Shallow burial dolomitization of an Eocene carbonate platform, southeast Zagros Basin, Iran

Dolostone-capped shallow-water carbonate successions have been reported throughout the geologic record (Bosence et al., 2000; Jones, 2007; Rameil, 2008; Geske et al., 2012; Zhao and Jones, 2012; Meister et al., 2013; Corbella et al., 2014). Amongst these, pervasive secondary dolomitization of shallow-water carbonates is the most abundant dolostone type, but also one that is genetically the least understood (Budd, 1997; Coniglio et al., 2003; Frazer et al., 2014). This is because many dolostone bodies, formed under different depositional and diagenetic environments worldwide, are characterized by similar fabrics and geochemical features (Sass and Katz, 1982; Machel and Mountjoy, 1986). The ongoing interest in the topic of pervasive dolomitization is due to the fact that many hydrocarbon reservoirs worldwide are located in dolomitized successions (Purser et al., 1994; Braithwaite et al., 2004; Ronchi et al., 2011; Rott and Qing, 2013; Wen et al., 2014).

Current understanding is that massive dolostones can be produced by fluids of various origin in different diagenetic environments. These include evaporitic marine brines in sabkha depositional settings (e.g., Geske et al., 2012; Bontognali et al., 2012; Wen et al., 2014), evaporated seawater in seepage reflux environments (e.g., Melim and Scholle, 2002; Al-Helal et al., 2012; Vandeginste et al., 2013), normal seawater in subtidal environments (e.g., Rameil, 2008; Maliva et al., 2011; Zhao and Jones, 2012), freshwater/seawater in mixing zones (e.g., Searl, 1988; Gaswirth et al., 2007; Azmy et al., 2009; Goldstein et al., 2012; El Ayyat, 2013; Li et al., 2013) and deep basinal fluids (e.g., Green and Mountjoy, 2005; Lonnee and Machel, 2006; Ronchi et al., 2011; Frazer et al., 2014).

One of the more often cited dolomitization models suggest that admixtures of evaporative brines with seawater acted as parent fluids for massive dolomite bodies in numerous shallow carbonate platforms (Mriheel and Anketell, 1995, 2000; Holail et al., 2005; Kirmaci, 2008; Salad-Hersi, 2011; Conliffe et al., 2012; Wen et al., 2014). This model, however, cannot be applied directly to extensive dolomitization of limestones that are not associated with important gypsum/anhydrite deposits. Simms (1984) has shown that the reflux of fluids of slightly elevated salinity on ancient shallow platforms during periods of hydrographic restriction and climatic aridity is favorable for dolomitization of thick carbonate sequences. According to Sibley (1991) and Vahrenkamp et al. (1991), the formation of massive dolomite requires a long residence time for the dolomitized body in a near sea-level position. Essentially, the most favorable setting for dolomitization is perhaps that of prolonged sea-level highstand under slow subsidence rates. On the other hand, Sun (1994) concluded that hydrographic restriction related to frequent pulses of sea-level fall under arid climate seem to be critical factors responsible for the massive dolomitization of Paleogene dolostones in several basins worldwide.

Massive replacive dolostones have been reported from the Zagros Basin of Iran, the basin under study here, and several of these dolostone bodies are important regional hydrocarbon reservoirs (Sun, 1995; Warren, 2000; Zohdi et al., 2011). Understanding the genesis of dolomitized limestone might aid the prediction of the distribution of dolostone bodies and shed light on geochemical fluxes of fluids in the subsurface. Here, we document and discuss a case example of an Eocene, pervasively dolomitized ramp (Jahrum Formation) from the Zagros Basin in Iran. Published data on dolomitized ramp carbonates from this region are rather limited (Zohdi et al., 2013), in comparison to the better studied Permian and Triassic dolomitized ramps in the Zagros Basin (Moradpour et al., 2008; Rahimpour-Bonab et al., 2009, 2010; Tavakoli et al., 2011; Esrafili-Dizaji and Rahimpour-Bonab, 2013; Mohammadi Dehcheshmehi et al., 2013). In general and particularly so in Iran, dolomitization of Permian–Triassic platforms has been mainly ascribed to seepage-reflux and/or evaporative mechanisms (e.g., Moradpour et al., 2008; Rahimpour-Bonab et al., 2009, 2010; Geske et al., 2012; Meister et al., 2013; Jiang et al., 2013).

In this paper we aim (1) to provide a comprehensive characterization of the petrography, geochemistry and spatial architecture of Jahrum Formation dolostones; and (2) to discuss the origin of this dolostone facies in its basin-wide context.

The Iranian plateau extends over a number of continental terranes welded together along suture zones of oceanic character (Berberian and King, 1981; Alavi, 2007). These terranes include the following provinces: (1) Zagros; (2) Alborz; (3) Central Iran; (4) Kopeh-Dagh; and (5) Makran sedimentary basin. The study area is located in the southeastern Zagros Basin (Figure 1). This basin constitutes a major structural feature within the Alpine-Himalayan Orogen, marking the transition between the Zagros Collision Belt to the west and the Makran and Oman Mountains to the east. Numerous salt diapirs characterize the southeastern Zagros Basin (Edgell, 1996; Jahani et al., 2009). These diapirs are composed principally of Upper Precambrian–Lower Cambrian Hormuz Salt. Where undeformed, the salt is overlain by more than four km of sedimentary rocks (Edgell, 1996; Jahani et al., 2009). The emergent diapirs provide an opportunity to study the diagenetic history of the affected units in association with the Hormuz Salt (Ghazban and Al-Aasm, 2010). The Hormuz Salt started to mobilize as early as Jurassic to Early Cretaceous based on geological evidence, but most diapirs did not reach the surface until the folding of the Zagros Mountains during the Paleogene (Ala, 1974; Jahani et al., 2009). At the initiation of the folding, diapirs had already been re-activated by earlier tectonic events and salt movement along faults resulting in evaporitic facies reaching the surface (Jahani et al., 2009). The overall geotectonic setting clearly affected the evolution of the Paleogene carbonate platforms in the southeast Zagros Basin (Zohdi et al., 2013). Hormuz salt diapirs are potentially of significance in the context of dolomitization processes as a source of saline ascending fluids (Ghazban and Al-Aasm, 2007, 2010).

In the study area, anticlines are built most often by competent limestone rocks of the Eocene Jahrum and the Lower-Middle Miocene Gurpi formations (Figure 1). Here, large-scale anticlines and synclines mostly have an E-W trending orientation that differs from other regions of the Zagros Basin, which have NW-SE oriented structures that trend parallel to the main Zagros Orogen (Molinaro et al., 2004). From north to south, exposures visited in the context of this study are located in the Faraghun (27°54′N and 56°24′E), Finu (27°48′N and 56°10′E) and the Genow anticlines (27°28′N and 56°12′E) (Figure 1).

Within the studied outcrops, the basal contact of the Jahrum Formation with the underlying Gurpi Formation is conformable. The top of the unit is placed at a regional disconformity (Zohdi et al., 2013; Figure 2). In all studied outcrops, the Jahrum Formation is overlain by red sandstones and conglomerates of the Oligocene Razak Formation (Zohdi et al., 2013; Figures 2 and 3). The facies changes from shallow-water carbonate deposition on a ramp to falling-stage and lowstand mixed carbonate-siliciclastic shelf of the Razak Formation near the Eocene/Oligocene boundary. This is probably best explained by a eustatic sea-level fall at that time (Haq et al., 1987; Zachos et al., 2001; Sharland et al., 2001; Miller et al., 2005; Erhardt et al., 2013). Generally, there is no biostratigraphic evidence for a Late Eocene depositional age of sedimentary successions in the southeastern Zagros Basin such as found in many other localities on the Arabian Plate (e.g. Hughes Clarke, 1988; Brannan et al., 1999; Sharland et al., 2001). Consequently, in all studied outcrops, Upper Eocene strata are missing, a feature that seems plausible in the context of high-amplitude sea-level fall and subaerial emergence of the carbonate seafloor (Zohdi et al., 2013).

