Research Article Quantitative Characterization of Tidal Couplets in Oil Sands Reservoir, the Upper Mcmurray Formation, Northeastern Alberta, Canada

GeoScienceWorld Lithosphere Volume 2021, Article ID 2993537, 14 pages https://doi.org/10.2113/2021/2993537

Research Article Quantitative Characterization of Tidal Couplets in Oil Sands Reservoir, the Upper McMurray Formation, Northeastern Alberta, Canada

Hao Chen ,1,2 Jixin Huang ,1 Zhaohui Xia ,1 Zhiquan Nie ,2 Xiaoxing Shi ,1 and Jiuning Zhou 1

1Research Institute of Petroleum Exploration and Development, PetroChina, Beijing 100083, China 2China National Oil and Gas Exploration and Development Company, Beijing 100034, China

Correspondence should be addressed to Jixin Huang; [email protected]

Received 20 May 2021; Accepted 17 July 2021; Published 23 August 2021

Academic Editor: Zhongwei Wu

The McMurray Formation, NE Alberta, Canada, is one of the most significant bitumen bearing deposits worldwide. This formation deposited and reworked in fluvial, tidal, or estuarine environments results in a huge number of tidal couplets (TCs) which is consisted of mm-cm scale sandy and muddy interlayers. These couplets not only increase the geologic heterogeneity of the oil sand reservoir but also make it hard to predict the performance of in situ thermal processes. In this paper, based on literatures, lab analysis, core photos, logging, and drilling data, a quantitative characterization procedure for mm-cm scale tidal couplets was proposed. This procedure, which includes identification, classification, quantitative description, and spatial distribution prediction, was presented. Five parameters, thickness, mud volume, laminae frequency, spatial scale, and effective petrophysical properties, were selected to describe the TCs quantitatively. To show the procedure practically, TCs in the oil sand reservoir of McMurray Formation, Mackay River Project, and CNPC, were selected to demonstrate this procedure. The results indicate that the TCs are in mm-cm thickness, densely clustered, and in a variety of geometries. Based on geologic origins, these couplets were divided into four types: tidal bar couplets (TBCs), sand bar couplets (SBCs), mix flat couplets (MFCs), and tidal channel couplets (TCCs). The thickness, mud volume, and frequency were calculated by mathematical morphological processed core photos. The spatial scale of TCs was estimated by high-density well correlations. The effective petrophysical properties were estimated by bedding scale modeling and property modeling via REV. Finally, the spatial distribution of TCs was predicted by object-based modeling.

1. Introduction to inject steam in the reservoir zone, thereby slowly heating and mobilizing the bitumen. The heated bitumen, now liquid, The oil sands or tar sands, which play an important role in flows via gravity down the margins of the developing steam the world energy market, are one typical unconventional chamber and into the production well below [7]. Although hydrocarbon resource [1, 2]. Large amount of oil sand reser- the recovery ratio of SAGD is high, it is badly influenced by voirs is found and in production in Athabasca, Cold Lake, strong reservoir heterogeneity, especially in the McMurray and Peace River, northeastern Alberta, Canada [3, 4]. The Formation which bears plenty of tidal couplets [8, 9]. bitumen in these reservoirs below 75 m only can be extracted The tidal couplets deposited and reworked in tidal, flu- commercially by in situ thermal methods, steam drive, and vial, or estuarine environments are mm-cm scale sandy and steam-assisted gravity drainage (SAGD) etc. [5, 6]. SAGD muddy laminae interlayers [10, 11]. It is necessary to under- utilizes an 800-1000 m long horizontal well pair, with the stand the origins, characteristics, and spatial distributions of two horizontal wells vertically aligned and placed at a vertical the tidal couplets to enhance SAGD well performance. distance of approximately 5 m apart. The upper well is used Numerous studies have been done to characterize these tidal

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/doi/10.2113/2021/2993537/5389492/2993537.pdf by guest on 27 September 2021 2 Lithosphere

Township 90 Township 90 Study area Range 14 Range 13 Alberta 100131409014W400 GR AI10 Sedimentary Depth–13 gAPI 139 1.0ohm.m 3,000.0 cycle m AI90 Lithology 27 26 25 Unit Epoch Period Short Middle

Formation 1.0ohm.m 3,000.0

150 Lower 155 N 22 23 24 Clearwater Fort McMurray 160 Upper

Wabiskaw 165 Lower

Athabasca 15 14 13 Lower 170

oil sands Cretaceous

175

Calgary Upper 180

10 11 12 McMurray 185 100 km

190

1 Middle 32500 m Carboniferous 195

(a) (b) (c)

Oilsands Muddy sands

Wells Mud

Study area Carbonate

Fine sands Fining upwards

Very-fne sands Coarsing upwards

Figure 1: Location maps showing (a) regional and (b) detailed study area. (c) A typical well (marked the red spot in (b)). Mackay River Project is located northwest of Fort McMurray, Alberta. Grey numbers in boxes indicate section of the LSD.

