ABSTRACT

The Middle Permian to Lower Triassic Khuff Formation is one of the most important reservoir intervals in the Middle East. This study presents a sequence stratigraphic analysis of the Khuff Formation of a well-exposed outcrop in the Oman Mountains, which may provide a reference section for correlations across the entire Middle East. On the Saiq Plateau of the Al Jabal al-Akhdar, the Permian Upper Saiq Formation is time-equivalent to the Lower and Middle Khuff Formation (K5–K3 reservoir units in Oman). The Permian section is dominated by graded skeletal and peloidal packstones and cross-bedded grainstones with a diverse marine fauna. The Lower Mahil Member (Induan Stage), time-equivalent to the Upper Khuff Formation (K2–K1 reservoir units in Oman), is dominated by grainstones composed of microbially-coated intra-clasts and ooids. In general, the studied outcrop is characterized by a very high percentage of grain-dominated textures representing storm-dominated shoal to foreshoal deposits of a paleogeographically more distal portion of the Khuff carbonate ramp.

A sequence-stratigraphic analysis was carried out by integrating lithostratigraphic marker beds, facies cycles, bio- and chemostratigraphy. The investigated outcrop section was subdivided into six third-order sequences, named KS 6 to KS 1. KS 6–KS 5 are interpreted to correspond to the Murgabian to Midian (ca. Wordian to Capitanian) stages. KS 4-Lower KS 2 correspond to the Dzhulfian (Wuchiapingian) to Dorashamian (Changhsingian) stages. Upper KS 2–KS 1 represent the Triassic Induan stage. Each of the six sequences was further subdivided into fourth-order cycle sets and fifth-order cycles. The documentation of this outcrop may contribute to a better regional understanding of the Khuff Formation on the Arabian Platform.

INTRODUCTION

The Khuff Formation and its time-equivalents cover most of the Arabian Platform with hydrocarbon production in Bahrain, Iran, Oman, Qatar, Saudi Arabia and the United Arab Emirates (UAE), and exploration potential in Kuwait, Iraq and beyond (e.g. Sharland et al., 2001). It is a classical example of a flat epeiric carbonate ramp (e.g. Aigner and Dott, 1990; Al-Jallal, 1995) extending for more than 2,500 km in SE-NW strike-direction and more than 1,500 km in SW-NE dip-direction. This setting led to the formation of a layer-cake type platform with certain m-scale marker beds traceable for hundreds of km across the Arabian Platform (Al-Jallal, 1995).

Deposition of the Khuff Formation on the Arabian Plate started in the Mid-Permian and accompanied the detachment of the Cimmerian terranes from the Pangea Supercontinent (Konert et al., 2001). It was mostly deposited as post-rift cover on a passive continental margin of the newly forming Neo-Tethys Ocean during a period of relative tectonic quiescence and steady subsidence (Stampfli, 2000). The climate during Khuff time is interpreted as transitional from icehouse to greenhouse with sea-level oscillations of moderate wavelength and amplitude (Al-Jallal, 1995). The temperature regime was probably similar to the arid conditions of the present day Arabian Gulf (Strohmenger et al., 2002).

The most important facies in the Khuff hydrocarbon reservoirs are grainstones, variably composed of ooids, peloids or skeletal components. Their prediction and characterization is key to economic production from Khuff reservoirs. To investigate the stratigraphic architecture and composition of primary reservoir facies, an outcrop study was initiated at the Saiq Plateau, Al Jabal al-Akhdar in the Sultanate of Oman (Figures 1 and 2). There, the Saiq and Mahil formations (Permian – Triassic Akhdar Group) are exposed and form an accessible outcrop of Khuff Formation time-equivalent strata (Glennie, 2006). During the Late Permian, the study area was most likely located some 150 km away from the interpreted Arabian Platform margin (Figure 3). Paleogeographic maps place the study area just south of the Equator within or close to the unrestricted, open-marine carbonate shelf (Ziegler, 2001).

The present paper provides an initial description of facies types and facies sequences. It proposes a stratigraphic framework of Khuff Formation time-equivalent strata on the Saiq Plateau, Oman Mountains, based on various stratigraphic methods. Currently, further work is underway on sections in other parts of the Oman Mountains in a more regional perspective, on which will be reported separately.

GENERAL STRATIGRAPHY

The studied section is located on the Saiq Plateau (Al Jabal al-Akhdar) some 150 km southwest of Muscat. The entire succession in the mountain range exposes Proterozoic to Cretaceous strata (Figures 2 and 4). The Saiq and Mahil formations that are the focus of this study unconformably overlie Proterozoic beds (Mistal, Hajir and Mu’aydin formations) variably composed of metasediments and diamictites with granite boulders (Rabu et al., 1986). The Carboniferous – Lower Permian strata represented by the Al Khlata, Saiwan, and Gharif formations, encountered in subsurface of Oman (Osterloff et al., 2004), are not present in the Oman Mountains. The absence of the latter formations is attributed to the uplift during the “Hercynian Orogeny”. The Permian – Triassic strata in the Oman Mountains was subdivided into the Permian Saiq and Triassic Mahil formations by Glennie et al. (1974).

The Permian Saiq Formation consists of two members (Montenat et al., 1977, Rabu et al., 1986; Baud et al., 2001) (Figure 5):

  • (1) The Lower Saiq Member (Figure 6) is up to 17 m in thickness and mainly composed of reddish to yellowish siltstones and shales that may contain ostracods (Rabu et al., 1986).

  • (2) The Upper Saiq Member (Figure 4) has a thickness of 624 m. Its lower part, 120 m thick, consists of limestones while its upper part is completely dolomitized (Figure 5). The Upper Saiq Member contains a very rich Permian invertebrate and microfossil fauna including brachiopods, corals, bivalves, crinoids and gastropods. Stratigraphically important are large miliolid foraminifera.

The Triassic Mahil Formation conformably overlies the Permian strata. In contrast to adjacent sections in the Wadis of the Oman Mountains, its base is well marked by a whitish colored step in the slope profile on the Saiq Plateau (Figure 4). There, the Saiq/Mahil Boundary was placed by Rabu et al. (1986) (Figure 2). In this paper, we follow this delineation of the Saiq-Mahil Boundary and propose to further subdivide the Mahil Formation into three informal members (Figure 5):

  • (1) The Lower Mahil Member, 101 m in thickness, is completely dolomitized and dominated by abiotic components, notably ooids, peloids and microbially-coated litho-clasts. Very few fossils are present. This unit is capped by several layers with polymict breccia and soft-sediment deformation features that are possibly associated with thrusting (J. Mattner, personal communication, 2009).

  • (2) The Middle Mahil Member, time-equivalent to the Sudair Formation, is 260 m in thickness and completely dolomitized with few fossils. Colored claystones and argillaceous dolomites appear at the base of the Middle Mahil. They constitute the only clastic deposits in the Triassic section. The middle and upper part of the Middle Mahil Member consists of trough cross-bedded oolitic-peloidal grainstones. These are interbedded with burrowed and graded mudstones and wackestones. Top Middle Mahil is marked by a dissolution breccia and a paleo-karst horizon.

  • (3) The Upper Mahil Member, time-equivalent to the Jilh Formation, is 194 m thick and completely dolomitized with very few fossils. It consists of stacked microbial laminites, burrowed mudstones and intercalated trough cross-bedded oolitic-peloidal grainstones. Top Upper Mahil is marked by a red-colored dissolution breccia.

The Triassic section is unconformably overlain by a several tens of m-thick sequence of Lower and Middle Jurassic strata (Sahtan Group) (Figure 4). These beds are in turn unconformably overlain by Lower Cretaceous deposits (Kahmah Group) (Glennie, 2006).

METHODOLOGY

Four individual sections on the Saiq Plateau were logged sedimentologically (scale 1:100) using standardized logging sheets (Figure 2). They were tied together to one complete 725 m thick section, time-equivalent to the Khuff Formation, by using prominent lithostratigraphic marker beds traceable over the whole Saiq Plateau (Figure 7). Texture, lithology, sedimentary structures, components and grain- sizes were recorded. Facies were classified in a facies scheme (Table 2). The rock character was documented with numerous outcrop photographs.

An outcrop spectral gamma-ray survey was run in the outcrop using a portable spectral GR spectrometer (model GS-512, manufactured by Geofyzika, Czech Republic). The spectrometer is equipped with a 3×3″ NaI(TI) scintillation detector collecting natural gamma-radiation at the rock surface. Total counts were measured within a time interval of 15 seconds with a sample point spacing of 50 cm, producing separate logs for total bulk-GR, Uranium (U), Potassium (K) and Thorium (Th). To detect overall GR trends usable for stratigraphic correlations and sequence interpretation, a sampling time interval of 15 seconds was found to be sufficient after test measurements of 180, 90, 30 and 15 seconds. The concentrations of each of the elements are automatically calculated by the instrument and displayed in ppm (U, Th) and % (K). Test measurements also showed that virtually no variations are recorded in the Thorium-Log throughout the sections. Thus it was not plotted and analyzed.

Carbon and oxygen stable-isotope analyses were performed on 170 dolomitic samples at the University of Bochum. Sample material was carefully removed with a dental driller from hand specimens. About 1 mg of untreated sample powder of each sample was reacted with 100% H3PO4 at 70°C for 2 hours in an off-line vacuum line using a Finnigan Gasbench II. Carbon and oxygen isotope ratios of the generated CO2 were measured on a Finnigan Delta S mass spectrometer at the University of Bochum. For this reaction an acid fractionation factor of 1.00993 was used. Data was reported in the usual δ-notation in permille (‰) relative to the known isotope reference standard Vienna Peedee Belemnite Standard (V-PDB) (Coplen, 1994). The precision for the carbon (δ13C) and oxygen (δ18O) isotopic composition of the dolomite is better than 0.08‰ and 0.14‰, respectively. Data was not corrected for differential fractionation of calcite and dolomite during the dissolution by phosphoric acid (Land, 1980) as rock samples were only collected from the dolomitized part of the Saiq Plateau section.

A total of 236 thin sections were manufactured from rock samples collected in the field and analyzed biostratigraphically. Facies types were analyzed in thin sections and interpreted in terms of vertical facies successions.

Logged sections were digitized with WellCAD. Interpreted facies and sequence stratigraphic data were compared to similar studies (Insalaco et al., 2006; Maurer et al., 2009) to tie the Saiq/Mahil Formations into a regional framework.

FACIES ANALYSIS AND INTERPRETATION

Eight principal lithofacies types (LFT) and their sub-types were distinguished in the investigated section (Table 2).

The Permian Upper Saiq Member is dominated by low-angle laminated to trough cross-bedded pack-to grainstones (Figures 10 and 11). These are mainly composed of peloids and bio-clasts of a diverse marine fauna. The beds represent storm-dominated foreshoal to shoal deposits. Interbedded are burrowed/rooted mud- to wackestones (Figure 8) and microbial laminites (Figure 9) representing a more protected backshoal environment.

The Triassic Lower Mahil Member is dominated by cross-bedded peloidal-oolitic grainstones (Figure 11) and graded storm beds (Figure 10) with a dramatically reduced fossil content. These formed in a foreshoal to shoal environment.

Lithofacies Types

Burrowed to Vertically Rooted Mudstone to Wackestone

Description: This facies type consists of light gray, whitish weathered dolomite showing intense bioturbation with normal grading in places (Figures 8a, b and d). Burrows cause a cloudy appearance of the rock texture with particle- and mud-rich patches. In some instances, ichnofabrics such as spreiten structures and burrows (e.g. Diplocraterion, Thalassinoides) can be identified. In some cases vertical shafts occur. Grain-size ranges from siltite to very fine-grained arenite. The mudstone to wackestone is poorly to moderately well sorted. Peloids are the dominant grain types with some intermixed intra-clasts, gastropod shells and skeletal debris. The microfauna is dominated by foraminifera (staffellids, small miliolids and few biseriamminids) as well as dasycladacean algae. Bed thickness of this facies type varies between a few cm to several dm. Its main distribution is within the Permian part of the section.

