Introduction

The ever-increasing demand of energy requires a deep insight into all possible ways of fulfilling it. One of the major components of the global energy mix comes from the exploitation of conventional resources (Bocora 2012; Zou et al. 2016). The increasing rate of the hydrocarbon production from conventional resources show continuous depleting trend for over 50 years (Miller and Sorrell 2014). Hence, there is a strong impetus to explore the hydrocarbon potential of unconventional resources i.e. oil/gas shales. Furthermore, the rise of unconventional resource exploitation could present a new era in the overall global energy mix (Kefferpütz 2010; Sunjay 2011; Bocora 2012; Rezaee 2015).

Conventionally, shale is a source rock for petroleum systems, showing low permeability (Gluyas and Swarbrick 2009; Liu et al. 2012; Rezaee 2015). Contemporary geophysical and geological techniques have been focused on extracting hydrocarbons from clastics and carbonates due to their favorable permeability. With advances in hydrocarbon exploration techniques in the last few decades, it is now possible to extract hydrocarbons directly from source rocks (Kefferpütz 2010; Bocora 2012; Rezaee 2015).

The Total Organic Content (TOC) in shale, its thermal maturity and good porosity are key parameters for the generation and accumulation of hydrocarbons (Rybach 1986; Wood 1988; Wandrey et al. 2004; Gluyas and Swarbrick 2009; Rezaee 2015). Additionally, the vertical and lateral extent of source rocks can be identified and mapped by using seismic and well data (Løseth et al. 2011; Rezaee 2015). In fact, TOC content can be precisely determined at boreholes using different techniques such as Passey et al. (1990), Schwarzkopf (1992) and Bowman (2010) etc., and via the analysis of rock samples. This TOC distribution can be further mapped and extended with confidence away from control points on the reservoir scale using seismic data (Løseth et al. 2011; Rezaee 2015).

The objective of the paper is to evaluate the hydrocarbon potential of Sembar Formation in the Badin area of the Lower Indus Basin, Pakistan (Fig. 1), by mapping the TOC content using seismic and well log data within the context of unconventional resource production. Pakistan is endowed with 51 Trillion Cubic Feet (TCF) of recoverable gas shale reserves (Rezaee 2015). Shale of Sembar Formation is considered to be the most potential candidate in the region, with favorable characteristics, in terms of thickness, TOC values, and thermal maturity (Wandrey et al. 2004). The workflow followed in the present study is given in Fig. 2. The data used in this study comprises of 3-D seismic data (13 × 13 km2) and well log/vertical seismic profiling (VSP) data recorded in three wells (A, B and C) (Fig. 3).

Fig. 1
figure 1

Map showing tectonics and regional geological setting of Pakistan including our study area (highlighted) situated in the Lower Indus Basin (Kazmi and Snee 1989)

Full size image
Fig. 2
figure 2

Flowchart of methodology followed in this study

Full size image
Fig. 3
figure 3

Base map of the Badin area with in-lines (gray) and cross-lines (green) including three wells (Well A, Well B and Well C)

Full size image

Geological setting

The study area is located about 160 km East of Karachi in the Lower Indus Basin, between Latitude 24°5′N–25°25′N and longitude 68°21′E–69°20′E (Fig. 1; Kadri 1995; Alam et al. 2002; Ahmad et al. 2011; Munir et al. 2014). The evolution of the Lower Indus Basin began in the Cretaceous, when the tectonic activity occurred in the Indus Basin forming a rift zone (Farah et al. 1984). This rift zone exists near the Thar Platform in the form of Sargodha High dividing the Indus Basin into the Upper and the Lower Indus Basins (Kadri 1995). As a consequence of this rifting and doming, mainly horst and graben structures are present below the Paleogene unconformity including Cretaceous and older strata (Alam et al. 2002). The effect of rifting and doming on our study area (eastern side of Lower Indus Basin) is relatively small, as it lies away from the main deformational zone. The extensional deformational rate increases from east to west and a large area of the eastern Lower Indus Sub-Basin (Badin) is dominated by tilted normal faults (Farah et al. 1984; Kadri 1995). These faults can be seen and mapped on seismic data which have deformed the Cretaceous and older strata (Kadri 1995).