couplets. Six parameters, laminae types, thickness, sandy vol- dicted by object-based modeling. The aim of this paper is to ume, the ability allowing steam to go through, vertical per- propose a quantitative characterization process for mm-cm meability, and spatial scale, were proposed [12], but some scale couplets and to clarify the tidal couplets’ influences on of these parameters are so ideal that it is not easy to get in SAGD production. the oilfield. However, most specialists pay great attention on estimating the vertical permeability because of the SAGD 2. Geologic Background production mechanism. To do so, based on deposition process-based bedding-scale geomodeling strategy [13] and The bitumen bearing formations in Athabasca were devel- 2D stream-line method [14], the representative elementary oped in Alberta subbasin of Western Canadian Basin [20]. volume (REV) to estimate effective petrophysical properties The McMurray Formation is a member of the Manville of tidal couplets was calculated under the criterion of variat Group that was deposited as fluvial-estuarine-marginal suc- ion coef f icient ðCvÞ <0:5 [15, 16]. Further, a vertical perme- cessions due to the Boreal sea level rising during the middle ability estimation workflow for the McMurray Formation to late Cretaceous. It is bounded unconformably below by was proposed [17, 18], and an effective permeability estima- Carboniferous and above by the Wabiskaw Member, Creta- tion workflow for the upper McMurray Formation was pre- ceous (Figure 1). sented [19]. The study area, Mackay River Project (MRP), is located c. However, effective permeability estimation is not enough 30 km northwest of Fort McMurray, Alberta. The initial to demonstrate the influence resulted from tidal couplets on development area (IDA) of the MRP covers townships 90 SAGD production. The spatial distribution of tidal couplets and ranges 13 to 14 W4M. The McMurray Formation in between these SAGD well pairs must be taken into consider- Alberta is roughly subdivided into lower, middle, and upper ation, either. Therefore, this paper differs from previous stud- intervals [21]. Although there are three intervals in the ies in (1) the origin-based classification of TCs that was McMurray Formation, the lower counterparts is missing in discussed via core photos, logging, and literatures; (2) five MRP because of erosion. The middle and upper McMurray practical and easy-to-get parameters were proposed to char- intervals characterized by a series of clean, linear tidal sand acterize TCs; and (3) the spatial distribution of TCs was pre- bars contain most of the exploitable bitumen resources

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/doi/10.2113/2021/2993537/5389492/2993537.pdf by guest on 27 September 2021 Lithosphere 3

Ofshore sediments Onshore sediments

Shallow sea Channel bar

Tidal bar Salt marsh

Sand bar Erosion

Mixed fat

Figure 2: Sedimentary model of the McMurray Formation, MRP [24].

[22]. Our focus was the TCs in the middle and upper and frequency. Then, based on the origins of TCs, a deposi- intervals. tion process-based modeling strategy was utilized to build The middle and upper McMurray Formation in the the bedding-scale model. Further, according to REV, the Mackay River Project, CNPC, is dominated by six estuary- effective petrophysical properties were estimated under the originated microfacies (Figure 2). The tidal channel, salt criterion of CV <0:5 [15]. Besides, the spatial scale of each marsh, sand bar, and tidal bar are deposited in the middle type of couplets was estimated by high-density well correla- McMurray Formation [23]. Meanwhile, the mixed flat, sand tions. Finally, the spatial distribution of these couplets was bar, and tidal bar are dominated in the upper McMurray For- predicted by object-based modeling. mation. The features and distributions of TCs are controlled by the geometry and scale of these microfacies. 3.1. Mathematical Morphology Process. MMP was applied to quantify the features of TCs in core photos because of its sim- 3. Methods plicity and reliability [26]. The MMP uses a certain shape of structure element to analyze, measure, and extract the infor- The quantitative characterization of mm-cm scale TCs mation of images. In MMP, the binary value image, regarded includes following aspects: identification, classification, as a data set, was handled by a structure element which is a quantitative description, and spatial distribution prediction. data set, either. A series of calculation was conducted when Fortunately, previous literatures and some practical methods the structure element was moving through the image. The can be applied to accomplish these tasks. To begin with, the information remained in image depends on the shape of identification and classification of TCs could be done by core the structure element. photos and logging. Next, for quantitative description, firstly, the mathematical morphology process (MMP) [25] was 3.2. Modeling Strategies. The characterization and spatial applied to calculate the interlayers’ thickness, mud volume, prediction of TCs require different modeling strategies. For

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/doi/10.2113/2021/2993537/5389492/2993537.pdf by guest on 27 September 2021 4 Lithosphere

50 m) ⁎

� 40

RT ( 30

0 (a) (b) (c) (d) (e) (f) 0 20406080 100 120 140 GR/API Sands TCs

Figure 3: Core images in McMurray (left) and GR vs. RT crossplot (right) (blue: sands, orange: TCs). (a) 0803089, 181.9 m. (b) 0227090, 178.6 m. (c) 0413090, 186.3 m. (d) 0812090, 196.15 m. (e) 0605091, 182.9 m. (f) 0614090, 187.8 m.

West East

5 m

200 m

TCCs SBCs MFCs TBCs

Carbonate Shelf

Tidal channel Bioturbation

Tidal bar Rhythmics

Mixed fat Tidal bedding

Sand bar

Figure 4: The TCs’ classiﬁcation in middle and upper McMurray Formation, Mackay River Project.

