Air Emissions Damages from Municipal Drinking ... - ACS Publications

Aug 23, 2017 - Air Emissions Damages from Municipal Drinking Water Treatment Under Current and Proposed Regulatory Standards. Daniel B. Gingerich† a...
0 downloads 16 Views 1MB Size
Subscriber access provided by University of Victoria Libraries

Policy Analysis

Air Emissions Damages from Municipal Drinking Water Treatment Under Current and Proposed Regulatory Standards Daniel B Gingerich, and Meagan S Mauter Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b03461 • Publication Date (Web): 23 Aug 2017 Downloaded from http://pubs.acs.org on August 25, 2017

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 free 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 accessible to all readers and 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.

Environmental Science & Technology 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 25

Environmental Science & Technology

1 1

Air Emissions Damages from Municipal Drinking Water Treatment Under Current and

2

Proposed Regulatory Standards

3 4 5 6 7 8 9 10 11 12

Daniel B. Gingerich1 and Meagan S. Mauter1,2,* 1

Department of Engineering and Public Policy, Carnegie Mellon University, Pittsburgh, PA 15213 2 Department of Civil and Environmental Engineering, Carnegie Mellon University, Pittsburgh, PA 15213 *Contact Author: [email protected] Abstract Water treatment processes present inter-sectoral and cross-media risk trade-offs that are

13

not presently considered in Safe Drinking Water Act regulatory analyses. This paper develops a

14

method for assessing the air emission implications of common municipal water treatment

15

processes used to comply with recently promulgated and proposed regulatory standards,

16

including arsenic, lead and copper, disinfection byproducts, chromium (VI), strontium, and

17

PFOA/PFOS. Life-cycle models of electricity and chemical consumption for individual drinking

18

water unit processes are used to estimate embedded NOx, SO2, PM2.5, and CO2 emissions on a

19

cubic meter basis. We estimate air emission damages from currently installed treatment

20

processes at US drinking water facilities to be on the order of $500 million USD annually. Fully

21

complying with six promulgated and proposed rules would increase baseline air emission

22

damages by approximately 50%, with three-quarters of these damages originating from chemical

23

manufacturing. Despite the magnitude of these air emission damages, the net benefit of currently

24

implemented rules remains positive. For some proposed rules, however, the promise of net

25

benefits remains contingent on technology choice.

26 27

ACS Paragon Plus Environment

Environmental Science & Technology

Page 2 of 25

2 28 29

1.0 Introduction The Safe Drinking Water Act (SDWA) provides regulatory authority to the US federal

30

government to create enforceable standards for drinking water quality and safety.1 These

31

standards have enabled extraordinary gains in public health,2 and retrospective analyses of

32

historic drinking water quality regulations indicate that the benefits have far exceeded any

33

compliance costs.2, 3 Fully assessing these benefits and costs for emerging drinking water

34

contaminants is more challenging, particularly when contaminant removal processes impose

35

systemic risk tradeoffs.4

36

The most notable example of these risk trade-offs in drinking water regulatory analyses is

37

for disinfection and disinfection byproducts. In establishing the Long-Term Enhanced Surface

38

Water Treatment Rules,5 the EPA set regulatory standards to balance the competing risks of

39

acute illness from microbial contamination with the risks of cancer from disinfection

40

byproducts.6 There is also a growing body of research on inter-sectoral risk trade-offs involving

41

drinking water, including the use of methyl tert-butyl ether (MTBE) as a gasoline additive,7

42

bromide for mercury control at power plants,8, 9 and atrazine as a pesticide.10 But the

43

incorporation of inter-sectoral and inter-media (e.g. water and air or water and soil) trade-offs

44

into regulatory analyses is often limited by the “bounded oversight” of the regulators responsible

45

for designing drinking water regulations.7, 11 Researchers have argued that benefit-cost analysis

46

can address bounded oversight to express the benefits and damages to a variety of media.12, 13

47

Even with benefit-cost analysis, however, an analyst must make a deliberate choice to include

48

cross-media impacts in the scope.

49 50

Drinking water regulatory analyses have also neglected to consider trade-offs that occur over the life-cycle of the water treatment system, most notably the operation phase. Drinking

ACS Paragon Plus Environment

Page 3 of 25

Environmental Science & Technology

3 51

water systems comply with SDWA rules by installing separation processes that rely on

52

electricity and chemicals.14, 15 Life-cycle assessments (LCAs) of drinking water systems have

53

demonstrated that the majority of impacts associated with drinking water treatment occur as a

54

result of electricity and chemical consumption during plant operation.15-18 These impacts include

55

emissions of greenhouse gasses (GHGs) and the criteria air pollutants (CAPs) NOx, SO2, and

56

particulate matter from electricity generation and chemical manufacturing. These indirect

57

emissions of GHGs and CAPs produced over the life cycle of water treatment impose human

58

health, environmental, and climate (HEC) damages.19-23

59

While there are a multitude of drinking water treatment LCAs that quantify the energy

60

intensity or GHG and CAP emissions from water treatment,15-17, 24-29 we are unaware of any that

61

quantify the human health damages associated with these emissions. Furthermore, the vast

62

majority of these LCAs were performed for a single treatment train, rather than for the diverse

63

set of processes currently installed at drinking water treatment facilities around the US. Accurate

64

estimates of the air emission damages from US water treatment plants under current and future

65

regulatory scenarios is further confounded by the high regional variability in the emissions

66

factors associated with chemical and electricity process inputs. In the absence of quantitative

67

methods for estimating air emission damages from existing and emerging water treatment

68

processes, rulemaking activities have been unable to consider competing risks from diminished

69

air or water quality.

