The Middle Eastern Infracambrian Debate offers specific choices between profoundly different tectono-stratigraphic models that have important scientific and petroleum exploration implications worldwide. A crucial first step in the Debate is choosing between the interpretations of zircon geochronology (Cryogenian Model) or regional chrono-stratigraphy based on much younger age-dating by alternative radiometric techniques (e.g. K-Ar, Rb-Sr, Ar/Ar; Ediacaran Model). The interpretation of zircon geochronology implies Oman’s oldest diamictites of the Abu Mahara Group represent the Sturtian (ca. 720–700 Ma) and Miranoan (ca. 663–636 Ma) glaciations of the Cryogenian Period (850–630 Ma) separated by the ca. 50 My Fiq-Ghubrah Hiatus. The Cryogenian Model implies three phases of rifting in the same regions between ca. 723–530 Ma, and another younger but disputed ca. 30–40 My Shuram-Khufai Hiatus occurring in a tectonically quiescent platform setting (post-glacial Nafun Group’s fine clastics and carbonates).
This combined essay and book review of Global Neoproterozoic Petroleum Systems disputes the interpretation of zircon geochronology to establish absolute time for Oman’s oldest rocks. It argues for the single-rift-without-hiatus Ediacaran Model based on ages of basement and volcanic rocks using alternative radiometric techniques in Jordan, Oman and Saudi Arabia. Oman’s Hadash Formation and coeval Mirbat Cap Carbonate are believed to provide an important correlative marker that recorded the start of the great sea-level rise of the Nafun Transgression at ca. 572 Ma, not 636 Ma. The Transgression was due to the melt-out of the late Ediacaran Varanginian Glaciation (represented in Oman by the diamictites of the Ayn, combined Fiq-Ghubrah and subsurface Ghadir Manqil formations, all of the Abu Mahara Group, deposited between ca. 585–572 Ma), not the Sturtian and Miranoan glaciations. The deep-marine organic-rich shales and siliciclastics of Oman’s Masirah Bay Formation (coeval Arkahawl Formation of Mirbat Group) are syn-rift and reflect the Nafun Transgression spilling beyond the rift basins and their surrounding lowlands. As proposed in the Middle East Geologic Time Scale 2010 and GeoArabia’s Infracambrian Debate, the time interval ca. 585–530 Ma can best be cast in terms of transgressive-regressive chrono-sequences in a regional tectono-stratigraphic extensional framework.
Until the turn of the 20th Century it was widely accepted that Oman’s oldest syn-rift glaciogenic rocks of the Abu Mahara Group were deposited after ca. 600 Ma in the Ediacaran Period (ca. 630–542 Ma; based on radiometric ages inGorin et al., 1982; Hughes Clark, 1988; Debreuilh et al., 1992; Ediacaran Model in Figure 1after e.g. Al-Husseini, 1989, 2000; Loosveld et al., 1996; Droste, 1997; see review in Sharland et al., 2001). Then in 2000 Brasier et al. dated the oldest Abu Mahara glaciogenic rocks using zircon geochronology at ca. 723–712 Ma in the Cryogenian Period (850–630 Ma), and by 2009 the Ediacaran radiometric age data were discarded by many as unreliable (main authors in alphabetical order: P.A. Allen, S.A. Bowring, M.D. Brasier, A. Cozzi, J. Grotzinger, J. Leather, E. Le Guerroué, R. Rieu and others; see references and Cryogenian Model in Figures 1 to 3). The revised Cryogenian age implies Oman’s tectono-stratigraphic evolution is highly complex and its regional ramifications invite broader participation in GeoArabia’s Infracambrian Debate (Al-Husseini, 2010a, b).
Shortly after announcing the Debate in early 2010, the Geological Society (London) coincidently invited GeoArabia to review their Special Publication 326 Global Neoproterozoic Petroleum Systems (Craig et al., 2009a; SP 326). This must-read book substantially catalyzes the Debate’s importance, not only for the Middle East, but worldwide. As highlighted in SP 326 Neoproterozoic petroleum systems are commercially proven in many regions of the world and frontier basins await exploration, particularly in North Africa (e.g. Craig et al., 2009b; Ghori et al., 2009; Lottaroli et al., 2009; Lüning et al. 2009).
Many SP 326 papers present untested exploration plays by analogy to the producing Infracambrian petroleum system of Petroleum Development Oman (for review see Al-Siyabi, 2005). Some of its papers inconclusively consider aspects of the Cryogenian versus Ediacaran models for Oman and elsewhere. This essay-book review highlights issues that arise from Oman’s Cryogenian Model, some of which may impact the interpretations and conclusions of SP 326. It argues in favor of the Ediacaran Model by correlating Oman’s Abu Mahara Group to the Ediacaran Varanginian Glaciation between ca. 585–572 Ma (Figure 1).
GLOBAL AND OMAN’S NEOPROTEROZOIC GLACIATIONS
The Infracambrian Period, although not precisely defined, is considered a useful term for North Africa and the Middle East by Smith (2009, SP 326). It is informally used here for the Middle East’s late Ediacaran and Early Cambrian interval between ca. 585–530 Ma (Figure 1). Smith reviews the evidence, and numerous conflicting age dating and correlation problems for the three main Cryogenian and Ediacaran (EC) glaciations: (1) Sturtian (741–734 or 720–700 Ma), (2) Miranoan (663–636 Ma) and Varanginian (sometimes Gaskiers, 584–582 Ma) (Figure 1). In near-polar Neoproterozoic South America, Hlebszevitsch et al. (2009, SP 326) recognize the three main EC glaciations and a few more (with some differences in ages): Caigas (also spelled Kaigas, 771–741 Ma), Sturtian (717–684), Miraonoan (635 Ma), Varanginian (Gaskiers, 584–582 Ma), Moelv (560 Ma) and Vingerbrook (548 Ma).
