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