Advertisement

3D geological modeling of the Upper Cretaceous reservoirs in GPT oil field, Abu Sennan area, Western Desert, Egypt

  • Mohamed F. Abu-HashishEmail author
  • Hamdalla A. Wanas
  • Emad Madian
Open Access
Original Paper - Exploration Geophysics
  • 217 Downloads

Abstract

This study aims to construct 3D geological model using the integration of seismic data with well log data for reservoir characterization and development of the hydrocarbon potentialities of the Upper Cretaceous reservoirs of GPT oil field. 2D seismic data were used to construct the input interpreted horizon grids and fault polygons. The horizon which cut across the wells was used to perform a comprehensive petrophysical analysis. Structural and property modeling was distributed within the constructed 3D grid using different algorithms. The workflow of the 3D geological model comprises mainly the structural and property modeling. The structural model includes fault framework, pillar girding, skeleton girding, horizon modeling and zonation and layering modeling processes. It shows system of different oriented major and minor faults trending in NE–SW direction. The property modeling process was performed to populate the reservoir facies and petrophysical properties (volume of shale (Vsh), fluid saturations (Sw and Sh), total and effective porosities (Φt and Φe), net to gross thickness and permeability) as extracted from the available petrophysical analysis of wells inside the structural model. The model represents a detailed zonation and layering configuration for the Khoman, Abu Roash and Bahariya formations. The 3D geological model helps in the field development and evaluates the hydrocarbon potentialities and optimizes production of the study area. It can be also used to predict reservoir shape and size, lateral continuity and degree of interconnectivity of the reservoir, as well as its internal heterogeneity.

Keywords

Geological model Petrophysical analysis Upper Cretaceous Seismic interpretation Western Desert Egypt 

Introduction

Abu Gharadig Basin is one of the most prolific basins in the northern part of the Egyptian Western Desert. It is attracted the attention of geologists and geophysicists for oil and gas exploration. This is due to the huge subsurface sedimentary section which exists in the basin which includes considerable reservoir rocks within the members of the Upper Cenomanian to Maastrichtian (i.e., Khoman, Abu Roash formations), Lower Cenomanian (i.e., Bahariya Formation) and the Albian (i.e., Kharita Formation). In addition, potential source rocks have been encountered within the Khatatba Formation and Abu Roash “F” member (El-Shaarawy et al. 1994).

GPT oil field (General Petroleum T oil field) is located at the southeast of Abu Gharadig Basin which includes Abu Sennan highs (Fig. 1). It lies about 200 km west to the Nile Delta and 15 km to the southeast of Abu Gharadig oil field. It covers an area of about 54 km2 and is located between latitudes 29°36′ and 29°39′N and longitudes 28°36′ and 28°42′E.
Fig. 1

a Location map of the study area. b Shot point and base map of the studied wells

Methodology

The geological model comprises mainly the structural and property modeling. Well log and seismic data were used to construct the structural model for GPT oil field including the fault framework, horizon modeling and zonation and layering modeling. The property modeling was performed to populate the reservoir facies and petrophysical properties such as effective porosity, permeability and water saturation inside the structural model.

Cutting’s description, cores, well log data, formation tops were integrated to construct both the petrophysical and facies models. The structural map for the Upper Cretaceous rocks in GPT filed is constructed. Facies modeling was done by using of the sequential Gaussian simulation (SGS) method. And the petrophysical model of the porosity, effective porosity, permeability and water saturation is obtained. The flowchart for the 3D geological model is shown in Fig. 2. The final geological model will be helpful for better development of the undrilled areas as well as for volumetric assessment of hydrocarbon reserves (Abu-Hashish and Said 2016).
Fig. 2

