Accepted Manuscript

Sedimentary basin geothermal favourability mapping and power generation

assessments

K. Palmer-Wilson, J. Banks, W. Walsh, B. Robertson

PII: S0960-1481(18)30503-2
DOI: 10.1016/j.renene.2018.04.078
Reference: RENE 10036

To appear in: Renewable Energy

Received Date: 7 December 2017

Accepted Date: 27 April 2018

Please cite this article as: Palmer-Wilson K, Banks J, Walsh W, Robertson B, Sedimentary basin
geothermal favourability mapping and power generation assessments, Renewable Energy (2018), doi:
10.1016/j.renene.2018.04.078.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

1 Sedimentary Basin Geothermal Favourability Mapping and Power

2 Generation Assessments
3
4 K. Palmer-Wilsona*, J. Banksb, W. Walshc, B. Robertsona
5
a
6 Institute for Integrated Energy Systems, University of Victoria

PT
7 PO Box 1700 STN CSC, Victoria BC V8W 2Y2, Canada
8
b
9 Earth & Atmospheric Sciences, University of Alberta

RI
10 4-02 ESB, Edmonton AB T6G 2E3, Canada
11
c
12 British Columbia Ministry of Energy, Mines & Petroleum Resources

SC
13 PO Box 9314 Stn Prov Govt, Victoria BC V8W 9N1, Canada
14
15 * Corresponding author: 1-250- 721-6292, kevinpw@uvic.ca

16 Abstract
U
AN
17 Globally, sedimentary basin geothermal energy could prove a significant source of low-carbon
18 electricity, but regional resource assessments require collecting high cost sub-surface data. This study
19 applies freely available petroleum production data in a comprehensive approach to: 1) identify
M

20 favourable locations for geothermal energy development and 2) to estimate electric power generation
21 potential at those locations. A geothermal favourability map identifies favourable locations by
22 overlaying geological and economic criteria. Power generation estimates are based on the Volume
D

23 Method, which derives power capacity from the thermal energy present in a reservoir. As a method case
24 study, the northeastern British Columbia section of the Western Canada Sedimentary basin is analyzed.
TE

25 Here, four favourable areas are identified (Horn River, Clarke Lake, Prophet River and Jedney) and have
26 a total power capacity of 107.3 MW (distribution mode). Values normalized by reservoir volume range
27 from 1.8 – 4.1 MW/km3. Geothermal brine flow rates required to produce 1 MW of electric power range
28 from 27.5 – 60.4 kg/s. Reservoir size is derived from stratigraphic cross sections and natural gas pool
EP

29 outlines. Uncertainty in reservoir parameters are modeled using Monte Carlo simulations.

30 Keywords
C

31 geothermal energy; power generation; volume method; favorability map; resource assessment;
32 sedimentary basin
AC

1
 ACCEPTED MANUSCRIPT

33 1 Introduction
34 Geothermal energy is a baseload, renewable power source that provides ~13 gigawatts of electrical
35 power and ~70 gigawatts of thermal power for global human consumption (Bertani 2015; Lund & Boyd
36 2015). The majority of this power is generated from convection-dominated geothermal systems, where
37 point heat sources in the shallow subsurface elevate the geothermal gradient above the global mean
38 and create circulation cells of hot, easily accessible ground water (Moeck 2014). Convection-dominated

PT
39 geothermal systems are located predominantly in, or near, tectonically and volcanically active regions,
40 and offer limited opportunities for long-term industry expansion. Recent trends in geothermal energy
41 development have focused on conduction-dominated and low-enthalpy systems. Such geothermal

RI
42 systems are most commonly found in sedimentary basins and continental interiors, and are often more
43 proximal to potential geothermal power end-users than convection-dominated systems (Moeck 2014).

44 A major challenge in developing new geothermal fields is the high up-front cost of exploration and

SC
45 resource characterization (Salmon et al. 2011). These risks are compounded in conduction-dominated
46 and low enthalpy geothermal systems, which typically have no surface expression and host only low to
47 mid-grade resources. Realizing the potential of these types of geothermal resources requires the

U
48 development of inexpensive and expedient methods for identifying and evaluating unmeasured
49 geothermal reservoirs.
AN
50 This paper presents a two-step method for locating and quantifying the power generation potential of
51 conduction-dominated geothermal systems hosted in sedimentary basins. First, a favourability map is
52 produced by considering the influence of several geological and economic criteria on the feasibility of
M

53 constructing a geothermal power plant in a given locality. Second, the Volume Method (Williams et al.
54 2008) quantifies the power generation potential of areas deemed most favourable. The results reveal
55 both the total gross power available to local end users, as well as power generation potential per unit
D

56 reservoir volume.
TE

57 The British Columbian portion of the Western Canadian Sedimentary Basin (WCSB) serves as a proxy for
58 sedimentary basins with similar geothermal regimes and as a case study for this work. A nation-wide
59 evaluation of Canada’s geothermal resources identified the WCSB as an area of moderate commercial
60 potential (Grasby et al. 2012). Localized research in the WCSB has included qualitative evaluations of
EP

61 specific Cambrian and Devonian formations (e.g. Weides, Moeck, Majorowicz, et al. 2013; Weides,
62 Moeck, Majorowicz, et al. 2014; Weides, Moeck, Schmitt, et al. 2014) and power potential assessments
63 in the Alberta foothills (Banks and Harris, unpublished results). The British Columbian portion of the
C

64 WCSB has received less attention, but several areas of high geothermal potential were identified here by
65 Fairbank et al. (1992) and Kimball (2010). Fairbank et al. (1992) applied a geothermal gradient > 45 °C to
AC

66 identify areas of high potential, while Kimball applied several geological and economic factors to map
67 geothermal favourability. These studies offer relatively coarse resolution due to their Provincial scale,
68 but warrant regionally focused research.

69 Two previous studies have quantified geothermal power generation potential in the British Columbian
70 section of the WCSB by applying the Volume Method (Williams et al. 2008). Walsh (2013) investigated
71 the Clarke Lake gas field, near the town of Fort Nelson. Geoscience BC evaluated 18 different regions
72 throughout British Columbia (Geoscience BC 2015). That study included two sites in the WCSB, namely
73 Clarke Lake and Jedney. Neither of these studies chose their locations based on systematic criteria; they
74 were simply hypothesized to be potential resources. Through the two-step process of favourability

2
 ACCEPTED MANUSCRIPT

75 mapping and quantitative resource assessment, this study is the first to provide a focused examination
76 of the geothermal resource potential of the entire British Columbian portion of the WCSB. Similar two-
77 step analyses of the more densely populated island of Sicily, Italy, are not applicable in the WCSB,
78 because they do not consider proximity to energy consumers or transmission infrastructure (Trumpy et
79 al. 2015; Trumpy et al. 2016).

80 The paper outlines the geological background of geothermal energy in the WCSB in Section 2. Methods

PT
81 for producing a favourability map, and input data derived from the case study area are described in
82 Sections 3.1 to 3.3. Power generation potential estimates, methods and data are described in Sections
83 3.4 to 3.6. Results are described in Section 0 and followed by a discussion in Section 5. Conclusions from

RI
84 this work are drawn in Section 6.

85 2 Geologic Background of the Western Canada Sedimentary Basin

SC
86 The WCSB is a 600 – 1200 km wide NW – SE trending wedge of Proterozoic to modern sediment that
87 spans over 3000 km from the Northwest Territories to Montana. The WCSB is bordered to the east by
88 the Canadian Shield and to the west by the front ranges of the North American Cordillera, where it is

U
89 greater than 5500 m deep adjacent to the mountain belt. Figure 1 shows the geographic location of the
90 WCSB and surrounding geologic terranes. The study area is delimited to the west and south by the
AN
91 Cordilleran deformation front, to the north by the Yukon Territory border (60 °N latitude) and to the
92 east by the Alberta border (120 °W longitude).
M
D
TE
C EP
AC

93
94 Figure 1 Modern tectonic and geographic setting of the Western Canadian Sedimentary Basin (adapted from Mossop & Shetsen
95 1994; Banks & Harris, unpublished results).