The Jahrum Formation carbonate platform was attached along the northern margin of the Arabian Plate. This formation was deposited as an Eocene carbonate ramp facies in the southeastern Zagros Basin. Facies and depositional environments of the Jahrum Formation in the southeastern Zagros Basin are described in detail in Zohdi et al. (2013). Eight facies were recognized from the shallower to deeper part of the platform. These are as follows: Orbitolites facies, Alveolina facies, Coskinolina-Dictyoconus facies and Miliolid red algae facies were deposited in inner-ramp settings; Nummulites facies and Nummulites-Discocyclina facies were recognized in the middle-ramp setting; and lime mudstone facies and re-sedimented deposits were determined in the outer-ramp setting (Zohdi et al., 2013). Thin-section evidence indicates that most of these facies are micrite supported (packstone to wackestone) with the groundmass being fine-grained carbonate mud). Marine and other cement phases are comparably rare and only recorded where more frequent in rare instances. There is a southward dipping inner-ramp-to-basin transect and dolomitization occurs in the proximal part of the Jahrum carbonate ramp, within facies characterized by a restricted fauna (Orbitolites, Alveolina, miliolids) (Zohdi et al., 2013).

The Jahrum Formation is the time-equivalent of the Radhuma and Dammam formations in Saudi Arabia, Kuwait and southeastern Iraq (Sharland et al., 2001; Ziegler, 2001; Zohdi et al., 2013). In the study area, the Jahrum Formation reaches a cumulative thickness about 600 m and can be divided into two major stratigraphic units: (1) A lower unit, reaching a stratigraphic thickness of about 300 m and consisting mostly of medium bedded dolostone (Figure 4a) with abundant fossil moulds of the larger benthic foraminifera and green algae (Figure 4b); (2) an upper unit composed of dolomitic limestone intervals alternating with very thickly-bedded limestone with abundant shallow-water benthic foraminifera (Figure 4c). Here the focus is on dolostone facies in both units.

Thinly to medium-bedded dolostones within the Jahrum Formation display numerous depositional structures and fabrics (e.g., lamination, bioturbation and burrows, microbial lamination; Figures 4d, e). The presence of a shallow-water biota including Alveolina and miliolid foraminifera suggests that the depositional environment of the precursor limestone, now dolostone, is best assigned to an inner to mid-ramp setting. This notion is in agreement with the observations made of the limestone portions of the Jahrum Formation as described in Zohdi et al. (2013). The transition between dolomites of the Jahrum Formation to limestones (with lagoonal sediments) is sharp and is easily recognized in all studied outcrops (Figure 4c). The focus of this paper is on the lower, dolostone unit of the Jahrum Formation.

For the present study, three major outcrops exposing Jahrum Formation dolostones in the southeastern Zagros Basin were logged and studied bed-by-bed (Figures 1 and 2). The sections measured a total length of ca. 2,000 m. Lateral and vertical facies variations were described and attention was paid to discontinuities observed in the field and particularly so to the regional unconformity between the Jahrum and Razak formations. A total of 1,200 rock samples were collected based on selection criteria such as stratigraphic position, facies and petrography.

Almost 1,000 thin sections of pervasively dolomitized intervals where studied using transmitted light microscopy. The aim was to assess precursor depositional fabrics, general dolomite crystal properties and textures. In order to differentiate ferroan and non-ferroan calcite from ferroan and non-ferroan dolomite in thin sections, the staining method of Dickson (1965) was applied. According to Sibley and Gregg (1987) and based on petrographic characteristics (e.g., crystal size and shape, extinction, characteristics of allochems and fabric), several main phases of dolomite development are recognized.

Nine samples were investigated using a hot stage HC1-LM cathodoluminescence microscope (CL) and gold-coated thin sections polished on both sides (cf. Neuser et al., 1996) at the Institute of Geology, Mineralogy and Geophysics, Ruhr-University Bochum, Germany. The CL microscope is connected to a high-resolution spectrograph (EG&G “digital grating spectrograph”, N2-cooled CCD detector) via a quartz light transmission cable for the spectral analysis of the measurements (spatial resolution: ca. 30 μm, electron beam energy: 14kV).

Morphologies and internal structures of selected dolomite crystal types were determined by use of a VEGA\TESCAN Scanning Electron Microscope (SEM) at the Razi Metallurgical Research Center, Iran. For this purpose, small fresh samples were broken, polished and etched with 30% HCl for 10–15 seconds then sputtered with gold prior to being examined on the SEM with an accelerating voltage of 15 kV. For more details refer to Jones (2005).

Powder x-ray diffraction (XRD) was carried out at the Ruhr-University Bochum, Germany on eight dolomite samples in order to determine: (1) the approximate mole% of CaCO₃, and (2) degree of ordering of different dolomite phases. For XRD analyses, powdered samples with quartz as an internal standard were smear-mounted on glass slides. Peak areas for calcite and dolomite were determined using a Phillips X’Pert MPD Theta-Theta X-Ray Diffractometer with Cu Ka radiation, include high voltage of 45 kV, a current of 40 mA, an angular range from 20 to 40, a step size of 0.025°2θ and a counting time of 11 seconds.

Mineral concentrations were calculated from peak area ratios assuming that each sample was composed only of calcite and dolomite. Peak area ratios were calibrated and the concentrations calculated using calibration curves prepared from results using a series of pure mineral standards (verified by XRD analysis). The mole% CaCO₃ content of each dolomite type was determined using the equation N_CaCO3 = md + b (b = -911.99, m = 333.33 and d is the main diffraction peak (104) in Angstrom) of Lumsden (1979). The degree of dolomite ordering reflects the cation distribution in the Ca and Mg layers of the lattice, and is generally expressed as the quotient of the intensity of the superstructure reflection d(015) to the neighboring reflection d(110) and calculated by using the equation: R=Id(015)/Id(110). For more details refer to Füchtbauer and Richter (1988).

Oxygen (δ¹⁸O) and carbon (δ¹³C) isotope analyses were performed on 33 bulk-rock dolostone and 60 associate limestone samples at the Institute of Geology, Mineralogy and Geophysics, Ruhr-University Bochum, Germany. Limestone samples were taken from sites near limestone-dolostone contacts. Samples for geochemical analysis were selected based on thin-section characterization of different facies types. For carbon- and oxygen-isotope analysis, 0.3 ± 0.03 mg dolomite material was weighed in and analyzed on a Thermo-Finnigan MAT 253 coupled to a Gasbench II and a PAL auto-sampler (Gas-IRMS). All isotope ratios are reported as per mil (‰) deviation from the Vienna Pee Dee Formation belemnite (V-PDB) standard. The certified carbonate standards NBS 19, IAEA CO-1 and CO-8 and an internal standard (RUB standard) were used. The standard deviations of 10 RUB standards measured within these sample sequences are 0.06‰ and 0.08‰ for C and O isotopic measurements, respectively. Please refer to Geske et al. (2012) for analytical details.

⁸⁷Sr/⁸⁶Sr ratios can be diagnostic of the precipitating pore fluids and shed light on alteration effects. Two to six mg of dolostone sample powder was dissolved at 125°C in PFA-beakers using 1 ml 6 M HCl. Before 1 ml of the solution was moved onto quartz glass columns, filled with Bio-Rad ion exchange resin AG50W-X8, the solution was centrifuged for 10 min to partition the insoluble silicate material. The Sr fraction was separated twice according to column calibration pattern. After final evaporation on a hot plate at > 100°C the samples (approximately 100 ng Sr) were re-dissolved for mass spectrometry in 1 μl of ionisation enhancing liquid (Birck, 1986) and loaded on Re-filaments.

The ⁸⁷Sr/⁸⁶Sr ratio was determined with a 7 collector Thermal Ionisation Mass spectrometer (TIMS) MAT262 in 3 collector dynamic mode at the Ruhr-University Bochum, Germany. As standard reference materials NIST NBS 987 and USGS EN-1 were chosen. The total blank for Sr isotope analysis, including chemical separation and loading blank, is 1.5 ng. The repeatability was tested with the USGS EN-1 modern bivalve carbonate, which passed through the identical procedures as all samples. The average Sr value is 0.709160 ±0.000027 2s (n=209). The reproducibility of Sr measurements represented by NBS987, which was directly loaded onto a Re filament, is 0.710240±0.00034 2s (n=233). Please refer to Faure and Powell (1972) and Geske et al. (2012) for analytical details.

In order to complement isotope data, major and trace elemental analysis was performed from aliquots of all samples (29 bulk-rock dolostone and 56 associate limestone samples). Approximately 1.5 ± 0.15 mg of the dolostone samples were dissolved in 3 M HNO₃. Subsequently, the solution was diluted with 2 ml of deionized H₂O (> 18.2 M cm^-1). Using an inductively coupled plasma optical emission spectrometer (ICP-OES, Thermo Fisher Scientific iCAP 6500 DUO) the concentration of Ca, Mg, Fe, Mn and Sr of samples was measured at the Ruhr-University Bochum, Germany. Within each set of samples, eight samples of certified reference material (BSC-CRM-512, dolostone and BSC-CRM-513, limestone) were analyzed. All major and trace elemental results are reported in ppm (parts per million) and errors are given as ±%RSD. Please refer to Geske et al. (2012) for analytical details.