Flowchart of the 3D geological modeling

Lithostratigraphy

Abu Gharadig Basin comprises rock units ranging from Cambrian to Recent resting on the basement rocks. The target formations in that area are the Khoman, Abu Roash and Bahariya formations (Fig. 3). The Lower Cretaceous (Albian) Kharita Formation is composed of sandstone interbedded with shales and siltstone. The Upper Cretaceous (Cenomanian) Bahariya Formation consists mainly of sandstone, with thin streaks of shales and limestone (Kandil 2003). The sandstones are the main gas and/or condensate pay zone. The Abu Roash Formation is subdivided into several members which are termed alphabetically from A to G (Norton 1967). The members C, E and G are made up of calcareous sandstone with argillaceous intercalations, while the members A, B, D and F are composed mainly of limestone with argillaceous intercalations. The Khoman Formation is Santonian to Maastrichtian in age and is composed mainly of chalky limestone and shale. It is subdivided into two members: “A” and “B.” Khoman “A” consists mainly of chalky limestone, whereas Khoman “B” consists of shale with limestone streaks. Khoman “B” is absent in the over high areas.
Fig. 3

General lithostratigraphic column of the North Western Desert

Modified by Abdelmaksoud et al. (2019) after Schlumberger (1995)

A well correlation chart was constructed in (NE–SW) direction passing through the wells: GPT-10, GPT-14 and GPT-5 and shows that the upper formations (Moghra, Dabaa and Apollonia) reveal the same thickness and gradual increase in the northeast and southwest directions (Fig. 4). However, three deeper formations (Khoman, Abu Roash and Bahariya) are thickening toward the southwest direction. This pattern of facies variation may be due to the depositional environment conditions as well as structural and tectonic settings.
Fig. 4

Geologic cross section along the profile (NE–SW)

Seismic data interpretation

Detection of the seismic surfaces and fault picking are the main two steps in the seismic data interpretation. During these processes, the seismic horizons and the structural and stratigraphic features could be recognized (Bacon et al. 2005). The eight seismic reflectors (tops of Dabaa, Apollonia, Khoman, Abu Roash (A, D, F and G members) and Bahariya formations) are picked out in ten seismic sections after being tied to each other. The seismic sections have been selected to highlight the structural pattern of the area.

Seismic sections run in northwest–southeast and northeast–southwest directions (Fig. 1). The interpretation of the seismic sections shows that the reflector sequences of these lines are strong with gradual dipping and thinning in a northeast–southwest direction. The main faults appear to be in two sites: northeast–southwest main faults and northwest–southeast minor faults. Khoman reflectors appear strong and continuous, dissected with few numbers of the normal faults giving rise to a form of horst structures. For the other reflectors (Abu Roash and Bahariya formations), there are main and minor faults that affect the area forming the horst, step faults and half graben block area. The main normal faults and the structural closures of the Cretaceous levels are observed (Fig. 5).
Fig. 5

NW–SE interpreted seismic section GP96

TWT and depth structure contour maps were constructed for the top of Abu Roash (A, D, F, G members) and Bahariya formations, which indicate the asymmetrical faulted anticline trending in a northeast–southwest direction. Also, it shows that the area is affected by several parallel faults that have the trends of northeast–southwest forming horst and step faults. These normal fault elements are mostly dissected the northeast–southwest mid-basin arch of the Syrian Arc System. Moreover, most of these faults are throwing northward toward Abu Gharadig Basin forming synthetic faults, while the other fault group is throwing toward the Sitra platform forming antithetic faults (Figs. 6, 7, 8, 9).
Fig. 6

