The microscopic model consists in a 1D sedimentary column that accounts for sedimentation, compaction, Darcy flow and Diffusion flow. It is particularly dedicated to the study of the evolution of the sedimentary organic matter in the first hundred meters below the water– sediment interface, over a period of time of a few hundreds of thousand years. The organic carbon is the only nonsoluble element considered in this version of the model. The dissolved elements are O₂, SO₄^2-, H₂, CH₃COOH, and CH₄. Methane is dissolved in water or present as a free phase if its concentration exceeds its solubility at given pressure and temperature.

Like other living organisms, microbes require for life and reproduction (Sherr and Sherr, 2000): i) energy, ii) elements for biosynthesis (C, N, P, S, Fe, Mn…) and iii) electrons for producing organic structures.

The energy is provided through redox reactions that could be described as follows (e.g.Hedin et al., 1998).

(1)

Where Organic Matter (OM) is the carbon source: organic matter or CO₂ (electron donor); B_ox is the electron acceptor (e.g. O₂, SO₄^2-); OM’ is a degraded form of the organic matter; B_red is the reduced form of B; ΔG is the free energy that can be used by the microbes.

The reactions accounted for in this first version of the model are (see Figure 1):

• The aerobic oxidation of the OM into CO₂

• The anaerobic sulfate-reduction of the OM, the acetate, and methane into carbon dioxide and sulfur

• The hydrolysis and fermentation of the OM into acetate, hydrogen and CO₂.

• The two pathways for methanogenesis of (1) acetate and (2) hydrogen and CO₂ into methane.

FIGURE 1
Reaction scheme used in the 1D microscopic model.

The substrate (S) and the biomass (B) concentration are supposed to be linked by the following equations (modified from Lawrence and McCarty, 1969; Rosso, 1995; Cherruault, 1998). In case of dissolved substrate, it is assumed that its concentration follows a convective and diffusive mixing process.

(2)

The term B stands for the biomass concentration [mass/volume], S is the substrate concentration [mass/ volume], and “a” represents the growth yield coefficient. The term B_max is the maximum concentration of the biomass or the carrying capacity. The term U is the fluid Darcy velocity; Terms D and τ are respectively the diffusion coefficient in water and the tortuosity of the porous medium. The reaction rate μ is given by the Michaelis–Menten formula:

(3)

Where μ_max is the maximum rate of substrate utilization per unit weight of micro-organisms [1/time], and Ks is the Monod coefficient equal to the substrate concentration when dS/dt is equal to one-half of the maximum rate. The temperature dependence of the reaction rate is given by the Rosso et al. (1993) relationships (see Figure 2).

FIGURE 2
Example of temperature dependency for μmax.

In the previous equations, “aμ” is the microbial growing rate and “aμB/B_max” is the microbial death rate. It can be easily checked that these two numbers are equal when the microbial population reaches the carrying capacity. These two parameters control the growth and the decrease of the microbe population as shown in the example presented in Figure 3.

FIGURE 3
Example of simulated evolutions of the substrate content and the microbe population in a waste treatment reactor (data from Converti et al., 1998).

The first results of the model are consistent with the observations in terms of nutrients and microbes concentrations. In the example presented on Figure 4, the oxidation zone is less than 10cm thick, the sulfate reduction zone is some meters thick, and the fermentation zone could be defined between 3–4mbsf (meters below sea floor) and 20–30mbsf. The bottom boundary of the fermentation zone is correlated with the disappearance of the hydrolysable organic matter.

FIGURE 4
Synthetic example of species and microbes concentrations calculated by the 1D microscopic model.

It should be noticed that each equation is defined with free parameters (e.g. a, B_max, μ_max, K_s…) that have to be calibrated in geological situations. This task is under development (Schneider, in prep) using, in particular, the results from the Ocean Drilling Program (ODP)/DSDP programs and some existing lab experiments.

In order to produce methane, the organic matter, once buried, should be preserved partially from the whole oxidation zone. A set of sensitive tests has been carried out in order to evaluate the preservation of the HOC through the oxidation and sulfate reduction zone. The results indicate that the preserved HOC is a function of the initial HOC, and that the preserved HOC increases as the sedimentation rate increases (see Figure 5).