Dolomite crystals range in size from microcrystalline to less than 150 μm, most being in the tens-of-microns range. Based on the degree of dolomitization of the precursor limestone facies, thin-section observations such as visual estimation of the abundance of dolomite crystals, and the crystal size and shape, characteristics of allochems and fabric, four types of dolostone are distinguished (Figures 5–8) and described below.

Type I dolomites consist of idiotopic, isolated rhombs of subhedral to anhedral, finely crystalline dolomite crystals that mimetically replaced limestones with locally well-preserved skeletal remnants and sedimentary features (Figures 5a–f). Dolomite crystals are generally fabric retentive. Based on thin-section observations of relict features, the pre-dolomitization facies has attributes of wackestones and packstones. Associated grains include benthic foraminifera, bivalves, algae and peloids. Some floating dolomite rhombs are cross-cut by stylolites, suggesting that they formed prior to stylolitization. Fine crystalline dolomite (< 20 μm) is generally non-luminescent, but idiotopic, isolated rhombs commonly reveal zonation (different shades of red) (Figure 5f). Dolomite Type I forms an estimated 10% of the dolomitized rock volume in the measured sections (Table 1).

Dolomite Type II is composed of anhedral to subhedral crystals, ranging from 20 to 100 μm in size (Table 1; Figures 6a–d). These dolomites commonly contain both fabric-retentive and fabric-destructive textures. No undulose extinction was observed. Scanning electron microscope photographs display dissolution pits in the cores of pervasive dolomite rhombs and thin, relatively pit-free rims (Figures 6c, d). Type II dolomites are mostly zoned with a characteristic succession of three zones (1–3), from center to rim, increasingly bright orange to yellow luminescent zones fringed by an outermost (4), dark brown luminescent zone (Figures 6e–g). Cathodoluminescence spectra of greenish-yellowish dolomite Type II are slightly asymmetrical with a second peak at 575 nm (Figure 6g).

Dolomite Type II is present as calcareous dolostones characterized by a former lime mudstone matrix with skeletal and non-skeletal grains now largely replaced by dolomite. Some of the large agglutinated benthic foraminifera and other large bioclasts are characterized by a low susceptibility to replacement and dissolution, compared to their micritic matrix (Figure 6a). Where dolomitization becomes spatially more intense, nearly all intergranular micrite has been replaced by dolomite and bioclasts and peloids exhibit varying degrees of dissolution ranging from minimal corrosion to complete dissolution to form mouldic pores. Matrix dolomite is the most abundant type in the dolomites of the Jahrum Formation, forming about 70% of the total dolomite by volume.

The third type of dolomitization consists of former limestones that are now pervasively dolomitized (Figure 7a). In general, dolomite Type III is texturally destructive (fabric-destructive) and modifies or obliterates earlier diagenetic features. Minor calcitic remnants appear mainly as inclusions within single dolomite crystals. The dolomite ranges from a dense, interlocking, non-sucrose texture with medium to coarse crystalline dolomite crystals (> 100 μm; Table 1; Figure 7b), to a medium-crystalline dolostone with abundant biomouldic porosity (e.g., benthic foraminifera, bivalves and algae; Figure 7a). Crystals are unimodal with a medium and coarse crystalline planar-s texture and lack undulatory extinction. Dolomite crystals are typically ‘dirty’ in transmitted light because of sub-micrometer mineral inclusions, fringed by clear rims (cloudy core-clear rim). Cathodoluminescence revealed that the dolomite crystals formed in four subsequent phases including (1) blotchy to blotchy brown; (2) dark-red; (3) yellowish green/or yellowish orange with an asymmetrical luminescence-spectra which contains two peak positions of the distribution of Mn²⁺ at the Ca position (575 nm) and the Mg position (656 nm); and finally (4) dark-red to red luminescence (Figures 7c–e).

Dense and crystalline dolomites of the Jahrum Formation (n = 8) range from 50 to 55 mole% CaCO₃, i.e. are slightly calcitic, but most of them display a stoichiometric to near-stoichiometric composition and contain 50 to 52 mole% CaCO₃ in their lattice. These dolomites have moderate to high degree of lattice ordering as revealed by d(015) / d(110) quotients of 0.47 to 0.94.

Type III dolomite crystals tend to be tightly intergrown (and/or packed together), a feature that has been amplified by compaction. Crystals of this type of dolomite generally have compromised curved, serrated or irregular intercrystalline boundaries, display slightly undulatory extinction, and show a turbid appearance because of abundant inclusions (Figure 7c). In places, interlocked dolomite crystals are closely packed and may have poorly defined boundaries. The matrix-replacing dolomite and dense and crystalline replacement dolomites are cross-cut by stylolites and by late calcite veins not further investigated in this context. Based on estimates based on point counting, dense and crystalline dolomite comprises about 18% of the total dolomite in the Jahrum Formation.

Type IV dolomites or dolomite cements occur as euhedral to subhedral crystals occluding secondary pores formed due to the leaching of bioclasts (biomoulds) such as larger foraminifera, pelecypod shells and green algae (Figures 8a, b). The crystals of the dolomite cement are clear (limpid) but in places may contain inclusions and range from medium to coarse sizes (50–200 microns). Under CL, this dolomite type reveals similar zonation patterns as present for dolomite Type I to III, but dolomite Type IV crystal diameters are larger compared to Types I and III (Figures 8c, d). Type IV occurs in only a few samples and is volumetrically insignificant. Point counting suggests that void-filling dolomite cements contribute, on average, less than 2% of dolomites of the Jahrum Formation (Table 1).

Dolomitized limestones of the Jahrum Formation occur entirely within shallow-water carbonate facies. Because the dolomite is dominantly matrix-replacive, its abundance is proportional to the amount of precursor lime mud present in the rock. Dolomites of the Jahrum Formation are most abundant in Orbitolites and Alveolina facies with packstone to wackestone texture. By contrast, the dolomite is rare in Nummulites layers. Lateral variations in distribution or abundance of dolomite have been observed in the area. Different types of dolomites of the Jahrum Formation (Types I, II and III) have been controlled by sedimentary textures, as indicated by their bedding-parallel distribution.

In order to gain an acceptable level of understanding of diagenetic pathways that affected the Jahrum Formation, the limestone layers that are interbedded with - and adjacent to - dolomitized intervals were investigated. Major diagenetic processes affecting the limestones of the Jahrum Formation include: (1) micritization of bioclasts; (2) carbonate cementation (fibrous or bladed isopachous and equant calcite cements); and (3) neomorphism and dissolution of aragonitic bioclasts (e.g., bivalves and gastropods). In many cases, micrite envelopes have been fragmented prior to calcite precipitation, indicating relatively early dissolution of these grains. Under cathodoluminescence, two generations of calcite cement are observed in the drusy pore-filling spar: an early fringe of stubby, dull-orange cements lining pores and a second, blocky cement type filling the majority of the remaining pore space (Figure 8e). Blocky cement is characterized by dull-orange luminescence followed by bright orange-yellow zoning (I and II in Figure 8f). The bright orange-yellow luminescent zone is generally much thicker compared to the early dull-orange luminescent fringe phase (Figure 8f).

Carbon- and oxygen-isotope ratios of dolomite within the Jahrum Formation are shown in Table 2 and Figure 9. Oxygen-isotope ratios range from -7.8 to +0.2‰ (n = 33; mean -2.6‰), and δ¹³C values range from -1.8‰ to +2.0‰ (n = 33; mean -0.1‰). The fine crystalline dolomite phase (dolomite Type I) is characterized by values ranging from -2.7‰ to +0.2‰ for δ ¹⁸O (n = 12; mean -1.7‰) (Figure 9), whilst those of the medium to coarse crystalline dolomite phases (dolomite Types II and III) vary from -7.2‰ to -0.6‰ (n = 21; mean -3.2‰). There is no specific pattern of carbon-isotope ratios related to specific dolomite types.

Sixty limestone samples (including micrite and Nummulites skeletons) from localities adjacent to dolostones were also analyzed for their carbon- and oxygen-isotope ratios (Table 3, Figure 9) and placed in context with dolomite isotope values. Nummulites tests yield depleted δ¹⁸O values (n= 11; mean -4.0‰) and enriched δ¹³C values (n= 11; mean 2.2‰; Figure 9). Micrite yielded negative δ¹⁸O (n= 49; mean -5.3‰) and fairly positive δ¹³C values (n= 49; mean +0.4‰) (Figure 9).