Interpretation: Burrowed to vertically rooted mudstones to wackestones are interpreted as deposits of a low-energy shallow, subtidal setting of a restricted lagoon or backshoal environment. Shallow and sheltered water possibly caused by the baffling action of sediment shoals is indicated by the abundance of peloids and by burrowing, which points to an intense activity of sediment feeding organism within a low-energy, oxidized setting. Bioturbation is not always discriminable from rooting. Root traces, very low fossil content and diversity are important criteria to place this muddy facies in a landward position and distinguish it from muddy offshoal deposits. This facies type is probably equivalent to facies type F5 of Insalaco et al. (2006).

Bioturbated Mudstone to Wackestone

Description: This facies type consists of gray-beige to dark-bluish gray weathered limestone or dolomite with occasional wavy bedding (Figures 8c, e and f). These show a variety of intense undefined bioturbation and a minor amount of preserved burrows, particularly feeding and spreiten structures (Zoophycus, Thalassinoides and Chondrites). Mudstones and wackestones are mainly siltites to very finely-grained arenites, moderately well to poorly sorted. The dm to a few dm thick beds contain undefined skeletal debris, peloids and rarely bivalve or brachiopod shells. Green algae, particularly gymnocodiacens, separates this facies type from protected lagoonal deposits. It mainly occurs within the lowermost part of the Permian section.

Interpretation: Bioturbated mudstones and wackestones are interpreted as open-marine deposits of a low-energy outer ramp setting (offshoal). This environment is characterized by strongly varying oxygen levels, reduced circulation and low sedimentation rates. The dominance of lutitic components indicates low-energy conditions with background sedimentation. Changing Eh-conditions are reflected by changing color and ichnofabrics which highlight strong variations between well oxygenated outer ramp environments (Zoophycus, Thalassinoides) and less well oxygenated conditions (Chondrites). This facies type may correspond to facies type F11 of Insalaco et al. (2006).

Microbial Laminites

Description: This facies type is made up of light gray to whitish weathered dolo-mudstone with mm to cm flat, crinkly to wavy laminations (Figure 9). Faint normal grading occurs together with desiccation cracks and rare cm-scale tepee structures. Mat-like structures are interbedded with thin streaks of peloidal packstones to grainstones and reworked clasts (flat pebble conglomerate). Bioturbation is weak to absent. The main components are peloids, undefined skeletal debris, reworked laminite clasts and rarely ooids. Biseriamminid foraminifera are characteristic constituents of this microbial boundstone facies. The boundstone is poorly sorted with variable grain-sizes ranging from lutite to medium-size arenites. The up to a few dm-thick beds are encountered in almost the entire outcrop section.

Interpretation: Microbial laminites are interpreted as intertidal microbial mudflat deposits exposed to periodical storm reworking. Bioturbation is generally weak or absent leading to the preservation of laminated structures. This facies can be best understood as intercalations of thinly bedded detritus stabilized by microbial mats (e.g. cyanobacterial filaments). The micro-graded laminae are the product of episodically occurring turbulence due to major storms or spring tides. Such events lead to reworking of partly lithified microbial laminites. This facies type is equivalent to facies type F6 of Insalaco et al. (2006).

Graded Packstone to Mudstone

Description: This light gray to blackish colored weathered facies type shows a variety of physical sedimentary features (Figures 10a, b, c and e). Mostly a thin interbedding of grainy and muddy beds on a cm- to dm-scale is observed. Sedimentary structures include scoured bases, low-angle to wavy lamination, hummocky cross-stratification (HCS), normal grading, bed amalgamation and muddy (bioturbated) tops. Post-event bioturbation includes spreiten traces (Teichnichus), grazing and crawling traces, vertical burrows (Skolithos) and escape structures. Grain-size varies between siltite to fine rudite. Sorting is generally poor. Main components are peloids, intra-clasts, skeletal debris and bivalve or brachiopod shells. The diversity of foraminifera is low with rare biseriamminids, miliolids and staffellids. The cm to several dm-thick beds are present throughout the investigated section. The thickest graded packstones to grainstones occur in the lowermost part of the Mahil Formation.

This facies type is further subdivided into two subtypes based on the amount of bioturbation and texture. Graded wackestones to mudstones are intensely bioturbated and finely-grained (siltite). Graded packstones to wackestones only show rare bioturbated tops and are coarsely-grained (arenite to rudite).

Interpretation: This facies type shows diagnostic signatures of storm deposition. It is interpreted as moderately low to locally high-energy storm beds, deposited above storm wave base (SWB). Storm sheets represent deposits of an outer ramp, foreshoal environment. Reworked intra-clasts and sharp erosive bases point to temporary high-energy storm events causing reworking of lithified sediment. Rare oscillation rippled tops suggest the influence of wave activity. Bored and bioturbated tops are indicators for sediment starvation and discontinuous sedimentation. There is no description of an equivalent facies type in Insalaco et al. (2006).

Intra-clastic Grainstone/Rudstone

Description: The light gray to yellowish weathered dolo-grainstone/rudstone appear as massive beds in outcrops (Figures 11e and f). Facies is low-angle cross-laminated and shows scoured bases and normal grading. Rounded to elongated black intra-clasts, composed of grainy material, are aligned along foresets or concentrated at the bases of cross-beds. Bioturbated tops occur occasionally. Grain size usually ranges from fine arenite to coarse rudite. The poorly to moderately well sorted grainstone/rudstone contains abundant micritized to microbially coated grainy lithoclasts, commonly peloids, oncoidal flat pebbles and rarely skeletal debris and bivalve or brachiopod shells. Foraminifera are not observed. Bed thickness ranges from few dm to several dm. This facies is common especially within the Triassic part of section.

Interpretation: Intra-clastic grainstones/rudstones are interpreted as high- to moderate energy proximal shoal deposits. High-energy storm events cause reworking of sediment and formation of intra-clasts, resting upon sharp erosive bases. Microbial stabilization and micritization around lithoclasts represent periods of sediment starvation and discontinuous sedimentation. Locally oncoids and grapestone are developed. This facies type may be the equivalent of facies type F16 in Insalaco et al. (2006).

Poorly-sorted Packstone to Grainstone

This facies is subdivided into 2 sub-types (a, b) based on different characteristic particles and sedimentary structures:

(a) Peloidal Packstone to Grainstone

Description: This facies includes beige to blackish weathered dolomite with a packstone to grainstone texture. The beige graded packstones to grainstones show common dm-scale low-angle lamination, massive to faint normal grading, HCS, scoured bases as well as top down bioturbation. Dark to black packstones are intensly mottled and faintly graded. The moderately well to poorly sorted rocks are finely-grained arenites to fine rudites, commonly several dm thick. The main components are peloids and bioclasts dominated by brachiopod and bivalve shells as well as rare corals, lithoclasts, crinoids and gastropods. The foraminiferal assemblage is moderately diverse and dominated by large miliolids, nodosariids, biseriamminids and staffellids. This facies type is very abundant in the Upper Permian section.

Interpretation: Peloid-rich packstones to grainstones are interpreted as foreshoal to shoal margin deposits of a moderate-energy, open-marine environment representing a transition between mid- to outer ramp. Rapid sedimentation is due to high-energy storm events cause the development of erosive bases and normal grading of rudite components. Common amalgamation and poor sorting point to slightly reduced accommodation. The scarcity of ooids and micritic envelopes and the abundance of crinoids and corals suggest open-marine steno-haline conditions. This facies type is most likely equivalent to facies type F10 of Insalaco et al. (2006).

(b) Bioclastic Packstone to Grainstone

Description: The bluish to dark gray weathered dolomite or limestone facies consists of packstones, less commonly grainstones (Figure 10d). Sedimentary structures include normal grading, erosive bases, scour surfaces, low-angle lamination, umbrella structures and bed amalgamation. Packstones are intensly mottled. The better sorted graded packstones to grainstones show top down bioturbation in places. Burrowing is visible due to a ‘cloudy’ appearance of this facies with particle- and mud-rich patches. Grain-size is fine- to coarse arenite. Packstones to grainstones are moderately-well to poorly sorted. Main components are bioclasts dominated by undefined skeletal debris, corals, crinoids, gastropods and fusulinid foraminifera. Foraminifera are common and dominated by paleotextulariids, fusulinids and nodosariids. The several dm-thick beds exclusively occur within the Permian part of the section.

Interpretation: Bioclastic packstones to grainstones are interpreted as moderate-energy deposits. Grainy textures and cross-bedding indicate storm reworking. The differences in texture between different sets point to variations in energy-levels. The diverse fossil assemblage including larger benthic foraminifera and crinoids suggests fully marine conditions. Based on the relative abundance of normal marine fauna, this facies type represents a near-shoal subtidal setting, either foreshoal or backshoal. This facies type is probably analogous to facies type F8 of Insalaco et al. (2006).

Well-sorted Grainstone

Based on compositional variations, this well-sorted, cross-bedded facies type is subdivided into either peloidal (a) or oolitic (b) grainstone:

(a) Well-sorted Peloidal Grainstone

Description: Dolo-grainstones are light beige in color (Figures 11c and d). They are trough cross-bedded with a sharp erosive base, overlain in places by intra-clasts. Locally sets of coarse skeletal material such as coral rudstone layers occur along scour surfaces. Bioturbated tops are rare. The very well sorted grainstones are fine to coarse arenitic. They contain abundant peloids, rarely ooids and bioclasts such as corals, bivalve and brachiopod shells. Foraminifera are common and show a low to moderate diversity with large miliolids, nodosariids, and biseriamminids. This facies type forms dm-thick beds throughout the entire investigated section. Thick intervals of massive peloidal grainstones are especially recognized within the Upper Permian section.

Interpretation: Cross-bedded, well-sorted peloidal grainstones are interpreted as deposits of proximal amalgamated incipient shoal or bar complexes within the high- to moderate-energy mid ramp. The diverse open-marine fauna suggests a position on the seaward fringe of the shoal. Well-sorted beds and erosive features like scoured bases, common amalgamation and missing bioturbation indicate extensive reworking in a low accommodation setting. Although there is no direct equivalent, this facies type may correlate to facies type F10 of Insalaco et al. (2006).

(b) Well-sorted Oolitic Grainstone

Description: Oolitic dolo-grainstones are light gray to white in color and commonly show planar to trough cross-bedding and erosive bases (Figures 11a and b). Microbial laminites are rarely observed on top of the grainstone. Grain-size of the well to moderately well sorted deposit ranges from fine to coarse arenite. This facies type comprises abundant ooids and coated grains, whereas peloids and clasts are only rarely observed. It occurs as dm-thick laminae sets to beds mainly towards the top of the Triassic section.

Interpretation: Cross-bedded, well-sorted oolitic grainstones are interpreted as shoal or bar complex deposits. They represent high-energy mid ramp shoal bodies. Components and sedimentary structures point to frequent high-energy conditions in a low accommodation setting. Grainstones with similar features have been described from several modern environments and are produced in shallow water (< 5m) or detached shoal settings, for instance in the Arabian Gulf (e.g. Purser and Seibold, 1973). Thin patches of grainstones within lagoonal sediments might represent spillover lobes or tidal channel sands induced by storm surges. This facies type is equivalent to facies type F9 of Insalaco et al. (2006).