Smith concludes that much more accurate age data is required in order to correlate far-apart glaciogenic rocks. He argues that the evidence for a Snowball Earth with the oceans completely frozen over is unlikely. The interpretation of zircon geochronology for Oman’s oldest glaciogenic rocks falls in the younger 720–700 or 717–684 Ma age windows of the Sturtian Glaciation (Figures 1 to 3).
Mirbat Diamicitites: Which Glaciations?
GeoArabia’s Infracambrian Debate can be somewhat simplified by starting with the two undated glaciogenic formations (Ayn and Shareef of the Mirbat Group) at outcrop in southern Oman (Figures 2 and 3). These were tentatively correlated to the Miranoan and Sturtian glaciations (Rieu and Allen, 2008; see references therein). The proposed correlations, however, did not consider either of the two Paleozoic glaciations in Arabia represented by: (1) Oman’s Late Carboniferous – Early Permian Al Khlata Formation and equivalents (e.g. Osterloff et al., 2004), and (2) Saudi Arabia’s Late Ordovician Sarah Formation and equivalents (e.g. Vaslet, 1990).
The older Ayn Formation cannot be either of these two. It cannot be Ordovician because: (1) Paleozoic glaciogenic rocks are not capped by carbonates as evident by the Mirbat Cap Carbonate (Figures 1 to 3), and (2) the Sarah Formation is represented in eastern Arabia by fluvial and marine clastics but not diamictites. The Ayn Formation cannot be correlated to the Carboniferous – Early Permian Al Khlata (no cap carbonate) because the overlying stratigraphy completely excludes such an option.
To better constrain the likely stratigraphic correlation consider the localities shown in Figure 2: (1) cross-section that skirts the Mirbat area (modified afterDroste, 1997, Figure 4a), (2) distribution of the Al Khlata in South Oman (Osterloff et al., 2004, Figure 5), and (3) Mirbat outcrop map (Figure 6a, Rieu and Allen, 2008). Figures 4b and 6b sketch the rock units by geography and chrono-stratigraphic position. The Shareef and Al Khlata formations are both eroded by sub-Cretaceous and Tertiary unconformities. The Masirah Bay, Marsham and Arkahawl formations are all eroded by the Middle Carboniferous Unconformity (Sub-Khlata Unconformity). The Sub-Khlata is a major Upper Paleozoic unconformity (sometimes referred to as “Hercynian”) associated with the great uplift of southern Oman (Al-Husseini, 2004; Faqira et al., 2009). By regional distribution, stratigraphic position and correlation of the two unconformities it seems apparent that the four glaciogenic formations correlate in a simple manner: Al Khlata to Shareef, and Abu Mahara Group (Ghadir Manqil Formation) to Ayn. Now the cap carbonates of the Mirbat Group and Hadash Formation correlate by stratigraphic position.
The suggested correlation would eliminate the Late Carboniferous – Early Permian Shareef Formation from the Infracambrian Debate. It also potentially clarifies another important aspect of the Debate by leaving only one Neoproterozoic glaciogenic formation (Ayn) above the Mirbat Proterozoic basement. A single Neoproterozoic diamictite implies the Ayn Formation cannot be Sturtian (Figure 3) because it passes continuously via the Mirbat Cap Carbonate (= Hadash Formation) to the Ediacaran Arkahawl and Marsham formations (together = Masirah Bay Formation). So at the Mirbat outcrop the suggested correlation only implies either a Miranoan Glaciation (Cryogenian Model) or Varanginian one (Ediacaran Model).
Abu Mahara Group: Two or One Glaciations?
The Cryogenian Model is primarily based on zircon geochronology in the outcrop section in Al Jabal al-Akhdar (Figures 1 to 3) where the Abu Mahara Group is divided into the older Ghubrah and younger Fiq formations both containing syn-rift diamictites. In places the formations are in contact and elsewhere separated by the undated Saqla Formation. To better appreciate the correlations of Abu Mahara diamictites to global glaciations, reviewing the ages and contact relationships between the Ghubrah, Saqla and Fiq formations in Al Jabal al-Akhdar is crucial (Figure 2).
Zircon Ages of Ghubrah Formation: In Wadi Mistal (Al Jabal al-Akhdar, Figure 2) a tuffaceous sandstone inter-bedded with rain-out diamictites of the Ghubrah Formation yielded a U-Pb-zircon age of 723 +16/-10 Ma (Brasier et al., 2000) and 712 ± 1.6 Ma (Leather, 2001). Similar U-Pb-zircon ages of ca. 714–712 Ma have been reported by others (see reviews inAllen, 2007, and Bowring et al., 2007). Further analysis on zircon grains from the same tuffaceous sandstone bed by Bowring et al. (2007) arrived at an average age of ca. 713 Ma. The latter authors stress, however, that even though this unit is a distinctive, uniform layer within the diamictite, it is a clastic rock and could contain a still younger population of zircons. These age estimates led to the correlation of the Ghubrah to the Sturtian Glaciation in the younger 720–700 Ma interval (Figure 1).
Zircon Ages of the Fiq Formation:Bowring et al. (2007) reported: (1) syn-glacial turbiditic sandstones yielded a suite of detrital zircon ages ranging from 920 to 664 Ma, and (2) U-Pb dating of zircons from Lahan-1 borehole (Figure 2) from possible volcanic ash in a succession of shale and siltstone containing dropstones (Fiq Formation, 9 m below a dolostone bed, Hadash Formation) yielded ages between 718–645 Ma. Taking the stratigraphic position of the dolostone and dating, they suggested that the Fiq glacial deposits were widespread in Oman, with an age of 645 Ma or younger and possibly consistent with a Miranoan Glaciation (ca. 663–636 Ma, Figure 1).
