Introduction

The growing global demand for oil, natural gas, and other sources of energy has moved exploration to challenging and complex provinces. The East Mediterranean region is one of those frontier provinces. Recent gas discoveries in Oligocene and Miocene sandstone reservoirs offshore Israel and Cyprus (e.g. Zohr, Tamar, Dalit, Leviathan, Karish, Tanin, Dolphin, and Cyprus-A) (www.nobleenergyinc.com; Esestime et al. 2016) have proven the hydrocarbon potential of the Levant Basin. The source of this gas is still unclear, although it has been reported to be of biogenic origin comprising 99 % methane (Needham et al. 2013). Oil discoveries in Mesozoic structures off the coast of Israel were also recorded in drilling campaigns during the 1990s (Gardosh 2013). Additionally, new seismic data acquired in the last decade revealed the large thickness of the sedimentary column and the presence of direct hydrocarbon indicators (DHI), thus suggesting the presence of working thermogenic petroleum systems rendering the under-explored Levant Basin one of the most promising hydrocarbon provinces in the region.

Onshore, the Palmyra Basin in Syria produces oil and gas from Early Cretaceous and Middle Triassic reservoirs, respectively (Nader 2014). Gas production is also ongoing from Carboniferous sandstone (Markada Formation) in several fields onshore Syria (Lučić et al. 2003). In Lebanon, seven exploration wells have been drilled between 1947 and 1966 penetrating only down to the Upper Jurassic which is the oldest surface-exposed rock formation. Drilling failed to encounter commercial oil and gas volumes in Lebanon, probably due to (i) meteoric water washing affecting the Jurassic and post-Jurassic rock succession and (ii) off structure position of drilled wells (Nader 2014). The pre-Jurassic succession might, however, include a working petroleum system similar to the Triassic petroleum system in Syria and sealed by the Kurrachine evaporites presumed to be present in Lebanon (Beydoun and Habib 1995; Brew et al. 2001; Nader 2014). These sediments are thought to be the extension of the Upper Triassic evaporites known in the Palmyra Basin (Renouard 1955; Beydoun and Habib 1995; Brew et al. 2001; Nader 2014).

The East Mediterranean offshore, onshore, and intermediate margin seem to include several promising petroleum systems. The thermal history of the region, as well as the distribution, quality, and maturity of major source rocks, is yet unknown. Thus, we constructed a 3D thermal history model (TemisFlow™, v. 2013.2) covering an area of 315 × 315 km (grid resolution 5 × 5 km) including the Levant Basin, margin, and onshore. New source rocks’ kinetic data were produced and used in the model in order to assess the maturity and the timing of hydrocarbon generation of major source rocks and to discuss the implications on potential thermogenic petroleum systems in the study area.

Geological setting

An extensional rift phase along the northern margin of Gondwana started in the late Paleozoic and resulted in the formation of the Levant and the Palmyra basins (Fig. 1) (Garfunkel 1989; Hawie et al. 2013a; Montadert et al. 2014; Nader 2014). Successive rift pulses affected the Levant Basin until the Late Jurassic (Gardosh et al. 2010; Hawie et al. 2013a). A calm post-rift cooling and subsidence phase prevailed until the Late Cretaceous (Hawie et al. 2013a). In the Palmyra Basin, subsidence started in the Late Permian and continued till the Late Cretaceous (Ponikarov 1966; Chaimov et al. 1992). During the Late Cretaceous, the closure of the Neotethys Ocean and the collision of the Afro-Arabian and Eurasian plates started. The compressional regime persisted until the Miocene leading to the emplacement of the Syrian Arc fold belt and the inversion of Mesozoic extensional structures in the Palmyra Basin (Ponikarov 1966; Chaimov et al. 1992; Robertson 1998; Walley 1998). The two major phases of Syrian Arc folding occurred in the Coniacian (Late Cretaceous) and Late Eocene (Garfunkel 1998; Walley 1998). The initiation of the Red Sea rifting during the Oligocene/Miocene resulted in the propagation of a series of strike-slip faults northward forming the Levant Fracture System which extends from the Gulf of Aqaba to the Taurus Mountains (Beydoun 1999). The Levant Fracture System is a sinistral fault system that includes the N–S striking Dead Sea segment, the NNE–SSW central segment (Yammouneh Fault and fault splays) forming a restraining bend in Lebanon, and the N–S Ghab Fault segment (Ghalayini et al. 2014).

