Sedimentological and facies characterization in the modelling of the Turonian sandstone reservoir package, Jubilee field, Ghana

Abstract

Assessment of conditions that favoured hydrocarbon accumulations necessitates sedimentological characterization and facies identification as it elucidates the detailed records of the depositional environment and its evolution. Integration of various data sources becomes imperative for the establishment of the facies architecture which enables the evaluation of reservoir quality for Jubilee field in the deep Tano Basin. Early stage realization of decrease in production prediction of the field was attributed to early water breakthrough, baffle bridges and reservoir heterogeneity. This development requires its sedimentological evaluation through textural characterization, lithofacies identification and reservoir quality indication. The identified seventeen lithofacies allowed the deduction of deep-marine turbidite depositional environment. Our findings highlight the significance of sedimentological and facies development processes in the overall evolution of deep-water channel systems. The history of the Jubilee channel complex sets almost definitely characterizes many other emerging undersea channel systems and can be used to evaluate the distribution, shape, and structuring of the resulting reservoir sands. In turn, the Jubilee data set adds to the library of channel-levee-splay-lobe systems, aids as a prospective predictive model for hydrocarbon exploration and development, and as an outcrop explanatory analogue of a channel-splay system.

Introduction

Sedimentological and facies characterization is critical for assessing the prerequisites that favoured hydrocarbon accumulation. The establishment of facies distribution in reservoir architecture aids in perceiving reservoir heterogeneity for hydrocarbon field development. The studying of rock outcrops can provide full knowledge of the facies architecture and their quality (Martinius et al. 2013; Siddiqui et al. 2019, 2020). In the absence of outcrops, core sample, Gamma ray (GR) pattern and 3-D seismic volume are critical for understanding reservoir sandstone geometry, architecture, and quality, which are essential components for hydrocarbon field development and exploitation (Donselaar and Schmidt 2005; Farrell et al. 2013; Lowe et al. 2019; Radwan 2021; Omisore et al. 2022). Many researchers have used outcrops, core samples, and well logs to better understand sedimentary facies architecture at the regional and subsurface levels to estimate reservoir properties (e.g., Shukla 2009; Tullius et al. 2014; Siddiqui et al. 2020; Ali et al. 2021). The analysis of facies and establishing their paleoenvironment can also be based on descriptions of core samples in terms of sedimentary textures and elemental composition (e.g., Martinius et al. 2013; El-Tehiwy et al. 2019; Huang et al. 2020; Solórzano et al. 2021). The combination of these tools (outcrop, core data, well log, 3D seismic volume) can provide crucial insights into sand body thickness, shape, particle size distribution, lithology, and depositional environment. Integration of these tools has been used in sedimentological and facies modelling (e.g., Islam et al. 2017; Al-Mimar et al. 2018; Pacheco et al. 2019; Pattison 2019; Shahbazi et al. 2020; Ahmed 2021; Xu et al. 2022;) to understand reservoir sand distribution and its reservoir quality.

The Turonian clastics serve as the main reservoir in the Jubilee field, Ghana (Fig. 1) which is one of the most important producing areas within the Deep Tano contract Area. The Jubilee field in its early stage of production saw a decrease in production prediction, which was attributed to early water breakthrough, baffle bridges, increasing gas–oil ratio, inadequate pressure support and reservoir heterogeneity. Due to these factors, studies have described the stratigraphic blueprint for the Basin (Atta-Peters et al. 2013; Rüpke et al. 2010; Bempong et al. 2021a, b) to help address some of these problems but neither of the authors focused on the sedimentological and facies characterization modelling of the Jubilee reservoir. The current study, therefore, is an assessment of the reservoir characterization through textural analysis, facies interpretation and depositional modelling of the Jubilee Turonian reservoir.

Fig. 1

Location map of the study area showing the wells used for the studies

The ensuing proposed depositional model will provide insight into the fields paleogeographic evolution, while the textural assessment of the reservoir architecture will aid in the evaluation of the flow behaviour. To achieve this purpose, the objective of the study is to (1) characterize the textural relationship of grain size distribution and their genetic significance, (2) create a lithofacies identification from core image, and (3) use a high-resolution 3D seismic volume of the subsurface to elegantly assist in revealing the depositional architecture of the reservoir. The outcome of the study will not only serve as a link within the Jubilee contract area but also by extrapolations be useful in the other parts of the basin thereby reducing the uncertainty of field development.

Geological setting

The study area is located south of the Deepwater Tano Basin (Fig. 1). Tano Basin is a coastal basin in Ghana's southwestern region that is part of the West African Transform Margin basin family. It is bounded by a greater portion offshore and a lesser portion onshore. The basin is bounded to the north by the African mainland, to the west and east by St Paul and Romanche transform zones, and to the south by the ocean. It evolved as a pull-apart basin (Atta-Peters et al. 2013) as a result of continental rifting during the Mesozoic Era. Subsequent extension and subsidence of the rift system's thinned continental crust and created space for lithic fill into the basin as the South Atlantic Ocean opened up and deepened (Bempong et al. 2021a). Wrench tectonics in the Cretaceous, on the other hand, alters the basin through a large-scale transform fault associated with sagging and fracture (Rüpke et al. 2010). The basin's lithic fill began in the Aptian, with significant continental clastic from rivers Tano and Ankroba (Bempong et al. 2021a). According to Dailly et al. (2013), the submarine canyon system served as a clastic sediment delivery system into the Tano Basin. The relationship between tectonic phases and associated sedimentation is defined by four patterns as enumerated below (Fig. 2):

• 1. The pre-rift phase of a stable platform was identified as Paleozoic to lower Mesozoic deposits.

• 2. The syn-rift phase, which began in the Jurassic and progressed to the Lower Cretaceous, was characterized by a continental rifting between Africa and South America. The drifting forced the continental margin into a transpressional setting, which was recorded in the basin by a series of folds and flow structures. These records indicate that the continental margins played a role in the overall shearing between the two plates.

• 3. The top syn-rift unconformity, denotes the end of a period of intense continental rifting and deformation. It is distinguished by a noticeable angular unconformity across the continental platform, as evidenced by a Top Syn-rift reflection on the seismic profile. The unconformity corresponds to the last continental contact between the two plates, which occurred near the end of Albian or early Cenomanian. According to Antobreh et al. (2009), the episode of thermal uplifting of the outer Ghana margins at the end of the Albian is plausible to coincide with magnetic underplating reported in deep seismic gravity and magnetic records across the margin Rüpke et al. (2010).

• 4. The post-rift phase denotes marine sedimentation from the Cenomanian to the recent, which occurred following the final breakup of the American and African plates. Full-scale ocean transform faults (St Paul and Romanche transform) developed along the trace of the older continental transforms during this phase (Fig. 2). Although some deformation may have continued into the continental margins, the final break up generally marks a period of rapid subsidence and thermal subsidence.