Skeletal Floatstone

Description: These dm-thick beds are dark gray to black dolomites or limestones containing abundant corals and compound coral heads, partly in life position (Figures 10f and g). These are interbedded with diverse fossil fragments in a muddy matrix. Common biota are allochthonous solitary rugose corals, colonial rugose and tabulate corals, gastropods, brachiopods (e.g. Productus, Spiriferina, Pentamerus, Richthofenia, Terebratula), bivalves and crinoids as well as rare undefined skeletal debris and peloids. Foraminifera are very common. Particularly paleotextulariids, fusulinids and nodosariids were recorded. Grain-size in floatstone ranges from fine to coarse rudite. Sorting is generally very poor. The facies exclusively occurs in the Permian part of the section.

Interpretation: The poor sorting of this facies type, the muddy matrix and the high amount of articulated open-marine fauna point to an allochthonous to parautochthonous origin of the skeletal components and a relatively low-energy setting. Coral floatstones are interpreted as deposits of open-marine coral patches originated as outer ramp deposits around or below SWB. Floatstones with coarse shell and crinoid debris possibly represent storm-reworked deposits. There is no facies equivalent interpreted by Insalaco et al. (2006).

Lithofacies Associations

Facies types were grouped into lithofacies associations (LFA) based on their interpreted depositional environment (Table 3). The LFA-scheme is based on the Saiq Plateau outcrop and Khuff cores across the Middle East. Eight lithofacies associations and their depositional environment are defined:

LFA 1: Sabkha / Salina (evaporitic supratidal setting);

LFA 2: Coastal marsh (intertidal setting);

LFA 3: Tidal flat (intertidal setting);

LFA 4: Backshoal (low-energy, shallow subtidal lagoon);

LFA 5: Shoal (high-energy, shallow subtidal setting above fair-weather wave base);

LFA 6: Beach / Barrier island (high-energy subaerial setting);

LFA 7: Foreshoal (moderate-energy, deeper subtidal setting between fair-weather wave base and storm wave base);

LFA 8: Offshoal to basinal (low-energy, deep subtidal setting below storm wave base).

The lithofacies types observed on the Saiq Plateau were classified in terms of these lithofacies associations and depositional environments. Within the section, mostly open-marine facies associations occur (LFA’s 5, 7 and 8). LFA’s 3 and 4 were rarely interpreted. Although LFA’s 1, 2 and 6 are absent in the studied section, they were included in the LFA scheme to link outcrop to subsurface sections.

Depositional Model

A conceptual 3-D depositional model of the Khuff carbonate ramp is presented in Figure 12. Most lithofacies types found in the outcrop section represent the foreshoal, storm-dominated section of a carbonate ramp. Fully open-marine conditions are present in large parts of the succession.

The Saiq Plateau revealed a number of features that are unlike producing Khuff reservoirs but are important for the regional understanding of the Khuff. For instance, the Saiq Plateau succession contains a higher percentage of open-marine facies types than most other documented Khuff sections (e.g. Al-Jallal, 1995; Strohmenger et al., 2002; Vaslet et al., 2005; Insalaco et al., 2006; Maurer et al., 2009). These grainy textures are mainly composed of skeletal and peloidal components with limited occurrence of ooids and indicate a high-energy, shallow-marine setting. Ubiquitously storm beds, interpreted as amalgamated tempestites (Aigner, 1985) and abundant open-marine fossils suggest deposition above the SWB. These are more abundant than trough cross-bedded grainstones. Coral patches are common particularly at the base and top of the Permian section. These together with incipient shoals and skeletal-peloidal bars occupy the foreshoal setting. There is a general scarcity of subaerial exposure features (e.g. dissolution breccias). Striking feature of the section is the complete absence of structures that would indicate the presence of evaporites (e.g. stratabound dissolution vugs or cavities). This interpretation fits well with the assumed paleogeographic location within the open-marine carbonate shelf near the edge of the Arabian platform (Figure 3) (Ziegler, 2001).

VERTICAL STACKING OF FACIES

Facies types are stacked into facies cycles of four hierarchies. The terminology to describe facies cycles was adopted from Kerans and Tinker (1997). Accordingly, cycles (fifth-order) are stacked to form cycle sets (fourth-order) that are arranged in sequences (third-order), which form the overall Khuff supersequence (second-order).

Fifth-order Cycles

Facies types are stacked to fifth-order transgressive-regressive cycles or parasequences (van Wagoner et al., 1990), approximately 2–8 m in thickness. These cycles represent the finest scale of cyclicity within the studied section. A time estimation based on their number within the outcrop section may suggest that they probably record a 100,000 year Milankovitch signal (after Vail et al., 1977) (Table 4).

The most significant characteristics of these cycles are regular changes in texture, grain-size and fossil content. Differences in color and weathering angle tend to mimic rock textures. They provide a useful proxy for the identification of facies cycles in the outcrop. Brown-gray rocks with steep weathering angles of 80–90° tend to represent peloidal-oolitic grainstones (LFA 5). In contrast, dark-gray to black peloidal mudstones to packstones (LFA 7 and 8) have weathering angles of 60–80°. They contain the most open-marine fauna. Light colored mudstones (LFA 3 and 4) and wackestones have weathering angles of 30–60°.

Small-scale cycles can be subdivided into a transgressive and regressive hemicycle. They are separated by turnarounds, i.e. zones of maximum and minimum accommodation (Cross and Lessenger, 1998). The facies stacking pattern of small-scale cycles varies along the depositional gradient and bathymetry as a function of lateral shifts in accommodation space. As individual cycle types represent end-members related to a certain depositional environment, transitions between cycle types are possible. Similar to the hierarchical subdivision of facies associations and facies types, we recognize four general cycle motifs. These are built by a number of more specific cycle types. Examples of individual cycle types are shown in Figures 13 to 17.

Foreshoal Cycle Motif

Description: This asymmetric cycle motif is 2–5 m thick. It is variably composed of stacked open-marine, normal graded mid- to outer-ramp facies types (LFA 7 and LFA 8) (Figure 13). The thinner lower part, later interpreted as transgressive hemicycle, usually consists of bioturbated mudstone to wackestone with various ichnofabrics or skeletal floatstone. The thicker upper part, later interpreted as regressive hemicycle, consists of graded, commonly bioturbated packstone to mudstone and bioclastic packstone showing erosive bases, scour surfaces and HCS. These may turn upwards into massive, low-angle laminated peloidal packstone to grainstone.

Interpretation: Dark burrowed mudstone and coral-rich skeletal floatstone at the base of the cycle motif indicate fully open-marine conditions and maximum relative water depth in a low-energy depositional environment around SWB. The rise to fall turnaround (zone of maximum accommodation) is defined at intervals with a maximum percentage of open-marine components and fossils. The facies stacking pattern of the regressive hemicycle suggests a transition from the outer ramp to distal to proximal foreshoal. This typical shallowing-upward trend is associated with an increase of energy indicators, grain-size and sorting. Small-scale cycles of the foreshoal cycle motif are most common in the transgressive to early regressive part of the composite sequences. They occur in the lower part of the Upper Saiq Member as well as in the Triassic portion of the investigated section.

Shoal Margin Cycle Motif

Description: Cycles of the shoal margin motif are 3–5 m thick and asymmetric. They can be related to a proximal shoal to shoal fringe setting (LFA 5–LFA 7) (Figure 14). The lower part usually consists of bioturbated bioclastic packstone with erosive bases and locally interclasts at the bases of the beds. Intercalated are graded storm beds with HCS or bioturbated mudstone to wackestone. The upper part mainly starts with amalgamated, graded bioclastic packstone to grainstone with erosive bases and frequent scouring. These may be overlain by thicker, low-angle laminated peloidal packstone to grainstone or coarsely-grained intra-clastic grainstone/rudstone generally poor in fossil content. They show an increase in sorting compared to the lower, bioturbated bioclastic packstone.

Interpretation: This cycle motif represents the transition from a storm-dominated foreshoal environment to a higher-energy, shallower shoal fringe setting. Common bed amalgamation and scouring reflects lower accommodation, accumulations of peloids and low-angle lamination suggest high-energy conditions and sediment input from the adjacent shoal complex. The fall-to-rise turnaround or regressive maximum occurs at the top of the packstone to grainstone that represent the time of maximum depositional energy and minimum accommodation. During the transgressive hemicycle (lower part), deeper-water foreshoal conditions are indicated by a higher degree of bioturbation and abundant open-marine bio-clasts. Storm influence is interpreted from the hummocky-cross stratified graded beds. Abundant shoal margin cycles types are present in the upper part of the Saiq Formation.

Shoal Cycle Motif

Description: Individual shoal cycles are commonly 1–5 m thick and strongly asymmetric. They are mainly composed of mid ramp facies types (LFA 5 and LFA 6) (Figure 15). The very thin lower part of the cycle motif is represented by sheets of open-marine facies types such as skeletal floatstone and graded storm beds. The thicker upper part starts with thick beds of bioclastic packstone to grainstone, thin layers of scoured graded beds or graded low-angle laminated peloidal packstone to grainstone. Upward these sediments may pass into massive amalgamated intra-clastic grainstone/rudstone with micritic envelopes and coated grains. In most cases, facies grade into well sorted and cross-bedded peloidal or oolitic grainstone. In some cases the grainstone is overlain by microbial laminites (Figure 16).

Interpretation: The shallowing-upward trend is associated with an increase of energy, sorting and the change from skeletal to peloidal or oolitic grains. The upward increase in non-skeletal grains depicts an increase in depositional energy with its maximum towards the shoreline fringing shoal belt. This cyclicity style is interpreted as a prograding shoal body (cf. Aigner, 1985). The fall-to-rise turnaround or regressive maximum mainly occurs at the top of the peloidal-oolitic grainstone that represents the time of maximum depositional energy. Cycle caps are rarely composed of microbial laminites that represent further shoaling into a lower accommodation, possibly intertidal setting. During the transgressive part, flooding is represented by outer ramp facies types (LFA 7).

The shoal cycle motif is most common during times of enhanced depositional energy and tends to develop during early transgressive and middle to late regressive part of the composite sequences. Specific cycle types of this motif are observed throughout the investigated section.

Shoal- to-Backshoal Cycle Motif

Description: This cycle motif is usually 2–4 m thick and consists of stacked mid- to inner ramp facies types (LFA 3 – LFA 5) (Figure 17). The lower part of the cycle motif, up to several m-thick, starts with a sharp erosive base, possibly interclast covered (flakestones) that pass into peloidal-bioclastic, poorly sorted packstone to grainstone or high-angle cross-bedded peloidal-oolitic grainstone. The dm-thick upper part of the cycle motif consists of burrowed to vertically rooted mudstone to wackestone. Microbial laminites with occasional tepee structures are observed at the very top.

Interpretation: This stack of mid- to inner ramp facies types indicates restricted conditions and lower accommodation. The basic theme of this cycle motif is the migration of low-energy lagoonal facies over transgressive skeletal-peloidal shoal facies types. The top of the regressive hemicycle is marked by extensive microbial mats. This facies indicates shallow water and calm sedimentation conditions. Cycles of the shoal- to backshoal cycle motif typically occur around peak regression of composite (third-order) sequences and were mainly documented in the lower part of the Permian section (Figure 17).

