FROM ATMOSPHERIC
CONTAMINANT DEPOSITION

PROCEEDINGS OF A LEGAL WORKSHOP
HELD AT THE UNIVERSITY OF WINDSOR
WINDSOR, ONTARIO

OCTOBER 26-27, 1999

Report to the International Joint Commission

by the International Air Quality Advisory Board

and the Great Lakes Water Quality Board

October 2000

ISBN 1-894280-23-7

This report was prepared from the written program description distributed prior to the workshop, the written legal submissions, the oral presentations, and the questions and discussion. The content is designed to appeal to both the legal community and others interested in environmental protection.

The views and opinions expressed in this report are those of the workshop participants and not necessarily those of the International Joint Commission.

FOREWORD Leonard Legault 1

EXECUTIVE SUMMARY 5

INTRODUCTION 11

Background and Context 13

Technical Scenario 16

UNITED STATES CASE STUDY 27

Laying the Groundwork and Setting the Legal Scene 29

Plaintiff Friends of Lake Hypo. Motion for Summary Judgment 33

Defendant Acme Widget. Motion for Summary Judgment 39

Plaintiff Friends of Lake Hypo. Rebuttal 45

Questions and Discussion United States Case Study 47

CANADIAN CASE STUDY 51

Laying the Groundwork and Setting the Legal Scene 53

The Applicant's Submission to the Ontario Environmental Appeal Board 57

The Director's Submission to the Ontario Environmental Appeal Board 59

The Intervener's Submission to the Ontario Environmental Appeal Board 61

Questions and Discussion Canadian Case Study 67

COMPARING AND CONTRASTING THE CANADIAN

AND UNITED STATES LEGAL LANDSCAPES 73

KEY FINDINGS 77
APPENDICES 83

A. Reports and Background Information Consulted 85

B. Summary of Comments Received after the Workshop 87

C. Workshop Participants 89

D. Plaintiff Friends of Lake Hypo's Motion for Summary Judgment

Neil S. Kagan and Paul Benington 93

E. Defendant's Brief in Support of Motion for Summary Judgment
Peter D. Holmes and Karen H. Schodowski 117

F. Appeal to the Environmental Appeal Board by the Applicant
Acme Widget Limited John L. Martin 139

G. Ontario Ministry of the Environment The Director's Submissions
Pat Moran 151

H. The Intervenor's Factum
Paul Muldoon, Marcia Valiante, Laura Shaw, and Toshi Takishita 159

Welcome to our legal workshop. For those of you who are not familiar with the Great Lakes Water Quality Agreement and the International Joint Commission, permit me to provide background and context.

More than a quarter century ago, in 1972, Canada and the United States entered into the Great Lakes Water Quality Agreement. The Agreement committed both countries to restoring and protecting the environmental quality of our Great Lakes. Amendments in 1987 provided a specific focus on airborne toxic substances, recognizing that the atmosphere is a significant pathway by which contaminants reach the lakes. As a result of this focus on air contamination, considerable time and effort have been devoted to better understanding the atmospheric transport and dispersion of contaminants and, through the application of technology, to reducing contaminant releases to the atmosphere. Air quality today is much improved, but there is still a great deal to be done.

The Commission's role under the Agreement is to review and assess the governments' programs and progress "to restore and maintain the chemical, physical, and biological integrity of the waters of the Great Lakes Basin Ecosystem." The Commission is the conscience of the governments and the watchdog for the public, the citizenry of Canada and the United States in this region.

This legal workshop corresponds to one element of the Commission's mandate. To more fully address the air contaminant issue, the Commission's Great Lakes Water Quality Board and the International Air Quality Advisory Board identified the adequacy of the legal framework in both countries to reflect the input of atmospheric deposition as a topic warranting consideration. Given that the atmosphere is a source of contaminants to the waters of the Great Lakes, the question is:

What legal tools are available, under current Canadian and United States law, to control atmospheric deposition of persistent toxic substances to the Great Lakes originating from sources within and outside the basin?

Summary

What legal tools are available, under current Canadian and United States law, to control atmospheric deposition of persistent toxic substances to the Great Lakes originating from sources within and outside the basin?

A hypothetical scenario was developed in which Acme Widget, a proposed new facility, would emit mercury from its stack. A portion of that mercury would be deposited into nearby Lake Hypo, increasing the loadings to the lake by 3%. As a result, levels of methyl mercury in fish were predicted to increase by 0.6 to 3%. Legal teams from the United States and Canada presented arguments from the perspective of Acme and the environmental group, Friends of Lake Hypo. A panel provided reactions to the legal arguments and all participants engaged in open discussion.

UNITED STATES CASE STUDY

Under the division of powers in the United States, both the federal and state governments have authority for environmental protection, with primary authority generally residing with the federal government. An overview of the U.S. legal framework covered relevant sections of the Clean Water Act, the Clean Air Act, and the Resource Conservation and Recovery Act.

In the scenario, Acme applied for and received approval from the state of Michigan for an air emissions permit. The state conducted air quality and deposition modelling and a multi-media risk analysis in support of its permitting decision.

Friends of Lake Hypo argued that Acme should obtain an NPDES (National Pollutant Discharge Elimination System) effluent discharge permit under the Clean Water Act and, furthermore, should be denied a permit, since Lake Hypo was already impaired due to high levels of mercury. Key points in their submission included:

• Issuance of an air emission permit does not necessarily meet the goals and requirements of the Clean Water Act.

• Acme's stack should be considered a point source under the Clean Water Act. The definition of point source is broad enough to include any conveyance from which pollutants are discharged. Congress explicitly identified certain exceptions to the definition but did not specifically exempt stacks.

• Modelling predicts that mercury from Acme's stack would be deposited into Lake Hypo. Courts have accepted the results of modelling such as that used here as a basis for their rulings.

• The U.S. Environmental Protection Agency (EPA) has traditionally defined non-point source pollution as diffuse pollution. This case is different from diffuse pollution, in that the emissions deposited to the lake can be traced back to their source. U.S. EPA has never adequately addressed this kind of situation.

In rebuttal, Acme argued that it is a non-point source and therefore not subject to NPDES permit requirements. Key points:

• Congress intended the Clean Air Act to address air deposition, including deposition to water bodies. Congress distinguishes between emissions and discharges, using the former under the Clean Air Act and the latter under the Clean Water Act. Other provisions of the Clean Air Act, such as the Great Waters Program, indicate that Congress envisioned the statute to address air deposition to water.

• There are no cases in which stacks have been considered point source discharges. The Clean Water Act definition of conveyance does not include stacks.

• The modelling uses unrealistic, overly conservative assumptions and is not validated. Even given such conservative assumptions, modelling cannot predict any more than a very small, unmeasurable increase in mercury in Lake Hypo.

CANADIAN CASE STUDY

Under the division of powers in Canada, the environment is largely within provincial jurisdiction. In Ontario's legal framework, water sources are regulated under the Ontario Water Resources Act. Environmentally significant releases other than to water must get a Certificate of Approval under the Ontario Environmental Protection Act.

In the scenario, Acme applied for and was granted a Certificate of Approval; however, the Director imposed a more stringent mercury emission limit as a condition. Acme appealed the condition to the Ontario Environmental Appeal Board. Key points in their appeal included:

• The Director, Ontario Ministry of the Environment (MOE), does not have the legal authority to impose the condition.

• There is no reasonable link between the mercury levels in Lake Hypo and the emissions from Acme, nor is there a link from the emissions to any environmental impact in the lake.

• There is no economically achievable control technology capable of meeting the required emission rate. Acme's technology is the best available and meets the legal requirements of the Act and its regulations.

Ontario MOE responded as follows:

• The Director may impose emission limits, to prevent or alleviate an adverse effect if there are probable grounds for such an effect to occur. If contaminants in the air cause an adverse effect in the water, then the Director may limit the emission of contaminants into the air.

• Even if the mercury from Acme meets regulations, the Director must take into account total mercury input to the lake and the amount that Acme would add.

• In addition to stack source controls, Acme needs to examine other mechanisms to reduce mercury, for example, changes to its production process.

Friends of Lake Hypo supported the Director's position but contended that the Ontario MOE condition was not sufficiently stringent. Key points:

• Mercury is a persistent, bioaccumulative toxic substance that poses a known threat to human and ecosystem health. Therefore, the Director has probable grounds to impose a condition of zero discharge.

• Many international and governmental institutions, including the Canadian and the United States federal governments through the Great Lakes Water Quality Agreement, have committed to virtually eliminate releases of persistent, bioaccumulative toxic substances. The Board must consider this broader policy and legal context and impose zero discharge, which is the only way to achieve virtual elimination.

AVAILABILITY AND USE OF LEGAL AUTHORITIES

Based on the written briefs submitted and the oral arguments presented, the panel offered answers to two specific questions regarding the adequacy of the legal frameworks in Canada and the United States.

Canada

1. Does the Canadian legislative framework allow Canada to effectively address the problem of water pollution via the atmosphere?

A qualified yes. The combination of federal and provincial frameworks is patchy, not comprehensive, and frequently appears to be ad hoc and subjective.

2. Does Canada's domestic legislative framework provide for effective fulfillment of its international obligations?

A qualified no. The framework is not sufficiently prescriptive.

United States

1. Does the United States legislative framework allow the United States to effectively address the problem of water pollution via the atmosphere?

A qualified yes. Section 112 of the Clean Air Act is adequate. However, a more aggressive and creative approach needs to be taken for application of that framework.

2. Does the United States' domestic legislative framework provide for effective fulfillment of its international obligations?

There is a disconnect between the U.S. legislative framework and the international commitment to virtual elimination, specifically in the definition of the goal. If virtual elimination is defined as absence of harm, that is, removal of fish-consumption advisories, the answer is a qualified yes. If virtual elimination is defined as zero discharge, the answer is a qualified no.

KEY ISSUES

Fulfillment of Agreement Requirements

In order to ascertain whether programs are on track to meet Agreement requirements in general and the requirements of Annex 15 (Airborne Toxic Substances) in particular, the underlying science must be improved and air and water monitoring programs expanded.

The Science

Technical challenges must be addressed to better link loadings to a particular water body back to specific pollutant sources. Also, the scientific community must be better engaged so their work informs policy making. The science to link atmospheric deposition back to individual sources, where reductions are required, can proceed in parallel with work on the institutional framework to achieve those reductions.

The TMDL (total maximum daily load) process under the U.S. Clean Water Act offers promise as a legal authority and conceptual approach to deal with air deposition of contaminants to a water body. However, the time frame to implement the process, much less to realize environmental improvements, is too long. Appropriate legal and institutional frameworks must be developed in a more timely manner.

Virtual Elimination and Precaution

The concept of virtual elimination for persistent toxic substances has been adopted at the national level in both Canada and the United States reflected, for example, in domestic federal legislation. Their commitment is further reflected as signatories to the Agreement. The precautionary principle has been articulated at the international level and ascribed to by both countries. However, both Canada and the United States need to more explicitly incorporate virtual elimination and precaution into specific domestic programs and decision-making processes. Further, both countries need to better align their existing legislative and institutional frameworks and better use their authorities to help meet their international obligations.

Other Legal Authorities

The U.S. attorneys based their cases on selected provisions of the Clean Air Act and the Clean Water Act. Other sections of the Clean Air Act could provide additional tools to protect water quality from air contaminant deposition and should be explored.

States have the option of exercising their independent sovereign authority to act or regulate in the area of toxic air deposition, to account for persistent, bioaccumulative toxic substances and for multiple exposure pathways. Legal arguments around the application of state authorities, such as Michigan's Rule 228, would also be beneficial.

This workshop addresses whether the United States and Canada have the legal authority to adequately deal with the atmospheric deposition of persistent toxic substances into the Great Lakes. The International Joint Commission first recognized a quarter century ago that atmospheric emissions from sources both within and outside the Great Lakes basin contribute significantly, through deposition, to the total burden of persistent toxic substances in the lakes. Since first determining that the air is a pathway by which contaminants reach the Great Lakes, the magnitude of atmospheric transport has become increasingly better defined. This is especially significant when considering the magnitude of the associated economic activity.

Canada and the United States are each country's largest trading partners. Over the past 25 years, the scale of human economic activity has become better understood, especially the downside one consequence of human activity are toxic contaminants. These contaminants respect no boundary, but it has taken years to realize and appreciate that their impact on, and consequences for human and ecosystem health is not trivial. The atmospheric deposition of persistent toxic substances is a barrier to restoring and protecting the lakes.

To ensure that the border stays friendly and that citizens of both nations continue to enjoy a high standard of living, contaminant issues, among others, need to be addressed in an amicable way.

When the United States and Canada amended the Great Lakes Water Quality Agreement in 1987, both countries formally acknowledged for the first time that atmospheric deposition had to be dealt with. Specifically, the addition of Annex 15 Airborne Toxic Substances committed Canada and the United States to determine contaminant loadings into the Great Lakes via atmospheric deposition and to develop, adopt and implement measures to control toxic substance emission sources and eliminate persistent toxic substance emission sources and, hence, eliminate the loadings.