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/doi/10.2113/2021/2993537/5389492/2993537.pdf by guest on 27 September 2021 Lithosphere 5

40 m) ⁎ 30 � RT (

0 0 20 40 60 80 100 120 140 GR/API TBCs SBCs MFCs TCCs

Figure 5: Logging classiﬁcation of each type of TCs by GR vs. RT crossplot.

(a) (b) (c) (d) (e) (f)

Figure 6: Process of MMP (041309014 well, 186.3-186.85 m). (a) Core photo. (b) Greyed photo. (c) Binary image. (d) Erosion. (e) Expansion. (f) Calculated area (yellow box).

the effective petrophysical property estimation, a bedding cies, lithofacies associations, intervals etc., seems to be less scale model is essential [13]. We recommend the process- flexible and variable. So, in this paper, the object-based based method for two reasons. First, the origins and features method was used to predict the spatial distribution of TCs. of each type of tidal couplet are analyzed through sedimentary interpretation and description. Secondly, process-based 3.3. REV Theory. The representative elementary volume modeling strategy has been applied successfully in the similar (REV) is adopted to estimate the effective petrophysical reservoir conditions. properties. The intrinsic variability in rock properties and As for the prediction of spatial distribution, two methods, geological characteristics at all scales is commonly referred object-based one and surface-based one, are both important to as “heterogeneity” [29]. Replacing a heterogeneous prop- and useful [27, 28]. The geobody in the model built by the erty field with a hypothetical homogeneous one is the notion object-based method can be in a variety of geometries and referred to as REV. It denotes a volume of the rock property spatial scales, while the surface-based method that utilizes that is large enough to capture a representative amount of the the surfaces and boundaries in each level, bedding, lithofa- heterogeneity [30]. The determination of this volume is

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/doi/10.2113/2021/2993537/5389492/2993537.pdf by guest on 27 September 2021 6 Lithosphere

70 25

60 20 50

15 40

30 10 Frequency (%) Frequency Frequency (%) Frequency 20 5 10

0 0 123456789101112 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 >24 Tickness (mm) Tickness (mm) (a) (b)

60 40

35 50 30 40 25

30 20

15 Frequency (%) Frequency 20 (%) Frequency 10 10 5

0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 8910 11 12 13 14 15 16 17 18 19 20 21 22 23 >24 Tickness (mm) Tickness (mm) (c) (d)

Figure 7: Thickness distribution of diﬀerent tidal couplets. (a) TBCs. (b) SBCs. (c) MFCs. (d) TCCs.

associated with the length scale. When the sample volume is thinnest mud laminae can be 2 mm, while the thickest one small compared to the length scale of heterogeneity, the mea- can be 5 cm (Figures 3(a)–3(d)); (2) they are in a variety of sured property will vary with small changes in the sample shapes and geometries because of different deposition micro- volume. At some volume, namely, the REV, the fluctuations environments and the degree of bioturbation (Figures 3(b) are minimized, and a representative amount of heterogeneity and 3(e)); and (3) they are densely clustered with high fre- can be confidently averaged in the measurement. Nordahl & quency sandy and muddy interlayers (Figure 3(f)). Ringrose [16] have shown that a lithofacies-scale REV can be In logging, TCs were identified by GR and RT crossplot achieved at the c. 0.3 m length scale for models of tidal het- because of their thickness and oil saturations. TCs identified erolithic bedding deposits. in logging are in high GR (40-120 API) and low RT (10- 50 Ω·m), while for the bitumen bearing sand intervals, gener- 4. Results ally, the value of GR is less than 40 API, and RT is more than 50 Ω·m (Figure 3 right). Based on core photos, lab analysis, logging, and drilling data, the identification, classification, characterization, and distri- 4.2. Classification. Many classification schemes were pro- bution prediction of TCs were done through mathematic posed for TCs from different point of views, such as mud vol- morphology process, bedding-scale modeling, REV, and ume, frequency, thickness of mud laminae, and the degree of object-based modeling. The details of each step were demon- bioturbation [31]. In the literatures, the classification scheme strated as follows. of mud volume is widely accepted. In our point of view, although the mud volume reflects lots of information, like 4.1. Identification. The 88 core wells in IDA or nearby pro- the sedimentary supply and hydrodynamic condition, it is vide a solid foundation for TC observation and description. not representative because the TCs originated from different The identification of TCs largely depends on high resolution environments may have the same mud volume [32]. There- core photos and logging. fore, according to sedimentary microfacies analysis, this In core photos, TCs offer the following characteristics: (1) paper proposed a geologic origin-related classification they are in mm-cm scale thickness with great variation. The scheme for TCs. In the origin-based classification scheme,

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/doi/10.2113/2021/2993537/5389492/2993537.pdf by guest on 27 September 2021 Lithosphere 7

60 51.06 50 45.29

30 26.54

20 Mud volume (%) volume Mud

9.57 10

0 TBCs SBCs MFCs TCCs (a)

70 64 60

30 25

20 Frequency (no./m) Frequency

10 8 3 0 TBCs SBCs MFCs TCCs (b) (c)

Figure 8: Mud volume and frequency (No./m) of each type of TCs and mud logs in varied intervals. (a) Mud volume. (b) Frequency. (c) Mud logs of 1 cm, 5 cm, 10 cm, and 30 cm intervals (060509114).