Fourth-order Cycle Sets

Stacks of 3 to 10 cycles form transgressive-regressive cycle sets or parasquence sets (van Wagoner et al., 1990), some 5–25 m in thickness. In large parts of the outcrop, these cycle sets are the most obvious and easiest to recognize order of cyclicity. The estimated average duration of each of these cycle sets is about 400,000 years, assuming an overall duration of the Khuff of around 17.7 My (Table 4). Thus they are classified as fourth-order cycles, which may record a Milankovitch signal (after Vail et al., 1977). They display the lateral movement of facies associations or belts according to Walther’s Law, most apparent by the repeated retrogradation and progradation of shoal complexes (LFA 5). The studied section is subdivided into 36 of those cycle sets, termed Khuff Cycle Sets (KCS) 1.1 to 6.4 from top to bottom (Table 5).

Within the cycle sets, the shoal LFA shows the lowest average GR values (19.5 API) with a narrow range of 7 API (Figure 18b). The tidal flat LFA shows the highest average GR values of 28.8 API. The presence of these indicator facies associations was used to objectively calibrate the LFAs within each cycle set motif to the measured GR logs. Offshoal (LFA 8), foreshoal (LFA 7) and shoal (LFA 5) facies associations generally show lower average GR values compared to muddy backshoal (LFA 4) and tidal flat (LFA 3) facies associations. However, it may be impossible to differentiate between offshoal (LFA 8), foreshoal (LFA 7) and grainy backshoal (LFA 4) facies associations based on GR-log data alone. Four principal cycle set motifs were identified (Figure 18a).

Cycle Set Motif 1: Offshoal to Foreshoal

Description: This motif shows a strongly serrated GR pattern of moderate absolute GR values ranging between 17 to 27 API. There is a complete absence of grainstones in this cycle set motif (Figure 18a).

Interpretation: The serrated GR pattern is caused by thinly-interbedded grainy and muddy offshoal and foreshoal sediments (LFAs 7 and 8). The absence of high and low GR values is due to low rates of marly background sedimentation and the general lack of grainy shoal-associated deposits.

Cycle Set Motif 2: Offshoal to Shoal

Description: Cycle sets of this motif commonly show a moderate to highly variable GR log response with values ranging from 16–27 API. A weakly developed ‘dirtying-upward’ trend during the transgressive hemi-cycle set and a moderately to well-developed ‘cleaning-upward’ trend in the regressive hemi-cycle set can be recognized (Figure 18a).

Interpretation: GR log response is due to interbedded wackestones to grainstones of the offshoal and foreshoal setting. Upward-cleaning of GR values is caused by the presence of variably thick grainy shoal deposits that commonly display low overall GR values.

Cycle Set Motif 3: Foreshoal to Backshoal

Description: This motif typically displays a threepart GR log pattern (Figure 18a). After a weakly developed ‘dirtying-upward’ trend during the transgressive hemi-cycle set (17 to 27 API), GR values decrease in the lower part of the regressive hemi-cycle set down to 16 API and strongly increase again in the uppermost part (17–38 API).

Interpretation: The foreshoal LFA of the transgressive hemi-cycle set shows a large scatter in moderate GR values due to the interbedding of grainy and muddy sediments. Muddy backshoal cycle set caps above variable well-developed grainy shoal unit present in the regressive hemi-cycle induce the distinctive cleaning-upward followed by a dirtying-upward trend in the regressive portion of the motif.

Cycle Set Motif 4: Shoal to Tidal Flat

This motif was not observed on the Saiq Plateau, but is present in more landward sections of the Khuff platform in Oman. It may be very useful for the overall understanding of lateral cycle set variations on the Khuff platform in Oman.

Description: This cycle set motif is characterized by a well-developed cleaning-upward trend during the transgressive hemi-cycle set towards the MFS and a dirtying-upward trend throughout the regressive part up to the cycle set boundary (Figure 18a).

Interpretation: The MFS of this cycle set motif is placed within the thickest-developed grainstone units (LFA 5) marked by the lowest GR values. Towards the cycle set boundary, these high-energy deposits are in turn overlain by muddy backshoal and tidal flat deposits showing moderate to high GR values.

Third-order Sequences

Regional stratigraphic analyses by Strohmenger et al. (2002), Alsharhan (2006) and Insalaco et al. (2006) suggest that the Khuff Formation can be subdivided into six or seven third-order sequences. Six sequences, composed of a variable number of fourth-order cycle sets (KCS), were defined at the Saiq outcrop bound by stratigraphically significant marker beds (Figures 19 to 23). They were termed Khuff Sequence 6 (KS 6) to Khuff Sequence 1 (KS 1) from bottom to top. These are sequence stratigraphic units not to be confused with Khuff reservoir units in Oman (K1 – K5) or elsewhere (e.g. Sharland et al., 2004; Alsharhan, 2006).

Each of the six Khuff sequences (KS 1 to KS 6) is composed of one transgressive and one regressive hemisequence bound by a zone of maximum flooding. Sequence boundaries are interpreted on top of each regressive unit.

Khuff Sequence 6 (KS 6)

Sedimentological Description and Interpretation: The KS 6, 167 m in thickness (Figure 19), comprises four fourth-order cycle sets (KCS 6.1 – KCS 6.4) and 12 cycles (fifth-order). It conformably overlies reddish-white, rooted siltstones of the middle part of the Lower Saiq Member (Figure 6). The lower 122 m of the sequence consist of limestones while the upper 45 m are dolomitized. The onset of dolomitization is sharp and bed-parallel.

The basal transgressive part of the KS 6 (uppermost part of the Lower Saiq Member) is dominated by bioclast-rich limy storm beds interbedded with thinly laminated ostracod siltstones possibly representing the transition from a continental-lacustrine to a shallow-marine environment (Rabu et al., 1986). The mixed carbonate-siliciclastic basal unit is overlain by massive to low-angle laminated bioclastic packstones to grainstones some 20 m in thickness with common scouring and erosive bases. They contain a diverse marine fauna (e.g. fusulinids, bivalve, brachiopod and gastropod shells). Bioturbation is visible due to ‘cloudy’ particle- and mud-rich patches. This part of the sequence consists of coarsening-upward cycles of the foreshoal and shoal margin cycle motif (Figures 13 and 14). Further up in the section, thinly bedded mud-rich deposits become more frequent. These finely-grained burrowed mudstones and wackestones are pale gray to yellowish-weathered and contain diverse ichnofabrics, most notably Zoophycus traces pointing to a low-energy deeper water setting. Intercalated are graded storm sheets with abundant skeletal debris and crinoid ossicles. In some cases, massive low-angle laminated bioclastic packstones to grainstones mark the top of individual small-scale cycles of the foreshoal cycle motif. This part of the succession is interpreted as an overall deepening-upward trend from shallow-marine to open-marine carbonates.

The zone of maximum flooding of the sequence is interpreted within the KCS 6.2, some 117 m from the base in a 15 m thick unit that contains thick stacks of massive dark-blue, bioturbated mudstone (“Muddy Marker”), in cases heavily stylolitized. It is the lowest energy zone mainly characterized by suspension settling, starvation and background sedimentation below storm wave base. The dark color suggests less oxygenated waters in a deeper water setting.

The dolomitic regressive part of KS 6 is 50 m thick and dominated by bioturbated peloidal packstone and peloidal to oolitic grainstone. The grainstone is medium- to coarsely grained and shows well-developed low-angle lamination or trough cross-bedding. Bioclasts are reduced to some shell lags at the base of individual graded storm sheets. This part of the KS 6 is interpreted as an overall shallowing-upward/coarsening-upward regressive hemi-sequence in which cross-bedded to low-angle laminated peloidal shoals prograded over open-marine bioturbated mudstones and graded foreshoal facies types.

The KS 6/KS 5 sequence boundary is placed on top of a single 0.5 m thick microbial laminite (“Microbial Marker 1”) showing crinkly lamination and possibly indicates restricted intertidal conditions.

Gamma-ray Pattern: The base of the Upper Saiq Member (equivalent to base Khuff) is marked by a sharp decrease in total GR readings. The values drop from 50 API in the mixed clastic-carbonate unit below to around 20 API in the massive bioclastic limestone of the KS 6 above (Figures 19 and 24). The lower calcitic part of the KS 6 shows an upward increase in GR values towards the zone of maximum flooding and subsequently a decrease in the upper, dolomitized part towards the KS 6 sequence boundary.

Stable-isotope Pattern: Wthin the KS 6, δ13C values get progressively higher from around +2 ‰ at the base to +5.1 ‰ at the KS 6 sequence boundary (Figure 24). The measured δ18O values within the limestone section of the KS 6 average around -4‰. At the limestone-dolomite transition close to the KS 6 MFS, a sudden major increase of δ18O values from -6.2‰ to as high as +2.7‰ is observed. Thus the δ18O record seems to be strongly altered by the dolomitization of the Saiq Plateau section.

Khuff Sequence 5 (KS 5)

Sedimentological Description and Interpretation: KS 5 is 214 m thick (Figure 20) and consists of 12 cycle sets (KCS 5.1 – KCS 5.12) and 57 cycles (fifth-order). The sequence begins with a transgressive surface reworking the microbial unit on top of KS 6. The 90 m thick transgressive hemicycle of the sequence is characterized by dark gray, medium-grained bioclastic packstone to grainstone and skeletal floatstone containing brachiopod shells (e.g. Productus, Spiriferina, Richthofenia). These facies types are interbedded with bioturbated mudstone to wackestone and cross-bedded peloidal grainstone. Further up in the section, beds become progressively muddier. Facies are mainly arranged in shallowing-upward cycles of the foreshoal and shoal margin motif (Figures 13 and 14). The transgressive part of the KS 5 forms an aggradational stack of open-marine to proximal foreshoal facies types representing a low-energy setting periodically perturbed by storms. Locally incipient shoals developed, which are represented by low-angle laminated to cross-bedded peloidal grainstone. Periodic shoaling is also indicated by the presence of shoal- to backshoal cycles with microbial laminite caps (Figure 17).

The maximum flooding surface is placed in the KCS 5.8, some 90 m from the base, within a 10 m thick zone containing burrowed mudstone to wackestone with abundant dm-sized, dark gray to reddish chert nodules (“Chert Marker”), common hardground development and skeletal floatstone.

The regressive part of KS 5, immediately above the “Chert Marker” bed, starts with a prominent, approximately 10-m-thick yellow-colored bed with a scoured/loaded base and several internal concave erosional surfaces. The lower unit of the bed is a dark gray coarse-grained fusulinid floatstone. It is erosively truncated by beige finely-grained bioclastic grainstone with corals and fewer fusulinids. Relict channel structures are preserved on the top. Above this conspicuous yellow bed, the regressive hemicycle consists of a bioclastic packstone to grainstone and coral floatstone. The dark packstone contains an open-marine fauna including megalodon bivalves, corals, brachiopods and large miliolid and staffellid foraminifera (e.g. Shanita amosi, Sphaerulina spp.). Packstones pass upward into cross-bedded peloidal and oolitic grainstone. Cycles are of the shoal and shoal-to backshoal motif (Figures 15 to 17). The uppermost part of the regressive hemicycle is muddy consisting mainly of light gray to white burrowed/rooted mudstone to wackestone interpreted as low-energy lagoonal/backshoal deposits and intertidal microbial laminites with occasional tepee structures. Intercalated are higher-energy facies types such as amalgamated bioclastic storm beds with erosive bases and thin oolitic grainstone sheets. Grainstone units are generally interpreted as spillover lobes deposited during small-scale transgressive events.

The KS 5/KS 4 sequence boundary is placed within a microbial laminite unit (“Microbial Marker 2”). This bed shows wavy and crinkly laminae and tepee structures indicating subaerial exposure with various wetting and drying cycles. The microbial laminites are brecciated in part and re-cemented by dolomite cement. Mudstone clasts within the dolomite breccia-matrix are angular and mm-cm sized. The sequence boundary coincides with the final occurrence of a distinctive large foraminifer (Shanita amosi).

