Sustainability Concepts in Decision-Making: Tools and Approaches for the US Environmental Protection Agency (2014)

4

Case Studies of Applications of Sustainability Tools and Approaches

INTRODUCTION

Activities undertaken by the US Environmental Protection Agency (EPA) are driven by congressional mandates, presidential directives, and voluntary or discretionary initiatives (such as research-grant programs and initiatives involving partnerships with other organizations) that stem from policy priorities. As described in Chapter 2, EPA’s activities include program development, development of internal and external guidance, strategic planning, research planning, budgetary decision-making, regulatory and standards development, enforcement, knowledge transfer, permitting, communication and education, and a wide variety of voluntary programs. To facilitate the consideration of sustainability concepts in relation to this broad array of activities, EPA has developed a report, Sustainability Analytics: Assessment Tools and Approaches (EPA 2013a), often referred to as the Analytics report, that presents various analytic tools and approaches.

This chapter illustrates the application of various tools and approaches listed in EPA’s report by using five case studies. Some case studies involve the use of only a few of the tools described in EPA’s report, and others involve the use of multiple tools that encompass the environmental, economic, and social dimensions in the sustainability assessment and management process. The committee selected case studies whose histories are well known to the committee members to consider the use of sustainability tools in a variety of agency activities, including voluntary initiatives and activities that EPA is required to undertake according to major legislative initiatives, such as the Clean Water Act. The case studies also illustrate a range in the depth and scope of tool applications—from screening-level assessments to more rigorous quantitative analysis. Table 4-1 lists the case studies considered, the relevant law and type of agency activity, and the sustainability aspects that are discussed. Table 4-2 identifies which of the tools listed in the EPA’s report are included in the case studies. (A glossary of tools and approaches that was developed from the Analytics report is presented in Appendix D of this report.) It is important to note that the committee is aware that, for each case study, there are likely other sustainability tools which are often used in such activities at EPA. The committee did not intend to identify all the potentially important sustainability tools that already may be used–or could be used–in the context of a particular case study.

Each case study summarizes the context for the relevant decision or other activity, traditional analytic approaches used to support the activity, other tools that have been or could be applied to advance the consideration of sustainability concepts, and the expected value to be gained by applying the other tools. After the presentation of case studies, the chapter discusses the increasing use of natural gas for electricity generation as an exemplar that would benefit from applying sustainability tools in a systems (value-chain) context. Finally, the chapter provides general conclusions and recommendations derived from the case studies.

THE DESIGN FOR ENVIRONMENT PROGRAM

Through the Design for Environment (DfE) program, EPA partners with manufacturers to help consumers, businesses, and institutional buyers to identify products that perform well and are cost-effective

and safer for the environment.(EPA 2014c). The program is voluntary but uses many of the tools used in Pre-Manufacture Notice (PMN) evaluations performed in accordance with the Toxic Substances Control Act (TSCA).

Traditional Approach

Companies that manufacture, import, or process any new chemical substance are required to report the chemical name and molecular structure, categories of use, amounts manufactured or processed, byproducts from manufacture, processing, use, disposal, potential environmental or health effects of the chemical and its byproducts, and exposure information. EPA has 90 days from the submission of a notice to assess the risks posed by the new chemical or by a new use of an existing chemical. If the risks are deemed to be unreasonable, EPA is required to take steps to control them. If data contained in the notice are insufficient, EPA may require the submission of additional information.

In its screening analyses under TSCA, EPA makes extensive use of quantitative structure–activity relationships (QSARs). QSARs can be used to estimate environmental persistence, bioaccumulation potential, and toxicity on the basis of the structure of the chemical under consideration. Those attributes can be compared with attributes of the thousands of chemicals previously evaluated in accordance with TSCA (Zeeman, et al., 1993) to make a decision as to whether steps are necessary to control risks.

Tools for Including Sustainability Concepts

Although the DfE program is independent of TSCA, it uses many of the same tools. The goal of determining whether the chemicals are safe for the environment mirrors the goal of PMN evaluations under TSCA, but the goals of assessing cost effectiveness and performance go beyond the TSCA evaluations. To achieve the additional goals in the DfE program, EPA uses an analysis framework, referred to as a chemical-alternatives assessment, in which alternative products are screened, a small number of promising alternatives are identified from the screening, and the screened alternatives undergo additional evaluations. Several recent DfE evaluations illustrate the process.

