Time Value of Greenhouse Gas Emissions in Life Cycle Assessment

Apr 23, 2019 - Life cycle assessment is a fundamental tool used to evaluate the environmental impact of products. Standard life ... View: PDF | PDF w/...
0 downloads 0 Views 290KB Size
Subscriber access provided by UNIV OF LOUISIANA

Energy and the Environment

Time Value of Greenhouse Gas Emissions in Life Cycle Assessment and Techno-Economic Analysis Evan Sproul, Jay Barlow, and Jason Quinn Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.9b00514 • Publication Date (Web): 23 Apr 2019 Downloaded from http://pubs.acs.org on April 25, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27

Environmental Science & Technology

1

Time Value of Greenhouse Gas Emissions in Life Cycle Assessment and Techno-

2

Economic Analysis

3

Evan Sproul†, Jay Barlow†, Jason C. Quinn†*

4

†Mechanical

Engineering, 1374 campus delivery, Colorado State University, Fort Collins, CO 80524-1374

5 6

*Corresponding

author: 1374 Campus Delivery

7

Fort Collins, CO 80524

8

Email: [email protected]

9

Ph: 970-581-7992

ACS Paragon Plus Environment

Environmental Science & Technology

11

Abstract

12

Life cycle assessment is a fundamental tool used to evaluate the environmental impact of products.

13

Standard life cycle assessment methodology ignores the impact of greenhouse gases relative to when

14

they are emitted. In this paper we present a method for leveraging the social cost of greenhouse gases

15

to account for the temporal impacts of emissions in life cycle assessment and techno-economics. To

16

demonstrate, we use this method to analyze the present value of the monetized impacts of emissions

17

across multiple electricity generation technologies. Results show that accounting for time increases the

18

present value across all but one of the technologies considered. Carbon intensive technologies show the

19

highest increase, with coal rising between 26% and 62% depending on social cost scenario. Additionally,

20

we demonstrate a second method that combines temporally resolved greenhouse gas emissions with

21

techno-economic analysis. Considering temporal impacts of emissions within techno-economic analysis

22

increases the levelized cost of electricity (LCOE) across all technologies considered. Carbon intensive

23

technologies increase significantly, with the LCOE from coal rising between 37% and 263% depending on

24

social cost scenario. The proposed methods show that temporal resolution in life cycle assessment is

25

critical for comparing the monetized impacts of greenhouse gas emissions across technologies.

ACS Paragon Plus Environment

Page 2 of 27

Page 3 of 27

27 28

Environmental Science & Technology

1. Introduction Life cycle assessment (LCA) is a tool used to investigate and compare the potential

29

environmental impacts of products. In LCA, mass and energy flows are tracked throughout production,

30

use, and disposal of a product. These flows are then translated into environmental indicators such as

31

greenhouse gas emissions. Techno-economic analysis (TEA) uses the same mass and energy flows to

32

estimate operational and capital costs. These costs and other economic inputs, such as an internal rate

33

of return, can be used to estimate economic performance and determine areas where further research

34

may be able to increase economic viability. LCA and TEA are often used together to inform research and

35

development of new products and engineering processes1–4.

36

In recent years, the expansion of LCA to integrate with decision making and policy has caused

37

certain limitations to become apparent5–8. One such limitation is the lack of temporal resolution

38

considered in LCA greenhouse gas emissions. A key reason for this limitation is the use of global

39

warming potential (GWP) as the foundational metric for greenhouse gas assessment. GWP is the ratio

40

of cumulative radiative forcing (CRF) of a greenhouse gas emission to the CRF of a CO2 emission. GWP

41

allows us to compare the CRF of different greenhouse gases and relate emissions to an equivalent

42

amount of CO2 (CO2-eq). Despite being the current standard for comparing greenhouse gases, GWP is

43

acknowledged by the International Panel on Climate Change (IPCC) and others to be a simplified metric

44

that does not account for a range of dynamic factors9–12.

45

One specific point of debate has been the use of a constant GWP analytical time frame (often

46

100 years) within LCA. A number of researchers have raised issues with this practice, noting that it

47

ignores the actual timing of emissions13–15. They highlight that an emission released early in the LCA time

48

frame will be in the atmosphere for a longer period of time than an emission released in a later year. As

49

a result, the earlier emission will generate more CRF than the later emission. The proposed solution for

50

this discrepancy is the implementation of a LCA time horizon. The time horizon is a cutoff year after

ACS Paragon Plus Environment

Environmental Science & Technology

51

which the CRF of an emission is no longer considered. To demonstrate the concept, we can consider a

52

hypothetical time horizon of the year 2100. If we consider an emission released in the year 2020, the

53

CRF of this emission will be analyzed over 80 years before reaching the time horizon. If we consider an

54

equivalent emission released in the year 2050, this emission will be analyzed for only 50 years and result

55

in a lower value of CRF. The impact of implementing a time horizon has been explored by a variety of

56

researchers.

57

O’Hare et al. 13 were some of the first to implement the time horizon approach by developing

58

the Fuel Warming Potential (FWP). The FWP compares the CRF between biofuel emissions and

59

petroleum fuel emissions up to a time horizon. In addition, O’Hare et al.13 also extended modeling to

60

translate CRF into economic damages that could be discounted to a present value. This work showed

61

that biofuels have an increased impact when accounting for time, largely due to the heavy burden of

62

land use changes occurring during early stages of biofuel production. Kendall et al. 14 also developed a

63

Time Correction Factor (TCF) for adjusting amortized emissions to account for actual emissions timing.

64

Similar to the FWP, the TCF utilized CRF up to a time horizon as the driving metric for impact. In this

65

case, Kendall et al.14 showed that correcting amortized emissions to consider time increased the impacts

66

of corn-ethanol.

67

Building upon these efforts, Levasseur et al.15 developed a more generalized form of dynamic

68

LCA. Like previous efforts, this dynamic LCA utilized CRF to gauge impact up to a time horizon. Unlike

69

previous efforts, this method was not limited to the specific cases of fuels or amortized emissions. Since

70

this introduction of dynamic LCA, an iterative development of methods has continued16–23. Throughout

71

this development, the underlying methods have remained much the same with efforts largely focused

72

on adapting and applying a time horizon to specific scenarios where temporal impacts can play a

73

significant role. This includes assessments of advanced vehicles, buildings, biogenic carbon and

74

temporary storage, gasification of crop residues, cellulosic biofuels, and photovoltaics.