Stratigraphically, the Indus Basin ranges from Cambrian to Holocene in age with erosional and non-depositional intervals, while in study area it ranges from Jurassic to Paleogene as shown in Fig. 4 (Alam et al. 2002). The depositional environment of the Cretaceous Sembar and Goru Formations is shallow marine in nature (Iqbal and Shah 1980). These formations are deposited over the regional erosive surface of Chiltan limestone (Jurassic age) along passive continental margins (Iqbal and Shah 1980). The Sembar Formation is a clastic rock with a large quantity of shale and a minor amount of siltstone and sandstone while Goru formation is a mixture of interbedded shale and sandstone (Fig. 4; Iqbal and Shah 1980; Raza et al. 1990). The shelf to shallow-marine depositional environment persisted till Late Cretaceous time, represented by the carbonates of the Parh, Moghal Kot and Fort Munro Formations. The Pab Formation hosts regressive sand, thus representing a change to a nearshore environment (Wandrey et al. 2004).

Fig. 4
figure 4

Generalized stratigraphic chart of the study area (Alam et al. 2002; Abbasi et al. 2014, 2016)

Full size image

The Sembar, Goru and Ranikot Formations are major potential conventional/unconventional sources of the Lower Indus Basin (Ahmad et al. 2011; Abbasi et al. 2014). The hydrocarbon potential of the Sembar Formation in the northern Kirther range is low but improves towards the south and southeast (Eckhoff and Alam 1991). Lithologic composition of the Sembar Formation is variable in different areas (Ahmad et al. 2011). The average composition includes Quartz 42%, Clay 47%, Calcite 10% and Pyrite 1% (Ahmad et al. 2011). The gas content is also variable in the basin with 1% < CO2 < 70%, 1% < N < 80% and 0.1% < H2S < 13%. The Sembar Formation extends towards the southern parts of Pakistan and into the offshore area (Eickhoff and Alam 1991; Shahzad et al. 2017). In the western side of the Lower Indus Basin, units of the same age (Kzhdumi, Garau and Gadvan Formations) continue into Iran in the Zagros fold-belt and the Persian Gulf Basin (Opera et al. 2013). These Formations possess potential source character in their respective basins within the context of unconventional resource exploitation (Opera et al. 2013). On the eastern side of the Lower Indus Basin, the Indian Rajasthan Basin comprises of lithological units, owing to the same depositional and tectonic history (Biswas 1987; Klett et al. 2011).

Data and approach

In this research, we have devised a methodology for TOC estimation in the Sembar Formation in Badin area, Lower Indus Basin, Pakistan using 3-D seismic data together with well log data of the three wells (A, B and C). The 3-D seismic cube is acquired and processed using bin size of 25 × 25 m2 and frequency range of 8–45 Hz, however this data can best resolve the target depth at frequency range of 10–35 Hz. The well data used, comprises of gamma ray (GR), spontaneous potential (SP), caliper (CALI), resistivity, density (RHOB), sonic (DT) and neutron porosity logs.

An indirect method is used for the estimation of TOC using well log data, which is an integral part for our proposed methodology. There exist many techniques for estimation of TOC from well log data such as Ferlt and Rieke (1980), Schmoker (1981), Meyer and Nederlof (1984), Fertl and Chilinger (1988), Passey et al. (1990), Myers and Jenkyns (1992), Schwarzkopf (1992) and Guidry and Walsh (1993). The contrast in geophysical response of a mature source rock, as compared to a non-source rock, provides the basis for these indirect techniques (Meyer and Nederlof 1984; Passey et al. 1990; Myers and Jenkyns 1992; Schwarzkopf 1992).

A substantial amount of organic matter in a mature source rock is converted to gaseous or liquid hydrocarbon, displacing formation water from the pores, and later revealing anomalous behavior on well logs (Philippi 1968; Nixon 1973; Meissner 1978). For example, resistivity, DT, neutron and GR logs show relative increase in the interval of mature source rocks, while density log response shows a decrease in mature source rock intervals, as described in Table 1 (Passey et al. 1990). In this study we have followed the approaches presented by Passey et al. (1990) and Schwarzkopf (1992) for the estimation of TOC from well log data.