To assist the governments in this endeavor, the Commission's Great Lakes Water Quality Board and International Air Quality Advisory Board sponsored a workshop, held May 21-22, 1997 in Romulus, Michigan, that provided opportunity for dialogue between scientists and managers. Specifically, the technical presentations at that workshop communicated to managers a distinct need to more fully develop the tools necessary to address contaminant air transport and deposition. The boards began to wrestle with the question of how, given the technology available and the existing legal structure within each country.

From a technological perspective, contaminant emissions to the atmosphere can be reduced and eliminated through application of control technology and changes in production processes. The boards, however, also saw an opportunity to explore the legal landscape, specifically the question posed in the foreword by Chairman Legault, what legal tools are available to protect water quality from contaminant air deposition? If Canada and the United States already have the authority to protect water quality from contaminant deposition from the air, through regulations and other legal tools, is that authority being adequately applied? If authority is not in place, should the governments ask Parliament and Congress for such authority?

The purpose of this workshop was to explore what legal tools are available in an animated and interactive setting resembling a "moot court," in an environment of mutual trust and respect. The workshop

The workshop was designed to provide a fact-finding environment to identify and demonstrate legal tools that could mitigate contributions of atmospheric contaminants subject to deposition and to assess the relative capabilities of such tools. Recognizing the myriad technical, legal, and policy aspects that could be considered, as well as the distinct regulatory structures in Canada and the United States, three hypothetical scenarios were created:

• The technical scenario was constructed largely from "real world" information. It describes a hypothetical water body, Lake Hypo, which could be adversely impacted by mercury emissions from a proposed new source, Acme Widget, located near the lake shore. Human health could be adversely impacted through consumption of fish whose mercury concentrations could be increased by the release from Acme. The technical scenario underpins both the U.S. and the Canadian legal scenarios.

• The United States legal scenario was constructed under U.S. federal and Michigan state law as a challenge to the operation of a new air source. Under this scenario, a permit has been issued to a new air source, and an environmental non-government organization (ENGO) has filed a suit against the source, among other legal claims, for potential violation of environmental protection laws. A hearing is held on cross motions for a summary judgment.

• The Canadian legal scenario was constructed under Ontario law where a Certificate of Approval has been issued by the Ontario Ministry of the Environment for a new air source. A hearing is held in response to the source's appeal of the condition imposing a mercury emission limit in the Certificate, and the ENGO has been granted intervention status.

Although the scenarios are realistic, a number of simplifying assumptions were made to ensure focus on the legal question at hand. For each legal scenario, a permitting process presently in place under the prevailing regulatory structure in each country was used.

To be able to address contaminants that cross the border, it is essential that citizens of one country understand and appreciate not only the governmental and legal structure of their own country, but the infrastructure and arrangements of their neighbor across the border. The Detroit incinerator case of the late 1980s provides a prominent example of the importance of such understanding. The facility in that case was built on one side of the border but impacted people on both. The resulting legal issues and court case demonstrated the similarities and dissimilarities in each government's administrative and legal structures and a lack of understanding of important differences between them. This present workshop was intended to further improve the understanding of such processes and differences.

Specifically, the workshop and these proceedings are structured as follows:

• Technical scenario.

• United States case study.

• Canadian case study.

• Key findings.

Given the distinct legislative structure in each country, the Canadian and the U.S. legal scenarios must be treated separately. To present arguments from the perspective of the source proponent and the ENGO, four legal teams two U.S. and two Canadian prepared and submitted written briefs prior to the workshop. At the workshop itself, each side presented an oral summary of its case and, to ensure that the issue was fully aired, a panel composed of Commissioners and members of the two sponsoring boards asked clarifying questions; and dialogue was stimulated through audience interaction. These written briefs, as submitted, are appended along with a summary of the oral arguments and a summary of the questions and discussion.

PANEL MEMBERS

• Alice Chamberlin, International Joint Commission

• Wayne Draper, International Air Quality Advisory Board

• Leonard Legault, International Joint Commission

• Ed Piché, International Air Quality Advisory Board

• Vic Shantora, Great Lakes Water Quality Board

• David Ullrich, Great Lakes Water Quality Board

No decision on the cases presented is forthcoming. The workshop purpose was to lay out, not judge, legal and policy issues associated with atmospheric deposition of contaminants to a water body. The desired outcome was to gain a deeper understanding of available legal tools and their viability.

The workshop and these proceedings are a means, not an end. They provide a basis for follow-up, specifically, to stimulate further consideration by the boards, the Commission, and others about the efficacy of existing legal tools to further control atmospheric deposition of persistent toxic substances. That is, given the present technical and legal landscape, what must be done to effectively mitigate atmospheric contaminant deposition to the Great Lakes? The boards envision this workshop as one in a series of events to better understand how well legal processes function on both sides of the border to address atmospheric deposition, and to explore how the boards and the Commission can contribute to improving the processes.

The technical scenario is constructed from realistic and appropriate "real world" information. Since the intent is to explore legal instruments, the scenario is not unduly complex. It incorporates a number of assumptions, for instance, in regard to source emissions, deposition, and fish contaminant uptake; the merits of each could be debated. However, as constructed, the technical scenario allows a focus on the purpose of the exercise and provides the desired latitude for the legal teams to present their cases and contend the extent to which the proposed source should be controlled.

Understanding of the ecosystem can be enhanced through the use of models, such as for contaminant dispersion, deposition, transformation in the water column, and movement among ecosystem compartments, including fish uptake. Multi-pathway risk assessment and the concept of total maximum daily load (TMDL) also draw upon models. Each model relies on a series of assumptions, some of which have been validated by subsequent real-world measurements. In addition, models are continually evolving, and no one model will ever be available to provide a definitive answer. Notwithstanding the uncertainty associated with models, they can nonetheless be useful predictive tools.

The technical scenario includes:

• Why mercury?

• Physical characteristics of the hypothetical receiving water, Lake Hypo.

• Emission characteristics of the hypothetical source, Acme Widget.

• Background concentrations.

• Dispersion models.

• Chemical forms of mercury.

• Mercury deposition estimates.

• Multi-pathway risk assessment.

• Mercury transformation and fish uptake.

• Human exposure and mercury uptake.

• Fish consumption advisories.

• Summary of mercury inputs to Lake Hypo.

• Questions and discussion expanding on relevant technical aspects.

WHY MERCURY?

Mercury was selected, first, because elevated mercury levels have been shown to compromise environmental integrity and to pose a threat to human health. For example, mercury is responsible for 60% of fish-consumption limitations for all water bodies in Ontario and for 99% of the limitations issued for

Mercury was also selected because a large body of "real world" emission data is available, as are background data for mercury in the ambient atmosphere and in the water column. Further, dispersion and deposition modelling has previously been done for this contaminant, thereby providing a comfortable level of confidence to the scenario.

PHYSICAL CHARACTERISTICS OF LAKE HYPO

The physical characteristics presented for Lake Hypo are used in various calculations to determine the impact of the hypothetical new source, Acme Widget, on Lake Hypo. The lake's characteristics are patterned on a reoriented Lake St. Clair.

• Shoreline length: 272 km

• Maximum natural depth: 6.5 m

• Maximum cross section length: 45 km (running diagonally northwest to southeast)

• Maximum width: 40 km (perpendicular to length)

• Surface area: 1115 km ²

• Elevation above sea level: 174.65 m

• 98% of water enters the lake via a large river at the northwest corner of Lake Hypo; the average discharge is 5200 m ³ /sec and ranges between 3000 and 6700 m ³ /sec

• Outflow from the lake is via another large river at the southeast corner

• Average flush time: 5-7 days

The source Acme Widget is located near the southeast corner of Lake Hypo, on the river that drains the lake. Acme Widget has the following characteristics:

• Mercury emission concentration: 0.075 mg/std m ³ @ 25 ^o C and 101.3 kPa and 7% O ₂

• Stack height: 90 m (295 feet)

• Stack exit diameter: 2.5 m (9 feet)

• Stack gas exit actual temperature: 90 ^o C

• Stack gas velocity: 16 m/sec

• Stack gas flow rate: 78.5 m ³ /sec @ 25 ^o C and 101.3 kPa and 7% O ₂

• Mercury emission rate: 21.2 g/hour (0.508 kg/day = 193.1 kg/year = 0.21 tons/year assuming near continuous operation)

BACKGROUND CONCENTRATIONS

Consistent with measurements from the Upper Great Lakes Connecting Channels Study, the concentration of mercury in the water column both the inflowing river and Lake Hypo is assumed to be 2 ng/L. This value includes all upstream sources, both direct and indirect. The concentration is more than the GLI (Great Lakes Water Quality Initiative) mercury water quality criterion of 1.3 ng/L to protect wildlife and the 1.8 ng/L criterion to protect human health. Since promulgation of these criteria in 1995, the U.S. Environmental Protection Agency (EPA) has recommended that states and tribes use a revised human health criterion of 2.9 ng/L; studies of health effects continue.

Lake Hypo is treated as a flow-through system. Using the average river flow given above, this translates into an average annual mercury loading of 328 kg. This estimate will vary with the river discharge volume.

DISPERSION MODELS

Dispersion models were used to predict the point-of-impingement concentration of mercury and to provide annual average air concentrations over Lake Hypo.

Half-Hour Point-of-Impingement Concentration

Regulation 346 of the Ontario Ministry of the Environment (MOE) contains a half-hour point-of-impingement ¹ standard of 5 g/m ³ for mercury and references the use of a specific dispersion model to predict the atmospheric concentration, for comparison with the standard. Regulation 346 also permits the use of other dispersion models. For the Lake Hypo scenario the U.S. EPA SCREEN3 model, which considers a wider range of meteorological conditions, was also run.

• The maximum half-hour point-of-impingement concentration predicted by the Regulation 346 model was 10 ng/m ³ , which is 1/500th of the standard.

• The SCREEN3 model predicted a maximum one-hour average concentration of 17 ng/m ³ .

The dispersion model used to estimate annual average air concentrations of mercury across Lake Hypo was developed by Ontario MOE and used in the Windsor Air Quality Study to estimate air concentrations in the greater Windsor area. The model and the work done are described in the report, Mathematical Modelling and Source Apportionment . During the Detroit incinerator court case in the late 1980s, the Ontario MOE model and the U.S. EPA Industrial Source Complex model were both run, and both yielded similar average atmospheric mercury concentrations.

For the Lake Hypo scenario, the results from the Windsor Air Quality Study were scaled to correspond to differences in stack characteristics and emission mercury rates for the Detroit incinerator, compared to the values assumed for Acme Widget. Also, in that study, the source was located some distance from the lake, whereas Acme is located near the shore of Lake Hypo. In the real world, some dispersion and dilution as well as land deposition would occur prior to the plume reaching the lake. For the purposes of this scenario, however, given the 40+ km distance that the plume travels across the open lake, these factors do not appreciably alter the deposition estimates presented below.

To maximize the potential for mercury deposition to Lake hypo, the wind rose was adjusted so that the prevailing (most frequently occurring) wind, from the south and southeast, carried the plume from Acme Widget across the lake a major fraction (30 - 40%) of the time. This translates into 60 - 80 kg of the annual mercury emissions from Acme Widget being transported across Lake Hypo. Application of the dispersion model for the source characteristics of Acme Widget provides an estimate of the average annual concentration across the area of Lake Hypo of 30 - 40 pg/m ³ .

For comparison, the background mercury concentration due to global emissions is ~1500 pg/m ³ , measured at locations remote from mercury sources. The human lung volume is approximately 3 L, which translates to a human intake of approximately 5 pg of mercury with each breath.

Wind direction is the major atmospheric variable to estimate atmospheric mercury concentration over the lake. Other factors such as temperature and the extent of cloud cover are relatively non-critical.

CHEMICAL FORMS OF MERCURY

Mercury emissions can occur in several forms. The most likely are elemental mercury, mercury chloride, and particle-bound mercury. These behave quite differently in the atmosphere in terms of chemical reactivity, dry deposition, and wet scavenging. The major uncertainties in calculating the deposition of mercury directly to the surface of Lake Hypo are knowing the chemical form of mercury and its affinity with other contaminants, especially particulate matter.

• Mercury chloride has a lifetime in the atmosphere ranging from hours to a day. It is very soluble and is therefore readily deposited in both dry or wet form.

• Elemental mercury has a lifetime ranging from months to a year. It has a very low solubility and, because of very low deposition rates, can be transported around the globe several times before deposition occurs.

• Particle-bound mercury , with a lifetime of days, is largely scavenged by precipitation and has a much lower dry deposition rate than mercury chloride. The chemistry of water droplets containing scavenged mercury is complex. For example, water droplets which evaporate release particulate mercury which can be re-scavenged later.

The total deposition to Lake Hypo can therefore vary by one-to-three orders of magnitude, depending on the chemical form of the emitted mercury.

The estimate of mercury deposition to Lake Hypo is based on a number of associated assumptions, including the form of mercury, the modelled annual air concentration over the lake, and dry and wet scavenging rates. No deliberate attempt was made to either over- or under-estimate the deposition values as a result of variability in these factors.