TCs can be divided into four types. They are tidal bar cou- mud volume, high frequency, and strong rhythmic. The plets (TBCs), sand bar couplets (SBCs), mixed flat couplets mud volume of MFCs is high with moderate bioturbations. (MFCs), and tidal channel couplets (TCCs). Each type of The BI ranges from 2 to 4. The most obvious feature is the TCs offers the following features (Figure 4). strong rhythmic sandy and muddy interlayers which indicate a rhythmic deposition process and abundant deposition 4.2.1. Tidal Bar Couplets (TBCs). TCs in tidal bars were supply. named tidal bar couplets (TBCs). The mud volume of TBCs is low. The dispersed mud laminae in TBCs are in mm-scale 4.2.4. Tidal Channel Couplets (TCCs). TCCs are found in thickness and low frequency with few bioturbations. The bio- tidal channels. Its mud volume is high with clustered thick turbation index (BI) [33] ranges from 0 to 1. The sedimentary mud laminae and high degree bioturbation. The thickness supply of sand is abundant, and the bedding structures well of mud laminae in TCCs can be up to 63 mm. The BI can preserved indicate a strong hydrodynamic condition. be 3-5. The structures of thick mud reworked by bioturbation form lots of irregular pores and holes. The mud volume and 4.2.2. Sand Bar Couplets (SBCs). The SBCs are TCs developed remained structures indicated a sedimentary environment in the sand bars. The mud volume of SBCs is moderate. The with low hydrodynamic condition. mud laminae are in varied thickness and clustered with few Based on logging identification of TCs and classification bioturbations. The thinnest laminae can be 2 mm, while the scheme, four types of TCs can be identified by GR and RT thickest one can be 42 mm. The BI ranges from 1 to 2. The crossplot. The results indicate that TBCs are in low GR (20- fine sediments supply increased, and there are wavy bed- 30 API) and high RT (40-50 Ω·m); SBCs are in moderate- forms in surrounding deposits with subtle slope. high GR (84-110 API) with moderate RT (25-35 Ω·m); MFCs are in moderate-high GR (85-115 API) and low RT (15- 4.2.3. Mixed Flat Couplets (MFCs). The laminae of MFCs 25 Ω·m); TCCs are in high GR (90-120 API) and low RT existed in mixed flat are easily recognized because of their (10-15 Ω·m) (Figure 5).

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/doi/10.2113/2021/2993537/5389492/2993537.pdf by guest on 27 September 2021 8 Lithosphere

ML1 1AB142209014 1AA102209014 1AA032309014 1AA141409014 1AA151409014 1AA161409014 1AA121309014 ML1′

ML2 1AA041309014 1AB131209014 1AA141209014 1AB121209014 1AA091209014 1AA081209014 1AA021209014 ML2′

ML01 N

ML01′

ML02

IDA ML02′ Cored wells Profles 2.5 km

TBCs MFCs SBCs TCCs

Figure 9: Two north-south well correlation proﬁles of IDA, Mackay River Project.

Table 1: The estimated spatial scale results of each TCs.

Thickness (m) Lateral length (m) Length/width Types Min Mean Max Min Mean Max Min Mean Max TBCs 0.2 1.2 3.8 125 235 732 1.88 2.81 5.86 SBCs 0.2 2.3 10.2 135 341 578 1.69 2.5 4.29 MFCs 0.2 0.35 0.5 64 106 167 1.58 1.65 2.61 TCCs 0.2 1.34 2.1 103 214 392 1.83 2.07 3.81

4.3. Quantitative Description. Five parameters, i.e., thickness, mud. Secondly, to ensure the accuracy, we only count the mud volume, frequency, spatial scale, and eﬀective proper- mud layers which go across the vertical axis of the core photo. ties, are used to describe each type of couplets quantitatively. Take SBCs as an example, (1) obtaining the pixels matrix Thickness, mud volume, and frequency are calculated by core of depth-corrected core images by gray processing and (2) photos. The spatial scale of each couplets is estimated by using a local pixel threshold to gain binary images. Pixels high-density well correlations. As for the petrophysical prop- above the threshold are set to be sand, while others are set erties, the porosity and permeability are estimated by the to be mud. The dark pixels denote sand while white one bedding scale model and REV. The details of results are as denotes mud; (3) erasing the noise pixels by erosion algo- follows. rithm via a rectangular structure element; (4) using expansion algorithm by the same structure element and for image 4.3.1. Thickness, Mud Volume, and Frequency. The thickness, restoration; and (5) counting the thickness and numbers of mud volume, and frequency of each type of TCs are calcu- the white pixels in yellow box. Further, the mud volume lated by core images processed by ﬁve steps. There are two and frequency can be calculated by the total pixels and the prerequisites of the MMP application in the study area. First, length of the core images. Additionally, the mud volume of it is assumed that only two kinds of deposits, i.e., sandy inter- TBCS and SBCs was calculated in every 50 cm intervals, while vals and muddy laminae, are in the target formation, while MFCs and TCCs’ are in 30 cm. The frequency was calculated the dark pixel represents sand and the white pixel represents by every 50 cm (Figure 6).