TABLE 4-1 Relevant Laws, EPA Activities, and Sustainability Considerations for the Case Studies

TABLE 4-2 Sustainability Tools and Approaches Considered in Case Studies^a

^aEach row represents a tool or approach listed in EPA’s Analytics report (EPA 2013a). Each column corresponds to a case study in this chapter. A square indicates selection of the tool for consideration in the case study. The table is not intended to provide a comprehensive list of all the sustainability tools or approaches that are being used or could be used.

In September 2013, EPA issued a draft DfE evaluation of Flame Retardant Alternatives to hexabromocyclododecane (HBCD). HBCD is used primarily as a flame retardant in insulation products, such as expanded polystyrene, but it is a persistent organic pollutant and has been detected in breast milk, adipose tissue, and blood. Alternatives that reduce that environmental footprint were sought. In collaboration with industry, government, and academic experts, EPA performed a chemical-alternatives assessment screening. EPA notes that this screening “along with LCAs [life-cycle assessments], risk assessments, and other tools can be used to improve the sustainability profiles of chemicals and products…. DfE Alternatives Assessments establish a foundation that other tools can build on” (EPA 2014l, p. 1-4). In the preliminary screening, potential alternatives to HBCD were compiled on the basis of the scientific literature and input from experts in chemical manufacturing and product development in industry, government, and academe (EPA 2014l, p. 3-6). Two alternatives emerged, and EPA identified a series of toxicity, ecotoxicity, bioaccumulation, and environmental-persistence metrics for HBCD and the two alternatives. Although EPA does not make a specific recommendation regarding the choice of flame retardant, the alternatives assessment identifies potential substitutes, compares hazards, and supports decision-making by a variety of stakeholders (EPA 2014l, p. iv).

The HBCD and similar alternatives assessments (EPA 2014m, 2012d) illustrate the use of key sustainability analytic tools, including chemical alternatives assessments, collaborative problem-solving, green chemistry, and green engineering. To evaluate sustainability implications more fully, other analytic tools could also be applied. For example, the HBCD assessment does not consider the footprints of the manufacturing processes for HBCD and the alternative products. Coupling of such tools as LCAs and risk assessments with alternatives assessments would explore sustainability factors more fully.

EPA has performed LCAs through its DfE program, although typically not for the same products as for the alternatives assessments. For example, a life-cycle evaluation of current and emerging energy systems used in plug-in hybrid and electric vehicles was conducted through the DfE/ORD Li-ion Batteries and Nanotechnology Partnership (EPA 2013d). LCAs identified which materials and processes were likely to have the greatest impacts or have the greatest potential for improvement.

Expected Value Added by Applying Sustainability Tools

By convening public–private partnerships and using a variety of screening-level and quantitative analytic tools (such as alternatives assessments and LCAs) and indicators (such as ecotoxicity, human toxicity, bioaccumulation, and environmental persistence) that are relevant to sustainability, the DfE program has built well-accepted approaches that help consumers, businesses, and institutional buyers to identify products that perform well and are cost-effective and safer for the environment (EPA 2014c).

COMBINED-SEWER OVERFLOW

Combined-sewer systems are designed to collect precipitation runoff, domestic sewage, and industrial wastewater in a common pipe system that is usually linked to a treatment system. During periods of heavy precipitation runoff, the capacity of the combined-sewer system can be exceeded in such a way that untreated wastewater flows directly into a nearby body of water. EPA has established a combined-sewer overflow (CSO) policy to provide guidance to municipalities in meeting National Pollutant Discharge Elimination System limits under the Clean Water Act. EPA encourages municipalities to incorporate green-infrastructure approaches to help to reduce CSO discharges in their long-term control plans.

Traditional Approach

Under CSO policy, EPA typically expects a municipality to plan for the occurrence of four overflow events (or fewer) in a typical year within 20 years or less. In sensitive areas,¹ EPA has required higher levels of control (for example, Washington, DC, and Cleveland, OH). Municipalities and regulatory agencies have favored the use of “gray infrastructure”—sewer separation, storage tunnels, and additional treatment units—because they are considered to provide a high level of certainty that the allowable number of overflows will not be exceeded. In response to exceeding the allowable maximum, EPA (or a delegated state agency) typically issues an enforcement action, and the municipality is required to construct gray infrastructure by a particular date.

Tools for Including Sustainability Concepts

Green infrastructure—such as dry basins, wet basins, constructed wetlands, rainwater harvesting, infiltration basins, bioretention swales, green streets, pervious or porous pavements, vacant-lot repurposing, green roofs, impervious surface removal, and reforestation—may provide more benefits than those obtained with gray infrastructure. Such sustainability tools as collaborative problem-solving and environmental-justice analysis can be used to assess the benefits associated with those alternatives; the benefits include reduced capital expenditures, improved water quality, and more flexibility. Green-infrastructure initiatives can be used to improve areas where low-income or minority-group communities have been disproportionately exposed to environmental pollution, for example, by providing additional CSO control or transforming abandoned properties into recreational areas.