Fig. 2

Paleogeological setting during the tectonic phase of the breakup of Africa and South America (Rüpke et al., 2010). a Early Cretaceous 125 Ma, continental rifting and transform faulting initiation, b Aptian 110 Ma, rapidly forming isolated rift basins, c latest Albian 100 Ma, final contact between Africa and America, and d Upper Cretaceous, fully developed oceanic basin in the central and south Atlantic

Reservoir zonation

The Jubilee field has two main reservoir intervals: the Upper and Lower Mahogany. The intervals are separated by extensive regional shale section (Table 1). The lower Mahogany is classified into LM4 to LM1 in succession and overlain by shale, while the Upper Mahogany is classified as UM4 to UM1 in succession. Of all these successions of sandstone bodies, the volumetrically significant are LM2A and LM2B which are denoted as MH1 (seismic stratigraphic section), as well as UM3A to C, denoted as MH4 (on the seismic stratigraphic section), as indicated in Fig. 3. The two reservoirs were considered with several producers and injector wells in the initial development program.

Table 1 Turonian reservoir zonation of the Jubilee field

Fig. 3

Seismic depth section of MH1(green) and MH4 (white) reservoir section through selected well of the Jubilee field

Materials and methods

The study derives from exploration, appraisal and development wells from the Jubilee field, Ghana that penetrated through the Turonian reservoir (from LM3 to MH5 of the Lower and Upper Mahogany). A total number of six well log suits is involved and are denoted as W1 to W6. Three of the wells have core data (i.e., W1 to W3), while W2 and W3 are employed for petrographic assessment. Since this study is focused on the volumetrically significant reservoirs of MH1 and MH4, penetrations of wells into other reservoirs only serve as complementary information for improving the knowledge of sand quality and fluid distribution.

The workflow for this study includes petrographic observation, lithotype/lithofacies determination, depositional package identification and modelling of the depositional setting (Fig. 4). Core images are used for lithofacies identification and reservoir quality assessment. Gamma ray log is eventually employed for facies correlation and upscaling the lithofacies to bed-stack facies association. Sieve analysis data allowed for the determination of textural characteristics using statistical grain size parameters based on calculations of Ghaznavi et al. (2019). Bivariate plots of the textural parameters as well as Linear Discriminate functions (Eqs. 1–4) (Ghaznavi et al. 2019) are used to interpret the depositional environment with the equation below;

$$Z_{{\text{1Aeol:Beach}}} = - 35688M + 3.7013{\text{SD}}^{2} - 2.0766SK + 3.1135K$$

(1)

Fig. 4

Flowchart adopted for the work

For a beach subenvironment (B), Z₁ > − 2.741, while Z₁ < -− 2.7411 represent aeolin environment (Ae)

$$Z_{{\text{2Beach:Shallowmarine}}} = 15.653M + 65.7091{\text{SD}}^{2} - 18.1071{\text{SK}} + 18.5043K$$

(2)

Z₂ < 63.3650 for the beach subenvironment (B), while Z₂ > 63.3650 for the shallow-marine subenvironment (SM).

$$Z_{{\text{3Shallowmarine:Fluvial}}} = 0.2852M - 8.7604{\text{SD}}^{2} - 4.8932{\text{SK}} + 0.0428K$$

(3)

Z₃ > − 7.4190 for the shallow-marine subenvironment (SM), whereas Z₃ < − 7.419 for the Fluvial subenvironment (F).

$$Z_{{\text{4Fluvial:Turbidity}}} = 0.7215M + 0.403{\text{SD}}^{2} + 6.7322{\text{SK}} + 5.2927K$$

(4)

Z₄ > 10 denotes the marine turbidity subenvironment (T), while Z₄ < 10 represents the fluvial subenvironment (F).

Results

Petrographic properties of the sandstone samples

Frequency curve and histogram

Peak constituents and consistency of sediment fractions for the study area are depicted by frequency distribution curves from the two wells (W2 and W3). Figure 5 represents the assemblage of the descriptive statistical curves for the two wells. The frequency distribution on the frequency curves is unimodal (Fig. 5a–g) and bimodal (Fig. 5h). The curves' unimodality is due to less variation in the depositional velocity and a constant depositional process over which sediment settled (Boggs 2012). The assertive peak in the bimodal distribution was at 3 phi, and the passive peak was at 7 phi (Fig. 5h). The fine and silt components of the bimodal grain-size distribution could be attributed to several depositional processes and/or mixing of different size populations from the source area. Bimodality can also be attributed to the low energy of the transporting medium and the difference in modes of transportation. The histogram (Fig. 5a–h) revealed a range of very fine to medium sand, with a low percentage of the very coarse-grained sand present only in conglomerate samples.

Fig. 5

Grain size distribution and cumulative frequency curve for some sandstone facies for the Turonian reservoir unit; a unimodal histogram for Sd; b unimodal histogram for Ss; c unimodal histogram for Sc; d unimodal histogram for Sxr; e unimodal histogram for Sp; f unimodal histogram for Sm; g unimodal histogram for Cgl and h bimodal histogram for Sc

Furthermore, phi values are plotted against cumulative frequencies, indicating an impact of different mechanisms of sediment supply and deposition in the formation of sandstone units. When plotted on an arithmetic scale, the curve typically exhibits an S-shaped trend (Fig. 5–h). Results for the cumulative percentage curve indicate many sediment transport and deposition were attributed to saltation, passive suspension, and minor traction (Fig. 5a–h). The slope of the middle portion of the curve can predict sorting (Ghaznavi et al. 2019). A broad and gentle slope indicates low kinetic energy and velocity, indicating poor sorting (Fig. 5a, c and h). A very steep slope, on the other hand, indicates good sorting (Fig. 5d–g). On the cumulative percentage plot, the value of phi ranges from 0 to 4 phi, with most curves exhibiting an almost identical broad and gentle slope S-shaped trend (Fig. 5). As a result, they can be reliably classified as fine to medium-grained (e.g., Spencer 1963; Ghaznavi et al. 2019; Siddiqui et al. 2020). The gentle slope indicates that such grains have been very poorly sorted (Fig. 5).

Statistical parameters—graphical method

Graphical mean: the graphical mean ranges from 1.12 to 10.34, with an average of 3.53 (Table 2). The average mean of the reservoir unit denotes fine-grained sand sediments. The distinct nature of the size distribution with fine sand may be attributed to the low-energy environment that existed during deposition (Ghaznavi et al. 2019). The values of certain sand samples, such as Sb, Sf, Sm, and Sxr, are quite high due to the dominance of a particular mode of sediment class (Fig. 6a). However, a few sample values, such as Cgl, Sc, and Sd, are low due to a nearly equal proportion of coarse and fine sediments, as indicated in Fig. 6a and Table 2. It is a poorly sorted type due to the wide variation in grain size.