70

This work fills this gap by developing and applying a quantitative method for evaluating

71

the life-cycle air emission damages from drinking water treatment. We build life-cycle models

72

of electricity and chemical consumption for drinking water unit processes, estimate the

73

embedded smokestack emissions of NOx, SO2, PM2.5, and CO2, and provide spatially resolved

ACS Paragon Plus Environment

Environmental Science & Technology

Page 4 of 25

4 74

estimates of associated damages to human health and the environment. We apply this

75

methodology to estimate the air emission damages from installed water treatment processes at all

76

US drinking water facilities. Finally, we evaluate the air-water risk tradeoffs for six proposed

77

and promulgated drinking water regulations.

78

2.0 Methods

79

2.1 Estimating Air Emission Damages from Drinking Water Treatment

80

To calculate the air emission damages associated with drinking water treatment processes

81

we combine treatment level-facility data with state-level data on electricity consumption,

82

chemical manufacturing, and air emission damages (Figure 1).

83 84

Figure 1. Method for calculating the drinking water air emission damages and net benefits or

85

net costs for the regulations included in this analysis.

86

For this analysis, we focus on six different contaminant/regulations and compliance

87

technologies: Arsenic,30 Lead and Copper,31 Disinfectant By-Products (DBPs),5 Hexavalent

88

Chromium,32 Strontium,33 and Perfluorooctanoic Acid (PFOA) and Perfluorooctyl Sulfonate

ACS Paragon Plus Environment

Page 5 of 25

Environmental Science & Technology

5 89

(PFOS)34 (Table 1, Supporting Information Section 1.0). Three of these rules (Arsenic, Lead and

90

Copper, and DBP) are finalized. The other three rules, Hexavalent Chromium, Strontium, and

91

PFOA/PFOS, are at various stages in the regulatory process. These rules will require the

92

installation of new treatment processes at non-compliant facilities. Analysis on the air emission

93

impacts of regulations considers only the effects of installing new processes at non-compliant

94

plants to meet proposed standards. If a facility already has a compliance technology installed

95

and is compliant with the standard, we do not consider it in evaluating the air emissions

96

associated with complying with that rule.

97 98

Table 1. Rules and Compliance Technologies Rule

Compliance Technologies

Arsenic Iron Co-Precipitation Lead and Copper Corrosion Inhibitor DBP

GAC Adsorption

Chromium (VI) Strontium PFOS/PFOA

Reverse Osmosis Lime Soda Ash Softening Reverse Osmosis / GAC Adsorption

Standard 10 ug/L Lead 15 ug/L Copper 1.3 mg/L TTHM 80 ug/L HAA5 60 ug/L 10 ug/L 1.5 mg/L Total 70 ng/L

Number of Non-Compliant Systems 1,086 2,757 1,697 25 113 27,000

99 100

Drinking Water System Data

101

We use data from the Safe Drinking Water Information System (SDWIS),35 Unregulated

102

Contaminant Monitoring Rule (UCMR),36 and Information Collecting Rule (ICR) 37 (SI Section

103

2.0) to identify the (1) state, (2) installed treatment technologies (SI Section 3.0), and (3) volume

104

of water. In total, 47,500 drinking water systems serving 288 million people were analyzed.

105 106

We use the location of each facility reported in SDWIS to identify the appropriate statelevel marginal emissions factor38 for generating electricity used during water treatment. SDWIS

ACS Paragon Plus Environment

Environmental Science & Technology

Page 6 of 25

6 107

also identifies currently installed treatment processes that we use in developing a model of

108

baseline emissions from drinking water treatment. SDWIS does not report water production, and

109

so we estimate total annual water production for system i, Vi [m3/yr], as the product of SDWIS

110

reported population served, Popserved [person], and the state-level average per capita water

111

consumption per year, Apc [m3/person·yr],39 as shown in Equation 1. 𝑉𝑖 = 𝐴𝑃𝐶 𝑃𝑜𝑝𝑖,𝑠𝑒𝑟𝑣𝑒𝑑 (1)

112 113

We determine facilities that will need to install additional treatment technologies to

114

comply with the six water quality rules (Table 1) by comparing the MCL for each rule to water

115

quality data reported in the UCMR and the ICR. If water quality data is not available for a

116

facility, we apply state-averaged water quality data for that facility. We assume that any

117

compliance technologies that are installed are operated to in order to achieve compliance with

118

the regulation.

119

Life-Cycle Inventory of Air Emissions from Drinking Water Systems

120

For each facility in our dataset, we use SDWIS35 to inventory the water treatment

121

technologies currently installed in the “baseline” treatment train. For non-compliant systems, we

122

also inventory technologies necessary to bring a “regulated” treatment train into compliance with

123

each of the six pending or proposed water quality rules (SI Section 1.0). Average electricity27, 40

124

and chemical inputs15, 27, 41 for each of these treatment processes is obtained from the published

125

literature and reported in SI Sections 2.0 and 3.0 and Tables S1-S3. Finally, we estimate the air

126

emissions embedded in these electricity and chemical inputs in the year 2014.

127

𝐸𝑙𝑒𝑐 The mass of annual air emissions associated with electricity consumption, 𝑀ℎ,𝑖,𝑗 [g/yr] is

128

calculated for each pollutant j from drinking water facility i under scenario h of either the

129

baseline scenario or new regulation scenario using Equation 2. Emissions are the product of the

ACS Paragon Plus Environment

Page 7 of 25

Environmental Science & Technology

7 130

volume of water produced at facility i, Vi [m3/yr], the total electricity consumption of the unit

131

processes, g, that are installed in the baseline or newly-regulated scenario, h, Eg,hW [kWh/m3], and

132

the marginal emissions factors per kWh for air pollutant j in state l, emf,j,l [g/kWh].38, 42 The

133

emissions factors for CO2, NOx, and SO2 are marginal emission factors, as only a small

134

percentage of US electricity generation is consumed by water treatment processes.43, 44 Marginal

135

emission factors are not available for PM2.5 and so we use state-level average emission factors