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/doi/10.2113/2021/2993537/5389492/2993537.pdf by guest on 27 September 2021 Lithosphere 9

16010 100000

14010 y = 0.9001x – 624.8 R2 = 0.9145 12010 10000

10010

8010 1000

Kv (mD) Kv 6010

4010 (mD) Permeability 100

2010

10 10 10 2010 4010 6010 8010 10010 12010 14010 16010 0.27 0.29 0.31 0.33 0.35 0.37 0.39 0.41 Kh (mD) Porosity (%) Kv Kh (a) (b)

Figure 10: Petrophysical property of core plugs in IDA. (a) Core plug Kh and Kv crossplot on a log scale and in original units (sample No. 328). (b) Core plug porosity and permeability crossplot (blue: Kv, orange: Kh).

Y (cm) X (cm) Y (cm) X (cm) Y (cm) X (cm) 0 22 0 2 0 2 684 4 6 4 22 4 Permeability (mD) 6 4 4 Porosity (fraction) 8 8 6 6 8 8 6 8 0 0 0 0 5500 0 0

0.35 5000

10 10 10 10 4500 10 10 0.3 4000

3500 0.25 20 20 20 20 20 20 3000 0.2 Z (cm) Z (cm) 2500 Z (cm) 30 30 30 30 30 30 2000 0.15

1500 40 40 40 40 40 40 0.1 1000

500 0.05

8 688 8 4 22 4 6 6 4 4 6 8 6 4 6 8 5 cm 0 2 0 2 2 0 2 4 Y (cm) X (cm) Y (cm) X (cm) Y (cm) X (cm) (a) (b) (c) (d)

Figure 11: Bedding scale models of MFCs (mud volume =36:44%). (a) Core images. (b) Geometry model. (c) Permeability. (d) Porosity realization. All models are presented in core view.

4.3.2. Thickness. Results indicate that thickness of each type 52.13% with the average of 51.06%. As for TCCs, its thickness of TCs varies. The thickness of TBCs ranges from 1 to ranges from 36.47% to 49.17% with average of 45.29%. 12 mm with the average value of 3 mm. For SBCs, the thick- Meanwhile, mud volume logs of 1 cm, 5 cm, and 10 cm inter- ness ranges from 2 to 42 mm with the average value of vals can be calculated by changing moving windows through 11 mm. The thickness of MFCs ranges from 1 to 16 mm with the whole well (Figure 8). the average value of 2 mm. As for TCCs, its thickness ranges from 8 to 63 mm with average value of 18 mm (Figure 7). 4.3.4. Frequency. The frequency changes greatly. The frequency of TBCs is relatively low, ranging from 1 to 5 4.3.3. Mud Volume. Similarly, the mud volume of diﬀerent No./m with the average of 3. For SBCs, the frequency ranges types of TCs changes. The mud volume of TBCs ranges from from 5 to 10 No./m with the average of 25. The thickness of 1% to 12.13% with the average value of 9.57%. For SBCs, the MFCs ranges from 32 to 66 No./m with the average of 64. value ranges from 13.24% to 31.02% with the average of As for TCCs, its thickness ranges from 3 to 10 No./m with 26.54%. The thickness of MFCs ranges from 43.97% to the average of 8 (Figure 8).

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/doi/10.2113/2021/2993537/5389492/2993537.pdf by guest on 27 September 2021 10 Lithosphere

70 5000 4500 60 4000 y = –4547.9x + 4363.7 50 3500 R2 = 0.9517 3000 40 2500 30 2000

Permeability (mD) Permeability 1500 20 y = –855.3ln(x) – 406.91 1000 R2 = 0.9648 10 500 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 0 0.1 0.2 0.3 0.4 0.5 0.6 Kv/Kh Mud volume (%) (a) (b) KVE Kv KC Kh

Figure 12: Petrophysical properties from the bedding model. (a) Kv/Kh histograms (blue: Kv/Kh of core plugs with mean value 0.74; range: eﬀective Kv/Kh with mean value 0.36. (b) KE and mud volume crossplot.

Figure 13: Estimated horizontal and vertical permeability on a log scale, 1AA141209014 well. Gamma ray (track 1), RT (track 2), and depth in MD (track 3); facies division (track 4); core images (track 5); Vshale for 1 cm (track 6); Vshale for 5 cm (track 7); Vshale for 10 cm (track 8); binary sand-mud division (track 9); Kve log (track 10, purple dots denote core plug Kv) at a vertical resolution of 5 cm; Khe at a vertical resolution of 5 cm (track 11, red dots denote core plug Kh); and core plug porosity (track 12).

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/doi/10.2113/2021/2993537/5389492/2993537.pdf by guest on 27 September 2021 Lithosphere 11

A A′

A B′

B B′ B

A′

Figure 14: Spatial distribution prediction of TCs (purple: TBCs; green: SBCs; blue: MFCs; black: TCCs).

