatiale des faciès carbonatés en termes de porosité et perméabilité.

Because of its location in an active margin context, the sandrich Orinoco turbidite system is controlled morphologically and tectonically by the compressional structures of the Barbados prism, and as a consequence, the sedimentation system does not exhibit a classic fan geometry. The sea-floor geometry between the slope of the front of the Barbados prism and the slope of the Guyana margin induces the convergence of the turbidite channels toward the abyssal plain at the front of the Barbados accretionary prism. Also, whereas in most passive margins the turbidite systems are commonly organized upstream to downstream as canyon, then channel levee, then lobes, here, because of the control by active tectonics, the sedimentary system is organized as channel levee, then canyons, then channelized lobes. In shallow water, landward of the prism, the system has multiple sources with several distributaries, and progressively downward, the channel courses are more complex with frequent convergences or divergences that are emphasized by the effects of the undulating sea-floor morphologies. Erosional processes are almost absent in the upper part of the turbidite system shallower than 1500 m (4921 ft). Erosion along channels develops mostly between 2000 and 4000 m (6562 and 13,123 ft) of water depth, above the compressional structures of the Barbados prism. Incisions show irregular meandering and sinuous courses in the low-relief segments and less sinuous courses where channels incise the structures. Larger incisions (canyons) are 3 km (1.9 mi) wide and 300 m (984 ft) deep. The occurrence of different phases of successive incisions is responsible for the development of morphologically correlative terraces in both flanks of the canyons. This might be the consequence of two mechanisms: the tectonic activity of the deformation front characterized by progressive uplift and thrusting of recent sediments, and the superimposition of the cycles of the Orinoco turbidite system. Piston-core surveys have demonstrated that turbidite sediments moving through the channel and canyon system and deposited in the abyssal plain are mostly coarse sandy deposits covered by recent pelagic planktonic-rich sedimentation, which indicates that sand deposition slowed down during the postglacial sea level rise.

The Paris Basin appears to be appropriate in terms of CO2 capture and storage when considering both the amount of CO2 produced and the availability of depleted fields and deep saline aquifers. The purpose of this study is to investigate the petrophysical properties in relation with the sedimento-diagenetic environment of the "Oolithe Blanche" formation (Dogger), a deep saline aquifer considered as a potential candidate for CO2 storage. Regarding the scarcity of the core data in the Paris Basin, our investigation was firstly based on the study of field analogs in the south-east of the Paris Basin. The Oolithe Blanche Formation is composed of very shallow marine oolitic and bioclastic limestones, mainly grainstones. Three main sedimentological fabrics were defined: tide-dominated, wave-dominated (oolitic shoal) and prograding oolitic shoal fabrics. These three fabrics show subtle variations in composition (ooliths vs bioclasts) and in the distribution of the carbonate textures. The Oolithe Blanche Formation (Bathonian) is 70-80 meters thick and water salinity ranges from 1 to 4 g NaCl/l. 18 oriented blocks were collected on outcrops selected for the different sedimentological facies recognised in the formation: 9 blocks in the Bierry Lès Belles Fontaines quarry (Yonne), 5 blocks in the Ravières quarries (Côte d'Or) and 4 blocks in the Massangis quarry (Yonne). On each block, several petrophysical properties were measured on three perpendicular plugs: porosity, pore size distribution derived from mercury injection tests, permeability, capillary imbibition parameters, electrical conductivity and acoustic velocities. The sedimento-diagenetic fabric and microstructural analysis was defined by using optical microscope and scanning electron microscopy, coupled with image analysis on thin sections. The porosity measured using the water saturation triple weight method ranges from 6% to 34%. The permeability values are low, between 0.1 mD and 9 mD. The data from the mercury intrusion porosimetry show that the distribution of the pore-throat diameter is either unimodal (microporosity only) or bimodal (macro- and microporosity). The microporosity is intraparticles (intraoolitic) and macroporosity is interparticles and is related to processes of dolo-dedolomitization. Variations of the capillary imbibition parameters acoustic velocities and electric conductivity can be explained by the microstructure. So, our investigations show a relationship between the sedimento-diagenetic fabrics and the petrophysical fabrics. The variation of reservoir properties are mainly controlled by two microstructural characters : the cement (type, quantity and distribution) and the microporosity distribution inside the oolithe grains. Now, these latter results calibrated by studying field analogs must be extended to bore-hole data and core data to optimize our knowledge of the deep saline aquifer of the Oolithe Blanche. Without these investigations, it will be impossible to estimate the good or poor quality of this deep saline aquifer with respect to CO2 storage.