136

from the EPA Emission Inventories.42 Electricity is assumed to be generated in the state where

137

the drinking water facility is located. 𝐸𝑙𝑒𝑐 𝑊 𝑀ℎ,𝑖,𝑗 = 𝑉𝑖 𝐸𝑔,ℎ 𝑒𝑚𝑓,𝑗,𝑙 (2)

138 139 140

𝑉𝑙 ) 𝑙 𝑉𝑙

𝐶 𝐶 𝐷 𝐷 𝑀ℎ,𝑖,𝑗 = ∑𝑓 𝑉𝑖 𝑄𝑖,𝑓,𝑗 (𝑒𝑐𝑚,𝑓,ℎ + ∑𝑑 𝐸𝑑,𝑓 𝑒𝑑,ℎ + 𝐸𝑓𝑊 ∑𝑙 𝑒𝑚𝑓,𝑙,ℎ ∑

(3)

Similarly, the mass of annual air emissions associated with chemical consumption by

141

𝑐 each water treatment facility, 𝑀ℎ,𝑖,𝑗 in [g/yr], is calculated using Equation 3. Air emissions from

142

chemical manufacturing originate from three sources. The first source is emissions released

143

𝐶 directly during the manufacturing of chemical f, 𝑒𝑐𝑚,𝑓,ℎ in [g-emissions/g-chemicals]. The

144

second source of emissions is from the combustion of thermal energy to drive chemical

145

𝐷 manufacturing, which is the product of the energy consumed from fuel source d, 𝐸𝑑,𝑓 in [MJ/g-

146

𝐷 chemical], and emission per unit of thermal energy, 𝑒𝑑,ℎ in [g-pollutant/MJ]. The final source is

147

emissions from electricity consumed in chemical manufacturing which is the weighted average

148

of the grid emissions, ∑𝑙 𝑒𝑚𝑓,𝑙,ℎ ∑

149

to produce a unit of chemicals, 𝐸𝑓𝐶 in [kWh/g-chemical]. This weighting is done according to a

150

state’s chemical manufacturing sector size and we assume chemical manufacturing follows the

151

nationwide distribution in chemical production.45 The sum of these emissions per unit of

𝑉𝑙 𝑙 𝑉𝑙

in [g-emissions/kWh] multiplied by the electricity consumed

ACS Paragon Plus Environment

Environmental Science & Technology

Page 8 of 25

8 152

chemical is then multiplied by the water required to be treated at facility i, Vi in [m3/yr], and the

153

chemical dosage, Qi,f [g-chemical/m3]. Additional detail about the methods used to calculate

154

chemical emissions can be found in our previous work.46 We also perform sensitivity analyses

155

(SI Section 4.0) on assumptions about the geographic distribution of chemical manufacturing and

156

binding air pollution regulations.

157

Damages from Drinking Water Systems

158

To calculate the associated air emission damages from drinking water systems at the

159

facility-level, we use AP2,21 EASIUR,47 and the social cost of carbon23 to estimate the damages

160

per ton of additional pollutant j occurring in state l, dj,l in [$/g]. In the main manuscript we report

161

health and environmental damages from CAPs using AP2. Note that AP2 includes both human

162

health and environmental damages of CAPs, but the human health damages are several orders of

163

magnitude larger than the environmental damages due to reduced agricultural crop and timber

164

sales. The SI reports results using the EASIUR model, which exclusively models human health

165

damages. Both AP2 and EASIUR account for atmospheric transformation of pollutants,

166

including damages associated with the formation of secondary aerosols. The disagreement

167

between AP2 and EASIUR health damage estimates stem from differences in underlying

168

atmospheric transport models and atmospheric transformations. Damages from these models are

169

calculated in 2014 USD with a value of a statistical life of $8.7 million.48

170

Air emission damages associated with electricity consumption are calculated at the

171

𝐸𝑙𝑒𝑐 facility-level, 𝐷𝑖,ℎ in [$/yr], by multiplying the amount of electricity consumed at facility i by

172

𝑑𝑗,𝑙 (Equation 4). Air emission damages associated with the production of chemicals consumed

173

in water treatment are calculated for each state l as the product of 𝑑𝑗,𝑙 and estimated mass of air

174

emissions from chemical manufacturing in that state (Equation 5).

ACS Paragon Plus Environment

Page 9 of 25

Environmental Science & Technology

9 175

𝐸𝑙𝑒𝑐 𝐸𝑙𝑒𝑐 𝐷ℎ,𝑖 = ∑𝑗 𝑑𝑗,𝑙 𝑀ℎ,𝑖,𝑗

176

𝐶 𝐷ℎ,𝑙 = ∑𝑗 𝑑𝑗,𝑙 𝑀𝐶ℎ,𝑖,𝑗 ∑𝑙=𝐿 𝑉

𝑉

(4) (5)

𝑙 𝑙

177

The total damages nationwide for treatment train h, Dh in [$/yr], is given by Equation 6.

178

𝐸𝑙𝑒𝑐 𝐶 𝐷ℎ = ∑𝑖 𝐷𝑖,ℎ + ∑𝑙 𝐷ℎ,𝑙

179

2.2 Benefit-Cost Analysis of Drinking Water Regulations

(6)

180

We compare the damages calculated using Equation 6 to the benefits reported in the

181

regulatory documentation for Arsenic,30 the Lead and Copper Rules,31 and the Disinfection

182

Byproduct Rules.5 Costs (C) considered in these regulatory analyses include technology

183

installation and regulatory oversight. The costs do not include damages from embedded air

184

emissions, as the EPA did not calculate them during the BCA process. Benefits (B) include

185

health benefits and other environmental benefits. The net benefits (N) of a rule are calculated

186

using Equation 7.