Table 2 Graphical statistical parameters of the grain size distribution for the Mahogany reservoir sandstones of the Jubilee field, calculated in phi units

Fig. 6

Histograms of all samples plotted with respect to statistical parameters calculated by the graphical method: a mean grain size; b standard deviation; c skewness and d kurtosis

Phi standard deviation: the kinetic energy variation associated with the mode of deposition concerning the particle size distribution is represented by the standard deviation. The standard deviation is an important parameter in sediment analysis, because it reflects the energy situation of the depositional environment (Ghaznavi et al. 2019), but it does not primarily assess the quality to which the sediment has been blended. The reservoir units' phi standard deviation ranges from 1.03 to 6.58 phi, with an average of 2.50 phi, indicating that the sediments are very poorly sorted (Fig. 6b, Table 2). The poorly sorted nature of the sediment depicts a blending of sediments of various sizes. The poorly sorted samples indicate that the current velocity of the transporting media varies slightly (Fig. 6b).

Graphic skewness: skewness is a measure of sediment distribution symmetry. Skewness is useful in environmental forecasting, because it is proportional to the fine and coarse tails of particle size and thus indicative of deposition energy. For the sand samples, the skewness values range from 0.26 to 0.57 phi, with an average skewness of 0.35 phi indicating finely skewed or positive skewness (Fig. 6c, Table 2). The positive skewness suggests that population growth is taking place in a low-energy environment (Fig. 6c). The fine tail of the sediments contains a higher concentration of material.

Graphic kurtosis: the samples peak from 1.31 to 2.74, with an average of 1.62 (Table 2). The vast majority of grains are leptokurtic, with a few notable exceptions that are extremely leptokurtic (Fig. 6d). This demonstrates that the tails portion is usually sorted equally.

Statistical parameters—moment measures method

1st moment—mean: the graphic mean ranges from 0.97 to 2.17, with an average of 1.47 (Table 3), indicating medium grain size. Only two samples, Sx and Sxr, are coarse and fine-grained, respectively; the rest have a medium-grained texture (Fig. 7a).

Table 3 Graphical statistical parameters of the grain size distribution for the Mahogany reservoir sandstones of the Jubilee field, calculated by the moment method

Fig. 7

Histograms of all samples plotted with respect to statistical parameters calculated by the moment method: A mean grain size; B standard deviation; C skewness and D kurtosis

2nd moment—standard deviation: this metric measures grain sorting or uniformity, indicating the energy conditions that existed during transport and deposition. Its value ranges from 0.76 to 1.18, with a mean of 1.05 (Table 3). Ultimately, the samples show poor sediment sorting, with only two samples, Sm and Sx, showing moderate sorting (Fig. 7b).

3rd moment—skewness: the skewness ranges from 0.72 to 2.45, with an average of 1.89 (Table 3). All samples are very finely skewed (Fig. 7c), indicating a population accumulation in a low-energy environment.

4th moment—kurtosis: Kurtosis values range from 3.78 to 10.52, with an average of 6.61. All of the samples are of the leptokurtic type (Fig. 7d), with equal sorting in the tails.

Bivariate grain size parameter

Multiple bivariate analysis seeks to evaluate a wide range of data analyses, which aids in clearly determining the depositional environment (Ghaznavi et al. 2019). According to Moiola and Weiser (1968) and Azidane et al. (2021), bivariate is most effective in distinguishing between the beach and river sands, as well as river and coastal dune sands, and the distinction works well whether quarter, half or whole phi data are used. Bivariate plots are used by Friedman (1961), Moiola and Weiser (1968) and Tan et al. (2019) to distinguish between the river, coastal dune, and beach subenvironments. Therefore, to confirm the current paleoenvironmental conditions, statistical properties derived using both graphical and moment techniques were plotted in various bivariate diagrams. To distinguish between the beach and river sedimentary subenvironments, Friedman (1961, 1967), Rajganapathi et al. (2013) and Su et al. (2021) developed a bivariate by plotting skewness vs. standard deviation. The sample points are all drawn to the river environment (Fig. 8a). The bivariate plot between kurtosis and skewness (e.g., Folk and Ward 1957; Rajganapathi et al. 2013 and Ghaznavi et al. 2019) was used to further distinguish between coastal areas, dune environments, and fluvial environments which confirmed that of Friedman (1961, 1967) bivariate plot. In the case of both the graphical and moment method, all samples represent that of a fluvial environment (Fig. 8b). The bivariate plot reveals that all of the samples are from the fluvial subenvironment (Fig. 8a, b).

Fig. 8

a Bivariate plot of skewness against inclusive graphic standard deviation (after Friedman, 1967) and b bivariate plot of skewness against kurtosis (after Folk & Ward, 1957)

Linear discrimination function analysis

Based on the results of the linear discriminate function using statistical parameters obtained by the graphical method, it was discovered that Z₁ indicated that all samples belonged to the beach subenvironment (Table 4 and Fig. 9a). The Z₂ values indicate that the sediments are shallow-marine (Table 4 and Fig. 9b). In a comparison of fluvial and shallow-marine dominance using Z₃, the former prevailed over the latter (Table 4 and Fig. 9c). Z₄ shows that the sediments were deposited primarily by fluvial action, with turbidity action being a component of both the turbidity and fluvial processes (Table 4 and Fig. 9c).

Table 4 Linear discriminate function analysis to interpret variation in energy and fluidity factors using the graphical method

Fig. 9

Linear discriminate function plot for Turonian sandstones. a Z₁ vs. Z₂ discriminates between the beach and aeolian environment; b Z₂ vs. Z₃ discriminates between the beach and shallow-marine subenvironment; c Z₃ vs. Z₄ discriminates between marine turbidity and fluvial environment

When the statistical parameters were used, there was no difference between the LDF calculated using the moment method and the LDF calculated using the graphical method.

Thin section analysis

The petrographic analysis of the conventional core samples from the three wells is presented in this section. The following are the photomicrograph and description details for the reservoir sandstones:

Burrowed sandstone (Sb)

Sb is constituent by abundant monocrystalline quartz (white) framework grain, with less plagioclase (PF) and polycrystalline quartz (PQ) (Fig. 10a). The most common type of matrix contains only a small amount of syntaxial quartz overgrowths (qo). This poorly sorted sandstone has a medium grain size. Due to trace amounts of authigenic clays in this sandstone, the intergranular pores (Pi) are large and appear to be well-interconnected (Fig. 10b).

Fig. 10

a, b Photomicrograph of burrowed sandstone sample at 1 mm and 0.2 mm, respectively; c, d SEM photomicrography at 100 µm and 10 µm, respectively

Scanning electron microscopy (SEM) analysis revealed the presence of qo, which reduces intergranular pore sizes in this sandstone (Fig. 10c). Patches of localized authigenic kaolinite (kao) were also present, but no authigenic chlorite was detected (Fig. 10d).