Expected Value Added by Applying Sustainability Tools

The value added through this approach is illustrated by activities undertaken in Cleveland. EPA entered into a consent decree with the Northeast Ohio Regional Sewer District (NEORSD), which serves 62 communities in the greater Cleveland metropolitan area (NEORSD 2012). The decree requires NEORSD to eliminate an estimated 4 billion gallons of CSO annually and achieve a level of control equating to 98% capture (and treatment) of combined sewage. The control measures are estimated to cost the district $3 billion in capital expenditures and will take 25 years to complete. During the consent-decree negotiations, an additional level of control (62.39 million gallons in a typical year) to be accomplished by upsizing tunnels at an estimated cost of $182 million was proposed. The parties agreed to a combination of cost-effective gray and green infrastructure to capture 44.18 million gallons in a typical year through green infrastructure at a prescribed expenditure of at least $42 million within 8 years (Figure 4-1). The NEORSD evaluated green-infrastructure control measures that addressed storage and treatment; stormwater storage, infiltration, and treatment; stormwater source reduction; and stormwater conveyance and separation (NEORSD 2014b).

EPA has just begun to evaluate CSO or sanitary-sewer overflow (SSO) issues from a sustainability perspective. Municipalities have had to make substantial investments in gray infrastructure to achieve targeted levels of control. Current systems have not been optimized, particularly with respect to nutrient-removal issues or tradeoffs between CSO–SSO control and stormwater control. That is particularly important because many of the CSO control programs are targeted at protecting recreational uses and many water bodies are not used for recreation during large storm events and can self-purify within a day or two (EPA 1977, p. 236).

In discussions among the US Conference of Mayors, EPA, and others, it became clear that existing approaches were resulting in capital expenditures beyond the point of commensurate benefits. EPA re-

____________________

¹Examples of sensitive areas are waters with threatened or endangered species or public drinking-water intakes.

sponded by issuing an Integrated Municipal Stormwater and Wastewater Planning Approach Framework in June 2012 (EPA 2012e). Several communities have prepared integrated plans.

In advocating integrated planning, EPA has the opportunity to help communities to incorporate sustainability analyses into plans, which can include a more comprehensive analysis of long-term CSO control plans or extending control projects for SSOs that appear to pose relatively low risks to human health and the environment.

SITE REMEDIATION

Through the Resource Conservation and Recovery Act and the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA or Superfund), EPA regulates site remediation and establishes guidelines for evaluating and selecting remedies to address soil and groundwater contamination. For Superfund sites, the process is clearly defined in the National Contingency Plan (NCP) (Federal 40 CFR, Part 300).

FIGURE 4-1 Green-infrastructure high-priority areas for the Northeast Ohio Regional Sewer District. Source: Courtesy of the Northeast Ohio Regional Sewer District, August 19, 2014. Reprinted with permission; copyright 2014, Northeast Ohio Regional Sewer District.

Traditional Approach

As directed by the NCP, remedial alternatives are selected and evaluated on the basis of nine criteria. Overall effectiveness in protecting human health and the environment and compliance with applicable, relevant, and appropriate requirements (ARARs) are considered “threshold criteria” that must be met by any alternative. Five criteria are considered “balancing”: long-term effectiveness; reduction in toxicity, mobility, or volume of wastes; short-term effectiveness; implementability; and cost effectiveness. Finally, any remedy must meet state and community acceptance. The results of the evaluations are used by EPA to identity a recommended alternative. Sustainability factors are not designated explicitly as criteria to be considered but are implicit in some of the “balancing” criteria.

Each alternative must, at a minimum, be capable of meeting the two threshold criteria. Remedial alternatives are then assessed and compared with the balancing criteria.

Tools for Including Sustainability Concepts

Collaborative problem-solving, ecosystem-services valuation, and LCA can be incorporated into the remedy selection primarily as part of the assessment of alternatives based on the balancing criteria. That is illustrated by two case studies: a large coal-tar–contaminated site (Pitt-Consul) and a contaminated uranium-processing facility (Fernald).