(1) Spatial Scale Estimation. The spatial scale estimation of Kv/Kh is corrected from 0.74 to 0.38. Further, the mud vol- TCs is mainly based on high-density well correlations ume log with a 10 cm moving window was used to estimate (Figure 9). The distance between wells in IDA ranges from the horizontal and vertical permeability from the bedding 250 to 1500 m. Two section profiles were selected carefully. scale model results. The upscaled permeability in each inter- The results are in Table 1. val is much lower than core-based permeability (Figure 13). This is generally true when we take small scale bedding struc- (2) Effective Property Estimation. To populate the bedding tures and mud volume into consideration. scale models, the lab analysis data of core plugs from the 13 wells were assembled. In Figure 10, the relationships between 4.4. Spatial Distribution Prediction. According to the thick- core-based horizontal and vertical permeability and between ness, mud volume, frequency, and spatial scale of each TCs, the permeability and porosity are presented (328 samples). an object-based geomodeling strategy was applied to forecast Because the tidal deposits in the McMurray Formation are the spatial distribution of TCs. Take the SBCs for instance, unconsolidated and the core plugs are often taken in pure firstly, the fraction of SBCs (1.89%) is set according to the sand intervals, the measured Kv, Kh, and Kv/Kh are com- upscaled well data. Secondly, the body shape, orientation, monly biased to higher values. The Kv and Kh range from minor width, Maj/Min ratio, and thickness of SBCs are set 10 to 14000 mD with a mean value of 4500 mD while Kv/Kh based on the spatial scales estimated by high-density well ranges from 0.1 to 1.3 with a mean value of 0.74 (Figure 10). correlations. Finally, SBCs can only be inserted in sand bars. Besides, there is no linearity between porosity and permeabil- Similarly, the models for the other types of TCs can be built ity. High permeability and the lack of a direct relationship (Figure 14). between porosity and permeability mean that estimations Spatial prediction results indicate that TCs are developed are necessary before property modeling. in the middle-bottom parts of the target formation. TBCs are in the middle of the formation. SBCs are mainly in the south- Therefore, to prepare the bedding scale modeling, a mean ern area, and MFCs are in the eastern area. Further, the spa- input porosity value for sand is 0.32, while for mud, it sets as tial distribution of TCs in 8 production pads can be predicted 0.1. A mean permeability value of 4500 mD, with standard (Figure 15). deviation of 200 mD, was assigned to the sand laminae. The permeability of mud laminae was assigned a mean input 5. Conclusions value of 100 mD, with standard deviation of 20 mD. Besides, the quantitative description data of each type of TCs were In this paper, a quantitative characterization procedure for also set in the model. Additionally, the size of the bedding mm-cm scale tidal couplets (TCs) was proposed. Details of scale model was in 10 × 10 × 50 cm3. Figure 11 gives an exam- this procedure, which includes identification, classification, ple of the MFC bedding scale model. quantitative description, and spatial distribution prediction, Finally, based on REV, the bedding scale modeling and are presented in the case study of the oil sands reservoir of property modeling process were done by SBED®. The prop- the McMurray Formation, Mackay River Project, and CNPC. erty models in this process are upscaled by single phase TCs in photos and logging are in mm-cm thickness, flow-based simulation and fixed boundary condition to densely clustered, and in a variety of geometries. There are obtain effective properties for a given bedding structure. four types of tidal couplets in the target formation. They The criterion for REV is Cv <0:5. To ensure the accuracy are named tidal bar couplets (TBCs), sand bar couplets and reliability, 50 realizations of each TC bedding scale (SBCs), mix flat couplets (MFCs), and tidal channel couplets model were done. The results show that the TCs reduce the (TCCs), respectively, according to surrounding microfacies. effective petrophysical properties greatly (Figure 12). The Five parameters were selected to describe TCs. Thickness, TBCs, SBCs, MFCs, and TCCs reduce 9%, 25%, 47%, and mud volume, and frequency were calculated by MMP core 44% of the porosity; 10%, 27%, 51%, and 45% of the Kh; photos. Spatial scale was predicted by high-density well cor- and 64%, 84%, 96%, and 94% of the Kv of the reservoir. relations, and the effective properties were estimated by a

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/doi/10.2113/2021/2993537/5389492/2993537.pdf by guest on 27 September 2021 12 Lithosphere

340 340 320

330 330

310 310

320 320

310 310 300 300

300 300

290 290

290 290 430500 430750 431000 431250 431500 431750 432000 432250 432500 6297000 6296750 6296500 6296250 6296000 AA pad AB pad

310 310

300 300

290 290 290 290

430750 434000 431250 6295500 6295750 6296000 6296250 AC pad AD pad

320 320 320 320

310 310 310 310

300 300 300 300

290 290 290 290

6295600 6295200 6294800 6298000 6298250 6298500 6298750 6299000 AE pad AF pad

320 320 310 310

310 310

300 300

300

290

290 433500 433250 434250 434000 433750 6297000 6296750 6296500 AH pad AJ pad

Figure 15: Distribution of TCs in 8 pads of IDA (purple: TBCs; green: SBCs; blue: MFCs; black: TCCs).

