Climatic Change (2015) 133:169–178 DOI 10.1007/s10584-015-1471-6

Effect of methane leakage on the greenhouse gas footprint of electricity generation Nicolas Sanchez II 1,2 & David C. Mays 1

Received: 10 December 2014 / Accepted: 8 July 2015 / Published online: 23 July 2015 # Springer Science+Business Media Dordrecht 2015

Abstract For the purpose of generating electricity, what leakage rate renders the greenhouse gas (GHG) footprint of natural gas equivalent to that of coal? This paper answers this question using a simple model, which assumes that the comprehensive GHG footprint is the sum of the carbon dioxide-equivalent emissions resulting from (1) electricity generation and (2) natural gas leakage. The emissions resulting from electricity generation are taken from published lifecycle assessments (LCAs), whereas the emissions from natural gas leakage are estimated assuming that natural gas is 80 % methane, whose global warming potential (GWP) is calculated using equations provided by the Intergovernmental Panel on Climate Change (IPCC). Results, presented on a straightforward plot of GHG footprint versus time horizon, show that natural gas leakage of 2.0 % or 4.8 % eliminates half of natural gas’s GHG footprint advantage over coal at 20- or 100-year time horizons, respectively. Leakage of 3.9 % or 9.1 % completely eliminates the GHG footprint advantage at 20- and 100-year time horizons, respectively. A two-parameter power law approximation of the IPCC’s equation for GWP is utilized and gives equivalent results. Results indicate that leakage control is essential for natural gas to deliver a smaller GHG footprint than coal.

1 Introduction Recent advances in directional drilling and hydraulic fracturing have increased the development of unconventional natural gas, which includes tight sand gas, shale gas, and coal bed methane, each of which are 70–90 % methane by volume (Moore et al. 2014). Compared to coal, electricity generation from natural gas has a smaller greenhouse gas (GHG) footprint (Brandt et al. 2014), where GHG footprint is defined as grams of carbon dioxide-equivalent emissions per kilowatt-hour of electricity generated [gCO2e/kWh]. However, because methane

* David C. Mays [email protected] 1

Department of Civil Engineering, University of Colorado Denver, Campus Box 113, PO Box 173364, Denver, CO 80217-3364, USA

2

Present address: Geo-Solutions Inc., 610 Garrison Street, Unit D, Lakewood, CO 80215, USA

170

Climatic Change (2015) 133:169–178

is a potent greenhouse gas, leakage could diminish or eliminate natural gas’s GHG footprint advantage over coal (Tollefson 2013). Methane leakage estimates vary widely, from 0.4 % to 10 % of total production (Table 1), and there is particular concern that unconventional natural gas may have higher leakage than conventional natural gas (Schneising et al. 2014). For example, Howarth et al. (2011) estimated leakage of 3.6–7.9 % for unconventional natural gas, compared to 1.7–6.0 % for conventional natural gas. Victor (1992) estimated globallyaveraged methane leakage from 1978 to 1987 using isotopic data reported by Cicerone and Oremland (1988), resulting in an estimated leakage of 3.6 %, which is approximately the median of Table 1. More recently, Moore et al. (2014) contrasted the U.S. Environmental Protection Agency’s revised leakage estimate of 1.7–2.8 % with leakage estimates by others of 3.3–5.6 %. To place these leakage estimates in context, this paper answers the following question: What is the threshold rate of natural gas leakage that renders the GHG footprint of natural gas equivalent to that of coal for the purpose of generating electricity? Using a variety of methods, several answers to this question have been reported (Table 2), with threshold estimates of 2.0– 14 % depending on the time horizon chosen. The importance of this question is indicated by its discussion by policymakers. For example, U.S. Senator Ron Wyden of Oregon cited a threshold of 3 % (Showstack 2013), which is close to Alvarez et al.’s (2012) result of 3.2 % estimated using technology warming potential (TWP), their metric that accounts for technology substitution and integrates results over all time horizons. More recently, Zhang et al. (2014) showed how leakage from 0 to 9 % and generation plant efficiency together control the cumulative GHG footprint of electricity generation, reporting cumulative global mean temperature changes over time horizons from 0 to 100 year. Considering the scientific and policy importance of natural gas leakage, this paper offers an alternative—and very simple—model that shows the effect of leakage in terms of global warming potential (GWP). Table 1 Reported estimates of methane leakage Reference