Pitt-Consol Site

The Pitt-Consol site required extensive remediation to achieve site before sale of the property. Sustainability factors were evaluated for eight remedial options by using LCA software. Sustainability evaluation followed the Sustainable Remediation Forum nine steps for footprint and LCA: define the study goals and scope, define the functional unit, establish the system boundaries, establish the project metrics, compile the project inventory, assess the impacts, analyze the sensitivity and uncertainty of the impact-assessment results, interpret the inventory analysis and impact-assessment results, and report the study results (Favara et al. 2011). Social and economic factors were considered in the remedy-selection matrix by assessing the impacts on the local community and by considering the cost of the remedy. Remedy evaluations are summarized in Table 4-3.

TABLE 4-3 Ranking of Remedies Evaluated by Using LCA Model Results

The remedy chosen for the site, smoldering combustion, was selected because it projected lower material and energy requirements than the other options. The activities for implementing smoldering combustion were determined to have a low impact on the surrounding community in that smoldering combustion required much less transport of materials to the site than other remedies, and this would reduce traffic concerns. As shown in Table 4-3, smoldering combustion had lower life-cycle water use, eutrophication potential, particulate-matter emissions, and greenhouse-gas emissions than the other alternatives. Smoldering combustion also constituted a rapid and permanent solution to the contamination at the site and so would make the property more available for reuse.

This case study illustrates the use of LCA for evaluating remediation options. It demonstrates that, in addition to the traditional balancing criteria used to evaluate remedies, EPA could use criteria that explicitly evaluate environmental, economic, and social sustainability by applying tools that quantify the environmental footprints of alternative remedial strategies.

Fernald Site

The Department of Energy (DOE) 1,050-acre Fernald site, in Crosby, Ohio, processed uranium ore for nuclear weapons. During the middle 1980s, contamination of wells and soils was found off site. Residents were concerned, and the issues were reported in the local mass media. In 1989, uranium manufacturing ended at the site; it was the last of nine US uranium-processing sites to end production of high-purity uranium. Waste management became the focus at Fernald, with a total cleanup cost of $4.4 billion. DOE, federal and state regulators, and the community agreed that it was imperative to involve local parties in remediation and future-use decisions about the site that the community, which is not far from Cincinnati, would need to live with in the foreseeable future. A sustainably protective system was deemed essential.

In 1993, the parties agreed to constitute a special advisory committee (see below) that would advise about a preferred future for the site, allowable residual risk and appropriate remediation levels, disposal options for onsite wastes, and remediation priorities. The group provided recommendations and for more than 20 years it has continued to play a role as the site has changed from uranium production to cleanup and then to a nature preserve and education center while the DOE has continued to manage the legacy of underground uranium contamination. The advisory committee has been involved in some difficult decisions, most notably how to transport waste off site (rail was selected) and how to manage onsite wastes (sequestration by cementation was selected). With the assistance of a charrette process² (see below), the advisory group created a vision of the site that has been realized with the opening of a multipurpose education center, walking trails, legally binding institutional controls, and continuing remediation of the legacy waste.

In making decisions associated with Fernald, some of the tools were required by laws and regulations (such as CERCLA) and were based on risk-related science, engineering and economic costs and benefits, exposure assessment, epidemiology, some version of ecosystem-services valuation, and collaborative problem-solving. Government and private-property owners and managers are required to follow federal and state site-closure requirements that demand modest public participation. DOE has closed hundreds of small sites around the United States that were in remote locations, and its actions were settled

____________________

²Design charrettes are a type of stakeholder engagement tool to develop a mutually agreed-on vision of future development, usually regarding land-use planning decisions.

through negotiations with the federal EPA and state environmental-protection agencies. Some public involvement was part of the process; larger DOE sites and sites near substantial populations have created site advisory boards, which advise DOE about a variety of risk-related and future-use decisions.

The difference at Fernald was that DOE and its government partners recognized that more than just an engineered solution and mandated public involvement was needed; substantive input from an advisory committee with strong local representation was essential. DOE and other government agencies chose Eula Bingham, a resident of the state and a highly respected former health and safety official, to suggest members of an advisory committee. Using personal contacts, mailings, public meetings, and citizen networks, she recommended 14 members that were agreed on by the three government agencies to constitute a Fernald Citizens Task Force (FCTF). Representing a diverse set of skills and interests, the group began working in 1993 and was assisted by four ex officio members from DOE, US EPA, the Ohio Environmental Protection Agency, and the Agency for Toxic Substances and Disease Registry.

The FCTF used the charrette tool to develop “what if” scenarios to work through options. In the Fernald case, the charrette tool (FUTURESITE) allowed FCTF members to visualize and manipulate pieces that represented alternative land-use options for the site. Using charrettes and other risk-analysis tools, the FCTF was able to assess tradeoffs among alternative locations on the site and among options for transporting waste and the final form to store waste.