Dolomite Type II and III from the Jahrum Formation (n = 4) yielded ⁸⁷Sr/⁸⁶Sr ratios of 0.7079 to 0.7086 with an average of 0.7081 (Table 2). Throughout the Eocene, the reconstructed oceanic strontium-isotope curve is invariant with ⁸⁷Sr/⁸⁶Sr ratios fluctuating between 0.7077 and 0.7079. During the latest Eocene through Oligocene times, the Sr ratio increased with one of the highest rates known in Earth’s history (about 0.000050 per Myr) and continues to increase, albeit at a lesser rate, up to the Present (McArthur et al., 2001).

Table 2 summarizes the major and trace element concentrations of all dolomites of the Jahrum Formation for overview. The mean Mg content of bulk dolomitized limestone from the Jahrum Formation rises continuously from 105,300 ppm (dolomite Type II) to 130,000 ppm (dolomite Type I; mean 121,303 ppm). Conversely, Ca abundances decrease with increasing Mg abundances from mean values of 233,620 (dolomite Type III) to mean values of 198,720 ppm (dolomite Type I; mean 228,305 ppm). Dolomites of the Jahrum Formation have low Sr (61–189 ppm), Mn (30–120 ppm) and Fe (67–991 ppm) contents. There is no correlation between Fe and Mn (Table 2, Figure 10a), with Fe contents being about seven times enriched compared to Mn elemental abundances. The Mn content in dolomites of the Jahrum Formation is considerably enriched relative to that of limestones, whilst Jahrum limestones and dolostones both exhibit low amounts of Fe (generally less than 450 ppm). Limestones yield Fe and Mn contents (n = 56; mean 148 ppm and 26 ppm, respectively) that are low compare to the adjacent dolostones (Figure 10a), whilst Sr contents (mean 241) are higher (n = 56; 140 ppm; Figures 10b, c).

Matrix replacive dolomite (dolomite Type II) and dense, interlocking dolomite (dolomite Type III) textures (Figures 6c, d and 7b), typical for the Jahrum Formation, have been documented from other Eocene locations worldwide (e.g., Gaswirth et al., 2007; Salad-Hersi, 2011; Maliva et al., 2011; El Ayyat, 2013). Dolomite preferentially nucleated within and replaced microcrystalline matrix, partially replacing adjacent fossils (e.g., impingement replacement, Lucia, 1962), and occluded, in part, open void space. Pervasive replacement of the calcitic matrix by very finely crystalline, planar-s type dolomite has been documented for numerous case examples in the literature and is most often attributed to early diagenesis or during shallow burial (Sibley and Gregg, 1987; Gregg and Shelton, 1990; Al-Aasm and Packard, 2000). Replacement of a limestone fabric by dolomite under low to moderate temperatures, below a presumed critical roughening value (Gregg and Sibley, 1984), produced generally euhedral and subhedral crystal faces resulting in idiotopic texture (e.g., Figures 5c, d and 7a, b). Additional evidence for early formation of the dolomite is that the matrix replacive dolomite clearly predates medium crystalline dolomite cement, which commonly lines the walls of vugs and smaller pores in the dolomite host rock (Figure 6a) and is interpreted as shallow burial diagenesis. Matrix replacive dolomite also occurs as mineral inclusions and predates coarse medium crystalline dolomite cement (dolomite Type IV) (Figure 8a).

The co-existence of fabric-retentive and fabric-destructive textures is common in dolomites of the Jahrum Formation and many ancient dolomites (Lee and Friedman, 1987; Machel, 2004; Zhao and Jones, 2012). Dolomites of the Jahrum Formation are dominated by finely crystalline fabric-retentive textures (Figures 5a, b) that commonly contain numerous bioclasts and peloids. The preservation of subtle depositional structures and skeletal grains represents independent evidence for an early, shallow-burial dolomitization. Some of these samples, however, also include coarsely crystalline fabric- destructive, more anhedral dolomites. Various textures and cathodoluminescence signatures of dolomites in the Jahrum Formation (e.g., variety of fabrics including polymodal distribution of dolomite crystals size and blotchy to bright orange CL patterns) indicate that dolomitization was probably a multiple-stage of recrystallization during diagenesis. Geochemical data from dolostones of the Jahrum Formation are consistent with the notion of recrystallization because there is a depletion of δ¹⁸O and Sr as would be expected from dolomite recrystallization (Mazzullo, 1992).

Several diagenetic features in the interbedded limestones (e.g., micrite envelopes, syntaxial overgrowth cement) provide evidence for near-seafloor marine and shallow-burial diagenesis (Figure 11). Evidence for this is given by the presence of micritic envelopes around aragonitic skeletal grains and early pre-compaction dissolution of aragonitic grains as indicated by the collapse of these envelopes. Furthermore, the presence of fibrous and columnar calcite cements of marine phreatic origin followed by thin to medium crusts of equant calcite cements (Figure 11) took place during burial diagenesis (Asadi Mehmandosti et al., 2013; Turpin et al., 2014).

The fine to medium crystal size of much of the dolomite, planar crystal boundaries (mostly subhedral crystals), a “cloudy-center-clear-rim” petrography and visually distinctively zoned CL patterns (Figures 5e, f) provide evidence for moderately elevated fluid temperatures in the order of 50–60°C and less (Sibley and Gregg, 1987; Warren, 2000). Furthermore, a cloudy-center-clear-rim texture is considered as typical for many sparry dolomites that precipitate from warm, but not hot, burial fluids (Warren, 2000). The cloudy center often represents pene-contemporaneous, or (very) early post-depositional, dolomite precipitation while the clear rim forms by continuous growth during later diagenesis (Machel, 2004). Alternatively, the clear rim has been interpreted as a low-temperature, early-burial syntaxial overgrowth cement (Choquette and Hiatt, 2008; Gillhaus et al., 2010).

The complex zonation in CL for Type II and III dolomite (Figures 6f, g and 7d, e) is the same in all thin sections, suggesting that all studied field areas experienced comparable dolomitization pathways. These CL patterns reflect the habit (and its changes) of the dolomite crystals and the lack of dolomite dissolution-reprecipitation features suggests that the paragenetic succession is completely present in thin sections studied. Many dolomites throughout the rock record appear red and mottled under CL and have inclusions of yellow and yellow-orange-luminescent dolomite (e.g., Cander, 1991; Choquette and Hiat, 2008). Similar features have been observed in the dense and crystalline Jahrum Type III dolomite (Figures 6f, g). Some studies have shown that the dull red luminescence observed in the dolomite (Figures 5f and 7d, e) is commonly associated with early-formed dolomite during shallow burial environment (Figure 11; Machel, 1985; Machel and Burton, 1991; Conliffe et al., 2012; Jiang et al., 2013). The zonation in CL indicates fluctuations in fluid composition and/or redox conditions. The activation of luminescence, and the high Fe (mean 450 ppm) would all suggest sub-oxic to reducing conditions. Moreover, a contribution of saline fluids ascending from numerous salt diapirs in the southeastern Zagros Basin cannot be ruled out as contributers to the dolomitizing fluid and enhanced rates of dolomitization (Ghazban and Al-Aasm, 2007, 2010). Evidence for this might include the presence of yellow-greenish dolomite luminescence spectra in the dolomite Types II and III (Figures 6f, g and 7d, e) that have previously been described to be characteristic for dolomitization under the influence of saline fluids (Gillhaus et al., 2010). Cathodoluminescence spectroscopy reveals that the dolomites of the Jahrum Formation with yellowish-green CL emission have a very high percentage of Mn²+ on the Ca lattice position which results in a visually yellowish-green CL emission (Figure 7e). Since the emission bands have high FWHM (full width at half maximum), the greenish part of the spectrum (between 520 nm and 575 nm) is also increased. Ascending saline fluids must not be confused with evaporated pore water from the inner platform, which was ruled out due to the conspicuous absence of syn-sedimentary evaporitic minerals (Zohdi et al., 2013).

Cathodoluminescence petrography of calcite cements in nearby limestones documents that bladed calcite crystals are mostly blotchy to non-luminescent suggesting precipitation in a marine-phreatic environment (Richter et al., 2003; Flügel, 2010; Figures 8e, f). Blocky calcite cements fill the remaining pore space and are characterized by CL patterns that are typically dark ochre to orange in appearance (Figures 8e, f). Following Banner et al. (1988) and Cander (1994), this complex CL pattern implies that these calcite cements precipitated from ambient fluids, that in terms of their chemical properties varied significantly with time.

