The Al Jabal al Akhdar and Saih Hatat culminations in the central Oman Mountains expose the complete mid-Permian to Late Cretaceous (Cenomanian) passive shelf and margin carbonate sequence beneath the allochtonous slope (Sumeini Group), basin (Hawasina complex), distal ocean-trench (Haybi complex) facies rocks, and the Semail ophiolite thrust sheets that were emplaced from NE to SW during the Late Cretaceous. Reconstruction of the pre-thrust sequences shows that time-equivalent rocks occur in successively stacked thrust sheets from shelf to slope to basin. The Al Jabal al Akhdar structure is a 60 km wavelength anticline plunging to the northwest beneath the Hawasina Window and with a fold axis that curves from WNW-ESE (Jabal Shams) to NNE-SSW (Jabal Nakhl). The structure shows little internal deformation except for minor intra-formational thrust duplexing within the Cretaceous shelf stratigraphy along the northern margin. The upper structural boundaries around the flanks of the shelf carbonate culminations have been re-activated as late stage normal faults. The Semail thrust formed a passive roof fault during late-stage culmination of al Al Jabal al Akhdar such that the ophiolite rests directly on Wasia Formation top-shelf with the entire Sumeini, Hawasina and Haybi thrust sheets displaced around the margins. NE-directed backthrusting and intense folding in the northern part of the Hawasina Window affects all allochtonous units and is related to a steep ramp in the Late Cretaceous shelf margin at depth.

The Saih Hatat culmination is another 40 km half-wavelength anticline in the central Oman Mountains, but shows extreme deformation in the form of recumbent folds, sheath folds with NNE-trending axes and thrusting along the northern margin. High-pressure carpholite, blueschist and eclogite facies rocks are exposed at successively deeper structural levels, separated by high-strain normal sense shear zones. There is no evidence for a separate ‘North Muscat microplate’ or an intra-continental subduction zone, as previously proposed; all high-pressure units can be restored to show their pre-deformation palaeographic positions along the northern margin of the Arabian Plate. Both Al Jabal al Akhdar and Saih Hatat are Late Cretaceous culminations, folded after obduction of the Hawasina, Haybi and Semail ophiolite thrust sheets from northeast to southwest during the period Turonian to Campanian-Lower Maastrichtian. Maximum compressive stress along the central Oman Mountains was oriented NE-SW, parallel to the ophiolite emplacement direction, but a second compressive stress axis was oriented WNW-ESE, either concurrently or slightly later in time, resulting in a dome and basin structural geometry. The biaxial fracture pattern in the foreland, southwest of the Oman Mountains could be explained as a result of the WNW-directed emplacement of the Masirah ophiolite belt and Batain mélange during the Campanian-early Palaeocene.

Both Al Jabal al Akhdar and Saih Hatat were positive topographic features at the end of the Cretaceous with Upper Maastrichtian and Palaeogene sediments onlapping both flanks. In Jabal Abiad, these Palaeogene sediments have been uplifted by at least 2 km since the Late Miocene-Early Oligocene associated with minor NNE-SSW compression. Tertiary shortening, folding and thrusting increases to the north in the Musandam peninsula where the first effects of the Arabian Plate-Eurasian Plate (Zagros belt) continent-continent collision are seen.

Three major culminations of Permian - Cenomanian shelf carbonate rocks occur along the Oman Mountain range, from NW to SE the Musandam, Al Jabal al Akhdar and Saih Hatat culminations (Figure 1). Each is a very large-scale antiformal fold that also affects the structurally overlying Sumeini, Hawasina, Haybi and Semail ophiolite thrust sheets. Palinspastic reconstruction of the Oman continental margin has shown that the allochthonous thrust sheets contain time-equivalent units but of progressively more distal facies structurally up-section (Figure 2; Glennie et al., 1973, 1974; Searle et al., 1983; Rabu et al., 1993). The reconstructed shelf margin and basin for the Central Oman Mountains is shown in Figure 3. Several basin-wide stratigraphic markers have been traced from the proximal Sumeini Group to the most distal Hawasina radiolarian cherts (Searle et al., 1983; Figure 3).

Regional mapping of the Oman Mountains was carried out first by the Shell team of Glennie et al. (1973, 1974) who also defined the stratigraphy, and proposed a thrust emplacement model for the Semail ophiolite and Hawasina complex rocks, a model that is still widely accepted today. During the 1970–1980s, a USGS team mapped a strip across Saih Hatat and the Ibra ophiolite block from Muscat south to Ibra (Bailey, 1981) and an Open University team mapped much of the northern Oman Mountains, concentrating mainly on the Semail ophiolite complex (Lippard et al., 1986). The BRGM carried out detailed mapping at 1:100,000 scale of much of the Oman Mountains and summaries of their work have been published by Rabu et al. (1993) and Le Métour et al. (1990). The northern Saih Hatat region has been studied in some detail following the discovery of high-pressure (HP) eclogites, blueschists and carpholite-bearing rocks (Le Métour et al., 1986; Goffé et al., 1988; El-Shazli et al., 1990; Searle et al., 1994, 2004; Miller et al., 1998, 1999). Recently, debate has centered on the timing of uplift of the Al Jabal al Akhdar and Saih Hatat culminations (Late Cretaceous or Tertiary) and the tectonic evolution of the Oman Mountains and the foreland.

Two recent models in particular have drawn attention to the key questions of timing of high pressure metamorphism and subduction, and the polarity of the subduction system. Gregory et al. (1998), Gray and Gregory (2000) and Gray et al. (2000) proposed a new tectonic model involving ‘nascent’ SW-directed subduction of the Muscat microplate (As Sifah and Hulw units) at 130–95 Ma prior to, and unconnected with ophiolite emplacement. Sedimentation along the Oman shelf during this time (Kharaib, Shu’aiba, Nahr Umr and Natih formations) was clearly related to passive margin sedimentation, and there is no evidence of any igneous activity as would be expected above an active subduction zone. This model was largely based on old (pre-ophiolite) 40Ar/39Ar ages of white micas from the high pressure units and the occurrence of NE-directed shear zones and NE-facing folds. Arguments against this model have been put forward by Searle et al. (2003, 2004) and Warren et al. (2005). They have questioned the validity of the older Sm-Nd garnet ages and 40Ar/39Ar phengite ages in the light of new and precise U-Pb ages of zircons from the ophiolite and amphibolite metamorphic sole.

Breton et al. (2004) presented a tectonic model involving two NE-dipping subduction zones, an outer one beneath the Semail Ophiolite and an inner intra-continental subduction zone beneath the ‘North Muscat microplate’ comprising the outer part of the Arabian platform (As Sifah unit), the slope and the Hawasina basin. This model, like the Gregory et al. (1998) one, requires a suture zone to separate the As Sifah eclogite unit from overlying structural units. Structural mapping and restoration clearly shows that there was no suture zone above the As Sifah rocks, which are in fact outboard, time-equivalent units to the inner shelf (Searle et al., 2004, Figures 3 and 4).