Saqla Formation: This formation separates the Ghubrah and Fiq formations, but is not present everwhere in Al Jabal al-Akhdar outcrops. Rabu et al. (1986) described the Saqla Formation (30 m thick in the Ghubrah Bowl in the Nakhl quadrangle) as discontinuous flows of mafic pillow lava, hyaloclastic rocks and a doleritic sill. In Wadi Mu’aydin it is a massive unit (60 m thick) comprising partly brecciated pillow lava and hyaloclastite. The lavas display trachytic texture with an andesitic-dacitic composition. Le Guerroué et al. (2005) added that the magmatic activity was principally effusive and occasionally hypovolcanic, with basalt pillows suggesting subaqueous eruptions.
Saqla/Fiq Boundary: The Fiq Formation conformably overlies the Saqla Formation (Rabu et al., 1986; Le Guerroué et al., 2005) and in turn is overlain by the cap carbonate (Hadash Formation of the Nafun Group). Therefore the transition from the Saqla volcanics and Fiq diamictites (both syn-rift) to the post-glacial Hadash cap carbonate shows no evidence of significant stratigraphic gaps (Figure 1).
Ghubrah/Saqla Boundary: Where the Saqla Formation separates the two glaciogenic formations, Brasier et al. (2000) and Leather (2001) considered the Ghubrah/Saqla boundary as possibly an unconformity. Le Guerroué et al. (2005), based on several studied sections, concluded that the Ghubrah/Saqla boundary is an angular unconformity, with bedding measurements showing a discordance of up to 70o.
Ghubrah/Fiq Boundary: Critically for the Debate, where the Saqla is absent, the Ghubrah/Fiq boundary shows no evidence for a stratigraphic break. In the Nakhl quadrangle at Wadi Mu’aydin, Rabu et al. (1986),observed that the Ghubrah and Fiq formations are indistinguishable except that bedding is better developed in the Fiq; locally the Fiq sandstones inter-finger with conglomerate similar to that of the Ghubrah Formation. In the adjoining Rustaq quadrangle, Beurrier et al. (1986)reported that the Ghubrah/Fiq contact is difficult to identify because the passage between the two formations is very gradual and locally indiscernible.
The field reports summarized above indicate: (1) Ghubrah and Fiq (Abu Mahara Group) are nearly indistinguishable in localities where the Saqla is absent and therefore represent only one main glaciation, and (2) in some localities these formations were structurally disturbed by major volcanic activity represented by the Saqla Formation. Yet the Cryogenian Model invokes the Fiq-Ghubrah Hiatus representing ca. 50 million year (My) of missing section (Figures 1 and 3). The zircon geochronology (younger than 645 Ma) for just one glaciation constrains the Abu Mahara (Fiq-Ghubrah combined) to either the Miranoan or Varanginian glaciations. The correlation of the Ghubrah to either of these two younger glaciations is also supported by the age of the basement below the Abu Mahara Group in subsurface: (1) in Zafer-1 the bottom-hole alkaline granite was dated by Rb-Sr at 662 ± 13 Ma (Hughes Clark, 1988), and (2) by zircon geochronolgy in Makarem-3 the intrusive syenite at 699 Ma (Bowring et al., 2007).
In summary, in the Ediacaran Model does not adopt two glaciations separated by the Fiq-Ghubrah Hiatus. The Fiq, Saqla and Ghubrah formations are considered as one syn-rift unit (Abu Mahara Group) corresponding to one glaciation. They are correlated to the Mirbat Group: Fiq-Ghubrah to Ayn, with the Hadash to Mirbat carbonates capping the glaciogenic rocks (see caption of Figure 3, and Figures 4b and 6b). But like the Mirbat outcrop the suggested correlations only imply either a Miranoan Glaciation (Cryogenian Model) or Varanginian one (Ediacaran Model).
Do Cap Carbonates Discriminate between Neoproterozoic Glaciations?
Smith (2009 SP 326, see references therein) reports that cap carbonates (ca. 5–27 m thick) can occur below or above diamictites, and: “where several discrete glacial intervals occur in a single succession each is commonly overlain by a cap carbonate that may constitute the only carbonate within an otherwise siliciclastic succession. Thus, from a global viewpoint, Neoproterozoic cap carbonates showing a particular style of development cannot be used as a chrono-stratigraphic marker, even in cases where carbon isotope data are also available.” Indeed, J. Mattner (2009, personal communication) reported that the Ayn Formation in Mirbat contains a carbonate bed at the top and in some areas, also at the base of the thick diamictite section.
Smith also stated: “Glaskiers [Varanginian] glacial deposits are not followed everywhere by an isotopically uniform cap carbonate. Thin Glaskiers cap carbonates are known from occurrences in Newfoundland and North China.” His figure 13 shows a Varanginian (Gaskiers) cap carbonate occurs above diamictites with an age of 575 ± 3 Ma in Tasmania suggesting this Glaciation lasted longer than just 2 My (584–582 Ma), more like 584–572 Ma as calibrated in the Arabian Orbital Stratigraphy (AROS time scale; see Chart in Al-Husseini, 2010b; Matthews and Al-Husseini, 2010, in press). The caption for Smith’s figure 13 reports that Miranoan-style cap carbonates also occur in Sturtian and Varanginian (Glaskiers) diamictites.
These observations do not support correlating the Hadash Formation and here-proposed correlative Mirbat Cap Carbonate (and basal carbonate bed, J. Mattner) to a particular Neoproterozic glaciation. Moreover, if Oman’s Abu Mahara diamictites (Ayn, combined Fiq-Ghubrah, subsurface Ghadir Manqil formations) are all indeed Varanginian then Smith’s conclusions would require major revisions derived from Oman (e.g. his figures 4–6, 8 and 13).