TWT structure contour map of top Bahariya Formation

Fig. 7

Depth structure contour map of top of the Bahariya Formation

Fig. 8

TWT structure contour map of top Khoman Formation

Fig. 9

Depth structure contour map of top of the Khoman Formation

Petrophysical analysis

In order to construct the 3D geostatic model, a comprehensive petrophysical analysis was performed as it gives a full description of the main properties of the reservoir pore system, permeability and saturation (Krygowski 2003). Environmental corrections that aimed to minimize the effect of variable hole size and acquisition conditions, especially the mud properties such as mud weight and salinity, were done. Volume of shale, porosity and net-to-gross ratio were estimated. Because of the presence of shale Indonesian, formula (Eq. 1) was used to calculate the water saturation Sw (Poupon and Leveaux 1971):
$$\frac{1}{{\sqrt {R_{t} } }} = \left( {\sqrt {\frac{{\varPhi^{m} }}{{aR_{W} }}} + \frac{{V_{\text{cl}}^{{\left( {1 - {{V_{\text{cl}} } \mathord{\left/ {\vphantom {{V_{\text{cl}} } 2}} \right. \kern-0pt} 2}} \right)}} }}{{\sqrt {R_{\text{cl}} } }}} \right)S_{W}^{n}$$
(1)
where Ф = porosity, Rt = true resistivity, Rw = formation water resistivity, Vcl = shale volume, Rcl = resistivity against shale, a = tortuosity factor.
The results of the petrophysical analysis in three wells in GPT oil field are tabulated in Table 1. The petrophysical properties of both Khoman and Abu Roash formations in GPT-1 well is presented in Fig. 10.
Table 1

Petropysical properties in GPT oil field

Parameters

Wells

GPT-1

GPT-14

GPT-17

Formation

Member

Volume of shale (Vsh) %

Khoman

15

19

15

Abu Roash

B

12

6

20

D

18

20

18

E

11

16

16

F

18

14

19

G

14

21

16

Bahariya

17

13

18

Total porosity (PhiT) %

Khoman

20

18

13

Abu Roash

B

16

13

10

D

26

19

13

E

25

21

22

F

21

18

18

G

28

21

24

Bahariya

25

23

21

Effective porosity (Phie) %

Khoman

15

16

12

Abu Roash

B

13

12

8

D

22

18

11

E

22

20

21

F

17

17

16

G

20

20

22

Bahariya

21

23

21

Water saturation (Sw) %

Khoman

23

30

41

Abu Roash

B

27

19

23

D

25

23

13

E

16

18

12

F

3

6

6

G

16

18

17

Bahariya

9

8

13

Hydrocarbon saturation (Sh) %

Khoman

77

70

59

Abu Roash

B

74

81

77

D

75

77

87

E

84

82

88

F

97

94

94

G

84

82

83

Bahariya

91

92

87

Gross thickness

Khoman

23.80

36.60

37.20

Abu Roash

B

6.70

2.00

2.10

D

3.70

3.00

6.20

E

6.4

8.20

6.40

F

10.00

17.70

16.80

G

13.8

10.90

14.20

Bahariya

19.8

16.1

32.9

Net thickness

Khoman

15.80

25.25

5.95

Abu Roash

B

2.95

0.60

1.60

D

2.50

1.35

1.80

E

4.35

4.45

3.85

F

3.90

16.80

4.85

G

8.05

4.65

9.7

Bahariya

15.3

16.10

28.80

Gross/net thickness

Khoman

0.66

0.69

0.16

Abu Roash

B

0.44

0.30

0.76

D

0.68

0.45

0.29

E

1.41

1.23

1.24

F

0.39

0.95

0.29

G

1.747

1.321

1.364

Bahariya

2.371

2.00

0.91

Fig. 10

Lithosaturation cross-plot of Khoman Formation and Abu Roash Formation (A, B, C and D members) in GPT-1 well

3D geological modeling

The 3D geological model mainly includes the structural framework of the area (faults and surfaces) in addition to zonation and layering of Khoman, Abu Roash (A, B, C, D, E, F and G members) and Bahariya formations, while the property model includes the population of the shale volume, porosity, permeability, net to gross thickness and fluid saturation as extracted from the available petrophysical analysis.