This scenario assumes deposition of mercury both background and from Acme to the lake surface only and not to the watershed, thereby excluding such factors as land use characteristics and the size of the watershed. The scenario also assumes no other mercury sources are large or close enough to Lake Hypo to contribute a significant amount of mercury chloride to the background deposition to the lake.

Acme Widget

For the model runs, it was assumed that the emitted mercury (193.1 kg/year = 0.21 tons/year) was entirely in one chemical form and remained in that form until deposited to Lake Hypo. The U.S. EPA SCREEN3 model yielded the following estimates of annual mercury deposition:

Mercury Particulate Elemental

Chloride Mercury Mercury

Dry Deposition 15 1.2 ~0.01

Wet Deposition 2.5 2.5 < 0.01

Total Deposition 18 4 0.02

Each number has an associated uncertainty range, due to variations in such factors as deposition velocity, concentration of mercury across the lake, and deposition variability associated with meteorological conditions (such as rainfall occurrences and cloud cover), especially for dry deposition.

In reality, the amount and form of mercury emitted from the stack depends on the emission control technology used. This information is crucial to estimating mercury deposition to Lake Hypo. The scenario assumes that maximum achievable control technology (MACT) has been applied to Acme, resulting in mercury emission in the ratio of 50:50 elemental:chloride. This estimate is not inconsistent with mercury emissions from a municipal waste incinerator with non-aqueous control technology and equates to an annual deposition to Lake Hypo of 9 kg.

The scenario also assumes no particle-bound mercury in the emission. Unless the proportion of particulate mercury present were quite large, the deposition of mercury chloride would predominate.

Another consideration is mercury transformation in the plume prior to deposition. The time between emission and deposition is considered too short for transformation of elemental mercury, but mercury chloride can be transformed to elemental mercury or attached to particles in the plume. However, in-plume mercury transformations cannot be modelled easily. For simplicity, the above estimates assume no plume transformation. This results in an overestimate of mercury deposition, the extent of which is difficult to quantify.

The wet deposition of background mercury can be estimated from monitoring sites in the area. The wet deposition estimate includes all forms of mercury present chloride, particulate-bound, and elemental in precipitation. The background concentration of mercury in precipitation averages ~10 ng/L, which corresponds to an annual deposition of mercury to Lake Hypo via precipitation of ~8 kg.

In the dry phase, the background concentration of mercury is overwhelmingly dominated (>98%) by elemental mercury. Dry deposition of elemental mercury is not well known but is likely much smaller than wet deposition to the lake surface. For this scenario, the annual dry deposition of mercury to Lake Hypo is assumed to be ~1 kg.

This yields a total annual background input of mercury to Lake Hypo of 9 kg.

MULTI-PATHWAY RISK ASSESSMENT

In addition to direct deposition into Lake Hypo, mercury emitted from Acme can enter other ecosystem compartments, such as land within the basin watershed. Mercury sources, transport, deposition, and recycling in the ecosystem are schematically depicted below. The scenario assumes that a multi-pathway risk assessment was conducted as part of the permitting process and that, as a result of the analysis, the proposed source is predicted to contribute to an increase in the mercury levels in fish.

Factors such as mercury sources in addition to Acme, deposition to and runoff from the land portion of the basin, and release from the sediment are necessary considerations of the risk assessment. For example, U.S. EPA's soil erosion model could be employed. However, to avoid undue complexity in the scenario and to ensure that the focus remains on the applicability of legal instruments to Acme, only three sources were considered direct deposition to Lake Hypo attributable to Acme and from the background, and inflow of the major river. Other sources and pathways, including indirect inputs of mercury from Acme via deposition to the watershed, were excluded from further consideration.

The form of mercury that poses the greatest threat to human and ecosystem health is methyl mercury. It is readily absorbed regardless of the exposure route, and it has the ability to bioaccumulate through the food web, including humans. Virtually all of the mercury detected in fish is in the form of methyl mercury.

The transformation of mercury into a methylated form depends on many factors and can vary dramatically from lake to lake. The scenario assumes that methylation occurs within Lake Hypo and that all forms of mercury undergo methylation at the same rate. Therefore, any increase in mercury deposition to Lake Hypo as a result of emissions from Acme Widget will manifest itself through an increase in the concentration of mercury in fish.

The relationship between change in mercury deposition to the water and mercury concentration in fish requires further study. Information presented in U.S. EPA's Mercury Study Report to Congress indicates that the relationship may be 10:2, that is, a ten-fold deposition increase translates into a two-fold fish increase. Other information indicates that a ten-fold deposition increase may translate into a greater than two-fold fish increase. This scenario assumes that the ratio is between 10:2 (lower bound) and 10:10 (upper bound).

HUMAN EXPOSURE AND MERCURY UPTAKE

The reference dose is the amount of mercury ingested daily over a lifetime that is anticipated to be without adverse health effects to humans, including sensitive sub-populations, for example, the developing fetus and young children. Of the total amount of mercury taken in by an individual (the delivered dose), the amount actually absorbed depends on the route of exposure and the form of mercury the person is exposed to. Potential exposure pathways include inhalation, dermal contact, and ingestion.

The U.S. EPA reference dose for methyl mercury is currently 0.1 micrograms per kilogram of body weight per day. The Agency for Toxic Substances and Disease Registry recommends a minimum risk level of 0.3 g/kg.bw/day. The World Health Organization has set a tolerable daily intake from all sources of 0.47 g/kg.bw/day for methyl mercury and 0.71 g/kg.bw/day for total mercury. The estimated intake for an adult Canadian of all forms of mercury via all routes is 7.7 g/day (0.11 g/kg.bw/day), equivalent to an absorbed dose of 5.3 g/day (0.076 g/kg.bw/day). Of these amounts, fish contribute about 27% of the intake and 40% of the absorbed dose. Drinking water and respiration of ambient air contribute very little to overall human exposure.

Aboriginal and certain subsistence populations may be at higher risk for methyl mercury exposure via their diet. Recreational anglers may also be at greater risk than the general population. Studies are ongoing to quantify the extent of human exposure.

FISH CONSUMPTION ADVISORIES

Since the fetus may be more susceptible than the mother to the adverse effects of methyl mercury, health experts in both Canada and the United States advise prudence to minimize the consumption of fish that have higher levels of methyl mercury. This advice covers both pregnant women and women of childbearing age. Authorities have issued the following advice regarding mercury levels in fish:

• The U.S. Food and Drug Administration (FDA) limit for methyl mercury in commercial fish is 1 part per million (1 mg/kg). U.S. FDA advises restricted consumption for pregnant women and women of child-bearing age, as well as of fish with higher mercury concentrations.

• Michigan currently has a statewide advisory for mercury, with a limit of 0.5 ppm:

­ No one should eat more than one meal per week of rock bass, perch, or crappie over 9 inches in length; or any size large- or small-mouth bass, walleye, northern pike, or muskie.

­ Women who are pregnant or nursing should not eat swordfish or shark.

• The Health Canada guideline for total mercury in the edible portion of fish is 0.5 ppm.

• The Ontario MOE guideline for mercury in fish flesh is <0.45 ppm for unrestricted consumption. Ontario MOE advises restricted consumption for children under age 15 and women of child-bearing age, and no consumption of fish with mercury levels above 1.57 ppm.

The Michigan 1999 Fish Advisory and the Guide to Eating Ontario Sport Fish 1999-2000 were consulted to ascertain for which fish that consumption advisories had been issued because of mercury contamination. Both books present information in regard to consumption by the general population and by women and children. The bases for advisories are not unambiguous in many cases, fish were tested for a group of contaminants and an advisory issued when one of the contaminants exceeded a criterion, but this does not mean that the fish necessarily contain all of the contaminants sought.

In general, the Michigan and Ontario books indicate unlimited consumption for smaller sizes of most fish species, increasingly restricted consumption for increasingly larger fish and, in one case (muskellunge), Michigan advises no consumption. Because of uncertainty whether Michigan and Ontario advisories are concerned with the same contaminants, the extent of agreement on an advisory for a given fish species is difficult to gauge.

Based on the information presented, the scenario assumes that, to protect human health of both the general population and women and children, increasingly restrictive consumption advisories are in place on Lake Hypo for increasingly larger members of various fish species which may include brown bullhead, carpsucker, large- and smallmouth bass, northern pike, muskellunge, walleye, white bass, and white perch.

It is assumed that, if mercury levels in fish were to increase significantly, consumption advisories would apply to smaller-sized fish and would be more restrictive for at-risk populations.

SUMMARY OF MERCURY INPUTS TO LAKE HYPO

Based on the above assumptions and modelling, the annual mercury inputs to Lake Hypo are:

• River: 328 kg

• Atmospheric background: 9 kg

• Total (without Acme): 337 kg

• Acme: 9 kg

The proposed source would increase mercury loading to Lake Hypo by 9 kg, or 3%. Assuming that an increase in mercury deposition to the water will translate to an increase in mercury concentration in fish by a ratio that can vary from 10:2 and 10:10, Acme's contribution of a 3% increase in the mercury load to the waters of Lake Hypo would translate thusly:

• 10:2 ratio: 3% water increasev0.6% fish increase.

• 10:10 ratio: 3% water increasev3% fish increase.

Recognizing the other assumptions outlined for this scenario, the 3% fish increase value is taken as the upper bound of the impact of the mercury contribution from Acme.

To gain a better understanding of the technical tools and procedures, questions and discussion centered on reference dose, risk assessment, and where harm occurs.

Reference Dose

1. How is a reference dose developed and a "safe" level established, for instance, contaminant intake through fish consumption?

U.S. EPA has established reference doses that are assumed to be protective for everyone, including sensitive sub-groups. The reference dose takes into account information such as the nature of adverse effects caused by the contaminant under consideration, as well as when a person is likely to be at greatest risk. It also takes into account such factors as cumulative impacts, time frame, and bioaccumulation. In the case of methyl mercury, the reference dose is based on studies of in utero exposure and testing for neurobehavioral effects as children mature. To take variations in human response into account, the reference dose incorporates a margin of safety determined, in part, by the knowledge base for the contaminant.

Risk Assessment

2. Are there set criteria that can be applied to determine whether a given contaminant load increase is or is not significant?

There are no set criteria and no clear line of demarcation. The question is not so much the percent contaminant increase in water or fish but, rather, what risk results from that release and is that risk acceptable. For example, the source could increase the mercury level in fish by 20% but, if a risk analysis shows that the subsequent risk to exposed humans is still acceptable, say, 1 in 1,000,000 then the Director could decide to approve the application, regardless of the percentage increase.

3. What factors go into a risk assessment to establish acceptable risk?

The risk assessment associated with Acme's contribution is basically a risk characterization describing the level of impact to the lake, the fish, and to humans consuming the fish. In Michigan, a risk assessment considers all routes of human exposure, the impacts of the contaminants on humans, fish, farmland, among other factors. The risk assessment would consider existing contaminant levels and the anticipated increase that the new source would contribute. Models and other tools would be applied to reach a reasoned decision.

If existing exposure already exceeds desirable levels and established limits expressed, for instance, in fish-consumption guidelines, the decision-maker must consider whether additional impact should be allowed through approval of another source. Among other subjective considerations are that many people do not read fish-consumption advisories, or are not aware of them, or choose not to follow them, so that their mercury uptake may approach or exceed the reference dose.

Water quality criteria and existing water quality are also part of the risk characterization and decision-making for an air permit. In this scenario, the concentration of mercury in the water column exceeds the U.S. GLI water quality criteria set to protect wildlife and human health, and Acme proposes to add more via the air pathway. Protection of wildlife and human health through compliance with water quality criteria contributes to the legal arguments to deal with air sources. However, to simplify the scenario, such compliance was not made a critical factor.

Present Michigan policy requires multi-pathway risk assessment only for municipal waste incinerators. Conduct for other sources is optional, with decisions made on a case-by-case basis with consideration to such factors as the nature of the contaminant, the amount to be emitted, and present environmental

When a proposed air permit has transboundary implications, jurisdictions on one side of the border endeavor to provide, as early as possible, information to their counterparts on the other side.

Relative Significance of Health Problems

4. Does the mercury emitted from a stack cause more health and environmental harm in the air or the water?

Work is under way to strengthen source-receptor relationships, as part of the evolution of the TMDL approach proscribed in the U.S. Clean Water Act. Through two projects at Devil's Lake, Wisconsin and the Florida Everglades U.S. EPA is constructing the science to relate mercury emissions to an endpoint within fish. Specifically, the projects will characterize the magnitude of local, regional, national, and global atmospheric emissions, then transport those elemental and chloride mercury emissions into the watershed, followed by transformation into the methyl mercury form. The models used are primarily those contained in the Mercury Study Report to Congress . Devil's Lake has a small watershed and no point sources; impact is entirely by the air pathway. After establishment of the scientific connection, atmospheric emissions might have to be regulated.

Studies in the Experimental Lakes area of northern Ontario are looking at atmospheric deposition of mercury and bioaccumulation through the food chain, as well as re-entry into the atmosphere and whether fish are more susceptible to mercury in the sediments. Different mercury isotopes are being used as tracers.