Fig. 1
figure 1

Regional map of the east Mediterranean showing major structural elements. Green square marks the modelled area. PS1–PS8 are pseudo wells created for calibration, yellow pseudo wells have vitrinite reflectance data, and green pseudo wells have temperature data. Orange circle marks the position of 1D extract reported in Figs. 11 and 12. (Modified after Hawie et al. 2013a; Bou Daher et al. 2014; Ghalayini et al. 2014)

Full size image

Lithostratigraphic framework

The onshore Jurassic–Quaternary stratigraphy was constrained based on field investigations and biostratigraphic studies (Müller et al. 2010; Hawie et al. 2013b; Nader 2014; Bou Daher et al. 2015). The offshore Jurassic–Quaternary stratigraphy was postulated mostly based on seismic interpretation (Hawie et al. 2013a). The pre-Jurassic stratigraphy was inferred from regional correlations based on published data from neighbouring countries (Brew et al. 2001; Nader 2003; Gardosh et al. 2008; Naylor et al. 2013).

The Mesozoic lithostratigraphic succession of the Levant Basin and that of onshore Lebanon realm are relatively similar and mostly dominated by carbonates (Fig. 2) (Nader 2014). In the late Cretaceous, the Afro-Arabian and Eurasian plate convergence and the emplacement of a flexural basin resulted in differential subsidence and deposition of hemipelagic/pelagic and clastic Neogene basinal fill while the marginal realms remained predominated by carbonate platforms (Fig. 2) (Hawie et al. 2013a&b). Sedimentation in the basin was almost uninterrupted while in the margin and onshore several erosion events are recorded (Müller et al. 2010; Hawie et al. 2013b). The latest ongoing erosion event is related to the uplift of Mount Lebanon and Anti-Lebanon, reaching its acme during the middle/late Miocene as a result of a transpressive regime at the Lebanese segment of the Levant Fracture System (Beydoun 1999; Gomez et al. 2006, Hawie et al. 2013a). This event led to the exposure of the Jurassic cores of the Lebanese mountains while deposition continued in the adjacent topographic lows, e.g. Bekaa Valley and the coastal areas (Fig. 1) (Khair et al. 1997; Hawie et al. 2013a; Nader 2014). The erosional thicknesses were estimated based on regional correlations and on maturity data when available (Fig. 2) (Brew et al. 2001; Nader 2003; Gardosh et al. 2008; Naylor et al. 2013).

Fig. 2
figure 2

Stratigraphic chart showing the onshore sedimentary facies and their extrapolation into the offshore basin, the main petroleum system elements, and the major tectonic events (modified after Hawie et al. 2013a)

Full size image

Several volcanic episodes are recorded along the eastern margin of the Levant Basin (Fig. 2). Late Jurassic–Early Cretaceous alkaline volcanism is found in north Lebanon and completely absent in south Lebanon (Dubertret 1955; Nader 2014). Wilson (1992) and Garfunkel (1992) attributed this regional intermittent volcanism to mantle plume activity in the Levant region. Another episode of volcanism occurred in the late Cenozoic (Fig. 2). These alkaline volcanics are linked to open cracks related to the Levant Fracture System permitting local deep decompression and magma ascent (Adiyaman and Chorowicz 2002). Abdel-Rahman and Nassar (2004) attributed the Pliocene alkali basalts of northern Lebanon to a transtensional regime at the junction of the Lebanese restraining bend and the Ghab segment of the Levant Fracture System.

Materials and methods

Source rocks

Twenty-six samples were collected from potential source rock intervals within the Kimmeridgian, Neocomian (Lower Cretaceous), Albian, Cenomanian, and Upper Paleocene in different outcrops and shallow boreholes onshore Lebanon. TOC, TIC, and Rock-Eval pyrolysis were performed on all samples using a LiquiTOC II and a Rock-Eval VI device, respectively. Additionally, Campanian–lower Maastrichtian Rock-Eval data is taken from Bou Daher et al. (2014, 2015). Total sulphur (TS) content was measured using a Leco S200 analyzer (detection limit 20 ppm; error < 5 %).

Vitrinite reflectance measurements were conducted on nine Neocomian coaly shales and two Cenomanian calcareous shales. Sample cuttings were prepared according to guidelines described by Taylor et al. (1998). Random vitrinite reflectance was measured at a magnification of ×500 in a dark room with a Zeiss Axio Imager microscope for incident light equipped with a tungsten–halogen lamp (12 V, 100 W), a 50X/0.85 Epiplan-NEOFLUAR oil immersion objective, and a 546-nm filter and using Zeiss immersion oil with refraction index ne = 1.518; 23 °C. For calibration, a leuco-sapphire mineral standard (0.592 %) was used. Details of sample preparation and microscopic equipment are described in Littke et al. (2012). Vitrinite reflectance data for the Campanian rocks was taken from Bou Daher et al. (2014, 2015).

Bulk hydrocarbon generation kinetics were determined for five samples. Kerogen was isolated from the samples by elimination of minerals using non-oxidizing acid treatment under a continuous nitrogen flow (Durand and Nicaise 1980; Behar et al. 2008), followed by a soxhlet extraction for 1 h using DCM to eliminate the bitumen fraction. A Rock-Eval VI apparatus was used to conduct the experiments. A pyrolysis run was realized on a portion of each sample to ensure that the S1 peak was successfully eliminated. XRD analysis was also performed prior and post-acid treatment to insure the elimination of mineral matter. Bulk hydrocarbon generation rates were then measured at five different heating rates (2, 5, 10, 15, 25 °C/min) starting at 200 °C for 15 min and reaching to 700 °C.