FIGURE 5
Preservation of the HOC through the oxidation and sulfate reduction zone. Left: the preserved HOC is a function of the initial HOC. Right: The preserved HOC increases as the sedimentation rate increases.

These results suggest that the lower boundary of the Clayton chart (Clayton, 1992) is not universal but is a function of the initial HOC. Another result is that a minimum HOC is required, at a given sedimentation rate, in order to enter the fermentation zone and then to produce methane.

Another set of numerical experiments has been carried out in order to determine the Expulsion Depth (ED). The expulsion depth is defined as the depth where the first buble of free methane is formed. This depth corresponds to the point where the methane concentration in water exceeds the methane’s solubility (e.g.Lu et al., 2008). It is assumed that if the vapor phase appears too close to the surface, a fraction of the methane will be likely lost and not preserved in the geological system. In the example presented in Figure 6, the expulsion depth is around 50mbsf.

FIGURE 6
A) The Expulsion Depth (ED) is defined when the methane concentration in water becomes higher than its calculated solubility. B) The expulsion depth increases with the increase of the sedimentation rate. C) The expulsion depth increases when the initial HOC decreases.

The sensitivy analysis shows that the expulsion depth increases as the sedimentation rate increases, and as the initial HOC decreases (see Figure 7): the degazing of methane occurs closer to the surface in case of low sedimentation rate or high TOC (preservation in the geological system less likely).

FIGURE 7
Left: Results of a simulation with a thermal gradient of 10ºC/km. Right: Result of a simulation with a thermal gradient of 50ºC/km. The comparison of the two simulations shows that one when the methane production increase when the thermal gradient increases because the methane solubility in water decreases when the temperature increases.

The tests performed with the model also showed that the production of methane in vapor phase increases as the thermal gradient increases because the methane solubility in water decreases as the temperature increases. These results justify the lower boundary of the Clayton’s chart.

“Conventional” thermogenic petroleum system models are designed to geological processes occurring at large scale and over long period of times. Each cell of such models is 10 to one 1000 meters thick and 0.5 to 5km wide, with main time steps generally exceeding 0.1My. Biological processes occurring close to the surface are not directly taken into account. Results of the microscopic model have to be adapted: the objective is to define a biogenic source rock potential which is not only generated but also preserved in the geological system. The amount of HOC consumed by concurrent reactions must be removed from the petroleum system model, as well as the amount of HOC corresponding to the biogenic methane lost at the surface during the early generation. In the same way the potential must be reduced if a fraction of the generation potential is not expressed before the pasteurization.

The sensitivity analysis carried out with the microscopic model indicates that free methane generation is conditioned by the initial TOC, the sedimentation rate and the thermal gradient. A modified Clayton chart (Clayton, 1992) has then been derived in which the lower boundary is now a function of the initial TOC. It should be noted that the upper boundary of the Clayton chart is not justified by the microscopic model in its present version. One reason is put forward: the biogenic potential cannot be fully expressed if the rock crosses too quickly the “biogenic window”, the pasteurization may occur before the complete “consumption” of the HOC by bacteria. Another indirect reason is that the organic matter is often too diluted in case of high sedimentation rate. Further studies are required to valid or invalid this “upper boundary” concept.

For a specific study, the first step is to identify the shaly sediment that contains enough TOC to be considered as a potential biogenic source rock. As explained previously, in the optimal situation the HOC is neither too low (organic matter fully consumed by concurrent reactions) nor too high (liberation of methane in vapor phase too close to the surface). Then, according to the estimated sedimentation rate and the estimated thermal gradient during its deposition, the rock is classified as a good biogenic gas source rock or not, through the definition of a “Biogenic Gas Potential Index”.

For the sake of simplicity when the sedimentation rate and the thermal gradient have some spatial variations, it could be easier to convert the Clayton’s chart into two single heating rate values. In the example presented the lower boundary is equivalent to an heating rate of 7ºC/Ma and the upper boundary equivalent to and heating rate of 18ºC/My (Figure 8). The optimal biogenic window is in between the curves.