Table 1 Responses of five conventional logs in the organic matter of source rock, in which only density response is low and other four logs (GR, DT, neutron and resistivity) have high response (Fertl and Chilingar 1988; Passey et al. 1990)
Full size table

Reflection seismology (seismic methods) is one of the most important methods for the delineation of subsurface rock properties over larger extents despite its comparatively low resolution. On the other hand, well log/core data with relatively high resolution provides reliable information, but only for very small areal extents i.e. near the boreholes. The uncertainty in the rock property modeling increases as the distance from borehole increases. That is the point where seismic measurements in conjunction with rock physics modeling are very useful and its calibration with available well data can provide us a reliable model at the reservoir scale. The workflow followed in the current study is presented in Fig. 2 and detailed below.

Demarcation of Sembar Formation

Seismic reflection data is used to understand the regional distribution of the Sembar Formation in combination with well log data (Fig. 3). The top and base of the Sembar Formation are identified/marked on seismic data using the available well tops and is confirmed through synthetic seismograms (well-to-seismic tie) generated from well log and VSP data. Time and depth contour maps are prepared from the top and bottom of the Sembar Formation to identify the lateral and vertical trend in the study area. Horizon based Relative Acoustic Impedance (RAI) attribute for the Sembar Formation is applied to trace any acoustic impedance variations and to map TOC at reservoir scale.

Passey overlay technique

In the ΔLogR method, resistivity logs are superimposed to one of three porosity logs (neutron, DT and RHOB) (Passey et al. 1990; Liu et al. 2012). Resistivity logs are plotted on logarithmic scale and other logs (neutron, RHOB and DT logs) are displayed on linear scales (Rider 2002). Any combination of two logs can be taken i.e. resistivity/DT, resistivity/neutron or resistivity/RHOB. In this method, the neutron and RHOB logs are mostly affected by borehole wall rugosity, and it is better to use a combination of resistivity/DT for more precise results (Liu et al. 2012). Resistivity and DT curves are relatively scaled so that 100 µs/ft of sonic data is equal to two logarithmic cycles of resistivity data (i.e. 100 µs/ft = 2 Ωm or 50 µs/ft = 1 Ωm). Jianliang et al. (2012) also used GR logs by keeping similar scale factors for sonic data (50 API = 1 Ωm).

The interval in which both curves are superimposed is known as a baseline. The baseline is mostly established in a fined grained non-source rock where both curves are on the top of each other with small or zero separation. The separation between the curves in the source rock is called ΔlogR which can be calculated mathematically as:

$$\Delta {\text{Log}}{{\text{R}}_{{\text{DT}}}}=~{\text{Lo}}{{\text{g}}_{{\text{10}}}}\left( {\frac{{\text{R}}}{{{{\text{R}}_{{\text{Baseline}}}}}}} \right) - ~0.02~ \times ~\left( {\Delta {\text{t}}+~\Delta {{\text{t}}_{{\text{Baseline}}}}} \right).$$
(1)

here \(\Delta {\text{Log}}{{\text{R}}_{{\text{DT}}}}\) reflects the separation between DT and resistivity logs, \({\text{R}}\) is the true resistivity from LLD or ILD resistivity logs, \({{\text{R}}_{{\text{Baseline}}}}\) is the value of resistivity corresponding to sonic transit time at baseline, \(\Delta {\text{t}}\) is the sonic transit time and \(\Delta {{\text{t}}_{{\text{Baseline}}}}\) is the value of sonic transit time corresponding to resistivity at the baseline. The constant value 0.02 is the scaling factor. Using \(\Delta {\text{Log}}{{\text{R}}_{{\text{DT}}}}\), TOC can be estimated as:

$${\text{TO}}{{\text{C}}_{{\text{DT}}}}=~\Delta {\text{Log}}{{\text{R}}_{{\text{DT}}}} \times ~{10^{\left( {2.297 - 0.1688~ \times {\text{LOM}}} \right)}}.$$
(2)

The \({\text{TO}}{{\text{C}}_{{\text{DT}}}}\) is the TOC calculated using DT log, and \({\text{LOM}}\) is the level of organic maturity. LOM values can be best obtained from core data. Whenever core data are not available, then conodont alteration indexes and Hood et al. (1975) charts can be used for LOM calculation, provided that vitrinite reflectance data is available (Passey et al. 1990).

It is important to remember that TOC content and the maturity of the source rocks are directly proportional to the ΔlogR separation. A large separation indicates large amounts of organic matter and high maturity for source rocks (Passey et al. 1990).