187

𝑁 = 𝐵 − 𝐶 − 𝐷ℎ

188

For the three rules that are still in development, we use published estimates of

(7)

189

compliance technology costs to calculate the health benefits from improved water quality

190

necessary for the rule to provide a net benefit to society.49

191

2.3 Evaluating Regulatory Compliance Options

192

The EPA recommends two compliance technology options for PFOA/PFOS: reverse

193

osmosis or granular activated carbon. In the main analysis laid out in Section 2.1, we model

194

plants installing granular activated carbon adsorption processes to comply with the PFOA/PFOS

195

health advisory. We repeat the calculations in Section 2.1 using reverse osmosis as the

196

compliance technology to explore the difference in damages resulting from selecting an

197

alternative technology.

ACS Paragon Plus Environment

Environmental Science & Technology

Page 10 of 25

10 198 199

2.4 Forecasting Emissions from Electricity Consumption The marginal emission factors from electricity generation are expected to decrease as

200

coal fired power plants are replaced by natural gas plants and renewable energy sources. This

201

transition may, in turn, reduce the embedded air emission damages from drinking water

202

treatment. We use EIA Annual Energy Outlook 2017 Reference case with the Clean Power

203

Plan50 forecasts of electricity generation, Gk,m [kWh], to calculate national average shares of

204

electricity from fuel m (coal, natural gas, diesel, and zero operating emission energy sources) in

205

year k between 2015-2050. This reference case assumes that the US upholds its CO2 reduction

206

commitments as outlined in the Paris Agreement. We also include two additional cases, the EIA

207

AEO Reference case without the Clean Power Plan and the EIA AEO high oil and gas resource

208

and technology scenario, as alternative future energy portfolios. We then calculate national

209

average emission factor forecasts, eaf,m [g/kWh], over this interval for SO2, NOx, PM2.5, and CO2

210

using average emission factors from fuel source coal, natural gas, and diesel.51 We then multiply

211

national average emission factor forecasts by the amount of electricity consumed to drive

212

baseline and regulatory treatment trains and average damages per mass of air pollution values in

213

order to calculate expected damages through 2050 (Equation 8).

214

𝑊 𝐷𝑘,ℎ = ∑𝑗(𝑑𝑗 ∑𝑖 𝑉𝑖 ∑𝑔( 𝐸𝑔,ℎ ∗

𝐺𝑘,𝑚 𝑒𝑎𝑓,𝑚 ∑𝑙 𝐺𝑘,𝑚

))

(8)

215

Because we use marginal emission factors for the main analysis and average emission factors for

216

forecasting, Equation 8 overestimates the damages. We correct for this overestimation, 𝐷̃ 𝑘,ℎ

217

[$/yr], by multiplying the ratio of calculated results for 2014 from Equation 5, Dh [$/yr],

218

forecasted results for 2014 from Equation 8, Dk,h [$/yr] (Equation 9).

219

𝐷̃ 𝑘,ℎ =

𝐷ℎ 𝐷ℎ,𝑘=2014

𝐷𝑘,ℎ (9)

220

ACS Paragon Plus Environment

Page 11 of 25

Environmental Science & Technology

11 221

3.0 Results and Discussion

222

3.1 Air Emission Damages from Operating Baseline and Compliant Treatment Trains

223

We estimate that baseline operation of water treatment plants consumed 2800 GWh of

224

electricity, consumed 1.5 million tons of chemicals, and generated $493 million dollars in air

225

emission damages in 2014 (expressed in 2014 dollars and using a value of a statistical life of

226

$8.69 million) (Figure 2A, 2C, and 2E). Emissions from chemical manufacturing contributes

227

73% of the damages in the baseline treatment train. For comparison, air emission damages

228

calculated using EASIUR amount to $533 million ($135 million from electricity generation and

229

$398 million from chemical manufacturing) as shown in SI Section 5.2, Figure S1, and Table

230

S10 .The damages associated with operating baseline water treatment are several orders of

231

magnitude lower than the benefits of avoided illness and death from untreated drinking water.2

232

These air emission damages are also several orders of magnitude smaller than the approximately

233

100,000 deaths from air emissions that occur annually in the US.52

234

ACS Paragon Plus Environment

Environmental Science & Technology

Page 12 of 25

12

235 236 237

Figure 2. Per capita air emission damages associated with operating drinking water treatment

238

processes under current and future regulatory scenarios. State-level damages embedded in

239

electricity consumption by water treatment processes for the (A) baseline treatment train and the

240

(B) regulated treatment train assuming compliance with all promulgated and proposed SDWA

241

regulations. State-level damages embedded in chemical consumption by water treatment

242

processes for the (C) baseline treatment train and the (D) regulated treatment train. Total air

ACS Paragon Plus Environment

Page 13 of 25

Environmental Science & Technology

13 243

emissions damages for the (E) baseline treatment train and the (F) regulated treatment train.

244

Emissions and damages are tabulated in SI Tables S8 and S9 and SI Section 5.1. Results

245

calculated using EASIUR to price CAP damages can be found in SI Section 5.2, Figure S1, and

246

Tables S10 and S11.

247 248

Achieving compliance with the six drinking water regulations increases air emission

249

damages by $242 million (in 2014$) annually using the compliance technologies listed in Table

250

1 and GAC for PFOA/PFOS removal (Figure 2, B, and D). This represents an increase of 49%

251

from the emissions associated with the baseline treatment train. Of this $242 million in

252

damages, 76% are damages from chemical manufacturing. Over 84% of these damages, $205

253

million (in 2014$), stem from compliance with the PFOA/PFOS Health Advisory at 27,000

254

drinking water facilities nationwide. EASIUR modeled air emission damages from regulatory

255

compliance estimate that total damages equal $796 million, or an additional $263 million beyond

256

baseline damage estimates using EASIUR ($41 million from electricity generation and $222

257

million from chemical manufacturing) as shown in SI Section 5.2, Figure S1, and Table S11.