Estimate

Allen et al. (2013)

0.42–0.49 %

Moore et al. (2014)1a

1.7–2.8 %

Jiang et al. (2011)

2–3 %

Wigley (2011) O’Donoughue et al. (2014)

2.5 % 1.3–4 %

Hayhoe et al. (2002)2a

2–4 %

Burnham et al. (2012)

0.97–5.47 %

Victor (1992)

3.6 %

Howarth et al. (2011)3a

1.7–6.0 %

Moore et al. (2014)1b

3.3–5.6 %

Hayhoe et al. (2002)2b

1–10 %

Howarth et al. (2011)3b

3.6–7.9 %

Schneising et al. (2014)

9.1–10.1 %

1

(a) EPA estimate or (b) non-EPA estimate

2

(a) US estimate or (b) world estimate

3

(a) conventional or (b) unconventional natural gas

Climatic Change (2015) 133:169–178

171

Table 2 Threshold leakage rates at which natural gas and coal have equal greenhouse gas (GHG) footprints for electricity generation Time Horizon 20 year

100 year

Jiang et al. (2011)

7%

14 %

Alvarez et al. (2012)1

3.2 %

3.2 %

Zhang et al. (2014)

2%

10 %

This Study

3.9 %

9.4 %

1

based on Technology Warming Potential, which integrates across time horizons

2 Methods The conceptual model (Fig. 1) is that natural gas, assumed to be 80 % methane, is either (1) combusted for electricity generation, or (2) leaked into the atmosphere. Here, leakage includes all natural gas not combusted during production, refining, and transmission, including flaring, which releases approximately 2 % of the natural gas, and venting, which releases 100 % of the natural gas (Jiang et al. 2011). The carbon dioxide-equivalent emissions resulting from electricity generation are estimated using published LCAs. The additional emissions due to natural gas leakage are approximated using the GWP of natural gas. Assuming the 20 % nonmethane components rapidly oxidize to CO2 (Zhang et al. 2014), the GWP of natural gas (NG) is modeled as GW PN G ðtÞ ¼ 0:20GW PCO2 þ 0:80GW PCH 4 ðtÞ;

ð1Þ

where GWPCO2 =1 by definition and GWPCH4(t) depends on the time horizon t as defined by the Intergovernmental Panel on Climate Change (IPCC) in (Myhre et al. 2013): GW PCH 4 ðt Þ ¼

AGW PCH 4 ðt Þ ; AGW PCO2 ðtÞ

ð2Þ

where AGWP is the absolute GWP in units of [W yr kg−1 m−2] and t is the time horizon [yr] measured from the time of emission. For methane, AGWP is defined by

Fig. 1 Conceptual model for estimating the greenhouse gas (GHG) footprint of electricity from natural gas

172

Climatic Change (2015) 133:169–178

Myhre et al.’s (2013) equation (8.SM.19) with coefficients taken from Table 8.A.1 and page 8SM-17:    W ⋅yr  ð3Þ AGW PCH 4 ðtÞ ¼ 2:61  10−12 2 1−e−t=12:4yr ; m ⋅kg whose decay term reflects methane’s atmospheric time scale of 12.4 years. For carbon dioxide, AGWP is defined by Myhre et al.’s (2013) equation (8.SM.11) with coefficients taken from Tables 8.A.1 and 8.SM.10:      −16 W −13 W ⋅yr −t=394:4yr ⋯ t þ 1:55  10 1−e AGW PCO2 ðt Þ ¼ 3:82  10 m2 ⋅kg m2 ⋅kg     ð4Þ    W ⋅yr W ⋅yr þ 1:81  10−14 2 1−e−t=36:54yr þ 2:09  10−15 2 1−e−t=4:304yr ; m ⋅kg m ⋅kg which includes one linear growth term and three decay terms with time scales of 394.4, 36.54, and 4.304 years. IPCC’s Eqs. (2)-(4) also include the unit conversions described on Myhre et al. (2013) page 8SM-15, specifically 3.52×10−10 ppb/kg for CH4 and 1.28×10−10 ppb/kg for CO2. As reported in Myhre et al. (2013), GWPCH4 =84 for t=20 year and GWPCH4 =28 for t=100 year. The goal of this paper, in large part, is to simplify the discussion of how natural gas leakage influences its GHG footprint for electricity generation. Toward that end, we sought a simplification of the IPCC equations presented above. To simplify Eqs. (2)-(4), a search was performed for an approximation with no more than two adjustable parameters fitted to two significant figures. After identifying power law equations as the most promising class, a grid search of two-dimensional parameter space was used to construct the following approximation of IPCC Eqs. (2)-(4) for the GWP of methane: GW PCH 4 ðtÞ≈790t−0:73 :