Schwarzkopf’s technique

For application of Schwarzkopf’s technique, we have followed the approach presented by Myers and Jenkyns (1992). This approach requires a few assumptions before the application of its mathematical relations. The assumptions are: (1) the source rock comprises mud rock as the matrix with density of 2.70 g/cm3, water filled density of 1.05 g/cm3 and kerogen density of 1.1 or 1.2 g/cm3, (2) porosity and density of the source and non-source interval are same. The porosity for water filled pores (\({ \upphi _{{\text{fl}}}}\)) and kerogen (\({ \upphi _{{\text{ker}}}}\)) are estimated using the relations given below:

$${ \upphi _{{\text{fl}}}}=~\frac{{{\uprho _{{\text{ns}}}} - ~{\uprho _{{\text{ma}}}}}}{{{\uprho _{{\text{fl}}}} - {\uprho _{{\text{ma}}}}}},$$
(3)

and

$${ \upphi _{\ker }}=~\frac{{{\uprho _{\text{s}}} - ~{\uprho _{{\text{ma}}}}}}{{{\uprho _{\ker }} - {\uprho _{{\text{ma}}}}}}.$$
(4)

The \({\uprho _{{\text{ns}}}}\) is the density of non-source interval (average from log) in g/cm3, \({\uprho _{\text{s}}}\) is the density of source interval (from log) in g/cm3, \(~{\uprho _{{\text{ma}}}}\) is the assumed mud rock density, \({\uprho _{{\text{fl}}}}\) is the assumed water filled density and \({\uprho _{\ker }}\) is the assumed kerogen density. Finally, TOC in wt% can be estimated using the following relation:

$${\text{TOC}}=~\frac{{0.85~ \times ~{\uprho _{\ker }} - ~{ \upphi _{\ker }}}}{{{\uprho _{\ker }}~\left( {{ \upphi _{\ker }}} \right)+~{\uprho _{{\text{ma}}}}~\left( {1 - ~{ \upphi _{{\text{fl}}}} - ~{ \upphi _{\ker }}} \right)}}.$$
(5)

Seismic based TOC estimation

Since acoustic impedance depicts elastic behavior of a material, it therefore should provide information about variations in lithology, mineralogy, and fluid constituents etc. across the rock unit (Løseth et al. 2011). Moreover, the organic rich source rocks have low densities and acoustic impedances due to the high percentage of TOC compared to the non-source rock with a similar composition (Løseth et al. 2011). To evaluate the source rock potential of the Sembar Formation in Badin area, the zero phase band limited 3-D seismic data of the area is converted from amplitude to RAI using the relation given by Nissen et al. (2009)

$$\ln (\uprho V)=2\mathop \int \limits_{{t=0}}^{{t=T}} f\left( t \right)dt,$$
(6)

where \(f\left( t \right)\) is the band limited reflectivity series, \(\uprho\) is density and \(V\) is the compressional wave velocity within the rock layer. The curves of estimated TOC and acoustic impedance (AI) from wells are calibrated with the RAI section. Subsequently, horizon/reflection having low RAI and high TOC is identified throughout the seismic cube within the Sembar Formation with the help of calibrated AI and TOC curves derived from the well log data. After that, RAI is extracted along that interpreted horizons on a grid interval of 25 × 25 m. The choice of this grid interval is based on the fact that the bin size used in the processing of the 3-D seismic volume is 25 × 25 m and this grid interval is able to capture all the finer variations (in terms of RAI) within the data. Furthermore, RAI extracted in this way comes only from the rich TOC zone having low acoustic impedance within the Sembar Formation.

Similarly, for the sake of comparison, a particular horizon/reflection within the Sembar Formation having high acoustic impedance and low TOC is identified throughout the seismic cube with the help of AI and TOC from well log data, and RAI is extracted and gridded (25 × 25 m) along this horizon as discussed above.

In the next step, for the estimation of TOC from seismic data, a relationship describing the general trend between TOC (well derived property) and RAI (seismic derived property) is required. A relationship is established via a cross-plot between TOC and extracted RAI at the location of the wells for the high/low zones of TOC and RAI within the Sembar Formation, which is then used for the conversion of RAI at reservoir scale to TOC.

Results

In this research, we have devised a methodology for TOC estimation in the Sembar Formation in Badin area, Lower Indus Basin using 3-D seismic data together with well log data. The results are presented in the relevant sub headings below.