FIGURE 8
Modified Clayton’s chart showing the equivalence in term of heating rate. The optimal geological condition for the biogenic gas generation and preservation in the geological system is comprised between both curves.

In order to get a macroscopic model usable in a basin model; the set of logistic reactions is up scaled into a set of first order kinetics assuming that the biomass is a catalyst (Pfeffer, 1974). It should be noted that the kinetic parameters calibrated from lab experiments are not able to reproduce the nature of the fluids generated at low maturity. Indeed, the kinetic parameters are mainly calibrated with kerogen samples for which Vitrinite Reflectance (Ro) is taken to be approximately 0.6%.

As mentioned before, kinetic parameters are determined in laboratory with samples that are at the limit between diagenesis and catagenesis, at the onset of the oil window hydrocarbon generation. Let us consider a synthetic which is characterized with constant partial HI (Hydrogen Index) of 1mg HC/g TOC for each of the activation energy from 34 to 64kcal/mol (Arrhenius coefficient= 3.1 10¹⁵s^-1). The model performs a numerical maturity experiment on this kerogen in order to place it at a maturity level equivalent to Ro= 0.6%. The results show that all the partial potentials have been totally consumed for activation energy below 50kcal/mol (Figure 9). Considering activation energies between 50 to 54kcal/mol, the partial potentials have been all but partially consumed.

FIGURE 9
Artificial maturity experiment on a synthetic sample (A) with uniform distribution of partial potential and A= 3.1 10¹⁵s^-1. At maturity equivalent to Ro= 0.6% (B) all the partial potentials have been totally consumed for activation energy below 50kcal/mol.

As a matter of fact, for an initial Type III kerogen (“humic” organic matter mostly formed of terrestrial plants: “continental” kerogen), the lowest activation energy that has a positive partial potential is 54kcal/mol (Figure 10, Schneider et al., 2012).

FIGURE 10
A) Original kinetic parameter for a Type III kerogen calibrated at Ro= 0.6%. B) Recalculated Type III kerogen at the beginning of the diagenesis window.

By considering the initial Type III kerogen as a starting point and if by observing natural data such as gas compositions, a new set of kinetic parameters may be derived to account for low maturities in the gas fields. The derived Type III OM differs from the previous one by a 13% increase of the HI from 235 to 271mg HC/g TOC (Figure 10).

The simulations carried out with new fitted Type III kerogen parameters enables to reproduce quite well the filling of the fields as well as the general composition and state of the hydrocarbons (Schneider et al., 2012). In the simulations, most of the gas is generated at a very low maturity: Vitrinite Reflectance Ro less than 0.6%. In this domain, it has been stated that the thermal C-C–bound breakage is unlikely to be significant without catalysis. Because microbial enzymes are the most efficient lowtemperature catalysts known, this kinetic model could be described as a microbial enhanced thermal model.

The Offshore Lebanon comprises the continental margin (Mesozoic carbonate shelf and slope) and a large deep water area (e.g.Hawie et al., 2013). They are part of the Levant sedimentary Basin bounded to the East by the Arabian Plate, to the south by Africa, to the west by Eratosthenes Continental Block and to the north by the Cyprus tectonic arc (subduction). This Basin resulted from the fragmentation of the Pangea by rifting and extension from Triassic to Middle Jurassic. While the margins are typical carbonate platforms with steep slopes, the 12 to 14km sedimentary infill of the basin comprise deep water pelagic and re-deposited clastic sediments at least since the Upper Jurassic and until Present time. It was affected by the Messinian Salinity Crisis with sub-aerial erosion of the margins and deposition of about 2000m thick evaporites in the deep basin, a major regional seal partly responsible for the under-compaction of the basin.