Gamma-ray Pattern: The KS 5 can be subdivided into two parts based on the overall GR pattern (Figures 20 and 24). The lower part, representing the transgressive hemisequence, shows constantly a smooth GR curve with average values of around 20 API. Minimum GR values occur just above the interpreted KS 5 MFS within a massive fusulinid floatstone. Above, the GR trend reverses and values increase again towards the KS 5 sequence boundary in the upper regressive part. The highest GR values of the entire carbonate section, averaging around 28 API, are reached around the KS 5 sequence boundary. This may be caused by the high amount of microbial laminites and burrowed to rooted mudstones to wackestones that occur in this position.

Stable-isotope Pattern: The δ13C-curve of the KS 5 shows a rather straight pattern with values ranging from +3.8‰ to + 5.5‰ with an average of around +4.9‰ (Figure 24). δ18O values of the KS 5 also show a highly serrated pattern and range from +2.5‰ to -2‰ with average values around -0.5‰ A major drop in δ18O from +1.5‰ to -3‰ is apparent in the uppermost part of the sequence towards the KS 5/KS 4 transition.

Khuff Sequence 4 (KS 4)

Sedimentological Description and Interpretation: KS 4 is 170 m thick and the most grain-rich unit in the outcrop (Figure 21). It is made up of 11 cycle sets (KCS 4.1 – KCS 4.11) and 66 cycles (fifth-order). The lower 114 m thick transgressive hemisequence starts with a unit of burrowed/rooted wackestone and microbial laminites (m-level 340–320), possibly representing the lateral equivalent of an anhydritic interval (“Middle Anhydrite”) described in various wells in Bahrain, Oman, Qatar, Saudi Arabia, UAE, and large parts of the Arabian Peninsula (e.g. Al-Jallal, 1995). Above this muddy interval, hummocky-cross stratified bioclastic beds grade upwards into thick stacks of massive, low-angle laminated to trough cross-bedded peloidal grainstone. Cycles are of the shoal cycle motif (Figures 15 and 16). Peloidal grainstones thicken-upward and bio-clastic storm beds become thinner and less frequent. Muddy cycle caps are rarely preserved. The aggradation of grain-dominated peloidal grainstone with occasional muddy caps represents a stack of shallow-water incipient shoals or sandwave complexes. Dark bioclastic packstones or skeletal floatstones represent significant marine flooding and opening-up of the platform over the shoal to muddy inter- to backshoal environments.

The zone of maximum flooding of the KS 4 is interpreted within the KCS 4.5 within a 6 m thick unit composed of thinly bedded, graded packstone to wackestone, interpreted as tempestite sheets. They contain open-marine fauna mainly consisting of crinoids, rugose corals and undefined shell debris.

The regressive part of KS 4 is 56 m thick (Figure 21). Beds turn back into grainy textures with cross-bedded and low-angle laminated peloidal packstone to grainstone. Rare bioclastic beds contain open-marine species (e.g. rugose horn corals, brachiopod shells and crinoids). Cycles are 3–5 m in thickness and of the shoal motif (Figures 15 and 16). Within the thick grainstone piles, the Permian fauna is progressively less abundant. The KS 4/KS 3 sequence boundary is placed on top of a 0.5 m thick light gray weathered microbial laminite unit (“Microbial Marker 3”) interbedded with burrowed/vertically rooted mudstone to wackestone.

Gamma-ray Pattern: The lower, transgressive part of the KS 4 shows a very serrated pattern in the GR curve with values ranging from 17–35 API (Figures 21 and 24). Due to the high amount of stacked peloidal-oolitic grainstones in the middle part of the KS 4, the area around the KS 4 MFS does not show a well-developed GR signal. In the regressive part of the KS 4, average GR values are 22 API. They gradually increase up to 28 API around the KS 4/KS 3 sequence boundary due to the renewed occurrence of abundant microbial laminites.

Stable-isotope Pattern: A serrated pattern of the δ13C signature is observed within the KS 4 with highly variable numbers between +2.9‰ to +6.1‰ (average: +5.1‰) (Figure 24). A first strong negative δ13C-excursion appears around the interpreted KS 4-KS 3 sequence boundary. The δ18O-curve of the KS 4 also shows a very serrated pattern. Values scatter from +1.5‰ to – 5‰. Around the KS 4/KS 3 sequence boundary, δ18O values rapidly increase up to a maximum of +0.5‰.

Khuff Sequence 3 (KS 3)

Sedimentological Description and Interpretation: KS 3 is 68 m thick (Figure 22), comprising four cycle sets (KCS 3.1 and KCS 3.4) and 19 cycles (fifth-order). Its lower transgressive part (basal 40 m) consists of dark coral floatstones and bioclastic packstones dominated by foraminifera, bryozoans, crinoids and bivalves. These turn into beds of graded, low-angle laminated peloidal packstone to grainstone and well sorted cross-bedded peloidal grainstone. Beds are mainly organized in cycles of the shoal margin motif (Figures 15 and 16).

Towards the zone of maximum flooding, situated within the KCS 3.2, bioturbated bioclastic and peloidal packstone with upwards increasing open-marine rugose horn corals, bivalve shell debris and rare crinoids dominates. Gastropod shells occur in places. Maximum flooding of this sequence is picked at a distinctive 1-m-thick coral floatstone with abundant rugose fasciculate corals (Waagenophyllum) (Figures 10f and g). This bed, informally referred to as “Coral Marker”, is a marker bed traceable on the Saiq Plateau for at least 10 km.

The regressive part of the sequence, 25 m thick, is characterized by beds of bioturbated and poorly sorted peloidal and bioclastic packstone to grainstone (Figure 22). Sharp erosive bases and low-angle cross-stratification are well developed within these beds. Muddy caps are absent.

The KS 3/KS 2 sequence boundary is placed on top of a 7 m massive, well sorted peloidal-oolitic grainstone. It marks the return of well-developed shoal-associated carbonate sands.

Gamma-ray Pattern: Within the KS 3, the GR drops from 28 API at the base to 16 API at the KS 3/KS 2 sequence boundary, marked a thick peloidal grainstone unit (Figures 22 and 24).

Stable-isotope Pattern: The KS 3 shows a gradual decrease in δ13C from +6.1‰ at the base to +3.7‰ at the top (Figure 24). Average carbon-isotope values are +3.9‰. After a strong negative shift down to -3.3‰ within the lowermost part of the KS 3, δ18O-values within the sequence show a straight pattern with average values of around -2.2‰. Around the interpreted KS 3/KS 2 sequence boundary, a strong negative shift in δ18O from -2‰ to a minimum of -4.5‰ is noted

Khuff Sequence 2 (KS 2)

Sedimentological Description and Interpretation: Three cycle sets (KCS 2.1 – KCS 2.3) composed of 11 cycles (fifth-order) stack to form the KS 2, 55 m in thickness (Figure 23). The lower transgressive hemisequence, measuring some 23 m, is mainly composed of skeletal floatstone and peloidal grainstone. Beds are stacked to 2–3 m thick cycles of the shoal margin motif (Figure 14). The interval around the “Saiq/Mahil Formation Boundary” is marked by a disrupted/brecciated mudstone to grainstone showing synsedimentary deformation fabrics. It is tentatively interpreted as seismite deposit.

Maximum flooding is picked within pale gray to blackish, finely-grained dolomitic mudstones of the KCS 2.2. These thinly bedded deposits are interpreted as distal open-marine graded storm beds. They represent storm-influenced foreshoal deposits just above the SWB in an outer to mid ramp setting.

The thicker upper regressive part of the KS 2 is 32 m thick and shows a clear coarsening-up, thickening-up trend (Figure 23). Graded storm beds pass upwards into intra-clastic grainstone/rudstones with microbial coated clasts and further into m-thick beds of cross-bedded peloidal grainstone. Facies are arranged in cycles of the shoal motif. Shallowing proceeds during the hemisequence into the development of thin intra-clastic-peloidal shoal complexes.

The KS 2/KS 1 sequence boundary is placed at the top of the thickest developed peloidal grainstone that indicates the highest depositional energy.

Gamma-ray Pattern: This sequence shows a well-develop cleaning-upward GR pattern (Figures 23 and 24). Values constantly shift from 20 API in the lower part to 14 API around the KS 2 sequence boundary. Average GR values within the KS 2 are 15 API.

Stable-isotope Pattern: A second strong negative δ13C excursion is apparent around the KS 2/KS 1 sequence boundary (Figure 24). Values drop from +3.7‰ at top KS 3 down to +0.7‰ within the middle part of the KS 2. Generally, the KS 2 sequence is defined by a smooth δ13C curve with average values around +2.2‰. δ18O-values show a straight pattern with average of around -1.9‰ throughout the sequence.

Khuff Sequence 1 (KS 1)

Sedimentological Description and Interpretation: The 51 m thick sequence is the thinnest of all Khuff sequences and is built by two cycle sets (KCS 1.1 and KCS 1.2) composed of 8 fifth-order cycles (Figure 23). The sequence is mainly composed of high-energy facies types arranged in shallowing-upward small-scale cycles of the shoal margin and shoal cycle motif (Figures 14 to 16). The lower transgressive part of the KS 1, some 10 m thick, consists of beds of graded mudstone to packstone, grading upwards into layers of cross bedded peloidal-oolitic grainstone.

The zone of maximum flooding is placed in thinly bedded, graded storm beds with interbedded dark bioturbated mudstones and wackestones (KCS 1.2). They represent the lowest energy, most intense burrowing and fully open-marine conditions.

The upper regressive 40-m-thick hemisequence (Figure 23) is dominated by up to 5-m-thick cycles of the shoal motif that are build of graded packstone, intra-clastic grainstone/rudstone and cross-bedded peloidal-oolitic grainstone intercalated only by thin transgressive storm sheets. Towards the top of the sequence, the thickest ooid grainstone deposits of the Upper Khuff equivalent (KS 2 – KS 1) are observed. They are interpreted as high-energy shoal deposits. They are laterally persistent and traceable over the whole study area (up to 10 km).

The KS 1 upper sequence boundary is marked by an up to 5 m thick polymict breccia (“Top Breccia”) with a dolomitic grainstone matrix. Brecciation was most likely followed by rapid cementation of the mud clasts. The origin of the breccia is not yet fully understood. Brittle thrusting together with cohesive soft-sediment deformation features due to dewatering may hint at a structural origin (J. Mattner, personal communication, 2009). Close proximity to thrust faults is indicated by isoclinal folds in m-scale and boulders up to m-size (Figure 30b). The strong mechanical contrast between the competent dolomites of the Lower Mahil Member below, and the incompetent basal shales of the Middle Mahil Member above, could have caused bedding parallel displacement and sediment brecciation due to thrusting during the Late Cretaceous.

The boundary between the Khuff-equivalent (Lower Mahil Member) and the overlying Middle Mahil Member (Sudair Formation equivalent) is picked on top of this brecciated bed just below a conspicuous thrombolite bed and the first occurrence of red and gray-green shales.

Gamma-ray Pattern: GR values generally show a very smooth pattern within the KS 1 with average values of around 16 API (Figures 23 and 24). An important GR marker occurs at the top of the Lower Mahil Member. It is characterized by the first appearance of shales in the entire investigated section. These argillaceous beds cause a strong increase in the GR readings, reflecting a regionally important GR marker at the base of the Middle Mahil Member (Sudair Formation time-equivalent) (Sharland et al., 2004; Osterloff et al., 2004).