Pebbly sandstone (Sc)

As shown in Fig. 11a, the maximum framework of this poorly sorted coarse-grained sandstone is composed of monocrystalline quartz (white) and polycrystalline quartz (PQ; also, white). As shown in Fig. 11b, this sandstone contains minor amounts of qo. Some authigenic chlorite (cl) linings and trace amounts of kao are present in isolated patches. In this sandstone, Pi appears to be well-interconnected. This cl lines the intergranular pore in this sandstone, according to SEM analysis (Fig. 11c).

Fig. 11

a, b Photomicrograph of pebbly sandstone sample at 1 mm and 0.2 mm, respectively; c, d SEM photomicrography at 100 µm and 10 µm, respectively

Figure 11d depicts minor amounts of discontinuous qo coating a detrital quartz grain. This sandstone also contains trace fibres of authigenic illite (I) (Fig. 11d).

Fine-grained sandstone (Sf)

Sf is primarily composed of monocrystalline quartz (mostly white grains) and minor amounts of plagioclase (PF). On quartz grains, moderate amounts of qo cement are observed, with kao being the most common authigenic clay recorded (Fig. 12a). As shown in Fig. 12b, three types of macropores have been identified: intergranular, secondary intragranular, and grain moldic. Pi is the most numerous and appears to be well-interconnected. Secondary intragranular pores (Ps) are small, isolated pores found mostly in feldspars (Table 5).

Fig. 12

a, b Photomicrograph of fine-grained sandstone sample at 1 mm and 0.1 mm, respectively; c, d SEM photomicrography at 100 µm and 10 µm, respectively

Table 5 Linear discriminate function analysis to interpret variation in energy and fluidity factors using the moment method

In the SEM analysis, intergranular pores are highlighted in Fig. 12c. Figure 12d depicts an intergranular pore that has been reduced by qo. Intergranular areas are also occupied by booklets of kao. Authigenic chlorite traces coat portions of grains and appears to have formed before quartz cement.

Planar stratified sandstone (Sp)

The Sp is composed of medium-grained sandstone with poor grain size sorting. As shown in Fig. 13a, the framework grains are abundant monocrystalline quartz (white). Figure 13b depicts the authigenic minerals present, which include cl, qo, and patchy kao. Primary intergranular, grain-moldic (Pm), and intercrystal pores are the three most common pore types found in chlorite and kaolinite authigenic clays.

Fig. 13

a, b Photomicrograph of planar stratified sandstone sample at 1 mm and 0.2 mm, respectively; c, d SEM photomicrography at 100 µm and 10 µm, respectively

Because of the abundance of authigenic minerals, intergranular pores (red arrows in Fig. 13c) are relatively small. The high-magnification view (Fig. 13d) shows authigenic cl coating grains and qo that appear to have formed after the cl.

Structureless sandstone (Ss)

Granule-sized grains coexist with sand-sized grains in this sandstone. The majority of the grains are monocrystalline quartz (mostly white grains), with minor amounts of polycrystalline quartz (PQ) and plagioclase grains (PF) (Fig. 14a). Figure 14b depicts authigenic constituents, such as qo, kao, and cl, as well as trace amounts of authigenic pyrite (py). The most common macropore type observed is well-interconnected Pi.

Fig. 14

a, b Photomicrograph of structureless sandstone sample at 1 mm and 0.1 mm, respectively; c, d SEM photomicrography at 100 µm and 10 µm, respectively

Primary intergranular pores (red arrows) are visible in the photomicrograph as a result of the SEM analysis (Fig. 14c). Quartz overgrowth cement has reduced some of these pores. Pore-filling authigenic clays shown in the photomicrograph (Fig. 14d) include kao booklets and cl blades.

Conglomerate (Cgl)

This sandy conglomerate is poorly sorted. The majority of framework grains are monocrystalline (most white grains) and polycrystalline (PQ) quartz, with minor feldspar grains. Mollusk fragments (Mol, stained red) are present in trace amounts (Fig. 15a). The most noticeable authigenic component is qo cement; authigenic clays such as kao are minor to rare (Fig. 15b). The most common pore type is well-interconnected Pi.

Fig. 15

a, b Photomicrography of conglomerate sample at 1 mm and 0.1 mm, respectively; c, d SEM photomicrography at 100 µm and 10 µm, respectively

Intergranular pores are highlighted in Fig. 15c as a result of the SEM analysis (arrows). Figure 15d shows qo, kao, cl, and py as the authigenic constituents present in this sample.

Ripple cross-laminated sandstone (Sxr)

Even though quartz constitutes the majority of the framework grains in these facies, the proportion of plagioclase (PF) in this sample is higher than in shallower samples (Fig. 16a). Quartz overgrowths are the most common intergranular cement, with minor amounts of authigenic kao and trace amounts of chlorite also filling some pores (Fig. 16b). Primary intergranular pores are the most common pore type; secondary intragranular pores and grain-moldic pores are minor.

Fig. 16

a, b Photomicrography of ripple-cross laminated sandstone sample at 1 mm and 0.2 mm, respectively; c, d SEM photomicrography at 100 µm and 10 µm, respectively

Because of the average grain size and qo cement, the intergranular pores (arrows) present are relatively small, as shown in Fig. 16c. Figure 16d depicts a minor patch of authigenic kao and trace amounts of bladed grain-coating cl partially filling some intergranular areas.

Sedimentary facies

The cored succession comprises thick intervals of sandstones and conglomerates interbedded with intervals of mudstones and heteroliths. These lithologies have been divided into sixteen depositional lithofacies, based on grain size, lithology, depositional and synsedimentary deformational structures. Detailed descriptions of the characteristic features of the lithofacies are discussed below and summarized in Table 6 with a coding scheme adopted by Ghana National Petroleum Commission (GNPC). The bed-stack facies for the cored wells are presented in Fig. 23.

Table 6 Core lithofacies scheme for the Wells

Facies A: conglomerate

The conglomerate unit observed in the core units composed more than 50% clasts with a mud-poor sandy matrix. The conglomerate facies observed can be divided into two (2) subfacies. These are identified as: (1) conglomerate composed largely of extrabasinal clast (Fig. 17a) and (2) Conglomerate composed largely of mudstone clast (Fig. 17a).

Fig. 17

Conglomerate lithofacies. a Extrabasinal-clast conglomerate occurring as a single layer and a mud-cast conglomerate observed at 4037–4037.28 m and 4046.8–4047.75 m, respectfully, in W1, b thin units of mud-clast conglomerate form numerous beds observed (3201–3202 m) in W3. c Sedimentological log of the conglomerate facies for W1

Extrabasinal clast-conglomerate is composed of 40–60% granules to small pebbles of lithified sedimentary, igneous, or metamorphic rock in a sandstone matrix (Fig. 17a). W1 contains several beds of the extrabasinal clast-conglomerate (Fig. 17a). Similar beds can be found in minor components of the W2 and W3 core.