Geostatistical techniques have been used for modeling of the different geological properties which are helpful to estimate the values of the geological properties at the undrilled locations (Deutsch 2002). These variables include petrophysical properties such as shale volume, porosity, permeability and fluid saturation. Seismic information such as acoustic values or geological attribute as the thickness of the sand or shale is also used to describe the reservoirs. The GPT oil field is about (54 km2), and due to the presence of relatively thin layers and zones we subdivided the model into different small xyz cells (25 × 25 × 5 m). It was noted that the pay zones in GPT field are separated by some intervals which have relatively poor reservoir properties, so we decide to subdivide the model according to the stratigraphic settings not the geographic settings. A 3D geological model using multiple parameters (e.g., shale volume, porosity, facies, fluid saturation, net to gross thickness and permeability) was created using Petrel software.

Structural framework

Structural modeling is the first and the most important step in building a 3D geological model. It was mainly performed in two parallel ways: construction of maps for structural configuration of the formation tops and detecting the sets of faults running across the reservoir. The geometry definition process is used to define the geometry of the structural framework by specifying its coverage, defining the xyz resolution. The default x and y coverage and resolution are taken from the seismic, e.g., the same coverage and orientation as the seismic survey. This model is finely layered and has relatively small xyz cell dimensions (25 × 25 × 5 m). A coarsening factor can be used to create a geometry that honors the input grid extents but with multiples of the grid increments (Fig. 11). The 2D seismic survey was used to interpret the sticks or boundary faults in the GPT oil field (four main faults). Then, the fault model was built in time domain and converted to depth domain using the velocity model. The 2D seismic survey was used to interpret the sticks or boundary faults in the GPT oil field (four main faults). Then, the fault model was built in time domain and converted to depth domain using the velocity model (Fig. 12).
Fig. 11

Base map showing the modeled faults in the study area (top Abu Roash D member

Fig. 12

3D view of the final fault model in the study area

Property model

The step of the property model includes the process of filling the grid cells with discrete or continuous properties (Schlumberger 2010). The target is to use all geological information available to build a realistic property model. Property modeling is split into three separate processes: geometrical, facies and petrophysical. The next step will be the analysis of data and scale-up well logs as it helps to assign the values of the well log data to the 3D grid cells in the model; every cell will hold one value only so the well log values will be averaged and scaled up to help in the distribution of the reservoir properties among the wells (Fig. 13).
Fig. 13

Scale-up logs for facies modeling in GPT-1 well

Facies modeling

For better understanding of the sedimentology, it is required to realize high-resolution stratigraphic correlations (Wescott 1993). Facies analysis is very important and helpful to guide the decisions concerning the proper modeling technique (Henriquez et al. 1990; Hirst et al. 1993). Facies modeling techniques can be divided into interactive, deterministic and stochastic methods where the stochastic methods are often used in conditions where sparse data are available. These methods produce a possible result and can be used to produce multiple equally probable realizations (Petrel 2009). It involves creating a series of equally probable realizations, or models, of a random variable. Each model, or stochastic image, depends on the assumptions that have gone into the simulation procedure. The most popular type of simulation is called conditional simulation, which honors the hard data values at measured locations (Srivastava 1994).

Although there are several algorithms used to perform stochastic simulation, the most common is called sequential indicator simulation (SIS) and sequential Gaussian simulation (SGS). SIS allows a stochastic distribution of the property, using the pre-defined histogram. It is used to produce random facies but with rugged fuzzy boundaries between the different facies as it visits each cell in the model in a random order and assigns it the facies value that has the highest probability of existing in that location. The probability of a facies prevailing at a given location is known as the indicator transform for that facies and is determined from aerial and vertical geologic trend information, spatial statistics that define the degree of correlation with nearby points, and facies values both original and modeled in nearby model cells (Deutsch 2002). We use SIS to construct the three-dimensional facies modeling in GPT oil field which makes a random distribution of the facies with rugged fuzzy boundaries between the different facies (Fig. 14). Facies type described in the study area was given a numerical code (0–11), and facies logs were created from the lithology descriptions. Recall that the model vertical resolution is high in the other geologic zones as shown in Table 2.
Fig. 14

Facies model example for the different pay zones

Table 2

Lithological code description in GPT oil field

Common facies environments are carbonates, shale, siltstone and sandstone. The target is to build a 3D facies model containing the reservoir architecture with flow units and barriers. The facies model was constructed for the different reservoir zones within the Khoman, Abu Roash (A, B, C, D, E, F and G members) and Bahariya formations.