Spatial extent of the Sembar Formation

A synthetic seismogram is generated using DT and RHOB logs of Well A, and is correlated with the 3-D seismic data of Badin area with a correlation coefficient of R2 = 0.93 (Fig. 5). The top (1.6 s TWT) and bottom (1.94 s TWT) of the Sembar Formation is marked (Fig. 5) and the time structural maps from the top/bottom of the Sembar Formation are prepared using contour intervals of 0.01 and 0.05 s respectively (Figs. 6, 7). As the generated contour maps from both top/bottom of the Sembar Formation reflect the same deformational style (Figs. 6, 7), therefore, only top and bottom of the Sembar Formation is mapped. However, it’s not a usual case to map the top/bottom of the target zone only, if the structural complexity exists within the target zone (Bacon et al. 2007). For delineation of exploitation depth, the time structural map generated from the top of the Sembar Formation is converted using velocity derived from seismic and well data (Fig. 8). Furthermore, the time and depth contour maps show the spatial variation of Sembar Formation along with its deformational style. The thickness of Sembar Formation is interpreted to be ≈ 366–546 m based on the isopach grid map (Fig. 9), starting from the depth of around 2200 m with reference to the datum surface in the study area (Fig. 8). It is further inferred from the contour maps that the depth of Sembar Formation increases from east to west with normal faulting resulting in horst and graben structural patterns. Due to the depth and thickness of the Sembar Formation, it has high temperature and overburden pressure, which makes it an ideal source rock within the context of shale gas exploration.

Fig. 5
figure 5

Superimposing synthetic seismogram over the seismic section of in-line 5821 along with GR log of well A. GR log is used here for differentiating/demarcating shaley intervals over seismic section. The AI from well A is also displayed for calibration purpose. F1, F2 and F3 represent the faults having normal sense of shear

Full size image
Fig. 6
figure 6

Time contour map from the top of Sembar Formation with contour interval of 0.01 s. F1, F2, F3 and F4 represents faults traced over the base map. The color legend provides the information about shallowest and deepest parts of the Sembar Formation

Full size image
Fig. 7
figure 7

Time contour map from the bottom of the Sembar Formation with contour interval of 0.05 s. F1, F2, F3, F4 and F5 represents faults traced over the base map. The color legend provides the information about shallowest and deepest parts of the Sembar Formation

Full size image
Fig. 8
figure 8

Depth contour map from the top of Sembar Formation with contour interval of 15 m. F1, F2, F3 and F4 represent the faults traced over the base map. The color legend provides the information about shallowest and deepest parts with of the Sembar Formation

Full size image
Fig. 9
figure 9

Isopach map of the Sembar Formation showing its thickness variation. F1, F2, F3 and F4 represent the faults traced over the base map. The color legend provides the information about lateral variation in thickness of the Sembar Formation

Full size image

Logs based TOC estimation

The well log data of three wells (A, B and C) are used for TOC (wt%) estimation of Sembar Formation by using Passey ΔlogR and Schwarzkofp methods as discussed in Sections "Passey overlay technique and Schwarzkopf’s Technique" (Table 2; Figs. 10, 11 and 12). In the source rock interval, only the density log shows a decreasing trend, while the response of other logs such as GR, resistivity, DT and neutron shows an increasing trend, marking a rich TOC interval (Table 1; Passey et al. 1990, 2010). A baseline is established in all the three wells based on small or no separation between DT and resistivity logs. There is a 250–300 m interval at the center of the Sembar Formation having TOC values in the range of 1–4 wt% (Figs. 10, 11 and 12). Based on the estimated TOC within the Sembar Formation, the source rock interval can be designated as having TOC > 1 and the non-source interval as having TOC < 1.

Table 2 Estimated values of TOC from Passey’s and Schwarzkopf’s methods for all the three wells
Full size table
Fig. 10
figure 10

TOC values (wt%) estimated from Passey’s ΔlogR and Schwarzkopf’s methods in Well A (track 8 and 9). Track 1 is a lithological track, track 2 is a resistivity track and track 3 is a porosity track. The rest of the tracks represent the derived reservoir properties (shale volume, water saturation, porosities, and Delta LogR) that are used to calculate TOC (wt%) curves from both methods (tracks 8 and 9)