ð5Þ

When compared to IPCC Eqs. (2)-(4) at 17 logarithmically-spaced time horizons from 20 to 500 year, chosen to avoid over-weighting longer time horizons, Eq. (5) has a root mean squared error (RMSE) of 3.8 % and absolute error less than or equal to 6.1 % (Fig. 2). As shown below, this simpler approximation gives equivalent results as IPCC Eqs. (2)-(4). LCA results are taken from Turconi et al. (2013), who compiled LCA studies on electricity generation over the previous 15 years and reported results in terms of greenhouse gases, NOx, and SO2. Here we assume that natural gas leakage generates negligible emissions of the combustion byproducts NOx and SO2. Moreover, NOx and SO2 are considered local pollutants while greenhouse gases are considered global pollutants (Jaramillo et al. 2007). Specifically, we focus on the two greenhouse gases CO2 and CH4 that drive essentially all of the climatic change resulting from electricity generation (Zhang et al. 2014). For the purpose of this study, the best available efficiencies were used for both coal and natural gas as to compare the best available technologies. From 18 studies on combined cycle natural gas plants (the best available technology), Turconi et al. (2013) reported LCAs ranging from 380 to 590 gCO2e/kWh with a median of 450 gCO2e/kWh. From seven studies on coal gasification plants (again the best available technology), they reported LCAs ranging from 686 to 950 gCO2e/kWh with a median of 800 gCO2e/kWh, including an update described below. By way of comparison, the renewable energy source with the largest GHG footprint was solar power

Climatic Change (2015) 133:169–178

173

by photovoltaics, for which 22 studies gave LCAs ranging from 13 to 190 gCO2e/kWh with a median of 53 gCO2e/kWh. The best-case GHG footprint of coal reported in Turconi et al. (2013) was 660 gCO2e/kWh, based on a study by May and Brennan (2003), whose corrected Table 2 (May and Brennan 2004) indicates CO2 emissions of 660 g/kWh and CH4 emissions of 0.974 g/kWh. Accordingly, we compute the updated LCA as 686 gCO2e/kWh, estimated by adding the fuel-related GHG footprint of 26 gCO2e/kWh reported in Table S1 of Turconi et al. (Turconi et al. 2013). This best-case scenario forms the basis for the following model for the GHG footprint (GF) from coal: G F coal ðtÞ ¼ GF o;coal þ GW PCH 4 ðt ÞRF CH 4 ;

ð6Þ

where GFo,coal =660 gCO2/kWh, GWPCH4(t) is the global warming potential of methane [gCO2e/gCH4] as a function of time horizon t, and RFCH4 =0.974 g/kWh is the methane reference flow (RF) or a factor to ensure that products are compared equivalently. In this equation, the first term represents CO2 emissions and the second term represents non-CO2 GHG emissions, assumed to be exclusively CH4 emissions during mining. Using this equation, with a time horizon of t=20 year, GFcoal =742 gCO2e/kWh, of which 11 % results from methane emissions; with a time horizon of t=100 year, GFcoal =687 gCO2e/kWh, of which 4 % results from methane emissions. These percentages are similar to those reported by 100 IPCC (2013) approximation

90

Global Warming Potential (GWP) [-]

80

70

60

50

40

30

20

10

0 20

30

40

50

100

200

300

400

500

Time Horizon [yr]