The bulk hydrocarbon generation curves were then used to calculate mean activation energy (Ea) and a pre-exponential factor (A) for every sample using the least squares method according to Schenk et al. (1997). A discrete model with a spacing of 2 kcal/mol between consecutive activation energies was adopted, and an IFPEN software, GeoKin, was used to calculate the activation energy distributions using the previously calculated pre-exponential factor (A).

3D data set

Regional isopach maps for the Levant Basin and margin representing the Jurassic to Plio-Quaternary units were provided by Hawie (2014) who constructed the maps based on available literature (Ponikarov 1966; Ukla 1970; Dubertret 1975; Brew et al. 2001; Hardenberg and Robertson 2007; Gardosh et al. 2008; Powell and Moh’d 2011; Zilberman and Calvo 2013; Hawie et al. 2013a&b). The thicknesses of the pre-Jurassic units were also inferred from data published in neighbouring countries (Brew et al. 2001; Nader 2003; Gardosh et al. 2008; Naylor et al. 2013). However, for simplicity, constant values as shown in Fig. 2 were used for the pre-Jurassic units, except where specific data was available.

Lithological information was based on published data (e.g. Dubertret 1975; Saint-Marc 1972, 1974; BouBou Dagher-Fadel and Clark 2006; Gardosh et al. 2008; Bowman 2011; Hawie et al. 2013a&b) according to the lithostratigraphic framework illustrated in Fig. 2.

Bathymetry maps were defined using sedimentological and biostratigraphical investigation in the onshore and seismic facies interpretation in the offshore (Hawie et al. 2013a&b; Hawie 2014).

Boundary conditions

As an upper thermal boundary, the bottom water temperature (for the offshore) and the surface temperature (for the onshore) and its evolution through the basin history was derived from Wygrala (1989). The lower boundary condition and its evolution through time were determined using a McKenzie-type crustal model with the TemisFlow™ “Advanced basement” tool. This crustal model was constructed based on the assumption that the crust flooring the Levant basin is attenuated continental rather than oceanic (Beydoun 1977; Khair et al. 1993, 1997; Makris et al. 1983; Netzeband et al. 2006; Segev et al. 2006). Present-day crustal thicknesses were set to 8 km under the basin’s central axis and 24 km under the onshore realm based on seismic refraction data presented in Netzeband et al. (2006). The depth of the lithospheric–asthenospheric boundary (LAB) with the thermal value of 1333 °C was adjusted to provide a best fit with the available calibration data and was set at 90 km under the stretched continental crust in the basin and 115 km under the onshore. Pre-rift crustal thickness was assumed to be 24 km. Thus, a Beta factor map was calculated and three rifting events were introduced in the Permian, Triassic, and Early Jurassic. The heat flow (HF) at the base of the sediments was then produced as an output of the crustal model.

Vitrinite reflectance and transformation ratio

The calculation of vitrinite reflectance was based on the EASY Ro% kinetic after Sweeney and Burnham (1990). Transformation ratios were calculated using the measured kinetics for each reported source rock and using the following formula:

$$ \mathrm{Tranformation}\ \mathrm{Ratio}\%=\frac{\left(\mathrm{Initial}\ HC\ \mathrm{Potential}-\mathrm{Current}\ HC\ \mathrm{Potential}\right)\times 100}{\mathrm{Initial}\ HC\ \mathrm{Potential}} $$

Results and discussion

Calibration

Calibration data used in this study consists of borehole temperature data and vitrinite reflectance data (Fig. 3a, b). Temperature data for Delta 01 and Tamar wells was derived from Dubille and Thomas (2012). Due to the scarcity of calibration points in many parts of the basin, especially the offshore, temperature data from calibration wells was used not only at the wells’ location but also for calibration in pseudo wells in areas of the basin with similar crustal thickness, lithologies, salt thickness, and bathymetry to the location of the wells to which the calibration data belongs (Fig. 1). Temperature data from nearby Nir-Am field (SW-Israel) (Shalev et al. 2013) was similarly used for calibration in pseudo well PS8 (Fig. 1). Vitrinite reflectance data (Fig. 3b) was measured on outcrop and near-surface samples collected from Mount Lebanon (Table 1). 1D burial analysis showed that those rocks have reached their deepest burial during the middle Miocene while their adjacent equivalents in the Bekaa Valley and along the coastal areas have reached their deepest burial at the present day due to continuous sedimentation since the Miocene. However, the increase in the calculated VRr for all layers between the middle Miocene and present day is small where the Miocene–Quaternary rock succession is not of a great thickness (<300 m). Thus, the vitrinite reflectance data collected from Lower Cretaceous, Cenomanian, and Campanian rocks on and around Mount Lebanon were used for calibration in several pseudo wells (Fig. 1) where the Miocene–Quaternary rock succession is not thicker than 300 m (Fig. 3b). The use of surface-collected vitrinite reflectance data for calibration in pseudo wells is further explained in Noeth et al. (2001, 2002).