Case Study

LEGAL CONTEXT AND BACKGROUND

Under the division of powers in the United States, both the federal and state governments have authority for environmental protection, with primary authority generally residing with the federal government. Administration of particular federal laws and programs is often delegated to the states, subject to federal oversight. The states also have independent authority to enact more stringent environmental legislation to complement and extend federal laws, except in limited circumstances where state authority is pre-empted by federal regulation.
The U.S. regulatory regime is highly compartmentalized, with media-specific statutes. Taken together, the statutes comprise hundreds of pages and the accompanying regulations thousands. The key statutes are the Clean Air Act and the Clean Water Act, which deal with air and water, respectively, and the Resource Conservation and Recovery Act and the Comprehensive Environmental Response, Compensation, and Liability Act, which deal with land concerns. Since these statutes and subsequent amendments were adopted over a period of years, they reflect the thinking of Congress and the administration at the time of their adoption, resulting in different approaches to environmental problems. Insufficient connectivity among the statutes may contribute to difficulties in resolving issues such as the one outlined in the scenario.

All four acts confirm federal and state authorities, but the balance of authorities differs from one to the other and also for different programs within a particular act. In general, federal statutes set bounds, and the states seek approval to receive delegation to implement programs and requirements. In some instances, delegation agreements are quite prescriptive; others are broader in regard to program implementation. Each act also contains provision for citizen suits.

U.S. statutes incorporate both technology- and management-based approaches. Ambient air and water quality standards have been established at levels that would protect, inter alia , public health and the environment. As a point of departure to achieve these standards and, hence, a safe environment, most federal statutes impose technology-based requirements on sources, the efficacy of which is subsequently determined. To ensure that the ambient standards are met, mechanisms are in place to implement, as necessary, measures that go beyond technology-based controls.

The U.S. legal scenario draws upon U.S. federal and appropriate Michigan state legislation. The relevant provisions of the acts are reviewed below.

U.S. Clean Air Act

The Clean Air Act, enacted in 1970 with major revisions in 1977 and 1990, is both complex and proscriptive. Unlike other key federal environmental legislation, the states have first opportunity to set control approaches and strategies which, in turn, must be approved by the U.S. Environmental Protection Agency (EPA). If they are disapproved, U.S. EPA then sets the requirements.

The interplay between a technology-based approach and risk assessment helps ensure achievement of the objectives of the Clean Air Act. Under Section 112(F), after MACT standards are in place, residual risk must be considered. Within eight years of establishment of a MACT standard, U.S. EPA must demonstrate that there are not significant environmental or human health effects for the source categories to which the controls apply. If there are effects, then the MACT standards must be strengthened in order to reduce risk to acceptable levels.

U.S. EPA has finalized its methodology for performing residual risk assessments, which include multi-media and multi-pathway considerations. The risk assessment standards will consider benchmarks for acceptable and excessive health risk, with a grey or undefined area in between. The residual risk assessment tool has not yet been applied.

The Clean Air Act provides U.S. EPA with statutory authority to address impacts of air emissions on other media, such as soil and water. The Great Waters provisions, added to the Act in 1990, require U.S. EPA to monitor air deposition and conduct an assessment of the extent to which that deposition is responsible for environmental or human health effects or exceedences of water quality standards. The U.S. EPA Administrator is required to review all rules under Section 112 and determine whether the authorities are sufficient to mitigate contamination arising from air sources; if problems exist under existing authorities, the Administrator can make more stringent MACT standards under Section 112, applicable to either an individual source or to broad-based source categories. The Administrator is also required to recommend to Congress whether other acts need to be revised to address water quality problems arising from air deposition. U.S. EPA's most recent review has concluded that the existing Section 112 authority is adequate to address such concerns.

Mercury, one of the 188 substances on the hazardous air pollutant list, was particularly problematic to Congress when the Clean Air Act amendments were passed in 1990. Congress included additional provisions to further study the effect of mercury and to study mercury emissions from electrical utilities. In addition to MACT standards, Section 129 of the Act calls for special standards applicable to medical waste incinerators and municipal combustors which, together, are the largest U.S. sources of mercury to the atmosphere. Under Section 129, U.S. EPA establishes guidelines, which must be adopted by the states before they are federally enforceable. The National Academy of Sciences has undertaken a review of the mercury study, but has not yet prepared a final report.

Section 115 requires U.S. EPA to respond to any finding by an international body and rule whether the alleged harm needs to be addressed. Section 126 addresses state transboundary considerations.

U.S. Clean Water Act

A keystone to the U.S. Clean Water Act is the NPDES (National Pollutant Discharge Elimination System) permit program, under which all point source effluent dischargers are subject to permit limitations. The Act also incorporates technology- and management-based approaches. The technological perspective is manifested through such release limit requirements as new source performance standards, best practicable technology, and best available technology.

The Clean Water Act contains a prohibition regulation where an existing use is impaired, no further discharge of the contaminant contributing to the impairment is allowed. The concept of a total maximum daily load (TMDL) is a practical means to address contaminant input. It can also be used to regulate air sources in order to protect water quality. Included as part of the Act, a TMDL establishes the total allowable loading of a contaminant to a water system via all pathways without a resulting violation of water quality standards. Public consultation is also provided for in the Act.

The application of TMDLs is evolving from conventional pollutants, such as phosphorus, to persistent toxic substances, such as mercury. Some Great Lakes states have committed to TMDL development generally for smaller water bodies wholly within their boundaries. However, application of a TMDL to a Great Lake or the Great Lakes as a whole would be a significant program departure and has not been done yet.

In its proposed revisions to the TMDL rule, U.S. EPA would allow for offsets for both point and non-point sources. A new point source wishing to discharge a contaminant, or an existing source wishing to increase its discharge, could negotiate with another source point or non-point to decrease its contaminant contribution. That requirement would become an enforceable condition of the NPDES permit issued to the point source, providing enforceable limits, albeit indirectly, on a non-point source. This assumes, of course, that there are other point or non-point sources to offset new or existing sources.

TMDLs relate to water quality standards, which the states are responsible for setting. In the 1990 amendments to the Clean Water Act, Congress called for a Great Lakes Water Quality Initiative (GLI) to establish standards that protect the Great Lakes. Among the contaminants addressed were bioaccumulative chemicals of concern, one of which is mercury. To deal with these contaminants in a more stringent way, specific regulatory provisions include antidegradation requirements that deal with preserving existing water quality, how existing sources will decrease discharges, and how new sources will be permitted.

The Clean Water Act's Non-Point Source Program encourages pollutant load reductions from non-point sources. When Section 208 guidance was issued in the late 1970s, non-point sources were agriculture, silviculture, and storm water runoff. The impact of atmospheric deposition and inclusion of this pathway in the non-point source category is more recent.

Resource Conservation and Recovery Act

The Resource Conservation and Recovery Act establishes enforceable federal requirements with regard to waste management. States can apply for approval for state programs. The Act contains a provision allowing any citizen to sue an entity that handles, stores, transports, or discharges any solid or hazardous waste in such a manner that it causes or poses an imminent and substantial endangerment.

APPLICATION OF THE LEGAL FRAMEWORK TO THE SCENARIO

To provide opportunity to consider technical, legal, and policy issues, the U.S. legal scenario is constructed as a challenge to the operation of a new air source, Acme Widget. In the U.S., different types of sources may be subject to different regulatory limits, for example, emission limits stipulated in MACT rules. In order to provide an overall consideration of U.S. legal provisions, the nature of the source is not identified.

• Under Section 303(d) of the U.S. Clean Water Act, the state has included, and U.S. EPA has approved, the inclusion of Lake Hypo on its list as a water-quality-limited water body, because of fish-consumption advisories for mercury. Approval requires the state to include Lake Hypo in its planning under Section 303(e).

• Lake Hypo has been designated for use as a warmwater fishery, that is, a water body that contains fish species that thrive in relatively warm water. Warmwater fish species include bass, pike, and walleye.

• Lake Hypo is subject to state-issued mercury-based fish consumption advisories for bass, pike, walleye, and several other fish species. If the designated uses of a surface water are not attained because of mercury contamination, the state's antidegradation rule prohibits the lowering of water quality with respect to mercury. The rule applies to a new or increased loading of pollutants from any source to state surface waters for which independent regulatory authority exists requiring compliance with water quality standards.

Through a new source permitting process under the U.S. Clean Air Act, Acme Widget applied to the Air Quality Division of the Michigan Department of Environmental Quality for an air use permit for a new source to be located near Lake Hypo. Technical staff reviewed the application, with consideration of the technical information presented above, and applied MACT standards from facilities with comparable characteristics to ensure that the source complied with Clean Air Act requirements. The technical review included a multi-pathway risk assessment. After the conclusion of its review, a public notice was posted for a period of 30 days.

Because Acme would emit mercury to the atmosphere, a portion of which is predicted by modelling to be deposited into Lake Hypo, thereby increasing the total amount of mercury in the lake, a public hearing was requested and held. After considering all the information received, the Division Director or his delegate approved the request as submitted, concurring with his staff's recommendation for applicable conditions. He deemed as acceptable a maximum 3% increase in the level of mercury in fish, and concluded that the 0.1 µg/kg.bw/day reference dose for mercury would be met, even for those sub-populations whose consumption of fish from Lake Hypo would result in their approaching the limit. The division notified the applicant and the opponents of its decision.

The ENGO, Friends of Lake Hypo, then filed a suit against Acme Widget for potential violation of environmental protection laws, specifically, the discharge of pollutants without an NPDES permit, as required under the U.S. Clean Water Act. The suit also included provision to make other legal claims under state or federal law. A hearing was held on cross motions for a summary judgment, ² with each side presenting the technical and legal bases for their positions. These bases are presented below.

The technical scenario, above, presents a synopsis of the information that supports the permit. The technical scenario should be viewed as credible rather than exhaustive. The legal scenario assumes that expert witnesses have been called and expert testimony provided.

MOTION FOR SUMMARY JUDGMENT ³

Neil Kagan and Paul Benington

INTRODUCTION

Mercury A Contaminant of Concern

The presence of mercury in air and water has increased dramatically over the past century due to industrial activity in the Great Lakes region and elsewhere. Mercury poses a real not hypothetical danger. Those who eat contaminated fish especially subsistence populations, women of childbearing age, and children are at risk. Mercury is a potent neurotoxin and even at very low levels can cause subtle but permanent damage to the brain and to the central nervous system. For instance, infants exposed to elevated levels of methyl mercury can suffer from cerebral palsy, mental retardation, and delayed walking and speech. In wildlife, mercury can cause infertility, interfere with reproduction and development, and lead to behavioral impairments and death.

The Issue

The overarching issue in this case is whether existing laws designed to protect water quality provide the means to control atmospheric deposition of mercury traceable to a particular facility. The plaintiff, Friends of Lake Hypo will show that:

• Acme Widget may not operate its facility without getting a permit pursuant to the Clean Water Act.

• Acme may not get a water quality permit because mercury from its stack would exacerbate the already existing impairment of Lake Hypo's warm-water fishery and reduce the lake's water quality.

Friends is not pursuing a claim under the Resource Conservation and Recovery Act (RCRA). Under the agreed facts, all the mercury emitted from Acme's stack will be in a gaseous form, and RCRA applies only to solid waste.

Judgment Sought

Friends contends that Acme's proposed addition of mercury by the air pathway to Lake Hypo from its stack constitutes a pollutant discharge under the Clean Water Act. Therefore, Acme is obligated to obtain a Clean Water Act permit before discharging mercury into Lake Hypo. Friends seeks the following relief from the court under the Act:

Factual Background

Consider the following facts:

• The U.S. Environmental Protection Agency (EPA) has established the safe level for mercury in water at 1.8 ng/L for people and 1.3 ng/L for wildlife. The present concentration in Lake Hypo is 2 ng/L. Even without contributions attributable to Acme, mercury levels are unsafe for both people and wildlife.

• Mercury-based fish-consumption advisories are already in effect for bass, pike, walleye, and several other fish species found in Lake Hypo. Because of the advisories, Michigan has included Lake Hypo on its list of water-quality-impaired segments needing a TMDL (total maximum daily load) for mercury.

• Through its air permitting process, Michigan predicted that the mercury contribution from Acme would increase methyl mercury levels in the fish in Lake Hypo by 0.6 to 3%. Friends considers this significant fish-consumption advisories would apply to smaller-sized fish and would be more restrictive for at-risk populations.

THE CLEAN WATER ACT OBJECTIVES, GOALS, AND SCOPE

The Clean Water Act is a comprehensive water quality statute, with broad environmental and public health purposes and ambitious pollution control goals. Its provisions should be construed broadly to give the Act full effect. Friends contends that Acme's proposed addition of a toxic pollutant to Lake Hypo via the air pathway clearly falls within the domain of the Act. The mercury deposited to Lake Hypo via the air pathway can be traced back to the Acme stack, which Friends contends is a point source where technology-based controls can be applied and industry-tailored limitations effectively implemented and enforced to protect water quality.