The correlation of the Abu Mahara Group to the Cryogenian Period (Cryogenian Model) transmits problems up-section. It not only invokes the problematic Fiq-Ghubrah Hiatus but also introduces the disputed Shuram-Khufai Hiatus (Shuram SB). Two of the most widely accepted ages for the Huqf Supergroup constrain the top of the Nafun Group (Ara SB) at ca. 547 Ma and the mid-Buah Formation at ca. 550 Ma (Figure 1; Bowring et al., 2007). These ages constrain the Nafun Group between ca. 547 and 636 Ma (top of the undated Fiq Formation if assumed to be Miranoan) implying the Nafun Group (ca. 2,000 m maximum thickness) was deposited in ca. 89 My at an average rate of ca. 23 m/My (uncompensated for compaction).
Bowring et al. (2007), however, based on ages across the lower part of the Ara Group to mid-Buah Formation estimated the uncompensated subsidence rate for the upper part of the Nafun Group (Buah and Shuram formations) to be of the order of 100 m/My. To keep the Fiq Formation correlated to the Miranoan Glaciation they interpreted a ca. 30–40 My Shuram-Khufai Hiatus between these formations (Figure 1).
Le Guerroué et al. (2009), however, disputed the inference for such a hiatus by pointing out that the Shuram SB shows no evidence of exposure or time gap. Forbes et al. (2010) reported this unresolved discrepancy and followed Bowring et al. (2007) by placing the base Shuram at ca. 558 Ma and the top Khufai at ca. 590 Ma – implying a 32 My Shuram-Khufai Hiatus. The Ediacaran Model bypasses this hiatus problem of the Cryogenian Model but retains the ca. 100 m/My rock-time conversion as an important rate (as explained below).
Multi-Rifting versus Single-Rifting Models
Depositing the Abu Mahara and Nafun groups between 723–547 Ma not only requires filling time with one and/or two hiatuses but also invokes multi-rifting.Craig et al. (2009b, p. 12, and their communication with P.A. Allen) raised a concern over the very long time duration attributed to the deposition of the Nafun and Ara groups (ca. 96 My between 636–540 Ma, or ca. 106 My to account for the syn-rift Nimr Group). They explain away this unlikely phenomenon as possibly due to low strain rates stretching an over-thickened crust. But the multi-rifting Cryogenian Model extends even further back in time starting at ca. 723 Ma (not just 636 Ma) and implies (Figure 1):
(1) Rift-1 phase (ca. 723–712? Ma) with deposition of the syn-rift Sturtian Ghubrah diamictites,
(2) A ca. 50-My Fiq-Ghubrah Hiatus of unknown geological significance (ca. 712?–663 Ma),
(3) Rift-2 phase started with Saqla volcanics followed by deposition of syn-rift Miranaoan Fiq diamictites (ca. 663–636 Ma),
(4) An 89-My phase (636–547 Ma) of tectonic quiescence and deposition of the Nafun Group, with or without the problematic ca. 32-My Shuram-Khufai Hiatus,
(5) Rift-3 phase (ca. 547–530 Ma) involving the deposition of the syn-rift Ara Group cycles (including massive halite) and syn-rift clastics of the Nimr Group,
(6) End of rifting at ca. 530 Ma (Angudan Unconformity).
The likelihood of three syn-rift rock units, separated by hiatuses and/or inactive tectonic periods each representing many 10s of million years, being deposited/preserved in or around congruent rift basins over a period of some ca. 193 My (ca. 723–530 Ma) is possible. But it is unnecessarily complicated when the alternative simpler Ediacaran Model fits all the data but not the interpretation of zircon geochronology.
REGIONAL CHRONO-STRATIGRAPHIC CONSIDERATIONS
Jibalah, Abu Mahara and Nafun Groups
A Miranoan correlation, let alone a Sturtian one, for one and/or two Abu Mahara glaciogenic units seems highly unlikely when regional correlations and their implied age constraints are considered. In Saudi Arabia’s Proterozoic Shield the Infracambrian Jibalah Group crops out in numerous pull-apart basins along the Najd Fault System (Delfour 1970; Hadley, 1974, 1986; see references inAl-Husseini, 1989, 2000, 2010b). The rapidly subsiding Jibalah basins sampled the prevailing regional sedimentary (and igneous) lithology in ascending stratigraphic order: Rubtayn Volcanic Conglomerate Member (ca. 700 m), Rubtayn Polymictic Conglomerate Member (up to 1,500 m thick), Rubtayn Sandstone Member (ca. 1,000 m) and massive carbonates and fine clastics (300–500 m thick) of the Muraykhah Formation (Delfour 1970; Hadley, 1974, 1986). In some basins andesite-basalt layers occur within, below or above the Rubtayn Formation (e.g. Badayi Formation between Rubtayn and Muraykhah formations, ca. 100–150 m thick, Hadley, 1974; dated 558 ± 6.6 Ma by Ar/Ar inBrown et al., 1989). The Jibalah Group is definitely younger-than 610 Ma (e.g. Nehlig et al., 2002, Jibalah basins in their figure 2). The coeval Saramuj Conglomerate of Jordan is younger than ca. 585 Ma (Ibrahim and McCourt, 1995). These ages narrow the coeval Jordanian and Saudi Arabian conglomerates to younger than ca. 585 Ma in late Ediacaran.
Smith (2009 SP 326, see references therein) supports the view that all the major syn-rift Neoproterozoic diamictites can be considered pene-contemporaneous with nearby glacial activity even if they are strictly of non-glacial origin (e.g. evidence for ice contact). Following his suggestion (and in discussions regarding the Saramuj Conglomerate with J. Mattner and J. Powell in 2009), the Rubtayn Polymictic Conglomerate and Saramuj Conglomerate are interpreted as melt-out deposits from nearby glaciers implying Varanginian glaciers covered parts of Arabia between ca. 585–572 Ma. This interpretation offers NW Arabia’s syn-Varanginian conglomerates as correlatives to Abu Mahara diamictites. The correlation is more evident when the overlying formations are correlated by stratigraphic position, lithology and thickness: (1) Rubtayn Sandstone Member to Oman’s Masirah Bay Formation, and (2) Muraykhah Formation to Oman’s Khufai, Shuram and Buah formations of the Nafun Group (see Chart inAl-Husseini, 2010b). Thick layers of andesite-basalt occurring in the Rubtayn Formation is not unlike Saqla andesite-basalt separating the Ghubrah and Fiq diamicites. These proposed correlations imply the Rubtayn, Badayi and Muraykhah formations of the Jibalah Group are equivalent to the Abu Mahara and Nafun groups – but all in late Ediacaran.