This paper describes the structures in the Al Jabal al Akhdar, Hawasina Window and Saih Hatat culminations with the aim of unraveling the Late Cretaceous and Tertiary structural evolution. Maastrichtian and Tertiary rocks are only exposed around the margins of each culmination, not over the main structures, so the Tertiary jabals are also examined along the flanks of the mountains. Firstly, a brief history of the evolution of the Oman shelf and margin is described.

Permian-Triassic Rifting Phase

The earliest rifting deposits along the shelf margin are dated as Early Permian in the Jabal Qamar exotic in the Dibba zone United Arab Emirates (UAE) and mid-Permian (Saiq Formation) around the Al Jabal al Akhdar (Figure 5a,b,c) and Saih Hatat shelf carbonate culminations (Glennie et al., 1973, 1974). The Jabal Qamar exotic in the Dibba zone is the only exotic in Oman to show a pre-Permian basement (Ordovician Rann Formation, and Lower Carboniferous Ayim Formation) indicating that it was a rifted block of the Arabian margin (Hudson et al., 1954; Pillevuit et al., 1997). The earliest sedimentary rocks in the slope facies Sumeini Group in the Jabal Sumeini area are late Permian - Triassic Maqam Formation (Watts and Garrison, 1986; Watts 1990), equivalent to the Khuff-Saiq Formation on the autochthonous shelf. The oldest radiolarian cherts in the Hawasina complex are also late Permian (255 Ma; De Wever et al., 1988; Cooper, 1990), suggesting that the shelf-slope-basin distinction along the Oman margin was already in place by Late Permian time (ca. 268 Ma).

The Middle Permian unconformity around Al Jabal al Akhdar and Saih Hatat abruptly truncates fold axes and bedding in the Late Proterozoic - Early Palaeozoic rocks below, indicating a mid-Palaeozoic deformation event (without a major metamorphic imprint). Early Permian continental siliciclastics of the Gharif Formation thin to the north and are overlapped by the Late Permian transgression (Blendinger et al., 1990). Drowning of the Arabian margin occurred as the ‘Fusulinid sea’ covered the rifted continental margin across Oman (Glennie et al., 1974; Rabu et al., 1990). Whereas in Al Jabal al Akhdar, the Permian is inner shelf facies, around Saih Hatat the facies is more outer-shelf and continental-slope facies (Pratt and Smewing, 1990). Mildly alkaline rift-related volcanic sills and flows occur in the Saiq 1 Formation in Saih Hatat (Le Métour et al., 1986). These volcanic flows are interpreted to form the protolith of the As Sifah eclogites at the deepest structural levels of the Oman margin exposed today (Searle et al., 1994, 2004).

The considerable thickness of Upper Permian along northern Saih Hatat indicates that the true continental margin must have been located some way outboard (NE) of the present coastline. Permian exotic limestones are exposed in the thrust sheets beneath the Semail ophiolite associated with minor alkali basalts, interpreted as within-plate ocean-island-type volcanics (Searle et al., 1980). The most prominent Permian exotic block is the Baid exotic south of Saih Hatat, an isolated piece of rifted Late Permian platform underlain by alkali basalts along the base. Permian rifting was followed by sea-floor spreading in the Hawasina (Tethys) Ocean. The earliest pelagic radiolarian cherts and red ammonoid limestones at the base of the allochthonous Hamrat Duru Group (proximal Hawasina complex) are Late Permian (De Wever et al., 1990; Béchennec et al., 1990). These deep-sea sediments were probably deposited on alkaline oceanic crust (Searle et al., 1980) rather than stretched continental crust (Béchennec et al., 1990; Rabu et al., 1993) or true mid-oceanic-ridge basalt (MORB) oceanic crust (Glennie et al., 1974; Bernoulli and Weissart, 1987; Pillevuit et al., 1997).

Permian-Triassic Volcanism

Permian volcanic rocks occur within the shelf carbonates (Saiq 1 Formation) in Saih Hatat, along the base of many of the Hawasina thrust sheets, and in the distal Haybi complex (Searle et al., 1980; Maury et al., 2003). The lavas and flows within the shelf sequence and in the proximal Hawasina are all high-titanium alkaline lavas related to the rifting of the Arabian margin. True MORB volcanics are not present and it is unlikely that true MORB oceanic crust was formed until at least Late Triassic. Maury et al. (2003) related the Permian volcanism in Oman to a mantle plume, but the widespread distribution extending over a distance of 350 km from the Dibba zone in the UAE to the eastern Oman Mountains, and the palaeogeographic distribution from the shelf margin outboard to Haybi complex preclude a plume origin. It seems far more likely that the Permian volcanism was related to continental rifting and break-up, with more localised within-plate, off-axis volcanism related to ocean-island guyots formed off the passive continental margin during the Late Permian and Late Triassic (Glennie et al., 1974; Searle et al., 1980).

In the most distal Haybi complex thrust sheets nephelinites, ankaramites, alkali basalts and trachytes represent extremely alkaline lavas, with rare, but widespread intrusions of highly alkali peridotites (jacupirangites, biotite wehrlites) and kaersutite gabbros beneath exotic limestones (Searle, 1984). These rocks are characteristic of present-day ocean islands offshore passive continental margins, such as the Comores, Tristan da Cunha, Canary Isles etc. A similar tectonic position is thought to be comparable for the origin of the Oman volcanics. The Haybi volcanics formed the protolith of the sub-ophiolite metamorphic sole amphibolites (Searle and Malpas, 1980, 1982). Thus the footwall of the ophiolite was old, cold Triassic and Jurassic ocean floor (Searle and Cox, 1999, 2002), not Cretaceous ocean floor equivalent to the Semail ophiolite (Boudier et al., 1988).

Oman Exotics: Baid and Misfah Platforms

Two types of ‘Oman Exotic limestones’ occur in the allochthonous stack structurally above the shelf carbonates. The first group includes large-scale, well-bedded carbonate banks frequently with an alkali basalt basement. Examples are Jabals Misfah, Kawr and Misht (Figures 5c and 6a) around the SW flank of Al Jabal al Akhdar, and the Baid exotic south of Saih Hatat. The second group includes numerous smaller exotics, frequently intimately associated with within-plate, highly alkaline volcanics, often occurring in mélanges and imbricate slices immediately beneath the Semail ophiolite (Searle et al., 1980; Searle and Graham, 1982; Pillevuit et al., 1997).

