Introduction

Complexity of depositional environment, geological settings and heterogeneity of reservoir parameters result in intricacy of resource identification and prospect generation, which affect directly well production performance, success of field development and cost-effectiveness (Ma and Pointe 2011; Adeoti et al. 2014; Seyyedattar et al. 2020). A comprehensive 3D reservoir model is essential to address the spatial reservoir property distribution (Abdel-Fattah et al. 2018; Adesoji et al. 2018; Oyeyemi et al. 2018). It involves interpretation of necessary and available geological and geophysical data to characterize, among others, structural elements, facies architecture and critical petrophysical properties like porosity, permeability, net-to-gross thickness and water saturation (Yu et al. 2011; Abdel-Fattah et al. 2018; Adelu et al. 2019). As drilling and field development progress, the initially built reservoir models are updated with additional drilled well data, and further refinement as well as complexity in the geological interpretation is introduced (Abdel-Fattah et al. 2010). This work attempted to establish the first 3D static reservoir model of the Q-Field, situated in the Niger Delta, Nigeria. The Niger Delta is an established prolific petroleum system, with extensive hydrocarbon accumulation in the clastic sequences of the Eocene Agbada Formation.

Researchers have utilized subsurface modeling workflows to address fault geometry (Adagunodo et al. 2017; Adelu et al. 2019), facies analysis (Aigbadon et al. 2017; Jika et al. 2019), petrophysical analysis (Alao et al. 2013; Adiela 2018; Ndip et al. 2018; Farouk et al. 2017; Kalu et al. 2020), hydrocarbon prospectivity (Omoja and Obiekezie 2018), hydrocarbon volume estimation (Oluwadare et al. 2017; Okpogo et al. 2018), reservoir production history matching (Ahmed et al. 2011) from the various oil and gas fields in the Niger Delta. This work provides all these deliverables and presents a robust fit-for-purpose 3D reservoir model of the onshore Q-Field by integrating seismic data as well as wireline logs from drilled wells. The principal objectives of the work were: (1) to interpret faults from the seismic volume and prepare a structural model; (2) to interpret key petrophysical properties from well logs and populate those in the 3D reservoir model; (3) to assess the reservoir properties and their spatial distribution within the facies model at the D6200, D7000, D9000 and E2000 reservoir intervals by 3D reservoir petrophysical models; and (4) to analyze the hydrocarbon distribution, fluid contact levels and estimate the hydrocarbon-in-place volumes based on the initial modeling results.

Geology of the study area

The Niger Delta Basin is a large rift basin located on the reactive continental margin near Nigeria's west coast, mainly in the Niger Delta and the Gulf of Guinea, with suspected or confirmed access to Cameroon, Equatorial Guinea and São Tomé and Príncipe (Agbasi et al. 2018, 2020). The Niger Delta Basin lies within a wider tectonic structure in the south-westernmost part, situated within 4°–9°E longitude and 4°–9°N latitude. This basin is very complex and has high economic value because it contains a proven petroleum system, hosted by more than 25,000 ft (1 ft = 0.3048 m) of sedimentary column, affected by regional tectonics (Inyang et al. 2018). The vast Niger Delta Basin is flanked by other basins in the offset region, influenced by similar structural evolution. The studied Q-Field is situated in the onshore central swamp depobelt region. Figure 1 presents a generalized location map of the study area, along with the base concession map with the available seismic inline-crosslines and studied well locations. The extraordinarily thick sedimentary succession is characterized by three principal lithofacies: (a) thick shales of the Akata Formation at the bottom; (b) marginal marine clastics of the Agbada Formation; and (c) extensive coastal sandstones of the Benin Formation. The Paleocene Akata Formation is characterized by marine shales with source rock qualities, and in the central part of the delta, its maximum thickness reaches up to 20,000 ft. The Eocene Agbada Formation is the principal hydrocarbon-bearing unit, which was deposited in a deltaic to fluvio-deltaic system (Ejedavwe et al. 2002; Burke 1972; Tuttle et al. 1999) above the Akata Formation. A maximum Agbada Formation thickness of approximately 9000 ft was recorded in the Niger Delta Basin. The Benin Formation is the youngest stratigraphic unit, which becomes thinner toward offshore (Doust and Omatshola 1990) and composed mainly of fluviatile sediments (Koledoye and Aydin 2000).

Figure 1
figure 1

Location of the studied Q-Field in the Niger Delta (left). Base concession map showing seismic inline-crossline coverages and well locations (right)

Full size image

Materials and Methods

This work integrates 3D seismic volume, checkshot data, lithological information, geological reports, deviation data and wireline logs (comprising of gamma ray, resistivity, density, neutron porosity and compressional sonic slowness) from drilled wells in the Q-Field. Four wells were available: Q-1, Q-25, Q-29 and Q-32. A well log correlation is presented in Figure 2. Details of the methods and the workflow applied in this work are discussed below.