Congressional Intent

The Act's stated objective is to restore and maintain the chemical, physical, and biological integrity of the waters of the United States. Congressional intent supporting this objective was expressed by the House Committee on Public Works in the broadest possible terms, "Any change induced by man which overtaxes the ability of nature to restore conditions to natural or original is an unacceptable perturbation." To accept Acme's request to further perturb Lake Hypo's already polluted waters would frustrate Congress' clearly stated intent.

In addition, Congress declared a national goal to eliminate the discharge of pollutants into navigable waters and to prohibit the discharge of toxic pollutants in toxic amounts. Further, Senator Edmund Muskie, the chief proponent of the Act, stated that the Act's goals "are not merely pious declarations that Congress so often makes in passing its laws. On the contrary, this is deliberately a life or death proposition for the nation."

Judicial Interpretation

In various judicial cases, courts have found that the Act was designed to regulate, to the fullest extent possible, those sources emitting pollution into rivers, streams, and lakes and that its provisions be liberally construed to achieve the Act's objectives. For example, the courts interpreted the definition of navigable waters to include tributaries and wetlands that are not navigable.

Congressional changes to the Clean Water Act in 1972 re-focused enforcement efforts on point sources of pollution. Prior to 1972, enforcement focused on ambient receiving water quality rather than on pollution sources. That was unworkable and ineffective in large part because it was difficult to link any particular polluter to a violation of an ambient water quality standard. To protect water quality, the 1972 changes aimed to enforce limitations wherever the addition of a pollutant could be linked to a point source through imposition of technology-based controls and industry-tailored limitations at the point source.

Friends contends that Congress intended that the 1972 amendments apply to air sources such as Acme. For example, a Senate Committee on Public Works report states that, "Where the Administrator can identify a direct link between a discharge source and water quality, the Administrator is authorized to tighten controls on the polluter." Since a direct link between Acme's proposed mercury discharge and Lake Hypo's water quality has been established, the Clean Water Act should be triggered.

Senator Muskie also stated, "Based on the history of consideration of this legislation, it is obvious that its provisions and the extent of application should be construed broadly." Friends of Lake Hypo and Acme both agree that the Clean Water Act is broadly remedial, and both accept that the call for the liberal interpretation of remedial statutes is consistent with the Act's purposes. Further, Friends and Acme both agree that some line must demarcate the limit of the Act's reach, but disagree about where the line is drawn and whether the line can move as advances in science and technology are achieved. The line currently includes pollutants added to navigable waters that are traceable to a point source.

The question is, on which side of the line does Acme's addition of a toxic pollutant to Lake Hypo fall?

THE CLEAN WATER ACT APPLICATION TO ACME WIDGET

Friends contends that Acme's stack is a point source under the Clean Water Act and that the mercury emitted from that stack is a discharge.

Is Acme's Stack a Point Source?

The Clean Water Act defines point source as any discernible, confined, and discrete conveyance, including but limited to a short list of such conveyances from which pollutants may be discharged. By stating "including but not limited to," Congress meant that its list of point sources was not exhaustive. With an open-ended definition, the courts have recognized that the term be construed expansively.

• In the Earth Sciences case, the court said the concept of point source was designed to further the statutory scheme establishing the permit requirement by embracing the broadest possible definition of any identifiable conveyance from which pollutants may enter the waters.

• In the Hecla Mining case, the court said the touchstone for defining a point source is the ability to identify a discrete facility from which pollutants have escaped.

Accordingly, courts have held a broad range of sources to be point sources, from piles of mined materials to sand to bulldozers to backhoes to pressurized sprays from irrigation systems and hoses. Acme contends that the term point source should not apply to its discernible, confined, and discrete conveyance. Its stack, 90 metres high and with an exit diameter of 2.5 metres, is easily identified as a discrete facility from which pollutants would be discharged into Lake Hypo and is, without question, a point source.

Acme claims that U.S. EPA has authoritatively determined atmospheric deposition to be a non-point source exempt from the permit requirement. Although the atmosphere is undeniably the route by which the mercury will reach Lake Hypo, Acme's stack, not the atmosphere, is the source. The indirect nature of the discharge through the atmosphere does not make it any less a discharge. EPA has never squarely addressed the issue this case presents, and has never said that an air source emitting a pollutant

• When EPA said that the proposed revisions to the TMDL regulations include atmospheric deposition as a non-point source, it meant that it will not allow the states to ignore pollution that comes from the atmosphere when establishing TMDLs. EPA did not mean that an identifiable source of pollution deposited into a specific water body is a non-point source simply because the pollution travels through the air.

• When EPA said in the proposed revisions to the antidegradation regulations that it would consider an offset from an air pollution source only where the air pollution source directly affects the water body, it said nothing to suggest that it considered such an air source to be a non-point source.

Acme contends that no other air source has ever been required to get a Clean Water Act permit. Friends contends that this argument has never been taken seriously by those who seek creative solutions to improve the human condition.

In the Clean Water Act, the term point source is very broad. Congress has enumerated express exceptions to its application, and the courts have held that making exceptions lies only with Congress. In NRDC v. Costle, U.S. EPA attempted to exempt discharges of pollutants from agricultural and silvicultural activities as well as discharges from certain animal confinement facilities, arguing that the number of permits from all such sources would be overwhelming. The court rejected EPA's argument, holding that EPA may not on its own authority exclude a point source from the permit program.

Acme argues a similar justification to exclude air sources. Although Congress has exempted certain sources from the definition of point source, it has not exempted air sources like Acme, which add a pollutant to a specific water body from a stack.

Regarding the administrative burden of requiring NPDES (National Pollutant Discharge Elimination System) permits for additional source categories, such as air sources, the courts have told U.S. EPA that, unless Congress changes the law, it has no choice.

Is the Mercury from Acme's Stack a Discharge?

The Clean Water Act defines discharge as any addition of any pollutant to navigable waters from any point source. The key word is "any," which is defined with terms like all, regardless of kind, without restriction or exception. Friends contends that "any addition" includes both direct and indirect discharges, and that Congress did not limit the Clean Water Act but chose to use "any addition" and not just any direct addition. Acme's projected addition of mercury to Lake Hypo by the air pathway has been established and is traceable to its point source stack.

Acme distinguishes between the Clean Water Act which addresses discharges into water, and the Clean Air Act, which addresses emissions into air. Friends claims that a discharge can include the emission of a pollutant indirectly deposited into a water body. The Clean Water Act defines discharge expansively and unconditionally to mean any addition of any pollutant to navigable waters from any point source. Each point source is thus subject to control, regardless of the pollutant route from the point source to the water and regardless of the medium of conveyance.

Senator Muskie explained Congress' intent in the Clean Water Act to prohibit pollution from both indirect and direct sources, that is, discharges. The courts have also interpreted discharge to include indirect discharges, rejecting the argument that pollutants must be discharged directly into navigable waters, upholding a conviction for violating the Clean Water Act where a concentrated sand-and-concrete mixture was sprayed through the air before it was deposited into navigable waters, and holding that spraying too much waste water on the fields, which then ran off into a nearby stream, was a point source discharge.

Friends contends that Acme cannot receive a discharge permit pursuant to Section 402(A) of the Clean Water Act. Under U.S. EPA's prohibition regulation, Michigan may not issue a Clean Water Act permit to a new source that will cause or contribute to a violation of any one of the three components of the water quality standard, that is:

• Designated uses of the navigable water body.

• Water quality criteria based on those designated uses.

• Anti-degradation policies and implementation procedures.

A violation of Lake Hypo's water quality criteria already exists because the lake's designation as a warm water fishery is not being attained, that is, fish in Lake Hypo are not safe to eat because of mercury contamination. That prompted Michigan to list the lake as a water-quality-limited segment in need of a TMDL. Friends contends that the mercury to be discharged from Acme's stack and deposited into Lake Hypo will translate into an increased mercury concentration in the lake's fish, thereby exacerbating the existing violation by making the fish even less safe to eat.

A state may permit a new source or discharge of a pollutant responsible for limiting water quality only if it has calculated a pollutant load allocation, a TMDL for the water body in question. The TMDL must contain an allocation for future growth and establish compliance schedules for current permit holders. If the state has not completed a TMDL, a new source or discharge cannot get a permit.

No TMDL has been developed for Lake Hypo. In the absence of a TMDL, U.S. EPA's prohibition regulation prohibits the issuance of an NPDES permit to Acme because its discharge will worsen the existing impairment to a designated use of the water body.

Acme's discharge should also be prohibited on a separate but related consideration. Mercury is present in the water column of Lake Hypo at a concentration which already exceeds the safe levels set out in the Great Lakes Water Quality Initiative (GLI) to protect people and wildlife. The deposition of additional mercury from Acme's stack will lower the lake's water quality even further, violating the anti-degradation policy for the lake.

Acme contends that its discharge will not lower Lake Hypo's water quality with respect to mercury. It argues that U.S. EPA approved Michigan's anti-degradation provisions, which provide that increased loadings of mercury do not constitute a lowering of water quality, so long as the loadings are within the authorized levels of a limit in an existing control document, and that its permit is a control document. Acme therefore concludes that the increased mercury loading predicted by the modelling conducted in the air permit proceeding does not constitute a lowering of Lake Hypo's water quality.

Friends notes that, although Michigan submitted anti-degradation regulations, U.S. EPA has neither reviewed nor approved them. Friends' position is that EPA should not approve the regulations because they are not consistent with the GLI antidegradation policy and U.S. EPA should not open a loophole for increased mercury loading. The GLI antidegradation policy prohibits a point source from discharging into a water body any additional amount of a pollutant that exceeds the minimum water quality standard.

Even if U.S. EPA reviews and approves Michigan's anti-degradation regulations, Friends contends that Acme's air permit is not a control document. A control document is any Michigan authorization to any source of pollutants to surface water. Acme's air permit is not an authorization to discharge pollutants to surface waters but to air. Even if the permit were a control document, it is not an existing control document, because it was not in effect when Michigan adopted its anti-degradation provisions in 1997.

Models have become indispensable to the courts and agencies in attempting to make complex decisions regarding complex systems. The courts have endorsed U.S. EPA's extensive reliance on computer models. In one case, petitioners claimed that U.S. EPA had relied on a model that had not been calibrated with real-life data, but the court upheld U.S. EPA's determination that the model was a satisfactory predictive tool on which to base its action.

Acme challenges the models used to predict that their discharge will degrade Lake Hypo's water quality and increase the mercury concentration in fish. Specifically, Acme challenges use of the dispersion model and the mercury methylation model, each on a single study. There is no indication, however, that either study has been peer reviewed or published. The evidence is too slim to dismiss the models used.

Further, the dispersion model used in this case has previously yielded results similar to those generated by U.S. EPA's industrial source ISC2 model and is thus particularly well suited to model the deposition of Acme's mercury emissions. In addition, the source operating parameters and the air transport distances are appropriate for both models. Thus, the predicted mercury deposition has a high degree of confidence.

Acme claims that actual mercury levels in rainfall are lower than predicted by models. Friends has documented data from U.S. EPA and major universities that mercury contaminates rain near Detroit as much as 65 times U.S. EPA's safe levels for mercury and, in more remote locations like Sleeping Bear Dunes, as much as 35 times. The need to clean the rain itself must precede the issuance of any permit that would allow Acme to add more mercury to an already stressed system.

CONCLUSION

In passing the Clean Water Act, Congress wisely crafted a law capable of reaching new point sources of pollution. In this case, the pollution travelling through the atmosphere has been traced from a particular source to a particular water body. The court must provide the following relief under the Act:

A judgment declaring that Section 301(A) prohibits Acme from discharging mercury from its stack into Lake Hypo without getting a permit pursuant to Section 402(A), and an order permanently enjoining Acme from operating its facility at least until it applies for and receives a Clean Water Act permit.

MOTION FOR SUMMARY JUDGMENT ⁴

Peter Holmes and Karen Schodowski

Friends of Lake Hypo's claims are contrary to law and should be rejected. Their claims are in effect an attack on the U.S. Environmental Protection Agency's (EPA) MACT (maximum achievable control technology) standards and the Michigan Department of Environmental Quality's (DEQ) permit, but Friends either opted not to challenge those standards or the permit, or made a challenge and failed.

AIR PERMIT CONDITIONS

Michigan DEQ issued an air permit to Acme after a full opportunity for public comment and after conducting a comprehensive multi-pathway risk assessment that demonstrated, despite overly conservative assumptions, that the emissions would not pose an unacceptable risk to public health. To meet the terms of the permit, Acme used state-of-the-art technology to reduce mercury emissions and comply with EPA's MACT standards under the Clean Air Act. MACT standards are set at the maximum degree of reduction in emissions, including a prohibition on those emissions where achievable, with consideration, inter alia , of a broad range of non-air quality health and environmental impacts, including the kind of impacts claimed by Friends. For a new source, the MACT standard can be no less than the emission control achieved in practice by the best controlled similar source.

MODELS

Assumptions and Limitations

An air dispersion model was used to estimate deposition to the lake, along with a second model to link mercury in the water column through the food chain to concentrations in fish tissue. Friends treats the model output as proven fact, but models can only predict or estimate, and the predictions in this case are scientifically flawed because of several key assumptions.