ACS Paragon Plus Environment

Page 4 of 27

Page 5 of 27

75

Environmental Science & Technology

The methods presented across these analyses show specific implications of accounting for

76

emissions timing within LCA. However, these methods are often limited to only correcting the analytical

77

time frame and do not account for other dynamic factors. One such factor is the ongoing increase in

78

atmospheric greenhouse gas concentrations. This increase will slow down the rate at which future

79

emissions are removed from the atmosphere and decrease the radiative efficiency of those emissions24.

80

While there is potential for these two factors to counteract one another25, current LCA practice

81

completely ignores these effects. Recognizing the exclusion of dynamic climate factors, Farquharson et

82

al.26 used the Model for the Assessment of Greenhouse-gas Induced Climate Change (MAGICC) to

83

compare life cycle greenhouse gas emissions of natural gas and coal power generation. In this dynamic

84

analysis, the results of MAGICC are compared across several climate metrics including GWP, Global

85

Temperature Change, and CRF. Results highlight that each metric can be useful for decision making with

86

a given set of societal values or targets.

87

The work by Farquharson et al.26 addresses a broad range of dynamic climate variables through

88

the use of MAGICC. However, downstream of these climate impacts lie other dynamic factors embedded

89

in socio-economic systems such as economic productivity and future technology deployment. These

90

variables will impact how climate change manifests into economic damage. Many integrated assessment

91

models (IAMs) used to model climate change and related economic impacts account for these sort of

92

dynamic variables. Often, the models utilize a non-linear trend in which economic damage increases

93

exponentially with the rise of global temperature27. This increase means that radiative forcing in future

94

years may have a significantly higher monetized impact than radiative forcing in the present. Although

95

this topic is under debate28, using metrics such as CRF, GWP, or temperature change totally excludes

96

these considerations.

97 98

While many LCA researchers recognize the potential for non-linear socio-economic impacts, there have been few attempts to account for them in life cycle assessment. A rare example of such an

ACS Paragon Plus Environment

Environmental Science & Technology

99

attempt is Delucchi’s29,30 development of the Lifecycle Emissions Model (LEM). Within the LEM,

100

emissions are modeled with increasing background atmospheric greenhouse gas concentrations. Then,

101

using a climate sensitivity factor and non-linear damage function, an emission’s radiative forcing is

102

translated into temperature change and economic damage. By comparing the present value of damage

103

from a greenhouse gas emission to an equivalent CO2 emission, the model can yield Carbon Equivalency

104

Factors (CEFs). This comparison represents a significant expansion compared to previous methods. Yet,

105

the CEFs are strictly a comparison of economic damage from a greenhouse gas emission and CO2

106

emission that are released in the same year. The CEFs do not compare the damage of greenhouse gas

107

emissions and CO2 emissions that are released in different years. Therefore, the CEFs do not provide a

108

comparison of changing impacts across different years.

109

The work in this paper seeks to address the shortcomings of current dynamic LCA by developing

110

new methods to compare the monetized impact of greenhouse gas emissions in the future with

111

monetized impacts of emissions in the present. This work requires temporally resolved LCA coupled

112

with the social cost of greenhouse gases. The social costs of greenhouse gases used in our methods are

113

derived from multiple IAMs used by the Interagency Working Group on Social Cost of Greenhouse

114

Gases27,31,32. These costs include temporally resolved damages for CO2, CH4, and N2O. This damage is

115

discounted back to a present value using an economic discount rate. Four economic discount rate

116

scenarios are considered to define a range of results, including a baseline scenario. Using the annual

117

social cost of CO2 (SC-CO2), social cost of CH4 (SC-CH4), and social cost of N2O (SC-N2O) values, we derive

118

a dynamic global warming impact (DGWI) which allows us to compare the monetized impacts of

119

greenhouse gas emissions relative to today’s environmental and economic conditions. Analogous to the

120

present value of money commonly discussed in economics, this approach weights the value of emissions

121

based on their monetized impact at a given time. In addition to this new LCA method, we also explore a

122

second method that is intended for applications where LCA results are incorporated into TEA. This

ACS Paragon Plus Environment

Page 6 of 27

Page 7 of 27

Environmental Science & Technology

123

method uses the social costs of greenhouse gases combined with temporally resolved LCA to generate

124

time resolved economic damages from emissions. Treating these damages as costs in a TEA cash flow,

125

we can then gauge their impact on the levelized cost of production.

126

The novelty of this work has two major components. First, our methods account for changing

127

atmospheric carbon concentrations and non-linear monetized impacts to society. Second, they include

128

the dynamic monetized impacts of greenhouse gas emissions occurring in different years. To

129

demonstrate, we apply these methods to electricity generation technologies and review the effect on

130

LCA and TEA results. We compare our results with previous LCA methods and discuss how our findings

131

contradict current dynamic LCA practice. Lastly, we discuss the limitations and ideal applications of

132

these new methods within the broader context of LCA.

133 134

2. Methods

135

2.1

Social Costs of Greenhouse Gases

136

The methods in this paper seek to address the shortcomings of current dynamic LCA by

137

following Delucchi’s29,30 general approach, but altering the methods to compare the monetized impact

138

of greenhouse gas emissions in the future with monetized impacts of CO2 emissions in the present. The

139

monetized impacts considered in this paper are the social costs of greenhouse gases developed by the

140

Interagency Working Group on Social Cost of Greenhouse Gases27,31,32 . These costs are derived using

141

IAMs that model climate and economic systems on a global scale. These systems include consideration

142

for the carbon cycle, climate sensitivity, future technology deployment, economic productivity, and a

143

variety of other variables. To develop the social cost of greenhouse gases, the IAMs are run under two

144

different scenarios. The first scenario tracks the global gross domestic product (GDP) for a specific future

145

global emissions pathway up to the year 2300. The second scenario runs the same pathway, but also

146

includes one extra pulse of greenhouse gas emitted in a specific year. The difference in global GDP

ACS Paragon Plus Environment

Environmental Science & Technology

Page 8 of 27

147

between these two scenarios represents the monetized social damage due to a marginal greenhouse

148

gas emission.

149

This procedure is repeated across three separate IAMs (PAGE, DICE, and FUND) for a range of

150

five potential socio-economic emissions scenarios. The values of each model and scenario are averaged

151

to arrive at a single damage value. This damage is then discounted back to a present value using an

152

economic discount rate. The use of a specific discount rate has been a subject of much debate. As a

153

result, current estimates for the social cost of greenhouse gases include results from a range of low

154

(2.5%), middle (3%), and high (5%) economic discount rates. Additionally, a fourth cost has been

155

developed to represent higher than expected damages. In this scenario, low probability high impact

156

damages outside the 95th percentile are discounted back at a rate of 3%, generating the low probability

157

high impact (3%-95th) social cost. Table 1 displays the social costs of greenhouses gases for the four

158

discounting scenarios. The table shows values on a five-year incremental basis for 2020-2050. A full list

159

of yearly values is presented in Table S3 of the supporting information.

