Mid-Permian Khuff Sequence KS6: Paleorelief-influenced facies and sequence patterns in the Lower Khuff time-equivalent strata, Oman Mountains, Sultanate of Oman

The Khuff Formation represents one of the most important carbonate reservoir units across the Middle East (Figure 1). Six sequences, KS1 to KS6 in descending order, have been identified in Khuff-equivalent strata in the Oman Mountains (Figures 2 and 3; Koehrer et al., 2010). The Upper Khuff sequences KS1 and KS4 contain several reservoir zones and are well explored across the Gulf region. The lower sequences KS5 and KS6 are so far only very sparsely investigated, and fewer penetrations leave these units relatively underexplored.

This study on Sequence KS6 is part of a larger research project on Khuff time-equivalent strata in the Al Jabal al-Akhdar area (Oman Mountains) in the Sultanate of Oman (Figures 1 and 2). Its aim is to unravel the geometries and distribution of Khuff grainstones as potential reservoir bodies. Initially a one-dimensional (1-D) facies and sequence-stratigraphic framework was proposed by Koehrer et al. (2010) for the Khuff time-equivalent strata at a type locality on the Saiq Plateau. Subsequently, Khuff grainstone geometries were documented on the Saiq Plateau from near-well-scale (< 2 km) by Zeller et al. (2011) to field-scale (< 10 km) by Koehrer et al. (2011), and to subregional scale (< 60 km) by Koehrer et al. (2012). These studies revealed an overall “layer-cake”-type geometry of shoal grainstone bodies in Upper Khuff sequences KS4 to KS1.

Only subtle lateral heterogeneity of facies associations is observed in Khuff sequences KS1 to KS4 on a subregional-scale (60 x 40 km) in the Al Jabal al-Akhdar by Koehrer et al. (2012). Cycle sets and sequences were found to be highly correlatable, pointing towards a rather uniform gross depositional environment and absence of significant tectonic activity in this area during the post-rift phase of the Neo-Tethys Ocean. Post-depositional sediment deformations and structurally-related breccias and discontinuity surfaces are however described in sequences KS1 to KS4 from the Saih Hatat Window (Figure 2), paleogeographically located more proximally to the Arabian platform margin (e.g. Chauvet et al., 2009; Weidlich and Bernecker, 2011; Weidlich and Bernecker, 2012).

However, heterogeneities markedly increase in KS5 and KS6, possibly due to deposition during a syn-rift phase. Walz and Aigner (2012) and Walz et al. (2013) report significant thickness variations and apparent downlaps on cycle and cycle set scale, which are interpreted to result either from differential subsidence or initial topography. This paper focuses on the architecture of the grainstones in Khuff Sequence KS6 on a scale of 30 x 50 sq km. It aims at assessing the possible influence of paleotopography on potential reservoir facies distribution and continuity. It follows a systematic 1-D/2-D/3-D approach: after a detailed documentation of 1-D outcrop sections, 2-D stratigraphic cross-sections are constructed, leading to 3-D geocellular models.

The study area is located in the Oman Mountains, about 130 km west of Muscat (Figures 1 and 2). The northern Oman region experienced several phases of long-term subaerial exposure between the Proterozoic and Early Permian, which resulted in times of non-deposition and erosion (e.g. Forbes et al., 2010). In the Oman Mountains, a distinctive angular unconformity between Neo-Proterozoic and Middle Permian strata is preserved, which is here termed as the “Sub-Saiq Unconformity” (Figures 3 and 4).

The initial deposition of Khuff sediments started as a result of the drift of Cimmerian terranes away from Gondwana and the accompanying opening of the Neo-Tethys Ocean (e.g. Pillevuit, 1993; Stampfli and Borel, 2002). Subsequent flooding of the Arabian Plate resulted in the development of an epeiric carbonate ramp and the unconformable deposition of Khuff and equivalent sediments on folded Proterozoic strata (e.g. Glennie et al., 1974; Rabu et al., 1993; Searle, 2007).

According to the paleogeographic reconstructions of the Permian–Triassic by Konert et al. (2001a, b) the Arabian Plate drifted northwards, from 30° to 15° south. Sea-level oscillations of moderate amplitude and wavelength, and a transitional icehouse to greenhouse climate (Al-Jallal, 1995), created conditions very similar to those observed along the present-day Arabian Gulf coast (Strohmenger et al., 2006). According to paleogeographic reconstructions from Ziegler (2001), the deposition of Khuff strata took place on a shallow-marine to open-marine epeiric carbonate ramp (Figure 1).

Permian–Triassic strata in the Oman Mountains (Al Jabal al-Akhdar) are subdivided into two formations by Glennie et al. (1973) (Figure 3):

• Saiq Formation (Permian), time-equivalent to the Lower and Middle Khuff Formation (K7–K3 reservoir units),

• Mahil Formation (Triassic), time-equivalent to the Upper Khuff (K2 and K1 reservoir units), Sudair and Jilh formations.