It is assumed that none of the mercury chloride emitted by Acme will be chemically transformed. The stipulated facts acknowledge that this assumption overestimates the amount of mercury deposited to Lake Hypo from Acme, but recognizes that not enough is currently known to quantify the extent of the overestimation. It could be as high as 5 or 10 fold.

The model unrealistically assumes that no other sources contribute a significant amount of mercury to Lake Hypo. This overstates the predicted percent increase of mercury in the lake attributable to Acme.

The model further assumes that all forms of mercury deposited in the lake will be transformed to methyl mercury at the same rate. In its 1997 Mercury Study Report to Congress , U.S. EPA states that there is a

The model unrealistically assumes "near continuous" operation of Acme's plant.

Regarding accumulation of mercury in fish tissue, the model assumes, as an upper bound, that all of the mercury deposited in the lake ends up in fish tissue and, as a lower bound, that only 20% ends up in fish tissue. This shows a large degree of scientific uncertainty. Also, Acme argues that there is inadequate scientific information to set the lower bound at 20%. In its review of the Mercury Report, U.S. EPA's Science Advisory Board stated that, "EPA's modeling of the post-deposition fate of mercury is over-simplified, neglects available information, and does not reflect recent advances in the science. The deficiency is one of the most serious problems in the report."

The Science Advisory Board also stated that it is not presently possible to establish a quantitative relationship between sources and actual exposures in humans or wildlife. Both the report and the review acknowledge that, because of various scientific deficiencies, the models used by U.S. EPA cannot predict wildlife criteria, which are based on water column concentrations of mercury, with less than an error factor of at least 100 to 1,000 times. The Science Advisory Board stated that the human and wildlife exposure models are so weakened that they have little predictive capability. U.S. EPA recognizes that it cannot support the existing human health criteria and has recommended an upward revision of the mercury criterion from 1.8 to 2.9 ng/L.

In its Mercury Report, U.S. EPA stated that there is "no validated air pollution model" to estimate wet and dry deposition of vapor phase compounds close to an emissions source, precisely what was attempted in this case. The ISC2 model is a preferred U.S. EPA model. U.S. EPA guidance states that if a modification is made to one of its preferred models, that model loses its preferred status and its use must be justified on a case-by-case basis. In this case, the Ontario Ministry of the Environment (MOE) model was run rather than the ISC2 model. That the two models produced "similar results" in a prior situation is insufficient. Thus, reliance on the Ontario MOE model is ill-founded and Friends' assertion of great confidence in the prediction that Acme's stack emissions would add up to 9 kg of mercury to Lake Hypo is refuted by the criticisms of the Science Advisory Board and by statements of U.S. EPA itself.

The model, which contains multiple over-estimations, predicts that Acme's stack emissions will increase mercury levels in fish tissue by less than 0.6 to 3%. The "less than" is important the stipulated facts acknowledge that Acme's mercury contribution is over-estimated. The true model output is less than 0.6% to less than 3%, but the science is inadequate to determine by how much. Therefore, for purposes of Friends' motion for summary judgment, the disputed facts must be construed against that motion, which must be judged against a standard of a less than 0.6% increase, not 3%.

Judicial Findings

The courts and, in particular, the U.S. Court of Appeals for the Sixth Circuit (which includes Michigan), have scrutinized the use of computer models and have held that use of models for purposes analogous to those claimed by Friends was arbitrary and capricious, specifically:

• In Ohio v. U.S. EPA, the Sixth Circuit rejected use of an unvalidated model to predict sulphur dioxide emission diffusion at two power plants.

• In an earlier related case, the Sixth Circuit concluded that U.S. EPA had not validated or empirically tested the model at sites and circumstances where lakeshore effects on dispersion were significant but unknown.

• In Columbus and Southern Ohio Electric v. Costle, the Sixth Circuit upheld computer modelling to determine whether a single county was in attainment with the Clean Air Act standards but, in a predecessor case, Cincinnati Gas and Electric v. U.S. EPA, struck down the use of a U.S. EPA dispersion model to make such a designation.

Friends has no scientific basis to claim an increase in mercury levels in fish tissue. A 0.6% or even a 3% change cannot be determined with sufficient precision. An Australian government laboratory study by Lewai and published in Analyst demonstrated that at a 1 ppm mercury level in fish tissue, measurements had only a 5% precision and, at 0.1 ppm, only 8 to 9% precision. This is insufficient to even confirm whether a predicted increase of 3% has occurred, must less a predicted increase of less than 0.6%.

ACME'S STACK EMISSION IS NOT A POINT SOURCE DISCHARGE

Friends argues that Acme's mercury emissions into the air, in compliance with the Clean Air Act permit, constitute a point-source discharge into the water and, thus, are also subject to an NPDES (National Pollutant Discharge Elimination System) permit under the Clean Water Act, whose requirements would apply at the identical point at which the mercury is emitted from the stack. It is highly unlikely that Congress ever intended this. Friends' argument is contradicted by the language and U.S. EPA's authoritative interpretation of the Clean Air and Clean Water Acts, by nearly 30 years of permitting history under the two statutes, and in the only reported case. Clearly, NPDES jurisdiction does not extend beyond a discharge to a receiving water to include the emission of gases to ambient air. Acme's emissions to the air do not constitute a point-source discharge to water.

The Clean Air Act and the Clean Water Act Their Interpretation and Application

U.S. EPA has consistently interpreted the Clean Water Act to classify atmospheric deposition as a non-point source. Under U.S. Supreme Court precedent, starting with the Chevron case in 1984, judicial deference must be granted to U.S. EPA's interpretation. That interpretation need not even be the interpretation that the court would adopt, only that it be reasonable.

That atmospheric deposition is a non-point source is stated in U.S. EPA's Non-Point Source Guidance in 1987, the proposed Great Lakes Water Quality Initiative (GLI) in 1993, the final GLI in 1995, and the draft Lakewide Management Plan for Lake Michigan in 1995. U.S. EPA has codified its current interpretation in the proposed August 29, 1999 revisions to the NPDES program, specifically the federal anti-degradation policy and the total maximum daily load (TMDL) regulations. In its proposed anti-degradation policy, U.S. EPA states, "These proposed revisions explicitly include atmospheric deposition as a non-point source of pollutants." Friends claims that U.S. EPA meant to say that, as long as air deposition from a particular source cannot be modelled, atmospheric deposition is a non-point source. U.S. EPA has never articulated such an interpretation.

Actions by U.S. EPA and the states are consistent with the determination that air emission sources do not equal point-source discharges. Clearly, U.S. EPA distinguishes between an air emission source subject to the Clean Air Act, and a water discharge subject to an NPDES permit. In its proposed anti-degradation policy, because air deposition contributes to some of today's existing water quality problems, U.S. EPA is considering a possible offset from an air pollution source that directly affects a water body in the vicinity of a new or expanded discharge. This proposal is in the context of the companion TMDL rule that interprets atmospheric deposition as a non-point source.

Among the tens of thousands of NPDES permits issued by U.S. EPA and the states, not one has ever been applied to an air emission source. Not one of the more than 50 Clean Water Act effluent limitation guidelines issued by U.S. EPA has ever been applied to an air emission source. U.S. EPA's recent cluster rule embodying both water and air regulations for the pulp and paper industry deal with air deposition issues under the air portion of the rule, not under the NPDES permit program, again demonstrating that air deposition is the province of the regulatory scheme under the Clean Air Act.

Even Friends admits that issuance of an NPDES permit for an air emission has never been done in the nearly 30-year history of the Clean Water Act. Contrary to Friends' argument, the reach of the NPDES program does have limits. In two court cases, the National Wildlife Federation argued that dams

The courts have stated their duty to harmonize potentially conflicting statutes wherever possible. Consistent with that duty, the U.S. Supreme Court, in Train v. Colorado Public Interest Research Group, limited application of the NPDES program by excluding discharges of radioactive materials, which are regulated under the Atomic Energy Act, stating that a clear expression of contrary legislative intent would be necessary to recognize a substantial policy change that would alter the "pervasive regulatory scheme" established in the Atomic Energy Act. That case, like this one, involved the Clean Water Act's broad definition of pollutant. Despite the absence of language in the Clean Water Act to expressly exclude radioactive materials, the court relied on U.S. EPA's interpretation and the legislative history that supported the exclusion.

Congressional Intent

There is no Congressional expression of intent to undermine the pervasive regulatory scheme of the Clean Air Act by so drastically expanding authority under the Clean Water Act. Congress clearly made regulation of air emissions the province of the former and regulation of surface water discharges the province of the latter. The Clean Water Act, which followed the Clean Air Act by scarcely two years, contains many provisions that are expressly patterned after its predecessor. Therefore, the language used in the two statutes is significant. The Clean Air Act defines an air pollutant as a substance "emitted" into ambient air. The Clean Water Act definition of pollutant borrows from the Clean Air Act definition, but refers to materials "discharged" into water, distinguishing between air emissions and water discharges. Congress continued the distinction between emissions and discharges in the Comprehensive Environmental Response, Compensation, and Liability Act, which defines a release using, among other terms, both emitted and discharged.

Yet another illustration of Congress' differentiation between emissions under the Clean Air Act and discharges under the Clean Water Act is contained in the Senate report on the 1972 Clean Water Act. That report states, "The Clean Air Act amendments of 1970 provided for a comprehensive attack on air pollution that included new source performance standards and controls over the emission of toxic substances. This bill would provide for new source performance standards and controls over the discharge of toxic substances." (emphasis added)

When Congress considered the precise issue at hand air emissions that may affect water quality through atmospheric deposition it chose to regulate through the Clean Air Act, not the Clean Water Act. Section 103 of the Clean Air Act requires U.S. EPA to perform a research program to evaluate the effects of air pollution on water quality. Title IV of the Clean Air Act establishes an entire regulatory program to address the effects of acid rain, air emissions of sulphur and nitrogen oxides on surface water and other resources.

Most significantly, when Congress established the Great Waters Program in 1990, it chose the Clean Air Act as its vehicle. Section 112(M) directed U.S. EPA to evaluate the effects from air deposition on water quality of the Great Lakes and other great waters. It also required U.S. EPA to issue further emission standards and control measures as appropriate under Section 112, not under the NPDES program or the Clean Water Act, to prevent such effects. Section 112 also established a technology-based permit program for emission sources of hazardous air pollutants that is expressly patterned after the NPDES permit program for toxic effluent discharges. The fact that Congress never once cited the NPDES program speaks volumes.

In March 1998, U.S. EPA expressly determined that the legal authorities in Section 112 are adequate to address the effects of atmospheric deposition on the Great Lakes and other great waters.

Friends also argues that a stack is a point source under the Clean Water Act. Even though the term

Friends argues against the overwhelming weight of this authority, claiming that the Clean Water Act is a broad remedial statute and should be broadly interpreted. However, even broad remedial statutes have their limits. The courts, and the Sixth Circuit in particular in the Cordova Chemical company of Michigan case, have warned against using the broad remedial statute mantra as a way to extend the scope of a statute's jurisdiction indefinitely to cover virtually every conceivable situation. The limits of the Clean Water Act are exceeded when Friends argues that an emission into the ambient air, in compliance with a Clean Air Act permit, constitutes a point-source discharge into surface water requiring a duplicative Clean Water Act permit.

Judicial Case Law

In the only reported decision on this issue, Chemical Weapons Working Group v. Department of Army, the court rejected plaintiff's logic. The emissions involved chemical warfare agents whose discharges into surface waters is absolutely barred by Clean Water Act Section 301(F). That section, as Friends concedes, is not applicable here. However, the court also rejected plaintiff's argument on grounds that are applicable here and are not unique to section to 301(F). It held that application of NPDES requirements to air emissions would be "irrational" and would have imposed the Clean Water Act discharge requirements on countless emission sources, including cars and chimneys.

The courts have clearly emphasized that the water quality effects of atmospheric deposition are expressly considered and regulated under the Clean Air Act, relying on the common sense view that a stack emits pollutants into the air, not into the water. The courts have also rejected attempts to draw an analogy with point-source discharges to groundwater that is hydrologically connected to a surface water body. Case law is evenly split on whether the NPDES program applies in such circumstances. Clearly, NPDES jurisdiction does not extend to emissions to the air, as opposed to discharges to water.

In summary, the jurisdictional scope of the Clean Water Act is clearly limited and not applicable to emissions from Acme's stack. Further, there is no precedent for two different media-specific regulatory regimes to apply permit requirements to the same source, in this case, emissions from Acme's stack. Not only would such a drastic result require clear expressions of Congressional intent, but Congress has, in fact, expressed contrary intent. In addition, Michigan DEQ performed a multi-pathway risk assessment and concluded that the air emissions would not pose any unacceptable risk. Even if there were an unacceptable risk, manifested through contaminant uptake by fish, the jurisdictional limitations of the Clean Water Act would not authorize imposition of NPDES permit requirements.