160 161

Table 1. Social Costs of Greenhouse Gases: Social cost of one metric ton of greenhouse gas (2020 US

162

dollars) based on 2.5%, 3%, 5%, and 3%-95th percentile IAM discount rates27,31,32. Year of Emission

Social Cost of CO2

Social Cost of CH4

Social Cost of N2O

5%

3%

2.5%

3%-95th

5%

3%

2.5%

3%-95th

5%

3%

2.5%

3%-95th

2020

23

62

85

181

1,018

1,762

2,206

2025

26

68

94

203

1,226

2,056

2,481

4,699

8,863

22,028

30,327

57,273

5,434

10,371

24,965

33,084

64,615

2030

30

73

101

223

1,433

2,350

2,757

6,168

11,880

27,902

37,220

71,958

2035

34

81

108

247

1,697

2,643

3,171

7,196

13,954

30,839

39,977

80,769

2040

40

88

116

269

2045

43

94

123

289

1,886

2,937

3,584

8,077

15,839

33,776

44,112

88,112

2,263

3,378

3,860

8,958

17,914

36,713

46,869

96,923

2050

49

101

131

311

2,451

3,671

4,273

9,839

20,742

39,650

51,005

105,734

163 164 165

As shown in Table 1, social costs increase for greenhouse gases emitted in future years. This increase is a direct result of exponential damage functions representing scenarios in which global

ACS Paragon Plus Environment

Page 9 of 27

Environmental Science & Technology

166

systems become more stressed over time31. Using the increasing SC-CO2, SC-CH4, and SC-N2O values, we

167

derive a dynamic global warming impact (DGWI) which allows us to compare the monetized impacts of

168

greenhouse gas emissions relative to today’s environmental and economic conditions. Analogous to the

169

present value of money commonly discussed in economics, this approach weights the value of emissions

170

based on their monetized impact at a given time.

171

2.2

172

Deriving the Dynamic Global Warming Impact Leveraging the social cost of greenhouse gases, we generate a Dynamic Global Warming Impact

173

(DGWI) to compare of the monetized impact of a marginal emission released in a future year, to the

174

monetized impact of a marginal CO2 emission released in the present year. This comparison represents

175

the change in monetized impact of emissions due to dynamic greenhouse gas concentrations and socio-

176

economic damage functions. As shown in Equation 1, we derived the DGWI from a ratio of the social

177

cost of a particular greenhouse gas (GHG) in the future year (i), to the social cost of CO2 in the present

178

year. A full list of terms used within equations of this paper is located in the supporting information.

179 180

𝑆𝑜𝑐𝑖𝑎𝑙 𝐶𝑜𝑠𝑡 𝐺𝐻𝐺,𝑖

𝐷𝑦𝑛𝑎𝑚𝑖𝑐 𝐺𝑙𝑜𝑏𝑎𝑙 𝑊𝑎𝑟𝑚𝑖𝑛𝑔 𝐼𝑚𝑝𝑎𝑐𝑡𝐺𝐻𝐺,𝑖 = 𝑆𝑜𝑐𝑖𝑎𝑙 𝐶𝑜𝑠𝑡 𝐶𝑂 , 2020 2

(1)

181 182

DGWI values were generated for CO2, CH4, and N2O using the SC-CO2, SC-CH4, and SC-N2O,

183

respectively. The DGWI compares the monetized damage of one gas to another and includes

184

consideration for when that gas is emitted. Applying the DGWI converts individual gases (CO2, CH4, and

185

N2O) to a CO2 equivalent, similar to the GWP. For the purposes of this paper, we considered a present

186

year of 2020 in all analyses. Due to the range of discount rates used to estimate social costs, we

187

generated the same range of DGWI values. Generating these values required applying Equation 1 to the

188

four separate discount rates in each year. In each calculation, the social cost based on a specific discount

ACS Paragon Plus Environment

Environmental Science & Technology

189

rate was compared to the social cost of CO2 in the present based on the same discount rate. While the

190

impacts of all four discount rates are included in this paper, the 3% discount rate is the central value of

191

social cost estimates31 in the literature, and is thus considered the baseline for our analysis.

192

2.3

193

Developing Temporally Resolved LCA Data In order to apply the DGWI and demonstrate the monetized impact of emissions, we developed

194

temporally resolved LCA emissions of conventional electricity-generation technologies. These

195

technologies included coal, natural gas, nuclear, solar photovoltaic (PV), concentrating solar power

196

(CSP), and wind. Additionally, we established temporally resolved LCA emissions for post-combustion

197

carbon capture and sequestration (CCS) applied to both coal and natural gas. To generate temporal

198

emissions, we defined two primary phases over the lifetime of a technology. The first is construction,

199

which is assumed to occur in the first year of the lifetime. The second is operation, which occurs over

200

the entire 30-year span of the lifetime. We obtained emissions data for construction and operation from

201

existing literature20,33–41. Results of this data collection effort are summarized in Table 2. For the

202

purposes of this paper, CH4 and N2O operational emissions are considered negligible for PV, CSP,

203

nuclear, and wind due to their expected impact being less than 6 gCO2-eq per kWh in all scenarios. This

204

decision aligns with findings from a number of previous studies39–41. Details of references used to derive

205

emissions and an example emission profile for coal with CCS (Figure S2) are presented in the supporting

206

information.

207 208

ACS Paragon Plus Environment

Page 10 of 27

Page 11 of 27

Environmental Science & Technology

209

Table 2. Life Cycle Emissions: Construction and operational greenhouse gas emissions used to generate

210

temporally resolved LCA results for energy generation technologies20,33–41.

Construction Emissions (g/kWh) CO2 CH4 N2O 41 0.04 0.001 56 0.06 0.002 23 0.03 0.001 30 0.04 0.001 949 5.58 0.015 606 1.53 0.175 89 0.18 0.001 97 0.32 0.008

Technology Coal Coal CCS (90%) Natural Gas Natural Gas CCS (90%) PV CSP Nuclear Wind 211

Operational Emissions (g/kWh-year) CO2 CH4* N2O* 936 3 0.0001 152 5 0.0001 384 3.87 0.0001 69 4.53 0.0001 0.1 17 4 0.1 -

*Operational CH4 and N2O emissions considered negligible for low emissions technologies.