Full size image
Fig. 11
figure 11

TOC values (wt%) estimated from Passey’s ΔlogR and Schwarzkopf’s methods in Well B (track 8 and 9). Track 1 is a lithological track, track 2 is a resistivity track and track 3 is a porosity track. The rest of the tracks represent the derived reservoir properties (shale volume, water saturation, porosities, and Delta LogR) that are used to calculate TOC (wt%) curves from both methods (tracks 8 and 9)

Full size image
Fig. 12
figure 12

TOC values (wt%) estimated from Passey’s ΔlogR and Schwarzkopf’s methods in Well C (track 8 and 9). Track 1 is a lithological track, track 2 is a resistivity track and track 3 is a porosity track. The rest of the tracks represent the derived reservoir properties (shale volume, water saturation, porosities, and Delta LogR) that are used to calculate TOC (wt%) curves from both methods (tracks 8 and 9)

Full size image

Analysis of TOC from seismic data

To analyze TOC from seismic data, the amplitude information in the seismic cube is converted into RAI values using Eq. (6) and calibrated with the AI and TOC curves calculated from well log data (Fig. 13). The identified low and high TOC zones with the Sembar Formation are gridded in terms of RAI as discussed in Section "Seismic based TOC estimation". From the RAI maps, one can clearly see the distribution of low and high acoustic impedances at well locations in Figs. 14, 15, which can be related to high and low TOC. The estimation of TOC from the RAI converted maps needs a relationship in the form of cross-plot between logs derived TOC and seismic derived RAI. The cross-plot of TOC and RAI gives inverse relation with a correlation coefficient R2 = 0.72 (Fig. 16). In particular, the data in this cross-plot comes from both high/low TOC and RAI zones and the resulting linear relationship is used for the estimation of both high and low TOC zones within the Sembar Formation. In addition to that, we have used the estimated values of TOC from the Passey’s method in cross-plot analysis. The Passey’s method is used, as the estimated TOC values involve the incorporation of laboratory/core data along with well log data making it more reliable than Schwarzkopf’s method, which depends only on the well log data i.e. RHOB log. The Equation of the best fit line representing the general trend between TOC and RAI is given as

Fig. 13
figure 13

Interpreted seismic section of in-line 5941 with a display of GR (green), AI (red) and TOC (black) curves at the location of well B and C for calibration purpose. The red horizon is picked at maximum TOC (low RAI) interval, while the blue horizon is picked minimum TOC (high RAI) interval within the Sembar Formation

Full size image
Fig. 14
figure 14

RAI extracted at a 25 × 25 m grid along the maximum TOC zone of Sembar Formation. The map shows low RAI values at the location of wells A, B and C

Full size image
Fig. 15
figure 15

RAI extracted at a 25 × 25 m grid along the minimum TOC zone of Sembar Formation. The map shows high RAI values at the location of wells A, B and C

Full size image
Fig. 16
figure 16

A linear relationship established via a cross-plot between TOC (estimated from well data) and extracted RAI (from seismic) at the location of all the three wells. In particular, both high/low TOC and RAI zones within the Sembar Formation are used. The linear relationship established in the cross-plot is then used for the estimation of TOC from seismic based extracted RAI values

Full size image
$${\text{TOC}}= - {10^{ - 5}}\,{\text{RAI}}+1.7356.$$
(7)

Finally, TOC maps are generated from RAI maps using Eq. (7) (Fig. 17). The estimated TOC values from seismic data show good calibration with well log data and suggest that Sembar Formation has suitable potential within the shale gas perspective (Table 2; Fig. 17). Moreover, we have also shown an estimation of TOC using RAI extracted at a low TOC and high acoustic impedance zone (Fig. 18). By examining the behavior of RAI and TOC at well locations in Figs. 14, 15, 17 and 18, it can be observed that high and low TOC indeed manifests itself in terms of RAI anomalies. In particular, high TOC at well locations in Fig. 17 show low RAI in Fig. 14 and low TOC at well locations in Fig. 18 show high RAI in Fig. 15.