Stable-isotope Pattern: In the KS 1, δ13C again only show little variation and vary between +1.8‰ to +3.1‰ (average: +2.3‰) (Figure 24). A positive δ13C shift from +2‰ to +3.5‰ appears in the first shale beds of the Middle Mahil Member (Sudair Formation-equivalent). δ18O-values within this sequence again scatter widely between -1‰ to -4.2‰ (average -3‰). At the base of the overlying shaley beds of the Middle Mahil Member, the curve shows a positive shift of δ18O values from -2‰ to -0.5‰.

Second-order Supersequence

The Upper Saiq Member and Lower Mahil Member possibly comprise a single second-order transgressive-regressive supersequence. During the second-order transgressive hemi-supersequence, basal clastics (Lower Saiq Member) are overlain by open-marine limestones and dolomites. The location of the 2nd-order MFS however is inconclusive based on the present dataset. In the subsurface, the MFS of the entire Khuff Formation is commonly interpreted in the middle part of the KS 4 (e.g. Alsharhan 2006; Insalaco et al., 2006). The Saiq Plateau outcrop however shows very little evidence for a major transgression within this interval. Possible candidates to place the overall MFS include the KS 6 MFS (“Muddy Marker”), the KS 5 MFS (“Chert marker”) as well as the KS 2 MFS. The different interpretations of the second-order MFS might be due to differential tectonic movements in Oman during the Dzhulfian (Wuchiapingian) (KS 4). The “Top Breccia” zone on top of the Lower Mahil Member (top Khuff time-equivalent) is interpreted as second-order sequence boundary coinciding with a major fall in relative sea-level.”

BIOSTRATIGRAPHY

Stratigraphic Distribution of Foraminifera and Regional Biostratigraphic Correlation

A discussion of the biostratigraphic correlation of the studied outcrop sections with other surface and subsurface units across the Arabian Platform necessitates a short summary of previous work on biostratigraphy and stratigraphic subdivision. The biostratigraphy of the Khuff and its correlatable formations is mainly based on brachiopods (Angiolini et al., 1998; 2003), ostracods (Crasquin-Soleau et al., 1999, 2006), smaller foraminifera and paleoflora (including palynomorphs) (Stephenson, 2006; Berthelin et al., 2006). Vachard et al. (2005) and Gaillot and Vachard (2007) highlighted the importance of smaller foraminifera as a potential tool for a sequence eco-biostratigraphic subdivision of the Upper Khuff in the Middle East Gulf region, subsequently applied in Insalaco et al. (2006). Data on the stratigraphic distribution of smaller foraminifera in the Lower Khuff are sparser and a biostratigraphic subdivision has not been established so far.

Onset of sedimentation on the Arabian Platform above the pre-Khuff unconformity is generally assumed to range diachronously from the Murgabian to the Midian (ca. Wordian – Capitanian). A widespread transgression led to the deposition of Lower Khuff carbonates, followed by an evaporitic interval (“Median Anhydrite”) over vast parts of the Arabian Platform. Carbonate deposition was reestablished in the Late Permian to Early Triassic forming the Upper Khuff strata.

The sections on the Saiq Plateau yield macro- and microfossil fauna throughout the Permian interval (Plates 1 and 2). Whereas microfossils in the lower part of the Permian section (KS 6 – KS 5) are fairly well preserved, the upper interval of the Permian part (KS 4 – lower KS 2) is affected by a pervasive late diagenetic dolomitization. In many samples, the primary microfacies are destroyed, and an unequivocal determination of fossils is often difficult. This hampers direct biostratigraphic correlations with the usually rich and generally well preserved fauna from the same stratigraphic interval in the subsurface of Oman and elsewhere on the Arabian Platform.

Lower Triassic deposits are generally characterized by a low diversity fauna following the end-Permian mass extinction. The pervasive dolomitization of the section prevents a biostratigraphic interpretation of the presumed Triassic interval.

The basal part of the KS 6 (samples 2–17 in Table 1) yields a fairly high-diversity fauna including common fusulinid (Chusenella sp., Schubertella sp., Globivalvulina aff. bulloides, Climacammina sp., Tetrataxis sp.), miliolid (Neodiscus sp.), and lagenid (Pseudolangella sp., Pachyphloia ovata, Nodosinelloides potievskayae, Geinitzina chapmani) foraminifera. In spite of the diverse open-marine fauna, neoschwagerinid and verbeekinid species, mentioned in previous studies from the Oman Mountains (Montenat et al., 1977; Weidlich and Bernecker, 2007), have not been encountered in this section.

The upper part of the KS 6 (samples 18–43 in Table 1) displays a low biodiversity dominated by recrystallized thalli of gymnocodiacean algae (Permocalculus spp.) in a wacke-/packstones matrix. The impoverished foraminiferal fauna consists mainly of staffellid and small miliolid forms (Hemigordiellina regularis, Midiella? sp.), generally with completely recrystallized shells.

A narrow zone within KS 5 (samples 48–51 in Table 1) yields Neoendothyra cf. parva, Parafusulina sp. and Yangchienia? sp., together with the enigmatic Sphairionia sikuoides, indicating the presence of Midian (Late Wordian – Capitanian) deposits. A similar faunal interval with Parafusulina (Monodiexodina?) sp. and Dunbarula sp. have also been encountered in the subsurface of Oman.

A burst of new faunal elements, including Shanita amosi and Paraglobivalvulina mira appears in the uppermost KS 5 (samples 83–106 in Table 1), related to the development of extensive backshoal environments. The beds with Shanita amosi are a well traceable marker just below the Median Anhydrite in Oman wells and are also reported from equivalents of the Nar Member of the Dalan Formation in Iran (Insalaco et al., 2006).

KS 4 (samples 109–157 in Table 1) is characterized by the disappearance of schwagerinids and presence of Neodiscopsis ambiguus and Rectostipulina quadrata, which have their first appearance in Upper Khuff strata (Insalaco et al., 2006; Gaillot and Vachard, 2007). Rare occurrences of Neomillerella mirabilis in a nearby section in the upper KS 4 may hint at similar assemblages in the Late Dzhulfian (Wuchiapingian) of the Zagros-Fars area (Insalaco et al., 2006).

A Dorashamian (Changhsingian) age is based on the presence of a zone with large miliolids (Glomomidiellopsis uenoi) accompanied by rare Nodosinelloides sagitta around the KS 4/KS 3 boundary (Vaslet et al., 2005; Insalaco et al., 2006; Gaillot and Vachard, 2007). Rare, indeterminable biseriamminids and staffellids persist into the upper KS 3 (samples 15–172 in Table 1). Insalaco et al. (2006) have described diverse assemblages including Paradagmarita and allied genera in uppermost Permian Khuff equivalents. Conspicuously, these foraminiferal assemblages have not been found in the studied sections and it is currently uncertain, whether the apparent absence is related to the strong dolomitization, sampling bias, or to unfavorable ecologic conditions in the KS 3 interval.

  • (1) Hemigordiusaff.schlumbergeri(sample 158)

  • (2) Hemigordiusaff.irregulariformis(sample 134)

  • (3) Midiellaex gr.reicheli(sample 119)

  • (4) and (5) Neodiscopsis ambiguus (sample 119)

  • (6) Neodiscopsissp. (sample 137)

  • (7) Glomomidiellopsis uenoi(sample 153)

  • (8) Globivalvulinacf.vonderschmitti(sample 119)

  • (9) Dagmaritasp. (sample 119)

  • (10) Dagmarita? shahrezaensis(sample 137)

  • (11) Retroseptellina decrouezae(sample 119)

  • (12) Biseriamminid foraminifera (cf.Globivalvulina?)(sample 175)

  • (13) (a, b)Rectostipulinan. sp. aff.syzranaeformis(sample 119)

  • (c) R. pentamerata(sample 137)

  • (d) R. quadrata(sample 153)

  • (14) Ichtyofrondinasp. (sample 153)

  • (15) “Endoteba”cf.controversa(sample 119)

  • (16) Earlandia?sp. (sample 206, Mahil Formation, early Triassic)

  • (17) Nodosinelloides sagitta(sample 159)

The Permian Faunal Extinction Event (PFE) on the Saiq Plateau is marked by the sudden disappearance of Permian fauna in the basal KS 2 (samples 175–177 in Table 1).

Biostratigraphic control is very poor for the Triassic interval due to the pervasive dolomitization and low biodiversity following the end-Permian mass extinction. Only few Earlandia? sp. have been encountered in the upper part of the KS 2 (samples 188–206 in Table 1). Several samples in KS 1 show microbially induced carbonate precipitation with oolites and aggregate grains constituting a grapestone facies. However, the thrombolitic facies with earliest Triassic fauna (Rectocornuspira kalhori, Spirorbis phylctaena), common in many Tethyan outcrop sections and subsurface wells right above the PTrB boundary (Insalaco et al., 2006; Groves and Altiner, 2005), have not been found in the section of the Saiq Plateau.

The reappearance of foraminiferal fauna with sporadic occurrences of Hoyenella sinensis and H. tenuifistula occurs in the lower part of the Middle Mahil Member (Sudair Formation time-equivalent). In the Musandam area, an association of Hoyenella sinensis together with Meandrospira pusilla is interpreted to yield a Late Induan to Olenekian age (Maurer et al., 2008, 2009).

Discussion of Proposed Stage Boundaries

Fossil groups (conodonts, larger benthic foraminifera), which are preferably used for Tethys-wide or global correlations, are largely absent on the Arabian Platform. The classical Tethyan Late Permian index species are likewise rarely reported.

Paleobiogeographically, the Arabian Platform represents the southeastward prolongation of the Southern Biofacies Belt (Altiner et al., 2000), which is characterized by the Late Midian (Capitanian) Shanita fauna and Late Permian (Lopingian) Paradagmarita fauna. The absence of key index fossils in the Middle – Late Permian and the paleobiogeogeographic differences in the faunal composition (Kobayashi, 1999; Kobayashi and Ishii, 2003; Ueno, 2003; Gaillot and Vachard, 2007) hamper straightforward correlations of the sections from the Peri-Gondwana margin with other Tethyan type sections. Correlations are based on those sections, where smaller foraminifera have been reported to co-occur with larger forams and are therefore afflicted with different degree of uncertainty.

Problems in correlating Khuff-equivalent strata with regional and global stratigraphic scales have been repeatedly stressed in several publications. Some of the problems are rooted in the imprecise definitions of stages (e.g. base of Midian) and the calibration of foraminiferal biostratigraphy with other faunal groups (conodonts, ammonoids, brachiopods). According to the foraminiferal, brachiopod, and sparse conodont data (Montenat et al., 1977; Lys, 1988; Rabu et al., 1990; Angiolini et al., 2003) from the Saiq Plateau and Al Huqf – Haushi areas, the lower part of the Khuff has been correlated with the Wordian (Murgabian) (“Wordian transgression” in Angiolini et al., 2003). In contrast, Vachard et al. (2002) updated the earlier work of Montenat et al. (1977) on the Saiq Plateau and assumed a Midian (Capitanian) age based on foraminiferal biostratigraphy (“Midian transgression”).

Recent work in Tunisia (Angiolini et al., 2008) and data from Sicily (Kozur and Davydov, 1996) indicate that part of the Midian might actually belong to the Wordian. However, the base of the Midian itself is biostratigraphically not strictly defined (Leven, 2003) and parts of the late Murgabian overlap with the early Midian. Due to the absence of Yabeina-Lepidolina assemblages in western Tethyan sections (including the Midian type section), the FAD of Dunbarula and Kahlerina instead has been used to characterize lower Midian strata. But Dunbarula nana has already been reported from the Afghanella schencki Zone in Iran (Kobayashi and Ishii, 2003), which is located well in the Murgabian. Dunbarula nana has been mentioned from the Saiq Plateau (Montenat et al., 1977), from the Lower Dalan Formation in Iran (Insalaco et al., 2006) and might be present in subsurface wells of Oman.