212 213

Integrating the temporally resolved LCA emissions with the DGWI values, we generated a

214

present value of emissions for each year represented by a mass of CO2-eq. As shown in Equation 2, this

215

present value represents emissions that are weighted by their impacts based on when they are released

216

relative to the impact of CO2 in the present. To compare with traditional LCA methods, we also

217

generated a mean present value of emissions over all years (n) of a technology’s lifetime as displayed in

218

Equation 3. This equation yields a single emissions value, but unlike standard methodology this value

219

includes temporally resolved impacts quantified in gCO2-eq per kWh.

220 221

𝑃𝑟𝑒𝑠𝑒𝑛𝑡 𝑉𝑎𝑙𝑢𝑒𝐺𝐻𝐺,𝑖 = 𝐸𝑚𝑖𝑠𝑠𝑖𝑜𝑛𝑠𝐺𝐻𝐺,𝑖 × DGWI𝐺𝐻𝐺,𝑖

(2)

222 ∑𝑛 223

𝑀𝑒𝑎𝑛 𝑃𝑟𝑒𝑠𝑒𝑛𝑡 𝑉𝑎𝑙𝑢𝑒 =

𝑃𝑟𝑒𝑠𝑒𝑛𝑡 𝑉𝑎𝑙𝑢𝑒𝐶𝑂2,𝑖 + 𝑃𝑟𝑒𝑠𝑒𝑛𝑡 𝑉𝑎𝑙𝑢𝑒𝐶𝐻4,𝑖 + 𝑃𝑟𝑒𝑠𝑒𝑛𝑡 𝑉𝑎𝑙𝑢𝑒𝑁2𝑂𝑖

𝑖=1

∑𝑛

𝐸𝑙𝑒𝑐𝑡𝑟𝑖𝑐𝑖𝑡𝑦 𝐺𝑒𝑛𝑒𝑟𝑎𝑡𝑒𝑑𝑖

𝑖=1

224

ACS Paragon Plus Environment

(3)

Environmental Science & Technology

225 226

2.4

Page 12 of 27

Incorporating the Social Cost of Carbon in Techno-Economic Analysis In addition to the alternative LCA method, we also considered a combined LCA/TEA approach.

227

This combined approach included integration of dynamic SC-CO2, SC-CH4, and SC-N2O values with

228

temporally resolved LCA to generate a yearly cost of emissions. Integrating these costs into TEA allowed

229

us to gauge the impact of greenhouse gas emissions on the levelized cost of electricity (LCOE). We

230

started by generating a discounted cash-flow rate of return TEA for each of the energy generation

231

technologies defined in Section 2.3. We used the National Renewable Energy Laboratory’s (NREL)

232

Annual Technology Baseline to define the capital and operational costs as well as other economic

233

parameters for each technology42. Details of these costs and parameters are located in Table S1 and

234

Table S2 of the supporting information.

235

With the conventional TEA cash flows defined, we then added an additional yearly cost based on

236

the greenhouse gas emissions and the corresponding social cost. As displayed in Equation 4, this cost

237

was generated by taking the emissions occurring in a given year multiplying them by the corresponding

238

social cost. As with the DGWI, we considered a range of SC-CO2, SC-CH4, and SC-N2O values based on the

239

2.5%, 3%, 3%-95th percentile, and 5% economic discount rates. To remain consistent with an analysis

240

conducted in the year 2020, we used the 2.5%, 3%, and 5% economic discount rates depending on the

241

scenario to shift the social cost values originally given in 2007 US dollars to 2020 US dollars 31,32.

242 243

𝐸𝑚𝑖𝑠𝑠𝑖𝑜𝑛𝑠 𝐵𝑎𝑠𝑒𝑑 𝑂𝑝𝑒𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙 𝐶𝑜𝑠𝑡𝐺𝐻𝐺,𝑖 = 𝐸𝑚𝑖𝑠𝑠𝑖𝑜𝑛𝑠𝐺𝐻𝐺,𝑖 × 𝑆𝑜𝑐𝑖𝑎𝑙 𝐶𝑜𝑠𝑡𝐺𝐻𝐺,𝑖

(4)

244 245

After defining all costs associated with a technology we solved the 30-year cash flow for the

246

LCOE. The LCOE represents the minimum selling price of electricity to offset production costs (including

247

emissions damages) and yield an acceptable internal rate of return. By incorporating emission costs

248

directly into the LCOE, we have merged temporally resolved LCA with TEA to account for the dynamic

ACS Paragon Plus Environment

Page 13 of 27

Environmental Science & Technology

249

impacts of emissions over time. It is important to note that this dynamic TEA represents an analysis of

250

private-costs. As a result, we assume that the external social costs of greenhouse gases will be

251

translated into private costs through a mechanism such as a carbon tax.

252

3. Results and Discussion

253

3.1.

Expanding Life Cycle Assessment to Include Temporal Impacts

254

Incorporating temporal impacts into LCA is based on the development of DGWI values combined

255

with time resolved LCA. Table 3 summarizes DGWI every five years from 2020-2050. A full list of annual

256

values is located in Table S3 of the supporting information. In an effort to capture the uncertainty

257

associated with future impacts, the social costs of each gas include the four different IAM discount rates.

258

Reviewing Table 3 we see that all DGWI values increase over time, demonstrating that future emissions

259

will have greater impact. Looking specifically at CO2, the DGWI values in the year 2050 range from a low

260

of 1.53 to a high of 2.17, corresponding to the 2.5% and 5% SC-CO2 scenarios, respectively. The baseline

261

scenario resulting from the 3% SC-CO2 discount rate shows growth to a DGWI of 1.64 over the 30-year

262

period. This foundationally means a CO2 emission occurring in 2050 would have a 64% higher monetized

263

impact than the same emission occurring in the year 2020.

264

Similar increases in future years occur across CH4 and N2O DGWI values with each gas demonstrating

265

its own specific impact based on its role in future climate scenarios. Looking at the initial analysis year of

266

2020, the range of CH4 and N2O DGWI values are significantly different than current GWP values. This

267

difference is the result of comparing monetized impact instead of CRF. For example, Table 3 shows that

268

the monetized impact of N2O is expected to be 317-392 times greater than CO2. Meanwhile, IPCC’s 100-

269

year GWP method yields the smaller ratio of 2659. Comparing the two metrics demonstrates the

270

importance going beyond CRF to include monetized impacts, especially when assessing the

271

contributions of N2O and CH4 in an analysis.