Fig. 17
figure 17

Seismic based TOC using Passey’s method for RAI extracted along high TOC and low acoustic impedance zone within the Sembar Formation. The map shows high TOC values at the locations of wells A, B and C

Full size image
Fig. 18
figure 18

Seismic based TOC using Passey’s method for RAI extracted along low TOC and high acoustic impedance zone within the Sembar Formation. The map shows low TOC values at the locations of wells A, B and C

Full size image

Discussions

Overall, the estimated TOC values for the Sembar Formation in this study ranges from 1 to 4 wt% showing that it lies in the range of good to excellent source rock (Peters et al. 1986). The values of TOC of the Sembar Formation analyzed by Wandrey et al. (2004) from laboratory samples range from 0.5 to 3.5 wt%. So the result of the TOC obtained in this work shows a good agreement with study conducted by Wandrey et al. (2004). It is further inferred that Kerogen of the Sembar Formation is mostly of Type-II and III in the Lower and Middle Indus Basin (Wandrey et al. 2004; Ahmad et al. 2011). The TOC values calculated from seismic data for the Sembar Formation increases from eastern to western side (Fig. 17). Thermally, the rock is much more mature to the western part of the shelf due to deep burial depth and less mature to the eastern part due to shallow burial depth (Fig. 8; Wandrey et al. 2004).

Our analysis shows that the Sembar Formation can act as a good potential unconventional resource because it is much mature towards higher concentration of TOC (Fig. 17). The reason for the high concentration of TOC is the paleo-geographic location of the Sembar Formation during its depositional time. The age and depositional environment as discussed in Geologic settings suggest that during the late Jurrasic and Lower Cretaceous, the Indian plate drifted northward, entering to warmer latitude (Scotese et al. 1988). On its lower shelf, marine shales of the Lower Cretaceous Sembar Formation were deposited over Jurrasic age Chiltan Limestone. During this age, the paleo-geographic latitude of the Indian plate was at location of nearly Latitude 20°S and Longitude 68°E (Scotese et al. 1988). The higher concentration of TOC suggests that the existed paleo-environment at that time was suitable for high productivity rate of organic matter. Furthermore, it can also be further inferred that, it was the time when the decomposition rate was slow (means an anoxic environment suitable for preservation of the organic matter existed) as compared to the productivity rate of the organic matter resulting in such high concentration of TOC. The rich TOC potential of the Sembar Formation makes it a potential source rock and as well as a potential unconventional play.

For further evaluation of the unconventional potential of a particular unit, the depth of exploitation and its thickness is an essential target. In platform areas, such as the Lower and Middle Indus Basins, the general economic exploitation depths for shale gas are about 3500 m, while that in tectonically unstable fold-belts, it ranges from 1000 to 3000 m (Ahmad et al. 2011). The exploitation depth of Sembar Formation obtained for our study starts from around 2200 m (Fig. 8) having considerable thickness (Fig. 9), making it a feasible and suitable candidate for unconventional resource exploitation with higher concentration of TOC (Fig. 17).

The analogues of Sembar Formation are present in the Rajasthan Basin (eastern side of the study area) of the Indian subcontinent and in the Zagros fold-belt of Iran (western side of the study area) (Klett et al. 2011; Opera et al. 2013). These analogues units are the most productive source rocks of their respective basins within the context of shale gas exploitation (Klett et al. 2011; Opera et al. 2013). The values of the TOC in the Gurpi and Kazdumi Formation of Zagros fold-belt of Iran ranges from 1.7 to 2.4 wt% and 1.5 to 3.92 wt% respectively (Opera et al. 2013). The source potential of the Ghaggar–Hakra Formation of Rajasthan Basin is poorly known due to limited number of well penetrations (Farrimond et al. 2015). We expect that this methodology may be helpful in exploration of unconventional resources for such basins worldwide.

Conclusions

The Sembar Formation is an ideal shale gas prospect of the Lower Indus Basin in terms of depositional environment, thickness, organic geochemistry, thermal maturity, mineralogy and porosity. One of the key features for successful shale gas plays requires high TOC. In this study, we have devised a methodology for estimation of TOC from seismic and well log data for the Sembar Formation in Badin area, Lower Indus Basin, Pakistan. The extracted maps of RAI and estimated TOC (1–4%) within the Sembar Formation are evaluated on the reservoir scale with a good correlation coefficient. The estimated TOC values for Sembar Formation show that it lies in the range of good to excellent potential shale oil/gas reservoir. Our results suggest that high and low TOC values indeed manifest themselves in terms of RAI anomalies. Moreover, the results also suggest an excellent calibration of TOC values from seismic at well locations. We anticipate that this methodology can be helpful in the initial assessment and identification of optimum sweet spots (high TOC) for shale gas plays worldwide.