Fig. 3
figure 3figure 3

a Temperature calibration in wells Delta 01 and Tamar, and pseudo well PS8. b Vitrinite reflectance calibration in pseudo wells (see Fig. 1 for location). Colour-coded stratigraphy based on international chronostratigraphic chart

Full size image
Table 1 Summary of source rocks data gathered from onshore Lebanon
Full size table

Source rocks

The Rock-Eval data (Table 1; Fig. 4) proved the presence of several oil-prone and gas-prone source rocks, with varying bulk hydrocarbon generation kinetics (Fig. 5), along the eastern margin of the Levant basin. The Paleocene contains around 30 m of calcareous source rocks showing a wide range of HI and OI values, varying from 100 to 508 mgHC/gTOC and from 96 to 249 mgCO2/gTOC, respectively (Table 1; Fig. 4). The variation observed in HI and OI values of Paleocene source rocks can be due to changes in the relative input of marine and terrestrial organic matter and/or due to oxidation of marine organic matter resulting in a kerogen-type varying between types II and III. The Campanian–lower Maastrichtian contains around 150 m of calcareous source rocks that show a wide range of HI and OI values, varying from 257 to 872 mgHC/gTOC and from 22 to 242 mgCO2/gTOC, respectively (Table 1; Fig. 4). The variation observed in the HI and OI values of the Campanian–lower Maastrichtian source rocks is most probably due to differences in bottom water redox conditions at the time of deposition (Bou Daher et al. 2014, 2015). The Campanian–lower Maastrichtian source rocks are of marine origin and plot in the types I and II field in Fig. 4. Detailed geochemical and petrographic results confirm the presence of type IIS kerogen especially in the TOC-rich intervals of the Campanian–lower Maastrichtian source rocks (Bou Daher et al. 2015).

Fig. 4
figure 4

Pseudo Van-Krevelen diagram showing the type and the organic content of the source rocks sampled onshore Lebanon

Full size image
Fig. 5
figure 5

a–e Bulk hydrocarbon generation kinetics of five source rock samples and f their calculated transformation ratio for a heating rate of 1 °C/Ma

Full size image

The Cenomanian includes also marine calcareous source rocks showing very high HI and relatively low OI values (Table 1; Fig. 4). The vertical extent of the Cenomanian source rocks in Lebanon is restricted to a few beds (3–4 m) only, but this source rock might have a larger extent in the offshore as it might be related to an ocean anoxic event (Kolonic et al. 2002; Lüning et al. 2004; Sachse et al. 2011). Similar Cenomanian source rock quality has been reported in the Bahriya formation of the western desert, Egypt (Moretti et al. 2010). The Albian source rock shows a poor quality (type III/IV kerogen) and a low TOC (Table 1; Fig. 4). The Neocomian (Lower Cretaceous) consists of coaly shales of moderate source rock quality (Table 1; Fig. 4) interbedded within the sandstones of the Chouf formation. This source rock seems to be of regional extent, as it is also reported in the Alam El Bueib formation of the western desert, Egypt (Moretti et al. 2010). The total thickness of the organic rich beds in the Neocomian onshore Lebanon can reach up to 12 m. Kimmeridgian source rocks onshore Lebanon consists of thin (<0.5 m) lenses of shales with very high TOC, moderate HI, and low OI within a generally reefal carbonate formation (Table 1; Fig. 4). However, very prolific Upper Jurassic source rocks are known to exist in the Middle East (Murris 1980; Ayres et al. 1982). All the source rocks analysed from onshore Lebanon are immature (Table 1).

The activation energy distribution for the Paleocene source rock shows a typical type II bell-shaped distribution (Fig. 5a). The Campanian source rock shows a major difference in the activation energy distribution between Campanian 1 and Campanian 2 samples (Fig. 5b, c) which can be attributed to the presence of type II (Campanian 1) and type IIS (Campanian 2) kerogen. A depositional model for the Campanian–lower Maastrichtian source rocks was proposed by Bou Daher et al. (2015) suggesting that type II kerogen with high activation energies would be expected for the Campanian source rocks in the basin, while along most of the onshore type IIS kerogen with lower activation energies is more abundant. The Lower Cretaceous and Upper Jurassic coaly shales show a wide spectrum of activation energies tailing up to 66 kcal/mol which is observed in many type III kerogen bearing source rocks (Fig. 5d, e) (Schenk et al. 1997). The presence of these reported source rocks in the offshore part of the study area is an open question. However, molecular and isotopic data from gaseous hydrocarbons encountered in the Mesozoic–Cenozoic section at the continental margin offshore Israel suggest the presence of several genetic systems (Feinstein et al. 2002, 2013).