96 A collage of Precambrian crustal blocks and suture zones that formed the passive continental margin of
97 North America’s ancestral west coast underlies the WCSB. Sediments deposited upon this Precambrian

3
 ACCEPTED MANUSCRIPT

98 basement preserve a long and relatively undisturbed history of transgressive and regressive marine
99 sequences (Mossop & Shetsen 1994, chap.3). Prominent among these sequences are thick deposits of
100 Devonian carbonate reefs and platforms; shales and fluvial and beach margin sandstones (Mossop &
101 Shetsen 1994, chaps.10–13). The Devonian reefs and sandstones in the deeper parts of WCSB are
102 hypothesized to contain reservoirs of hot brine, depending on local porosity (Walsh 2013; Weides &
103 Majorowicz 2014; Banks & Harris, unpublished results). Formations of primary interest in this study are
104 the Slave Point, Keg River and Chinchaga. Other potential water-bearing formations include the Sulphur

PT
105 Point, Wokkpash and Stone. All of these formations are major constituents of the Upper Givetian to
106 Lower Frasnian Beaverhill Lake Group, as it is found north of the Peace River Arch (Mossop & Shetsen
107 1994, chap.11). The local thicknesses of these strata play an important role in the volumetric

RI
108 calculations of power generation potential, described in Section 3.5.

109 Extensive hydrocarbon development throughout the WCSB has created a robust set of thermo- and

SC
110 hydrodynamic data to evaluate the basin’s geothermal setting. Heat flow within and beneath the WCSB
111 has been studied for over half a century (e.g. Garland & Lennox 1962; Majorowicz & Jessop 1981; Lam et
112 al. 1985; Jones et al. 1985; Majorowicz et al. 2012). Figure 2 (Weides & Majorowicz 2014) shows the
113 geothermal gradients throughout the WCSB. Gradients range from 20 – 25 °C/km in the southern and

U
114 southwestern-most portions, to > 50 °C/km in the far north of the basin. Within the study area, the
AN
115 basin deepens from about 2000 m in the upper northeast corner to about 5500 m in the west-central
116 section, adjacent to the deformation front. Temperatures at the bottom of the basin (top of the
117 Precambrian surface) range from ~100 °C in the shallower sections of the basin, where the gradient is
118 high, to > 160 °C in the deepest parts of the area, where the gradient is lower (Weides & Majorowicz
M

119 2014).
D
TE
C EP
AC

120
121 Figure 2 Geothermal gradient in the Western Canada Sedimentary Basin. The study area is located in the Canadian province of
122 British Columbia. The center province is Alberta, adjacent to Saskatchewan to the east. The southern border (49 °N latitude)
123 delimits the United States of America (adapted from: Weides & Majorowicz 2014).

4
 ACCEPTED MANUSCRIPT

124 3 Methods and Materials

125 This study employs a two-step method to locate favourable sites and quantify their potential for
126 geothermal power generation in a sedimentary basin. First, a geothermal favourability map is produced
127 by geospatially overlapping data that relate the technical and economic potential of a geothermal power
128 project. Second, the electricity generation potential from areas of highest favourability is assessed via
129 the Volume Method (Williams et al. 2008). As a case study, the method is applied in the British

PT
130 Columbian portion of the WCSB.

131 3.1 Favourability Mapping Procedure

132 Selecting favourable sites for geothermal power development is a geospatial multi-criteria decision

RI
133 problem (Greene et al. 2011). This study evaluates favourability based on geologic and economic
134 criteria. Geologic criteria include reservoir temperature and hydraulic conductivity. Economic criteria

SC
135 include access to electrical transmission grids, potential for behind-the-fence electrification of the
136 upstream petroleum sector, and proximity to population centers.

137 A geographic information system (i.e. ArcGIS) overlays sets of geospatial data pertaining to the

U
138 aforementioned criteria into a ‘geothermal favourability map’. The map visualizes the ‘favourability
139 score’ at any given location. The favourability score is the weighted sum of individual criteria scores,
AN
140 which measure the degree to which criteria are satisfied by input data. Identifying areas with high
141 favourability scores narrows the regional scope for a detailed estimate of power generation potential.

142 As shown in Figure 3, the favourability score is generated by a two-stage weighted linear combination
M

143 (Malczewski 2000; Malczewski & Rinner 2015). The left side of the flow diagram shows the two
144 geological and four economic input layers; these contain the criteria scores. In the Weighted Summation
145 Process section input layers are combined to form geological and economic summary layers.
D

146 Aggregating the summary layers forms the final favourability map. The scores in each input and
147 summary layer are multiplied by the given weights in each summation step.
TE
C EP
AC

5
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
C EP

148
149 Figure 3 Flowchart of mapping geothermal favourability. Geological and economic criteria are represented by input layers which
AC

150 can consist of several data sets. Input layer weighting and summation produces summary layers. Summary layer weighting and
151 summation produces the favourability map. Weights shown are used for the northeastern British Columbia case study.

152 Input data cannot be summed directly because data units and measurement scales differ. To enable
153 weighted summation, the input data is normalized to a common unit-less scale, or ‘criteria score’
154 (Voogd 1983, Chapter 5; Massam 1988; Nyerges & Jankowski 2010, Chapter 7.1). Criteria scores range
155 from 0 to 1 and identify the degree to which a specific criteria is satisfied at a given location. Datasets
156 used to produce input layers are described in Section 3.2.

6
 ACCEPTED MANUSCRIPT

157 Input layers consist of georeferenced 100 m x 100 m cells containing individual criteria scores. The score
158 Sij is contained in the jth cell of the ith layer. Each weighted summation stage receives inputs from N
159 input layers that consist of J cells. The jth cell of every input layer is geospatially congruent. Scores of
160 congruent cells are multiplied by the weight wi of the ith layer and summed as follows:


̅ = 
  ∀ ∈ 1,2, … , 

(1)

PT


161

162 As shown in Figure 3, first-stage summations produce summary layer scores and the second-stage
summation produces favourability scores. Depending on the summation stage,
̅ is the summary score

RI
163
164 or favourability score of the th cell. Weights reflect a criteria’s importance to geothermal development.
165 The sum of all weights  in each stage equals 1.

SC
166 The favourability scores are plotted on a map using a colour scale. This favourability map highlights
167 locations where geothermal power development is most favourable. Areas with a high score are
168 selected for assessment of potential power generation.

169 3.2 Favourability Mapping Input Data

U
AN
170 3.2.1 Geologic Criteria
171 Geological data is derived from drill-stem tests (DST), bottom-hole temperature measurements (BHT),
172 and natural gas and oil producing well records (NGOW) taken from IHS Energy (2016). All three datasets
M

173 are filtered for lower middle-Devonian and Beaverhill Lake Group strata (Mossop & Shetsen 1994,
174 Chapter 11). Figure 4 shows locations where data was recorded. Red crosses mark temperature and blue
175 dots mark indicated aquifer data records.
D

176 The Temperature Input Layer uses DST and BHT data. BHTs are Harrison-corrected (Harrison et al. 1983)
177 to account for the cooling effect of drilling mud. Spatial interpolation and averaging of temperature data
TE

178 produces the underlying temperature map. Areas below 80 °C are excluded because this temperature is
179 the minimum value for geothermal resources under the British Columbian Geothermal Resources Act
180 (1996).
C EP
AC

7
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
C EP

181
AC

182 Figure 4 The two geological input layers are based on temperature (red crosses – BHT and DST) and indicated aquifer (blue dots
183 – DST and NGOW) data records. Spatial interpolation and averaging produces the temperature map. Temperature data records
184 vary in depths from approximately 1400 m in the northeast to 4000 m around Sikanni Chief in the southwest of the study area.
185 Locations of data records extend into British Columbia’s neighbouring provinces to avoid edge effects along provincial borders.
186 The map shows the northeastern section of British Columbia and adjacent to sections of Alberta to the east, the Northwest
187 Territories to the northeast and Yukon to the northwest.