Varanganian Melt-Out and Regional Nafun Transgression
The problematic Shuram-Khufai Hiatus, disputed between proponents of the Cryogenian Model, is in part due to the interpretation that the Nafun Group was deposited over a very long time period (636–547 Ma). It is also in part based on the observation that the Nafun Group is much more widespread than the older syn-rift Abu Mahara (combined Fiq-Ghubrah) and younger syn-rift Ara and Nimr groups. Indeed the Nafun’s widespread distribution is evident by the coeval deposition of the massive carbonates of the Muraykhah Formation over the Arabian Shield and Infracambrian Complex in Iran (Soltanieh Dolomite and Bayandor formations above Basement; see Chart inAl-Husseini, 2010b). But these formations were not deposited across the entire Middle East in a stable platform setting but rather were restricted to pull-apart and rift basins and nearby lowlands.
The distribution of the Nafun Group and equivalents beyond the rift basins more likely testifies to the great rise in relative sea level caused by the melt-out of the Varanginian glaciers (see Ediacaran Model in Figure 1). The Nafun Transgression would have found substantial accommodation space (100s+ meter post-glacial sea-level rise enhanced by some 100 m/My extensional subsidence). Its semi-regional flooding over Oman should have extended beyond the rift shoulders of the Abu Mahara rift basins.
The Ediacaran Model implies the Nafun’s deposition only lasted some 20 My. Converting to time its maximum known thickness (ca. 2,000 m) together with that of the Abu Mahara Group (715 m in Ghadir Manqil-1 borehole) at 100 m/My (as estimated for the Shuram-Buah interval by Bowring et al., 2007): the age for base Abu Mahara would be 547 + 27 = 574 Ma. A lower rate of 70 m/My would put the base Abu Mahara at 585 Ma and the Hadash cap carbonate at 575 Ma (comparable to Varanganian cap carbonate in Tasmania dated at 575 ± 3 Ma, Calver et al., 2004). In the Ediacaran Model the geoconceptual complexity involving multi-rifting, two glaciations and one or two hiatuses would be eliminated by a single rifting event between 585–530 Ma.
Sequence Stratigraphy: The Way Foreward?
Gámez Vintaned et al. (2009, SP 326) show how sequence stratigraphy can be used to interpret Ediacaran – Cambrian transgressive-regressive (T-R) sequences in far-away Spain. This is a very promising approach because T-R sequences have also been interpreted in Oman’s Ediacaran and Lower Cambrian successions (Figure 1): Nafun 1–5 sequences (Cozzi and Al-Siyabi, 2004) and Ara Cycles 1–6 (Amthor et al., 2005). Ediacaran – Early Cambrian T-R sequences could provide the primary chrono-stratigraphic framework upon which less-conclusive markers can be calibrated (biostratigraphy, cap carbonates, glaciogenic rocks, chemostratigraphy, etc.).
A good datum to start with is the age of the Vingerbrook Glaciation (548 Ma), which coincides with the well-constrained age of Oman’s Buah/Ara Boundary (Figure 1, Ara SB at 547 Ma, Bowring et al., 2007). This correlation implies the polar Vingerbrook Glaciation caused the sea-level drop of the Ara SB. Moreover the regularity of the Spanish and Omani (Nafun and Ara) sequences argues that Infracambrian glacio-eustasy was driven by long-period orbital forcing (ca. 4.9 My; see Arabian Orbital Stratigraphy AROS 2010 inAl-Husseini, 2010b; Matthews and Al-Husseini, 2010, in press). In the AROS 2010 time scale, the Moelv Glaciation (560 Ma) correlates to the boundary between Nafun Sequences 2 and 3 (Nafun SB 3) orbitally calibrated at ca. 562 Ma. Detailed studies of the oldest glaciogenic and coeval conglomeratic formations in Arabia may reveal the number of long-period cycles of the Varanganian Glaciation, here estimated between ca. 585–572 Ma.
ZIRCON GEOCHRONOLOGY: DETRITAL, PARENTAL OR ABSOLUTE?
There is no doubt zircon geochronology provides an important dataset that constrains chrono-stratigraphic frameworks. But to avoid similar pitfalls that occur in biostratigraphy, the non-expert must be unequivocally assured that the zircon ages pertain to the rock unit in question and not its parent rock. Detrital zircon ages only indicate the oldest age constraint (Fiq Formation is younger than 645 Ma). They are suspect when their ages coincide with, or are older than, the age of the basement (e.g. Ghubrah Formation). Reassurance of their reliability is particularly important when other dating techniques (e.g. K-Ar, Rb-Sr, Ar/Ar, etc.) give ages that are 100 or more million years younger. Reassurances become paramount when the revised ages contradict regional chrono-stratigraphic correlations, including the candidate Halfayn-Shammar Correlation.
Correlation of Halfayn and Shammar Rhyolites?
The Halfayn Formation is 35 m thick in Al Huqf and consists of volcaniclastic and carbonate clasts, pyroclastic flows and tuff that overlie granodioritic basement dated 825–822 Ma by Leather (2001; see review inAllen, 2007). Debreuilh et al. (1992) obtained an Rb-Sr 554 ± 10 Ma age for granodiorite dikes from the basement below the Halfayn Formation, and a whole-rock Rb-Sr 562 ± 42 Ma age from a Halfayn rhyolite sample. Allen (2007) correlated the Halfayn Formation to the bottom-hole volcanic rock (trachyte texture) encountered in Corehole-17 in the center of the Khufai Dome (Al Huqf), for which Gorin et al. (1982) reported a K-Ar age of 654 ± 12 Ma. These age estimates range by some 100 My (654–554 Ma) and bracket the age of the Shammar Rhyolite centered at ca. 620–585 Ma and Badayi Formation (558 Ma) in Saudi Arabia.