Heat flow

The modelled basal heat flow (HF) showed different trends in the basin, margin, and onshore areas (Fig. 6). In the basinal realm, an elevated HF during the Permian–Early Jurassic is the result of rifting pulses that occurred during that period. The HF values in the basin decrease slowly during the post-rift cooling and subsidence phase until the late Eocene/early Oligocene when a fast decrease in HF occurs, most probably as a result of fast sedimentation (Fig. 6a). Along the margin, the HF trend is relatively similar to the basinal area, but with lower calculated HF values for the Permian–Early Jurassic and slightly higher ones during the last 40 Ma (Fig. 6b), which is due to a lower β factor and a lower sedimentation rate along the margin, respectively. In the onshore area, our model shows relatively constant values through time, varying slightly above 50 mW/m2 with an increase in the last 10 Ma due to uplift and erosion of Mount Lebanon (Fig. 6c). The modelled present-day HF in this study shows higher values in the onshore than in the offshore, which is in accordance with the few data available on the present-day surface heat flow. Jiménez-Munt et al. (2006) reported HF values ranging from 40 to 60 mW/m2 with an increase from the thin continental crust flooring the basin to the thicker crust in the onshore. Makris and Stobbe (1984) also reported a similar trend from 0.9 HFU (38 mW/m2) in the basin to 1.6 HFU (67 mW/m2) in the onshore. Many other authors reported a similar decrease in the HF values from the onshore to the offshore as well as from the southern margin (southern Israel) northwards (northern Israel) (Eckstein and Simmons 1978; Roded et al. 2013; Shalev et al. 2013; Schütz et al. 2014). This trend is most probably the result of a decrease in the thickness of the radiogenic continental crust and an increase in the less radiogenic sedimentary cover. The resulting average present-day geothermal gradient varies from 19 °C/km in the offshore basin to around 25 °C/km in the onshore area.

Fig. 6
figure 6

Basal heat flow trends for the offshore basin, margin, and onshore Lebanon

Full size image

Burial history and maturity

The different geodynamic events experienced by the Levant Basin and the onshore Levant region have been recorded differently in the sedimentary successions of different parts of the study area, particularly during the Cenozoic. Since the middle Eocene, the northwestern part of the Arabian plate has been emerging out of the water (Bar et al. 2011), resulting in several regional erosion events (Hawie et al. 2013a), while the nearby Levant basin has been rapidly subsiding and accumulating more than 5 km of sediments (Hawie et al. 2013b). This differential vertical movement had an important effect on the thermal history and source rock maturity of the study area. To illustrate this effect, four burial history diagrams were extracted from four different areas of the model (Fig. 7). Up until the Late Cretaceous, the study area underwent a relatively similar burial history with only slightly higher thermal maturity in the offshore basin due to higher heat flow resulting from rifting pulses at that time.

Fig. 7
figure 7

a Extracted burial history and calculated vitrinite reflectance from a the deepest offshore, b margin, c uplifted mount Lebanon, and d Bekaa Valley

Full size image

The Cenozoic burial history, however, differs considerably in the onshore from the offshore (Fig. 7a–d). A fast burial in the deep offshore basin during the last 60 Ma (Fig. 7a) resulted in the thermal maturation of all Mesozoic and early Cenozoic source rocks. A similar fast burial occurred along the margin, but with a shallower depth due to the thinner Neogene cover, resulting in the thermal maturation of most of the Mesozoic with the exception of the Upper Cretaceous (Fig. 7b). In the onshore, deepest burial on Mount Lebanon was reached in the middle Miocene leading to a halt in maturation since that time (Fig. 7c). In the Bekaa Valley and along the coastal area and topographic lows onshore, sedimentation continued during the uplift of Mount Lebanon throughout the Miocene, Pliocene, and Quaternary resulting in the deepest burial at the present day (Fig. 7d). With the exception of the Jurassic in the Bekaa Valley reaching the early oil window, only Triassic and older rocks have reached the petroleum generation stage onshore, while the younger rocks are still immature.

The effect of the thick Cenozoic cover in the Levant basin on the thermal maturity of potential source rocks can be clearly seen in maturity maps of calculated vitrinite reflectance (Fig. 8). Permian and Triassic source rocks seem to have entered the oil window everywhere in the study area during the Late Cretaceous (Fig. 8a). The Permian and Triassic source rocks experienced deep burial and maturation reaching the wet gas window in most of the offshore basin during the mid to late Eocene and the dry gas window during the Oligo-Miocene. Since the late Miocene, the Permian and Triassic source rocks entered the overmature stage in most of the offshore Levant Basin (Fig. 8a). The maturity of Permian and Triassic source rocks decreases gradually along the margin towards the onshore where it has remained within the oil window throughout the Cenozoic, with the exception of some local deeper burial (e.g. in the Bekaa Valley) where the Permian source rocks have reached the wet gas window.