The Baid platform carbonates comprise Late Permian platform limestones, Triassic (Hallstatt facies, red ammonite-rich limestones) overlain by thin cherty limestones and radiolarites of Early Jurassic age (Blendinger et al., 1990; Pillevuit et al., 1997). This platform sequence probably rifted away from the Oman continental margin as it occurs northeast outboard of the true shelf and the proximal Hawasina rocks in the palaeogeographic reconstruction (Bernoulli and Weissart, 1987; Bernoulli et al., 1990). These exotics have been thrust to the SW above Hamrat Duru (Hawasina complex) thrust sheets that are structurally above the Cretaceous shelf and its cover of Aruma Group, and immediately beneath the Semail ophiolite.

In Jabals Misfah (Figure 5c), Kawr and Misht (Figure 6a), structurally above the Al Jabal al Akhdar shelf sequence, 700–800 m thick Late Triassic platform limestones (with Late Dogger Ammonico Rosso facies) up to 1,000 m thick, were deposited in 10–15 my. These limestones are similar to the Triassic shelf (Mahil Formation), but overlie an alkaline basaltic substrate indicative of off-axis, within-plate ocean-island volcanic rocks (Searle et al., 1980; Searle and Graham, 1982). On the Jabal Misfah exotic, foundering of the Triassic seamount occurred during the Late Oxfordian to Tithonian with deposition of thin deep-water pelagic limestones and cherts of the Nadan Formation at the top (Pillevuit et al., 1997). Thermal subsidence following sea-floor spreading during the Late Triassic, resulting in subsidence and drowning of the Triassic exotic carbonate platform.

These exotic limestones have been thrust over proximal Hamrat Duru Group and Al Aridh Formation (Hawasina complex) rocks and occur immediately beneath the Semail ophiolite, indicating their palinspastic restoration position well outboard of the shelf and shelf margin (Figure 3).

Triassic Ocean-Island ‘Exotic’ Limestones

Numerous Late Triassic exotic limestones occur throughout the Haybi complex thrust sheet immediately beneath the ophiolite (Glennie et al., 1973, 1974; Searle and Graham, 1982). These rocks were later termed part of the Umar Group (BRGM mapping; Béchennec et al., 1990; Rabu et al., 1993); however they span Late Permian to Middle Cretaceous, are time-equivalent rocks to the Hawasina and Sumeini thrust sheets and the autochthonous shelf, and are bounded by thrust faults, so the structural term Haybi complex is retained here. Many Triassic exotics show spectacular Megalodon sp. limestones representing reef limestones, with flanking mega-breccias representing ocean-island slope deposits. Blocks of shallow-water Triassic exotic limestone are sometimes encased in deep-water radiolarian cherts indicating base of slope facies.

The presence of alkali basalts, and some highly alkaline peridotite-gabbro intrusions into the Late Triassic exotic basements suggest that their origin was on a series of isolated within-plate, off-axis oceanic volcanic island seamounts. These Triassic oceanic floor rocks of the Haybi complex restore to a palaeogeographic position immediately SW of the Semail thrust subduction zone (Figure 3). The restoration shows that old, cold Triassic - Jurassic oceanic crust was subducted beneath the Semail ophiolite to form the amphibolite-greenschist facies metamorphic sole, very different from the Late Cretaceous of the Semail ophiolite (Searle and Malpas, 1980, 1982; Searle and Cox, 1999, 2002).

Late Jurassic - Early Cretaceous Drowning of Platform

The Jurassic - Cretaceous boundary (145 Ma) is marked by a hiatus along the Oman platform and preceded a drowning event that lasted ca 20 my (Pratt and Smewing, 1990). Shallow-water shelf carbonate deposition was resumed during the Albian-Aptian (Rabu et al., 1990). Increasing subsidence across the shelf and slope into the Hawasina basin may have occurred in response to stretching during sea-floor spreading. Along the southeastern margin however the NNE-trending Huqf High remained a positive feature into the Cretaceous. This basement high probably extended north into the eastern Oman Mountains where basement rocks exposed at Jabal Ja’alan crop out beneath the Tertiary unconformity. This basement high may have had a profound effect on the subsidence within the foreland basin and may also have divided the dominant NE-SW compressive stress regime to the NW from the WNW-ESE compressive stress regime along the Masirah-Batain coast.

Late Cretaceous Platform Margin Collapse

A large-scale gravity low over the Oman Mountains foreland has been interpreted in terms of elastic flexure of the Arabian foreland beneath the allochthonous thrust sheets (Ravaut et al., 1997). The Late Cretaceous Aruma foredeep basin includes up to 3.5 km thickness of Santonian - Campanian Fiqa Formation with up to 1.5 km thickness of Late Campanian Juweiza Formation thickening to the NE. Major changes in platform margin sedimentation occurred following the Cenomanian when the whole continental margin collapsed as recorded by large-scale megabreccias, intra-formational unconformities and transgression to deep water clastic sediments of the Qumayrah Formation along the margin and the time equivalent Aruma Group (Robertson, 1987; Watts, 1990). Collapse of the shelf margin and beginning of the foreland basin deposition following the ‘Wasia-Aruma break’ (Turonian) was accompanied by uplift along the peripheral bulge. Up to 1,100 m of uplift and erosion of the Natih Formation on the Lekhwair High may have been caused by flexural bending of the foreland (Warburton et al., 1990; Loosveld et al., 1996; Filbrandt et al., 2006) along the Suneinah Trough (Patton and O’Connor, 1988; Boote et al., 1990).

The Aruma (Suneinah) basin shows a rapid increase in thickness towards the NE with Coniacian-Santonian Muti Formation conglomerates passing up to deeper water shales and siltstones of the Campanian Fiqa Formation (Figure 4). The Late Campanian Juweiza Formation shows the first debris eroded from the Semail ophiolite (Glennie et al., 1974). Seismic sections across the foreland show the Hawasina thrust tips extending up into the Fiqa Formation (Warburton et al., 1990; Boote et al., 1990). The entire thrust process was completed by 70 Ma (Lower Maastrichtian) when the mountain belt just breached sea-level. Thin laterite soils overlie the ophiolite erosional surface in the southeast Oman Mountains.

Age constraints from the Semail ophiolite and the amphibolite-greenschist facies metamorphic sole are summarised in Figure 7. The age of crustal formation of the Semail ophiolite is known from the U-Pb zircon ages of plagiogranites (95.4–94.5 ± 0.5 Ma; Tilton et al., 1981, and 95.3 ± 0.2 Ma; Warren et al., 2005). The ages are consistent with Cenomanian radiolarian ages from cherts within the lower pillow lavas (Geotimes unit or V1) and Turonian ages from the middle volcanic member (Lasail unit or V2). The ophiolite complex has an inverted metamorphic sole along the base of the mantle sequence harzburgites and dunites comprising garnet and clinopyroxene amphibolites overlying greenschist facies metasediments (Searle and Malpas, 1980, 1982; Searle and Cox, 2002). These rocks are metamorphosed Triassic and Jurassic basalts and sediments subducted to 35 km depth (10–12 kbar) beneath the mantle at the same time as the ophiolite crust was forming (40Ar-39Ar ages of hornblendes of 94.9–92.6 ± 0.6 Ma; Hacker, 1994; Hacker et al., 1996).