THE COURT'S JURISDICTION OVER NPDES PERMIT ISSUANCE

Federal courts have jurisdiction under the Clean Water Act to entertain citizens suits to determine whether Acme is subject to NPDES permitting requirements and to require Acme to get a permit. However, the court lacks jurisdiction to enforce water quality standards that are not incorporated into a permit. The Supreme Court and other courts have held that water quality standards are enforceable only when incorporated into a permit. When Friends seeks a declaration and a permanent injunction against issuance of an NPDES permit, arguing that Acme's operations will violate water quality standards, they are asking the court to enforce water quality standards not incorporated into a permit.

Even if the court concluded that it had jurisdiction to enforce water quality standards outside the context of a permit, it should refrain from exercising that jurisdiction on prudential grounds. A court should so refrain if litigation presents issues that would be better resolved by an administrative agency. In a similar case, a citizens suit brought to enforce water quality standards, the court held that it lacked

Acme urges adopting the same procedure. Friends is trying to preempt the permitting process and deny Acme the opportunity to demonstrate that it can qualify for an NPDES permit, if the court determines that one is required. If the court dismisses Friends' claim and allows the permitting process to go forward, Friends would not be denied judicial review or be prejudiced in any way. It could fully participate in the permitting process and, if a permit were issued, it could also challenge the permit in the context of a petition for review by the court.

ANTIDEGRADATION

Friends has not established that Acme's operations would violate water quality standards under any circumstance. Acme contends that Michigan's anti-degradation rules are consistent with the GLI. Moreover, Acme's predicted air deposition does not constitute a "lowering of water quality" under those rules, which exclude increased loadings "within the authorized limit in any control document," such as a Michigan DEQ-issued air permit. In this case, Michigan has considered the impact to surface waters prior to issuing the air permit, and this exemption is appropriate. Thus, the predicted loadings would not violate the antidegradation rules.

Alternatively, Acme should have the opportunity to engage in the TMDL process itself. Michigan could identify potential reductions in point and non-point sources contributing mercury to Lake Hypo.

MERITS OF CLAIM UNDER THE RESOURCE CONSERVATION AND RECOVERY ACT

Since Friends is not pursuing a claim under the Resource Conservation and Recovery Act, Acme has not presented its case relevant to that statute. Acme's position is detailed in its written submission (Appendix E).

CONCLUSION

Due to scientific uncertainties surrounding the modelling of the increased mercury loadings allegedly attributable to Acme, it would be arbitrary and capricious for the state to deny a permit based upon the modelling. If a permit is denied, Acme is entitled to have the opportunity to argue before the Court of Appeals that such a denial is arbitrary and capricious.

REBUTTAL

Neil Kagan

Acme contends that Friends are bringing a collateral attack on the issuance of the air permit. The goals of the Clean Air Act and the issuance of an air permit, in this case, do not satisfy the goals of the Clean Water Act and the protection of water quality that would be achieved through an NPDES permit. In many situations more than one statute applies to a given environmental impact, for example, the Clean Water Act and the Resource Conservation and Recovery Act, or the Federal Power Act and the Clean Water Act. There is nothing inherently wrong or improper about saying that a source that emits into the air and regulated under the Clean Air Act also requires a Clean Water Act permit.

Acme contends that the U.S. Environmental Protection Agency (EPA) has interpreted atmospheric deposition to be a non-point source. A non-point source by definition is a diffuse source of pollution to a particular water body. That is not the case here, and U.S. EPA's interpretation is not relevant. U.S. EPA has never addressed a specific facility emitting pollutants into the air that are then traceable to a particular water body. U.S. EPA has never interpreted the Clean Water Act or a non-point source to mean deposition of pollution from a particular source because it is always understood that atmospheric deposition is a reference to diffuse pollution.

Acme has studiously avoided discussing the definition of the terms point source and discharge, because it is inescapable that the stack is an identifiable point source, and the modelling predicts that mercury discharged from the stack will be deposited into Lake Hypo. Friends contends that this sufficient to meet the requirements, under the Clean Water Act, of a point source.

In the Chemical Weapons case, the court rejected the claim that an emission from a stack had to be regulated under the Clean Water Act. That claim was quite different and is inapplicable here. It considered a generalized allegation that emissions from the stack would be deposited in water somewhere, and no connection had been established between the stack to any particular water body. In Acme's case, there is an identified pathway from their stack to Lake Hypo.

Acme argues that, if the court were to find the stack is a point source, then countless other sources would require regulation. Friends contends that Congress' intention was to regulate such sources under the Clean Water Act where pollution from a particular source can be traced to a particular water body. It would not apply to automobile exhausts, for example, where deposition to a particular water body cannot be established. There must be a specific connection.

The Great Waters Program is applicable to the Great Lakes, Lake Champlain, Chesapeake Bay, and coastal waters, but not Lake Hypo, and nothing in that program covers the situation in this case. The Great Waters program was meant to complement, not supplant the Clean Water Act. The Act requires U.S. EPA to advise Congress about whether changes or revisions must be made to other laws that might bear on the issue of atmospheric deposition. The Clean Water Act clearly applies to Acme a pollutant is emitted and falls into the water and, therefore, the stack is a discharge.

MODELS

The technical scenario was designed to provide a factual basis to explore legal tools. The use of models and the validity of their predictions are key factors. Acme and Friends agreed on what the models predicted, but Acme strongly challenged the validity of the model's predictions.

1. Was Acme attacking the models or the predictions?

Both. The modelling in the scenario was overly conservative. In addition, as both the U.S. Environmental Protection Agency (EPA) in its Mercury Study Report to Congress and the Science Advisory Board in its review of that report point out, there is considerable scientific uncertainty and no validated model to predict wet and dry deposition of vapor-phase compounds to a nearby water body. Therefore, one cannot rely on predictions with any degree of precision.

2. Can the model be validated, that is, its predictions agree with measured real-world conditions?

Given the current state of monitoring, it is unlikely that a study can be designed to validate a multi-component transport, deposition, and fate model such as the one used in the scenario. The scenario is actually composed of several smaller models. These components can be validated and one can obtain a quantifiable but large uncertainty to the scenario's overall prediction. Real-world variability accounts for part of the uncertainty. Even if the model is not wrong and the measurements are accurate, it is unlikely that a 3% increase in mercury in fish can be measured because of environmental variability. One can either reject the model and its predictions or follow a precautionary approach and take the worst case the model predicts. However, one must be aware of the implications of adopting either extreme as a policy choice. Further, it is not appropriate for a court to make that choice.

3. What bearing do the limitations of modelling have on a judicial decision?

Questions about the validity of a model and its predictions cannot be ignored. The court's confidence or lack thereof will be significant and may be the determining factor in viewing the case in its entirety. The Acme scenario has many simplifying assumptions. In an actual case, the court would probably ask for more expert testimony regarding what to attribute as important to the model.

4. In Ontario, the proponent would use the model to support its application. Who bears the onus regarding models in the United States?

Michigan only models small sources. For larger facilities, the state asks the applicant to supply the model, perform a demonstration, and submit the results. The Michigan Department of Environmental Quality's permitting group checks to ensure that the parameters are correct, and consults with their modelling counterparts to ensure that the modelling was proper. For toxic substances such as mercury, efforts are made to ensure that such aspects as deposition to water and land are covered.

For a federal air regulation such as PSD (prevention of significant deterioration), the applicant submits the model and the output. However, mercury is not a pollutant under PSD but is covered through other mechanisms such as MACT (maximum achievable control technology) standards or risk analysis.

It was not Acme's responsibility to identify a superior model. Acme contends that there is inadequate scientific information to make an accurate prediction and then base regulatory decisions on that prediction.

6. Can an example be provided where a model's prediction for mercury is or is not validated by actual environmental conditions?

Mr. Dennis Leonard provided the following illustration. The model used in U.S. EPA's Mercury Report predicted that annual atmospheric mercury concentrations would be lowest in the desert area of the U.S. southwest and highest in the industrial heartland. However, using monitoring data from the U.S. Geological Survey's (USGS) National Atmospheric Deposition Program, Mr. Leonard contended that:

• The amount of mercury deposited from global and natural sources is more than what the model predicted.

• There do not appear to be mercury deposition hotspots in the industrial heartland.

• There is no increasing west-to-east gradient, as predicted by the model.

He concluded that the model does not predict the extent that natural background sources are responsible for mercury deposition, over-predicts contributions from sources such as Acme, and shows wrong dispersion trends.

Others, however, questioned the number and geographical distribution of monitoring sites in the USGS network, for instance, only three in the west, few if any in the heartland, and none in Michigan. This could influence the perception of actual environmental conditions. Others questioned the selective use of data from only one monitoring network. Other real-world data are available that are consistent with the model's predictions.

Mr. Leonard acknowledged that the daily concentration of mercury can be quite variable. For that reason, he had presented annual flux data, because bioaccumulation has a time frame of months and years. The issue is how mercury ends up in the lake and enters and moves through the food web. The true test of a model is its ability to make regional predictions and predictions at sites not influenced by localized sources.

TOTAL MAXIMUM DAILY LOAD MANAGEMENT IMPLICATIONS FOR PERSISTENT TOXIC SUBSTANCES

7. The TMDL (total maximum daily load) concept has been advanced for water bodies where effluent limitations required by the Clean Water Act are not adequate to achieve applicable water quality standards. Please describe the concept, its evolution since introduction as part of the Act in 1977, and its present and projected application.

Section 303 of the Act requires states to identify and establish a priority ranking for water bodies for which effluent limitations are not adequate and, for pollutants identified by the U.S. EPA Administrator, establish a TMDL which recognizes seasonal variation and reflects a reasonable margin of safety.

The TMDL concept was initially applied almost exclusively to conventional, largely waterborne pollutants such as nitrates and phosphorus. The process has evolved to now include several persistent toxic substances, among them mercury and dioxin. States have also enlarged their listing of impaired waters to include several where the presence of mercury in fish has resulted in consumption advisories, and they have made commitments to develop appropriate TMDLs. In many cases, atmospheric transport and deposition contributes a significant amount of the contamination. Thus, the atmospheric pathway is a critical element in some TMDL calculations.

The direct impact, if any, of the TMDL concept on persistent toxic substance management of the Great Lakes is difficult to determine. Because Lake Hypo is under a fish-consumption advisory due to mercury contamination, the TMDL requirement would have been applicable were the lake totally within one U.S. state. However, the Great Lakes are under multi-state and/or binational jurisdiction, suggesting further U.S. federal action as well as Canadian commitment before the TMDL concept could be applied. Nevertheless, some interplay involving the techniques applied to TMDL determinations and the evolution of Lakewide Management Plans may occur.

ALTERNATIVE LEGAL MECHANISMS

8. To establish that an air emission from a stack is a point source discharge and, therefore, requires an NPDES (National Pollutant Discharge Elimination System) permit under the Clean Water Act appears rather tortuous. Are there legal alternatives?

The legal landscape presented earlier identified and described three regulatory provisions and tools that could deal with the issue more globally and more efficiently: MACT standards for air sources and associated residual risk assessment, the Great Waters Program under the Clean Air Act, and the TMDL concept under the Clean Water Act. Additional mechanisms include:

• Preparation of a state environmental impact statement (EIS).

• Michigan, unlike many other states, does not require an EIS but does allow, in certain circumstances, for plaintiff to argue, under the Michigan Environmental Protection Act, that a particular action is polluting, impairing, or destroying a natural resource of the state.

• The principle of ARAPs applicable or relevant and appropriate requirements could be factored into the decision-making process for air sources. ARAPs are used in Superfund to decide clean-ups. Clean-up standards, set by the Agency for Toxic Substances and Disease Registry, use a reference dose. Therefore, one need not be concerned about discrepancies among fish-consumption advisories.

• Lawsuit against Michigan for issuing a permit without applying the Michigan Environmental Protection Act.

9. This scenario is one permit from one company for one contaminant. However, there are hundreds or thousands of companies. How do we as a society approach this reality?

A case-by-case approach is inefficient. Application of regulatory alternatives, identified above, may be preferable. In addition, flexibility rather than strict command and control would provide opportunity for innovation and inventiveness and would allow the market to determine where necessary reductions could be made most economically and efficiently. Compliance assistance is another complementary alternative. Michigan is considering, for example, 100% compliance for medical and municipal waste incinerators, with provision of major assistance in the period leading up to the compliance date and with vigorous enforcement after.

LEGAL AND PRACTICAL CONSIDERATIONS

10. Do the Clean Water Act and the Clean Air Act provide the same permit requirements on sources?

The Clean Water Act has a blanket prohibition against point-source discharges without a permit. The permit is like a gate-keeping function. In order to lawfully discharge, a point source requires an NPDES

11. Does the Clean Water Act contain any requirements for non-point sources?

When Congress wrote the Clean Water Act, it specifically chose not to include enforceable requirements, certainly no permit requirements, on non-point sources, because (1) it is more difficult to devise a regime to address non-point sources and (2) politically, non-point source control moves into the area of land-use control which, in the United States, is a zealously guarded local responsibility rather than within the domain of the federal government. Rather than directly regulate non-point sources under the Act, Congress encourages the states to act.