272

ACS Paragon Plus Environment

Environmental Science & Technology

Page 14 of 27

273

Table 3. Dynamic Global Warming Impact: comparison of social cost of greenhouse gases in future years

274

to the year 2020 for five year increments from 2020-2050. Year of

DGWI of CO2

DGWI of CH4

DGWI of N2O

Emission

5%

3%

2.50%

3%-95th

5%

3%

2.50%

3%-95th

5%

3%

2.50%

3%-95th

2020

1.00

1.00

1.00

1.00

45

29

26

26

392

357

355

317

2025

1.17

1.10

1.10

1.12

54

33

29

30

458

405

387

358

2030

1.33

1.19

1.18

1.24

63

38

32

34

525

452

435

398

2035

1.50

1.31

1.26

1.37

75

43

37

40

617

500

468

447

2040

1.75

1.43

1.35

1.49

83

48

42

45

700

548

516

488

2045

1.92

1.52

1.44

1.60

100

55

45

50

792

595

548

537

2050

2.17

1.64

1.53

1.72

108

60

50

54

917

643

597

585

275 276

Applying the DGWI to temporally resolved LCA of electricity generation technologies yielded the

277

mean present value of emissions shown in Figure 1. The figure also includes standard LCA results for

278

comparison. These standard results are based on IPCC’s 100 year GWP values of 28 gCO2-eq and 265

279

gCO2-eq for CH4 and N2O, respectively9. Comparing the results of standard LCA to the new DGWI method

280

shows an increase across all technologies, except PV where the 2.5% and 3%-95th scenarios result in

281

slightly lower values than standard LCA. The decrease in these PV scenarios comes from the 2020 CH4

282

DGWI of 26, which is lower than the standard 100-year GWP value of 28. Across all technologies, the 5%

283

discount rate consistently produces the largest increase in present value, as it represents the most

284

aggressive increase in social costs of future emissions. Following the 5% scenario, the 3%-95th, 3%, and

285

2.5% scenarios have varying impacts based upon the mix of greenhouse gases in construction and

286

operational phases.

ACS Paragon Plus Environment

Page 15 of 27

Environmental Science & Technology

287

Mean Present Value of Emissions (gCO2-eq kWh-1)

2000

60 DGWI 5%

1600

DGWI 3%-95th

40

DGWI 3%

1200

DGWI 2.5%

20

Standard LCA

800

0 PV

CSP

Nuclear

Wind

400 0 Coal

288

Natural Coal CCS Natural Gas Gas CCS

PV

CSP

Nuclear

Wind

289

Figure 1. Monetized Impact of Energy Technologies: Mean present value of emissions for electricity

290

generation technologies based on DGWI values corresponding to 2.5%, 3%, 3%-95th percentile, and 5 %

291

social cost discount rates.

292

When considering construction compared to operational emissions, it is important to recognize

293

that the majority of emissions from fossil-based technologies occur during operation. This makes the

294

mean present value of fossil-based technologies especially sensitive to temporal impact. Thus the

295

technology with the largest increase in emissions impact is coal. When using the 3% DGWI as our

296

baseline scenario, emissions from a coal power plant increase from a standard LCA value of 1031 gCO2-

297

eq/kWh up to a new mean present value of 1362 gCO2-eq/kWh. This 32% increase is primarily due to

298

large operational emissions over the 30 years of plant operation. Technologies with lower operational

299

emissions show a much smaller increase. For example, applying the 3% DGWI to wind power changes a

300

standard LCA value of 3.74 gCO2-eq/kWh to a new mean present value of 3.81 gCO2-eq/kWh, which is an

301

increase of only 1.9%.

ACS Paragon Plus Environment

Environmental Science & Technology

302

Comparing the results of the baseline 3% scenario to the 3%-95th percentile scenario across coal

303

CCS, natural gas CCS, PV, and wind shows that in these cases the 3%-95th percentile have a slightly lower

304

impact than the 3% scenario. For coal CCS and natural gas CCS, this is the result of significant

305

operational CH4 and N2O emissions. These emissions are weighted by the CH4 and N2O DGWI values

306

which are lower for the 3%-95th scenario than the 3% scenario across all years as shown in Table 3. For

307

wind and PV, the majority of emissions occur during construction in the first year. Again, looking at the

308

DGWI values in Table 3, the CH4 and N2O values in the initial year are lower for the 3%-95th scenario,

309

leading to a slightly lower overall impact. In general, it is important to reiterate that the change in

310

emission values associated with the mean present value does not represent a physical change in the

311

quantity of greenhouse gases emitted. Instead, these values represent emissions that are scaled by their

312

impact relative to the present year of analysis.

313

3.2.

Incorporating the Social Cost of Greenhouse Gases in Techno-Economic Analysis

314

Adding the social costs of greenhouse gases into TEA increases the cost of the electricity production.

315

This work evaluated the four different discount rates to produce four separate dynamic LCOE results for

316

each electricity generation technology evaluated. Figure 2 displays these four LCOE results, as well as

317

the LCOE from standard TEA methods that do not include costs of greenhouse gases. Across all

318

technologies the 3%-95th percentile discount rate has the largest impact on LCOE followed by the 2.5%,

319

3%, and 5%, respectively. The largest rise in LCOE occurs in coal where a standard TEA LCOE of 9¢2020

320

kWh-1 changes to a LCOE of 16¢2020 kWh-1 for the baseline 3% scenario. This 88% rise is primarily due to

321

long-term operational emissions over the 30-year lifetime of the coal power plant. For comparison, a

322

technology with lower operational emissions, such as wind, shows just a 1% increase, changing from a

323

standard LCOE of 5.58¢2020 kWh-1 to a new LCOE of 5.64¢2020 kWh-1 when the costs of greenhouse gases

324

are included.

ACS Paragon Plus Environment

Page 16 of 27

Environmental Science & Technology

Levelized Cost of Energy (US2020$ kWh-1)

Page 17 of 27

0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00

3%-95th Costs Included 2.5% Costs Included 3% Costs Included 5% Costs Included Standard TEA

Coal 325

Natural Coal CCS Natural Gas Gas CCS

PV

CSP

Nuclear

Wind

326

Figure 2. Economic Impact of Integrating Emissions: LCOE of electricity generation technologies including

327

social cost of greenhouse gas based operational emissions cost.

328 329

3.3.