The geochemistry of the ophiolite lavas (Pearce et al., 1981) and the synchroneity of the ophiolite ages with the metamorphic sole ages strongly suggest that the ophiolite formed above a NE-dipping subduction zone (Searle and Malpas, 1980; Lippard et al., 1986; Searle and Cox, 1999, 2002), and not at a mid-ocean ridge (Boudier et al., 1988; Nicolas et al., 1988). The lack of any trace of an active continental margin during the Cretaceous along the Oman Mountains also strongly suggest that this subduction zone dipped to the NE away from the continent (e.g. Lippard et al., 1986; Searle et al., 1994, 2003; Searle and Cox, 1999, 2002; Breton et al., 2004), not to the SW beneath the continent (Gregory et al., 1998; Gray et al., 2000; Gray and Gregory, 2004).

During the latest stages of ophiolite obduction, the northeastern leading edge of the Arabian Plate was subducted beneath an oceanic (ophiolitic) hanging-wall. The entire northeast margin of Saih Hatat has been affected by high-pressure metamorphism with increasing pressure and temperature (P-T) conditions with depth. The deepest structural levels are exposed around As Sifah (Figure 1) where Ordovician Amdeh quartzites, and overlying Permian limestones with basaltic flows (Saiq 1 Formation) have been metamorphosed to eclogite facies (Searle et al., 1994, 2004). Crinoid ossicles are still preserved within eclogite facies calc-schists at As Sifah. Gray et al. (2005) obtained a U-PB SHRIMP (Sensitive High-Resolution Ion MicroProbe) age of 298 ± 3 Ma from felsic schist within an eclogite boudin, suggesting that the protolith rocks were Early Permian - Late Carboniferous.

A U-Pb zircon age of 78.95 ± 0.13 Ma from the As Sifah eclogites, interpreted as dating peak high-pressure metamorphism (Warren et al., 2003, 2005), shows that the leading edge of the Arabian continental shelf was subducted to ca 80 km depth beneath the oceanic crust and mantle during the Campanian, some 15 My after formation of the ophiolite (Figure 8). 40Ar-39Ar ages of phengites from the upper slab rocks in northern Saih Hatat 80-70 Ma are consistent with Campanian High-Pressure metamorphism and cooling (Miller et al., 1999, 2002). The geological and geochronological constraints (Figure 8) are consistent with a single NE-dipping subduction zone, starting as an intra-oceanic subduction zone during the Cenomanian along which old, cold Triassic - Jurassic crust was subducted beneath the ophiolite (Searle and Cox, 1999, 2002; Searle et al., 1994, 2004). When the passive continental margin reached the subduction zone, the Triassic - Cretaceous upper slab decoupled and folded, forming the NE-directed backfolds and backthrusts of northern Saih Hatat. The Permian and older lower slab was subducted rapidly down and back up the same zone.

The Late Cretaceous period of thrusting and ophiolite obduction ended by early Maastrichtian time with a regional unconformity over all allochthonous units (Figure 9). Conglomerates and shales of the Qahlah and Al Khod formations are overlain by shallow water rudist-bearing limestones, indicative of a return to stable shelf conditions (Simsima Formation). The abundance of branching corals, rudists, bivalves and gastropods in the Simsima Formation across all of north Oman is testament to the fact that tectonism had ceased by 70 Ma. North Oman became emergent again at the end of the Maastrichtian, due largely to a global eustatic fall in sea-level. Minor deformation and uplift of the mountains is indicated by the outcrop pattern of the two unconformities below and above the Simsima Formation. During the Late Palaeocene stable shallow-water limestone deposition resumed and lasted for 25 My until the Oligocene (Nolan et al., 1990; Mann et al., 1990).

A second orogenic phase began during the Late Oligocene - Miocene with renewed shortening and folding of the Palaeogene limestones. The style of folding in the Tertiary is generally large-scale open folds or box folds with kink band sets. These folds may have steep, even overturned limbs but flat tops characteristic of box folds with prominent kink-band sets. Shortening is minimal but uplift of at least 2,000 meters is known from Jabal Abiad, east of Saih Hatat (Figure 10a,b). The magnitude and extent of Tertiary folding increase to the north along the Oman foreland (Figure 10c). The Musandam peninsula shows large-scale re-folding and re-thrusting of the entire Pre-Permian basement up to Palaeogene cover with minimum east to west translation of 14 km along the Hagab thrust (Searle et al., 1983; Searle, 1988a,b). This is interpreted as the beginning of the Zagros phase of continent–continent collision when the Musandam shelf first collided with the central Iran continental block (eastward extension of the Zagros suture zone).

The Al Jabal al Akhdar anticline axis continues NW along the Hawasina Window (Figure 1), where complex folding and thrusting in Sumeini slope facies, Hamrat Duru proximal basin facies, and Haybi complex distal basin facies rocks is exposed (Searle and Cooper, 1986). The cross-section (Figure 11) shows that the central part of the window is dominated by steep NE-verging back-folding and back-thrusting, affecting all allochthonous units. Early SW-directed thrusts are folded by later NE-directed backfolds and thrusts. Following SW-directed emplacement of the allochthon, the Semail thrust has been folded and inverted such that Haybi complex rocks have been re-thrust NE over the Semail ophiolite. Along strike to the NW, normal SW-vergence is resumed in the Haybi corridor (Searle, 1985). Along the northeastern half of the window recumbent nappes in the Hamrat Duru Group at Jabal Milh are NE-facing, whereas along the southern half of the window all folds are SW-facing, creating a divergent ‘pop-up’ structure (Searle and Cooper, 1986; Figures 10, 13). The deepest structural levels are three culminations of Sumeini Group slope facies limestones (Jabals Rais, Rastun and Mawq). These jabals are all affected by NE-verging late-stage backthrusts that cut up-section as far as the Semail thrust (Figure 11).