The main lithology of the GPT oil field reservoir (sandstone, shale, siltstone and limestone) was modeled to produce 3D geological models. In order to construct facies models, the following inputs are required: (a) discrete facies logs from the wells, (b) facies probability maps that characterize the general aerial distribution of each facies, (c) vertical proportion curves that quantify the vertical trends exhibited by each facies, (d) vertical and horizontal variograms that establish quantitative measures of spatial correlation between cells in each of the vertical and major and minor horizontal directions. Volume of shale, porosity, permeability, water and hydrocarbon saturation and net to gross value control the distribution of the facies.

Petrophysical model

The petrophysical evaluation data have been used to populate the reservoir properties in the finalized structural model, which include evaluation of shale volume, effective porosities, fluid saturations and permeability (Petrel 2009). Effective porosity, shale volume, fluid saturation, net to gross thickness and permeability models for the area represent the distribution of these properties within the reservoir. These models are used to analyze the distribution of reservoir quality within the formation.

Effective porosity model

The main factor controlling the effective porosity is the facies; it is modeled separately for each rock unit. The model of both the effective porosity and facies is relatively similar. Effective porosity was populated into the model for the Khoman, Abu Roash (A, B, C, D, E, F and G members) and Bahariya formations. The porosity log curves were scaled up honoring the new fine layering approach using the arithmetic computation (Fig. 15). Sequential Gaussian simulation algorithm method was used for Khoman, Abu Roash (A, B, C, D, E, F and G members) and Bahariya formations (Fig. 16). The effective porosity model shows several variations ranging from 2 to 35% and are related to color from red to green; this may be due to changes in facies. The effective porosity of the studied formations is highest in the central parts of the study area and decreases considerably in the south which led to the recommendation of further drilling in the central parts.
Fig. 15

Scale-up logs for the effective porosity modeling in GPT-1 well

Fig. 16

Effective porosity model example for the some pay zones

Permeability and water saturation models

In addition to populating grid cells with facies and porosity, as discussed above, the workflow process involves populating the grid cells with permeability and water saturation. Their log curves were controlled by the porosity and facies. These log curves for all wells were scaled up honoring the new fine layering approach using the arithmetic computation. SGS algorithm was used for the Khoman, Abu Roash (A, B, C, D, E, F and G members) and Bahariya formations (Fig. 17). The permeability increases toward the central part of the field and decreases to the north and south directions, while the water saturation shows several variations from increasing and decreasing. It decreases toward the south and the center of the study area (Fig. 18).
Fig. 17

Permeability model example for the different pay zones

Fig. 18

Water saturation model example for the different pay zones

Conclusions

The constructed 3D geostatic model for GPT oil field shows that the main faults in the area appear in the northeast–southwest direction while the main faults are found in the northwest–southeast direction. Property and facies modeling indicates that the central northern part of the field is a promising area having a good reservoir quality so we highly recommend to drill more development wells in that area. The constructed 3D geological model is valid integrated with the production data and flow rate to make the dynamic models in order to enhance the total production of the GPT oil field. The constructed model built a better understanding of the heterogeneity and variability of geological condition as well as the hydrocarbon distribution within the Upper Cretaceous reservoirs in GPT oil field. The Khoman and Abu Roash G sand and Bahariya reservoirs are very promising because of their good to very good porosity values, low water saturation, high hydrocarbon saturation (Sh), good permeability and moderate net to gross value.

Notes

Acknowledgements

The authors wish to express their thanks and gratitude to the EGPC (Egyptian General petroleum Corporation) and GPC (General Petroleum Company) for permission and release of the data for this research.