The mud-clast conglomerate is composed of angular clasts of dark gray mudstone in a sandstone matrix (Fig. 17b). These units represent grains transported within sandy high-density turbidity currents. W3 contains several beds made up of granule- to small-pebble-sized mud clasts embedded in a sandstone matrix (Fig. 17b). Similar beds can be found in minor components of the W1 core. These units are typically thin (1 m) with small clasts. These have been included in either the thick-bedded sandstone or the interbedded thick-bedded sandstone and thin-bedded sandstone and mudstone lithofacies on the graphic logs (Fig. 17c).

Conglomerate occurs mostly within thick-bedded sandstone units in the cores and occurs as S3 divisions of Lowe (1982) deposited by the collapse of dense floating clasts and sand high-density currents. Current structured sandstone units are largely devoid of mud clasts, implying that they could not withstand abrasion and shearing during transport.

Facies B: sandstone

Facies B consists of more than 80% sandstone in beds that are mostly greater than 50 (cm) thick (Fig. 18). In general, the sandstone facies can be recognized as (1) Thick-bedded sandstone composed largely of sandstone beds deposited by high-density turbidity currents; structureless or showing water-escape structures (Fig. 18a); (2) Thick-bedded sandstone composed largely of current-structured sandstone deposited by low-density turbidity currents; dominated by Tb and Tc Bouma divisions (Fig. 18b); (3) Medium-bedded sandstone (Fig. 19), and (4) Thin bedded sandstone (Fig. 20).

Fig. 18

Thick-bedded, massive to dish-structured sandstones were observed in a W1 and b W3. The structural framework depicts deposit of the S3 divisions (Lowe 1982) by direct suspension sedimentation from high turbidity current. S3 divisions are formed as a result of a further fallout of suspended-load leading to sediments deposition directly through suspension sedimentation

Fig. 19

Medium-bedded sandstone intercalation with mudstone. Most involve various combination of the Bouma divisions: T_a, formed by direct suspension sedimentation; T_b, deposited by bed load sedimentation under high-velocity plane-bed conditions; T_c, deposited as bed load under low-velocity conditions with a rippled bed; T_d, deposited under very low-velocity condition; and T_e, silt and mud deposited after the flow has ceased. a Thick sandstone beds with cross lamination Bouma T_c division at the tops of beds and flat laminations T_b division occurring in W1. b Thick sandstone beds with cross lamination Bouma T_c division at the tops of beds and flat laminations T_b division occurring in W2

Fig. 20

Thin-bedded sandstone and mudstone intercalation. Most involve various combination of the Bouma divisions: T_a, formed by direct suspension sedimentation; T_b, deposited by bed load sedimentation under high-velocity plane-bed conditions; T_c, deposited as bed load under low-velocity conditions with a rippled bed; T_d, deposited under very low-velocity condition; and T_e, silt and mud deposited after the flow has ceased. a Fine-grained thin-bedded turbidites deposited by low-density turbidity current showing T_c, T_d, and T_e turbidite division for W1. b Fine-grained thin-bedded turbidites deposited by low-density turbidity current showing T_c and T_bc turbidite division for W2, and c thin-bedded turbidites deposited by low-density turbidity current showing T_c and T_bc turbidite division for W3

Facies B1 units are composed of amalgamated sandstone layers in which grading and well-defined sedimentation units may be absent to poorly developed, but most consist of stacked graded sedimentation units (Fig. 18a). Water-escape structures, especially dish structures form with this unit and are isolated vertical water-escape conduits with sets of subvertical water-escape sheets (Fig. 18a). These are similar to the S3 divisions of Lowe (1982) deposited by collapsing high-density turbidity currents.

Facies B2 unit comprises of many thick-bedded sedimentation units in the MH1 reservoir unit. They are composed of stacked, graded turbidites up to 150 cm thick made up largely of Bouma Tb divisions (Bouma 1962), which are deposited as a results of bed load sedimentation under high-velocity plane-bed conditions (Fig. 18b). These Tb divisions show alternating light quartzo-feldspathic laminations and dark laminations rich in mud grains and carbonaceous matter. They are in some instances underlain by thin massive Ta divisions and in most beds are overlain by thin Tcde divisions (Fig. 18b). These beds were deposited by direct suspension sedimentation (Ta) and low-density turbidity currents (Tc, Td, Te) but the flows may have been exceptionally thick and/or long-lived.

The medium-bedded sandstone and mudstone lithofacies are composed of 20–80% sandstone beds, the majority of which are 10–50 cm thick with sections of thin-bedded sandstone and mudstone (Fig. 19). Bouma T_abc divisions or combinations of these divisions are included in these beds (Fig. 19a, b). The bases are less erosive than those of sand beds in thick-bedded lithofacies, reflected in the common preservation of mud units representing T_de divisions between the sand beds (Fig. 19). These beds were deposited as a result of declining, late-stage high-density turbidity currents and successor energetic low-density turbidity currents.

Facies B4 is made up of more than 80% sandstone, which forms thin, very fine-to-medium-grained turbidites up to 10 cm thick (Fig. 20). Bouma T_b and T_c divisions dominate the thin sandy turbidites, which are separated by thin muddy laminations, most of which are 1 mm thick (Fig. 20a, b and c). Low-density turbidity currents deposited these beds, but the sparse development of capping T_de mudstone or post-flow hemipelagic mudstone divisions is unusual in the cores (Fig. 20). These low-density turbidites may be devoid of mud, or the majority of silt. In addition, the likelihood of mud bypassing and depositing father downslope may be likely. The lithofacies of thin-bedded sandstone and mudstone are well-developed in the WI (Fig. 7a). These are the dominant lithofacies in the out-of-channel levee and more distal splay deposits in most deep-water sequences.

Facies C: Mudstone

The mudstone lithofacies is made up of mudstone that contains 10–20% sandstone beds, the majority of which are fine-grained and less than 10 cm thick (Fig. 21). The deposition was generally accomplished through (1) Very low energy, mud-rich turbidity currents and, in some cases, hemipelagic sedimentation and (2) As a result of Mass transport deposits.