Baud and Richoz (2013) explained that some authors apply a different lithostratigraphic scheme, also attributed to Glennie et al. (1974), in which the Saiq Formation corresponds to the entire Permian–Lower Triassic Khuff Formation, and the Mahil Formation to the overlying Triassic formations. In the range of this research project (this paper as well as Koehrer et al., 2010, 2011, 2012; Pöppelreiter et al., 2011; Zeller et al., 2011; Obermaier et al., 2012; Walz et al., 2013), we refer to the lithostratigrahic Lower Saiq/Mahil Boundary of Glennie et al. (1974), defined in the type locality of the Saiq Formation on the Saiq Plateau.

In the Al Jabal al-Akhdar outcrops, the Khuff-equivalent strata are divided into six sequences: Khuff Sequence KS6 to Khuff Sequence KS1 from bottom to top (Koehrer et al., 2010). These sequences are further subdivided into fourth-order cycle sets and fifth-order cycles.

Based on the Lower Saiq/Mahil Boundary of Glennie et al., 1974, the following lithostratigraphic units can be identified in Khuff-equivalent strata in the Oman Mountains:

• Sub-Saiq Unconformity: The contact between Saiq Formation (KS6) and pre-Permian metamorphic basement strata, represented by the Mistal, Hajir and Mu’aydin formations, forms a rather spectacular angular unconformity in the Oman Mountains (Figure 4).

• Lower Saiq Member: Conglomerates, sandstones, siltstones with paleosoils and ostracods (Rabu et al., 1986) transitioning into marine carbonates of the Upper Saiq Member. This unit might be a time equivalent to the lowermost Khuff in the subsurface or the Gharif Formation.

• Upper Saiq Member: Shallow to open-marine carbonate deposits of an epeiric ramp. Above the first 120 m of limestones, the Upper Saiq Member is entirely dolomitized. It is time-equivalent of the Lower and Middle members of the Khuff Formation in subsurface Oman (Osterloff et al., 2004), and KS6 to basal KS2 (Koehrer et al., 2010).

• Lower Mahil Member Koehrer et al. (2010, 2012): Shallow- to open-marine carbonate deposits of an epeiric ramp (entirely dolomite). Time-equivalent of the Upper Khuff Member in subsurface Oman (Osterloff et al., 2004) or middle part of Sequence KS2 and Sequence KS1 (Koehrer et al., 2010). Baud and Richoz (2013) propose that this unit be named the “Saiq unit C”.

First biostratigraphic studies of KS6 deposits which have been carried out on the Saiq Plateau by Montenat et al. (1976) suggested the lower part of Sequence KS6 to be of a Middle Murgabian (Wordian) age due to the presence of Neoschwagerina schuberti. Data from basal strata in the deep-water sections of the Hawasina Basin indicate a Roadian (Henderson and Mei, 2003) or Wordian (Kozur and Wardlaw, 2010) age according to conodonts Hindeodus excavatus and Hindeodus wordensis. Recent biostratigraphic studies from Forke et al. (2012) report the presence of Verbeekina grabaui and primitive (Afghanella? cf. tereshkovae) species in the basal part of KS6 in the Al Jabal al-Akhdar area, which indicates a stratigraphic range from the base of the Murgabian (late Roadian?/Wordian) to the lower part of middle Murgabian (Leven, 1997, 2009).

This study focuses on Sequence KS6 in the wadis on the northern rim of Al Jabal al-Akhdar region (Wadi Sahtan, Wadi Hajir, Wadi Bani Awf and Wadi Mistal) and one section from the southern flank (Saiq Plateau) (Figure 2). In all logged sections the lower part of KS6 consists of limestones with some meters of basal clastics at the base, while the upper interval of about 50 m is entirely dolomitized. Each outcrop, where sections have been logged, is located by UTM coordinates at its base and representative figures (Table 1).

A standardized logging sheet was used to describe five sections of Sequence KS6 in detail. GPS coordinates were captured for all sections (Figure 2). The following properties were recorded and digitized in WellCAD 4.2: texture (Dunham, 1962), sedimentary structures (biogenic and physical), lithology, components and spectral gamma-ray, integrated into the facies types and facies associations according to Koehrer et al. (2010). The color code used in WellCAD, correlation panels and 3-D models is presented in Figure 5.

More than 250 samples were taken for microfacies analyses in thin sections, which were carried out with a transmission light microscope. The integration of thin section analysis and field observations led to a highly reliable facies characterization in KS6 with calibrated facies logs and a detailed microfacies atlas (Figures 17 to 24).

A spectral gamma-ray (GR) survey was run in outcrops using a portable spectral GR spectrometer (model GS-512, manufactured by Geofyzika, Czech Republic). The spectrometer is equipped with a 3 x 3” NaI(TI) scintillation detector, which measures natural gamma-radiation of rocks. To detect overall GR trends usable for stratigraphic correlations and sequence interpretations, a sampling time interval of 15 seconds every 0.5 m was found to be sufficient after test measurements of 180, 90, 30 and 15 seconds (Koehrer et al., 2010). The authors are aware that quantitative analysis cannot be carried out using 15 second measuring time. However it is sufficient to capture and reproduce clear GR trends. The concentration of each of the elements are automatically calculated by the instrument and displayed in ppm (U, Th), % (K) or counts per second (cps, total GR). Although spectral GR was recorded only total GR was used for correlation and cycle interpretation since trend curves of U, Th and K seem not reproducible with a measuring time interval of 15 seconds.