Then in 2001 Leather reported that welded tuffs in the Halfayn Formation have U-Pb-zircon ages of ca. 802 Ma (Figure 3). In 2007, Bowring et al. analyzed four zircon grains from a Halfayn rhyolitic ignimbrite with well-preserved volcanic textures, and obtained 207Pb/206Pb dates ranging from 850823 Ma. They noted that these zircon grains are indistinguishable from those of the basement, and given this possible inheritance and Pb loss, the eruptive age of the rhyolite cannot be determined. Four more zircon grains from an underlying sample taken from a rhyolitic ash-flow tuff were dated by 207Pb/206Pb at 840–803 Ma; they also showed evidence of basement inheritance. Four zircon grains taken from a sample of volcaniclastic and calcareous breccia and conglomerate, above the other two rhyolite samples, yielded ages of 850–827 Ma. Bowring et al. (2007) concluded: “All samples have variable U-Pb systematics that preclude precise age determination, however the data strongly suggests that these undeformed, low-grade rocks [Halfayn Formation] are approximately 823 Ma, or slightly older.”
In Saudi Arabia the Jibalah Group unconformably overlies the Shammar Rhyolite dated at ca. 620–585 Ma; see references inAl-Husseini, 1989, 2000). Rhyolites are attributed to mantle-derived magmas in intra-continental rifts (Brown et al., 1989) suggesting they heralded widespread extension in many regions in the Arabia Shield starting at ca. 620 Ma. In Oman, the discarded age for the Halfayn rhyolite (562 ± 42 Ma) originally offered a possible Shammar (or Badayi) correlation. Perhaps choosing between zircon geochronology and alternative age dating techniques requires increasing the scale-of-investigation to plate tectonics.
GLOBAL AND ARABIAN PLATE TECTONICS
Was Oman a Separate Microplate in Earliest Cambrian?
One geological branch of the Cryogenian Model considers most of South and Central Oman as a separate Proterozoic microplate that collided in the Early Cambrian (ca. 530 Ma) with the rest of Arabia along the Western Deformation Front (Figures 1 and 2; see suture in figure 10 inCraig et al., 2009b, SP 326). This model apparently derives from conference abstracts that attribute Rift-3 (Ara and Nimr groups) to the evolution of syn-collisional foreland basins and the Angudan Unconformity/Event to an Early Cambrian collision (Immerz et al., 2000; Koopman et al., 2007). Allen (2007) argues that Oman was attached to the rest of Arabia long before ca. 530, at least by 645 Ma. He attributes Rift-3 to the possibility that Oman was located in a retroarc setting between 547–525 Ma.
These models contrast to the Ediacaran Model, which considers the Angudan Event as ending intra-continental rifting (Al-Husseini, 1989, 2000; rift sag unconformity of Loosveld et al., 1996, and Droste, 1997). However, it remains unclear if the Angudan Event represents strike-slip adjustments in Oman (J. Mattner, personal communication, 2009). In the context of regional and global tectonics it seems unlikely that Oman was a separate microplate, even as far back as ca. 750 Ma as suggested by Scotese (2009, SP 326).
Arabia in the Rodinia to Pannotia Supercontinents
Scotese’s (2009) maps tentatively take us all the way back to 750 Ma when the Rodinia Supercontinent (lasting ca. 1,100–750 Ma) started breaking up. He describes how between 750–650 Ma North and South Rodinia may have drifted apart only to come back and collide between ca. 650–600 Ma to close the EW-trending Mozambique Ocean (Figures 7 and 8). He names the Ediacaran supercontinent Pannotia (meaning all southern land, sometimes Greater Gondwana, Vendian or Pan African Supercontinent). The closing suture of the North-South Rodinia collision (650–600 Ma) is here referred to as the Amar-Mozambique Suture (Figures 1, 7 and 8; also shown in some SP 326 papers; e.g. Lottarolli et al. and Lüning et al. 2009). Scotese’s reconstructions from 750 to 540 Ma show most of Gondwana’s usual continents were attached together as in the Paleozoic Era (see review inRuban et al., 2007). Importantly his 750–540 Ma reconstructions do not show an Oman microplate.
In greater tectonic detail the western part of the present-day Arabian Plate consists of the Midyan, Hijaz and Asir terranes that continue on the African side of the Red Sea, mainly in Egypt and Sudan (Stoeser and Camp, 1985). Scotese’s 700 Ma map groups them into the Hijaz Arc (consolidated by 715 Ma, Figures 7 and 8). They were located somewhere in the Mozambique Ocean and apparently accreted along NE Africa by ca. 700 Ma (Figure 8a). East of the Hijaz Arc, the middle part of the Arabian Shield consists of the Proterozoic Afif Terrane (Stoeser and Camp, 1985; Figures 7 and 8). It is bounded by the NS-trending Nabitah and Amar suture zones but is not explicitly labeled in Scotese’s maps. Between 680–640 Ma the Afif Terrane closed the Nabitah Sea and collided with the Hijaz Terrane along the Nabitah Suture (containing ophiolites, Nabitah Orogeny, Figure 8b). The elongated Afif Terrane may have connected southwards across Yemen with the Kalahari Terrane of East Africa (Figures 7 and 8).