Fig. 8
figure 8figure 8figure 8figure 8figure 8

Calculated vitrinite reflectance map for a Lower Triassic, b Kimmeridgian, c Cenomanian, and d Campanian. e, f Cross sections AA’ and BB’, respectively, showing calculated vitrinite reflectance at the present day (see Fig. 1 for location)

Full size image

Upper Jurassic to Lower Cretaceous source rocks entered the oil window in most of the offshore basin during the mid to late Eocene, the wet gas window during the late Miocene, and the dry gas stage at the present day (Fig. 8b). Only in the deepest part of the offshore basin are the Upper Jurassic and Lower Cretaceous source rocks overmature (Fig. 8b). Along the margin, these source rocks vary from early oil window to late oil window, while in the onshore, they remain immature or in the very early oil window with slightly higher maturities in the Bekaa Valley. The Upper Cretaceous Cenomanian source rocks entered the oil window during the late Eocene, the wet gas window in the northern offshore basin during the mid to late Miocene, and the dry gas stage at the present day (Fig. 8c). Along the margin, the Cenomanian source rocks are at the present day in the early oil window (Fig. 8c). The Campanian and Paleocene source rocks have entered the oil window in the offshore basin during the late Oligocene (Fig. 8d). In the northern segment of the offshore Levant Basin, the Campanian and Paleocene source rocks reached the wet gas window during the late Miocene while they remained in the oil window in the rest of the offshore basin. Campanian source rocks have reached the dry gas window in the deepest part of the basin offshore Lebanon (Fig. 8d). Along parts of the margin, both source rocks have just entered the oil window, while in the onshore they are still immature. Figures 8e, f shows two E–W and NNE–SSW cross sections, respectively, with calculated vitrinite reflectance that reflect the very different maturation trends along the onshore and offshore.

It should be noted that the maturity stages mentioned in this chapter are standard translations of calculated vitrinite reflectance into petroleum generation stages and do not consider the different kerogen types. Therefore, type IIS bearing source rocks in the onshore area might have generated some oil although they are classified here as thermally immature.

Transformation ratio

The very different source rock maturation trends in the Levant basin, margin, and onshore imply very different potential petroleum systems in each of these realms due to different timing and depth of hydrocarbon generation. In order to assess the petroleum generation timing of a certain source rock calculated, vitrinite reflectance alone is not sufficient, but transformation ratios are also needed.

Transformation ratio maps (Fig. 9) have been extracted for several potential source rocks. The Upper Jurassic Kimmeridgian source rocks started generating hydrocarbons in most of the offshore basin during the very Late Cretaceous and early Paleocene, and have reached 100 % transformation ratio by the late Miocene (Fig. 9a). Along the margin, hydrocarbon generation from the Kimmeridgian source rock continues to the present day (Fig. 9a). The Lower Cretaceous Neocomian source rocks show a similar trend, with a slightly younger onset of hydrocarbon generation (Fig. 9b). In the onshore, limited hydrocarbon generation have occurred from the Kimmeridgian source rocks in some of the coastal areas and inland topographic lows since the late Miocene (Fig. 9a), while the Neocomian source rocks did not generate any hydrocarbons onshore (Fig. 9b). The generated Kimmeridgian hydrocarbons onshore have most probably been washed by meteoric water which have been invading the Jurassic and the overlying rock succession since the emergence of Mount Lebanon in the late Cenozoic (Nader and Swennen 2004).

Fig. 9
figure 9figure 9figure 9figure 9

Source rock transformation ratio maps for a Kimmeridgian, b Lower Cretaceous (Neocomian), c Campanian, and d Paleocene

Full size image

Campanian and Paleocene source rocks have very similar hydrocarbon generation histories starting in the late Oligocene in the deepest part of the offshore basin and remaining restricted to the offshore basin while almost no hydrocarbon generation have occurred from those source rocks along the margin or in the onshore (Fig. 9c, d). The transformation ratio maps of the Campanian source rock were calculated using the kinetics of type II kerogen (Campanian 1) (Fig. 5b). Another simulation was run using the kinetics of type IIS kerogen (Campanian 2) (Fig. 5c) which resulted in a much earlier onset of hydrocarbon generation from Campanian source rocks in the offshore Levant Basin and considerable generation along the margin, while in the onshore, only minor hydrocarbon generation occurred in the Bekaa Valley. Based on the conceptual depositional model proposed by Bou Daher et al. (2015), suggesting that type II kerogen with high activation energies would be expected for the Campanian source rocks in the basin and margin, while along most of the onshore type IIS kerogen with lower activation energies is more representative, we suggest that the scenario reported in Fig. 9c is to our best knowledge the representative case for the Levant Basin and margin. The second scenario, however, is validated for the onshore realm through the presence of large amounts of immature solid bitumen south of the Bekaa Valley which is typical for early stages of hydrocarbon generation from type IIS kerogen (Bou Daher et al. 2015).