Fig. 2 Exact and approximate equations for the global warming potential (GWP) of methane. Exact equations are given by (2)-(4) provided by the Intergovernmental Panel on Climate Change (IPCC) reported in Myhre et al. (2013). The approximate equation is (5) in the text

174

Climatic Change (2015) 133:169–178

Steinmann et al. (2014), who found that methane emissions account for 12 % of the GHG footprint of electricity generation by coal for t=20 year and 6 % for t=100 year. The 100 year GHG footprint is nearly identical to the corrected LCA of 686 gCO2e/kWh calculated above (May and Brennan 2003, 2004). The best-case GHG footprint of natural gas reported in Turconi et al. (2013) was 380 gCO2e/kWh, based on independent studies by Voorspools et al. (2000) and Briem et al. (2004), the latter of which contained sufficient information to calculate a carbon mole balance from the reported emissions of CO2, CO, and CH4 but neglecting non-methane volatile organic compounds (Table 3): C → CO2 þ CO þ CH4

ð7Þ

Assuming the natural gas is 80 % methane CH4 and 20 % ethane C2H6 gives a calculated molecular weight of 18.85 g/mol and a molecular weight per mole of carbon of 15.71 g/molc. Using values shown in Table 3, this carbon mole balance gives a reference flow of RFNG =(8.60 molc/kWh)(15.71 g/molc)=135.2 gNG/kWh, which is similar to the standard estimate of 153 gCH4/kWh (U.S. Energy Information Administration 2014). Assuming the natural gas is 80 % methane, as a check the reference flow of methane is found to be RFCH4 =0.80(8.60 molc/kWh)=6.90 mol/kWh. Accordingly, the reported methane emissions of 0.0310 mol/kWh correspond to 0.45 % leakage, which is in the range of 0.42–0.49 % reported by Allen et al. (2013). Howarth (2014) described this leakage as Bthe best possible performance,^ which is consistent with this study being the best case LCA reported by Turconi et al. (2013). The GHG footprints from electricity generation and methane leakage (Fig. 1) are combined using the principle of mass balance: mt ¼ m c þ m l ;

ð8Þ

where mt is the total mass of natural gas produced, mc is the mass combusted, and ml is the mass leaked. The leakage rate is defined as l=ml/mt. The total GHG footprint as a function of t and l is G F NG ðt; lÞ ¼ GF o;N G þ GW PNG ðt Þ

ml ; E

ð9Þ

where GFo,NG =377 gCO2/kWh (Table 3), GWPNG(t) is the global warming potential of natural gas [gCO2e/gNG] as a function of time horizon t, and ml/E is the mass leaked per kilowatthour of electricity generated. The term ml/E is calculated from substituting mt =ml/l into the mass balance Eq. (8) and dividing by E: ml l ¼ RF N G ; E 1−l

ð10Þ

Table 3 Carbon mole balance for natural gas (NG) from Briem et al. (2004)

flow [g/kWh] MW [g/molC] flow [molC/kWh]

NG

CO2

135.2

377

15.71 8.60

44.01 8.57

CO

CH4

0.194

0.498

28.01 0.00693

16.04 0.0310

Climatic Change (2015) 133:169–178

175

where RFNG =mc/E is the natural gas reference flow. Accordingly, the GHG footprint of electricity generation by natural gas as a function of time horizon and additional leakage is G F N G ðt; lÞ ¼ GF o;N G þ GW PN G ðt Þ⋅RF N G

l ; 1−l

ð11Þ

where RFNG =135.3 gNG/kWh, GWPNG(t) is given by (1), whose term GWPCH4(t) can be calculated from IPCC Eqs. (2)-(4) or estimated from the simpler approximation Eq. (5). For example, using a time horizon of t=20 year and assuming leakage l=2.5 % gives GFNG =611 gCO2e/kWh, of which 38 % results from leakage. Using a time horizon of 100 years, leakage accounts for 17 % of the total GHG footprint. These figures are similar to Hauck et al.’s (2014), who reported that upstream emissions at 20 and 100 year time horizons constituted 31 % and 19 % of the total GHG footprint, respectively.