258

Emissions and damages increases estimated for compliance with each of the six regulations are

259

broken down in SI Section 6.0, Tables S12-S17, and Figures S2-S7.

260 261 262 263 264 265

ACS Paragon Plus Environment

Environmental Science & Technology

Page 14 of 25

14 266

3.2 Benefit-Cost Analysis of Drinking Water Regulations

267

Table 2. Benefit-Cost analysis for the six drinking water regulations after accounting for

268

estimated air emission damages. Benefits

Contaminant Finalized Rules Arsenic30 DBP

5

Lead & Copper31

Costs/Damages Compliance Electricity Costs Damages 3 ($/m in 2014$)

Net Chemical Damages

0.14

0.16

0.0005

0.000002

-0.023

3.13

1.08

0.0013

0.0080

2.04

8.36

1.35

0.0006

0.0003

7.01

Rules Under Consideration Chromium (VI)

≥0.25*

0.2049

0.024

0.0014

≥0

Strontium

≥0.43*

0.4053

0.0008

0.042

≥0

PFOA/PFOS**

≥0.61*

0.6049

0.0015

0.0078

≥0

*Rules that are under consideration do not have published estimates for their benefits and compliance costs. **Using Granular Activated Carbon as the compliance technology.

269 270

For the three finalized rules (Arsenic, DBPs, and Lead and Copper), the signs on the

271

benefit-cost analysis do not change with the inclusion of air emission damages (Table 2). The

272

DBP and Lead and Copper rules had large benefits to society, and accounting for air emissions

273

damages does not change this conclusion. The arsenic rule was a net cost to society without

274

incorporating air emissions from drinking water; including air emissions in the benefit-cost

275

analysis only increases the cost.

276

For the three rules that are not yet finalized, Table 2 presents the minimum benefits

277

required given estimates of air emission damages from this analysis and literature-based

278

compliance technology costs.49, 53 On a per cubic meter basis, air emission damages contribute 2-

279

11% of the total costs to society for these rules. Air emissions are therefore a small, but non-

ACS Paragon Plus Environment

Page 15 of 25

Environmental Science & Technology

15 280

negligible contributor to these overall costs. The benefits required for these rules to impart net

281

social benefits range from $0.25/m3 to $0.61/m3 (in 2014$).

282

There are several unquantifiable co-benefits of adopting the six drinking water

283

regulations that are not accounted for in this analysis. These include the reduction of unregulated

284

contaminant concentrations associated with the advanced treatment technologies that are

285

installed to comply with these rules. There is also the possibility that, in the process of

286

complying with new water quality regulations, monitoring and enforcement of existing water

287

quality regulations is improved.

288

3.3 Evaluating Regulatory Compliance Options

289 290

Figure 3. Air emission damages from reverse osmosis and granular activated carbon treatment

291

for PFOA/PFOS removal at non-compliant drinking water facilities across the United States

292

using state-level marginal emissions factors for NOx, SO2, and CO2 and state-level average

293

emissions factors for PM2.5. The error bars show the uncertainty in the location and presence of

294

binding regulations affecting emissions from the chemical manufacturing sector (SI Section 4.0)

295

and Tables S4-S7.

296 297

Quantifying the air emission damages of drinking water treatment processes can also assist in compliance technology selection. Of the two compliance options for PFOA/PFOS

ACS Paragon Plus Environment

Environmental Science & Technology

Page 16 of 25

16 298

removal, reverse osmosis is the highest damage alternative (Figure 3). Using reverse osmosis to

299

comply with the PFOA/PFOS standards imposes air emission damages of $1.2 billion annually

300

(in 2014$). The use of GAC to achieve compliance with PFOA/PFOS imposes air emission

301

damages of $205 million/yr (in 2014$), a sixth of the damages from reverse osmosis. The

302

primary source of air emission damages from RO is the electricity used to drive the process,

303

rather than from emissions embedded in chemical consumption. In contrast, granular activated

304

carbon leads to more damages from chemical manufacturing than reverse osmosis, with

305

subsequently more uncertainty.

306

Reverse osmosis and granular activated carbon have different removal efficiencies for

307

PFOA/PFOS, with RO producing higher quality water. This sets up trade-off between reducing

308

health risks from PFOA/PFOS in drinking water and minimizing the environmental and health

309

risks from criteria air pollutants and greenhouse gasses. To account for this additional $1.0

310

billion in air emission damages annually, reverse osmosis would need to save an additional 120

311

statistical lives through PFOA/PFOS concentration reduction. These additional statistical lives

312

could also be saved by the co-benefits of RO by producing higher quality water.

313

3.4 Future Air Emission Damages from Electricity Consumed in Drinking Water Treatment

314

ACS Paragon Plus Environment

Page 17 of 25

Environmental Science & Technology

17 315

Figure 4. Projected air emission damages from electricity generated to treat US drinking water

316

with the baseline treatment train and the additional emissions that result from compliance with

317

current and proposed drinking water standards. The grid mix follows the AEO 2017 Reference

318

case with the Clean Power Plan. Forecasts for the AEO 2017 Reference case without the Clean

319

Power Plan and the High oil and gas resource and technology scenario differ by less than 7%

320

(details reported in SI Section 7.0). N.B. That this figure does not include damages resulting

321

from the electricity consumed during chemical manufacturing.

322

As the grid evolves to rely less on fossil fuel sources, damages from air emissions are

323

expected to decrease (Figure 4). By 2050, annual damages associated with emissions from

324

electricity generation that result from compliance with these six standards decreases by $19

325

million per year from the 2015 level of $40 million. This reduction in damages from electricity

326

will lower the required benefits presented in Table 2 to achieve net benefits from these drinking

327

water rules. Projected air emission damages for two alternative electricity grid scenarios,

328

including scenarios without the Clean Power Plan and with high natural gas use differ by less

329

than 7% (SI Section 7.0 and Figure S8). The robustness of these results to changes in the

330

electricity mix reflect the relatively small fraction of damages that originate from electricity

331

consumption.