Figure 2
figure 2

Well log correlations between the studied wells. ‘FS’ indicates the interpreted major flooding surfaces and correlated across the wells [a1 ft = 0.3048 m]

Full size image

Seismic Interpretation

A brief summary of the available seismic vintage is presented in Table 1. Seismic data analysis involved co-ordinate conversion, data loading, well to seismic tie using synthetic seismogram, horizon and fault mapping, time attribute map generation and time to depth conversion (Abdel-Fattah and Alrefaee 2014). Because the well logs are indexed against depth and because seismic data are indexed against time, it was very critical to establish a tie between seismic and well data. This was achieved by using checkshot survey available from the well Q-1 (Fig. 3). Using density (RHOB) and sonic slowness (DT) logs, the acoustic impendence (AI) and reflection coefficients (RC) were generated. RC and Ricker wavelets were then integrated to create a synthetic seismogram, which was tied to seismic data by adjusting a bulk shift of 15 ms. After achieving a satisfactory well to seismic tie, as a next step, formation tops from well logs were correlated and marked in the seismic profile together with the structural discontinuities (i.e., faults). Primary seismic reflections corresponding to the top of the main reservoir sands were produced by seismic mapping data (Victor et al. 2019). The selected horizons and faults were later used in 3D reservoir modeling.

Table 1 Vintage seismic properties available in the studied Q-Field
Full size table
Figure 3
figure 3

Well to seismic tie using the well Q-1 (post-tie). TWT is two-way time (ms). The second track plots density (RHOB) and sonic slowness (DT), which were combined to generate the acoustic impendence (AI on third track) and then reflection coefficients (RC on fourth track). RC and Ricker wavelets were integrated to generate synthetic seismogram, which was then tied back to seismic data by adjusting a bulk shift

Full size image

Petrophysical Interpretation

Three principal petrophysical parameters were characterized in this work using the wireline logs—porosity, permeability and water saturation. Total porosity (ϕ) was estimated from density log assuming a matrix density value of 2.65 g/cc in the studied clastic intervals. A shale volume correction was applied to assess effective porosity from total porosity. Water saturation (Sw) was estimated by the Waxman Smits method, and in the absence of produced water analysis, Pickett plot was utilized to evaluate formation water resistivity (Rw), as presented in Figure 4. Permeability was generated using Timur’s equation (Ismail et al. 2020). The standard practice is to validate and calibrate the log-based petrophysical properties with core-based direct measurements. However, in this study, core-based petrophysical properties were unavailable; therefore, all these log-based petrophysical parameters were populated in the 3D static reservoir model to infer the field wide property distribution.

Figure 4
figure 4

Pickett plot for assessing formation water resistivity (Rw) in the studied wells. Interpreted Rw was used to estimate water saturation (Sw)

Full size image

Static Reservoir Modeling

3D geocellular modeling was performed using PETREL software to infer the vertical and horizontal dispositions of the reservoir facies units, structural features and petrophysical properties (Godwill and Waburoko 2016; Abdel-Fattah et al. 2018; Jika et al. 2020). The structural model was based on the 3D seismic interpretation. The input data consisted of reservoir surface tops, polygons and measured fault surfaces. The first phase in developing a field structural model was to select the horizons and mark the fault based on the vertical dispositions of the correlated horizons. A 3D grid divides the space into cells in which materials are considered to be almost identical (Edwin et al. 2011; Olubunmi and Olawale 2018). Depending on the definition, the defects were mainly modeled on the input fault surfaces (Adeoti et al. 2014). The structural skeleton consisted of top, mid and base grid levels, where 3D fault pillars were interpreted after sealing the fault connections with the horizons (Haque et al. 2016) and then utilized to generate consecutive structural maps at various target formation top levels.

In view of the definition of the reservoir, the areal dimensions of the grid cells were set at 50 ft by 50 ft. There were 252 × 125 × 65 cells in the 3D static model. Once the seismic-based structural modeling was achieved, a velocity model was constructed to convert the structural domain from time domain to depth domain before running facies and petrophysical modeling (Abdel-Fattah and Tawfik 2015; Abdel-Fattah et al. 2015, 2018). Simulation of Sequential Indicator (SIS) technology was used for facies modeling in 3D space by a stochastic approach (Journel 1982; Deutsch and Journel 1998; Remy et al. 2009) (see “Appendix”). The approach allows for simple modeling of facies where vertically, laterally or, otherwise, the volumes of facies vary. The wells were labeled on the basis of the log data available (mostly gamma ray log) in facies associations. A gamma ray cutoff of GR > 75 API was used to distinguish the non-reservoir facies. The facies realizations were conditioned to the wells during the facies modeling phase, and several realizations were performed in the facies modeling to catch the intrinsic heterogeneity, if any. Well-based petrophysical interpretations were up-scaled and stochastically distributed using Sequential Gaussian Simulation (SGS) method (Maleki Tehrani et al. 2012; Geboy et al. 2013) within the respective facies distributions in 3D space (see “Appendix”).