Charrettes are now widely used in urban and environmental planning and in architecture for helping clients and citizens to participate in the planning process. Visual impressions are critical, and environmental psychologists have found that visual impressions, smell, and touch are critically important in public reactions to alternatives. Most charrettes use or are at least enhanced by computer simulations. For example, in EPA’s role in the environmental-impact analysis, when a new road or bridge is proposed, the Department of Transportation may use charrettes to illustrate the visual impact of placing a road, bridge, or other transportation asset on various alternative routes.

A Fernald committee continues to work with DOE, EPA, and the state on issues related to the site, and DOE has used a similar approach at the Rocky Flats (Denver region) and Mound (Ohio).

Although an impressive array of sustainability tools have been used at Fernald, more formal versions of social-impact assessment, social-network analysis, environmental-footprint analysis, health-impact assessment, and environmental-justice analysis could be used at this and other remediation sites, depending on the specific case in question. Arguably, some of those tools were implicitly part of the Fernald process. For example, to build the advisory team, the parties probably knew the key parties and players in the region and apparently choose wisely to enhance the group rather than include incompatible people. In other words, key elements of social-network analysis were part of this case study even if the tool was not named and practiced as it might be today.

Expected Value Added by Applying Sustainability Tools

EPA and other federal, state, and local government agencies and private landowners could benefit from applying some of the tools for site-remediation decisions described in these case studies. Federal agencies have spent many years at major remediation sites. An issue to consider is how transferable the lessons learned at the Fernald site are to other sites where the expected cost of cleanup is not billions of dollars but tens of millions of dollars.

IMPLEMENTATION OF NATIONAL AMBIENT-AIR QUALITY STANDARDS

Section 109 of the Clean Air Act requires EPA to set national ambient-air quality standards (NAAQSs) for ambient air pollutants considered harmful to public health and welfare.³ EPA has set

____________________

³In the context of the Clean Air Act, welfare refers to the viability of ecosystems and agriculture, the protection of materials (such as monuments and buildings), and the maintenance of visibility.

NAAQSs for six "criteria" pollutants: ozone, particulate matter, nitrogen oxides, sulfur oxides, carbon monoxide, and lead (EPA 2012f). In designing plans to meet the NAAQSs, states develop state implementation plans (SIPs) that can involve the use of sustainability tools.

Traditional Approach

By law, EPA sets both primary and secondary NAAQSs that are based solely on protection of public health and public welfare. Primary standards are established to protect human health with an adequate margin of safety, and secondary standards are established to protect public welfare. As a result, EPA cannot consider issues related to economics and feasibility of achieving the primary, health-based NAAQSs in the standard-setting process. Once NAAQSs are set, EPA, states, and local agencies choose emission-reduction strategies, which are described in the SIPs. The SIPs can consider a variety of factors related to sustainability and other issues, including simultaneously addressing reductions in multiple pollutants.

Traditionally, SIPs have been pollutant-specific; that is, control strategies target only one pollutant at a time. The National Research Council (NRC 2004b, p. 130) discussed inefficiencies and other disadvantages of this single-pollutant approach in which the consideration of only individual pollutants causes the “relatively cumbersome SIP process [to be] undertaken for a pollutant such as ozone and then again for PM in a separate process and on a different timetable, despite the fact that the exposures are simultaneous, the sources are often the same, and the two pollutants share many common chemical precursors.”

As part of the traditional SIP process, state and local agencies generally follow similar procedures in which emission-control options are identified and the cost and feasibility of each control option are assessed. From the identified options, state and local agencies select an overall control strategy and use air-pollution computer models to determine whether the strategy is sufficient to meet the NAAQSs. Implicit in the development of SIPs is the sustainability tool “futures methods” because the SIPs involve making projections of emissions under various future scenarios of growth of and control of emissions. If federally mandated control measures are not sufficient to attain the NAAQSs, additional, region-specific control measures are incorporated into the pollution-reduction strategy until the strategy is shown to meet the NAAQSs. Once the overall strategy is determined, SIPS are submitted to EPA for approval. The process is generally sufficient for single-pollutant management plans and has been successfully combined with benefit-analysis tools (such as BenMAP) (EPA 2014n) and cost-effectiveness tools to evaluate the health benefits and costs of various control strategies further.