Due to the absence of unequivocal age-diagnostic index fossils in the studied section, it is currently difficult to precisely trace either the Murgabian/Midian, or the Wordian/Capitanian boundary. The lower part is herein conventionally attributed to the Murgabian (Wordian) and the Murgabian/Midian boundary most probably lies somewhere around the KS 6/KS 5 sequence boundary (Figure 25).

The end-Guadalupian faunal extinction selectively wiped out several fossil groups, which were assumed to host photosymbionts including the larger benthic foraminifera (Schwagerinidae, Verbeekinidae, Neoschwagerinidae) (Ota and Isozaki, 2006). This biotic event is associated with strong perturbations of the carbon-isotope signal during the Capitanian (“Kamura event” sensu Isozaki, et al., 2007) and a widespread regression in the latest Capitanian. A latest Midian (Capitanian) fauna is indicated by the last occurrence of schwagerinids and the FO/LO of Shanita amosi in the upper KS 5, which also corresponds to a widespread regression (KS 5/KS 4 boundary), that can be followed across the entire Arabian Platform (Al-Jallal, 1995; Alsharhan, 2006). The Guadalupian/Lopingian boundary is therefore assumed to approximately coincide with the top KS 5 sequence boundary (Figure 25).

Evidence for Lopingian deposits on the Arabian Platform generally relies on the presence of rare, primitive Colaniella and abundant Paradagmarita and its allied genera (Gaillot and Vachard, 2007). Deposits of Wuchiapingian age are herein assumed to enclose the KS 4 according to the above stated faunal correspondance with data from Insalaco et al. (2006) (Figure 25).

The rare and poorly preserved biseriamminids in the upper KS 3 do not provide sufficient data to give a specific assignment. The typical Late Changhsingian foraminiferal fauna including Paradagmarita monodi is absent in the studied section. The absence of this species is most likely associated with the general scarcity of leeward shoal facies in the outcrop. Strongly recrystallized Glomomidiellopsis uenoi in the basal KS 3 has been selected alternatively to confirm the presence of Changhsingian deposits. The Wuchiapingian/Changhsingian boundary is placed in accordance with Insalaco et al. (2006) at the top KS 4 sequence boundary (Figure 25).

The Permian/Triassic Boundary (PTrB) in more landward settings like Yibal (Figure 3) is characterized by bioclastic/oolitic grainstones with abundant latest Permian smaller foraminifera followed by a widespread occurrence of microbial sediments (Masaferro et al., 2004; Insalaco et al., 2006; Ehrenberg et al., 2008; Maurer et al., 2009). A stromatolitic interval contains the first occurrence of earliest Triassic fauna indicated by Rectocornuspira kalhori and Spirorbis phylctaena (Insalaco et al., 2006). On the Saiq Plateau however, the PTrB is indistinct as the classical thrombolite unit is absent. Biostratigraphically it can only be approximated by the drastic decline of invertebrate fauna in the basal KS 2 (Figure 25).

TENTATIVE LOCATION OF THE PERMIAN/TRIASSIC BOUNDARY

The Permian/Triassic Boundary (PTrB) is one of the most important markers for regional correlation of the Khuff. Its position was tentatively placed based on three independent stratigraphic methods: biostratigraphy, chemostratigraphy and sequence stratigraphy.

Biostratigraphy

Biostratigraphically, the PTrB is defined by the first occurrence (FO) of the conodont Hindeodus parvus (Krull et al., 2004). In this study, no conodont remains were detected in outcrop samples (D. Korn, personal communication, 2009).

The extinction of the Permian fauna, commonly referred to as ‘Permian-Faunal Extinction’ (PFE) or ‘event horizon’, is located 6 m above the KS 3 sequence boundary (Figure 26). It occurs at the base of a prominent disrupted/brecciated pack- to grainstone bed, just below the Saiq/Mahil Formation Boundary (Figure 22, photo C). Above the PFE, there is an interval with only rare fossil remains (Figure 26). On the Saiq Plateau, only rare echinoderm fragments, shell debris as well as indications of burrowing occur.

The prime marker for the Triassic faunal recovery in many parts of the Arabian Platform is the widespread occurrence of Triassic microbialite carbonate sheets and thrombolites (e.g. Baud et al., 1997; Insalaco et al., 2006; Weidlich and Bernecker, 2007). This marker is not developed on the Saiq Plateau. However, sample 201 (Figure 7) is reminiscent to this facies, showing alternations of laminated dark, micritic to light, sparitic layers with possible fenestrae fabrics.

Chemostratigraphy

Carbon Isotopes

Numerous studies suggest that the PTrB can globally be recognized on the sharp negative shift in carbon isotopes (δ13C). This is commonly defined as “Carbon Isotope Shift” (CIS) (e.g. Magaritz et al., 1988; Baud et al., 1989; Wang et al., 1994; Septhon et al., 2005; Ehrenberg et al., 2008). In the investigated section, the δ13C drops gradually from the KS 3 sequence boundary towards the end PFE and further up to a minimum δ13C-value 5 m above. This point is subsequently interpreted as PTrB (Figure 26).

The δ13C-log from the Saiq Plateau was compared with the δ13C-log measured by Richoz (2006) in Wadi Sahtan, some 25 km to the NW (Figure 27). The general trend of the δ13C pattern with a decrease across the PTrB and more negative values in the Triassic part of the Khuff time-equivalent is apparent. Less prominent intra-Triassic negative δ13C shifts above the PTrB are apparent and correlatable. Values become increasingly positive in the overlying Sudair-equivalent, coinciding with the appearance of the clastic shale beds.

Spectral Gamma-ray

Coincident with the negative δ13C shift, the PTrB is regionally marked by a significant negative shift in U (Uranium event) (e.g. Szabo and Kheradpir, 1978; Alsharhan, 1993, 2006; Al-Jallal, 1994; Sharland et al., 2001; Bashari, 2005; Insalaco et al., 2006; Ehrenberg et al., 2008; Maurer et al., 2009). This drop in U is possibly caused by a chemical oceanographic change in earliest Triassic seawater associated with the abrupt onset of deep-ocean anoxia (Wignall and Twichett, 1996).

Our U curve shows higher values and a serrated pattern in the Permian part of the section. In contrast, the Triassic section is generally characterized by a drop in U readings and a smoother U curve (Figure 24). Unlike published data from the subsurface, the transition from higher to lower values on the Saiq Plateau is not sharp but gradual occurring over an interval of c. 30 m within the KS 2.

The rather indistinctive general pattern of the spectral GR-curve around the Permian/Triassic transition might be due to the overall lack of detrital and shaley material in the investigated outcrop section or due to local diagenetic effects.

Sequence Stratigraphy

Regional studies suggest that the PTrB occurs globally throughout a transgression (Wignall and Twitchett, 2002; Insalaco et al., 2006). Thus the Permian/Triassic transition should be associated with an opening of the Khuff platform. This is confirmed by the facies and stratigraphic analysis of the investigated outcrop. Opening of the platform and a deepening is inferred from a re-occurrence of Late Permian open, normal-marine fauna (e.g. rugose horn corals, crinoids and brachiopod shells) right above the KS 3/KS 2 sequence boundary within the transgressive KS 2-hemisequence (Figure 26). Fauna indicates an agitated and open shallow shelf environment in a foreshoal setting. The transgression is also marked by a drastic and abrupt facies change at the Saiq/Mahil Formation boundary. Thickly-bedded packstones to grainstones (high-energy shoal) pass into thinly-interbedded mudstones to wackestones (distal foreshoal) (Figures 23 and 26).

SEQUENCE STRATIGRAPHIC SYNTHESIS

Khuff Sequence Stratigraphic Framework

The integration of facies cycles, lithostratigraphic marker beds, bio- and chemostratigraphy was used to establish a sequence stratigraphic subdivision of Khuff Formation time-equivalent strata in the Oman Mountains. This analysis builds on work by Mabillard et al. (1985), Coy (1997), Osterloff et al. (2004) and Insalaco et al. (2006). Accordingly, the Upper Saiq and Lower Mahil Member (Khuff time-equivalent) can be subdivided into six third-order sequences (KS 6–KS 1) (Figure 28):

Sequence KS 6: The KS 6 comprises a time-interval from the middle to the end of the Murgabian. The MFS (“Muddy marker”) probably corresponds to the P20 MFS of Sharland et al. (2004). This sequence is equivalent to the lower K 5 reservoir interval in the subsurface of Oman.

Sequence KS 5: Covering approximately the Midian stage, the top of the sequence is biostratigraphically correlated with the end-Guadalupian mass extinction. It encompasses the upper part of the K5 reservoir interval in the subsurface of Oman. The KS 5 MFS (“Chert marker”) is also interpreted as MFS of the second-order supersequence.

Sequence KS 4: This sequence coincides with the Dzhulfian (Wuchiapingian) stage and includes the P30 MFS of Sharland et al. (2004). It corresponds to the K 4 reservoir interval in the subsurface of Oman.

Sequence KS 3: Falling entirely within the Dorashamian (Changhsingian) stage, the KS 3 encompasses the lower and middle part of the K 3 reservoir interval in the subsurface of Oman. The MFS (“Coral marker”) of the KS 3 might correspond to the P40 MFS of Sharland et al. (2004). Not to scale.

Sequence KS 2: This sequence chronostratigraphically belongs to the uppermost Dorashamian (Changhsingian) and lower Induan stages. Its MFS most likely corresponds to the Tr10 MFS of Sharland et al. (2004). It represents the upper part of the K 3 as well as the entire K 2 reservoir interval in the Omani subsurface.

Sequence KS 1: This sequence coincides with the upper Induan stage. The KS 1 MFS may be correlated with the Tr20 MFS of Sharland et al. (2004). The KS 1 corresponds to the K 1 reservoir interval in the subsurface of Oman.

Regional Correlation

Figure 29 illustrates a tentative correlation between the Middle to Upper Khuff section of offshore Fars (figure 9 of Insalaco et al., 2006), the Musandam outcrop section (figure 14 of Maurer at al., 2009) and the measured section of the Saiq Plateau (this study).

“Cycle IV” of Insalaco et al. (2006) might correspond to the KS 4 of our scheme (Dzhulfian-Wuchiapingian stage) (Figure 29). For this sequence, similar high-energy facies, most notably oolitic grainstones, dominate in the offshore Fars section and on the Saiq Plateau. No data is avalaible from Musandam from this stratigraphic interval.

“Cycle III” of Insalaco et al. (2006) probably corresponds to the lower part of KS 3 in our stratigraphic nomenclature (lower Changhsingian stage) (Figure 29).

“Cycle II” of Insalaco et al. (2006) can possibly be correlated with the upper part of our KS 3 and KS 2. For this interval, the Saiq Plateau section significantly differs in facies and depositional environment compared to the other two data points. Whereas mixed lagoonal and shoal facies dominate in offshore Fars and Musandam, the KS 3 on the Saiq Plateau is mainly made up of foreshoal to shoal deposits indicating a more open-marine, distal position on the Khuff carbonate ramp (Figure 3). Within the upper part of “Cycle II” (KS 2) in Musandam and offshore Fars, oolitic grainstones are common whereas mud-dominated foreshoal to offshoal facies types dominate on the Saiq Plateau.

Probably equivalent to “Cycle I” of Insalaco et al. (2006), the KS 1 reflects one third-order depositional sequence (Figure 29). This sequence varies significantly in thickness between the three sections, possibly reflecting changes in accommodation space, differential subsidence histories and tectonic activity. However, similar depositional facies characterized by intra-clastic, oolitic and peloidal grainstones are present in all sections.