Offshore Lebanon is a frontier Basin with no drilled well in which hydrocarbon resources are not yet proven. However major gas discoveries have been made in the vicinity in Offshore Israel (Tamar, Leviathan) and Offshore Cyprus (Aphrodite) in sandy turbidites reservoirs of Lower Miocene age intercalated between Miocene and Oligocene shales. Moreover, the giant Zohr gas discovery in Egypt (close to Eratosthenes sea mount, announced in 2015) proved that shallow water carbonates of the margins connected to adjacent Miocene-Oligocene basinal shales can be gas reservoirs. When the study was carried out in 2011–2012 the biogenic nature of the gas was not yet confirmed and was still debated among the community (Dubille and Thomas, 2012; Dubille et al., 2013; Schneider et al., 2013).

The petroleum system study was preceded by an extensive geological and geophysical study. An integrated interpretation of regional 2D seismic surveys offshore Lebanon provided structural maps and an improvement of the geological knowledge. These data allowed the building of a geological model in 3 dimensions at the scale of the whole offshore Lebanon. The pressure field and the thermal field were calculated in this geological model through geological times, as well as the hydrocarbon generation / expulsion / migration. The thermal modeling included a modeling of the whole lithosphere. Thermogenic source rocks observed in surrounding basins were implemented and tested in the model.

The analysis of the thermal gradient and the sedimentation rate allows defining the potential biogenic gas source rock. For example in the offshore Magdalena Delta, according to the geothermal gradient values in the studied area ranging from 14 to 20ºC/km, the optimum sedimentation rate for biogenic gas preservation is in the range 300 to 1500m/My.

In the case where the thermal gradient and the sedimentation rate present spatial variations, such as in offshore Lebanon, a methodology is proposed for identifying which parts of a layer could develop a biogenic gas generation and preservation potential. The thermal gradient and the sedimentation rate calculated at deposition time by TemisFlow3D® (in the uppermost layer of the model at a given time step) are multiplied for obtaining a “heating rate map at deposition time” (Figure 11). The generated maps give, for a given layer a first overview of the biogenic gas generation potential.

FIGURE 11
Illustration of the methodology that identify, for a given layer, the area that may produce microbial gas. The thermal gradient is calculated in the uppermost layer of the model, the average thermal gradient in the basin is much lower around 22ºC/km.

A map of the “Biogenic Gas Potential Index” is then edited using criteria on the Heating Rate map (heating rate values within the “Biogenic Window”), but also on the TOC and on the sedimentation rate. It is also considered that even if the heating rate exceeds 18ºC/My, a reduced biogenic gas potential may remain in the system. Moreover, the evolution of the heating rate of the rock after its deposition is checked too: if the pasteurization temperature is reached too quickly, then the potential is decreased.

Resulting maps of biogenic gas generation and preservation potential are interpreted in terms of remaining or effective Total Organic Carbon TOCe available for the biogenic gas system, considering the following assumptions:

• When heating rate conditions are favorable for microbial generation and preservation (“highly probable” on the map Figure 12), the biogenic gas source rock is characterized with a full potential (TOC= TOCe).

• When the heating rate conditions are not favorable for the microbial generation and preservation (“no biogenic potential” on the map Figure 12), the concerned area is characterized with a null potential (TOC= 0%). Indeed, there is no effective biogenic potential.

• Intermediary Biogenic Gas Potential (0%

FIGURE 12
Definition of the possible biogenic generation zones for the 7 potential biogenic gas source rocks.

There was no geochemical data for calibrating the TOCe in Offshore Lebanon, so TOCe values have been estimated from analogs. In the Carupano Basin, the “TOCSR methodology” (see appendix of Schneider et al., 2012) has been used in order to define continuous TOC profiles from sonic and resistivity wire logs in poorly compacted sediments. The method has been applied to all the wells that have a sonic and a resistivity logs. As a result, the average TOC content of the possible biogenic gas source rock is in the range 0.5 to 1.2% with a maximum averaged value of 2.5%.

In the case of offshore Lebanon, we assumed that the full potential of the biogenic gas source rock is set to TOCe= 1%. In intermediate zones the TOC equals 0.3% or 0.6%. The resulting methane generation potential varies between 0 and 1.25kg CH4/m3 Rock.