Tools for Including Sustainability Concepts

Some sustainability tools are pervasive in the SIP and air-quality management process. For example, under Section 812 of the Clean Air Act, EPA performs estimates of the national costs and benefits of the Clean Air Act. The most recent assessment finds that the 1990 Amendments to the Clean Air Act provide $2 trillion in public health and welfare benefits at a cost of $85 billion (EPA 2013e). Similarly, states assess costs and emission-reduction benefits of control measures as they determine which control measures to incorporate into their SIPs. Other sustainability approaches selected case by case. For example, congestion-mitigation and air-quality improvement programs identify transportation projects, which are funded through the Federal Highway Administration, to simultaneously reduce transportation congestion and reduce emissions. Similarly, electric-utility demand-management programs seek to simultaneously reduce air-pollutant emissions at key times of day and the need for additional power-generation capacity by reducing electricity demand during peak periods.

States have encouraged the participation of stakeholders and consideration of environmental justice issues in their SIPs. Although still evolving, approaches that consider multiple pollutants in air-quality management commonly involve a variety of modeling (such as CMAQ (EPA 2014o)), energy modeling (such as MARKAL (EPA 2012g)), benefit-assessment tools (such as BenMAP), benefit–cost assessment (BCA) tools (EPA 2010a), and risk-assessment tools (EPA 2014p). A multipollutant approach and associ-

ated tools have been used by state agencies as in the case of Georgia (Cohan et al. 2007), by EPA as in Detroit (Wesson et al. 2010), and in cross-medium programs (such as biomass promotion in Massachusetts and mercury regulation in New Jersey).

Multipollutant approaches at the national level are less common, but several have been implemented by EPA, such as the nitrogen oxide cap-and-trade programs, or are being considered by EPA. For example, EPA plans to conduct multipollutant analyses in parallel with the traditional single-pollutant analyses used in setting NAAQSs. The parallel analyses would extend the integrated science assessments and health and welfare risk assessments that are used in setting individual pollutant NAAQSs to consider multiple pollutants simultaneously.

Expected Value Added by Applying Sustainability Tools

Examining the sustainability considerations of various concentration-reduction strategies has multiple potential benefits, including improvements in local and regional air quality regarding multiple pollutants simultaneously, minimization of potential adverse effects while maximizing benefits, consideration of multiple effects (environmental, health, sociologic, economic, and energy-related) of pollution-control strategies, development of cost-effective approaches that meet NAAQSs, ability to consider effects and tradeoffs to multiple media, and evaluations to determine expected amounts of emission reductions (credits) for inclusion of energy-efficiency measures in SIPs.

RENEWABLE-FUEL STANDARD

Through the Energy Policy Act of 2005 and the Energy Independence and Security Act (EISA) of 2007, EPA was given the authority to set regulations in support of a national renewable-fuel standard (RFS). EPA's role is to ensure that transportation fuels have at least a minimum content of renewable fuels, which are produced from renewable biomass. A national goal for 2022 of renewable-duel production of 36 billion gallons per year (about one-fourth of domestic transportation-fuel use) was established by EISA. Among its responsibilities under the RFS, EPA must ensure that renewable fuels meet lower life-cycle greenhouse gas (GHG) emission thresholds than traditional petroleum-based fuels.

The promotion and adoption of higher levels of biofuels caused substantial attention with respect to their sustainability effects compared with those of existing petroleum-based fuels (Jiang and Swinton 2009; Sheehan 2009; Williams et al. 2009; Gnansounou 2011; Lora et al. 2011). Economic issues included such aspects as the relative net economic benefits to consumers from the use of biofuels, the differences in location of fuel production fuel (domestic vs imported), moderation of oil prices, and job creation. Social issues included job creation, rural development, and the equity of using land and crops for fuel instead of food. Environmental issues included the relative energy and emission performance of the various fuels; potential water-quality effects, such as effects on hypoxia in the Gulf of Mexico from fertilizer runoff into the Mississippi River basin; and land use (NRC 2011c).

Traditional Approach

From the outset, EPA used a variety of tools to evaluate the environmental, economic, and societal effects of the RFS simultaneously. Among the environmental effects, EPA is required to ensure that lifecycle GHG emissions of renewable fuels are lower than those of petroleum-based fuels that they replace. For example, corn-based and cellulose-based ethanol must achieve 20% and 60% reductions in life-cycle GHG emissions, respectively, compared with gasoline. Making that comparison requires EPA to determine the baseline GHG emissions of petroleum-based fuels and of the renewable alternatives.

In support of the RFS, EPA's regulatory impact analysis (RIA) established deterministic estimates of life-cycle GHG emissions of petroleum-based gasoline and biobased feedstocks (EPA 2010b). An important issue addressed in the RIA for the RFS was the modeling of emissions from so-called indirect

land-use change (ILUC). ILUC considers broader effects of agricultural soil disruption around the world as a result of local decisions about how to use land (Fargione et al. 2008; Searchinger et al. 2008). For example, diverting corn to the fuel market (instead of food) in the United States would be expected to lead to decreased supply of corn for use as food in the United States. That would increase pressure to grow corn in other parts of the world by either displacing other local crops or converting land to agricultural use. Such conversions would lead to higher GHG emissions and could be linked to the source decision to produce fuel from corn. Effects of ILUC are highly uncertain but have large estimated effects on life-cycle GHG emissions.