The Hawasina Window rocks form a large-scale duplex bounded by the Hawasina thrust along the top of the shelf carbonates and the Semail thrust beneath the ophiolite. Whereas around the Al Jabal al Akhdar shelf most of the Haybi and Hawasina rocks have been faulted out by the late-stage normal faulting, culmination of Al Jabal al Akhdar may have displaced these rocks laterally to the NW beneath the passive roof fault defined by the Semail thrust. A lateral (NW-SE) section shows how the Hamrat Duru nappes have been squeezed out laterally, sandwiched between the Sumeini culminations below and the Semail thrust above (Figure 12). Searle and Cooper (1986) explained the structural geometry of the Sumeini backthrusts by the presence of a steep frontal ramp at depth possibly along the shelf edge. The structural evolution of the Semail thrust around the northern Hawasina Window is complicated; it must have started (stage 1) as an early SW-vergent thrust carrying the ophiolite above Haybi and Hawasina thrust sheets, then (stage 2) became re-activated as an out-of-sequence late-stage thrust truncating stratigraphy in the footwall (Searle, 1985). During culmination of the Sumeini Group structures at depth and folding of the Hamrat Duru rocks in the Jabal Milh nappes (stage 3) it became a normal fault bringing footwall Hawasina Window rocks up from beneath the ophiolite. Finally during further NE-SW compression (stage 4) the deep-level Haybi and Hawasina rocks were backthrust NE over the ophiolite in the central part of the Hawasina Window.

The Al Jabal al Akhdar culmination (Figure 13) shows a single very large-scale 30–35 km half-wavelength anticline with a flat top (Saiq plateau). Dips increase outward from the axis such that the style is one of a gentle rounded box fold. Along the southern flank dips are gentle (15–20°S) but increase towards the NW-plunging nose, where a distinct overturning, verging SW is apparent. This may be a kink band that progressively developed into a steep thrust from Al Jabal al Akhdar towards the Hawasina Window at depth. Along the northern flank dips are 25–35°N. The upper contact of the shelf carbonates is presently a listric normal fault along the north and south flanks of Al Jabal al Akhdar, and the west and east flanks of Jabal Nakhl (Figure 6b). Ophiolite mantle sequence rocks rest directly on Wasia Formation top shelf with most of the Haybi and Hawasina thrust sheets faulted down. Throw on these late-stage normal faults could be as much as 4–5 km (Figure 6b).

Continuity of Hawasina, Haybi and Semail ophiolite thrust sheets over Al Jabal al Akhdar prior to folding is apparent from the presence of ophiolite all around its south, west and northern flanks (Bahla – Sint – Hawasina – Rustaq ophiolite blocks) and south and eastern flank (Nizwa – Semail ophiolite blocks). West of the Wadi Nakhr ‘Grand Canyon’ and Jabal Shams summit (Figure 5c), Haybi thrust slices comprising Triassic alkali volcanics and Oman exotics (Jabals Kawr, Misfah and un-named exotics along the crest of the fold) are more-or-less continuous over the southwestern limb of the Al Jabal al Akhdar fold.

The large scale of the Al Jabal al Akhdar fold resulted in extensive bedding plane slip during flexure. Along the south flank of Al Jabal al Akhdar, little or no internal folding or thrusting has been observed in many of the deeply incised wadi sections (e.g. Wadis Nakhr, Tanuf, Mu’aydeen). Some small-scale intra-formational thickening occurred by internal duplexing locally along the north flank of Al Jabal al Akhdar (Breton et al., 2004; figure 7). These are very small-scale folds and thrusts, restricted to the Cretaceous part of the shelf sequence and are not continuous up to the underlying shelf or overlying allochthonous units. These duplexes are not regional north-verging folds (Gray and Gregory, 2004), but local features related to thickening along the flanks of the main Al Jabal al Akhdar fold. Similar internal thrusts and folds in the shelf sequence of northern Saih Hatat are south-vergent (for example Yiti duplex; Figure 6c). The north-vergent backfolds of the Hawasina Window are much larger scale, late-stage structures affecting the overlying Semail ophiolite thrust sheet as well (see next section).

Bouger gravity data suggest a crustal thickness beneath Al Jabal al Akhdar of 47 km (Ravaut et al., 1997), or as much as 48–51 km (Al-Lazki et al., 2002). This thick crustal root is surprising, being only 75 km inland from the present-day coastline. Even beneath the Batinah coast the crustal thickness is as much as 39–42 km (Al-Lazki et al., 2002). Because of this, it is suggested that the Arabian continental crust must have extended a long way offshore buried beneath the thick Tertiary sedimentary sequence. This would increase the distance of obduction of the Semail ophiolite and underlying thrust sheets even more than proposed from surface geology.

Two possible models for the structure of Al Jabal al Akhdar have been proposed. The first model is a simple fold with no thrusting at depth (Glennie et al., 1973, 1974; Al-Lazki et al., 2002). The second model shows the anticline as a thrust culmination, either with a flat-lying detachment at depth ramping up in front of the Jabal Salakh frontal fold (Searle, 1985; Figure 11; Bernoulli and Weissart, 1987; Figure 2; Hanna, 1990; Figure 11; Cawood et al., 1990, Figure 11), or with steeper basement faults (Mount et al., 1998). Although the dips of the Al Jabal al Akhdar range are relatively gentle there is a SW-verging asymmetry to the fold towards the west. Geometrical constraints suggest that a deep flat within the pre-Permian basement is likely albeit with little horizontal translation. Although the Al Jabal al Akhdar culmination must have been a positive feature during the latest stages of Late Cretaceous deformation, the main period of growth was probably Oligocene - early Miocene. Fission track apatite data suggests that Oligocene uplift and erosion was followed by slower cooling in the Neogene (Poupeau et al., 1998; Mount et al., 1998).

The Saih Hatat culmination is the most complex of the three major shelf carbonate folds (Figure 14). Systematic mapping of northern Saih Hatat was carried out by Le Métour et al. (1986), Searle et al. (1994, 2004) and Miller et al. (2002). The structures shown here are based on the cross-section of Searle et al. (2004) for the north and from BRGM mapping around the Ibra ophiolite block to the south. The restored section (Figure 15) shows that the shelf-slope section is still ca 3 km thick in the north, indicating that the true Mesozoic shelf margin remains buried offshore or is missing. The Saih Hatat culmination is a 40 km half-wavelength fold that affects the entire basement up to Tertiary cover sequence. A prominent middle Permian unconformity crops out all the way around the Saih Hatat window, and separates a folded and mildly metamorphosed pre-Permian basement from the Permian - Cretaceous shelf carbonate sequence. Along the southern margin of Saih Hatat the shelf carbonates are conformable and unaffected by major internal folding or thrusting. Several thrust and normal faults do cut the section along Wadi Dayqah, but offsets are minor. However, along the northern margin of Saih Hatat, the basement rocks (Late Proterozoic Hatat schists, Hijam dolomite and Ordovician Amdeh quartzites) and the Permian - Jurassic shelf carbonates are severely affected by folding and thrusting.