Petroleum systems

The results presented and discussed above illustrate the quality and quantity of known source rocks and the timing of hydrocarbon generation from key source rock intervals in different realms of the study area and indicate that different potential source rocks can be active at different times in the offshore, margin, and onshore. However, for a successful petroleum system to be present, several other factors need to come in place. These include migration paths, reservoir quality, trapping, sealing, and seal integrity. In the Permian–Triassic succession, rock units with source, reservoir, and seal potential have been reported in Israel (Gardosh et al. 2008) and in Syria (Brew et al. 2001; Lučić et al. 2010). Triassic source rocks have been proved to charge Middle Triassic, Middle Jurassic, and Upper Cretaceous reservoirs in Syria (Abboud et al. 2005). In the deepest part of the offshore Levant Basin, hydrocarbon generation from Permian–Triassic source rocks occurs between 90 and 34 Ma and petroleum can accumulate in tilted rift blocks (Fig. 10). Along the margin, hydrocarbon generation from the Permian to Triassic starts at later stages (75 Ma) and continues in the onshore till the present day. Onshore structures, such as the Qartaba box fold (Nader 2014), can be an excellent trap for Permian–Triassic hydrocarbons sealed by the Upper Triassic anhydrites that are expected to be present in Lebanon, as an extension of the Kurrachine anhydrites of the Palmyra Basin. Such evaporite deposits could protect the Permian–Triassic succession in Lebanon from water washing and karstification that seems to have affected the Jurassic and younger rock units (Renouard 1955; Beydoun 1977; Beydoun and Habib 1995; Brew et al. 2001; Lučić et al. 2010). Hydrocarbons from Jurassic and Lower Cretaceous source rocks were generated in the deepest basin between 75 and 24 Ma and might have accumulated in reactivated Mesozoic structures along the margin (Fig. 10) (Ghalayini et al. 2014). Jurassic and Lower Cretaceous rocks exposed onshore have excellent reservoir potential (Nader 2014) and can be sealed by the Upper Jurassic (Kimmeridgian) shales, marls, and volcanics and the Lower Cretaceous (Albian) marls, respectively (Nader 2014). Oil has been reported in Lower Cretaceous sandstones in the Mango field offshore Sinai and in the Euphrates Graben (OAPEC Annual Report 2002; Gardosh et al. 2008; Nader 2014). Oil and gas shows have been also reported from Middle Jurassic limestones Yam-2 and Yam Yafi-1 wells off the coast of Israel (Gardosh et al. 2008). One very promising structure that can trap Jurassic and Lower Cretaceous hydrocarbons is the box fold Ile du Palmier off the coast of Tripoli (Fig. 1). The Jurassic and Lower Cretaceous reservoirs in this structure would be protected from water washing by the distance from the uplifted inland intake area (Beydoun 1977; Nader and Swennen 2004).

Fig. 10
figure 10

Schematic diagram summarizing the potential petroleum systems of the Levant basin offshore, margin, and onshore Lebanon

Full size image

Upper Cretaceous source rocks are only mature in the deep basin, and in the deepest part of the basin they produced hydrocarbons between 34 and 16 Ma. Upward vertical migration from Upper Cretaceous (particularly Campanian) source rocks in the offshore basin would be impeded by the presence of a very good Paleocene seal. Thus, migration would most probably occur laterally into marginal pinchouts and reactivated Mesozoic structural traps (Fig. 10). Lower Maastrichtian calcareous turbidites were reported along the coast in Northern Lebanon (Hawie et al. 2013b). These can be potential reservoirs for Upper Cretaceous hydrocarbons, sealed by the Paleocene marls. Paleocene and Eocene source rocks produce hydrocarbons at slightly younger ages and might be charging Oligocene and Miocene reservoirs in offshore structures. Oligocene potential source rocks are in the oil window at the present day and have started to generate hydrocarbons at around 6 Ma and thus can charge Oligocene and Miocene reservoirs in young structures (Fig. 10), such as the NNE–SSW trending anticlines affecting the Oligocene–Miocene succession in the offshore basin (Ghalayini et al. 2014). Disseminated terrestrial organic matter in the upper Oligocene has been proved to be an important source rock offshore the Nile Delta (Villinski 2013).