Comparison to Existing Methods

330

The new DGWI LCA method produces results that are dissimilar to the majority of previous temporal

331

methods. In most previous CRF based methods, greenhouse gases occurring in early years of an LCA

332

timeframe have greater weight13–15,17,18. Our method shows that emissions occurring in later years

333

should be weighted with greater monetized impact. This difference is due to two major methodological

334

factors. The first factor is the time horizon. Previous methods often use a near term time horizon that

335

causes early emissions to have a greater cumulative impact than later emissions. By leveraging the social

336

cost of greenhouse gases, our method also inherently includes a time horizon. However, this horizon is

337

the year 2300, extending beyond the time horizon of many previous methods. Using the long-term time

338

horizon reduces the impact of residence time, but does not entirely remove it. The second factor

339

causing a difference is our inclusion of social damages. Previous studies have typically relied upon CRF or

340

global temperature change as the metric for impact. Our DGWI goes several steps beyond CRF and

341

temperature to include biological impact, economic impact, and other factors. The expected exponential

ACS Paragon Plus Environment

Environmental Science & Technology

342

increase in these damages causes later emissions to have a higher monetized impact. This increase

343

exceeds any reduction caused by our time horizon or decreased radiative efficiency due to increasing

344

greenhouse gas concentrations.

345

The TEA methods in this paper follow an increasing number of studies that include environmental

346

externalities in economic calculations. However, other studies typically do not include individual costs

347

for temporally resolved emissions of CO2, CH4, and N2O. Instead, they utilize 100-year GWP values to

348

equate greenhouse gas emissions to CO2 and then apply a single time independent external cost for CO2-

349

eq emissions43. One of the more comprehensive studies goes a step further to include temporal

350

resolution in CO2 and CH4 costs, but still utilizes CO2-eq within certain calculations 44. Determining which

351

methods are most appropriate for a cost-benefit analysis will largely depend upon how global social

352

damages are applied to industry through mechanisms such as a carbon tax.

353

3.4.

Limitations and Implications of Methods

354

The methods presented in this paper demonstrate two different approaches for quantifying

355

temporal impacts of greenhouse gas emissions. Both methods are based on the social costs of

356

greenhouse gases, which come with inherent limitations. The most prominent limitation is the

357

uncertainty of future climate, social, and economic systems. We have captured one major element of

358

uncertainty by including a range of economic discount rates provided with the social cost values. This

359

and other elements of uncertainty are addressed thoroughly within the technical documentation

360

provided by the Interagency Working Group on the Social Cost of Greenhouse Gases27,31,32. The

361

Interagency Working Group authors outline their use of multi-model ensemble, probabilistic analysis,

362

and scenario analysis to address uncertainty by obtaining frequency distributions for the social costs of

363

greenhouse gases. The numbers used in this paper are the central estimates of those distributions.

364

Future development of our methods could benefit by propagating the social cost distributions through

365

analysis of temporally resolved life cycle assessment.

ACS Paragon Plus Environment

Page 18 of 27

Page 19 of 27

366

Environmental Science & Technology

Beyond uncertainty, another limitation of the social cost estimates is that they represent the cost of

367

marginal emissions. As a result, they convey the damage of a one-ton emission on top of a

368

predetermined global emissions scenario. Therefore, we cannot use these values to predict the damage

369

from changes that would significantly alter global emissions. Performing such an analysis would require

370

varying the global emissions scenario within the IAM. This approach is possible, but it differs from the

371

intent of this paper, which is to demonstrate the monetized impact of including temporal resolution in

372

LCA. As a result, the methods of this paper are limited to analyzing scenarios that will not affect global

373

emissions trajectories. In the case of electricity production, these methods are used to compare the

374

construction and operation of specific power plants. The results should not be interpreted as the

375

average impact of altering the global or national electrical grid mix with these technologies.

376

Although limitations exist, the results of these new methods have some important implications for

377

the future of dynamic LCA. First, the current dynamic LCA practice of comparing CRF up to a given time

378

horizon results in findings that are opposite of monetized social damage estimates. This does not mean

379

current dynamic LCA practice is incorrect, but it does force us to consider the purpose of this

380

methodology. Considering impacts up to a given time horizon prioritizes the importance of certain

381

emissions relative to a given time-frame. This approach does make sense if the intent is limiting near

382

term global temperature rise to avoid climate tipping points. However, if the intent is to compare the

383

actual long-term impact of a technology then the social cost methods provide a more comprehensive

384

basis for comparison.

385

A second implication is the substantial effect of translating mid-point metrics such as quantities of

386

greenhouse gases into end-point metrics such as economic damage. In the example of electricity

387

generation, we see that using the DGWI in place of the GWP can change the result of coal power by up

388

to 62% in the most severe scenario. Change of this magnitude could significantly alter a comparison

389

between technologies. The inclusion of these end-point metrics undoubtedly comes with an increased

ACS Paragon Plus Environment

Environmental Science & Technology

390

level of uncertainty as mentioned above. However, this uncertainty is common within existing LCA,

391

climate science, and economic practice. As a result, new methods are under constant development to

392

address uncertainty in these fields.

393

The results of this work demonstrate that the inclusion of temporal considerations and monetized

394

impact can dramatically alter comparisons between technologies. Based on these results we have a

395

series of recommendations. First, it is important that LCA data and results include temporal resolution.

396

Averaging data or making assumptions regarding the timing of emissions can introduce significant

397

differences and skew results. Second, when comparing technologies in LCA or TEA, methods such as the

398

two presented in this paper should be used to compare temporal impact. The selection of a specific

399

method will be dependent upon the context of a particular comparison. Third, further work needs to be

400

carried out to identify other technologies beyond electricity generation where temporal resolution can

401

dramatically change results. There may be existing analyses where inclusion of temporal impact reverses

402

an established conclusion. Lastly, the methods in this paper should be expanded beyond marginal

403

emissions to include altered global emissions trajectories. Ultimately, there is need for a dynamic tool

404

that compares monetized temporal impacts of technologies and alters global emissions trajectories

405

based on those comparisons.

406

Supporting Information

407

Terminology definitions, explanation of temporally resolved LCA data, TEA data, expanded tables of

408

social costs of greenhouse gases and DGWI

409

Acknowledgments

410

The authors are grateful for support and text editing from the Quinn Research Group and Danna Quinn.

411

ACS Paragon Plus Environment

Page 20 of 27

Page 21 of 27

Environmental Science & Technology

412

Author Contributions: J.C.Q. conceived the study. E.S., J.B., and J.C.Q. designed the research. E.S. and

413

J.B. performed research. E.S., J.B., and J.C.Q. wrote the paper.

414

References

415

1. Cai, H., Markham, J., Jones, S., Benavides, P. T., Dunn, J. B., Biddy, M., Tao, L., Lamers, P. & Phillips, S.

416

Techno-Economic Analysis and Life-Cycle Analysis of Two Light-Duty Bioblendstocks: Isobutanol and

417

Aromatic-Rich Hydrocarbons. ACS Sustain. Chem. Eng. 6, 8790–8800 (2018).