Fig. 21

Mudstone lithofacies were observed in all wells a W1, b W2 and c W3. Most of these mudstones are characterized by thin silty or sandy interbeds and show evidence of deposition by a very fine, low-density turbidity current. Most involve various combination of the Bouma divisions: T_a, formed by direct suspension sedimentation; T_b, deposited by bed load sedimentation under high-velocity plane-bed conditions; T_c, deposited as bed load under low-velocity conditions with a rippled bed; T_d, deposited under very low-velocity condition; and T_e, silt and mud deposited after the flow has ceased

Most of the mudstone in the Upper Mahogany interval is turbiditic mudstone, representing the deposits of low-density, very low-energy turbidity currents (Fig. 21a, b). The deposition was generally by very low energy, mud-rich turbidity currents. In the cored mudstone section representing MH3 and in much of MH1 (Fig. 21c), mudstone is mainly fissile mudstone with few or no associated turbidites (Fig. 21). This mudstone may represent hemipelagic sedimentation with burrowing ranges from absent to intense.

Many mass transport units are largely composed of mudstone, and many of these mudstone units exhibit evidence of local mixing and recumbent folding. The majority of the mudstones appear to have been deposited as very loose, low-strength bottom sediments that were subject to widespread failure and downslope movement (Fig. 21a, b and c). In addition, units representing debris flows, slides and slide blocks, and slumps are among them. These are primarily muddy units, owing to the abundance of mud on the seafloor and the presence of significant local slopes (Fig. 22a). Many are transitional into mudstone sections and are made up of mudstone chunks and blocks in a homogenized mud matrix. In other sections, more heterogeneous sections failed and the mass transport units are mixtures of mudstone clasts and thin-bedded mudstone and sandstone clasts and blocks in a matrix of sandy mudstone (Fig. 22b, c and d). It was difficult to determine whether mudstone units associated with mass transport deposits were intact mudstone layers or mudstone blocks within the mass transport complexes in some cases. Slide or displaced blocks have tilted stratification, internal disruption, common micro faulting of internal laminations (Fig. 22b), and/or shearing along with their contacts with adjacent units (Fig. 22c).

Fig. 22

Mass transport deposits observed. a Sedimentological log for the mass transport deposit. b Feature of thin sandstone beds completely disrupted and sheared within mass transport deposit observed in W1. c Sheared and folded thin-bedded mudstone and sandstone observed in W1 and d sheared and folded thin-bedded mudstone and sandstone observed in W2 at different depths

Reservoir quality

In terms of reservoir quality sandstone, logarithmic cross-plots of porosity vs. permeability derived from data sets of each sandstone facies for all wells (W1, W2 and W3) were analyzed (Fig. 24). The logarithmic cross-plot, which is colour coded by facies type, does not precisely separate the sandstone facies, but it does indicate reservoir quality (Fig. 24). Based on these values, a high quality (> 10% porosity, > 10 mD permeability), moderate quality (> 10% porosity, 10 mD permeability), and low quality (porosity 5% and permeability 0.01 mD) reservoir sandstone facies class were identified. A significant number of data points are less than 10% porosity and 10 mD permeability. These are typically well-cemented layers with pervasive calcite cement that has occluded microporosity (Figs. 11c, 12c, 13c, 14c, 15c and 16c).

The low-quality sandstone represents sandstone samples that have been extensively cemented by calcite, resulting in occluded porosity and reduced reservoir potential (Figs. 11, 12, 13 and 15). This sandstone contains bioclasts, and the calcite may have been mined locally (Fig. 23). This is a diagenetic reservoir quality control as indicated in Figs. 10c, 11c, 12c, 13c, 14c, 15c, and 16c.

Fig. 23

Gamma ray upscaling of lithofacies to bed stack facies for well a W1, b W2 and c W3

A grey filling (Fig. 24) delineates moderate reservoir quality samples. These are primarily argillaceous sandstone (Sa), bioturbated sandstone (Sb), fine-grained sandstone (Sf), and weakly stratified sandstone (Sm). Reservoir quality is likely to be poor in these lithofacies. The porosity measured in these samples is most likely microporosity.

Fig. 24

Core analysis of porosity and permeability cross plot coded by sand facies

Pebbly (Sc), planar laminated (Sp), structureless (Ss), dewatered/deformed sandstone (Sd), and conglomerates (Cgl) appear to dominate the high-quality reservoir sandstones partitioning (black circle, Fig. 24). Sc, Ss, Sp, and Cgl samples have higher permeability values than bioturbated sandstone (Sb) and ripple cross-laminated sandstone (Sxr) types for any given porosity value (from about 10% to about 25%, as shown in Fig. 24). The majority of the sandstone samples have good reservoir quality, and many of them have permeabilities greater than 100 mD (Fig. 24). These findings suggest that reservoir quality is primarily governed by lithofacies type, with diagenetic controls also playing a vital role. Due to the thin and rare nature of these lithofacies, no core analysis data exists in siltstone (Stn) core lithofacies.

Discussion

Depositional model of the Turonian Reservoir

The 3D seismic volume (Fig. 3), Sum of negative Amplitude (SNA) maps (Figs. 25 and 26), and cores from wells W1, W2 and W3 documents the evolution of the MH1 and MH4 reservoir system. In general, two main evolutionary stages can be identified: (1) pre-channel events, and (2) formation and evolution of channel complex.

Fig. 25

Sum of negative amplitude (SNA) extraction showing the evolution of the complex turbidite system of the MH1 reservoir system. a Interval at the base of the MH1 unit marking an erosional surface of possible levee which is marked by abundant heteroliths and mudstones in well W1 (Fig. 21a) and mudstones in W3 (Fig. 21c). b Lower section of the reservoir unit representing the development of a central splay complex with the Jubilee contract area. The yellow polygon shows the channel system. c Interval of the MH1 indicating the development of a splay (dark blue) and crevasse (gold) system in the central to the southeastern section of the Jubilee contract area. Multiple spays are active in the central section and shows the beginning of the development of meanders of the channel complex (black line). d Upper section of the MH1 reservoir indicating a possible flooding surface with few dispersed sand bodies to the northern and southern section of the Jubilee contract area

Fig. 26

Sum of Negative Amplitude (SNA) extraction showing the evolution of the complex turbidite system of the MH4 reservoir system. a Interval at the base of the MH4 unit marking a flooding surface with predominate mud chips from core (Fig. 21b). b Lower section of the reservoir unit representing the development of a south-southeast branch of the splay complex. The black arrow-line shows the path of the channel system. c Interval of the upper section of the MH4 indicating the development of a single splay system to the Southwestern section of the Jubilee contract area. d Upper section of the MH4 reservoir indicating the well-defined southwestern splay development system

MH1/MH4 Pre-channel event

The lowest symmetric cycle at the base of MH1/MH4 in core is too small to be resolved unambiguously using seismic data (Figs. 25a and 26a). In the pre-channel event, the site of the channel complex was primarily covered by mudstone containing thin sandstone interbeds representing fine-grained muddy turbidites that have widely suffered post-depositional movement and flowage to form mass transport deposits (Figs. 25a, 21a, c).

Formation and evolution of the MH1/MH4 channel complex

MH1 channel complex

The Lower Mahogany interval is characterized by cored intervals from wells W1, W2, and W3 representing the lower fan channel complex (Fig. 25). The development and evolution of the MH1 channel complexes are discussed below.