The structurally highest units exposed show basement and Permian cover rocks folded around a 20 km long NNE-aligned sheath fold, exposed along Wadi Mayh (Figure 16). The compressive stress axis was primarily aligned NNE-SSW, but a second major axis was oriented WNW-ESE. Folds are NE-facing throughout. Above a major detachment termed the ‘Upper plate – Lower plate discontinuity’ (Miller et al., 1998, 2002) the Permian limestones are highly deformed by internal folding and imbrication. Below the discontinuity, metamorphism increases with depth through carpholite-bearing assemblages (ca 7–8 kbar, 280–350°C; Goffé et al., 1988) and blueschist facies (ca 12–15 kbar, 450–540°C) to deepest eclogite facies (As Sifah unit: 15–20 kbar, 540°C) rocks (Searle et al., 1994, 2004). Major normal-sense ductile shear zones separate these metamorphic zones. A 20 kbar (>80 km) depth profile is condensed to ca 7 km structural depth today, reflecting very large strains during rapid exhumation from the subduction zone at 78.95 ±0.13 Ma (U-Pb age of zircon from eclogite; Warren et al., 2005).

The southern flank of Saih Hatat along Wadi Tayyin is a large-scale, listric normal fault down-throwing the Semail ophiolite mantle sequence against the top shelf Wasia Formation. The lower part of the ophiolite thrust sheet together with the underlying Haybi and Hawasina complex thrust sheets has been down-faulted to the south. This normal fault is not a gravitational ‘collapse’ fault structure, but a normal fault operative during active compression. As the shelf carbonates were pushed up along the basal detachment the antiformal culmination was forced up through the overlying ophiolite. Little or no horizontal extension was involved in this process.

The minimum distance of emplacement of the Hawasina, Haybi and Semail ophiolite thrust sheets was 140 km from the tip line in the foreland to the present-day coast, but internal shortening within each duplex, and the unknown missing amount of continental margin offshore, must also be taken into account. Cooper (1988) estimated over 400 km of emplacement of the Hawasina complex rocks from palinspastic reconstructions. The protolith of the As Sifah unit eclogites were lateral equivalents of the Permian Saiq Formation limestones. Deepest-level quartz mica schists may be metamorphosed Ordovician Amdeh quartzites. The protoliths of the eclogites were basaltic flows and sills intruded into Permian Saiq 1 Formation limestones (Searle et al., 2004; Figure 14). There is no evidence for suture zone rocks above the As Sifah unit as required by the model of Gregory et al. (1998), Gray et al. (2000) and Gray and Gregory (2000). There is also no evidence of an active plate margin along the Cretaceous shelf in Oman, also required by their model of early nascent SW-directed subduction beneath the continental margin. Instead, passive margin sedimentation existed along the Oman margin throughout the Mesozoic up until ca 95 Ma when the ophiolite, Haybi and Hawasina thrust sheets were emplaced above the shelf into a rapidly deepening foreland basin.

Most workers agree that a single NE-dipping subduction zone existed offshore Oman from 95 Ma, and thrusting progressed generally in-sequence from NE to SW, from deep to shallow during the Late Cretaceous (Searle and Malpas, 1980; Lippard et al., 1986; Goffé et al., 1988; Searle and Cox, 1999, 2002; El-Shazli et al., 2001; Searle et al., 2003, 2004). There is no need for the double subduction zone postulated by Breton et al. (2004), as it can be proven that protoliths of the As Sifah unit were Permian and older rocks of the Arabian continental margin (Figure 15), and not an exotic plate.

Thrust emplacement of the Semail ophiolite, Haybi and Hawasina complexes onto the stable passive-margin foreland, exhumation of the deeply subducted continental margin and late-stage culmination of Saih Hatat was completed by the Upper Maastrichtian (Figures 8, 9). Upper Maastrichtian Qahlah and Simsima formations and Palaeocene - Eocene shallow water limestones unconformably overlie all allochthonous units around Saih Hatat. These shallow water limestones have been uplifted by about 600 meters along the southern margin of Saih Hatat and up to 2,000 meters on Jabal Abiad to the east of Saih Hatat (Figure 1). Jabal Abiad shows a giant box fold with cascade-type folds verging away from the fold axis along the flanks (Figure 10a). Post-Eocene shortening in the eastern Oman Mountains was very small. Along the northern margin of Saih Hatat in the Ras al Hamra area near Muscat, Tertiary folds show gentle folding along NNE-aligned axes, mimicking the Late Cretaceous sheath fold orientation. Post-Eocene normal faulting is also apparent along the Wadi Kabir fault at Ruwi where Late Cretaceous Ruwi mélange (with high pressure carpholite assemblages) and the Semail ophiolite have been down-faulted to the north against Triassic Mahil Formation limestones (Figure 14).

Two major orogenic events dominate the stress field in Oman during the Late Cretaceous, firstly the emplacement of the Semail ophiolite and underlying thrusts sheets from NE to SW onto the passive margin of Arabia, and secondly the emplacement of the Masirah ophiolite and underlying Batain mélange from ESE to WNW along the southeast Oman coast. The resultant bi-axial compressive stress field resulted in a dome and basin type fold pattern along the North Oman Mountains (Figure 17). The major anticline axis swings around from NW-SE in the Hawasina Window to ENE-WSW along Al Jabal al Akhdar, then swings around through 90° along the Jabal Nakhl NNE-SSW trend before swinging back to WNW-ESE along Saih Hatat. Four main NNE-SSW aligned anticline axes at right angles to the main trend are apparent from the map pattern: (1) Wadi Bani Umar–central Hawasina; (2) Rustaq to Bahla; (3) Saiq to Jabal Nakhl; and (4) Hubat-Baid-Ras al Hamra (Figure 17). Where two anticline axes meet, a structural dome is present (Hawasina Window, Al Jabal al Akhdar, Saih Hatat). The ophiolite blocks of Rustaq, Bahla and Semail are located in structural depressions where syncline axes intersect, in-between the shelf carbonate domes.

The bi-axial fold interference pattern is undoubtedly late stage because it affects all the allochthonous thrust sheets after their emplacement onto the shelf margin. Mid-Tertiary fold axes orientations are parallel to the Late Cretaceous fold axes indicating two phases of NE-SW compression. Folds in the Tertiary jabals along the southwestern flank of the mountains (e.g. Fahud, Natih, Ibri, Dhank structures; Figure 17) have a similar orientation to the Late Cretaceous structures along the mountains. West of Muscat in the Ras al Hamra area, fold axes in Tertiary rocks are aligned NNE-SSW parallel to the Late Cretaceous sheath folds in the Saih Hatat Window.