Five KS6 sections in the Oman Mountains were correlated on a scale of 30 x 50 km (Figure 2). All correlation scenarios use the top of KS6 as the datum since this surface is well traceable throughout the study area. Various correlation scenarios were tested, focussing on the gamma-ray logs, carbonate facies and cycles. The preferred correlation strategy was that correlation lines follow paleolandscape surfaces, i.e. elevation changes associated with depositional profile, integrating gamma-ray, facies, and sequence-stratigraphic interpretations.

Facies associations were used to create a 3-D facies model with standard industry software (Petrel). Several modeling techniques were applied based on different correlation strategies in order to provide a wide spectrum of possible reservoir geometries (cf. modeling section).

In general, Sequence KS6 is subdivided into four units from the base to the top: (1) Basal Saiq Clastics (clastics, Figures 16 to 18); (2) transgressive hemi-sequence (limestone, Figures 19 and 20), (3) maximum flooding interval (limestone, Figures 21 and 22), and (4) and regressive hemi-sequence (dolomite, Figures 23 and 24).”

Description: In all outcrops where the base Khuff is exposed, the Sub-Saiq Unconformity is overlain by clastic deposits, 2–10 m thick, which are mixed upwards with an increasing amount of carbonate clasts (Figures 16 to 18). They are termed Basal Saiq Clastics and vary in thickness, composition and grainsize from outcrop to outcrop. On the Saiq Plateau they are composed of rooted siltstones with some ostracods (Rabu et al., 1986), whereas cross-bedded sandstones and conglomerates with imbrications dominate in the Wadi sections on the northern flank of the Oman Mountains. A variety of sedimentary structures, grain sizes and lithologies can be observed at the different outcrops of the Basal Saiq Clastics throughout the Oman Mountains.

Interpretation: The basal clastics are interbedded with fossil-rich Permian carbonates, which are interpreted to represent the initial Khuff transgression, and a transition from a terrestrial (lacustrine/alluvial) to a shallow-marine environment. The fossil-rich carbonate clasts within the Basal Saiq Clastics contain typical Khuff fossils (Forke et al., 2012). They are interpreted as basal Khuff deposits and not pre-Khuff sediments (e.g. Gharif-equivalent strata). The Early Permian continental clastics of the Gharif Formation present in southwest interior of Oman pinch-out towards the Oman Mountains in the northeast (Blendinger et al., 1990).

Description: The transgressive hemisequence of KS6 is characterized by bioclastic-rich beds (bioclastic wackestones, packstones, grainstones and floatstones (Figures 19 and 20). They contain a very diverse marine fauna (e.g. crinoids, fusulinids, coral heads, bivalve, brachiopod and gastropod shells) and beds are often graded or show bioturbation. Well-sorted and sometimes trough cross-bedded crinoidal pack- to grainstones are common especially in the lowermost part. They are interbedded with coral-rich units such as coral floatstones and coral framestones.

Interpretation: The very diverse marine fauna indicates fully open-marine conditions in a foreshoal environment. The cross-bedded, well-sorted crinoid columnals can be interpreted as high-energy bioclastic deposits which were probably deposited within range of wave energy in a shoal-like environment. The graded beds are likely to represent storm sheets (tempestites) forming meter-scale coarsening-upward cycles of a foreshoal and shoal margin environment. This interpretation attributes the transgressive hemisequence of KS6 to a shoal to foreshoal environment.

Description: In this interval dark colored mud-rich carbonates become more frequent (Figures 21 and 22). Graded storm sheets with abundant skeletal debris interfinger with thinly bedded to very massive dark-blue mudstones, which contain diverse ichnofabrics such as Zoophycus burrows. In general this interval is characterized by a very low biodiversity. The main fossils are bivalve shells, gastropods and gymnocodiacean algae as well as some calcispheres; corals and crinoids are absent.

Interpretation: The increasing lime mudstone volume indicates that with progressing transgression in the lower part of KS6, the overall deepening-upward trend from shallow-marine to open-marine carbonates continued. The lack of fauna, as well as the dark color, suggests less oxygenated waters and an azoic environment. Texture (mudstones, wackestones) and possible calcispheres, together with abundant gymnocodiacean algae (Permocalculus), indicate deep water far below the storm wave base (SWB) characterized by suspension settling, sediment starvation and muddy background sedimentation. This mudstone interval forms the so called “Muddy Marker” (Koehrer et al., 2010) and is interpreted as the maximum flooding interval of the lowest Khuff Sequence KS6 (Figure 3).

Description: The upper part of KS6 is about 50 m thick and entirely dolomitized. Low biodiversity, a high percentage of ooids and peloids, and a lack of crinoids are observed (Figure 23 and 24). Fossil content is restricted to bivalve and brachiopod shells and few foraminifers. Massive, cross-bedded and graded oolitic and peloidal grainstones are abundant. The KS6/KS5 sequence boundary (SB KS5) is placed on top of an up-to-50-cm-thick microbial laminite, which can laterally develop into rooted mudstone on a 10 km scale. This interval was termed “Microbial Marker 1” by Koehrer et al. (2010). Baud et al. (2012) described from the upper KS6 on the Saiq Plateau paleokarst features and ferrigenous crusts; however, in this study such features were not observed in the KS6 section on the Saiq Plateau.