The Makran accretionary prism developed in the north-western part of the Indian Ocean as a consequence of the subduction of the Arabian Sea since Late Cretaceous times. It extends from southern Iran to the Baluchistan region of Pakistan where it joins the Chaman-Ornach-Nal left-lateral strike-slip fault systems to the north and the Owen Fracture Zone-Murray Ridge transtensional (right-lateral) system to the south in a complex triple junction near the city of Karachi. In September to October of 2004, we surveyed most of the accretionary complex off Pakistan with R/V Marion Dufresne. We achieved a nearly continuous bathymetric mapping of the prism and the subduction trench from 62°30′E to the triple junction near 62°30′E together with nearly 1000 km of seismic reflection (13 lines) and we took 18 piston cores in different geological settings. One of the main results is that the frontal part of the Makran accretionary prism is less two-dimensional than previously expected. We interpret the along-strike tectonic variation as a consequence of lateral variations in sediment deposition as well as a consequence of the under-thrusting of a series of basement highs and finally of the vicinity to the triple junction.

The Makran accretionary prism developed in the northwestern part of the Indian Ocean as a consequence of the subduction of the Arabian Sea since Late Cretaceous times. It extends from southern Iran to the Baluchistan region of Pakistan where it joins the Chaman-Ornach-Nal left-lateral strike-slip fault systems to the north and the Owen Fracture Zone-Murray Ridge transtensional (right-lateral) system to the south in a complex triple junction near the city of Karachi. In September to October of 2004, we surveyed most of the ac-cretionary complex off Pakistan with R/V Marion Dufresne. We achieved a nearly continuous bathymetric mapping of the prism and the subduction trench from 62°30'E to the triple junction near 65°30'E together with nearly 1000 km of seismic refl ection (13 lines) and we took 18 piston cores in diff erent geological settings. One of the main results is that the frontal part of the Makran accretionary prism is less two-dimensional than previously expected. We interpret the along-strike tectonic variation as a consequence of lateral variations in sediment deposition as well as a consequence of the under-thrusting of a series of basement highs and fi nally of the vicinity to the triple junction.

During Eocene-Oligocene times, the Gres d'Annot turbidite system (French Alps) was deposited in several tectonically controlled sub-basins, which were mainly fed from a southern major sediment source: the Corsica-Sardinia Massif. In order to establish regional correlations in the southern part of the basin, four kilometre-scale outcrop areas were studied in detail. From south to north these are: the St Antonin, Annot, Grand Coyer and Chalufy areas. The results are: (1) an updated chronostratigraphic framework, (2) a major SE-NW correlation panel, approximately 400m thick and 50 km long, parallel to palaeocurrent directions, within which all stratigraphic units are defined in terms of sedimentology and micropalaeontology and (3) some correlation panels at outcrop scale (around 5 km long and several hundred metres thick), within which all stratigraphic units are defined as before, but with the addition of a direct visual control on correlations, which enables the reconstruction of higher resolution geometry. Seven time-equivalent stratigraphic packages have been correlated from upstream to downstream, making use of micropalaeontologic constraints, and their geometric and facies evolution have been reconstructed through times. This evolution may be related to different stages in the basin deformation, induced by the east to west development of the Alpine foreland basin.

The Siluro-Devonian reservoirs of the Illizi-Berkine basins are some of the most prolific plays in Algeria and remain an attractive target for exploration. The main reservoir of this period is constituted by the F6 reservoir comprising the Ludlow / Pridoli stages of the Upper Silurian and the Lochkovian / Pragian stages of the Lower Devonian. During this period, the area can be subdivided in two sub basins: Southward, the Illizi basin, characterized by a low subsidence rate and limited eastward by the active Tihemboka High. Northward, the Berkine basin, widely open to the north and the East, was differentiated by a higher subsidence rate. The two sub basins were separated by the active uplift of the Ahara high acting as a barrier for the sediment supply coming from the south. The sedimentological environments and their sequential organization within this serie are illustrated in this poster with two wells: − GRA-1, located in the north east of the Illizi basin, shows the complete series: a wave dominated shoreface prograding northward on a low subsidence platform during the Upper Silurian, followed by an emersion period during the early Devonian terminated by the Emsian transgression. − Located in the north of the Berkine basin, RE-1, presents a thick accumulation of fluvial sandstone, characteristic of the higher subsidence rate of this basin during the Lower Devonian. For each well, a set of core photographies, core descriptions and wireline log signatures will be displayed to characterize the main facies associations. The dynamic evolution of the sedimentary environments will be summarized by a sequence stratigraphic interpretation and a correlation panel will be proposed.