The Rayn Terrane represents the easternmost part of the Arabian Shield and most if not all of the present-day Arabian Plate east of the Amar Suture and Amar Arc (Figures 7 and 8). The ophiolites of the Amar Suture represent the remains of the Rayn Sea between the Rayn and Afif terranes (Figure 8b). The Rayn-Afif collision caused the Amar Orogeny at ca. 640–620 Ma (Figure 8c; inNehlig et al., 2002, their figure 2 combines the Amar and Nabitah orogenies into the Suturing Orogeny between ca. 680–610 Ma). The Amar Suture almost linearly projects to the Mozambique Suture representing the North-South Rodinia collision that completely closed the Mozambique Ocean at ca. 650–600 Ma (Figure 8c). The linearity and overlapping collisional age windows suggest the Rayn Sea and Mozambique Ocean (Figures 7 and 8) were underlain by the same oceanic plate. The Rayn Sea’s subducting slab was directed below the Rayn Terrane to produce the Amar Arc (Figure 8b), which extended northwards to the Zagros Suture and may have continued into Yemen as the undated Al-Mukalla Island Arc of Windley et al. (1996).
Of relevance to choosing between the Miranoan versus Varanginian glaciations the Rayn Terrane (including Oman) was at low latitudes in the Pannotia Supercontinent (Figure 7) at ca. 650–600 Ma. Pannotia apparently formed an almost pole-to-pole supercontinent thus presenting an ideal setting for the Miranoan ice age (663–636 Ma): (1) large continents surrounding the South Pole, (2) blockage of warm currents from flowing EW along the closing equatorial Mozambique Ocean, and (3) high mountains raised around the Amar-Mozambique Suture. A Miranoan Glaciation in low-latitudinal mountainous Arabia would have been in an Alpine setting, but not compatible with the glacio-marine rift setting for the Abu Mahara diamictites. Nor was Pannotia’s mountain-building phase (650–600 Ma) compatible with the tectonic quiescence and depositional setting of the Nafun carbonates and marine clastics (presumed starting at 636 Ma). Arabian Plate and global plate tectonics do not suggest that these sedimentary rocks were deposited in the Cryogenian and early Ediacaran time, but more likely during the collapse of the Pannotia Orogen (after ca. 620 Ma).
Basins and NW-trending Fault Systems
The widespread pattern of parallel NW-trending strike-slip faults and lineaments that intersect pull-apart and sag basins, or terminate failed intra-continental rifts, was tentatively connected as the Arabian Infracambrian Extensional System (Husseini, 1988, Husseini and Husseini, 1989). This pattern is now emerging in SP 326 in Neoproterozoic Libya where seismic lines show the Al-Kufrah sag basin with a diameter of ca. 60 km containing thick sedimentary infill (Bashati et al. 2009; Aziz and Ghnia, 2009; Lüning et al., 2009). The pattern becomes more striking by the enhancement of digital elevation model data of the 6,000 km NW-trending Tibesti Lineament from Somalia, through the Al-Kufrah region that continues to Algeria (Le Heron et al., 2009, and references therein). The straight-and-narrow Tibesti Lineament is not only parallel to the Najd Fault System but nearly six times longer than its exposed segments. Detailed mapping of coeval tectono-stratigraphic domains, rift basins and discrete fault segments could provide a better tectono-geometric understanding for the collapse of the Pannotia Orogen.
Returning to Arabia, the Najd Fault System projects SE to near Salalah located along the coast of Oman (Figures 2 and 4) where the El Hota-Ain Sarit Formation crops out below Cretaceous and Tertiary unconformities (Platel et al., 1987; Roger et al., 1989). J. Mattner (personal communication, 2009) reported that the Formation is heavily folded and faulted, and suggested it could be the tectonized equivalent to the Arkahawl Formation in the Mirbat outcrops. His preliminary assessment of selected structural features in the El Hota-Ain Sarit Formation indicate it consisted of unconsolidated mainly turbiditc siliciclastics deposited in a transtensional tectonic setting. He interpreted its syn-depositional and structural setting started with listric normal faulting. His interpretations are consistent with the Najd Fault System leaving the Arabian Shield and continuing SE through the greater Salalah region. The deep-marine siliciclastics of the El Hota-Ain Sarit and Arkahawl formations would therefore correlate to the Masirah Bay Formation and represent the Nafun Transgession. But the Najd Fault System switched from transpressional to transtensional after ca. 600 Ma again arguing all these formations are late Ediacaran.
Which Najd Faulting Phase?
The chrono-tectonic phases of the Najd Fault System represent another issue in GeoArabia’s Infracambrian Debate. This issue is partly semantic because all NW-trending faults in Arabia are referred to as “Najd” regardless of the age of their movements. The Ediacaran Model proposes that the present-day Arabian Plate’s NS faults, folds and anticlines and conjugate NW- and NE-trending fracture and fault systems are consistent by age and geometry with the EW-directed compression during the Nabitah Orogeny (ca. 680–640 Ma) and Amar Orogeny (ca. 640–620 Ma; Al-Husseini, 2000). EW compression stopped between ca. 620–585 Ma (Shammar Rhyolite and coeval post-tectonic igneous rocks, The Collapse) and switched to left-lateral NW-strike-slip (Escape Tectonics of other authors) as tentatively sketched in the Arabian Infracambrian Extensional System (ca. 585–530 Ma, Al-Husseini, 2000).
The role of the Najd fault System in the Debate was explicitly raised 21 years ago when Brown et al. (1989) wrote on p. 103: “These reactivated and rotated older [Najd] faults give the unwarranted impression that the age of the Najd faulting event began long before 600 My, but we cannot agree with this interpretation.” Their statement emphasizes that the post-600 Ma tectonic regime is completely different from the pre-600 Ma one, even though both involved Najd-trending faults. The younger one exploited the weakened NW-striking crustal zones left over by the pre-600 Ma orogenies. Brown et al.’s position can be summarized by their statement: “The [post-600 Ma] Najd grabens [Jibalah grabens] formed in a crust that was fully continental and along newly formed, deep-crustal shear structures that tapped magma generated in the mantle.” They offer several age constraints including: “A high-quality, whole-rock Rb-Sr isochron of 577 ± 15 My (Baubron et al., 1976) on granite is probably slightly older than the Jibalah Group, because the granite intrudes the Murdama Group and disconformably underlies the basal Jibalah conglomerate.” The Murdama Group is dated 680–610 Ma and together with the Gneiss Domes represents the Suturing Orogeny of Nehlig et al. (2002; combined Nabitah and Amar orogenies). It is the post-600 Ma phase of Najd faulting that is relevant to the Infracambrian Debate.