Volumetric Analysis

Reservoir volumetric is the mechanism by which the hydrocarbon concentration in a reservoir is determined (Ali et al. 2020; Egbe et al. 2019). It is really important because it serves as a reference for exploring and developing fields. After a static field model was created, the structural model and the constructed petrophysical model were used to measure the reserves in terms of stock tank of original oil in place (STOOIP) and gas initially in place (GIIP) of the reservoir under study using the following equations:

$$\mathrm{STOIIP }\left(\mathrm{STB}\right)=7758 A h \phi (1- {S}_{w})\frac{1}{{B}_{o}}$$
(1)
$$\mathrm{GIIP}= \left(\frac{43{,}560 A h \phi (1-{S}_{w})}{{B}_{gi}}\right)$$
(2)

where A is area in acres, h is net pay thickness in feet, ϕ is porosity, Sw is water saturation, B0 is formation volume factor, and \({B}_{gi}\) is gas formation volume factor.

Results and Discussion

In the studied Q-Field, a confident well to seismic tie was achieved using the checkshot data of well Q-1, which provided great compatibility (Fig. 5). A seismic coherence attribute map (Fig. 6) provided critical information in identification of several faults and their extension. Faults and horizons were mapped in the seismic volume, guided by the well log-based interpretations (Figs. 7 and 8). Based on the horizon and fault mapping in the seismic volume, fault pillars were interpreted and structural modeling was performed (Fig. 9). Time and depth structure maps display network of development faults, which stretched up to 80% of the mapped area's whole breath. Series of normal faults dominated the whole mapped area, categorized as syn-depositional listric faults and minor antithetic faults. Principal normal faults were NW–SE trending and dipped toward SW. These faults assisted in hydrocarbon trapping mechanism in the Q-Field.

Figure 5
figure 5

TWT-depth plot using synthetic seismogram and checkshot data of well Q-1 [a1 ft = 0.3048 m]

Full size image
Figure 6
figure 6

Coherence seismic attribute map at 2256 ms indicating mapped faults and interpreted fault trends

Full size image
Figure 7
figure 7

Seismic section showing the correlated reservoir tops (D6200, D9000 and E2000 marked in this section) as confirmed by the well log-based stratigraphic interpretation. Black vertical sticks represent the well locations (Q-29, Q-25 and Q-1)

Full size image
Figure 8
figure 8

Mapped faults and horizons on traverse line along well Q-1

Full size image
Figure 9
figure 9

Structural model of the studied Q-Field. a Fault pillars as structural framework. b Stacked reservoir horizons (D6200–E2000) along with interpreted faults. c Depth map at D6200 reservoir top level [a1 ft = 0.3048 m]

Full size image

Petrophysical investigation identified four principal reservoir intervals; these are D6200, D7000, D9000 and E2000 from shallow to deep. Both the well logs and seismic horizon mapping indicated that these reservoirs are laterally extensive throughout the modeled area and affected by the interpreted normal faults. Reservoir interval distributions in the studied wells are documented in Table 2. Average porosity and permeability in these reservoirs vary in the range of 17.9–21.6% and 62–105 mD, respectively. Reservoir net-to-gross (NTG) varies in the range of 0.6–1. Water saturation varies from 36.3 to 74%, while hydrocarbon saturation values vary in the range from 26 to 63.7%. The 3D static reservoir modeling was performed to infer the lateral and vertical distributions of the facies assemblage (Fig. 10), porosity (Fig. 11), NTG (Fig. 12), permeability (Fig. 13) and water saturation (Fig. 14) throughout the Q-Field.

Table 2 Distribution of reservoir intervals in the four studied wells in the Q-Field
Full size table
Figure 10
figure 10

Results of facies modeling in the Q-Field. Left panel presents facies maps at a D6200, b D7000, c D9000 and d E2000 reservoir levels. e Structural base map and cross-sectional panel line. f Facies model in a cross section with the studied well locations [a1 ft = 0.3048 m]

Full size image
Figure 11
figure 11

Results of reservoir porosity modeling in the Q-Field. Left panel presents porosity maps at a D6200, b D7000, c D9000 and d E2000 reservoir levels. e Structural base map and the cross-sectional panel line. f Porosity model in a cross section with the studied well locations [a1 ft = 0.3048 m]