Development of the MH1 channel: SNA maps show that the channel complex was initiated by overbanking, crevasse development, and splay formation along the northern to the southeastern section of the Jubilee unit (Fig. 25). Flows moving down the eastern side of the sediment source diverted through an erosive breach and spilt westward into the Mahogany area (Fig. 25a). However, the existence of a small linear area of sand deposition inside the channel in this area before crevasse formation (Fig. 25a, b) suggests that it was a low spot along the levee. The uppermost parts of the large flows moving eastward, prior to crevasse erosion, may have topped the levee at this point (Fig. 25b). Sand thickness distribution (Fig. 27a) and flow patterns from the SNA maps (Fig. 25a, b) suggest that just up-system from the point of breakthrough on the northern bank, the flows moving westward went around a tight bend on the central side of the channel that would have diverted them to the southwest across the channel (Fig. 25c). The western and southern branches of the initial off-channel development were incorporated into the main MH1 reservoir development system as overbank splays that fed off the developing channel course at meander bends (dark blue of the frontal splay and the gold for the crevasse splay in Fig. 25c). The well-defined meandering channel coincides well with the route and specific meanders that characterized the MH1 channel (Fig. 25c). Flows coming out of this mender bend were aimed directly toward the point of breakthrough along the central levee (Fig. 28). This configuration would have enhanced the levee erosion and overbanking of the flows at this point (Figs. 25c and 28).

Fig. 27

Isochore (vertical depth thickness, in meters) map of the a MH1 reservoir and b MH4 reservoir unit

Fig. 28

Schematic representation of the key elements of the conceptual depositional model for the MH1 reservoir system

In addition, overbanking was preceded by the deposition of about 10 m of sandy strata (4050–4060.3 m) that includes a middle section of 5 m thick of thin-bedded sandstone (4053.15–4058.3 m), overlain and underlain by thin-bedded sandstone and mudstone (Fig. 23a). There is a single thick sandstone bed from 4055.0 to 4055.75 toward the middle of the cycle (Fig. 23a). This symmetric cycle lacks evidence for channels and erosion and represents deposition from strictly low-energy, low-density turbidity currents. It is interpreted to represent the downslope part of a small lobe or splay (Fig. 25b, c). The deposition of sediment within the channel may have been sufficient to trigger more extensive overbanking along the channel levee, erosion, and eventually, avulsion, as inferred from other systems (Kolla 2007; Lowe 2019; Stow and Smillie 2020). The earliest major overbanking and crevasse splay development along the main course of the channel complex (Fig. 25c) show the development of a small splay with southern and central lobes (gold polyline in Fig. 25c). Some sandy flows were diverted into a distal southern lobe running along the southern edge of the main channel levee (Fig. 25c). This distal lobe may be the distal tip of a larger sand body that is present at the base of the LM2/MH1 interval to the south (Fig. 25c), or it may represent a separate small lobe developed north of the main sand fairway during lower LM2/MH1 sedimentation.

As this splay developed, the locations and geometries of channels and lobes shifted across the depositional surface (Fig. 25b, c). Eventually, the outlines of the channel took shape, with enlargement of the middle part of the avulsion splay from the main channel flowing eastward (Fig. 25). However, the main channel itself was not yet well-defined toward the southeastern section. At this time, there was also localization of silt and mud deposition in more downslope areas into narrow splay-like features, similar to the description by Lowe et al. (2019).

The upper section of the core log (Fig. 23a) indicates two overall trends: (1) symmetrical pattern composed of upper and lower sections of medium-bedded sandstone and mudstone and a central core of thick-bedded sandstone meaning that the lowest and uppermost 4 to 5 m are made up of sandstone beds that are distinctly thinner and finer than those comprising the middle part of the cycle. (2) The lowest 5 m (4040.1–4045 m) includes a basal coarsening- and thickening-upward zone and an overlying minor symmetric cycle. All overlying small-scale internal cycles are fining-upward cycles, from 1.5 to about 3 m thick (Fig. 23a). These overall features suggest that this section represents a thick splay system (Fig. 25c). It indicates a broadly progradational at the base but the top was dominated by small channels that crossed the splay surface so that the waxing of splay activity is reflected in the coarseness of the fining-upward channel fill units (Fig. 28). Upslope this splay may show an increasing proportion of high-density flow deposits. Downslope, the Tb-dominated thick-bedded turbidites with poorly developed Tde divisions may grade into sections of thin-bedded sandstone. These two Lower Mahogany cycles probably represent more distal similar splay systems (Fig. 28).

Overall, the Lower Mahogany LM2/MH1 sand reflects the development of a large, Southwestern-trending sand fairway south of W1 (Fig. 28). During the evolution of this fairway, a splay- or finger-like sand body developed along its northwestern margin extending to the east off the channel area (Figs. 25 and 28). This splay may have developed toward the end of a distributary formed by the upslope bifurcation of the main channel or as an overbank splay along the side of the main channel, which lay to the south (Fig. 28).

MH4 channel complex

The MH4 reservoir unit shows sand accumulation as a series of three main northward offset stacked bodies that successively onlapped the south-western basin margin (Fig. 29). In the W1 core, the lowest sandstone bed, 3872.6–3878.7 m, overlies a dense, black fissile mudstone marking a regional interval of fine hemipelagic sedimentation. It lacks clear trends in grain size or bed thickness. The overlying sandstone from 3857.75 to 3871.1 m forms a well-defined lobe that was characterized by the abrupt flooding of the sediment fairway by high-density turbidity currents (Fig. 26). This unit probably represents lag deposits, deposited and worked by flows that were largely bypassing and transporting most of their sediment downslope (Fig. 26b). The dominance of amalgamated S3 divisions and the immaturity of the lower part of this section suggest that well W1 was located in a proximal setting on a lobe.

Fig. 29

Schematic representation of the key elements of the conceptual depositional model for the MH4 reservoir system

The development of these MH4 units can be interpreted in the context of the overall sedimentation in this area. The northeast–southwest-trending sand thick representing the lower MH4 sandstone unit (Fig. 27b) lies along the northwest edge but slightly north of the underlying thick sand body in MH1 (Fig. 29). At least one well-defined mud-filled in the lower MH4 channel can be seen crossing both the MH4 and the MH1 sand belts (Figs. 26a and 25a, respectively). These features suggest that during early MH4 sedimentation, a shallow topographic low existed parallel to the northeast–southwest trend of the MH1 sand belt but slightly offset to the north from the zone of maximum thickness of MH1. This shallow low was the site of lower MH4 sand accumulation (Fig. 26). It was bounded to the north by the northern basin slope and to the south by a high development along the trend of MH1 sedimentation. This high may have been accentuated by differential compaction, with the thick MH1 sands showing less overall compaction than muddier sections to the north.