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/doi/10.2113/2021/2993537/5389492/2993537.pdf by guest on 27 September 2021 Lithosphere 13

combined modeling method. The reductions of petrophysical [7] A. M. al-Bahlani and T. Babadagli, “SAGD laboratory experi- properties, especially the Kh, caused by TCs and the spatial mental and numerical simulation studies: a review of current distribution of TCs in each well pad explain why the SAGD status and future issues,” Journal of Petroleum Science and production performance in the target formation is not Engineering, vol. 68, no. 3-4, pp. 135–150, 2009. so ideal. [8] M. R. Gray, Upgrading oilsands bitumen and heavy oil, Can- Obviously, other sources of data, like Formation Micro- ada, University of Alberta Press, 2015. resistivity Images (FMI) [34], can be used in the process of [9] P. Yin, G. Liu, C. Liu, H. Li, W. Liu, and Y. Liu, “Depositional quantitative description of TCs. Besides, the bedding-scale patterns of oil sand layer in Northern Lights block, Alberta ” – model could also be done by surface-based modeling strategy Basin, Canada, Oil & Gas Geology, vol. 32, no. 6, pp. 807 [27, 28]. The results of effective petrophysical properties can 814, 2011. further be upscaled for reservoir simulation which is not [10] J. R. Boersma, “Internal structures of some tidal megaripples on a shoal in the Westerschelde Estuary, the Netherlands,” mentioned here. Therefore, more work needs to be done for – better understanding of the TCs in the McMurray Formation Geologie en Mijnbouw, vol. 48, no. 4, pp. 409 419, 2007. “ and its influence on SAGD production. [11] D. Fan, C. Li, M. Chen, D. J. Wang, and P. X. Ding, Quantita- tive analyses of diastems of the mudflat deposits in the Yangtze river delta,” Marine Geology and Quaternary Geology, vol. 21, Data Availability no. 4, pp. 1–6, 2001. The data comes from the Mackay River Oil Sands Project, [12] Y. Hu, M. Chan, S. Bharatha, G. Gurba, D. Lee, and S. Papadimitrio, “Mudstone interbeds in McMurray oil sand CNPC. reservoirs in Alberta, Canada,” Earth Science-Journal of China University of Geosciences, vol. 29, no. 5, pp. 550– Conflicts of Interest 554, 2004. “ fl [13] R. Wen, A. W. Martinius, A. Naess, and P. Ringrose, Three- The authors declare that they have no con icts of interest. dimensional simulation of small-scale heterogeneity in tidal deposits - a process-based stochastic method,” in Proceedings Acknowledgments of the 4th Annual Conference of the International Association of Mathematical Geology, pp. 129–134, Ischia De Frede, We would like to acknowledge the CNPC for allowing us to Naples, 1998. publish the data on Mackay River Oil Sands Project. It is also [14] H. Haldorson and L. Lake, “A new approach to shale manage- appreciated that researchers in Research Institute of Petro- ment in filed scales simulation models,” Journal of Petroleum leum Exploitation and Development (RIPED) help us in well Technology, vol. 24, no. 4, pp. 1–4, 1984. log analyzing, bedding scale modeling, and property model- [15] K. Nordahl and P. S. Ringrose, “Identifying the representative ing. Additionally, we are grateful for the financial support elementary volume for permeability in heterolithic deposits from the National Science and Technology Major Project of using numerical rock models,” Mathematical Geosciences, China (Grant No. 2016ZX05031) and the National Natural vol. 40, no. 7, pp. 753–771, 2008. Science Foundation of China (Grant No. 41472119). [16] P. S. Ringrose, A. W. Martinius, and J. Alvestad, “Multiscale geological reservoir modelling in practice,” Geological Society, London, Special Publications, vol. 309, no. 1, pp. 123–134, References 2008. “ [1] X. Tong, G. Zhang, Z. Wang et al., “Distribution and potential [17] C. V. Deutsch, Estimation of vertical permeability in the ” of global oil and gas resources,” Petroleum Exploration and McMurray formation, Journal of Canadian Petroleum Tech- – Development, vol. 45, no. 4, pp. 727–736, 2018. nology, vol. 49, no. 12, pp. 10 18, 2010. “ fl ” [2] C. Zou, G. Zhai, G. Zhang et al., “Formation, distribution, [18] P. Raats, Dynamics of uids in porous media, Engineering potential and prediction of global conventional and unconven- Geology, vol. 7, no. 4, pp. 174-175, 1972. tional hydrocarbon resources,” Petroleum Exploration and [19] H. Chen, L. Mu, J. Huang, and G. Chang, “Bedding-scale Development, vol. 42, no. 1, pp. 14–28, 2015. geomodelling for effective permeability estimation in the [3] Y. Liu, Y. Xie, Z. Zhang, X. Yang, and D. Wang, “Sequence upper McMurray Formation, NE Alberta, Canada,” Petroleum stratigraphy and evaluation of oil sands resources in block Geoscience, vol. 26, pp. 101–109, 2019. M, Western Canada Basin,” Science Technology and Engineer- [20] F. J. Hein, K. P. Parks, D. A. Leckie, C. Seibel, and D. Morrow, ing, vol. 17, no. 10, pp. 13–21, 2017. “Foreword — Oil-sands and heavy-oil deposits: Local to global [4] H. Wang, F. MA, X. Tong et al., “Assessment of global uncon- multidisciplinary collaboration, ” Bulletin of Canadian Petro- ventional oil and gas resources,” Petroleum Exploration and leum Geology, vol. 64, no. 2, pp. 99–105, 2016. Development, vol. 43, no. 6, pp. 925–940, 2016. [21] M. Fustic, S. M. Hubbard, R. Spencer et al., “Recognition of [5] S. K. Das and R. M. Butler, “Mechanism of the vapor extrac- down-valley translation in tidally influenced meandering tion process for heavy oil and bitumen,” Journal of Petroleum fluvial deposits, Athabasca Oil Sands (Cretaceous), Alberta, Science and Engineering, vol. 21, no. 1-2, pp. 43–59, 1998. Canada,” Marine and Petroleum Geology, vol. 29, no. 1, [6] G. Liang, S. Liu, P. Shen, Y. Liu, and Y. Luo, “A new optimiza- pp. 219–232, 2012. tion method for steam-liquid level intelligent control model in [22] F. J. Hein and D. K. Cotterill, “The Athabasca oil Sands — A oil sands steam-assisted gravity drainage (SAGD) process,” regional geological perspective, Fort McMurray area, Alberta, Petroleum Exploration and Development, vol. 43, no. 2, Canada,” Natural Resources Research, vol. 15, no. 2, pp. 85– pp. 275–280, 2016. 102, 2006.