Interpretation: Proximal storm deposits with some bioclasts but a generally low biodiversity, indicate high-energy conditions in a more restricted shallow-marine environment. The lack of crinoids, typical for regressive systems, can be observed in similar settings like in the Middle Triassic Muschelkalk in Germany (Palermo et al., 2010). Abundant cross-bedded peloidal and oolitic grainstones probably represent shoal bodies, which prograde over open-marine and foreshoal facies from the maximum flooding interval below. In general the regressive hemisequence can probably be placed into a backshoal to shoal environment with some influence from a rather proximal foreshoal.

In order to describe the development of the KS6 carbonate ramp two facies models are required. One for the transgressive part, a system with high biodiversity and no real shoal where the crinoidal grainstones form the most proximal facies (Figure 25). The other is for the regressive part with very typical shoal and backshoal facies (Figure 26). Thus Sequence KS6 shows a high degree of facies partitioning resulting in the need for separate facies models for the transgressive and regressive hemi-sequences.

Facies types show vertical stacking patterns that indicate a hierarchical cyclicity. The terminology to describe this cycle hierarchy was adopted from Kerans and Tinker (1997). Accordingly, cycles are stacked to form cycle sets, which can be grouped into high-frequency and composite sequences, which in turn build super-sequences. In general three different cycle orders can be detected in Sequence KS6, 1–5 m thick cycles, 6–15 m thick cycle sets and the overall Sequence KS6. Recent biostratigraphic studies indicate that the timeframe of KS6 ranges from late Roadian to late Wordian (Forke et al., 2012).

Since first possible Roadian fossil samples originate from around 15 m above the base of the KS6 the lowermost part could be even older. Thus it can be assumed that the time interval assigned to the ca. 190 m-thick KS6 approximately lies in between 4 and 7 million years (Myr) (Gradstein et al., 2012).

With a thickness of ca. 1–5 m these cycles represent the smallest scale of cyclicity within the studied section. The most significant characteristics of these small cycles are regular vertical changes in texture, grain-size and fossil content. Four basic cycle types defined by Koehrer et al. (2010) can be identified in the KS6 (Figure 27).

Cycles consist of a transgressive and regressive hemicycle. The facies stacking pattern varies along the depositional gradient. Due to the predominant foreshoal environment in KS6 they tend to have high grain content at cycle boundaries and highest mud content around the zone of maximum flooding. A maximum of 61 cycles are recorded in Sequence KS6, and given the 4 to 7 Myr duration, they would represent ca. 65–115 Kyr (thousand year) in duration. As in other Khuff sequences (Koehrer et al., 2012; Walz and Aigner, 2012) these 1–5 m cycles possibly represent a fifth-order or 100 Kyr Milankovitch signal.

Description: This cycle type is variably composed of stacked muddy, normally graded mid- to outer-ramp facies types. The thinner lower part usually consists of bioturbated mudstones to wackestones with various ichnofabrics. The thicker upper part consists of graded, commonly bioturbated packstones to mudstones and bioclastic packstones showing erosive bases and scour surfaces. These may turn upwards into massive, low-angle laminated peloidal packstones to grainstones.

Interpretation: Dark burrowed mudstones at the base of the cycle type indicate fully open-marine/deeper intra-shelf offshoal conditions and maximum relative water depth in a low-energy depositional environment around SWB. The rise to fall turnaround (interval of maximum accommodation) is defined at intervals with a maximum percentage of mudstone textures. 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.

Description: The lower part usually consists of graded storm beds or bioturbated mudstones to wackestones. The upper part mainly starts with amalgamated, graded bioclastic packstones to grainstones with erosive bases and frequent scouring. These may be overlain by thicker, low-angle laminated peloidal packstones to grainstones or coarse-grained intra-clastic grainstones and rudstones. They show an increase in sorting compared to the lower, bioturbated bioclastic packstone.

Interpretation: This cycle type represents the transition from a storm-dominated foreshoal environment to a higher-energy, shallower shoal flank 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. Storm influence is interpreted from normally graded beds. The regressive maximum occurs at the top of the packstone to grainstone that represents the time of maximum depositional energy and minimum accommodation.

Description: The thin lower part of the cycle type is represented by sheets of muddy mid- to outer-ramp facies such as graded/bioturbated storm beds. The thicker upper part starts with thick beds of bioclastic packstones to grainstones, thin layers of scoured graded beds or graded low-angle laminated peloidal packstones to grainstones. Upward these sediments may pass into massive amalgamated intra-clastic grainstones/rudstones. In most cases, facies grade into well sorted and cross-bedded peloidal or oolitic grainstones. In some cases the grainstones are overlain by microbial laminites or thin muddy lagoonal deposits.

Interpretation: The shallowing-upward trend, associated with an increase of energy, sorting and the change from skeletal to peloidal or oolitic grains, is interpreted as a prograding shoal body. The regressive maximum mainly occurs at the top of the peloidal-oolitic grainstone that represents the time of maximum depositional energy. Cycle caps are partly composed of muddy beds that represent further shoaling into a lower accommodation lagoonal/intertidal setting.