Sensitivity and uncertainty

The thermal history of a sedimentary basin is controlled by many parameters, most of which are often very poorly constrained, particularly in frontier basins such as the Levant Basin. Testing the sensitivity of the system to some of those parameters can help in quantifying the effects they have on the temperature of the basin and thus on source rocks maturation. In this study, we have extracted a 1D model from the deepest part of the offshore basin and tested for the sensitivity of the present-day geothermal gradient to lithologies’ variation, crustal thickness, and depth of the lithospheric–asthenospheric boundary (LAB). The pre-Messinian Cenozoic lithology was varied from carbonate dominated (low radiogenic heat production, high thermal conductivity) to siliciclastic dominated (high radiogenic heat production, low thermal conductivity). The resulting difference in the present-day geothermal gradient between the tested scenarios affects the pre-Miocene succession shifting the temperatures by up to 18 °C (Fig. 11a). This temperature variation seems to have no effect on the Miocene succession, most probably due to the chimney effect exerted by the high thermal conductivity of the thick Messinian salt (Fig. 11a).

Fig. 11
figure 11

Extracted 1D model from the deepest offshore basin showing the sensitivity of the system to changes in a lithology, b crustal thickness, and c depth of the LAB. See Fig. 1 for location. Colour-coded stratigraphy based on international chronostratigraphic chart. Note that the first 2300 m depth are water column

Full size image

The present-day crustal thickness was varied from 8 to 20 km (Netzeband et al. 2006). The resulting effect on the present-day geothermal gradient increases with depth, reaching a difference of 28 °C between end member scenarios for the deepest intervals (Fig. 11b). The effect of crustal thickness is minor for shallow intervals (Fig. 11b).

The depth of the LAB was varied from 80 to 120 km (Jiménez-Munt et al. 2006; Segev et al. 2006) which had a major effect on the present-day geothermal gradient (Fig. 11c). The effect of these three modified parameters on source rock maturation and petroleum systems is summarized in Fig. 12, showing a significant impact on the expected depth of the oil and the gas windows. The thickness of the pre-Messinian biogenic zone also varies between 700 and 1500 m, indicating that under any scenario, the pre-salt Miocene in the deep offshore basin would have a biogenic gas potential in a zone of at least 700 m, assuming a biogenic zone from 0 to 80 °C (Fig. 12).

Fig. 12
figure 12

Extracted 1D model from the deepest offshore basin showing the minimum and the maximum geothermal gradients and their effect on maturity and petroleum systems. See Fig. 1 for location. Colour-coded stratigraphy based on international chronostratigraphic chart. Note that the first 2300 m depth are water column

Full size image

In an underexplored basin such as the Levant Basin, uncertainties related to hydrocarbon generation timing can raise not only from uncertainty in thermal history but also from several other parameters. TOC values, source rock presence, quality and thickness, hydrocarbon generation kinetics, and lateral changes in organofacies are parameters that have an important impact on hydrocarbon generation and eventually on petroleum systems. Further uncertainties that should be considered when assessing potential petroleum systems in the Levant Basin, particularly offshore, are related to reservoir quality, migration paths, and seal integrity. A catastrophic event that had an enormous impact on seal integrity and petroleum systems is the Messinian salinity crisis. The fast drop of sea level by around 2000 m resulted in a drop in pressure, an increase in temperature, and fast salt deposition. The surface temperature change during the Messinian is minor and has little effect on the petroleum systems. The hydrostatic pressure drop, however, would result in higher pore pressure and possible seal fracturing within the shallow biogenic system (Wygrala et al. 2014). The effects of this event are thought to attenuate in deeper biogenic systems (e.g. Tamar, Leviathan, Aphrodite) and in speculative thermogenic systems.

Conclusions

The results presented in this study proved the presence of several potential petroleum source rocks along the eastern margin of the Levant Basin and suggest different prospective working petroleum systems in the offshore, margin, and onshore Lebanon.

An Upper Cretaceous–Oligo-Miocene mixed thermogenic and biogenic system is expected in the offshore. In the pre-Messinian salt succession, the thickness of the biogenic zone can vary between 700 and 1500 m throughout the offshore basin, rendering the NNE–SSW offshore anticlines which affect the Oligo-Miocene succession an excellent target for biogenic and thermogenic hydrocarbons. The biogenic hydrocarbon potential decreases towards the margin and onshore due to the lack of appropriate seal.

A Jurassic–Lower Cretaceous petroleum system is expected along the margin, where Lower Cretaceous sandstones and Upper Jurassic Carbonates, which have excellent reservoir quality in the onshore, could be charged with oil and gas from Jurassic and Lower Cretaceous shales.

A Permo-Triassic petroleum system is expected in the onshore, where Lower Triassic reservoirs sealed by Upper Triassic evaporites in Miocene structures would be filled with gas from Triassic (and possibly deeper) source rocks.

Kinetic results derived from Campanian source rocks showed a variation of bulk hydrocarbon generation kinetics attributed to lateral variation of organofacies within the same source rock. These lateral source rock quality changes have an important impact on the assessment of hydrocarbon generation particularly in a frontier basin. Thus, a detailed understanding of a source rock’s depositional environment at the sampling sites is essential for a better source rock extrapolation at a basin scale.