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/doi/10.2113/2021/2993537/5389492/2993537.pdf by guest on 27 September 2021 14 Lithosphere

[23] M. Tang, K. Zhang, J. Huang, and S. Lu, “Facies and the architecture of estuarine tidal bar in the lower Cretaceous Mcmurray Formation, Central Athabasca Oil Sands, Alberta, Canada,” Energies, vol. 12, no. 9, p. 1769, 2019. [24] R. W. Dalrymple, D. A. Mackay, A. A. Ichaso, and K. S. Choi, “Processes, morphodynamics, and facies of tide-dominated estuaries,” in Principles of Tidal Sedimentology, pp. 79–107, Springer, Netherlands, 2012. [25] K. Wisaeng and W. Sa-ngiamvibool, “Improved fuzzy C- means clustering in the process of exudates detection using mathematical morphology,” Soft Computing, vol. 22, no. 8, pp. 2753–2764, 2018. [26] H. Chen, L. Mu, J. Huang, G. Chang, J. Wu, and T. Sun, “Quantitative characterization of mud laminae based on mathematical morphology image process, a case study of Mackay River oil sands reservoir,” Science Technology and Engineer- ing, vol. 18, no. 26, pp. 77–83, 2018. [27] B. Y. Massart, M. D. Jackson, G. L. Hampson, H. D. Johnson, B. Legler, and C. A. L. Jackson, “Effective flow properties heterolithic, cross-bedded tidal sandstones: part 1. Surface-based modeling,” AAPG Bulletin, vol. 100, no. 5, pp. 697–721, 2016. [28] B. Y. Massart, M. D. Jackson, G. J. Hampson, and H. D. Johnson, “Effective flow properties heterolithic, cross-bedded tidal sandstones: part 2. Flow simulation,” Flow simulation, AAPG Bulletin, vol. 100, no. 5, pp. 723–742, 2016. [29] P. J. Fitch, M. A. Lovell, S. J. Davies, T. Pritchard, and P. K. Harvey, “An integrated and quantitative approach to petrophysical heterogeneity,” Marine and Petroleum Geology, vol. 63, pp. 82–96, 2015. [30] J. Bear, Dynamics of Fluids in Porous Media, Courier Corpora- tion, 2013. [31] S. M. Johnson and S. E. Dashtgard, “Inclined heterolithic stratification in a mixed tidal-fluvial channel: Differentiating tidal versus fluvial controls on sedimentation,” Sedimentary Geol- ogy, vol. 301, pp. 41–53, 2014. [32] E. R. Timmer, M. K. Gingras, M. L. Morin, M. J. Ranger, J. P. Zonneveld, and G. Plint, “Laminae-scale rhythmicity of inclined heterolithic stratification, lower Cretaceous McMur- ray Formation, NE Alberta, Canada,” Bulletin of Canadian Petroleum Geology, vol. 64, no. 2, pp. 199–217, 2016. [33] A. M. Taylor and R. Goldring, “Description and analysis of bioturbation and ichnofabric,” Journal of the Geological Society, vol. 150, no. 1, pp. 141–148, 1993. [34] O. Babak and J. Resnick, “A workflow for vertical and horizontal near-wellbore permeability modelling in the McMurray Formation,” Petroleum Geoscience, vol. 22, no. 2, pp. 177– 189, 2016.

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/doi/10.2113/2021/2993537/5389492/2993537.pdf by guest on 27 September 2021