3 Results Figure 3 shows the GHG footprint of electricity generation by natural gas, coal, and solar, based on the best-case LCAs reported in Turconi et al. (2013) for time horizons from 20 to 500 years. As shown on Fig. 3, leakage always reduces natural gas’s GHG footprint advantage 1000

900

GHG Footprint [gCO2 e/kWh]

800

700 COAL 600 8% 500

4% 1%

400

2%

0%

300

200

100 SOLAR 0 20

30

40

50

100

200

300

400

500

Time Horizon [yr]

Fig. 3 Greenhouse gas (GHG) footprint of electricity generation for coal, natural gas, and solar versus time horizon. For natural gas, leakage is indicated from 0 % to 8 %. Results are estimated with Eq. (11) and IPCC Eqs. (2)-(4) using the minimum GHG footprints of coal, natural gas, and solar listed in Turconi et al. (2013)

176

Climatic Change (2015) 133:169–178

over coal. With enough leakage, particularly at shorter time horizons, the GHG advantage is eliminated entirely, giving natural gas a larger GHG footprint than coal. The threshold leakage rate, l*, at which natural gas has a GHG footprint equal to that of coal, is calculated by setting GFNG =GFcoal, choosing a time horizon, and then solving for the threshold leakage rate: l* ¼

G F o;coal þ GW PCH 4 ðtÞRF CH 4 −G F o;N G : GW PN G ðtÞR F NG þ G F o;coal þ GW PCH 4 ðtÞR F CH 4 −G F o;N G

ð12Þ

From Eq. (12), any leakage over 3.9 % will eliminate the GHG advantage of natural gas for any time horizon less than 20 years, and any leakage over 9.1 % will eliminate any GHG advantage for any time horizon less than 100 year. Using the approximate Eq. (5), results are 3.7 % and 9.4 %, respectively. Considering the range of uncertainty in methane leakage (Table 1), these results are equivalent. Using the same approach, finding the leak rate l*1/2 at which GFNG =½(GFo,coal +GFo,NG) determines the leakage that eliminates half of natural gas’s GHG footprint advantage over coal: l*1=2 ¼

GF o;coal þ GW PCH 4 ðt ÞR F CH 4 −GF o;N G : 2GW PN G ðtÞR F N G þ G F o;coal þ GW PCH 4 ðtÞR F CH 4 −G F o;NG

ð13Þ

This leakage is 2.0 % for a time horizon of 20 years and 4.8 % at 100 years using IPCC Eqs. (2)-(4). Using the approximation Eq. (5), equivalent results are 1.9 % and 4.9 %, respectively.

4 Discussion The approach presented here is simple, so by construction, this analysis has a number of limitations. First, because the required equations are not provided in Myhre et al. (2013), the equations for GWP do not include (a) the additional GWP from burning fossil methane or (b) the additional GWP from considering climate-carbon feedbacks. Taken together, these effects increase methane’s 20 year GWP from 84 to 87, and methane’s 100 year GWP from 28 to 36. Accounting for these effects would increase the GHG footprint of natural gas relative to coal resulting in slightly lower threshold leakage rates. Second, this analysis omits variability in the reference flow (RF), and consequently this analysis omits the important effects of generation plant efficiency (Zhang et al. 2014). Third, this analysis omits two other factors associated with replacing coal-fired with natural gas-fired generation plants that were considered by Wigley (2011), specifically a reduction in coal mining-associated methane emissions, and a reduction in SO2 emissions. None of these limitations, however, is expected to have a qualitative impact on the results in Fig. 3. Fourth, because this analysis is strictly limited to GHG footprint, it does not consider other factors that could make electricity generation by natural gas preferable to electricity generation by coal, such as particulate air pollution, construction cost, and plant operation. The question addressed here is strictly about the climate benefits of substituting natural gas for coal. The results in Fig. 3 explicitly show the effect of the selected time horizon. This was a deliberate choice that sidesteps the debate between the 20-year time horizon, which is argued to avoid short-term climate tipping points (e.g., Howarth 2014), and the 100-year time horizon, which is argued to account for long-term climate dynamics (e.g., Fearnside 2002). Zhang et al. (2014) addressed this debate directly, emphasizing that tradeoffs between energy sources