Description: Cycles of this type are dominated by grainy textures, i.e. stacked physically stratified shoal flank to shoal complex beds. Layers of densely packed, often imbricated rip-up clast packstones pass rapidly into graded packstones that turn into high-angle trough cross-bedded peloidal/oolitic grainstones during an overall coarsening up trend. These rather massive grainstone bodies make up the gross of this cycle type. They are covered, often above a sharp boundary, by microbial laminites. These “microbial-caps” may be replaced by bioturbated “muddy-caps” in some cases.

Interpretation: A coarsening up trend from shallow-water muddy beds to thick shoal beds in the lower part of the cycle records a clear increase in accommodation. It is associated with a rapid landward stepping of shoal bodies. The arrival of laminites or bioturbated wackestones heralds the seaward stepping of low-energy, shallow-water backshoal facies deposits during falling relative sea level.

In many outcrop sections, 6–15 m-thick cycle sets are the most obvious and easiest scale of cycles to recognize. They are often well-reflected in the gamma-ray signal, which makes them the smallest possible units to correlate (cf. 2-D correlation section). Cycle sets display the lateral movement of facies associations or belts triggered by medium-term changes of relative sea level according to Walther’s Law. While the lower, transgressive part of most sections shows offshoal (LFA 8) to foreshoal (LFA 7) transitions within medium cycles, the upper regressive part of Khuff Sequence KS6 shows the prevalence of repeated progradation and retrogradation of shoal complexes (LFA 5). Sequence KS6 contains a maximum of 17 cycle sets. Dividing the time interval of 4 to 7 Myr by this number would result in 235–412 Kyr per cycle set. Thus cycle sets possibly represent the fourth-order or 400 Kyr Milankovitch signal.

The entire KS6 shows an overall, 190 m-thick transgressive-regressive sequence. The lower part starts with terrestrial facies (Basal Saiq Clastics, Figures 16 to 18) and transitions into foreshoal carbonates (Bioclastic Facies, Figures 19 and 20). Around the maximum flooding, offshoal facies predominates (deeper-water facies, Figures 21 and 22) whereas the regressive part is composed of shoal to backshoal facies (oolitic/peloidal grainstones, Figure 23 and 24). High-frequency sequences as detected in Sequence KS5 (Walz and Aigner, 2012) and KS4 (Koehrer et al., 2012) could not be identified unequivocally.

Base Sequence Boundary: The base of Sequence KS6 is the Sub-Saiq Unconformity, where Permian sediments (Basal Saiq Clastics and Saiq carbonates) overlie Proterozoic strata. The unconformity below the Basal Saiq Clastics is erosive where high-energy sediments such as sandstones (Wadi Sahtan) or conglomerates (wadis Bani Awf and Bani Hajir) predominate. On the Saiq Plateau the contact is not erosive but the siltstones that are present there show intense rooting. In the context of the increasingly open-marine overlying sediments, the Basal Saiq Clastics represent the most proximal, most shallow facies in KS6 and their base is a clear sequence boundary.

Top Sequence Boundary: Microbial laminites (Saiq, Wadi Sahtan and Hajir) and rooted mudstones form the top of Sequence KS6.

Cycle Set Boundaries: In the context of the shoal to foreshoal environment of Sequence KS6, cycle set boundaries are often erosive surfaces. Higher-energy facies, such as grainstones or proximal storm deposits, erode lower energy facies, like distal storm and mud rich deposits.

Five sections in the Oman Mountains were correlated on a scale of 30 x 50 sq km. A mainly W-E transect runs from Wadi Sahtan to Wadi Mistal, and a N-S transect from the Saiq Plateau to Wadi Mistal. In general the Khuff epeiric carbonate ramp dips towards the north (Ziegler, 2001) (Figure 1). The deposition of the lower part of KS6, however, was probably strongly controlled by the initial topography created by the Sub-Saiq Unconformity.

Various correlation scenarios were applied based on sequence-stratigraphic interpretations as well as on facies and gamma-ray trends. The effect of the different correlation scenarios on potential reservoir geometries are discussed, modeled and displayed in the 3-D modeling section. Most correlations use the top KS6 (“Microbial Marker 1”) as datum (Figures 28 to 30b), some use the maximum flooding surface (Muddy Marker) (Figures 31 and 32).

The thickness of KS6 varies considerably from west to east (Figures 28 to 32) with the lowermost part of the KS6 seems to be missing in the east. These thickness changes can be explained by a paleohigh in the east. Consequently correlation lines in all correlations onlap onto the interpreted paleohigh in the east (Figures 28 to 32).

Although GR values cannot be directly translated into Dunham texture, the values tend to increase around muddy facies and decrease around grainy facies. Mudstones form in low-energy conditions and are therefore more likely to contain clay, which creates higher GR values in the sedimentary record. Figure 28 illustrates how the cyclic pattern of the GR signal allows a simple peak-to-peak correlation. The correlation based on GR only results in a layer-cake pattern with the maximum flooding surface being interpreted at a GR high with good correlatability. Although the GR signal seems cyclic throughout most of Sequence KS6 a pure GR correlation probably reveals only lithostratigraphic time lines and does not follow paleolandscape surfaces.