From the data set it can be concluded that the KS 4 and KS 1 on the Saiq Plateau possibly provides an outcrop analog regarding the facies types to the gas-bearing Khuff Formation of the North Field area.

TECTONIC AND DIAGENETIC OVERPRINT

Tectonic Overprint

The Oman Mountains are one of the most tectonized and structurally complex areas on the Arabian Plate. We are aware of the fact that the logged outcrop sections on the Saiq Plateau and the resulting sedimentological interpretations are affected by this strong tectonization. Tectonic distortion of primary sedimentary fabrics in the study area mainly occurs on two different scales (Figure 30).

On a larger scale, brittle bed-parallel to low-angle thrust-faulting may have caused dm to several m of lateral displacement of individual beds (J. Mattner, personal communication, 2009). Minor thrustlike ramps also cross-cut primary sedimentary bedding planes in places (Figure 30a). The larger fault planes, mostly visible on satellite images and noted down in the geologial map of the study area (Rabu et al., 1986), were avoided when selecting the exact location of the logged sections A-D (Figure 2). However, less obvious low-angle thrust-planes along the section traces may be only unravelled by detailed structural mapping.

The boundary between the competent Lower Mahil Member (KS 2 – KS 1) and the overlying more shaley Middle Mahil Member (Sudair equivalent) is strongly affected by bedding-parallel anastomosing thrust faults, recumbent isoclinal folds up to a m-scale and horses of brittle material of up-to meter size (Figure 30b) (J. Mattner, personal communication, 2009).

On a smaller scale, stylolitization also modifies original sedimentary boundaries in the outcrop sections (Figures 30c to f). In general, nearly every sedimentary bed boundary is overprinted by (micro-) compaction stylolitization, leading to the development of columnar stylolites with low to moderate amplitudes of up to 3 cm. Along some of these sedimentary bedding planes, mm- to cm-thick reddish stylolitization seams with clay residues were observed. Altogether, stylolitization may have removed about 10-30% of the pre-compacted rock volume of the investigated section (J. Mattner, personal communication, 2009).

Diagenetic Modification

A detailed evaluation of the diagenesis of the Saiq and Mahil Formation on the Saiq Plateau is presented in Coy (1997). In this study, no attempt was made to perform closer paragenetic investigations as analyses on cement petrography and trace element content were not carried out. Coy (1997) concluded that is likely that the δ13C values of the Saiq and Mahil dolomites reflect the initial marine isotopic composition of the precursor whereas δ18O is more readily modified by subsequent diagenetic events. Thus oxygen stable isotope values of the investigated dolomites are expected to be significantly modified by diagenetic alteration.

Carbon isotope values for the dolomites of the Permian part of the Saiq Plateau section range from +6.4 to +2.1‰ with a median δ13C value of +4.5 ‰ V-PDB (Figures 24 and 31). Permian oxygen isotope values range from -4.9 to +2.35‰, with a median δ18O value of -1.5 ‰. The carbon isotope values for Triassic rock samples range from 0 to 4.5‰ (median +2.15‰ V-PDB), oxygen values vary from -4.0 to -1.2‰ (median -2.4‰). The strong depletion of the measured oxygen isotope values is consistent with the data presented in Coy (1997). No correlation was found between δ13C and δ18O values within the investigated section. However, average δ13C and δ18O values are generally higher for Permian than for Triassic samples (Figure 31).

CONCLUSIONS

The study of the Permian and Triassic carbonates on the Saiq Plateau, Al Jabal al-Akhdar, in the Sultanate of Oman yielded the following results:

  • (1) The investigated section is interpreted to be time-equivalent to the subsurface Middle Permian to Lower Triassic Khuff Formation.

  • (2) The outcrop is characterized by a very high percentage of grain-dominated textures. They represent the storm-dominated shoal to foreshoal section of the Khuff carbonate ramp. Most facies are open-marine and high-to moderate energy. There is a scarcity of peritidal deposits. Indicators for subaerial exposure and evaporites are absent.

  • (3) The interpreted depositional setting is in line with the established late Permian and lower Triassic paleogeographic location within the unrestricted marine carbonate shelf.

  • (4) Facies are stacked to transgressive-regressive cycles (fifth-order) of four general motifs: foreshoal, shoal margin, shoal and shoal- to backshoal.

  • (5) Stacks of these cycles form 36 transgressive-regressive cycle sets (fourth-order) clearly reflected in gamma-ray patterns. These are termed KCS 1.1 to 6.4 from top to bottom.

  • (6) The investigated section was subdivided into six transgressive-regressive sequences (third-order), termed KS 1 – KS 6. KS 6 – Lower KS 2 are interpreted to correspond to the Permian Upper Saiq Member. The Triassic Lower Mahil Member comprises Upper KS 2 – KS 1.

ACKNOWLEDGMENTS

This study is part of an extra-mural research project sponsored by Shell (Qatar). We are also grateful to Petroleum Development Oman (PDO, Muscat) for financial support and permission to publish this paper. The authors would especially like to thank Jan Schreurs, Gordon Forbes and Joachim Amthor (all PDO) for reviewing and improving earlier versions of this manuscript. We would also like to thank Claus von Winterfeld, Aly Brandenburg and Gordon Coy (all PDO) for assistance in many ways. We are grateful to Erwin Adams (Shell), Daniel Vachard (University of Lille), Joerg Mattner (GeoTech, Bahrain), Sylvain Richoz (University of Vienna), Heiko Hillgaertner (PDO), Henk Droste (Shell) and Deborah Bliefnick (Badley Ashton) for sharing their knowledge of the Khuff. Ulrike Schulte (University of Bochum) and Peter Swart (University of Miami) are thanked for stable isotope analysis. Per Jeisecke (University of Tuebingen) is thanked for the preparation of thin sections. We thank Shuram Oil and Gas (Muscat) for logistics of our field work. The final version of this manuscript greatly benefited from the comments by two anonymous reviewers and Moujahed Al-Husseini. GeoArabia’s Arnold Egdane is thanked for designing the final version of the figures. Access to the WellCAD software was kindly provided ALT (Luxembourg).

ABOUT THE AUTHORS

Bastian Koehrer studied Geosciences at the Universities of Tuebingen, Germany, and Bristol, UK, focusing on Sedimentary and Petroleum Geology. His MSc thesis (2007) was on reservoir characterization and 3-D modeling of a dolomite body from the Triassic Muschelkalk. Currently Bastian is a research and teaching associate at the Center for Applied Geosciences (Petroleum Geoscience Lab) in Tuebingen. The main objective of his PhD thesis, funded by Shell and Petroleum Development Oman (PDO), is a detailed description and characterization of the Khuff platform in outcrop and subsurface of the Sultanate of Oman. He aims to establish a regional valid sequence stratigraphic framework and conceptual geological model of the Khuff Formation that highlights nature and dimensions of potential reservoirs on exploration and production-scale. Among others Bastian is a member of the AAPG, SEPM, IAS and DGMK.

bastian.koehrer@gmx.net

Michael Zeller is currently enrolled as a PhD student at the Rosenstiel School of Marine and Atmospheric Sciences, University of Miami, Florida. He obtained an MSc in Sedimentary Geology from the University of Tuebingen (Germany) in July 2008. The main focus of his MSc thesis has been digital outcrop modelling of an Upper Khuff equivalent (Al Jabal al-Akhdar, Sultanate of Oman) and the resulting implications for static modelling in KS3-KS1 reservoir units. His study has been sponsored by Shell QSRTC, Shell EPI and Petroleum Development Oman.

mzeller@rsmas.miami.edu

Thomas Aigner studied Geology and Paleontology at the Universities of Stuttgart, Tuebingen and Reading/England. His diploma thesis was on the Geology and Geoarcheology of the Egyptian pyramides plateau in Giza (1982). For his PhD dissertation on storm depositional systems (1985) he worked at the Senckenberg-Institute of Marine Geology in Wilhelmshaven and spent one year at the University of Miami in Florida. He then became an exploration geologist at Shell Research in Rijswijk/Holland and Houston/Texas focussing on basin analysis and modelling (1985-1990). He worked as adjunct lecturer for applied sedimentology at the University of Wuerzburg (1988-1990). Since 1991 Tom is a professor and head of the sedimentary geology group at the University of Tuebingen. 1996 he was an ‘European Distinguished Lecturer’ for American Association of Petroleum Geologists. In 2007/8 he spent a sabbatical with PDO and Shell Qatar. His current projects focus is on sequence stratigraphy and reservoir characterisation/modelling in outcrop and subsurface.

t.aigner@uni-tuebingen.de

Michael Poeppelreiter studied at the Mining University of Freiberg, Germany, the Postgraduate Research Institute of Sedimentology, United Kingdom, and the University of Tubingen, Germany, where he earned a PhD in 1998. Since then, Michael has worked as sedimentologist/3-D modeller with Shell in Holland, as carbonate geologist/3-D modeller at Shell’s Bellaire Technology Center in Houston, USA and at present, he is senior carbonate geologist at the Qatar Shell Research and Technology Centre in Doha, Qatar where he is coordinating the Khuff/Sudair outcrop analogue study. Michael published numerous papers on carbonate reservoirs, reservoir modelling and borehole image log technology. He is guest lecturer at the University of Tuebingen, Germany. His research interests include structural control on reservoir distribution in carbonate reservoirs.

m.poppelreiter@shell.com

Paul Milroy has recently joined BG Group as Carbonate Technology Manager for BG Exploration and Production, Reading, UK. Prior to joining BG, Paul was a Senior Reservoir Geologist in the Shell’s Carbonate Research Team, Rijswijk, The Netherlands. Paul obtained a PhD in Geology at University of Bristol in 1998 before completing postdoctoral research at the University of Tokyo in 2001. He then worked as a reservoir geologist with Badley Ashton & Associates, before joining Shell in 2006 to work on the quantification of carbonate reservoir heterogeneity. He is a member of AAPG, SEPM and BGRS and has research interests in sedimentology, diagenesis, reservoir characterisation and modelling.”

paul.milroy@bg-group.com

Holger Forke studied Geology and Paleontology at the University of Erlangen. His diploma thesis and PhD dissertation (2001) focused on the biostratigraphic correlation of Carboniferous-Permian deposits from the Southern Alps (Austria) and Urals (Russia). He has then worked at the Senckenberg Research Institute in Frankfurt/Main and at the Institute of Geology (University of Erlangen) within the DFG Priority Programme 1054 ‘Late Paleozoic sedimentary geochemistry’. In recent years, he participated in expeditions and mapping campaigns to Svalbard and the Canadian Arctic in cooperation with the Norwegian Polar Institute, University of Bremen, and BGR Hannover. His work mainly deals with Late Paleozoic foraminifera and conodonts with emphasis on the application for sequence biostratigraphy. He is currently a guest researcher at the Museum of Natural History, Leibniz Institute for Research on Evolution and Biodiversity at the Humboldt University Berlin, Germany.

holger.forke@gmx.de

Suleiman Al-Kindi has a degree in Geology/Geophysics from the University of Durham, a PhD in Marine Geophysics from the University of Cambridge, and an MBA degree from the University of Hull. Sulaiman has joined the exploration Department in the Petroleum Development Oman (PDO) in 2002 as a hydrocarbon system analyst working on the deep gas petroleum system of North Oman and then moved into the regional team working on several gas exploration opportunities. He then joined the Qatar Shell Research and Technology Carbonate team as regional geologist in April 2007. He is currently working as regional geologist with PDO Exploration team.

sulaiman.a.kindy@pdo.co.om