The coarsely crystalline, euhedral Type III dolomite rhombs are interpreted as a product of partial recrystallization of the finely crystalline, subhedral dolomite rhombs that dominate the matrix (cf. Gregg and Shelton, 1990; Machel, 1997). This interpretation is supported by the following textural relationships: (1) coarsely crystalline rhombs are scattered throughout the finely crystalline matrix (Figures 6b, 6e and 11); and (2) the coarsely crystalline rhombs appear to overlap and cut across the finer crystals (Figures 6b, 6e and 11).

The carbon- and oxygen-isotope ratios of Type I to III dolomite are depleted relative to published Eocene dolomites precipitated directly from seawater or marine pore waters (Saller, 1984; Mansour and Holail, 2004; Maliva et al., 2011; Figure 9). The heaviest δ¹⁸O of the dolomite Type I falls within the range of calculated Eocene seawater dolomite (Figure 9), suggesting the dolomitizing fluids for dolomite Type I were probably Eocene marine pore waters. Early-formed dolomites were variably recrystallized during shallow burial as indicated by their lower δ¹⁸O, and larger crystal size. The depleted δ¹⁸O values could have originated from compaction fluids of Eocene seawater parentage during shallow burial at moderately elevated temperature. Such fluids would have δ¹⁸O values similar to seawater, or slightly heavier as a result of recrystallization of the rock matrix in pore fluids of seawater parentage (Machel and Anderson, 1989; Kirmaci, 2008; Maliva et al., 2011; Jiang et al., 2013).

Acknowledging the obvious limitations of oxygen-isotope thermometry based on bulk diagenetic mineral data, a very tentative estimate of the temperature of the diagenetic fluid from which dolomites of the Jahrum Formation precipitated is presented here. For this purpose, the Land (1985) equation was used T°C = 16.4 - 4.3 ([(δ¹⁸O_dolomite) – 3.8] - δ¹⁸O_fluid) + ([δ¹⁸O_dolomite) – 3.8] - δ¹⁸O_fluid)². The bulk δ¹⁸O value of Eocene marine waters is commonly believed to be in the order of 0‰ (V-PDB; Maliva et al., 2011). Applying the Land (1985) equation, a fluid temperature range of between 33.6°C and 85°C with a mean of 49.6ΰC results (Figure 12a). As an independent test, the Allen and Wiggins (1993) approach was used and the outcome is that a plot of δ¹⁸O and δ¹³C values of dolomites of the Jahrum Formation agrees with dolomite precipitation within a moderately elevated temperature domain. Pulses of warm to hot burial fluids (85°C), probably related to increasing burial depths, lead to mainly coarse-grained dolomite phases (Figure 12b).

The δ¹⁸O values of dolomite Type III are similar to all other dolomite types, except for three samples (Figure 9), which are more depleted than the typical range of Eocene values reported for marine dolomite. The more negative dolomite oxygen-isotopic composition is interpreted to result from the progressive recrystallization of dolomite at higher temperatures during burial. This suggests that an increase in the degree of alteration of the original dolomite (and thus crystal size) is concomitant with a decrease in δ¹⁸O values (Figure 9). A similar correlation is found in Eocene dolomites of the west-central Florida, USA, where increasing dolomite crystal size with decreasing δ¹⁸O values is also attributed to progressive chemical stabilization of dolomite crystals during shallow burial (Gaswirth et al., 2007).

The δ¹³C ratios of dolomites analyzed are confined to a narrow range that is equivalent or only slightly more negative than Eocene seawater δ¹³C_DIC ratios. The carbon in dolomite and limestone samples is therefore derived either from Eocene seawater or from fluids that were in equilibrium with Eocene seawater. Since the δ¹³C composition of the dolomite samples remains relatively constant, alteration can be assumed to have taken place in a system where carbon was buffered by the host rocks.

Under normal circumstances, diagenesis under the influence of meteoric fluids (Matthews and Allen, 1977; Allen and Matthews, 1982) results in the formation of carbonates with significantly depleted δ¹³C values beneath exposure surfaces. The relatively high δ¹³C values of all the dolomites may suggest that the rocks experienced alteration in the absence of soil zone CO₂-derived, isotopically depleted carbon. Alternatively, subaerial exposure under generally arid conditions may results in less depleted carbon isotope values (see discussion in Christ et al., 2012). The impact of subaerial exposure on these carbonate successions is probably not sufficiently well understood, although evidence for long-term subaerial exposure at the hiatal top of the Jahrum Formation, overlain by terrestrial clastic material of the Oligocene to Miocene Razak Formation, was observed (Figure 2). In all studied outcrops, Upper Eocene strata and much of the Oligocene successions are missing, likely due to a eustatic sea-level fall (Zohdi et al., 2013) at the Eocene/Oligocene boundary as proposed by previous workers (Haq et al., 1987; Zachos et al., 2001; Sharland et al., 2001). Evidence for ancient meteoric alteration is found in soil carbonate features (calcretes) found beneath exposure surfaces.

The Sr contents of the dolomites of the Jahrum Formation vary from 61 ppm to 189 ppm, but the vast majority of samples plot around values of 100 ppm (Figures 10b, c). The lower Sr concentration in dolostones relative to limestones (mean 400 ppm) is mainly due to the lower Sr partitioning in dolomite (Figures 10b, c) (Land, 1980; Veizer, 1983). The Sr contents of the dolomites of the Jahrum Formation do not appear to vary with crystal size, suggesting an overall stable diagenetic environment. Strontium concentrations in the different types of dolomites of the Jahrum Formation are relatively low compared to that of Paleogene and Holocene marine and evaporative dolomites (100 ppm to 1,000 ppm; Al-Aasm and Veizer, 1982). Dolomites with mean Sr values of less than 300 ppm are generally interpreted as having formed in either a water-buffered diagenetic system or a rock-dominated system in which the precursor phase contained little Sr (Budd, 1997). In the case of the dolomites of the Jahrum Formation, the Sr concentration is thus considered to be compatible with, but not necessarily indicative of, a marine origin. Nevertheless, Zhao and Jones (2012) pointed out that the Sr content of dolostones from the Cayman Formation, with a range of 80–278 ppm is consistent with dolomitization by seawater-like fluids. It should be also be mentioned that Sr-depleted dolomite may occasionally form as a primary phase, although this is not common (Vahrenkamp and Swart, 1990; Mazzullo, 1992).

Vahrenkamp and Swart (1994), Kyser et al. (2002) and Azmy et al. (2009) suggested that low Sr contents of 70–250 ppm, particularly when associated with depleted δ¹⁸O ratios, are indicative of mixing zone dolomitization. Nevertheless, average δ¹³C values of Sr-depleted dolomites of the Jahrum Formation are similar to estimated δ¹³C ratios of Eocene marine carbonate. This suggests that the fluids responsible for alteration or precipitation of Sr-depleted dolomites of the Jahrum Formation were buffered by Eocene marine carbonate.

Fe and Mn concentrations in dolomite can impart information about the redox state of the dolomitizing fluids and/or the availability of Fe and Mn (Budd, 1997). The incorporation of Fe or Mn into the dolomite lattice is favored because their distribution coefficients are greater than unity (Veizer, 1983). Manganese (30 ppm to 120 ppm) and Fe (67 ppm to 991 ppm) concentration of dolomites of the Jahrum Formation are similar to Mn (5 ppm to 275 ppm) and Fe concentrations (10 ppm to 2,000 ppm) of Paleogene and Holocene marine and evaporative dolomites (Al-Aasm and Veizer, 1982; Gregg et al., 1992), which are compatible with a seawater origin of these dolomites.

Manganese and Fe concentrations of dolomites of the Jahrum Formation (58 ppm and 450 ppm, respectively) are slightly greater than those of the Jahrum limestones (27 ppm and 221 ppm, respectively; Figure 10a), suggesting that either the pore waters were sub-oxic to reducing, or that the pore waters were reducing but the diagenetic environment lacked significant sources of these elements (Budd, 1997).