Erosion associated with the late-stage, largely bypassing currents may have cut or accentuated the channel shown in Fig. 26d. This channel cuts across both the northeastern–southwestern-trending depositional area of lower MH4 sands and the MH1 sand thick near well W7. This suggests that these underlying sand units did not form a significant topographic barrier on the sea floor following the deposition of the lower MH4 sands. Although a low barrier probably existed initially for the lower MH4 sands to be deposited in this area, their deposition erased this topography and allowed turbidity currents to spill to the south across the MH4 unit. The channel shown in Fig. 26c, d may have formed at this time as the flows spilt across to the southwestern section.

The MH4A sand body is overlain by a thick section of mass transport deposits (Fig. 23b, c). The top of the MH4 reservoir is marked by an interval of mass transport deposits which shows that most topographic and architectural features developed during the preceding interval of sand sedimentation had been obliterated and the sediment surface was a largely featureless plain.

Sedimentation Pattern of MH1 and MH4 reservoirs

The main architectural components of this depositional setting included: (1) levee-like sections of thin-bedded mudstone and sandstone; (2) thick mass transport deposits, with rare intervals of less mixed mudstone and interbedded mudstone and thin-bedded sandstone, that probably were deposited in large part as thin-bedded sandstone and mudstone levee-like sequences; (3) coarse-grained sandstone layers from less than 1 m to about 15 m thick, many of which coarsen and thicken upward and show sharp erosional upper contacts, that appear to represent crevasse and splay deposits. The abundance of mud-chip-rich sandstones suggests extensive erosion and downcutting through the levee or levee-like sequences; and (4) units of mainly fine-to-medium-grained turbiditic sandstone showing little or no evidence of scour, mud-chips, or erosional tops. These may represent more distal splay deposits or flows that topped the levees without extensive erosion or crevasse formation (Figs. 23, 25, 26, 28 and 29).

At the distal section of the Jubilee basin, sediments reflect deposition on a large splay/lobe or a series of amalgamated splays/lobes by sandy high-density turbidity currents (Figs. 28 and 29). These currents included both collapsing flows and quasi-steady, largely bypassing currents that deposited thick sections of current-structured sand (Fig. 23). The surfaces of such lobes are commonly covered by shallow, meandering channels, but few well-defined channels can be identified (Figs. 25c and 26b). This probably reflects pervasive bed amalgamation and erosion and the shallow, non-erosive, meandering-like character of most channels. This was a large splay but probably not the terminal splay in the overall transport and depositional system.

A general model for this system is shown in Figs. 28 and 29. Flows topping the sediment source may have split into two parts (Figs. 28 and 29): (1) the near-bed high-density loads were deflected to the west, following the regional paleoslope into the Jubilee basin, and (2) the upper low-density sediment clouds, which would have tended to continue on a more southerly course, depositing their mud, silt, and fine sand loads as the broad eastern area of sedimentation.

The Jubilee Mahogany sand depositional system is divided into several areas: flows cresting the sediment source probably moved downslope into the Jubilee basin through one or more erosional channels. These may have been strictly erosional or may have accumulated flanking levees. The W3 well was drilled in this very proximal area and did not intersect any significant channels. Upper and Lower Mahogany sediments reflect mud and mass transport deposits outside of major channels and cut by numerous, mainly small erosional crevasses that fed small-to-moderate-sized splays (Figs. 28 and 29). A few sand units appear to reflect splays developed by overbanking without significant erosion. If the main channels were large areas of sediment bypassing, there may be few or no channel sand deposits preserved.

Downslope, the channels widened, shallowed, and perhaps largely disappeared (Figs. 28 and 29). The flows spread out into a broad fairway bounded on the north by the southwest slope off of the sediment source and on the southeast by the slope off of the mud- and mass-transport-covered eastern part of the Mahogany depositional area (Figs. 25, 26, 28 and 29). This fairway accumulated a series of amalgamated lobes. Sand deposition in this area may have been triggered by (1) spreading and thinning of the flows as they exited the proximal channel area, and (2) slope decrease off of the sediment source.

The distal section of the Jubilee basin represents an amalgamated splays system, reflecting largely bypassing flows (Figs. 28 and 29). This seems likely that, downslope of the splay depositional area, the flows reformed as steady or even erosive flows. Figures 28 and 29 suggest that in more distal areas, the lobe may have been replaced by one or more channels within which the flows deposited some sand but largely bypassed areas still farther downslope.

Conclusions

The Turonian Formation is interpreted as a deep marine, marginal to a slope channel with periodic failure of levee margins and associated ingress of channel-sourced sediment into extra-channel locations as lateral crevasse splays. These were based on the recognition of seventeen sedimentary facies. Upscaling the identified lithofacies to bed-stack facies allowed the deduction of deep marine depositional environment. The identified seventeen lithofacies were grouped into six facies association depicting deep marine, bioturbated shallow marine, paludal crevasse splay and overbank sheet sand.

The cumulative frequency percentage curves and grain size statistics for Turonian sandstone are primarily indicative of the sediments' medium-grained nature. Furthermore, the majority of the sandstones have a unimodal grain-size distribution representing less variation in the depositional processes. The sandstones have an average sorting of 2.50 (very poorly sorted) and are fine-skewed or positive skewed in nature. In general, the poorly sorted nature of the sediment depicts a blending of sediments of different sizes, resulting in an influx of kinetic energy associated with the mode of deposition in a marine environment. In most cases, both the peak and tails are poorly sorted, resulting in leptokurtic grain size patterns dominating. Linear discriminant function analyses for the Turonian sandstone are primarily indicative of turbidity current deposits in a shallow-marine environment.

In terms of diagenetic offset of the sandstone facies, syntaxial quartz overgrowths, kaolinite, and chlorite were the main authigenic phases visible in all the sandstone samples. Syntaxial quartz overgrowths appear to be volumetrically reduced in samples containing detrital clays, ferroan calcite, or abundant chlorite. These authigenic minerals resulted in reducing reservoir potential in these samples. Minor percentages of Sa, Sb, Sf, and Sm are found in moderate reservoir quality samples. The reservoir sandstones of high quality appear to be dominated by facies types Sc, Sp, Ss, Sd, and Cgl. Overall, the majority of the sandstone samples have good reservoir quality, with permeabilities greater than 100 mD.

Our findings highlight the significance of sedimentological and facies development processes in the overall evolution of deep-water channel systems. The history of the Jubilee channel complex sets almost definitely characterizes many other emerging undersea channel systems and can be used to evaluate the distribution, shape, and structuring of the resulting reservoir sands. In turn, the Jubilee data set adds to the library of channel-levee-splay-lobe systems, serves as a prospective predictive model for hydrocarbon exploration and development, and as an outcrop explanatory analogue of a channel-splay system.