188 The Indicated Aquifer Input layer qualitatively infers hydraulically conductive strata by including DST and
189 NGOW that have recovered water. NGOW data are filtered for water production greater 0 m³/day, a
190 minimum production duration of 1 hour, a cumulative water production greater than 10 m³, and

8
 ACCEPTED MANUSCRIPT

191 production zones outside of shale formations. DST data is filtered by ‘blow test’ results. This test
192 analyzes the fluid exiting the drill pipe during the pressure-release phase of the drill-stem test. The three
193 fluid characteristics listed in the left column of Table 1 can infer aquifers. When a DST record contains a
194 noted qualifying description for at least 2 characteristics, an aquifer is inferred.
195 Table 1 Results from ‘blow tests’ are used to infer aquifers. The blow test analyzes the fluid exiting the drill pipe during the
196 pressure-release phase of the drill-stem test. When at least two of the three blow test characteristics (left column) include
197

PT
keywords from the qualifying description (right column), then a DST record indicates a potentially hydraulically conductive
198 aquifer. Qualifying descriptions are based on author experience.

Characteristic Qualifying Description

Strong

RI
Surface Blow
Good
Water
Recovery Salt

SC
Sulphur
Average
Good
Permeability
Excellent

U
High
199
AN
200 2.2.2 Economic Criteria
201 Economic input layer data identify potential locations for commercial power sales. These include regions
202 with upstream petroleum production, electrical infrastructure and population centers. Areas with
M

203 upstream petroleum production, and associated electrification opportunities, are identified from
204 historical natural gas rights sales. These areas include the Northern Montney and Heritage Field, Horn
205 River Pools A and D of the Muskwa Otter Park Formation, and the Cordova Embayment (Ministry of
D

206 Natural Gas Development 2016). Electrical Infrastructure data identifies locations with > 60 kV
207 transmission lines and substations (BC Hydro 2012b). A proposed transmission line between the Bennet
TE

208 Dam and Pink Mountain (ATCO Power 2015) is included as Proposed Electrical Infrastructure. Population
209 centers identify the approximate geographical centers of towns and communities that may provide
210 economic opportunity for geothermal resources.
EP

211 Figure 5 provides a map of economic input data locations. Areas with upstream electrification potential
212 are traced in green. Transmission lines are depicted in orange. The purple proposed transmission line
213 connects Pink Mountain to the Bennet Dam (not shown) in north-south direction. Existing substations
C

214 along the transmission lines are shown as yellow dots. Population centers cluster in the south of the
215 study area.
AC

9
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
C EP

216
AC

217 Figure 5 Economic input layers contain these data to represent potential for local electrification, proposed and existing electrical
218 infrastructure that permits electricity sales to the grid, and population centers that have potential heat demand. Transmission
219 infrastructure outside of the study area is excluded.

220 3.3 Criteria Scoring

221 Table 2 shows the conversion of data to criteria scores. Criteria scores can be: input value dependant
222 (e.g. temperature), distance dependant (e.g. electrical infrastructure, aquifers, population centers), or
223 location dependent (e.g. upstream petroleum electrification).

10
 ACCEPTED MANUSCRIPT

224 The maximum temperature (146 °C) is assigned a score of 1, and the minimum temperature (80°C) is
225 assigned a score of 0. Intermediate temperature scores are linearly interpolated between these two
226 points. Distance dependant scores are 1 at the location of a data point and linearly decline to 0 at or
227 beyond a maximum distance. Maximum distances are 20 km for indicated aquifers, substations, and
228 population centers. 10 km is used for transmission lines. The Upstream Petroleum Electrification Input
229 Layer is binary and has a score of 1 within electrification areas and 0 outside of electrification areas.
230 Rationale for the scoring methods are listed in the right column of Table 2.

PT
231 Table 2 Input data is converted to input layers by assigning criteria scores to each location. The center column lists the type of
232 score decay (linear or binary), the score determining parameter (temperature, distance or location) and examples that
233 determine a score of 1 and a score of 0. The right column provides the rationale for the scoring method.

RI
Input Layer Criteria Scoring Rationale
Temperature Linear decay based on temperature: Temperature map based on interpolated

SC
1 at 146 °C and spatially averaged DST and BHT
0 at 80 °C records (Figure 4).
Indicated Linear decay based on distance: Probability to recover water decreases
Aquifer 1 at indicated aquifer with distance from an indicated aquifer.

U
0 at 20 km distance from indicated aquifer
Upstream Binary decay based on location: Electrification opportunities may exist in
Petroleum 1 inside gas activity area petroleum production areas.
AN
Electrification 0 outside of gas activity area
(Proposed) Linear decay based on distance: Transmission line scores decay faster
Electrical 1 at transmission line or substation with distance due to higher connection
Infrastructure 0 at cost (BC Hydro 2012a).
M

a) 20 km distance from substation

b) 10 km distance from transmission line
(Choose greater score value where overlap occurs.)
D

Population Linear decay based on distance: Economical heat transport distance in

Centers 1 at town center district heating system (International
0 at 20 km distance from town center District Heating Association 1983; Ulloa
TE

2007; Danfoss 2014).

234

235 With the exception of Proposed Electrical Infrastructure, all criteria are equally weighted. The exception
EP

236 reflects the uncertain completion of proposed infrastructure. Input layer combination is performed by
237 calculating Equation 1 in the ArcGIS ‘raster calculator’. Favourability scores are plotted using a colour
238 scale to produce the favourability map.
C

239 3.4 Estimating Power Generation Potential

AC

240 The power generation potential of geothermal reservoirs are evaluated in regions of highest
241 favourability using the Volume Method (Williams et al. 2008). The reservoir’s gross thermal energy  is
242 derived from the reservoir’s volume, temperature and heat capacity as follows:

 = 1 − ϕ   + ϕ   ! " # − #$  (2)

243 where ϕ is the reservoir porosity,  is the reservoir rock density,  is the reservoir rock specific heat
244 capacity,  is the geothermal brine density,  is the geothermal brine specific heat capacity, ! is the
245 reservoir area, " is the reservoir thickness, # is the reservoir temperature and #$ is the rejection
246 temperature.

11
 ACCEPTED MANUSCRIPT

247 Equation 2 is modified from Williams et al. (2008) in two ways: 1) the porous fraction of the reservoir is
248 assigned a specific heat capacity of geothermal brine, and 2) the reservoir volume term % is separated
249 into the factors area ! and thickness " , so that:

% = ! " (3)

250 Converting thermal energy to exergy accounts for entropy generation, which thermodynamically
251 constrains the amount of useful work available from a temperature difference. Utilization efficiencies

PT
252 account for additional losses in a power plant. The total available electric energy is divided by project
253 lifetime to estimate the gross power capacity. Gross power capacity (&) is calculated as follows:

'(  +, #$ - − -$ 

RI
&= * +, − 0
) ℎ − ℎ$
(4)

254 where '( is the power plant utilization efficiency, ) is the power plant lifetime, +, is the recovery factor,

SC
255 #$ is the rejection temperature, - is the specific entropy of the geothermal brine at # , -$ is the specific
256 entropy of the geothermal brine at #$, ℎ is the specific enthalpy of geothermal brine at # , ℎ$ is the
257 specific enthalpy of the geothermal brine at #$ .