EPA's RIA supporting the RFS was perhaps the largest investment of time and effort by the US government to date involving the incorporation of LCA into public-policy decision-making. An important precedent set in the analyses was the selection of single-point values for life-cycle GHG emissions of various fuels. Although EPA’s analysis of the scientific literature found relatively large ranges of values for GHG emissions, EPA inevitably defined a series of life-cycle emissions factors for the relevant fuels. They were all deterministic, fixed-point values and formed the basis of future decisions on whether fuels met the RFS. For example, values of 93 g of carbon dioxide–equivalent emissions per megajoule (93 g/MJ) and 75 g/MJ were determined for the baseline of gasoline and corn-based ethanol, respectively. The corn ethanol value barely meets the 20% reduction called for in EISA. Additional work by EPA has since set values for various other fuels (such as grain sorghum for ethanol). Of course, ILUC has a dramatic effect on the carbon emissions of biofuels, changing a roughly 60% reduction for corn ethanol without ILUC into only a 20% reduction compared with a gasoline-only scenario.

Managing the interests of the various parties requires approaches to combine stakeholder concerns. BCA methods were used to consider net benefits, including differences in prices of fuels and vehicle fuel economy. Sufficiently appreciating the complexity of the carbon emissions of biofuels as a replacement for petroleum-based fuels requires LCA. Considering the uncertainty of life-cycle carbon emissions requires risk assessment.

Tools for Including Sustainability Concepts

A wide variety of sustainability tools could be applied to the decisions related to the RFS, but in this case study the committee focused its attention on the issues related to uncertainty analyses. There was underlying variability and uncertainty in the available life-cycle data used in setting standards according to the RFS, but only a single deterministic life-cycle GHG emission value was set and published. The results were not explicitly expressed as mean values of a probabilistic distribution or otherwise mentioned as probabilistically-based. Various practices in LCA, however, demonstrated how to consider uncertainty and variability robustly in system results.

Not only from a sustainability perspective but from a policy-analysis perspective, it is important to consider more than single deterministic "point values", because many components of the system have various possible resulting emissions rather than a single value. By using ranges and probability distributions that represent potential values, a simulation can be performed to assess the likely comparative performance of fuels. In the end, such analytic methods could support an assessment of the performance of a renewable-fuels policy better. For example, given the probability distributions that represent ranges of life-cycle emissions, a risk analysis could assess the likelihood that corn ethanol could meet the policy target threshold of a 20% reduction from the baseline of petroleum-based gasoline.

Figure 4-2 summarizes estimated probabilities of carbon intensity throughout the life cycle of various transportation fuels (Kocoloski et al. 2013). As noted above, LCAs often use or report only a single value from such a distribution (such as the mean) and might not include the underlying analysis to create the probability distribution. Given the "baseline" of gasoline emissions to be compared, Figure 4-2 shows that various particular sources of biofuels may end up with emissions close to those of gasoline in terms of carbon intensity but also shows that mean values may differ by more or less than the thresholds required in accordance with the RFS.

Figure 4-3 represents the likelihood that particular biofuels (various points on the distribution as in Figure 4-2) have life-cycle carbon intensities compared with that of gasoline that meet their threshold (Mullins et al. 2010). The resulting display of the "risk of policy failure" shows the effect of the uncertainty inherent in the life-cycle data and other data given the relatively large uncertainties in the emissions. RIAs are an appropriate vehicle for incorporating a robust quantification and discussion of uncertainty.

FIGURE 4-2 Probability distributions of estimated carbon intensity of various petroleum-based fuels and biofuels. For biofuels, two modeled cases are presented: full life cycle and life cycle without ILUC. Source: Kocoloski et al. 2013. Reprinted with permission; copyright 2013, Energy Policy.

FIGURE 4-3 Probability that biofuel emissions are below those of gasoline (at 0%) or are below some policy target. The EISA target for corn fuels is 20% reduction and for cellulosic fuels is 60% (shown with vertical lines). Two modeled cases are presented: full life cycle and life cycle without ILUC. Source: Adapted from Mullins et al. 2010.