Full size image
Figure 12
figure 12

Results of net-to-gross (NTG) modeling in the Q-Field. Left panel presents NTG maps at a D6200, b D7000, c D9000 and d E2000 reservoir levels. e Structural base map and the cross-sectional panel line. f NTG model in a cross section with the studied well locations [a1 ft = 0.3048 m]

Full size image
Figure 13
figure 13

Results of permeability modeling in the Q-Field. Left panel presents permeability maps at a D6200, b D7000, c D9000 and d E2000 reservoir levels. e Structural base map and the cross-sectional panel line. f Permeability model in a cross section with the studied well locations [a1 ft = 0.3048 m]

Full size image
Figure 14
figure 14

Results of water saturation (Sw) modeling in the Q-Field. Left panel presents Sw maps at a D6200, b D7000, c D9000 and d E2000 reservoir levels. e Structural base map and the cross-sectional panel line. f Sw model in a cross section with the studied well locations [a1 ft = 0.3048 m]

Full size image

In general, the 3D facies model depicted larger probability distribution of sand (dominantly) and minor silts in all the reservoirs (Fig. 10), although the D9000 reservoir was mostly silty and E2000 was, lithologically, more heterogeneous in both vertical and lateral scales. The top reservoir D6200 had an average thickness range of 104–155 ft, as encountered in the four wells. The facies model indicated a dominance of sands with minor silt distribution in the D6200 reservoir, which translated into high NTG value and a more homogenous porosity–permeability distribution. The static modeling results showed that the D7000 reservoir properties were very similar to that of the D6200 interval in terms of major facies distribution, porosity and NTG distribution; however, it had lower permeability and higher water saturation than the overlying reservoir. D9000 reservoir was dominated by silty facies distribution, resulting in lesser porosity and moderate to poor NTG. In the western part of the Q-Field, the D9000 reservoir showed higher abundance of sand facies comprising of higher porosity and NTG. The bottom most reservoir E2000 provided the maximum reservoir thickness of 171–213 ft, and it was found to be heterogeneous in terms of reservoir properties. Sand facies was less laterally extensive, rather discontinuous when compared to the upper reservoirs, resulting in a wide NTG distribution throughout, greater variability of water saturation and lower permeability in the E2000 reservoir interval.

The dominant reservoir fluid type is gas. Out of the four studied wells, only Q-1, being drilled in the anticline, had encountered hydrocarbon columns in all the reservoirs, while Q-32 was drilled in the synclinal position and passed through the reservoir water legs. The Q-25 and Q-29 encountered both gas and oil only in the E2000 reservoirs. The lateral and vertical distributions of the interpreted hydrocarbons were populated through the static reservoir modeling (Fig. 15). Gas-down-to exists in the upper three reservoirs (D6200, D7000 and D9000) of the well Q-1 at 10,577 ft, 10,756 ft and 11,389 ft, respectively. Gas–oil and oil–water contacts were interpreted in the deepest reservoir interval E2000 at 11,812 ft and 11,886 ft, respectively. The interpreted fluid contact depths and reservoir volumetrics (gas and oil in place) of the studied reservoirs are presented in Table 3.

Figure 15
figure 15

Results of hydrocarbon distribution modeling in the Q-Field. Left panel presents hydrocarbon extension maps at a D6200, b D7000, c D9000 and d E2000 reservoir levels. e Structural base map and the cross-sectional panel line. f Gas–oil–water distribution and fluid contacts in a cross section with the studied well locations [a1 ft = 0.3048 m]

Full size image
Table 3 Hydrocarbon contacts, average porosity, water saturation, and hydrocarbon volume estimates in the four studied reservoir intervals of the Q-Field
Full size table

Conclusions

The present 3D static reservoir modeling work presents a clearer understanding of the field-wide distribution of the reservoir facies, structural style and petrophysical parameters in the Q-Field, Niger Delta, Nigeria. Four clastic Eocene reservoir intervals (D6200, D7000, D9000 and E2000) were studied using four wells drilled across the Q-Field. Normal faults with NW–SE trends, which acted as hydrocarbon traps in the study area, were interpreted from the seismic data and 3D structural modeling. Facies modeling indicated extensive lateral continuity of the sand-dominated reservoir facies. Overall, a narrow porosity range of 18–22% was found from the 3D reservoir petrophysical modeling. The reservoirs exhibited medium to high NTG and heterogeneous permeability distribution across the four reservoir levels. Water saturation modeling and fluid distribution maps indicated that the hydrocarbon accumulation was restricted in the structural highs, i.e., anticlines, in the central part of the Q-Field, as confirmed by fluid contacts in three studied wells (Q-1, Q-25 and Q-29). The subsurface geological model, presented in this work, will be critical for a further development program of the reservoirs. It will help to better identify future well locations. Future work will involve further detailing and fine-tuning of the reservoir models based on future well data.