Early marine-diagenetic, low-temperature dolomites (e.g., sabkha type) are commonly poorly ordered (Geske et al., 2012), whereas most burial dolomites are more stoichiometric and hence better ordered. It is reasonable to assume that most high-temperature dolomites have undergone some degree of textural and crystallographic modifications leading to stoichiometric and ordered phases (McKenzie, 1981; Warren, 2000; Machel, 2004; Kaczmarek and Sibley, 2011). Shallow burial dolomites of the Jahrum Formation reveal, on average, a higher degree of stoichiometry than dolomites reported from many other burial settings (Land, 1973; Ward and Halley, 1985). The stoichiometry of dolomites of the Jahrum Formation is probably primary or earliest diagenetic in origin as petrographic observations suggest that dolomites of the Jahrum Formation escaped deep burial diagenesis. Where present, mild degrees of alteration occurred in the shallow burial zone where marine pore waters, mixing with thermal ascending saline fluids interacted with dolomites. It cannot be excluded that salt diapirs contributed saline ascending fluids and geochemically affected the dolomitization fluids. A possible, albeit circumstantial, line of evidence for ascending diagenetic fluids with moderately elevated salinities is found in the presence of yellow-greenish luminescence patterns with the highest percentage of Mn²+ in the Ca position of the crystal lattice (Figures 6 and 7). Direct evidence (evaporative minerals) for saline fluids during limestone deposition is lacking. Saline waters provide high Mg²⁺/Ca²⁺ ratios and at low crystallization rates, dolomites of the Jahrum Formation tend to be initially more stoichiometric. Nevertheless, it must be noted that dolomite samples from lower portions of the Jahrum Formation, formerly exposed to somewhat greater burial depths, indicate a better lattice ordering (0.94) and more stoichiometric composition (50 mole% CaCO₃) relative to dolostones from the upper, shallower burial Jahrum Formation (52 to 55 mole% CaCO₃). Hence, burial diagenesis did affect these dolomites to some degree.

We have documented several lines of evidence (i.e., circumstantial evidence such as fluid mean temperatures of 50°C as well as direct geochemical evidence) all pointing to an early diagenetic origin of the vast majority of Jahrum Formation ramp limestone dolomitization. Moreover, calcitic remnants of the proximal ramp depositional facies of the Jahrum Formation are indicative of an overall shallow-marine setting sensitive to even low-amplitude sea-level change. Below, we discuss these findings in the context of previous work with focus on different dolomitization models.

Initial dolomitization of the Jahrum Formation likely occurred in the marine realm by a moderately increased salinity (72‰ to 199‰; penesaline dolomitization of Simms, 1984; Qing et al., 2001) that were driven by a reflux mechanism (Figure 13). The lack of normal marine biota, where still recognizable, within dolomitized portions of the Jahrum Formation, and the common occurrence of shallow-water marine biota (green algae, gastropods and agglutinated benthic foraminifera; Figures 14a–f), suggest waters of variable or elevated salinity during the deposition of the previous dolomitized part of the Jahrum Formation (Figure 13). However, the absence of massive gypsum/anhydrite and/or evaporite solution collapse breccias suggests that evaporated seawater did not reach the salinity required for abundant gypsum precipitation. The relatively dense, penesaline seawater penetrated the underlying carbonate sediments by displacing marine pore water of lower density. During the passage of Mg-rich seawater along the regional slope down-dip, thermodynamically less stable CaCO₃ minerals were replaced by dolomite.

The most commonly invoked and best-documented models for the early dolomitization of large areas of shallow platforms and ramps are evaporative pumping in the reflux of mesosaline to hypersaline seawaters. Examples for this process have been documented for ancient carbonates, whereby reflux of mesosaline to penesaline seawater through shallow carbonate strata is driven by brine density and/or sea-level fluctuations (e.g., Qing et al., 2001; Harvey et al., 2004; Aqrawi et al., 2006; Vandeginste et al., 2009; Al-Helal et al., 2012; Rivers et al., 2012; Iannace et al., 2014). Moderately modified seawater as a dolomitizing agent can account for extensive dolomitization without concurrent precipitation of sulfates, a notion that is consistent with the conspicuous lack of depositional sulfate deposits in the Jahrum Formation.

The hydrological system and climate settings responsible for dolomitization of the Jahrum Formation are here considered to share many important similarities with the reflux model (Figure 13; Adams and Rhodes, 1960; McKenzie et al., 1980) including slightly elevated marine (penesaline) salinity levels. Fieldwork as well as petrographic and isotopic data shown here all point to the reflux model for initial dolomitization. Dolomites of the Jahrum Formation form spatially extensive rock bodies in platform-top environments. These dolostones clearly show meter-thick beds of stratabound dolomite and shallowing-upward cycles of restricted platform limestones capped by mixed carbonate-terrigenous siliciclastic shallow-marine to continental deposits of the Razak Formation. On the other hand, most of the dolomites of the Jahrum Formation consist of subhedral to euhedral crystals of which most are in the tens-of-microns range and the δ¹³C of dolomites of the Jahrum Formation fall within the range of expected values for Eocene seawater (Figure 9). These observations suggest that moderately altered seawater was the fluid responsible for dolomitization.

Reflux dolomitization is often ascribed to periods of sea-level fall under arid climate settings (Purser et al., 1994; Bosence et al., 2000) as brines are driven into limestone host rocks. Reflux of penesaline seawater flow, driven by sea-level fall leads to regionally important fluid flow patterns that affect areas as large as thousands of square kilometers in ancient carbonate platforms combined with climatic aridity. The combination of these factors results in formation of regionally important, massive dolomite bodies. Climatic aridity has been proposed for the Eocene of the northern Arabian Plate (e.g., Yemen, Saudi Arabia and Kuwait; Ziegler, 2001). Evidence for this is found in the abundance of primary evaporitic layers interbedded by dolostones formed in a shallow-marine evaporitic mudflat environment (Sharland et al., 2001). Along similar lines, Zohdi et al. (2013) suggested that the overall sequence-stratigraphic pattern of the Jahrum Formation carbonates is perhaps best interpreted as a large-scale regressive unit. Arguments include the stratigraphically underlying distal marine shale and marl deposits of the Gurpi Formation and the overlying mixed clastic-carbonate Razak Formation.

Matrix-replacive dolomite in the Jahrum Formation is petrographically similar to dolomite described from other Eocene intervals in the northeastern Arabian Plate, including the Rus Formation (Sharland et al., 2001). It is possible that dolomite in other Eocene strata in the northeastern Arabian Plate have a similar origin via shallow seepage reflux by evaporative brines and also underwent partial recrystallization during burial.

The majority of dolomites formed from evaporitic seawater are, at precipitation, disordered, non-stoichiometric and thermodynamically unstable. Recrystallization of early formed dolomite is believed to be a common process during progressive burial (e.g. Gregg and Shelton, 1990; Durocher and Al-Aasm, 1997).

Reconstructions of the burial history of the study area suggest that the Jahrum Formation was initially buried to a maximum depth of about 1,000 m beneath Neogene sediments before uplift and exhumation related to the Zagros orogeny took place (Kamali and Rezaee, 2003; Zamanzadeh et al., 2009). Applying a geothermal gradient of 30°C/km and assuming a fluid type and isotopic composition similar to that of seawater, the expected burial temperature is sufficient to explain recrystallization of matrix-replacive dolomite.

In agreement with the above considerations, dolomites of the Jahrum Formation are interpreted to have undergone recrystallization during shallow burial, based on a set of petrographic and geochemical evidence: (1) Dolomites of the Jahrum Formation display a variety of fabrics including polymodal distribution of dolomite crystals size, nonplanar texture and highly irregular or weakly defined intercrystalline boundaries. (2) The increase of average crystal size coincides with increasing in abundance of nonplanar crystal boundaries (Figure 7c). These textural trends suggest that planar crystal boundaries likely developed during initial dolomitization under near-surface conditions and nonplanar boundaries and increased crystal diameters formed as a result of recrystallization of early planar dolomite. (3) Dolomites of the Jahrum Formation are generally characterized by blotchy to bright orange CL patterns (Figures 7d and 8d) suggesting recrystallization during shallow burial diagenesis (Banner et al., 1988; Qing, 1998; Maliva et al., 2011). (4) The near-stoichiometric composition (50 to 52 mole% CaCO₃) and low Sr concentrations (mean 100 ppm) of the dolomites of the Jahrum Formation are interpreted to have resulted from recrystallization during shallow burial. Holocene sedimentary dolomites typically are nonstoichiometric (up to 62 mole% CaCO₃) and poorly ordered (McKenzie, 1981; Mazzullo et al., 1987; Aqrawi, 1995; Last et al., 2012; Turpin et al., 2012). (5) Slightly higher than Eocene seawater ⁸⁶Sr/⁸⁷Sr ratios of dolomites of the Jahrum Formation suggest that the matrix-replacive dolomite underwent variable recrystallization during burial. Finally, (6) δ¹⁸O ratios of dolomites of the Jahrum Formation are depleted relative to Eocene dolomites precipitated directly from seawater or marine pore waters. At first glance, the δ¹⁸O depleted dolomites of the Jahrum Formation might be explained by mixing zone diagenesis (Badiozamani, 1973; Humphrey and Quinn, 1989). The meteoric-marine mixing zone model, however, seems unlikely in the case of dolomites of the Jahrum Formation because there is no petrographical evidence and a dominant meteoric soil-zone CO₂ signal in the carbon-isotope record to suggest that the dolomites of the Jahrum Formation have undergone meteoric diagenesis. The rather depleted δ¹⁸O values of dolomites of the Jahrum Formation (Figure 9) are interpreted to reflect recrystallization at elevated temperatures during burial, and in the presence of connate evaporitic seawater.