418

2. Ahmed, A., Hassan, I., Ibn-Mohammed, T., Mostafa, H., Reaney, I. M., Koh, L. S. C., Zu, J. & Wang, Z. L.

419

Environmental life cycle assessment and techno-economic analysis of triboelectric nanogenerators.

420

Energy Environ. Sci. 10, 653–671 (2017).

421

3. Patel, M., Zhang, X. & Kumar, A. Techno-economic and life cycle assessment on lignocellulosic

422

biomass thermochemical conversion technologies: A review. Renew. Sustain. Energy Rev. 53, 1486–

423

1499 (2016).

424 425

4. Dutta, S., Neto, F. & Coelho, M. C. Microalgae biofuels: A comparative study on techno-economic analysis & life-cycle assessment. Algal Res. 20, 44–52 (2016).

426

5. Pennington, D. W., Potting, J., Finnveden, G., Lindeijer, E., Jolliet, O., Rydberg, T. & Rebitzer, G. Life

427

cycle assessment Part 2: Current impact assessment practice. Environ. Int. 30, 721–739 (2004).

428

6. Reap, J., Roman, F., Duncan, S. & Bras, B. A survey of unresolved problems in life cycle assessment.

429 430 431 432 433

Int J Life Cycle Assess 15 (2008). 7. Hellweg, S. & i Canals, L. M. Emerging approaches, challenges and opportunities in life cycle assessment. Science 344, 1109–1113 (2014). 8. McManus, M. C. & Taylor, C. M. The changing nature of life cycle assessment. Biomass Bioenergy 82, 13–26 (2015).

434

9. Myhre, G., Shindell, D., Bréon, F.-M., Collins, W., Fuglestvedt, J., Huang, J., Koch, D., Lamarque, J.-F.,

435

Lee, D., Mendoza, B., Nakajima, T., Robock, A., Stephens, G., Zhang, H., Aamaas, B., Boucher, O.,

ACS Paragon Plus Environment

Environmental Science & Technology

436

Dalsøren, S. B., Daniel, J. S., Forster, P., Granier, C., Haigh, J., Hodnebrog, Ø., Kaplan, J. O., Marston,

437

G., Nielsen, C. J., O’Neill, B. C., Peters, G. P., Pongratz, J., Ramaswamy, V., Roth, R., Rotstayn, L., Smith,

438

S. J., Stevenson, D., Vernier, J.-P., Wild, O., Young, P., Jacob, D., Ravishankara, A. R. & Shine, K.

439

Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth

440

Assessment Report of the Intergovernmental Panel on Climate Change. Chapter 8: Anthropogenic and

441

Natural Radiative Forcing. 82 (Intergovernmental Panel on Climate Change).

442

10.

Joos, F., Roth, R., Fuglestvedt, J. S., Peters, G. P., Enting, I. G., von Bloh, W., Brovkin, V., Burke, E.

443

J., Eby, M., Edwards, N. R., Friedrich, T., Fr?licher, T. L., Halloran, P. R., Holden, P. B., Jones, C.,

444

Kleinen, T., Mackenzie, F. T., Matsumoto, K., Meinshausen, M., Plattner, G.-K., Reisinger, A.,

445

Segschneider, J., Shaffer, G., Steinacher, M., Strassmann, K., Tanaka, K., Timmermann, A. & Weaver,

446

A. J. Carbon dioxide and climate impulse response functions for the computation of greenhouse gas

447

metrics: a multi-model analysis. Atmospheric Chem. Phys. 13, 2793–2825 (2013).

448

11.

Shine, K. P., Fuglestvedt, J. S., Hailemariam, K. & Stuber, N. Alternatives to the global warming

449

potential for comparing climate impacts of emissions of greenhouse gases. Clim. Change 68, 281–302

450

(2005).

451 452 453

12.

Shine, K. P. The global warming potential—the need for an interdisciplinary retrial: An editorial

comment. Clim. Change 96, 467–472 (2009). 13.

O’Hare, M., Plevin, R. J., Martin, J. I., Jones, A. D., Kendall, A. & Hopson, E. Proper accounting for

454

time increases crop-based biofuels’ greenhouse gas deficit versus petroleum. Environ. Res. Lett. 4,

455

024001 (2009).

456 457

14.

Kendall, A., Chang, B. & Sharpe, B. Accounting for Time-Dependent Effects in Biofuel Life Cycle

Greenhouse Gas Emissions Calculations. Environ. Sci. Technol. 43, 7142–7147 (2009).

ACS Paragon Plus Environment

Page 22 of 27

Page 23 of 27

458

Environmental Science & Technology

15.

Levasseur, A., Lesage, P., Margni, M., Deschênes, L. & Samson, R. Considering Time in LCA:

459

Dynamic LCA and Its Application to Global Warming Impact Assessments. Environ. Sci. Technol. 44,

460

3169–3174 (2010).

461 462 463 464 465 466 467

16.

Schwietzke, S., Griffin, W. M. & Matthews, H. S. Relevance of Emissions Timing in Biofuel

Greenhouse Gases and Climate Impacts. Environ. Sci. Technol. 45, 8197–8203 (2011). 17.

Kendall, A. & Price, L. Incorporating Time-Corrected Life Cycle Greenhouse Gas Emissions in

Vehicle Regulations. Environ. Sci. Technol. 46, 2557–2563 (2012). 18.

Kendall, A. Time-adjusted global warming potentials for LCA and carbon footprints. Int. J. Life

Cycle Assess. 17, 1042–1049 (2012). 19.

Levasseur, A., Lesage, P., Margni, M., Brandão, M. & Samson, R. Assessing temporary carbon

468

sequestration and storage projects through land use, land-use change and forestry: comparison of

469

dynamic life cycle assessment with ton-year approaches. Clim. Change 115, 759–776 (2012).

470 471 472 473 474 475 476 477 478 479 480 481

20.

Ravikumar, D., Seager, T. P., Chester, M. V. & Fraser, M. P. Intertemporal Cumulative Radiative

Forcing Effects of Photovoltaic Deployments. Environ. Sci. Technol. 48, 10010–10018 (2014). 21.

Yang, J. & Chen, B. Global warming impact assessment of a crop residue gasification project—A

dynamic LCA perspective. Appl. Energy 122, 269–279 (2014). 22.

Daystar, J., Venditti, R. & Kelley, S. S. Dynamic greenhouse gas accounting for cellulosic biofuels:

implications of time based methodology decisions. Int. J. Life Cycle Assess. 22, 812–826 (2017). 23.

Ravikumar, D., Wender, B., Seager, T. P., Fraser, M. P. & Tao, M. A climate rationale for research

and development on photovoltaics manufacture. Appl. Energy 189, 245–256 (2017). 24.