Climatic Change (2015) 133:169–178

177

depend on the time horizon chosen, while Alvarez et al. (2012) cleverly sidestepped this debate by using their technology warming potential, which integrates across all time horizons. It is also worth noting that the 100-year time horizon was adopted as Ba policy decision, not a scientific result^ (Fearnside 2002), and that the IPCC itself stated that time horizons of 20 or 100 years have no special significance (Myhre et al. 2013). The uncertainty surrounding actual leakage is well documented (Moore et al. 2014). As a case in point, Heath et al. (2014) harmonized a number of studies on the GHG footprint of shale gas to account for varying data sources, system boundaries, and models. But they did not harmonize for leakage because they considered the data too sparse and its treatment in the literature too inconsistent. Whatever leakage rates are, it is clear that minimizing them will maximize the GHG benefit of switching from coal to natural gas, and in this context Senator Wyden’s point bears repeating: If a certain leakage rate renders natural gas equivalent to coal, then the target leakage must be less than that rate (Showstack 2013). The goal is to gain climate benefits through fuel substitution, not to break even, nor (as shown possible on Fig. 3) to cause increased warming. Acknowledgments The authors thank Indrani Pal for reviewing a preliminary draft of this paper, and three anonymous referees whose feedback helped to improve the clarity and rigor of the presentation. Contributions NS developed the model, derived the equations, and created Figs. 2 and 3. DCM updated these results using the most recent IPCC report, created Fig. 1, and drafted the paper.

References Allen DT, Torres VM, Thomas J, Sullivan DW, Harrison M, Hendler A, Herndon SC, Kolb CE, Fraser MP, Hill AD, Lamb BK, Miskimins J, Sawyer RF, Seinfeld JH (2013) Measurements of methane emissions at natural gas production sites in the United States. Proc Natl Acad Sci U S A 110:17768–17773 Alvarez RA, Pacala SW, Winebrake JJ, Chameides WL, Hamburg SP (2012) Greater focus needed on methane leakage from natural gas infrastructure. Proc Natl Acad Sci U S A 109:6435–6440 Brandt AR, Heath GA, Kort EA, O’Sullivan F, Petron G, Jordaan SM, Tans P, Wilcox J, Gopstein AM, Arent D, Wofsy S, Brown NJ, Bradley R, Stucky GD, Eardley D, Harriss R (2014) Methane leaks from north american natural gas systems. Science 343:733–735 Briem S, Blesl M, Fahl U, Ohl M (2004) Fossil gefeuerte kraftwerke. Lebenszyklusanalysen ausgewählter zukünftiger stromerzeugungstechniken (in German). University of Stuttgart, Institut für Energiewirtschaft und Rationelle Energieanwendung, Stuttgart, pp 255–341 Burnham A, Han J, Clark CE, Wang M, Dunn JB, Palou-Rivera I (2012) Life-cycle greenhouse Gas emissions of shale Gas, natural Gas, coal, and petroleum. Environ Sci Technol 46:619–627 Cicerone RJ, Oremland RS (1988) Biogeochemical aspects of atmospheric methane. Glob Biogeochem Cycles 2: 299–327 Fearnside PM (2002) Why a 100-year time horizon should be used for global warming mitigation calculations. Mitig Adapt Strateg Glob Chang 7:19–30 Hauck M, Steinmann ZJN, Laurenzi IJ, Karuppiah R, Huijbregts MAJ (2014) How to quantify uncertainty and variability in life cycle assessment: the case of greenhouse gas emissions of gas power generation in the US. Environ Res Lett 9:074005. doi:10.1088/1748-9326/9/7/074005 Hayhoe K, Kheshgi HS, Jain AK, Wuebbles DJ (2002) Substitution of natural gas for coal: climatic effects of utility sector emissions. Clim Chang 54:107–139 Heath GA, O’Donoughue P, Arent DJ, Bazilian M (2014) Harmonization of initial estimates of shale gas life cycle greenhouse gas emissions for electric power generation. Proc Natl Acad Sci U S A 111:E3167–E3176 Howarth RW (2014) A bridge to nowhere: methane emissions and the greenhouse gas footprint of natural gas. Energy Sci Eng 2:47–60 Howarth RW, Santoro R, Ingraffea A (2011) Methane and the greenhouse-gas footprint of natural gas from shale formations. Clim Chang 106:679–690