An additional factor potentially affecting Jahrum dolomitization is the increased heat flow associated with salt domes (Figure 13). During deposition of the Jahrum Formation, salt diapirs moved up into the zone of shallow-marine carbonate production (Jahani et al., 2009; Zohdi et al., 2013). Therefore, the stratigraphy, platform development and diagenetic history of the Eocene ramp deposits in the southeast Zagros Basin is at least affected if not controlled by salt diapirism. The uniformity of the trace elemental abundances and carbon isotope ratios suggest that the dolomites of the Jahrum Formation probably crystallized from a single parent fluid in shallow burial environment being affected by the rising salt bodies. An association between salt domes and the circulation of saline fluids has been noted by several authors (Hovland et al., 2006; Lambert et al., 2006; Ghazban and Al-Aasm, 2010).

Salt movement is typically associated with faulting, forming a possible mechanism for the access of saline fluids into the sedimentary column of the Jahrum Formation. Evidence for this might include the presence of yellowish-green dolomite luminescence (Figures 6f, g and 7d, e) with an asymmetrical luminescence-spectra that contains two peak positions as a result of high percentage of Mn²⁺ in Ca positions of the dolomite Types II and III. Similar patterns have been described as a characteristic feature for dolomitization under the influence of saline fluids (e.g., Richter et al., 2003; Gillhaus and Richter, 2001; Gillhaus et al., 2010).

Considering these arguments, it seems likely, that the Jahrum Formation dolomitization was affected by salt movement. Fluids might have reached higher temperatures (85°C) as indicated in Figure 12 for Type III dolomites. Salt bodies conduct heat more rapidly than other sedimentary rocks and deep-seated salt structures conduct heat from depth (O’Brien and Lerche, 1987). This hypothesis shares similarities with that presented by Beavington-Penney et al. (2008), who concluded that dolomitization in offshore Tunisia - above a salt diapir - resulted from circulation of diapir-derived, hot fluids.

Summing up, a more likely scenario, and the interpretation presented here, is that fluids responsible for the initial dolomitization of the Jahrum Formation were penesaline Eocene seawater, that via reflux passed through the calcitic ramp deposits during sea-level fall and under arid climate settings. Evidence for this comes from the low-temperature fluids that characterize the vast majority of dolomites of the Jahrum Formation.

The ⁸⁷Sr/⁸⁶Sr ratios of dolomites of the Jahrum Formation range from 0.70796 to 0.70863 (Figure 15). It is possible that initial radiogenic Sr isotope ratios might have been affected by Sr from limestone precursors, detrital input, hydrothermal fluids, volcanic deposits and/or groundwater redistributing Sr (Jones and Luth, 2003). Moreover, in geochemically open systems, Sr abundances and ⁸⁷Sr/⁸⁶Sr ratios are often found to deviate from the values at deposition during diagenetic alteration. In contrast to this, Vahrenkamp et al. (1988) argued that the influence of precursor limestones on the ⁸⁷Sr/⁸⁶Sr of dolostones depends on the mineralogy of these precursors. According to modelling data, aragonitic precursor mineralogies (with about 7,000 ppm Sr), would have resulted in ⁸⁷Sr/⁸⁶Sr ratios of dolomite that to a significant degree reflect their precursor lithology. The mainly high-Mg calcite precursor mineralogies (with about 400 ppm Sr) suggests that the ⁸⁷Sr/⁸⁶Sr of dolomites of the Jahrum Formation reflects predominantly the chemistry of the dolomitizing fluid (Vahrenkamp et al., 1988).

In the case of the precursor original mineralogy of Jahrum Formation, calcite arguably formed the dominant mineral of the carbonates deposited (Zohdi et al., 2011). Petrographic features indicate that most, if not all, of the limited amount of mainly biogenic aragonite, which was potentially present at deposition, was altered to secondary low-Mg calcite prior to dolomitization. Based on petrographic evidence, syn-depositional diagenetic alteration took place under the influence of marine pore waters. Having considered the primary mineralogy of Jahrum precursor carbonate, alternative sources of Sr must be discussed that might have affected their geochemistry during diagenesis.

Volcanic successions or significant stratigraphic units of mature siliciclastic sandstones are unknown in the southeastern Zagros Basin. Therefore, an open-system radiogenic or hydrothermal source of ⁸⁷Sr/⁸⁶Sr ratios affecting Jahrum Formation dolostones seems unlikely. Acknowledging the limitations of bulk dolomite Sr chemostratigraphy, it is here tentatively proposed that ⁸⁷Sr/⁸⁶Sr ratios of mono-mineralic dolomites of the Jahrum Formation might represent approximate time markers for the dolomitization stage. Evidence for this also comes from the Sr elemental abundances in the dolomites of the Jahrum Formation (61–189 ppm, mean 100 ppm; Table 2). This value falls in the expected Sr range (< 300 ppm) of dolomites that precipitated from seawater or evaporated seawater assuming a seawater Sr source (Vahrenkamp and Swart, 1990; Banner, 1995; Suzuki et al., 2006; Zhao and Jones, 2012).

The reconstructed ⁸⁷Sr/⁸⁶Sr values for Cenozoic seawater (McArthur et al., 2001) are shown in Figure 15. The ⁸⁷Sr/⁸⁶Sr composition of seawater has risen from the Middle Eocene to the Quaternary (McArthur et al., 2001). The calculated ⁸⁷Sr/⁸⁶Sr value of 0.70796-0.70863 for dolomites of the Jahrum Formation tentatively indicates a post-Middle Eocene Sr isotope age. The ⁸⁷Sr/⁸⁶Sr values of dolomites of the Jahrum Formation partially overlap with the range of estimated Eocene values, whilst most samples are slightly radiogenic with respect to estimated Eocene values. Slightly radiogenic Sr-isotope signatures of the dolomites of the Jahrum Formation are probably related to recrystallization in the pore waters with somewhat elevated ⁸⁷Sr/⁸⁶Sr ratios. Burial fluids carrying radiogenic Sr might have modified the original isotopic signature of the early-formed dolomite during recrystallization at depth. Enrichment in ⁸⁷Sr is common in recrystallized burial dolomites relative to their unaltered marine equivalents (Mazzullo, 1992; Kupecz et al., 1993; Fu and Qing, 2011; Haeri-Ardakani et al., 2013), particularly in Cenozoic dolostones of the Zagros Basin (Ehrenberg et al., 2007; Moallemi, 2009). Although the dolomites of the Jahrum Formation precipitated during and after the Late Eocene, the duration of (one to several) dolomitization cycles is difficult to constrain. An obvious upper age limit is given by the carbonate-terrigenous shallow-marine to continental deposits of the Lower Miocene Razak Formation (Figure 2). Concluding, it appears that cumulative Jahrum dolomitization might have taken place at perhaps slow rates but over a rather expanded time interval of roughly 15 Myr (i.e., latest Eocene to Early Miocene; Figure 15).

Platform-scale dolomitization occurred during shallow burial of the Eocene Jahrum Formation in the southeastern Zagros Basin of Iran. During sea-level fall under arid climate settings, the Jahrum Formation (Eocene) carbonates were within platform-top, restricted settings characterized by moderately elevated salinities, providing favorable conditions for dolomitization.

Four specific dolomite types are characterized as based on petrographic evidence: Type I includes idiotopic, isolated rhombs or subhedral to anhedral, fine dolomite crystals that mimetically replaced limestones with locally well-preserved skeletal remnants and sedimentary features. Type II dolomites, volumetrically the most significant type, consist of anhedral to subhedral crystals, ranging from 20 to 100 μm in size and forming calcareous dolostones characterized by a former lime mudstone matrix with skeletal and non-skeletal grains now largely replaced by dolomite. Type III pervasively dolomitized limestone facies texturally ranges from a dense, interlocking, non-sucrosic texture with medium to coarse crystalline dolomite crystals, to a medium-crystalline dolostone with abundant biomouldic porosity. Type IV includes euhedral to subhedral crystals occluding secondary pores formed due to the leaching of bioclasts and are volumetrically insignificant. Petrographic evidence such as the preservation of subtle depositional structures and skeletal grains is in agreement with an early, shallow burial dolomitization from fluids with mean temperatures of 50–60°C or less.