258

U
The economics of geothermal energy development are highly dependent on the number of production
AN
259 wells necessary to achieve the required brine flow rates (Astolfi et al. 2014). The required brine flow rate
260 123456 is the amount of brine that must be extracted per unit time to produce the predicted gross
261 power capacity & (Equation 5). Equation 5 is derived from Williams et al. (2008, eq. 2) by dividing the
262 mass of brine extracted from the well head by project lifetime. The brine flow rate per well 127 then
M

263 determines the total number of production wells 87699: required to produce gross power capacity & as
264 follows:
 +,
D

123456 =
)ℎ − ℎ$ 
(5)
TE

123456
87699: =
127
(6)

265
EP

266 3.5 Input Data to Estimating Power Generation Potential

267 Four areas are selected from the favourability map for power potential estimates. As described in
C

268 Section 4.1, these areas are Horn River, Clarke Lake, Prophet River and Jedney. The Volume Method
269 presented in Section 3.4 is applied using the data inputs described in Table 3. Values for the area,
AC

270 thickness and temperature of the reservoir are specific to each favourable area. Due to the uncertainty
271 around temperatures, areas, thicknesses, recovery factors and reservoir porosities, Monte Carlo
272 simulations with 100,000 iterations are performed for each favourable area. Stochastic input
273 parameters are chosen from triangular probability distributions, consistent with work by Walsh (2013).
274 The volumetric heat capacities  of brine and reservoir rock are obtained from density  and mass
275 based heat capacity . The  of rock is approximated from dolomite. The  of brine is approximated
276 from an aqueous sodium chloride solution. Utilization efficiency, brine entropy and enthalpy are
277 temperature dependant. For these parameters, an equation of best fit is derived from data listed in the
278 ‘source’ column in Table 3.

12
 ACCEPTED MANUSCRIPT

279 Table 3 Summary of values applied to the Volume Method. A stochastic assessment of each favourable area is performed
280 separately using 100,000 iteration Monte Carlo simulations. Parameters defined by minimum, mode and maximum value are
281 selected from triangular probability distributions. Single-column values are deterministic.*

Value
Parameter Unit Source
Min. Mode Max.
ϕ
Reservoir Slave Point formation porosities (IHS Energy
0 3 20 %
Porosity 2016; Lam & Jones 1985; Weides et al. 2013)
 Dolomite:  0.928 kJ/kgK (Krupka et al.

PT
Reservoir Heat
Capacity 1985) and density  2870 kg/m³ (Gardner et
al. 1974). Brine:  3.6 kJ/kgK (Chen 1982)
kJ/m³K
Rock 2663
Geoth. Brine 4200 and  1166 kg/m³ (Garcia 2001).

RI
A< Reservoir Area
Lateral extent of geothermal reservoirs is
Horn River 8.0 10.0 12
unknown. Natural gas pool areas (BC Oil and
Clarke Lake 101.6 127.0 152.4 km²
Gas Commission 2016) serve as proxy. Details

SC
Prophet River 36.6 45.7 54.8
in Reservoir Area Section below.
Jedney 6.8 8.5 10.2
" Reservoir
Thickness

U
Derived from stratigraphic cross sections
Horn River 108 221 225
m (Ibrahimbas & Walsh 2005). Details in
Clarke Lake 162 295 428
Reservoir Thickness Section below.
AN
Prophet River 221 241 260
Jedney 221 316 410
# Reservoir
Temperature
Derived from BHTs and DSTs (IHS Energy 2016)
M

Horn River 116.4 129.6 142.9

°C located within favourable area. Details in
Clarke Lake 91.7 111.1 130.4
Reservoir Temperature Section below.
Prophet River 107 125.6 144.2
Jedney 122.8 142.8 162.7
D

#$
Rejection Annual average air temperature at Fort Nelson
0 °C
Temperature (Government of Canada 2017)
TE

Reservoir
+, Recovery 10 17.5 25 % (Williams 2007)
Factor
)
Project
EP

30 Years (Walsh 2013; Geoscience BC 2015)

Lifetime
Equation type: linear
'(
Utilization
Values are temperature % Data: subcritical plant efficiency (Augustine et
Efficiency
dependant and calculated al. 2009)
C

for each Monte Carlo Equation type: third order polynomial

-
Geothermal
iteration using an kJ/kgK Data: H2O saturation temperature tables
Brine Entropy
AC

empirically derived (Bhattacharjee 2017)

equation that best fits data Equation type: linear
ℎ
Geothermal
from ‘Source’ column. kJ/kg Data: H2O saturation temperature tables
Brine Enthalpy
(Bhattacharjee 2017)
kg/s Consistent with previous work (Majorowicz &
127
Brine flow rate
30 and 100 per Moore 2014; Majorowicz & Grasby 2014; Lam
per well
well & Jones 1985; Geoscience BC 2015)
282 * Detailed descriptions of sources are available in (Palmer-Wilson et al. 2017).

283 Additional information on reservoir parameters is provided below.

13
 ACCEPTED MANUSCRIPT

284 Reservoir Area A< 

285 This study uses Slave Point formation natural gas pools as proxy geothermal reservoir areas. Table 4 lists
286 gas pools, their associated fields and the total area (located within favourable areas). Area values are
287 produced by geospatial analysis of gas pool contours provided by the BC Oil and Gas Commission (2016).
288 The total area values are assigned to the mode of the probability distributions, with maximum and
289 minimum values defined by a ±20 % factor to account for uncertainty. No Slave Point gas pools are

PT
290 available at Horn River. Here, a mode geothermal reservoir area of 10 km² is chosen by the authors to
291 allow comparison.
292 Table 4 The areas of gas pools of the Slave Point formation are used as proxy geothermal reservoir areas. Potential power

RI
293 generation in each favourable area is estimated using Total Gas Pool Area values. Geospatial analysis of data from the BC Oil
294 and Gas Commission (2016) provides area values of individual gas pools.

Favourable Area Slave Point Gas Pools Gas Field Total Gas Pool Area [km²]

SC
a
Horn River - - (10.0)
Clarke Lake A Clarke Lake 127.0
Prophet River A, B, C, H, I, J, L, M, N, O, P Adsett 45.7
A, B, C Bubbles

U
Jedney 8.5
B, C Bubbles North
a
295 Value chosen by authors
AN
296 Reservoir Thickness " 

297 This study uses stratigraphic cross-sections to infer reservoir thickness (Ibrahimbas & Walsh 2005).
M

298 Carbonate and sandstone formations between Beaverhill Lake Group rocks and the Precambrian surface
299 are assumed to contain a geothermal reservoir. Selected formations include the Slave Point, Sulphur
300 Point, Keg River, Wokkpash, Stone, and Upper Chinchaga. The local thickness of these formations is the
D

301 reservoir thickness value reported in Table 3. This data is available at several locations across the study
302 area, so the three locations closest to each favourable area inform the respective maximum and
TE

303 minimum value of the probability distribution for that location. The mode of the probability distribution
304 is assigned the average of the maximum and minimum thickness values.

305 Reservoir Temperature # 

EP

306 The reservoir temperature distributions are based on DST and BHT data filtered for 1) geographical
307 location within favourable areas, 2) a favourability score above 0.43, and 3) minimum depth based on
308 the top of the youngest formation considered for reservoir thickness. The mode of the temperature
C

309 probability distribution is assigned the average value of filtered data. The minimum and maximum
310 values are assigned ± 1 standard deviation from the average.
AC

311 3.6 Power Generation Sensitivity

312 A ‘Change in Output Mean Analysis’ quantifies the sensitivity of Monte Carlo simulation results to
313 stochastic input parameter changes (Palisade 2017). Monte Carlo iteration datasets contain input and
314 related output parameter values. Input parameters are temperature, recovery factor, reservoir area,
315 reservoir thickness and porosity. Output parameters selected for analysis are gross power capacity per
316 unit reservoir volume &/% and brine flow rate to produce 1 MW of power 123456 /&. Iteration datasets
317 are sorted by ascending value of an input parameter. Datasets are distributed equally across twenty

14
 ACCEPTED MANUSCRIPT

318 bins. Each bin represents five percent of the input value distribution. The mean value of the selected
319 output parameter is computed for the iteration datasets contained in each bin.

PT
RI
U SC
AN
M
D
TE
C EP
AC

15
 ACCEPTED MANUSCRIPT

320 4 Results
321 4.1 Favourability Mapping Results
322 The results of the favourability mapping procedure are shown in Figure 6. To highlight areas of highest
323 favourability, the top 10 %, 20% and 30 % are shown. The highest computed score is 0.61; hence each 10
324 % interval represents a 0.06 score step. The top 10 % to 30 % areas occur in patches that range from
325 tens to several hundred square kilometers. Larger patches cluster in the central west section and in the

PT
326 petroleum production areas in the northern section of the study area. Smaller patches occur along the
327 Fort Nelson – Rainbow Lake transmission line and the southern-most section of the study area, 20 to
328 100 km north of Fort Saint John. Only three areas feature a score in the top 10 % interval. These are

RI
329 located approximately 10 km south of the town of Trutch, in the Jedney region and adjacent to Fort
330 Nelson.