178

Climatic Change (2015) 133:169–178

Jaramillo P, Griffin WM, Matthews HS (2007) Comparative life-cycle air emissions of coal, domestic natural gas, LNG, and SNG for electricity generation. Environ Sci Technol 41:6290–6296 Jiang M, Griffin WM, Hendrickson C, Jaramillo P, VanBriesen J, Venkatesh A (2011) Life cycle greenhouse gas emissions of Marcellus shale gas. Environ Res Lett 6:034014. doi:10.1088/1748-9326/6/3/034014 May JR, Brennan DJ (2003) Life cycle assessment of Australian fossil energy options. Process Saf Environ 81: 317–330 May JR, Brennan DJ (2004) Life cycle assessment of Australian fossil energy options (vol 81, pg 317, 2003). Process Saf Environ 82:81–81 Moore CW, Zielinska B, Petron G, Jackson RB (2014) Air impacts of increased natural Gas acquisition, processing, and use: a critical review. Environ Sci Technol 48:8349–8359 Myhre G, Shindell D, Bréon F-M, Collins W, Fuglestvedt J, Huang J, Koch D, Lamarque J-F, Lee D, Mendoza B, Nakajima T, Robock A, Stephens G, Takemura T, Zhang H D2013] Anthropogenic and natural radiative forcing. In: Stocker TF, Qin D, Plattner G-K, Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V, Midgley PM Deds] Climate change 2013: the physical science basis. Contribution of working group I to the fifth assessment report of the intergovernmental panel on climate change. Cambridge University Press, Cambridge, pp 659–740, http://www.climatechange2013.org/report/full-report/ Daccessed 10/20/2014] O’Donoughue PR, Heath GA, Dolan SL, Vorum M (2014) Life cycle greenhouse Gas emissions of electricity generated from conventionally produced natural gas systematic review and harmonization. J Ind Ecol 18: 125–144 Schneising O, Burrows JP, Dickerson RR, Buchwitz M, Reuter M, Bovensmann H (2014) Remote sensing of fugitive methane emissions from oil and gas production in North American tight geologic formations. Earth’s Future 2:548–558 Showstack R (2013) Shale gas development requires bipartisan path forward, U.S. Senator Wyden urges. Eos T Am Geophys Union 94:278–279 Steinmann ZJN, Hauck M, Karuppiah R, Laurenzi IJ, Huijbregts MAJ (2014) A methodology for separating uncertainty and variability in the life cycle greenhouse gas emissions of coal-fueled power generation in the USA. Int J Life Cycle Assess 19:1146–1155 Tollefson J (2013) Methane leaks erode green credentials of natural gas. Nature 493:12–12 Turconi R, Boldrin A, Astrup T (2013) Life cycle assessment (LCA) of electricity generation technologies: overview, comparability and limitations. Renew Sust Energ Rev 28:555–565 U.S. Energy Information Administration D2014] How much coal, natural gas, or petroleum is used to generate a kilowatthour of electricity? U.S. Energy Information Administration, Washington, http://www.eia.gov/tools/ faqs/faq.cfm?id=667&t=2 Daccessed 11/20/2014] Victor DG (1992) Leaking methane from natural-gas vehicles - implications for transportation policy in the greenhouse Era. Clim Chang 20:113–141 Voorspools KR, Brouwers EA, D’haeseleer WD (2000) Energy content and indirect greenhouse gas emissions embedded in ‘emission-free’ power plants: results for the low countries. Appl Energy 67:307–330 Wigley TML (2011) Coal to gas: the influence of methane leakage. Clim Chang 108:601–608 Zhang XC, Myhrvold NP, Caldeira K (2014) Key factors for assessing climate benefits of natural gas versus coal electricity generation. Environ Res Lett 9:114022. doi:10.1088/1748-9326/9/11/114022