The screening of the different shaly layers in term of thermal gradient and sedimentation rate allowed defining seven levels with a possibility of generating microbial methane (Figure 12)

The results of the TemisFlow® simulations show that the 2 shallow biogenic gas source rocks (Pliocene and Upper Miocene) did not expelled noticeable amount of microbial methane despite their initial potential (Figure 13). Main contributors to biogenic gas petroleum systems would be Middle-Lower Miocene and Lower Oligocene layers. The expulsion from this layer is also enhanced by the presence of closely interbedded reservoir sands and source shales in turbiditic sediments of the deep offshore Levant Basin (distal fan of the Nile Delta).

FIGURE 13
Cumulative expelled methane calculated by the model.

According to the model at present day, Lower Miocene and Oligocene sediments are beyond the pasteurization temperature in the deep basin: in these deep layers the biogenic methane generation ended after the Messinian Crisis (deposition of 2000m of salt). Large gas discoveries in offshore Israel are likely related to a biogenic system which is no longer active even if the gas expulsion from the source rocks continues up today at a lower rate.

The computed HC composition indicated that the gas, mostly biogenic, completely dominates the shallow plays from the Pliocene to the Oligocene. In particular, the Lower Miocene play which produces gas in Tamar and Leviathan is easily filled by biogenic gas. The oil/condensate content would increase in the Eocene, and overall in the Cretaceous– Jurassic. The gas content increases again in deep layers (mostly thermogenic gas), up to 100% in the Triassic play.

The results of the study indicate that there is petroleum potential in the offshore Lebanon. The most active systems would be:

• The Oligo–Miocene biogenic gas source rock, with Upper Oligocene and Miocene clastic reservoirs. The main risk is the absence of deep-sea fans related to the Nile Delta or to other local contributors (mainly in the North along the Latakia Ridge). The dense Oligo–Miocene fault network in offshore Lebanon (Hawie et al., 2013) may increase the vertical gas migration in shallower Middle and Upper Miocene reservoirs (below the Messinian Salt) but may also limit the size of drainage areas.

• The Upper Cretaceous thermogenic source rocks, with hypothetic Paleogene clastic reservoirs. The main risk is the absence of Upper Cretaceous SRs– which are well-known onshore Lebanon– in the deep offshore domain where the sedimentary context is not favorable (Bou Daher et al., 2015). This system is largely disconnected to the biogenic plays in the basin, even if traces of heavier components may locally reach Neogene reservoirs (dense Oligo–Miocene fault network in offshore Lebanon rooted in a Paleogene detachment level).

• The Mesozoic thermogenic source rocks (Upper Cretaceous and Jurassic), with Mesozoic (in particular Jurassic) shallow water carbonate reservoirs or Lower Cretaceous clastic reservoirs. Deep Jurassic source rocks are poorly documented in the northern Levant Basin but described in its southern part (Gardosh et al., 2006). In Lebanon their existence is possible on the Mesozoic carbonate platform (buried in the eastern half of the study area).

• The model also shows that the biogenic gas generated in Oligo–Miocene clastic sediments may locally migrate laterally within the Mesozoic carbonate platform. This possibility seems to be realistic as shown by the recent Zohr gas discovery (Offshore Egypt).

In the reference scenario the biogenic gas represents 40% of in place hydrocarbons at basin scale (and 60% of the gas), but only 20% of expelled hydrocarbons. Indeed the late expulsion (Mio–Pliocene) is favorable to methane accumulations in Neogene layers of the deep basin.

Many structures in Neogene reservoirs of the deep basin could be filled with biogenic methane, even without a thermogenic source rock contribution.

Coupling the Biogenic and the thermogenic model, the petroleum potential of the offshore Lebanon has been evaluated as a dry gas zone in its western part (dominant biogenic origin in the deep offshore basin), a reduced oil zone in its eastern part along the shore (limited petroleum potential), and an intermediate ribbon between the oil and the gas zones where both biogenic and thermogenic gas are expected, with the likely presence of condensates in Mesozoic reservoirs (Figure 14)

FIGURE 14
Synthesis of the results of the model in which the thermogenic generation model has been coupled with the biogenic generation model.