332

4.0 Implications

333

Stronger epidemiological evidence of health impacts from contaminated water, enhanced

334

public awareness of water quality issues, and improved analytical techniques for detecting

335

aqueous contaminants are each drivers for stricter water quality regulations.1, 54 While the

336

counter-argument to installing the advanced treatment technologies necessary for compliance has

337

historically been cost, we propose that the very real risks of air emission damages associated

ACS Paragon Plus Environment

Environmental Science & Technology

Page 18 of 25

18 338

with increased electricity or chemical consumption of water treatment processes should also be

339

considered. These incremental air emission damages may outweigh the incremental benefits of

340

cleaner water enabled by stricter drinking water regulatory limits, but establishing this balance of

341

cost and benefits is difficult to establish a priori.

342

Given that air emission damages are difficult to establish a priori and useful in informing

343

regulatory concentration limits, we conclude that there are merits to regulatory assessment

344

processes explicitly considers the life-cycle tradeoffs between improved water quality and

345

reduced air emissions. This process may inform the selection of concentration limits that

346

maximize human health benefits, technologies for regulatory compliance, or investment

347

priorities in the water, chemical, and electricity generation space. While performing this analysis

348

will increase EPA workload, screening approaches such as Economic Input/Output LCA may

349

allow the EPA to assess the need for the more robust analysis described in this manuscript.55

350

Adapting the regulatory assessment process to include life-cycle air emissions may also

351

broaden the scope of environmental justice analyses. This analysis demonstrates that spatial

352

distribution of the damages, and not simply the water quality benefits, varies across the United

353

States and should be considered in evaluating regulations. Areas with high-levels of coal-fired

354

electricity generation or chemical manufacturing may experience damages from air emissions in

355

excess of benefits from improved drinking water quality. Performing spatially-resolved air

356

emissions analyses allows the EPA to quantify these trade-offs as part of their standard

357

environmental justice analyses.

358

Finally, this analysis supports the ongoing transition from single-media focused

359

regulations (e.g. drinking water) towards more holistic analyses that account for the

360

countervailing risks analyzed in this paper. Previous examples of the EPA moving to holistic

ACS Paragon Plus Environment

Page 19 of 25

Environmental Science & Technology

19 361

regulatory activity includes the cluster rules for the pulp and paper industry56 and Executive

362

Order (EO) 13211 mandating that all regulations, including drinking water regulations, undergo

363

an energy inventory if they are deemed “economically significant.”43 Unfortunately, it is not

364

standard practice for analyses performed under EO 13211 to assess emissions associated with

365

this electricity generation or incorporate the damages into benefit-cost analyses performed under

366

EO 12866. Even if damages from the emissions associated with electricity generation were

367

included in EPA BCAs for drinking water treatment, the emissions from chemical manufacturing

368

would be overlooked. As chemical manufacturing damages contribute roughly three times as

369

much as electricity generation, their inclusion is important in ensuring a comprehensive analysis

370

of these trade-offs.

371

In short, more rapid adoption of environmental regulatory frameworks that account for

372

inter-sectoral and cross-media risk trade-offs should be a priority at both the national and state

373

levels.

374

Acknowledgements

375

This work was supported by the National Science Foundation under award number CBET-

376

1554117.

377

Supporting Information

378

The supporting information contains descriptions of 1) the six drinking water standards with

379

compliance modeled in this analysis; 2) data sources; 3) drinking water treatment unit process

380

description and inputs; 4) chemical manufacturing location sensitivity analyses; 5) emissions and

381

damages from drinking water treatment using AP2 and EASIUR; 6) emissions and damages

382

from the individual drinking water standards; and 7) air emission damages using alternative grid

383

forecasts.

ACS Paragon Plus Environment

Environmental Science & Technology

Page 20 of 25

20 384

Nomenclature

385

A:

Per capita annual water consumption [m3/person·yr]

386

B:

Benefits from a drinking water regulation [$/yr]

387

C:

Compliance costs from a drinking water regulation [$/yr]

388

D:

Air emission damages for drinking water treatment [$/yr]

389

̃: 𝐷

Corrected damages forecast [$/yr]

390

d:

Damages per unit of NOx, SO2, PM2.5, and CO2 [$/g]

391

E:

Energy consumption [kWh/m3], [kWh/g-chemical], or [MJ/g-chemical]

392

e:

Emissions factor per unit of energy [g/kWh] or [g/MJ]

393

G:

Annual electricity generation [kWh]

394

M:

Annual mass of NOx, SO2, PM2.5, or CO2 emissions [g/yr]

395

N:

Net benefits or costs from a drinking water regulation [$/yr]

396

Pop:

Population served by a facility [people]

397

Q:

Chemical dosage [g-chemical/m3]

398

V:

Annual value of products from the chemical manufacturing sector [$]

399

V:

Annual water production [m3/yr]

400

W:

Electricity consumption for water treatment process [kWh/m3]

401

Subscripts

402

af:

Average emissions factor

403

cm:

Chemical manufacturing

404

d:

Thermal fuel

405

f:

Chemical

406

g:

Unit process

ACS Paragon Plus Environment

Page 21 of 25

Environmental Science & Technology

21 407

h:

Treatment train (baseline or newly-regulated)

408

i:

Drinking water facility

409

j:

Air pollutant (NOx, SO2, PM2.5, or CO2)

410

k:

Year

411

l:

State

412

m:

Fuel source for electricity generation

413

\mf:

Marginal emissions factor

414

PC:

per capita

415

served:

Population served by a facility

416

Superscripts

417

C

:

From chemical manufacturing

418

D:

From thermal fuel combustion

419

Elec

420

W

421

References

422 423 424 425 426 427 428 429 430 431 432 433 434 435

1. Roberson, J. A., What's next after 40 years of drinking water regulations? Environ Sci Technol 2011, 45, 154-160. 2. Craun, G. F.; Craun, M. F.; Calderon, R. L.; Beach, M. J., Waterborne outbreaks reported in the United States. Journal of Water and Health 2006, 04, (Suppl 2), 19. 3. Gurian, P. L.; Tarr, J. A., The origin of federal drinking water quality standards. Proceedings of the Institution of Civil Engineers - Engineering History and Heritage 2011, 164, (1), 17-26. 4. Putnam, S. W.; Wiener, J. B., Seeking Safe Drinking Water. In Risk vs. Risk, Graham, J. D.; Wiener, J. B., Eds. Harvard University Press: Cambridge, Massachusetts, 1995. 5. U.S. Environmental Protection Agency Stage 1 and Stage 2 Disinfectants and Disinfection Byproducts Rules https://www.epa.gov/dwreginfo/stage-1-and-stage-2disinfectants-and-disinfection-byproducts-rules (March 25, 2017), 6. Richardson, S. D., Water Analysis: Emerging Contaminants and Current Issues. Analytical Chemistry 2009, 81, (12), 4645-4677.

:

:

From electricity generation Electricity consumption

ACS Paragon Plus Environment

Environmental Science & Technology

Page 22 of 25

22 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480

7. Stikkers, D., The Unintended Consequence of Reformulated Gasoline. In Improving Regulation: Cases in Environment, Health, and Safety, Fischbeck, P. S.; Farrow, R. S., Eds. Resources for the Future Washington, D.C., 2001. 8. Wang, Y.; Small, M. J.; VanBriesen, J. M., Assessing the Risk Associated wtih Increasing Bromied in Drinking Water Sources in the Monongahela River, Pennsylvania. Journal of Environmental Engineering 2017, 143, (3). 9. Good, K. D.; VanBriesen, J. M., Current and Potential Future Bromide Loads from CoalFired Power Plants in the Allegheny River Basin and Their Effects on Downstream Concentrations. Environmental Science & Technology 2016, 50, (17), 9078-88. 10. Tesfamichael, A. A.; Caplan, A. J.; Kaluarachchi, J. J., Risk-cost-benefit analysis of atrazine in drinking water from agricultural activities and policy implications. Water Resources Research 2005, 41, (5). 11. Wiener, J. B.; Graham, J. D., Resolving Risk Tradeoffs. In Risk vs. Risk: Tradeoffs in Protecting Health and the Environment, Graham, J. D.; Wiener, J. B., Eds. Harvard University Press: Cambridge, MA, 1995. 12. Arrow, K. J.; Cropper, M. L.; Eads, G. C.; Hahn, R. W.; Lave, L. B.; Noll, R. G.; Portney, P. R.; Russell, M.; Schmalensee, R.; Smith, V. K. S., Robert N., Is There a Role for Benefit-Cost Analysis in Environmental, Health, and Safety Regulation? Science 1996, 272, 221222. 13. Lave, L. B., Benefit-Cost Analysis: Do the Benefits Exceed the Costs? In Risks, Costs, and Lives Saved: Getting Better Results from Regulation, Hahn, R. W., Ed. The AEI Press: Washington, D.C., 1996; pp 104-134. 14. Santana, M. V.; Zhang, Q.; Mihelcic, J. R., Influence of water quality on the embodied energy of drinking water treatment. Environmental Science & Technology 2014, 48, (5), 308491. 15. Stokes, J. R.; Hovarth, A., Energy and Air Emission Effects of Water Supply Environmental Science & Technology 2009, 43, (2680-2687). 16. Stokes, J.; Horvath, A., Life-Cycle Assessment of Urban Water Provision: Tool and Case Study in California Journal of Infrastructure Systems 2011, 17, (1), 15-24. 17. Mo, W.; Wang, H.; Jacobs, J. M., Understanding the influence of climate change on the embodied energy of water supply. Water Research 2016, 95, 220-9. 18. Fang, A. J.; Newell, J. P.; Cousins, J. J., The energy and emissions footprint of water supply for Southern California. Environmental Research Letters 2015, 10, (11), 114002. 19. Muller, N. Z.; Mendelsohn, R., Efficient Pollution Regulation: Getting the Prices Right. American Economic Review 2009, 99, (5), 1714-1739. 20. Muller, N. Z.; Mendelsohn, R.; Nordhaus, W., Environmental Accounting for Pollution in the United States Economy. American Economic Review 2011, 101, (5), 1649-1675. 21. Muller, N. Z., Using index numbers for deflation in environmental accounting. Environment and Development Economics 2013, 19, (04), 466-486. 22. Heo, J.; Adams, P. J.; Gao, H. O., Public Health Costs of Primary PM2.5 and Inorganic PM2.5 Precursor Emissions in the United States. Environmental Science & Technology 2016, 50, (11), 6061-70. 23. Interagency Working Group on the Social Cost of Carbon, Technical Support Document: Technical Update of the Social Cost of Carbon for Regulatory Impact Analysis. In Washington, D.C., 2015.