Expected Value Added by Applying Sustainability Tools

Additional applications of LCA in support of public policy and decision-making could be expected, for example, in broader consideration of carbon dioxide limits for energy-generation units that consider upstream methane emissions of natural gas–fired power plants. Such analyses could follow expanded methods similar to those described above to provide clear and robust demonstrations (and set best practices) of merging life-cycle, risk, and other analytic tools.

There are additional opportunities for application of sustainability approaches on different geographic scales that could point to practices and policies for mitigating adverse environmental effects associated with renewable fuels.

Ecosystem-services valuations that address the tradeoffs between land-management practices and water-quality improvements on local, regional, and continental scales could potentially lead to the development of markets that provide incentives for both non–point-source and point-source reduction.

CONSIDERING GENERATION OF ELECTRICITY FROM NATURAL GAS IN A VALUE-CHAIN CONTEXT

A number of business sectors have recognized the growing interdependence of economic relationships that include economic, environmental, and societal effects that extend well beyond the boundaries of individual firms (see Chapter 5). The interdependence can be considered as a value chain, along which businesses add value to the initial input of raw materials through various functions that result in finished products.

In a sustainability context, a value chain consists of the following major functions:

• Product research and development.

• Extraction and consumption of raw materials.

• Transportation of raw materials for storage or for intermediate or direct processing.

• Manufacturing.

• Distribution and logistical operations to move a manufactured product to a business customer or consumer.

• Product use.

• Postcustomer use of a product.

EPA traditionally focuses on reducing emissions or waste releases from individual or regional source categories irrespective of their relationship to or effect on the sustainability performance of the larger value chains. The increased use of natural gas for electricity production in the United States serves as an example to illustrate additional opportunities for EPA to incorporate sustainability concepts into its decision-making. Figure 4-4 shows the infrastructure for natural gas in the United States and represents the value chain in a physical sense. Commercial activities along the chain include extraction of natural gas from underground reservoirs, processing to remove nonmethane components, shipping and trading of the cleaned gas, transmission (for example, through pipelines), storage, and distribution to end users for electricity generation, heating, industry feed stocks, and transportation fuel.

Driven by technologic innovations in horizontal drilling and hydraulic fracturing, natural-gas production is rapidly increasing in the United States. Domestic production is projected to increase 30% by 2035, and much of the gas will displace coal in electricity generation (EIA 2012). Some fuel-switching

has already begun. Because electricity generation is the largest sector of energy use in the United States, accounting for about 40% of primary energy production, shifts to natural gas raise the potential for broad systemic effects—similar in breadth to the effects associated with the use of renewable transportation fuels. As natural gas continues to displace coal for electricity generation, changes in an aggregate sense can be expected to occ r in

• Greenhouse gas emissions.

• Water use and water quality.

• Air quality (related to emission of criteria pollutants and air toxicants).

• Land use (for example, for extraction).

To support decision-making concerning the increased use of biofuels for transportation, EPA took a systems and LCA approach. Similar approaches are warranted for electricity generation, in which equally dramatic transformations are under way. Examining sustainability effects by using LCA in a natural-gas value-chain context provides important systematic perspective. For example, it is well established that when natural gas is burned to produce energy, HG emission per unit of energy released is lower than that of the other two principal fossil fuels, petroleum and coal. However, if methane, the primary constituent of natural gas and a potent GHG, leaks along the natural-gas value chain, much of or all the GHG advantage of natural gas in combustion can be lost.

Similarly, several million gallons of water can be used in fracturing at a natural-gas well. Over an electricity-generation life cycle, however, the water consumption can be offset by the lower water use in natural-gas combined-cycle electricity-generation facilities than in existing coal-fired electricity-generation units (Scanlon et al. 2013).

FIGURE 4-4 The natural gas infrastructure in the United States. Source: MIT 2010, p. 59. Reprinted with permission; copyright 2010, Massachusetts Institute of Technology.

As it develops regulations, such as those for carbon dioxide emission from electricity-generation units, EPA should consider applying sustainability tools, such as LCA, in a value-chain context. The benefits of doing so are expected to include

• Adoption of a more consistent or coordinated approach to the application of sustainability concepts in all major EPA decision-making programs.

• Development of a robust dataset to obtain a more comprehensive understanding of the total effects, which can guide decision-making that is less likely to have unintended consequences at different stages of the value chain.

• Identification of opportunities for cost-effective innovative approaches to reduce specific effects that go beyond what can be achieved with existing regulatory tools.

• Identification of new and more significant opportunities for collaboration with nongovernment organizations and the private sector.

CONCLUSIONS AND RECOMMENDATIONS

Key Conclusions and Recommendations

Other Recommendations