U SC
AN
M
D
TE
C EP
AC

16
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
C EP

331
AC

332 Figure 6 Geothermal Favourability Map of northeastern British Columbia. Scores range from the minimum 0 to the maximum
333 0.61. Colour gradients show 10 % score range intervals, each representing a 0.06 score step. To highlight areas of highest
334 geothermal favourability only the top three score intervals are shown. Coloured regions enclosed by red ellipses are selected for
335 estimating power generation potential.

336 The four areas highlighted in red on Figure 6 are deemed ‘Favourable Areas’. These areas include Horn
337 River, Jedney, Clarke Lake and Prophet River. Selection is somewhat qualitative and based on each
338 area’s unique characteristics. Horn River and Jedney gas fields were chosen due to remote upstream
339 natural gas electrification opportunities (Government of British Columbia 2016). The Clarke Lake area
340 features power export opportunities to Fort Nelson, and significant geological data is available from the

17
 ACCEPTED MANUSCRIPT

341 adjacent Clarke Lake and Milo gas fields. Finally, Prophet River is a First Nation, where geothermal
342 development may aid economic opportunities for the community.

343 Some areas with high favourability scores are excluded from further investigation in order to limit the
344 scope to areas where geothermal development is most likely. Several excluded patches of high
345 favourability in the northeastern study area feature economic opportunities similar to those at Horn
346 River, but lower overall favourability. Exclusions along the Fort Nelson – Rainbow Lake transmission line

PT
347 are based on temperature being generally relatively low. Areas southeast of Jedney display high
348 favourability, but indicated aquifers are scarce here. The excluded area ~10 km south of Trutch features
349 economic opportunities similar to those at Jedney.

RI
350 4.2 Estimates of Power Generation Potential
351 The Volume Method and data described in Section 3.4 – 3.6 is applied to assess the geothermal

SC
352 potential of the four favourable areas. Table 5 shows four metrics of evaluation. The gross power
353 capacity & is the potential electric power generation for 30 years at continuous production. ‘Gross’
354 disregards self-consumed electricity needed to serve a power plant’s internal load (e.g. pumps and
355 cooling fans). The impact of reservoir volume uncertainty is mitigated by calculating the capacity per

U
356 unit reservoir volume &/% . The 123456 /& metric states the required brine flow rate to produce 1 MW
357 of electrical power. The number of production wells is based on conservative (30 kg/s) and optimistic
AN
358 (100 kg/s) assumed brine flow rates per well (127 ). The feasibility of geothermal power increases with
359 higher values for & and &/% and lower values for 123456 /& and 87699: . All units of power (MW) refer
360 to electrical power.
M

361 Table 5 Four evaluation metrics characterize the potential geothermal electric power generation at the four favourable areas of
362 northeastern British Columbia. Stochastic Monte Carlo simulations of the Volume Method result in probability distributions. P90,
363 P50 and P10 values are lower than 90, 50 and 10 % of values in these distributions. The mode is the most likely value.
D

Exceedance
Probability or >2?
Evaluation Metric Horn River Clarke Lake Prophet River Jedney
TE

P90 3.6 43.9 21.6 7.7

&
Gross power
P50 5.2 71.3 31.5 11.3
capacity [MW]
Mode 4.8 63.2 28.9 10.4
EP

Gross power
&
P90 2.34 1.30 2.01 3.04
capacity per unit
%
P50 3.19 1.94 2.88 4.28
reservoir volume
[MW/km³] Mode 3.00 1.80 2.54 4.10
C

Required brine P90 31.3 45.1 32.0 22.0

123456 flow rate per P50 37.1 60.5 40.9 27.6
AC

&
gross power
capacity P10 44.5 84.5 53.7 35.2
[kg/sMW] Mode 37.0 60.4 40.8 27.5
Number of
87699:
at 30 kg/s per well 5 90 31 8
required
production wells* at 100 kg/s per well 2 27 10 3
364 *For & at P90 and 123456 /& at P50.

365 The P90 value is lower than 90 % of Monte Carlo simulation results. The P50 and P10 values are lower
366 than 50 % and 10 % of simulation results, respectively. The mode is the most likely value. The P90 values
367 of & and &/% and the P10 values of 123456 /& represent conservative estimates.

18
 ACCEPTED MANUSCRIPT

368 The power capacity is greatest at Clarke Lake and lowest at Horn River, with respective P90 estimates of
369 43.9 and 3.6 MW. The difference is largely due to the different reservoir volumes. The volume
370 normalized power infers greatest potential at Jedney and lowest at Clarke Lake, with P90 estimates of
371 3.04 and 1.30 MW/km3, respectively. Similarly, 123456 /& is lowest at Jedney and highest at Clarke Lake,
372 with respective P10 estimates of 35.2 and 84.5 kg/sMW.

373 Figure 7 shows & and 123456 /& histograms for Clarke Lake. The histograms plot the probability of
Monte Carlo simulation results to fall into one of fifty bins of equal width. Probability of & follow a

PT
374
375 gamma distribution. The 123456 /& distribution follows a skewed triangular shape, due to the required
376 flow rate’s exclusive dependence on reservoir temperature. The Clarke Lake histogram shapes are

RI
377 representative for all favourable areas.

U SC
AN
Probability

M
D
TE

378
Figure 7 Histogram of Monte Carlo simulation results for & (top) and 123456 /& (bottom) at Clarke Lake. Respective probabilities
EP

379
380 follow a gamma and skewed triangular distribution shape. Results at other favourable areas follow similar distributions.

381 4.3 Power Generation Potential Sensitivity Analysis

C

382 Figure 8 shows the results of the sensitivity analysis for Clarke Lake, which is representative of the four
383 favourable areas. The plot shows the input parameter percentiles on the horizontal axis and the mean
AC

384 output parameter values for the input percentiles on the vertical axis. For example, the ‘Reservoir
385 Temperature’ curve in the left plot shows that the lowest 5% of temperature inputs to the Clarke Lake
386 Monte Carlo simulation result in a mean & of 45.2 MW. The next 5% result in a mean & of 50.8 MW, and
387 so on up the curve.

19
 ACCEPTED MANUSCRIPT

per gross power capacity [kg/sMW]

Gross power capacity [MW]

Required brine flow rate

PT
RI
SC
388
389 Figure 8 A Change in Output Mean Analysis shows the sensitivity of & (left plot) and 123456 /& to stochastic input parameters at
390 Clarke Lake. Horizontal axes are percentiles of Monte Carlo input parameter value distributions. Vertical axes are mean output

U
391 values of iteration datasets sorted into 20 bins. The & depends on reservoir temperature, recovery factor, reservoir thickness,
392 reservoir area and porosity. The 123456 /& exclusively depends on reservoir temperature. Sensitivity at other favourable areas
AN
393 follows similar trends.

394 The sensitivity analysis shows that reservoir temperature, thickness and recovery factor significantly
395 affect &. The respective impact factors (maximum over minimum mean value) are 2.45, 2.1 and 2.0.
M

396 Reservoir area has a less pronounced effect on the output, with an impact factor of ~1.38. Reservoir
397 porosity has a negligible effect on &, with an impact factor of 1.1. The effect of reservoir porosity on & is
398 small because the total reservoir heat capacity increases only slightly when substituting a small porous
D

399 fraction of the lower heat capacity rock with the higher heat capacity of geothermal brine. The
400 123456 /& exclusively depends on reservoir temperature. At Clarke Lake the mean values resulting from
TE

401 the bottom and top five percentile inputs are 99.1 and 40.2 kg/sMW. The impact factor is 2.5.
C EP
AC

20
 ACCEPTED MANUSCRIPT

402 5 Discussion
403 This study describes a comprehensive approach for assessing geothermal resources in a sedimentary
404 basin. The method identifies areas of high geothermal potential via favourability mapping and
405 subsequently estimates power generation potential in these areas. Compared to the 12,635 MW of
406 global installed geothermal capacity (Bertani 2015) the total estimated mode value of 107.3 MW in the
407 WCSB case study is small. Globally, however, many sedimentary basins are deep enough to host

PT
408 geothermal resources (Laske & Masters 1997). The assessment method can be applied in any
409 sedimentary basin where similar data is available. This study is therefore a significant step in reducing
410 the time and cost involved with geothermal energy prospecting.