Semail Gap

The Semail Gap is the lineament of the Semail thrust, reactivated as a down-to-east normal fault during the uplift of Al Jabal al Akhdar-Jabal Nakhl shelf carbonates (Figure 1). The geometry of the main anticline along the northern end of Jabal Nakhl suggests that some sort of basal detachment must exist in the pre-Permian basement at depth. This detachment is likely to be similar to the one proposed at depth beneath Saih Hatat. Both deep detachments have more displacement in the north than the south, and may ramp up to higher levels along a dominant kink band set, as seen, for example, in the SW flank of Al Jabal al Akhdar and the Wadi Tayyin area of southern Saih Hatat. The Semail Gap fault would therefore be related to a deeper-level lateral ramp parallel to the main transport direction. Up to 3–4 km of throw occurs along this fault which juxtaposes ophiolite mantle sequence rocks above Wasia Formation shelf carbonates.

The Semail Gap fault is not a strike-slip fault because it does not laterally offset any geology, and does not extend south into the foreland. Instead it is a S-shaped transfer between the Al Jabal al Akhdar and Saih Hatat domes. The fault is most likely a listric normal fault curving around the Wasia Formation top shelf, decreasing in throw to the north and south away from the central sector. The lateral ramp would therefore have been active during the latest part of the Late Cretaceous tectonic event and enhanced by some unquantifiable amount of late Tertiary uplift. Prominent NNE-trending fold axes also occur along the Baid corridor, where the Semail ophiolite has been folded around a N-S-aligned fold axis, along the NNE-aligned Jabal Qirmadil and the Wadi Mayh sheath folds around the northern flank of the Saih Hatat culmination.

Foreland Fracture Systems

The fracture pattern in the Oman foreland south and southwest of the mountains is very different from the Oman Mountains (Figure 17a,b,c). The Late Cretaceous foreland fracture pattern, interpreted from 3-D seismic data as the result of NW-SE to WNW-ESE compressive stress (Filbrandt et al., 2006) is also very consistent across nearly 1,000 km from the Arabian Gulf to the Indian Ocean. South of the Hawasina basal thrust in the foreland there is very little evidence of S- or SW-directed thrusts or fractures associated with the Oman Mountains trend (Filbrandt et al., 2006). In the foreland, SW of the Oman Mountains, from Lekhwair to the Ghaba salt basin, two dominant fault sets are oriented WNW-ESE and NNW-SSE (Filbrandt et al., 2006). The highest fault density occurs in the competent Mesozoic shelf carbonate unit, whereas the clastic dominated Aruma Group absorbs strain by buckling and internal structural thickening. Several prominent faults, such as the Maradi and Burhan faults, developed during the Campanian and form en-echelon faults showing sinistral shear. Many of the north Oman Shu’aiba and Natih oil traps formed during the Campanian and are associated with large steep faults that were active during deposition of the Aruma foredeep sediments (Turonian to Campanian). The Fahud structure was formed by thickening of the Fiqa Formation towards the SW-dipping normal fault, and subsequent folding of the Fiqa Formation and overlying Palaeogene limestones (Filbrandt et al., 2006). Tertiary contraction caused local inversion of earlier faults such as the Fahud fault and reversal of slip along the Maradi fault, which shows a dextral sense of shear today. These faults may have provided vertical conduits for migration of hydrocarbons from deeper reservoirs (for example: Gharif Formation) up into shallower reservoirs (for example: Shu’aiba, Natih formations).

Filbrandt et al. (2006) suggested that the NW-SE to WNW-ESE maximum horizontal stress present along the Oman foreland was a result of an oblique collision of the Indian continent with the Arabian Plate during the Santonian-Campanian. Although there is certainly a fracture pattern that is consistent with this stress field, the fractures have very small or no offsets and several faults have been re-activated during an evolving stress regime. The deformation in the foreland and along the SE coast of Oman is not compatible with a continental collision such as presently seen around the northern margins of the Indian Plate (hundreds of km of shortening, regional metamorphism etc). Instead, it is suggested here that the fracture pattern in the Oman foreland is more likely to be related to WNW-directed emplacement of the Masirah ophiolite and Batain mélange.

Magnetic, seismic and structural data suggest that the Masirah ophiolite is part of an elongate ‘basement’ ridge of ophiolite that extends along the SE coastal margin of Oman (Mountain and Prell, 1990). This ophiolite belt, together with the Batain mélange was emplaced towards the WNW during the late Maastrichtian-early Palaeocene (Shackleton and Ries, 1990; Shackleton et al., 1990; Peters et al., 1997; Schreurs and Immenhauser, 1999). This timing slightly post-dates the final stages of emplacement of the North Oman ophiolite belt (ca 95–67 Ma), although the earlier stages of obduction of the Masirah ophiolite remain unknown, as the base is never seen. It is possible that the NNE-SSW trending fold axes in the North Oman Mountains were related to the Masirah ophiolite emplacement event. In this case, the biaxial fold interference pattern in Al Jabal al Akhdar-Saih Hatat could have been the result of refolding of the earlier NW-SE aligned folds (Al Jabal al Akhdar trend) by a slightly later set of NNE-SSW aligned folds (Jabal Nakhl-Semail Gap trend).

The structural evolution of the Oman Mountains can be related to the Late Cretaceous phase of thrusting of the Semail Ophiolite and underlying Haybi and Hawasina thrust sheets from NE to SW onto the passive continental margin of the Arabian Plate. This process closed an ocean at least 400-450 km wide (Cooper, 1988) and lasted ca 27 My from 95–68 Ma. Minimum convergence rates were of the order of 17 mm/year (Searle et al., 2004). The Semail ophiolite was formed above a NE-dipping subduction zone at 95 Ma, and thrusting propagated towards the SW with time from Semail to Haybi to Hawasina thrusts. Many of these thrusts were locally reactivated during the later phases of deformation as normal faults or backthrusts. There is no evidence of SW-dipping subduction beneath the Oman margin during the Middle Cretaceous as claimed by Gregory et al. (1998), Gray et al. (2000) and Gray and Gregory (2000). There are clearly no Andean-type granites or volcanic rocks typical of these active margins, and the Middle Cretaceous was a time of relatively stable passive margin carbonate sedimentation (Kharaib, Shu’aiba, Nahr Umr, Natih Formations) along north Oman. There is also no evidence of a separate plate origin for the ‘North Muscat microplate’ (Breton et al., 2004) or ‘Lower plate’ rocks of the Hulw and As Sifah units (Gregory et al., 1998; Gray et al., 2000). No suture zone rocks occur above the As Sifah eclogites or Hulw unit, and the protoliths rocks were attached to the Arabian margin, although show a slightly deeper stratigraphic level (Ordovician to Carboniferous-Permian protoliths).