The Triassic reservoir is one of the most prolific plays in Algeria since the beginring of the exploration. One of the exploration risk is linked to the distribution and heterogeneity of the reservoir: Triassic series were mostly deposited in continental to lagoonal depositional environments, and reservoirs mostly correspond to channelised fluvial sandstones interfingering with floodplain and sabbkha sediments.

Reservoir modeling of the Chaunoy field was performed by combining a sedimentological study, a sequence stratigraphic analysis, geostatistical simulations, and the analysis of production data and fluid-flow simulations. The reservoir corresponds to the distal part of a Middle Triassic alluvial fan system in the Paris basin (France), and is extremely heterogeneous and layered. The reservoir mostly consists of small ribbon channel deposits interbedded with flood-plain and lacustrine mudstones. The channel amalgamation rate varied with cyclic lake-level variations, which directly controlled the reservoir geometry. Within a base-level cycle, during periods of low accommodation, channels were amalgamated, forming highly heterogeneous sand sheets. As the accommodation increased, channels became progressively isolated within flood-plain mudstones. Finally, a lacustrine transgression deposited lacustrine mudstones and induced thin but widespread vertical permeability barriers across the field. As accommodation started to decrease, considerable pedogenetic alteration occurred, as shown by dolocretes and groundwater dolomites. Five cycles that constituted the reservoir layering framework were identified. Geostatistical simulations of lithotype distribution within these units were computed using the truncated random Gaussian function method. Horizontal and vertical lithotype proportion curves and variograms were calculated from well data. Because of the wide well spacing, it was not possible to determine the range of horizontal experimental variograms. Three lithotype realizations were simulated within a high-resolution grid to compare short, medium, and long correlation lengths. After assigning petrophysical properties to the lithotypes and upscaling, fluid-flow simulations were performed for the three realizations. The three flow simulations were then compared to the 10-yr production history of the field. The simulations showed quite a good match regardless of the variogram range, except in the northern part of the field, indicating a problem in the reservoir layering in this area. This relative insensibility of the flow simulation to the correlation length probably is due to the high net pay within the amalgamated channel reservoir units and to the high number of conditioning wells; however, the flow simulation performed with the longest correlation length showed the best fit with the production history.

The Cenomanian of the Arabian Peninsula comprises a carbonate platform setting with rudists, characterized by gradual lateral facies changes including the interfingering of carbonate reservoirs and source rocks. In order to be more predictive with regard to the distribution and the geometrical aspects of the reservoirs and source rocks, a high resolution sequence stratigraphic study has been carried out in the Adam Foothills of Northern Oman. Based on detailed field sections a correlation scheme covering a transect of 100 kilometers has been established. Three orders of stacked depositional sequences have been found based on the re-occurrence of facies. The distinct hierarchical organization of the depositional sequences in the Cenomanian, and the relative stability at that time of the Arabian Peninsula, implies a strong correlation potential and thus a broad regional similarity of the architecture of the petroleum systems at that time.

Lithofacies simulations using a high resolution 3D grid allow to enhance the geometries of internal heterogeneities of reservoirs. In this study the series simulated were the Ness formation, part of the Brent reservoir in the Dunbar field located in the Viking graben of the North Sea. Simulations results have been used to build the reservoir layering supporting the 3D grid used for reservoir engineering, and also used as a frame to study the effects of secondary diagenetic processes on petrophysical properties. The method used is based on a geostatistical study and integrates the following data: a geological model using sequence stratigraphic concepts to define lithofacies sequences and associated bounding surfaces, well data (cores and logs) used as database for geostatistical analysis and simulations, seismic data: a 3D seismic survey has been used to define the internal surfaces bounding the units, outcrop data: The Mesa Verde formation (Colorado, USA) has been used as an outcrop analog to calibrate geostatistical parameters for the simulations (horizontal range of the variograms). This study illustrates the capacity to use high resolution sequence stratigraphic concepts to improve the simulations of reservoirs when the lack of subsurface information reduce the accuracy of geostatistical analysis.

The Buntsandstein facies outcrops along a 12 km long, 150 m thick cuesta near Ayllon (Central Spain). The outcrop study is based on vertical sedimentological sections and continuous photo paneling, and demonstrates the presence of two depositional systems: an alluvial fan system in the lower half of the outcrop, and a straight and braided river system in the upper part of the outcrop. This overall evolution is probably related to base-level fall to base-level rise cycle, in which the reservoir architecture is linked to genetic units stacking pattern: during the base-level fall, the alluvial fan is prograding over sand flat and sandy alluvial plain deposits. Coarse and pebbly proximal sandsheets are interbedded with finer reddish distal deposits. Reservoirs units are laterally continuous, but silty alluvial plain deposits constitute vertical permeability barriers, during base-level stillstand, erosive channels and sandsheets are vertically amalgamated. Reservoirs units are laterally continuous and vertically connected, during the base-level rise, alluvial fan deposits are overlapped by straight river deposits. Reservoirs units are laterally connected but silty argillaceous alluvial plain horizons are preserved, at the end of the base-level rise, braided and straight river deposits are amalgamated. Fully connected, these reservoirs units have a very large lateral extension. A lithofacies database is compiled on this outcrop, and variograms, horizontal and vertical proportion curves are completed. Each stage of the base-level cycle is then quantitatively characterized by a specific heterogeneity pattern. The outcrop study will improve the prediction of reservoir extension and architecture in subsurface gas storage of the Paris basin.