Reisinger, A., Meinshausen, M. & Manning, M. Future changes in global warming potentials

under representative concentration pathways. Environ. Res. Lett. 6, 024020 (2011). 25.

Caldeira, K. & Kasting, J. F. Insensitivity of global warming potentials to carbon dioxide emissions

scenarios. Nature 366, 251–253 (1993).

ACS Paragon Plus Environment

Environmental Science & Technology

482

26.

Farquharson, D., Jaramillo, P., Schivley, G., Klima, K., Carlson, D. & Samaras, C. Beyond Global

483

Warming Potential: A Comparative Application of Climate Impact Metrics for the Life Cycle

484

Assessment of Coal and Natural Gas Based Electricity: Beyond Global Warming Potential. J. Ind. Ecol.

485

21, 857–873 (2017).

486

27.

Technical Support Document:- Social Cost of Carbon for Regulatory Impact Analysis - Under

487

Executive Order 12866. (Interagency Working Group on Social Cost of Carbon, United States

488

Government, 2010).

489 490 491

28.

Burke, M., Hsiang, S. M. & Miguel, E. Global non-linear effect of temperature on economic

production. Nature 527, 235–239 (2015). 29.

Delucchi, M. A. A Lifecycle Emissions Model (LEM): Lifecycle Emissions from Transportation Fuels,

492

Motor Vehicles, Transportation Modes, Electricity Use, Heating and Cooking Fuels, and Materials.

493

(Institute of Transportation Studies University of Califormia Davis, 2003).

494

30.

Delucchi, M. A. & Lipman, T. Appendix D: CO2 Equivalency Factors, An Appendix to the Report: A

495

Lifecycle Emissions Model (LEM): Lifecycle Emissions From Transportation Fuels, Motor Vehicles,

496

Transportation Modes, Electricity Use, Heating and Cooking Fuels, and Materials. (Institute of

497

Transportation Studies University of California Davis, 2003).

498

31.

Technical Support Document:- Technical Update of the Social Cost of Carbon for Regulatory

499

Impact Analysis - Under Executive Order 12866. (Interagency Working Group on Social Cost of

500

Greenhouse Gases, United States Government, 2016).

501

32.

Addendum to Technical Support Document on Social Cost of Carbon for Regulatory Impact

502

Analysis under Executive Order 12866: Application of the Methodology to Estimate the Social Cost of

503

Methane and the Social Cost of Nitrous Oxide. (Interagency Working Group on Social Cost of

504

Greenhouse Gases, United States Government, 2016).

ACS Paragon Plus Environment

Page 24 of 27

Page 25 of 27

505 506 507 508 509

Environmental Science & Technology

33.

Skone, T. & James, R. Life Cycle Analysis: Supercritical Pulverized Coal (SCPC) Power Plant. 108

(National Energy Technology Laboratory, 2010). 34.

Skone, T. & James, R. Life Cycle Analysis: Natural Gas Combined Cycle (NGCC) Power Plant. 151

(National Energy Technology Laboratory, 2010). 35.

Alvarez, R. A., Zavala-Araiza, D., Lyon, D. R., Allen, D. T., Barkley, Z. R., Brandt, A. R., Davis, K. J.,

510

Herndon, S. C., Jacob, D. J., Karion, A., Kort, E. A., Lamb, B. K., Lauvaux, T., Maasakkers, J. D.,

511

Marchese, A. J., Omara, M., Pacala, S. W., Peischl, J., Robinson, A. L., Shepson, P. B., Sweeney, C.,

512

Townsend-Small, A., Wofsy, S. C. & Hamburg, S. P. Assessment of methane emissions from the U.S. oil

513

and gas supply chain. Science eaar7204 (2018). doi:10.1126/science.aar7204

514

36.

Hou, G., Sun, H., Jiang, Z., Pan, Z., Wang, Y., Zhang, X., Zhao, Y. & Yao, Q. Life cycle assessment of

515

grid-connected photovoltaic power generation from crystalline silicon solar modules in China. Appl.

516

Energy 164, 882–890 (2016).

517 518 519 520 521

37.

Skone, T. J. Power Generation Technology Comparison from a Life Cycle Perspective. (National

Energy Technology Laboratory, 2013). 38.

Skone, T. J. Role of Alternative Energy Sources: Nuclear Technology Assessment. 143 (National

Energy Technology Laboratory, 2012). 39.

Burkhardt, J. J., Heath, G. & Cohen, E. Life Cycle Greenhouse Gas Emissions of Trough and Tower

522

Concentrating Solar Power Electricity Generation: Systematic Review and Harmonization. J. Ind. Ecol.

523

16, S93–S109 (2012).

524

40.

Hsu, D. D., O’Donoughue, P., Fthenakis, V., Heath, G. A., Kim, H. C., Sawyer, P., Choi, J.-K. &

525

Turney, D. E. Life Cycle Greenhouse Gas Emissions of Crystalline Silicon Photovoltaic Electricity

526

Generation: Systematic Review and Harmonization. J. Ind. Ecol. 16, S122–S135 (2012).

527 528

41.

Warner, E. S. & Heath, G. A. Life Cycle Greenhouse Gas Emissions of Nuclear Electricity

Generation: Systematic Review and Harmonization. J. Ind. Ecol. 16, S73–S92 (2012).

ACS Paragon Plus Environment

Environmental Science & Technology

529 530 531

42.

National Renewable Energy Laboratory. 2018 Annual Technology Baseline. (2018). Available at:

https://atb.nrel.gov/. (Accessed: 4th December 2018) 43.

Corona, B., Cerrajero, E., López, D. & San Miguel, G. Full environmental life cycle cost analysis of

532

concentrating solar power technology: Contribution of externalities to overall energy costs. Sol.

533

Energy 135, 758–768 (2016).

534

44.

Rhodes, J. D., King, C., Gulen, G., Olmstead, S. M., Dyer, J. S., Hebner, R. E., Beach, F. C., Edgar, T.

535

F. & Webber, M. E. A geographically resolved method to estimate levelized power plant costs with

536

environmental externalities. Energy Policy 102, 491–499 (2017).

537

ACS Paragon Plus Environment

Page 26 of 27

Page 27 Environmental of 27 Science & Technology

CO2 N2O

CH4

Social Costs of GHGs

S

New LCA Methods 200 0

Impact of Emissions

Temporal LCA

New Method Old Method

160 0

120 0

800

400

ACS Paragon Plus Environment 0

Coal

Natural G as

Coal CCS

Natural G as CCS

Technologies Evaluated