This correlation scenario focused on the Dunham carbonate textures, key components and marker beds (Figure 29). The main marker bed is formed by an interval that can be found in all logged sections: the so-called “Muddy Marker”. Although its appearance varies with respect to the abundance of mudstones, it is traceable through all sections; its thickness increases towards the west. A second marker is the microbial bed at the top of Sequence KS6 (“Microbial Marker 1”) in Wadi Sahtan and Wadi Bani Awf, which laterally changes into a rooted mudstone in Wadi Hajir and Wadi Mistal.

The occurrence of specific fossils was further used to correlate the KS6 sections. Examples are crinoidal floatstones (rather distal setting), which are correlated with crinoidal grainstones (more proximal setting). In the lower KS6 facies, texture and fossils apparently follow a proximal-to-distal trend from east to west. The upper portion of Sequence KS6 generally shows less lateral heterogeneity than the interval beneath the maximum flooding.

Compared to a pure GR correlation, correlation lines follow proximal-to-distal trends from east to west. However, the resulting correlation shows that some of the timelines dip in different directions. Such a pattern would be relatively unlikely since the time lines do not consistently follow paleo-landscape surfaces.

In this scenario, correlation lines were chosen to honor both GR and textural trends (Figures 30a and 30b). In addition, time lines strictly mimic proximal-to-distal paleolandscape surfaces, and follow sequence-stratigraphic principles. During the transgressive hemi-sequence, an aggradational-retrogradational stacking pattern with onlaps onto the paleohigh is apparent. In contrast, the regressive hemi-sequence shows a progradational stratigraphic architecture with shingles above the maximum flooding, with correlation lines dipping towards Wadi Sahtan in the west (dipping angle < 0.06°). Additional fifth-order cycles appear below the top of the KS6 towards the more distal sections in Wadi Sahtan to the west and the Saiq Plateau to the south forming toplaps. A fifth-order cycle pinch-out (downlap) around the maximum flooding surface can be observed in Wadi Sahtan. Toplap, downlap features and the pinch-out of cycles are indicated by correlation and can not be observed on an outcrops scale.

In the context of a syn-rift phase, cycles pinch-outs and thickness variations are probably influenced by a combination of differential subsidence and initial topography. The key factor for the correlation of KS6 appears to be the paleorelief associated with the Sub-Saiq Unconformity. The first deposition of Khuff carbonates occurred in paleolows forming onlapping geometries on paleohighs. This depositional pattern explains the strong thickness variations in KS6 with thicknesses of up to 190 m in paleolows and about 90 m around potential paleohighs. Initial topography also contributed to facies variations between different sections (e.g. rooted mudstones in Wadi Mistal and Wadi Hajir turn laterally into microbial laminites in Wadi Bani Awf and Wadi Sahtan).

The undulating initial paleotopography probably affected different depositional trends and the shapes of facies belts. More grainstones appear around the paleohigh in the east and more muddy facies types in Wadi Sahtan to the west. The maximum dipping angle of correlation lines of 0.06° appears steep compared to KS4 to KS1 (Koehrer et al., 2012) where the maximum dip angle is around 0.001°. The steeper-dipping correlation time lines in KS6 are probably a result of the underlying topography.

Previous studies in Khuff sequences KS1 to KS3 on the Saiq Plateau in Oman (Zeller et al., 2011) revealed that single-grainstone bodies may show pinching and swelling geometries on a 1 x 2 sq km scale, but reflect an overall layer-cake geometry. Correlations on an 8 x 8 sq km scale for sequences KS1 to KS3 on the Saiq Plateau (Koehrer et al., 2011) indicates single shoal bodies (LFA 5) to be traceable if they reach a certain thickness. The KS6 correlation in the Oman Mountains is on a subregional scale (30 x 50 sq km). Sparse data coverage (five sections with an average spacing of ca. 15 km) creates high uncertainty for the correlatability of single grainstone bodies. The upper, regressive part of KS6 contains more than 50% of grainstones throughout all sections. Based on this observation it can be assumed that the percentage of grainstones in between the sections is about as high. However this assumption does not reveal the lateral geometry of single grainstone beds.

In order to visualize the potential differences between laterally extensive and laterally limited grainstone bodies, two correlation scenarios for grainstone units are displayed in Figures 31 and 32. These scenarios visualize two possibilities on how laterally extensive or confined grainstone bodies could be. They mirror how the interpretation on this scale by the correlating geologist can influence reserve estimations and production forecasts. Observations in the field could only prove that individual grainstones beds are traceable over hundreds of meters. Both illustrations do not incorporate the basal part of KS6, and therefore the number of fourth-order cycle sets is different to the number in other figures. In Figure 31 laterally continuous grainstones (ca. 15 km correlation lengths) form large, connected reservoir bodies. In the lower part, grainstones mainly composed of bioclasts, do not correlate over wide areas; they are confined by the paleohigh to the east and pinch-out to the west. In the upper part of KS6 where peloidal/oolitic grainstones predominate in all sections, larger volumes and a high connectivity can be assumed.