Lithofacies simulations using a high resolution 3D grid allow to enhance the geometries of internal heterogeneities of reservoirs. In this study the series simulated were the Ness formation, part of the Brent reservoir in the Dunbar field located in the Viking graben of the North Sea. Simulations results have been used to build the reservoir layering supporting the 3D grid used for reservoir engineering, and also used as a frame to study the effects of secondary diagenetic processes on petrophysical properties. The method used is based on a geostatistical study and integrates the following data: a geological model using sequence stratigraphic concepts to define lithofacies sequences and associated bounding surfaces, well data (cores and logs) used as database for geostatistical analysis and simulations, seismic data: a 3D seismic survey has been used to define the internal surfaces bounding the units, outcrop data: The Mesa Verde formation (Colorado, USA) has been used as an outcrop analog to calibrate geostatistical parameters for the simulations (horizontal range of the variograms). This study illustrates the capacity to use high resolution sequence stratigraphic concepts to improve the simulations of reservoirs when the lack of subsurface information reduce the accuracy of geostatistical analysis.

The architecture of a major Transgressive Systems Tract has been studied in large outcrops and correlated with subsurface data in the Campanian series of the San Juan basin (Colorado and New Mexico). The TST shows a landward stepping pattern of seven units of shoreface and foreshore sediments. Each unit is 20 to 40m thick, and extends over 3 to 5 km in a dip direction, pinches out seaward in offshore shales and landward in coastal-plain sediments. Each unit is bounded by ravinement surfaces and has been deposited during a IVth order transgressive-regressive cycle. The transgressive phase was recorded by a ravinement surface, by transgressive deposits reworked seaward and by thick coastal-plain and estuarine sediments aggrading landward. The coastal-plain series was shale-rich, and showered marsh sediments alternating with coal beds and t-de-influenced meandering channels. The estuarine complexes were very thick (up to 20m) and consisted of stacked tidal channels, sand flats and wave-influenced shoals. The regressive phase consisted of a prograding wedge of storm-dominated shoreface, foreshore and coastal-plain sediments. This wedge showed a seaward stepping pattern of prograding units, related to the superimposition of several Vth order transgressive-regressive cycles. Each Vth order transgressive phase induced a landward shift of the facies associations and a minor ravinement surface. Then the progradation of a shingle of shoreface, foreshore and coastal-plain sediments occurred during the regressive phase. The turn around between the transgressive and regressive phase of the IVth order cycle was marked by the vertical stacking of thick shoreface, foreshore back-barrier and coastal-plain deposits.

Chaunoy field, the largest oil field of the Paris basin, is exploiting heterogeneous reservoirs deposited during the Triassic in a large alluvial fan/lacustrine complex. The construction of a realistic reservoir model is difficult in such a setting because of the highly complex architecture of single reservoir units. Geostatistical simulations therefore have been performed to take into account the reservoir heterogeneities in the fluid flow modeling. A first layering has been determined from sedimentological and sequence stratigraphic analysis. The series was deposited in an alluvial outer fan environment. A lower siliciclastic member shows four heterogeneous sand sheets (7 m thick), which have been correlated across the field. Each of them is made up of stacked single channel sequences. The sand sheets are separated by extensive lacustrine and flood plain mudstone layers acting as permeability barriers. An upper siliciclastic/dolomitic member has been divided into two units with porous conglomeratic channels interfingered with cemented lagoonal dolomites. Proportional curves in lithofacies have confirmed this layering, showing the continuity of the permeability barriers, and the variogram analysis has shown that the well spacing is larger than the channel width. Simulations in lithofacies have been performed with the Heresim software using three different variogram ranges (small, medium, and large values). Because a good correlation exists between the lithofacies and the petrophysical attributes, a transcription of the lithofacies simulations into petrophysical attributes therefore was easy and realistic. Scaling-up techniques have given fluid-flow models corresponding to the three correlation ranges. Comparison of the global results of the fluid flow simulations with the observed production history enabled us to choose the more relevant case. The the model using the selected correlation range helped determine optimum well spacing.