References

  1. Abdelmaksoud A, Amin AT, El-Habaak GH, Ewida HF (2019) Facies and petrophysical modeling of the Upper Bahariya Member in Abu Gharadig oil and gas field, north Western Desert, Egypt. J Afr Earth Sci 149:503–516 CrossRefGoogle Scholar
  2. Abu-Hashish MF, Said A (2016) Volumetric assessment through 3D geostatic model for Abu Roash “G” reservoir in Amana field-East Abu Gharadig Basin-Western Desert-Egypt. J Geol Geophys 2016(5):2.  https://doi.org/10.4172/2381-8719.1000242 CrossRefGoogle Scholar
  3. Bacon M, Simm R, Redshaw T (2005) 3D seismic interpretation. Cambridge University Press, Cambridge, p 212Google Scholar
  4. Deutsch CV (2002) Geostatistical reservoir modeling. Oxford University Press, New York, p 376Google Scholar
  5. El-Shaarawy OA, Zaafan S, Rashed R, Lelek J, El Leboudy M, Hadidy A (1994) Jurassic hydrocarbon potentiality in the Abu Gharadig Basin, North Western Desert of Egypt. In: 12th EGPC exploration and production conference Cairo, vol 2, pp 234–247Google Scholar
  6. Henriquez A, Tyler KJ, Hurst A (1990) Characterization of fluvial sedimentology for reservoir simulation modeling. SPE Form Eval 5:211–216CrossRefGoogle Scholar
  7. Hirst JP, Blackstock CR, Tyson S (1993) Stochastic modeling of fluvial sandstone bodies. Int Assoc Sedimentol Spec Publ 15:237–252Google Scholar
  8. Kandil M (2003) Reservoir characterizations of Bahariya formation in Khalda oil field, Western Desert, Egypt. M. Sc. Thesis, Zagazig University, Zagazig, p 168Google Scholar
  9. Krygowski D (2003) Guide to petrophysical interpretation. American Association of Petroleum Geologists, AustinGoogle Scholar
  10. Norton P (1967) Rock stratigraphic nomenclature of Western Desert, Egypt. Internal Report of General Petroleum CompanyGoogle Scholar
  11. Petrel Course Notes (2009) Computer software course manual. Schlumberger Information Solutions, HoustonGoogle Scholar
  12. Poupon A, Leveaux J (1971) Evaluation of water saturation in shaly formations. Log Anal 12(4):3–8Google Scholar
  13. Schlumberger (1995) Well evaluation conference, Egypt. Schlumberger Technical Editing Services, Chester, pp 58–87 Google Scholar
  14. Schlumberger (2010) Petrel 2010 introduction G & G manual. In: Schlumberger Technology Corporation Three Allen Center, Suite 1900, 333 Clay St, Houston, Texas 77002, USA. Sedimentology for reservoir simulation modeling. SPE Formation Evaluation, pp 211–216Google Scholar
  15. Srivastava RM (1994) An overview of stochastic methods for reservoir characterization. In: Yarus JM, Chambers RL (eds) Stochastic modeling and geostatistics: principles, methods and case studies. AAPG Computer Application in Geology no 3, American Association of Petroleum Geologists, Tulsa, OK, pp 3–16Google Scholar
  16. Wescott WA (1993) Geomorphic thresholds and complex response of fluvial systems some implications for sequence stratigraphy. AAPG Bull 77(7):1208–1218 Google Scholar

Copyright information

© The Author(s) 2019

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Authors and Affiliations

  1. 1.Geology Department, Faculty of ScienceMenoufiya UniversityShebin El-KomEgypt
  2. 2.Department of Petroleum Geology and Sediments, Faculty of Earth SciencesKing Abdulaziz UniversityJeddahSaudi Arabia
  3. 3.PetroGulf Misr Oil CompanyCairoEgypt

Personalised recommendations