RI
411 The case study of the British Columbian portion of the WCSB improves upon previous geothermal
412 favourability maps and power generation estimates. Previous geospatial studies did not estimate power
413 generation potential (Fairbank et al. 1992; Kimball 2010), or were not selected using a comprehensive

SC
414 favourability mapping approach (Walsh 2013; Geoscience BC 2015). The higher resolution analysis of
415 geological and economic criteria identifies Clarke Lake, Jedney, Horn River and Prophet River as
416 favourable areas for geothermal development, of which the latter two are new discoveries. Fairbank et

U
417 al. (1992) identified Clarke Lake by deducing high geothermal potential from thermal gradients greater
418 45 °C/km, but did not consider economic criteria. This study shows that geothermal gradient alone is
AN
419 neither an indicator of high favourability nor increased power production potential. As shown, areas
420 with high gradients (e.g. Clarke Lake) had both lower favourability and lower power potential per unit
421 reservoir volume than areas with lower gradients (e.g. Jedney).
M

422 Two favourable areas identified in this study have been the subject of previous power generation
423 estimates: Clarke Lake and Jedney (Walsh 2013; Geoscience BC 2015). At Clark Lake, this study
424 estimated P90 and P50 gross power capacities of 43.9 and 71.3 MW respectively; larger than respective
D

425 P90 and P50 values of 18.4 and 37.4 MW (Geoscience BC 2015), and the P50 value of 34 MW (Walsh
426 2013). Both previous works assumed smaller reservoir volumes. The Geoscience BC study uses
TE

427 proprietary information to justify reservoir volume, while Walsh (2013) uses the dolomitized section of
428 the Slave Point formation. This study assumes a larger reservoir present in several carbonate and
429 sandstone formations. At Jedney, this study’s respective P90 and P50 gross power capacities are 7.8 and
EP

430 11.3 MW; less than respective values of 12.2 and 24.7 MW (Geoscience BC 2015), based on that study’s
431 smaller reservoir volume.

432 The mode of the gross power capacity per unit reservoir volume &/% ranges from ~1.8 to ~4.10
C

433 MW/km3 at Clarke Lake and Jedney, respectively. These results are significantly higher than ~0.23 to 1.6
434 MW/km3 found in western Alberta (Banks & Harris, unpublished results). Several factors contribute to
AC

435 these differences. While this study focuses on the highest temperature reservoirs in the region, Banks &
436 Harris (unpublished results) evaluated reservoirs with temperatures as low as 80 ˚C. Additionally, they
437 applied a smaller recovery factor of 10%, while this study uses a range from 10 – 25 %. As shown in
438 Section 4.3, the selected temperature and recovery factor ranges can each affect gross power capacity
439 by a factor of ~2.

440 This study estimates the brine flow rate required to produce 1 MW of electric power 123456 /& and
441 deduces the number of production wells required to support the estimated gross power capacities.
442 These are proxies for the technical and economic viability of a geothermal energy project (Majorowicz &
443 Moore 2014; Ferguson & Ufondu 2017). The conservative P10 123456 /& estimates in this study range

21
 ACCEPTED MANUSCRIPT

444 from 35.2 to 84.5 kg/sMW at Clarke Lake and Jedney, respectively. A single production well might
445 sustain between 0.41 to 1.0 MW under P10 conditions. Optimistic P90 123456 /& values at Jedney are as
446 low as 22 kg/sMW. Here, a single production well might sustain 1.6 MW. Brine flow rates as high as 35
447 kg/s are seen in co-produced fluid situations in narrow diameter gas well bores throughout the WCSB
448 (Lam & Jones 1985; Ferguson & Ufondu 2017). If full-size geothermal wells were producing from the
449 same conditions, flow rates as high as 100 kg/s may be achievable (Walsh 2013). Thus, hydrogeologic
450 conditions in potential reservoirs throughout the WCSB appear favourable to geothermal energy

PT
451 production.

452 When estimating the power generation potential, assessing the reservoir size is a significant challenge.

RI
453 This study estimates reservoir thicknesses by spatially extrapolating stratigraphic cross-sections and
454 assuming several carbonate and sandstone formations to be brine-saturated and hydraulically
455 conductive. The lateral extent of geothermal reservoirs is derived from natural gas pools in the Slave
Point formation, and by varying that extent by ±20% to account for uncertainty. Estimates of & and

SC
456
457 &/% assume brine extraction from all formations, in spite of potential hydraulic separation and brine
458 temperature differences up to 20 °C due to varying formation depths. These assumptions introduce
459 uncertainty to estimated power generation potential and may over-estimate the gross power potential

U
460 of the area. In contrast, not all areas with high favourability scores are included in power potential
AN
461 estimates. These exclusions may lead to underestimating the total power potential in northeastern
462 British Columbia.

463 In the Volume Method, a geothermal reservoir comprises a finite heat resource without thermal
M

464 recharge. Some work suggests that heat flow in the WCSB is controlled primarily by conduction of heat
465 from the Precambrian basement, rather than by convection via groundwater flow (Bachu & Burwash
466 1994). In this case, thermal recharge can be neglected as conductive recharge is negligible on a human
D

467 lifespan timescale (Barbier 2002). Other studies suggest that groundwater flow influences the
468 geothermal regime (Majorowicz et al. 1999). If thermal recharge occurs the Volume Method produces a
TE

469 conservative power generation estimate.

C EP
AC

22
 ACCEPTED MANUSCRIPT

470 6 Conclusion
471 This study presents and demonstrates a comprehensive method to estimate geothermal power
472 generation potential of a sedimentary basin. The two-step process of favourability mapping and
473 volumetric power estimates significantly reduces the uncertainty surrounding the available geothermal
474 resource on a regional basis.

475 As a case study, the method is applied to the British Columbian portion of the WCSB and identified

PT
476 previously undiscovered areas where geothermal power development is favourable (Horn River,
477 Prophet River) and confirmed known locations (Clarke Lake, Jedney). Power generation potential
478 estimates apply the Volume Method (Williams et al. 2008) and utilize a Monte Carlo simulation to

RI
479 identify the possible ranges in predictions. The mode of total power generation potential is 107.3 MW.
480 The volume normalized mode of expected power potential for the four investigated favourable areas
481 range from 1.8 – 4.1 MW/km3. Mode brine flow rates of 27.5 – 60.4 kg/s are required to produce 1 MW

SC
482 of power, based on the temperatures present in these reservoirs.

483 The sensitivity study highlights the strong dependence of results on reservoir temperature, thickness,
484 and recovery factor, each of which can affect the gross power capacity by a factor ~2 between the

U
485 lowest 5% and highest 5% values selected from input distributions.
AN
486 This two-step method can be applied to sedimentary basins with similar characteristics and data
487 availability globally; thereby providing improved estimates of available low-carbon, dependable
488 geothermal energy.
M

489 Acknowledgement
490 Funding for this research was provided by Geoscience BC and the Pacific Institute for Climate Solutions.
D

491 In-kind contributions from the British Columbia Ministry of Energy, Mines & Petroleum Resources made
492 this work possible.
TE
C EP
AC

23
 ACCEPTED MANUSCRIPT

493 References
494 Arianpoo, N., 2009. The Geothermal Potential of Clarke Lake and Milo Gas Fields in northeast British
495 Columbia. Master Thesis. Department of Mining Engineering. University of British Columbia
496 (Vancouver).
497 Astolfi, M. et al., 2014. Binary ORC (Organic Rankine Cycles) power plants for the exploitation of
498 medium-low temperature geothermal sources - Part B: Techno-economic optimization. Energy, 66,

PT
499 pp.435–446.
500 ATCO Power, 2015. ATCO Ministerial Exemption Request, Available at:
501 http://prrd.bc.ca/board/agendas/2015/2015-32-478298400/pages/documents/14-b-CA-

RI
502 9ATCONMPSExemption.pdf.
503 Augustine, C. et al., 2009. Modeling and analysis of sub- and supercritical binary rankine cycles for low-

SC
504 to mid-temperature geothermal resources. Transactions - Geothermal Resources Council, 33,
505 pp.604–608. Available at: http://www.scopus.com/inward/record.url?eid=2-s2.0-
506 77952627774&partnerID=tZOtx3y1.