In the correlation shown in Figure 32, single potential reservoir bodies were assumed to form laterally more confined patches with horizontal correlation lengths of about 5 km. Even in upper KS6, where there is an overall very high percentage of grainstones, the single units seem often laterally disconnected in 2-D. Although single-grainstone units might not be very extensive, the overall chance of amalgamation is expected to be very high when the overall percentage of grainstones is sufficiently high (upper KS6 > 50%). This implies for the upper KS6 that even if grainstones are restricted in their lateral extent, they may still form connected potential reservoirs.

To illustrate possible reservoir facies distribution and depositional trends, facies associations (backshoal, shoal, foreshoal and offshoal) from outcrops logs were imported into Petrel to create a 3-D model.

Since outcrop observations have shown that the reservoir facies seem laterally continuous over kilometers, the horizontal cell size was set to 1 km x 1 km. After the correlation of reservoir bodies in 2-D (Figures 30a and 30b), tops of cycle sets were correlated in Petrel. The resulting surfaces where transformed into horizons and zones were assigned in between. Therefore most zones represent one complete fifth-order cycle. Only around the maximum flooding interval (“Muddy Marker”) down-lapping hemicycles are interpreted, which led to the described shingle-like architecture (Figure 30a, in between Wadi Sahtan and Wadi Bani Awf).

Subsequently zones were subdivided into layers. Depending on thickness, a high number of layers were allocated to each zone to ensure good vertical resolution. The resulting models comprise 16 zones, 1,380 layers and a total number of about 2 million cells. Each cell has an average size of 1.0 by 1.0 sq km and a thickness of 0.13 m. The smallest recorded bed in the field measures about 0.1 m in thickness.

Each cell in the model was populated with a value for each facies association by upscaling the well logs using the “most-of averaging” method, so the predominant facies fills the whole cell. Vertical facies proportions derived from the outcrop data were used for all models to automatically define the percentage of each facies in each zone. Thus the proportion of each facies association within one zone (equivalent to one cycle set) is derived from the proportions in outcrops.

The method “Truncated Gaussian with Trends” (TGS) modeling algorithm was used to areally distribute facies associations within the model. The TGS method was selected because it combines statistical methods with geological concepts such as facies trend maps. Specific trend maps were generated for each zone. Since a variogram cannot be calculated with the presented data density, an isotropic variogram was chosen (5 and 15 km range). As the correlation revealed (e.g. Figures 30a and 30b) facies belts in KS6 follow a proximal-to-distal trend from the paleohigh in the northeast (Wadi Hajir and Wadi Mistal) to south and west (Saiq Plateau, Wadi Bani Awf and Wadi Sahtan).

Previous studies suggest that thicker grainstone bodies are laterally continuous for more than 8 km (Koehrer et al., 2011). Other studies, e.g. the Triassic Muschelkalk in SW-Germany, (Palermo et al., 2010), showed that grainstone bodies could be traceable over distances of up to 10–20 km. To take the different possibilities into account, facies modeling was performed with different lateral facies variograms. Two simulations were run and compared with respect to the resulting connectivity of potential reservoir facies (Figures 33 and 34).

1. Lateral extent of grainstone bodies 15 km: isotropic variogram with 15 km horizontal facies range

2. Lateral extent of grainstone bodies 5 km: isotropic variogram with 5 km horizontal facies range

Both models portray the general vertical deepening-shallowing trend in the KS6 (Figures 33 and 34): Some shoal grainstone facies is present in the lower KS6, barely any in middle KS6 around the MFS while the upper part of KS6 is dominated by shoal facies association. Most offshoal facies can be observed near the maximum flooding interval. Some backshoal facies can be found at the top of the KS6. Laterally the most proximal facies concentrates around Wadi Mistal with a deepening trend towards southern and western sections (Saiq Plateau, Wadi Sahtan). Shingles can be observed around the maximum flooding and at the top of the KS6 (Figure 33 and 34, marked in yellow).

A potential reservoir volume was delineated, calculating the percentage of potential reservoir facies association in each zone. Approximately 40% of KS6 consists of potential reservoir facies (i.e. shoal LFA, grainstones). The regressive part of KS6 consists of more than 50% potential reservoir facies, the lowermost part of 33%, and the middle transgressive around the maximum flooding of only 6% (Table 2). Although the overall percentage of potential reservoir facies stays the same in all models because the vertical proportion is honored, the lateral extent of grainstones can make a difference when it comes to the connectivity between single grainstone shoal bodies.

Figure 35 illustrates how the lateral size affects the reservoir connectivity in single zones with different grainstone percentage. The models with 5 and 15 km facies range are filtered so that they show exclusively the shoal facies association (red). Two zones (A and B, see Figure 35) are presented in top view (A) from the top of the KS6, where grainstones have an abundance of over 50%; and (B) from the interval of maximum flooding where grainstones account for less than 10% of the facies. It can be observed that in zone A, where grainstones are abundant (> 50%), the lateral extent of single shoal bodies apparently does not have a significant effect on connectivity since grainstones amalgamate and form one, laterally connected unit. In zone B, where grainstones are rare they form single isolated volumes which are not inter-connected. Thus the lateral extent of single grainstone units does not effect the connectivity if the overall percentage of grainstones is high enough (approx > 50%).