The two giant shelf carbonate culminations of Al Jabal al Akhdar and Saih Hatat were both formed during the late stages of Late Cretaceous obduction, after thrusting of the Semail ophiolite, Haybi and Hawasina thrust sheets across the Arabian shelf margin. Thrusting along the northern margin of Al Jabal al Akhdar (Le Métour et al., 1990; Breton et al., 2004) is restricted to the Cretaceous (Natih and Nahr Umr formations) part of the shelf carbonate and is very minor intra-formational flexural slip-thickening associated with late-stage, large-scale folding of Al Jabal al Akhdar. Bedding is continuous below and above each duplex. In contrast, NE-vergent folds and NE-directed backthrusts across the Hawasina Window affect the Sumeini Group (Jabals Rais, Mawq, Rastun), the Hawasina and Haybi thrust sheets, as well as the Semail thrust (Searle and Cooper, 1986) and must be late-stage structures formed after emplacement of the entire allochthon onto the Arabian margin above a large-scale ramp along the basal detachment. The fact that these NE-facing folds affect all units rules out the model of early NE-directed thrusting associated with presumed craton/SW-directed subduction (Gregory et al., 1998; Gray et al., 2000; Gray and Gregory, 2003).

Saih Hatat is another 40 km half-wavelength anticline affecting all units from pre-Permian basement up to Semail ophiolite. Whereas the southern margin of the shelf carbonates is relatively undeformed, the northern margin shows extreme deformation: (1) in the form of major ductile shear zones, placing lower grade rocks onto higher grade rocks; (2) a series of large-scale asymmetric recumbent folds; (3) sheath folds with highly attenuated limbs; and (4) NE-facing folds affecting upper crustal rocks. These NE-directed folds and shear zones are late-stage backfolds (Searle et al., 1994, 2004) and not related to early SW-directed subduction as claimed by Gregory et al. (1998), Gray et al. (2000) and Gray and Gregory (2003).

The ‘North Muscat microplate’ comprising ‘lower plate’ Hulw and Sifah structures (Gregory et al., 1998; Gray et al., 2000; Breton et al., 2004) does not exist. The Hulw and Sifah structures are not exotic plates, but were connected to the Arabian continental margin prior to its attempted subduction (Searle et al., 1994, 2004). There is no evidence for a suture zone between the ‘lower slab’ Hulw-Sifah rocks and the overlying thrust sheets. The carbon isotope differences between the Sifah unit and the ‘upper slab’ rocks (Gray et al., 2005) are attributed simply to different stratigraphic levels, not to different structural ‘plates’. The As Sifah unit eclogite and blueschist facies rocks are metamorphosed Carboniferous-Permian limestones with basaltic flows and sills, that were part of the leading edge of the Arabian Plate (Searle et al., 2004). Quartz mica schists in the lowest structural level of the As Sifah unit are probably metamorphosed Ordovician Amdeh quartzites (Miller et al., 2002; Searle et al., 2004).

All stratigraphic, structural and U-Pb geochronological data suggest that a single NE-dipping subduction zone was operative during the Late Cretaceous (Searle and Malpas, 1980, 1982; Lippard et al., 1986; Searle and Cox, 1999, 2002; Searle et al., 1994, 2003, 2004; Warren et al., 2003, 2005). Protoliths of the As Sifah eclogites are Permian volcanic rocks intruded into the leading edge of the Arabian Plate, subducted to 20 kbar (ca 78–80 km depth) at 78.95 ± 0.13 Ma (U-Pb zircon age; Warren et al., 2005) and exhumed back up the same subduction zone (Searle et al., 1994, 2004).

Both the Al Jabal al Akhdar and Saih Hatat culminations may be underlain by deep blind-thrust flats, which may ramp to the surface along the foreland fold-thrust belt (Jabal Salakh-Madmar-Adam folds). Both culminations have large-scale listric normal faults bounding their flanks. These normal faults were active during uplift of the footwall shelf carbonates through their allochthonous cover (Hawasina-Haybi-Semail) and were operating in a wholly compressional stress field. The faults were likely active during the latest phase of Late Cretaceous deformation associated with ophiolite emplacement. It is also possible that the flank-bounding normal faults could have accommodated some amount of mid-Tertiary uplift of the Al Jabal al Akhdar and Saih Hatat culminations.

A second phase of folding occurred after the Late Eocene - Early Oligocene as evidenced by the gentle box fold of Jabal Abiad, east of Saih Hatat (Figure 10a,b), and the box folded Tertiary jabals along the SW flank of the Oman Mountains (Jabals Hafit, Awaynah, Dhank (Figure 10c), Ibri, Fahud, Natih, etc). Eocene limestones are now at 2,000 m altitude on the Jabal Abiad structure and up to 4,000 m of post-Eocene uplift may have occurred in northern Saih Hatat. Apatite fission track data is consistent with a compressional tectonic event beginning in the Burdigalian (Early Miocene) and culminating in the Late Miocene - Pliocene (Poupeau et al., 1998).

The three main shelf carbonate culminations in the Oman Mountains (Musandam, Al Jabal al Akhdar, Saih Hatat) all show different structural geometries, degree of thrusting and internal folding, and timing of culmination and uplift. The Musandam culmination resulted from deep level Miocene thrusting affecting the pre-Permian basement up to the Eocene - Early Oligocene neo-autochthonous cover. At least 15 km of westward thrusting of the complete sequence occurred along the Hagab Thrust, during the initial affects of collision between Arabia and the Central Iran Plate (Searle et al., 1983; Searle, 1988a,b).

Much of this work was funded by NERC grant NER/K/S/2000/00951. I thank John Smewing, Dave Cooper, Jon Cox, Clare Warren, Tom Jordan, Mohammed Ali, Tony Watts, Randall Parrish and Hilal Al-Azri for discussions, and Dave Sansom for excellent cartography. I am grateful to Dr. Hilal Al-Azri and Dr. Salim Al Bu-Saidi from the Omani Directorate General of Minerals (Ministry of Commerce and Industry) in Muscat for their support and to Judy and David Willis for their hospitality in Bawsher. Detailed reviews from Pete Jeans and two anonymous reviewers greatly improved the manuscript. The design and drafting of the final graphics by Gulf PetroLink is appreciated.

Michael P. Searle is a Lecturer in the Department of Earth Sciences, Oxford University, United Kingdom, and is a Senior Research Fellow at Worcester College, Oxford. He obtained his PhD in 1980 working on structures and metamorphism beneath the Semail Ophiolite in northern Oman. Since then he has worked along the length of the Oman Mountains from the Musandam to Masirah Island and the Batain coast. Michael has also worked over 25 years along the Himalayan and Karakoram Ranges in Pakistan, India, Nepal, Bhutan and Sikkim as well as in Tibet, Burma, Thailand and Vietnam. His research integrates field mapping, structural geology, metamorphic and igneous geology, and geochronology with the aim of unraveling the large-scale evolution of mountain belts.