ACS Paragon Plus Environment

Page 23 of 25

Environmental Science & Technology

23 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526

24. Stokes, J. R.; Hendrickson, T. P.; Horvath, A., Save water to save carbon and money: developing abatement costs for expanded greenhouse gas reduction portfolios. Environmental Science & Technology 2014, 48, (23), 13583-91. 25. Lam, K. L.; Kenway, S. J.; Lant, P. A., Energy use for water provision in cities. Journal of Cleaner Production 2016. 26. Qi, C.; Chang, N.-B., Integrated carbon footprint and cost evaluation of a drinking water infrastructure system for screening expansion alternatives. Journal of Cleaner Production 2013, 60, 170-181. 27. Bonton, A.; Bouchard, C.; Barbeau, B.; Jedrzejak, S., Comparative life cycle assessment of water treatment plants. Desalination 2012, 284, 42-54. 28. Capitanescu, F.; Ahmadi, A.; Benetto, E.; Marvuglia, A.; Tiruta-Barna, L., Some efficient approaches for multi-objective constrained optimization of computationally expensive black-box model problems. Computers & Chemical Engineering 2015, 82, 228-239. 29. Wakeel, M.; Chen, B.; Hayat, T.; Alsaedi, A.; Ahmad, B., Energy consumption for water use cycles in different countries: A review. Applied Energy 2016, 178, 868-885. 30. U.S. Environmental Protection Agency Chemical Contaminant Rules. https://www.epa.gov/dwreginfo/chemical-contaminant-rules (March 25), 31. U.S. Environmental Protection Agency Lead and Copper Rule. https://www.epa.gov/dwreginfo/lead-and-copper-rule (March 25), 32. U.S. Environmental Protection Agency Chromium in Drinking Water. https://www.epa.gov/dwstandardsregulations/chromium-drinking-water (March 25, 2017), 33. U.S. Environmental Protection Agency Regulatory Determination 3. https://www.epa.gov/ccl/regulatory-determination-3 (March 25, 2017), 34. U.S. Environmental Protection Agency Basic Information about Per-and Polyfluoroalkyl Substances (PFASs). https://www.epa.gov/pfas/basic-information-about-and-polyfluoroalkylsubstances-pfass (March 25, 2017), 35. U.S. Environmental Protection Agency Safe Drinking Water Information System (SDWIS). https://www3.epa.gov/enviro/facts/sdwis/search.html (March 25, 2017), 36. U.S. Environmental Protection Agency Occurrence Data for the Unregulated Contaminant Monitoring Rule. https://www.epa.gov/dwucmr/occurrence-data-unregulatedcontaminant-monitoring-rule#3 (March 25, 2017), 37. U.S. Environmental Protection Agency Information Collection Rule. https://archive.epa.gov/enviro/html/icr/web/html/index.html (March 25, 2017), 38. Siler-Evans, K.; Azevedo, I. L.; Morgan, M. G., Marginal emissions factors for the U.S. electricity system. Environmental Science & Technology 2012, 46, (9), 4742-8. 39. United States Geological Survey Water Use in the United States. (March 3, 2017), 40. Plappally, A. K.; Lienhard V, J. H., Energy requirements for water production, treatment, end use, reclamation, and disposal. Renewable and Sustainable Energy Reviews 2012, 16, (7), 4818-4848. 41. Crittenden, J. C.; Trussell, R. R.; Hand, D. W.; Howe, K. J.; Tchobanoglous, G., Water Treatment: Principles and Design. John Wiley & Sons, Inc.: Hoboken, New Jersey, 2012. 42. U.S. Environmental Protection Agency Emission Inventories. https://www3.epa.gov/ttn/chief/eiinformation.html (October 9, 2015), 43. Reiling, S. J.; Roberson, J. A.; Cromwell III, J. E., Drinking water regulations: Estimated cumulative energy use and costs. Journal American Water Works Association 2009, 101, (3), 4253.

ACS Paragon Plus Environment

Environmental Science & Technology

Page 24 of 25

24 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556

44. Liu, Y.; Hejazi, M.; Kyle, P.; Kim, S. H.; Davies, E.; Miralles, D. G.; Teuling, A. J.; He, Y.; Niyogi, D., Global and Regional Evaluation of Energy for Water. Environmental Science & Technology 2016, 50, (17), 9736-45. 45. United States Census Bureau Annual Survey of Manufacturers 2013. https://www.census.gov/manufacturing/asm/index.html (October 11, 2015), 46. Gingerich, D. B.; Sun, X. B., A. Patrick; Azevedo, I. M. L.; Mauter, M. S., Spatially resolved air-water emissions tradeoffs improve regulatory impact analyses for electricity generation. Proceedings of the National Academy of Science 2017, 1862–1867. 47. Heo, J.; Adams, P. J.; Gao, H. O., Reduced-form modeling of public health impacts of inorganic PM2.5 and precursor emissions. Atmospheric Environment 2016, 137, 80-89. 48. National Center for Environmental Economics, Guidelines for Preparing Economic Analyses. In Agency, U. S. E. P., Ed. Washington, D.C., 2014. 49. Plappally, A. K.; Lienhard, J. H., Costs for water supply, treatment, end-use and reclamation. Desalination and Water Treatment 2013, 51, (1-3), 200-232. 50. U.S. Energy Information Administration Annual Energy Outlook 2017 with Projections to 2050; U.S. Energy Information Administration: Washington, D.C., 2017. 51. National Renewable Energy Laboratory National Renewable Energy Laboratory - LifeCycle Inventory Database. . https://www.lcacommons.gov/nrel/process/show/50158 (October 3, 2015), 52. U.S. Burden of Disease Collaborators, The state of US health, 1990-2010: burden of diseases, injuries, and risk factors. Journal of the American Medical Association 2013, 310, (6), 591-608. 53. Bergman, R. A., Membrane softening versus lime softening in Florida. Desalination 1995, 103, (1), 11-24. 54. Roberson, J. A., The middle-aged Safe Drinking Water Act Journal - American Water Works Association 2014, 106, (8), 96-106. 55. Hendrickson, C. T. L., Lester B.; Matthews, H. S., Environmental Life Cycle Assessment of Goods and Services. Resources for the Future: Washington, D.C., 2006. 56. Gray, W. B. S., Ronald J, Multimedia Pollution Regulation and Environmental Performance: EPA’s Cluster Rule. In Resources for the Future: Washington, D.C., 2015.

557

ACS Paragon Plus Environment

Page 25 of 25

Environmental Science & Technology

TOC Image 80x47mm (300 x 300 DPI)

ACS Paragon Plus Environment