• 1) Five outcrop sections of Lower Khuff (KS6) time-equivalent strata (Saiq Formation) were investigated in the Oman Mountains (Al Jabal al-Akhdar, Sultanate of Oman). The KS6 can be subdivided into four different facies associations (backshoal, shoal, foreshoal, offshoal). The KS6 forms a third-order transgressive-regressive sequence, built by smaller-scale cycle sets and cycles.

• 2) The initial paleorelief apparently controlled the thickness and distribution of the Basal Saiq Clastics, and that of the overlying KS6 carbonates as well as their composition. The correlation strategy to follow paleolandscape surfaces using all available data shows that the stratigraphic architecture is aggradational with onlaps against the paleohigh during the transgressive hemi-sequence and displays subtle shingles during the regressive hemi-sequence.

• 3) This study revealed potential reservoir units in the KS6, commonly regarded as non-reservoir in the subsurface of Oman. In the transgressive part, the predominant reservoir facies are bioclastic crinoidal grainstones (with only poor diagenetic potential), concentrated around the margin of a paleohigh. In contrast, oolitic/peloidal grainstones in the upper regressive part (with a much higher diagenetic potential) have a much more widespread distribution.

• 4) A range of 3-D models were generated using different correlation scenarios and varying lateral extents of potential reservoir grainstones. The lateral continuity of shoal grainstones bodies seems to have no to little effect on their inter-connectivity if the overall percentage of reservoir facies is high enough (> 50%) as single reservoir units amalgamate.

This study is part of an extra-mural research project of the University of Tuebingen with Qatar Shell and Petroleum Development Oman. We would like to thank Shell and PDO for their financial support and M. Poeppelreiter, J. Amthor, A. Brandenburg, J.-M. Dawans, G. Forbes and J. Schreurs for their assistance. PDO and the Omani Ministry of Oil and Gas are thanked for permission to publish the paper. We are grateful to our Sedgeo members of the University of Tuebingen: L. Walz (now Shell), M. Haase (now ExxonMobil) and M. Bartenbach (now Statoil), in addition to P. Jeisecke who prepared the thin sections. H. Forke is thanked for biostratigraphic analysis of the thin-sections. Shuram Oil and Gas (Muscat) is acknowledged for fieldwork logistics. We are also very grateful to ALT and Schlumberger for providing access to WellCAD and Petrel software packages. GeoArabia’s Assistant Editor Kathy Breining is thanked for proofreading the manuscript, and GeoArabia’s Production Co-manager, Nestor “Nino” Buhay IV, for designing the paper for press.

Daniel Bendias studied Geosciences at the University of Tuebingen (Germany). His diploma thesis (2010) at the University of Tuebingen was on the Paleorelief-influenced facies and sequence patterns in the Lower Khuff time equivalent strata (Sultanate of Oman). He is currently working as Research Associate and PhD student at the Center for Applied Geosciences (University of Tuebingen). His PhD thesis, funded by Petroleum Development Oman (PDO), focuses on sequence stratigraphy, reservoir and seal geobodies of the Jurassic Mafraq Formation, a mixed carbonate siliciclastic system in outcrops and subsurface of Oman.

Daniel.bendias@gmail.com

Bastian Koehrer is a Development Geologist in Wintershall’s German business unit, working on mature oil field and tight gas sands development in the German North Sea and Lower Saxony. He has more than five years of E&P project experience in Germany, Oman, Qatar and the UAE with a professional track record in both carbonate and clastic reservoirs. Bastian obtained a PhD degree (2011) in Carbonate Sedimentology from the University of Tübingen (Germany) in research collaboration with Shell (Qatar) and Petroleum Development Oman. For his PhD dissertation on the Khuff Formation, Bastian spent 18 months of outcrop mapping in the Sultanate of Oman. Bastian is a member of the EAGE, AAPG and DGMK and has published several papers on carbonate sequence stratigraphy and reservoir outcrop analogs.

bastian.koehrer@wintershall.com

Michael Obermaier studied Geosciences at the Universities of Tuebingen (Germany), and Miami (Florida, USA). His PhD thesis (2013), a research cooperation between the University of Tuebingen and Petroleum Development Oman was on Triassic reservoir and seal characterization in outcrops and subsurface of Oman. Since 2013 Michael has been working as a Carbonate Geologist for Shell Global Solutions in Rijswijk, The Netherlands.

michaobm@gmail.com

Thomas Aigner studied Geology and Paleontology at the Universities of Stuttgart, Tuebingen/Germany and Reading/UK. For his PhD dissertation on storm depositional systems (1985) he worked at the Senckenberg-Institute of Marine Geology in Wilhelmshaven (Germany) and spent one year at the University of Miami in Florida (USA). He then became an Exploration Geologist at Shell Research in Rijswijk/Holland and Houston/Texas focussing on basin analysis and modelling (1985–1990). Since 1991 Tom has been a Professor and Head of the Sedimentary Geology Group at the University of Tuebingen. In 1996 he was a “European Distinguished Lecturer” for the AAPG. His current projects focus is on sequence stratigraphy and reservoir characterisation/modelling in outcrop and subsurface.

t.aigner@uni-tuebingen.de