U
507 Bachu, S. & Burwash, R.A., 1994. Geothermal Regime in the Western Canada Sedimentary Basin,
508 Available at: http://ags.aer.ca/publications/chapter-30-geothermal-regime.
AN
509 Banks, J. & Harris, N., unpublished results. Geothermal Potential of Foreland Basins: A case study from
510 the Western Canadian Sedimentary Basin. Geothermics.
511 Barbier, E., 2002. Geothermal energy technology and current status: An overview. Renewable and
M

512 Sustainable Energy Reviews, 6(1–2), pp.3–65.

513 BC Hydro, 2012a. 2013 Resource Options Report Update - Appendix 5 B - Roads & Power Lines Cost
514 Estimation using GIS, Available at:
D

515 https://www.bchydro.com/content/dam/hydro/medialib/internet/documents/planning_regulator
516 y/iep_ltap/2012q2/draft_2012_irp_appx_3A_24.pdf.
TE

517 BC Hydro, 2012b. Transmission System Map, Available at:

518 https://www.bchydro.com/content/dam/BCHydro/customer-portal/documents/accounts-
519 billing/electrical-connections/bchydro-transmission-systems-map.pdf.
EP

520 BC Oil and Gas Commission, 2016. Pool Contours. BCOGC Open Data Portal. Available at: http://data-
521 bcogc.opendata.arcgis.com/datasets/c9c7ebb8c12041ddab4c93685269ddec_0?geometry=-
522 123.09%2C57.286%2C-120.584%2C57.508 [Accessed April 7, 2017].
C

523 Bertani, R., 2015. Geothermal Power Generation in the World 2010-2014 Update Report. In World
AC

524 Geothermal Congress 2015. Melbourne, Australia.

525 Bhattacharjee, S., 2017. Saturation Temperature table (PC Model) of H2O. Available at:
526 http://www.thermofluids.net/ [Accessed April 5, 2017].
527 Chen, C.A., 1982. Specific Heat Capacities of Aqueous Sodium Chloride Solutions at High Pressures.
528 Journal of Chemical & Engineering Data, 27(2), pp.356–358.
529 Danfoss, 2014. The heating book: Instructions for designing district heating system, Available at:
530 http://heating.danfoss.com/Content/61078BB5-1E17-4CA2-AC49-
531 8A7CAC2BA687_MNU17424997_SIT54.html.

24
 ACCEPTED MANUSCRIPT

532 Fairbank, B.P. et al., 1992. Geothermal resources of British Columbia - Open File 2526, Available at:
533 http://geoscan.nrcan.gc.ca/starweb/geoscan/servlet.starweb?path=geoscan/fulle.web&search1=R
534 =133397.
535 Ferguson, G. & Ufondu, L., 2017. Geothermal energy potential of the Western Canada Sedimentary
536 Basin: Clues from coproduced and injected water. Environmental Geosciences, 24(3), pp.113–121.
537 Garcia, J.E., 2001. Density of Aqueous Solutions of CO 2. Lawrence Berkeley National Laboratory.

PT
538 Available at: http://escholarship.org/uc/item/6dn022hb.
539 Gardner, G.H.F., Gardner, L.W. & Gregory, A.R., 1974. FORMATION VELOCITY AND DENSITY—THE
540 DIAGNOSTIC BASICS FOR STRATIGRAPHIC TRAPS. GEOPHYSICS, 39(6), pp.770–780. Available at:

RI
541 http://library.seg.org/doi/10.1190/1.1440465.
542 Garland, G.D. & Lennox, D.H., 1962. Heat Flow in Western Canada. Geophysical Journal International,

SC
543 6(2), pp.245–262.
544 Geoscience BC, 2015. An Assessment of the Economic Viability of Selected Geothermal Resources in
545 British Columbia. Geoscience BC Report 2015-11, Available at:

U
546 http://www.geosciencebc.com/i/project_data/GBCReport2015-11/GBC2015-11_KWL_Geothermal
547 Economics_Project_Report_27Sep16.pdf.
AN
548 Geothermal Resources Act, 1996. (RSBC 1996) Chapter 171, Part 1 - Interpretation.
549 Government of British Columbia, 2016. Climate Leadership Plan - British Columbia, Available at:
550 https://climate.gov.bc.ca/wp-content/uploads/sites/13/2016/06/4030_CLP_Booklet_web.pdf.
M

551 Government of Canada, 2017. Canadian Climate Normals 1981-2010 Station Data - Fort Nelson A, British
552 Columbia. Available at:
553 http://climate.weather.gc.ca/climate_normals/results_1981_2010_e.html?searchType=stnProv&ls
D

554 tProvince=BC&txtCentralLatMin=0&txtCentralLatSec=0&txtCentralLongMin=0&txtCentralLongSec=
555 0&stnID=1455&dispBack=0 [Accessed April 5, 2017].
TE

556 Grasby, S.E. et al., 2012. Geothermal Energy Resource Potential of Canada, Geological Survey of Canada -
557 Open File 6914.
EP

558 Greene, R. et al., 2011. GIS-Based Multiple-Criteria Decision Analysis. Geography Compass, 5(6), pp.412–
559 432.
560 Harrison, W.E. et al., 1983. Geothermal resource assessment in Oklahoma. Oklahoma Geological Survey
C

561 Special Publication, (83–1), p.42. Available at: papers://a25d78e7-d6e0-4b2b-9b83-

562 8d9ae6ec0592/Paper/p103.
AC

563 Ibrahimbas, A. & Walsh, W., 2005. Stratigraphy and Reservoir Assessment of Pre-Givetian Strata in
564 Northeastern British Columbia. British Columbia Ministry of Energy and Mines, pp.1–14.
565 IHS Energy, 2016. AccuMap, V.23.05. Englewood, Colorado: IHS Energy.
566 International District Heating Association, 1983. District heating Handbook,
567 Jones, F.W., Lam, H.L. & Majorowicz, J.A., 1985. Temperature distributions at the Paleozoic and
568 Precambrian surfaces and their implications for geothermal energy recovery in Alberta. Can. J.
569 Earth Sci, 22, pp.1774–1780.

25
 ACCEPTED MANUSCRIPT

570 Kimball, S., 2010. Favourability Map of British Columbia Geothermal Resources. Master Thesis.
571 Department of Mining Engineering. University of British Columbia (Vancouver). Available at:
572 http://pics.uvic.ca/sites/default/files/uploads/publications/kimball_thesis_2010.pdf.
573 Krupka, K.M. et al., 1985. High-temperature heat capacities and derived thermodynamic properties of
574 anthophyllite, diopside, dolomite, enstatite, bronzite, talc, tremolite and wollastonite. American
575 Mineralogist, 70(3–4), pp.261–271.

PT
576 Lam, H.L. & Jones, F.W., 1985. Geothermal energy potential in the Hinton–Edson area of west-central
577 Alberta. Canadian Journal of Earth Sciences, 22(3), pp.369–383.
578 Lam, H.L., Jones, F.W. & Majorowicz, J.A., 1985. A statistical analysis of bottom-hole temperature data in

RI
579 southern Alberta. GEOPHYSICS, 50(4), pp.677–684.
580 Laske, G. & Masters, G., 1997. A Global Digital map of Sediment Thickness. EOS Trans. AGU, 78, p.F483.