The unwarranted impression objected to by Brown et al. (1989) has not been fully appreciated or accepted and taken a central role in the Debate. The Najd faults were indeed first formed during the Nabitah and Amar orogenies (ca. 680–620 Ma) and in some cases rotate into the NS-trending Nabitah and Amar sutures. These aspects have been taken to imply post-600 Ma Najd faults did not laterally dislocate the ca. 680–640 Ma Nabitah Suture (Passive Marker) by ca. 300 km (e.g. Johnson and Woldehaimanot, 2003). Some authors lump all Pannotia’s pre- and post-600 Ma NW-trending fault systems into broad mobile belts that fail to distinguish collision from extension: the ca. 650–530 Ma all-encompassing Pan African Orogeny (Jacobs and Thomas, 2002).
The rejection of the post-600 Ma ca. 300 km dislocation of the Nabitah Suture into a step-wise linear pattern of ophiolite belts by three main Najd strike-slip fault zones is inconsistent with the style of the Jibalah basins. They are mappable as discrete and narrow (1–10 km), NW-elongated (up to 100 km) fault-bounded basins that contain as much as 5,000 m of Jibalah sedimentary (including marine carbonates) and volcanic rocks. In the Arabian Shield the Jibalah basins resemble the pull-apart basins along the Aqaba-Dead Sea Fault. Aside from the 300-km dislocation of the Nabitah Passive Marker, the Jibalah basin’s ca. 10s to 100-km NW-elongated shapes indicate major strike-slip movement of a similar order of magnitude along each of its three zones.
Mapping the Arabian Infracambrian Extensional System in greater detail could support exploration by better explaining the origins, geometries and structural evolution of Infracambrian basins. They occur in a complex mosaic of NS-trending horsts-and-grabens and NE-trending rift basins terminated by multi-1,000 km NW-trending strike-slip faults. Mapping the subsurface parts of the system would require much geophysical and borehole data. Such an exercise could simultaneously predict their petroleum stratigraphy and systems.
The Geological Society’s SP 326 shows the scientific and commercial importance of Neoproterozoic rocks worldwide, with examples of producing fields and unexplored hydrocarbon habitats. The frequent reference to Oman’s Huqf Supergroup in SP 326 and other publications testifies to Petroleum Development Oman’s scientific and E&P leadership roles towards the understanding of Neoproterozoic and Early Cambrian geology. Indeed Oman’s Huqf Supergroup is one of most stratigraphically complete and best-documented successions in the public domain and therefore offers the type section for one part of the Earth’s history – but for what time interval? The answer not only impacts Earth Science but also has great commercial consequences.
As recognized by many SP 326 authors, the great sea-level rises and falls caused by the advances and retreats of the great Neoproterozoic glaciers presented suitable settings for the deposition of thick deep-marine organic-rich source rocks, many reservoirs and seals (including massive evaporites), especially in rift basins. But clearly there were many Neoproterozoic glaciations and distinguishing them in a Snowball Earth (even without frozen oceans) is challenging, whether in Oman or elsewhere. Each major Neoproterozoic glacial cycle will have different settings implying a highly diversified spatio-temporal distribution of petroleum systems; some of the oldest ones may have survived orogenies but most will undoubtedly have been destroyed by metamorphism and erosion.
Determining the age of the glaciogenic rocks of the Huqf Supergroup is crucial if it is to be used as a chrono-stratigraphic type section and exploration analogue worldwide. This is why it is important to resolve GeoArabia’s Infracambrian Debate. The adoption of the Cryogenian Model of Oman has already dominoed worldwide and is considered by some as a given-fact. It presents for Oman a highly complex scenario involving two Cryogenian glaciations, a missing late Ediacaran Varanginian Glaciation and multiple phases of rifting and hiatuses. It critically hinges on the Cryogenian zircon geochronolgy of the Ghubrah diamictite, which is inconsistent with much younger Ediacaran radiometric ages and regional considerations. The here-advocated Ediacaran Model for Oman offers a single-rift model that fits all data and regional considerations except for the interpretation of zircon geochronology.
The Geological Society (London) is thanked for inviting me to review their Special Publication 326 on behalf of GeoArabia. I hope my review, although prejudiced and incomplete, will nevertheless trigger greater participation in GeoArabia’s Infracambrian Debate. I thank S.I. Al-Husseini, J. Mattner, K. Romine and D. Vaslet for discussions and pointing out many papers and public domain maps and explanatory notes for the Arabian Shield and Oman. Arnold Egdane is thanked for designing the paper.
ABOUT THE AUTHOR
Moujahed I. Al-Husseini founded Gulf PetroLink in 1993 in Manama, Bahrain. Gulf PetroLink is a consultancy aimed at transferring technology to the Middle East petroleum industry. Moujahed received his BSc in Engineering Science from King Fahd University of Petroleum and Minerals in Dhahran (1971), MSc in Operations Research from Stanford University, California (1972), PhD in Earth Sciences from Brown University, Rhode Island (1975) and Program for Management Development from Harvard University, Boston (1987). Moujahed joined Saudi Aramco in 1976 and was the Exploration Manager from 1989 to 1992. In 1996, Gulf PetroLink launched the journal of Middle East Petroleum Geosciences, GeoArabia, for which Moujahed is Editor-in-Chief. Moujahed also represented the GEO Conference Secretariat, Gulf PetroLink-GeoArabia in Bahrain from 1999-2004. He has published about 50 papers covering seismology, exploration and the regional geology of the Middle East, and is a member of the AAPG, AGU, SEG, EAGE and the Geological Society of London.