Assuming predominantly closed system diagenetic behavior, the bulk dolomite ⁸⁷Sr/⁸⁶Sr data suggest that the dolomites of the Jahrum Formation predominantly contain Sr that was derived from a seawater source of post-Middle Eocene age.

Type I and II dolomites are interpreted as the product of near-surface dolomitization of mainly calcitic lithologies by seawater of penehaline to mesohaline salinity during periods of sea-level lowstand (reflux model). This interpretation is supported by the widespread, facies-independent distribution of the dolomite and the general absence of syn-sedimentary evaporite deposits in the studied Jahrum outcrops. The influence of ascending saline fluids from underlying salt diapirs seems likely. Circumstantial evidence for a saline fluid component is arguably found in the yellow luminescence spectra. Variations in dolomite crystal size, elevated ⁸⁷Sr/⁸⁶S ratios and depleted δ¹⁸O compositions, relative to the expected range of Eocene seawater derived dolomite, suggest that the dolomites of the Jahrum Formation underwent recrystallization during shallow burial. Again, the influence of ascending, diapir-related fluids seems likely. Moreover, Type III dolomites are indicative of higher fluid temperatures (85°C) typical of diapir-related fluids.

The data shown here are of relevance as process-oriented descriptions of Eocene dolostones from the southeastern Zagros Basin of Iran are as yet lacking. Moreover, the Jahrum Formation dolostone case example represents one of the not-so-frequent case examples of Eocene ramp deposits worldwide.

The authors would like to thank the Iranian Central Oil Fields Company (ICOFC), the Ferdowsi University of Mashhad, Iran and the Ruhr-University, Bochum, Germany who provided financial support for this research. We are grateful to the Iranian Central Oil Fields Company for permission to publish this paper. We thank Dr. A. Niedermayr, Bochum, who ran the ICP-OES and isotope analyses and Dr. R.D. Neuser, Bochum, who provided the cathodoluminescence photomicrographs used in this study. This manuscript has benefitted from constructive comments by GeoArabia reviewers. GeoArabia’s Assistant Editor Kathy Breining is thanked for proofreading the manuscript, and GeoArabia’s Production Co-manager, Arnold Egdane, for designing the paper for press.

Afshin Zohdi is presently a PhD student at the Ferdowsi University of Mashhad, Iran, and a Visiting Scientist at the Ruhr University Bochum, Germany. He won the first prize in the open PhD entrance exam at University of Mashhad, Iran, in 2008. He holds a BSc (Honors) in Geology from Damghan University, Iran, in 2004 as well as his MSc (top) in Geology (carbonate sedimentary rocks) in 2007 from Shahid Beheshti University in Tehran. In 2011, he received a grant from the International Association of Sedimentology (IAS) in order to attend at the 14th Bathurst Meeting of Carbonate Sedimentology in Bristol, England. His research interests include carbonate sedimentology and stratigraphy of the Paleogene, sequence stratigraphy, carbonate petrography, diagenesis and dolomitization. Afshin’s publications include studies on the sedimentary environment, sequence stratigraphy and diagenesis of the Early Paleogene successions in the Zagros and Alborz basins. He is an active member of Geological Society of Iran, Sedimentological Society of Iran and IAS.

[email protected]

Seyed Ali Moallemi is Assistant Professor and Deputy Dean of Faculty Research and Development in Upstream Petroleum Industry Research Institute of Petroleum Industry (RIPI), Iran. Ali completed his PhD in Sedimentary Geology at the Shahid Beheshti University, Tehran, Iran. He has worked for more than 21 years in research, industry and consulting in Iran. He received his BSc and MSc in Geology from the University of Zahedan, Iran, in 1988 and Islamic Azad University, Iran, in 1984, respectively. Ali has authored and co-authored 25 research papers published in national and international journals. His research interests are petroleum geology, sequence stratigraphy, carbonate diageneses and reservoir geology. He is an active member of AAPG, EAGE, and Geological Society of Iran.

[email protected]

Reza Moussavi-Harami is Professor of Geology in the Department of Geology, Faculty of Sciences, Ferdowsi University of Mashhad, Iran. He received his BSc in Geology from Mashhad University (1972), MSc from the University of Oklahoma, USA (1977) and PhD from the University of Iowa, USA (1980). From 1975 through 1980, he consulted for the Oklahoma and Kansas Geological Surveys. From 1980 through 1983, he was a Senior Sedimentologist with the Exploration Directorate, NIOC, Tehran, Iran. From 1994–1997, he was a Visiting Professor to the Department of Geology, University of Adelaide and consultant to the Oil, Gas and Coal Division of Mines and Energy South Australia. From 1998–2000, he was President of the Geological Society of Iran. From 1999–2001, he served as a Chairman of Geology Department at the Ferdowsi University of Mashhad, Iran. From 2000–2002, he was an Adjunct Professor to the Department of Geosciences at the University of Iowa, USA. He is consultant to RIPI, Tehran. From 2011 to present he is President of the Sedimentological Society of Iran. His main research involves basin analysis, sequence stratigraphy and the reconstruction of paleoenvironment in relation to oil and gas exploration and production. He is a member of AAPG, SEPM and EAGE.

[email protected]

Asadollah Mahboubi is a Professor at the Geology Department of Ferdowsi University of Mashhad in Iran since about 25 years. He received his BSc in Geology from Ferdowsi University of Mashhad in 1984, and his MSc and PhD in Geology (sedimentology and sedimentary petrology) in 1991 and 2000 from Kharazmi University in Tehran, Iran, respectively. He was previously a researcher at Research Center of National Iranian Oil Company (NIOC) in Tehran for two years. His main research interests include carbonate petrology, depositional environments and sequence stratigraphy. He has published more than 70 papers in journals and presented many papers in congresses.

[email protected]

Detlev K. Richter obtained his Geology diploma from the Technical University of Munich, Germany in 1967 working on stratigraphy and sedimentary petrology of alpine Mesozoic sequences. In 1970 Detlev was awarded by the University Bochum/Germany following studies on dolomite rocks in the Eifel mountains. After one year at the University of Berlin working on sedimentology aspects of the geology of Greece (rift phases in Triassic sequences, Quaternary marine terraces), Detlev returned to the Ruhr – University Bochum. His habitation thesis in 1984 related to the compositions and diagenesis of natural Mg calcites. Special interests are cathodoluminescence of carbonates, genesis of breccias, polyphase dolomitization and paleoclimate research of speleothems. Currently Detlev is working in a team with Adrian Immenhauser in the DFG founded project CHARON (Marine Carbonate Archives: Controls on carbonate precipitation and pathways of diagenetic alteration).

[email protected]

Anna Geske is a Geologist and Sedimentologist and obtained her MSc in Geology in 2010 from the Ruhr-University Bochum, Germany. She is currently a PhD student at the Ruhr-University Bochum and has over three years of experience in research in the field of isotope geochemistry of dolomites. Anna’s research interests include the magnesium isotope system as novel proxy for paleoenvironmental reconstructions in dolomites, carbonate sedimentology, petrography, diagenesis, biomineralization and geochemistry.

[email protected]

Abbas A. Nickandish is a Senior Geologist at the National Iranian Oil Company (NIOC), Iranian Central Oil Fields Company (ICOFC) with 15 years’ experience in surface and subsurface geology. He received his BSc in Geology from Shahid Beheshti University of Tehran (1995) and MSc in Hydrogeology from the Shiraz University (1998), Iran. From 1998 through 2003, he was a surface Geologist for the Jahad-e Keshavarzi Organization of Fars province, Shiraz, Iran, in the watershed management division. From 2003 to present, he is a subsurface Petroleum Geologist and oil and gas Reservoirs Geomodeller with NIOC. His main interests are 3-D geological modeling, reservoir characterization and petroleum hydrogeology.

[email protected]

Adrian Immenhauser is a Full Professor for Sediment and Isotope Geology at the Ruhr-University Bochum, Germany. He holds a PhD in Geology from the University of Berne, Switzerland. Adrian has a longstanding interest in the geology of Oman that goes back to 1990 when he commenced fieldwork as a PhD and later as a post doc on Masirah Island and in northeast Oman. Subsequently, Adrian spent 10 years as Assistant Professor at the Vrije Universiteit Amsterdam, The Netherlands, focusing on carbonate diagenesis. Presently, his research team applies conventional and nonconventional isotope systems and other geochemical and optical tools to various carbonate materials in order to resolve the diagenetic history of reservoir units. A special focus is on the evolution and impact of discontinuity surfaces in shoalwater settings that may represent flow conduits or seals resulting in reservoir compartmentalization.

[email protected]