SC
581 Lund, J.W. & Boyd, T.L., 2015. Direct Utilization of Geothermal Energy 2015 Worldwide Review. In
582 Proceedings World Geothermal Congress. pp. 19–25.
583 Majorowicz, J. et al., 2012. Implications of post-glacial warming for northern Alberta heat flow-

U
584 correcting for the underestimate of the geothermal potential. Geothermal Resources Council
585 Transactions, 36, pp.693–698. Available at: http://pubs.geothermal-
AN
586 library.org/lib/grc/1030303.pdf.
587 Majorowicz, J.A. et al., 1999. Present Heat Flow Along a Profile Across the Western Canada Sedimentary
588 Basin: The Extent of Hydrodynamic Influence. In Geothermics in Basin Analysis. Heidelberg,
M

589 Germany: Springer, pp. 61–79.

590 Majorowicz, J.A. & Jessop, A.M., 1981. Regional heat flow patterns in the Western Canadian
591 Sedimentary Basin. Tectonophysics, 74, pp.209–238.
D

592 Majorowicz, J. & Grasby, S.E., 2014. Geothermal Energy for Northern Canada: Is it Economical? Natural
TE

593 Resources Research, 23(1), pp.159–173.

594 Majorowicz, J. & Moore, M., 2014. The feasibility and potential of geothermal heat in the deep Alberta
595 foreland basin-Canada for CO2 savings. Renewable Energy, 66, pp.541–549. Available at:
EP

596 http://dx.doi.org/10.1016/j.renene.2013.12.044.
597 Malczewski, J., 2000. On the Use of Weighted Linear Combination Method in GIS: Common and Best
598 Practice Approaches. Transactions in GIS, 4(1), pp.5–22. Available at: http://www.blackwell-
C

599 synergy.com/links/doi/10.1111/1467-9671.00035.
600 Malczewski, J. & Rinner, C., 2015. Multicriteria Decision Analysis in Geographic Information Science,
AC

601 Available at: http://www.amazon.com/Multicriteria-Decision-Analysis-Geographic-

602 Information/dp/3540747567/ref=sr_1_1?ie=UTF8&qid=1430864854&sr=8-
603 1&keywords=Multicriteria+decision+analysis+in+geographic+information+science.
604 Massam, H., 1988. Multi-criteria (MCDM) Decision Making Techniques in Planning. Progress in Planning,
605 (30), pp.1–84. Available at: http://www.sciencedirect.com/science/article/pii/0305900688900128.
606 Ministry of Natural Gas Development, 2016. Summary of Shale Gas Activity in Northeast British
607 Columbia - Oil and Gas Report 2016-1, Victoria.
608 Moeck, I.S., 2014. Catalog of geothermal play types based on geologic controls. Renewable and

26
 ACCEPTED MANUSCRIPT

609 Sustainable Energy Reviews, 37, pp.867–882. Available at:

610 http://dx.doi.org/10.1016/j.rser.2014.05.032.
611 Mossop, G.D. & Shetsen, I., 1994. Geological atlas of the Western Canada Sedimentary Basin, Available
612 at: http://ags.aer.ca/publications/atlas-of-the-western-canada-sedimentary-basin.htm.
613 Nyerges, T.L. & Jankowski, P., 2010. Regional and Urban GIS: A Decision Support Approach, New York:
614 The Guilford Press.

PT
615 Palisade, 2017. 7.8. Interpreting Change in Output Statistic in Tornado Graph. Available at:
616 http://kb.palisade.com/index.php?pg=kb.page&id=248 [Accessed August 28, 2017].

RI
617 Palmer-Wilson, K., Walsh, W. & Banks, J., 2017. Techno-Economic assessment of Geothermal Energy
618 Resources in the Sedimentary Basin in Northeastern British Columbia, Canada, Geoscience BC.
619 Vancouver.

SC
620 Salmon, J.P. et al., 2011. Guidebook to Geothermal Power Finance, NREL/SR-6A20-49391. National
621 Renewable Energy Laboratory. Boulder, CO.
622 Trumpy, E. et al., 2016. Geothermal potential assessment for a low carbon strategy: A new systematic

U
623 approach applied in southern Italy. Energy, 103, pp.167–181. Available at:
624 http://dx.doi.org/10.1016/j.energy.2016.02.144.
AN
625 Trumpy, E. et al., 2015. Geothermics Data integration and favourability maps for exploring geothermal
626 systems in Sicily , southern Italy. Geothermics, 56, pp.1–16. Available at:
627 http://dx.doi.org/10.1016/j.geothermics.2015.03.004.
M

628 Ulloa, P., 2007. Potential for Combined Heat and Power and District Heating and Cooling from Wasteto-
629 Energy Facilities in the U.S. – Learning from the Danish Experience. Master Thesis. Department of
630 Earth and Environmental Engineering. Columbia University.
D

631 Voogd, H., 1983. Multicriteria evaluation for urban and regional planning. Available at:
TE

632 https://pure.tue.nl/ws/files/3744610/102252.pdf.
633 Walsh, W., 2013. Geothermal resource assessment of the Clarke Lake Gas Field, Fort Nelson, British
634 Columbia. Bulletin of Canadian Petroleum Geology, 61(3), pp.241–251. Available at:
EP

635 http://bcpg.geoscienceworld.org/cgi/doi/10.2113/gscpgbull.61.3.241.
636 Weides, S.N., Moeck, I.S., Schmitt, D.R., et al., 2014. An integrative geothermal resource assessment
637 study for the siliciclastic Granite Wash Unit, northwestern Alberta (Canada). Environmental Earth
C

638 Sciences, 72, pp.4141–4154.

639 Weides, S.N., Moeck, I.S., Schmitt, D.R., et al., 2013. Characterization of Geothermal Reservoir Units in
AC

640 Northwestern Alberta By 3D Structural Geological Modelling and Rock Property Mapping Based on
641 2D Seismic and Well Data. Proceedings, 38 Workshop on Geothermal Reservoir Eng, (198), pp.1–8.
642 Weides, S.N., Moeck, I.S., Majorowicz, J.A., et al., 2013. Geothermal exploration of Paleozoic formations
643 in Central Alberta. Canadian Journal of Earth Sciences, 50, pp.519–534. Available at:
644 http://www.nrcresearchpress.com/doi/full/10.1139/cjes-2012-0137.
645 Weides, S.N., Moeck, I.S., Majorowicz, J.A., et al., 2014. The Cambrian Basal Sandstone Unit in central
646 Alberta – an investigation of temperature distribution, petrography and hydraulic and
647 geomechanical properties of a deep saline aquifer. Canadian Journal of Earth Sciences, 51, pp.783–

27
 ACCEPTED MANUSCRIPT

648 796. Available at: http://www.nrcresearchpress.com/doi/abs/10.1139/cjes-2014-0011.

649 Weides, S.N. & Majorowicz, J., 2014. Implications of spatial variability in heat flow for geothermal
650 resource evaluation in large foreland basins: The case of the Western Canada sedimentary basin.
651 Energies, 7(4), pp.2573–2594.
652 Williams, C.F., 2007. Updated methods for estimating Recovery Factors for geothermal resources. Thirty-
653 Second Workshop on Geothermal Reservoir Engineering, p.7.

PT
654 Williams, C.F., Reed, M. & Mariner, R., 2008. A Review of Methods Applied by the U.S. Geological Survey
655 in the Assessment of Identified Geothermal Resources, U.S. Geological Survey. Open-File Report
656 2008–1296.

RI
657

U SC
AN
M
D
TE
C EP
AC

28
 ACCEPTED MANUSCRIPT

• We comprehensively map geothermal favorability & estimate electric power potential.

• Geological & economic criteria identify favorable areas for geothermal development.
• The volume method produces power generation & required brine flow rate estimates.
• We find a total power potential of 107 MW in northeast British Columbia, Canada.
• Our method can be applied to global sedimentary basins with similar available data.

PT
RI
U SC
AN
M
D
TE
C EP
AC