12. The term "discrete conveyance" is part of the definition of point source. At what point does a discrete conveyance become a diffuse conveyance? When the source is one mile, ten miles, 100 miles from the receiving water body?

Friends contends that "discrete conveyance" is not limited to a particular way that the contaminant reaches the water, nor that Congress intended such a limitation. Acme counters that the impact of air emissions on a surface water some distance away no longer qualifies as a discrete conveyance. Analysis must be on a case-by-case basis. For this scenario, a traceable connection shows that the emission from Acme's stack is predicted to have an impact on Lake Hypo. Just because there is not a discrete pipe directly into the water may not mean that the Clean Water Act does not apply.

Case Study

Under the division of powers in Canada, the environment is largely within provincial jurisdiction. Therefore, the legal scenario is structured under Ontario provincial law. In Ontario, different types of pollution sources fall under different legislative provisions. Whether a source will discharge contaminants to air or to water, or will be classified as a waste facility, and whether the proponent is a public- or a private-sector enterprise will determine the particular process requirements, rights of the parties, and criteria for decision-making on an approval application. In order to provide a realistic view of the legal provisions, the nature of the source is not identified but is assumed to be a new source of contaminants to the air, unrelated to a waste facility.

THE LEGAL LANDSCAPE IN ONTARIO

Environmental protection in Ontario is grounded in the Ontario Environmental Protection Act and the Ontario Water Resources Act, each with associated regulations. Together, sources within Ontario's boundaries are addressed through a system of permits and approvals. Primary responsibility lies with the Ontario Ministry of the Environment (MOE). Ontario MOE's business is environmental protection clean air, water and land. Its vision is a healthy and natural environment in which water, energy and material resources are conserved and wisely used. To fulfill this vision, the Ministry sets and enforces policies and standards, assesses compliance, and develops and encourages partnerships.

Certificates of Approval

One key instrument used to protect the environment and human health is the approvals process. Under the Ontario Environmental Protection Act, a Certificate of Approval is required for the construction or installation of equipment that discharges any contaminant to the atmosphere, and also for process changes that are anticipated to result in a change in the quantity or quality of emissions. Ontario MOE issues Certificates, each of which contains conditions. It can amend an existing Certificate and impose more stringent conditions. The conditions reflect the requirements of the Act, specifically Section 9, which reads:

"No person shall, except under and in accordance with a certificate of approval issued by the

[Ontario MOE] Director,

(a) construct, alter, extend or replace any plant, structure, equipment, apparatus, mechanism or thing that may discharge or from which may be discharged a contaminant into any part of the natural environment other than water; or

(b) alter a process or rate of production with the result that a contaminant may be discharged into any part of the natural environment other than water or the rate or manner of discharge of a contaminant into any part of the natural environment other than water may be altered."

"Despite any other provision of this Act or the regulations, no person shall discharge a

contaminant or cause or permit the discharge of a contaminant into the natural environment

that causes or is likely to cause an adverse effect."

An adverse effect is broadly defined as impairment of the natural environment; injury or damage to property, plants or animals; material discomforts and loss of enjoyment. Section 14 also provides for enforcement.

Regulation 346, developed under the Ontario Environmental Protection Act, deals specifically with air contaminants. Through Regulation 346, limits for specific contaminants are set as a concentration at the point of impingement. Dispersion models are used and aggregate emissions accounted for.
The approvals process is being streamlined. Among the reforms are:
• Exemption regulations, effective September 30, 1998, provide unconditional exemption from a Certificate of Approval for certain environmentally insignificant activities. Virtually every emission to the environment from a fixed source requires approval but many, such as vents, are insignificant.

• Standardized approvals regulations, presently under development, will focus on activities with predictable, easily controlled impacts and will provide a conditional exemption from a Certificate with regulatory requirements.

Other Legal Provisions

The Ontario Water Resources Act contains a general provision against discharge without a permit of any substance that may impair water quality. A source requires a Certificate of Approval and, if the source were in a sector covered by the MISA (Municipal-Industrial Strategy for Abatement) regulations, would have to meet those requirements as well.

The Ontario Environmental Protection Act and the Ontario Water Resources Act contain provisions in addition to a Certificate of Approval process. For example, to prevent adverse effects, the Director can issue an order to an existing source to reduce emissions. This is useful in cases where adverse effects are being monitored and action needs to be taken. To control any or all emissions, Ontario MOE has extensive regulation-making powers; the MISA regulations for water discharges are an example.

The Environmental Bill of Rights adds checks and balances. The Bill of Rights required Ontario MOE to develop a Statement of Environmental Values. The Statement includes the proposition that any new legislation and subsequent approvals be based on an ecosystem approach. The Statement has been filed with the Environmental Commissioner, an independent watchdog.

Under the Environmental Bill of Rights, citizens of Ontario can take a direct role. They can make specific requests to the Minister of the Environment to investigate an existing pollution problem, for instance, a discharge that appears to be unregulated. Citizens can also ask the Minister to review existing legislation, regulations, or policies and amend, revoke, or replace them with better ones. This tool requires a written submission by two or more citizens. It has, as yet, not been used extensively, but this is expected to change as people become more acquainted with it.

Prior to the Environmental Bill of Rights, only the recipient of a Certificate of Approval could appeal its provisions. Now citizens of Ontario may apply to the Ontario Environmental Appeal Board for leave to appeal. The threshold for granting permission is whether the decision to issue the Certificate was, for example, one that no reasonable person, having regard to relevant law and policy, would have made, and the decision would create the risk of significant environmental harm. This creates a checks-and-balance situation, since the applicant can appeal the Director's decision on the one hand, and others can seek leave to appeal on the other. The overall process has thus changed and, as it develops momentum, could yield further improvements.

Overview

Under Section 9 of the Ontario Environmental Protection Act, a new source applies to the Director for a Certificate of Approval. Under the Environmental Bill of Rights, Ontario MOE posts notice of the application for a minimum of 30 days for public comment. Simultaneously, the proposal undergoes technical review by Ontario MOE personnel. Regulation 346 is one of the tools used. After completion of the technical review and the close of the public comment period, the Director considers all information received and decides whether to approve the request as submitted or with conditions and issue a Certificate of Approval or to deny the request.

After the Director notifies the proponent of his decision, the proponent has 15 days from the date of receipt to appeal the decision. A decision to appeal triggers an Ontario Environmental Appeal Board hearing, to which the Director and the proponent are automatic parties. The Board must grant status for others, such as an environmental non-government organization (ENGO), to intervene.

Under the Environmental Bill of Rights, the Director's decision is also posted. An ENGO or member of the public has 15 days from the date of posting to ask the Board for leave to appeal the Director's decision. Leave can only be granted in very limited circumstances:

• Where a person has an "interest" in the decision, AND

• There is good reason to believe that "no reasonable person, having regard to the relevant law and to any government policies developed to guide decisions of that kind, could have made the decision," AND

• The decision could result in significant harm to the environment.

Generally, the Board will grant intervener status to people who may be affected by the proponent's activities and to ENGOs that can demonstrate a track record of involvement in the environmental issues raised by the proponent's application. However, only the Director and the applicant are guaranteed a place at the table.

Usually an Ontario Environmental Appeal Board hearing is before a single member. The Board has the same authority as the Director and sits in place of the Director, whose decision is under appeal. Based on the evidence and submissions heard from the parties, the Board decides whether to impose, modify, or revoke the condition, or to take a completely different approach to the issue. A Board decision can be appealed to the divisional court on a question of law, and to the Minister of the Environment on policy-related issues.

Acme Widget

For this scenario, the Ontario MOE Director has issued a Certificate of Approval to Acme Widget, a new source located near Lake Hypo. The Director considered not only the concentration of mercury predicted by the Regulation 346 model but also the subsequent uptake and bioaccumulation in fish. The Certificate acknowledges the emission of mercury from Acme and its subsequent deposition to Lake Hypo. Because mercury is a persistent bioaccumulative toxic substance, the technical review process provided the necessary close scrutiny. To protect human health from the adverse consequences of mercury deposition to the lake and subsequent uptake and bioaccumulation in fish, the Certificate contains a condition requiring Acme to meet a limit more stringent than the standard in Regulation 346. The condition is:

The operator shall ensure that the mercury emission rate does not exceed 11 grams per hour.

The Director's rationale for imposing the more stringent condition is that:

Acme had asked for approval to release mercury to the environment at a rate of 21.2 grams per hour. The Director approved a limit of 11 g/hour, and the ENGO, Friends of Lake Hypo, wants zero discharge. Acme appeals the condition and Friends is granted intervention status.

TO THE ONTARIO ENVIRONMENTAL APPEAL BOARD ⁵

John L. Martin

Acme Widget will manufacture devices for use in cardiac cases for both adults and children. The state-of-the-art facility will be located in a community that needs jobs, and the government provided, or could provide assistance for the company to locate there.

The only hazardous material associated with the manufacturing process is mercury. Some mercury is present in the stack emission. However, Acme is an environmentally sensitive company and will use proven technology to reduce emissions as much as possible, not only to protect the environment but also because of the economic savings of re-using the captured mercury. If better economically achievable technology becomes available, the Ontario Ministry of the Environment (MOE) can, through Section 9 of the Ontario Environmental Protection Act, revisit the Certificate of Approval and impose a more stringent requirement, which Acme will gladly accept.

Acme is appealing the Ontario MOE Director's decision to impose a more stringent condition on three grounds.

DOES THE DIRECTOR HAVE AUTHORITY TO IMPOSE A CONDITION?

Section 9 stipulates that all emitters of a contaminant into the natural environment other than water require a Certificate of Approval. The more stringent condition in Acme's certificate is designed to control the deposit of mercury into the water of Lake Hypo. Since the applicant is not required to obtain an air approval for contaminant emissions into water, the Director cannot impose the condition.

Legislative loopholes are never intended, but are often there. With respect to this case, Acme submits that it is the proper role of the Ontario Environmental Appeal Board to advise the legislature that, if it was their intention to control atmospheric deposition into water through Section 9, they do so clearly and modify the Act.

HAS ACME MET LEGAL REQUIREMENTS?

The Director may impose a condition if it is necessary to ensure that plant operation will comply with the Act and regulations. Regulation 346 requires the release to meet the standards set out in the regulation. The only legal requirement with respect to mercury is a health-based maximum half-hour point-of-impingement concentration of 5 µg/m ³ . The applicant used the Regulation 346 dispersion model to confirm that the release would be only 1/500th of the standard.

Ontario MOE has no general stack emission concentration criteria or guidelines for mercury but, in 1996, adopted Guideline A-7 specific for new or expanded municipal waste incinerators in Ontario. The proposed guideline went through a rigorous public consultation process and, in response to the comments received, changes were made prior to adoption of a mercury emission guideline of 57 µg/m ³ . Using the emission characteristics for Acme, one can calculate that the 21.2 g/hour emission value is comparable to 53.4 µg/m ³ . Thus, Acme would meet this requirement as well, and Ontario MOE wishes to reduce the 21.2 g/hour value down to 11 g/hour.

DOES THE DIRECTOR HAVE PROBABLE GROUNDS TO PREVENT OR ALLEVIATE AN ADVERSE EFFECT?

The Director may refuse to issue a Certificate or may attach terms and conditions considered necessary to regulate contaminant release if there are probable grounds to prevent or alleviate an adverse effect, as defined in the Act. There are two key points on probable grounds and to prevent or alleviate . The Act does not say may prevent or may alleviate. Probable grounds do not relate to the possibility of an adverse effect.

Because of the present state of the science, models must be used to connect the emission of mercury from Acme's stack to the potential for mercury ingestion by humans, including models for air dispersion, deposition into the water, and methylation. Because of the concern about mercury, a conservative stance must be used when applying models. However, there is a danger with the compound use of models as in this case. Testimony has amply shown that the safety factors are into the thousands.

One major assumption is that methyl mercury production and bioaccumulation is the same in all ecosystems. In U.S. EPA's Mercury Study Report to Congress , the variability in mercury methylation among ecosystems was neglected because of the difficulties in modelling this process. Methyl mercury does not constitute the same fraction of total mercury in sediment, soil, and water across all ecosystems. Methylation varies among ecosystems by many orders of magnitude.

Because of the error associated with fate, transport, exposure and other models, wildlife criteria cannot be predicted within two or three orders of error. When the uncertainty associated with methylation is factored in, clearly the Director does not have probable grounds to impose a condition to prevent or alleviate an adverse effect. The proposed condition would reduce emissions by one half, but the built-in safety factors are many, many times more than that.

According to Paragraph 58, Ontario MOE, when making decisions, will consider cumulative effects on the environment, the interdependence of air, land, water, and living organisms and the interactions among the environment, economy, and society. Acme submits that the Director, when setting the condition in the Certificate of Approval, failed to consider economy and society.

CONCLUSION

Acme asks the Ontario Environmental Appeal Board to consider the evidence and agree that this applied control technology is the best available and, on balancing between environmental consequences and the economy and society in general and the benefits of this project in particular, that the Director and, now, this Board, impose a condition of 21.2 rather than 11 g/hour for mercury.

TO THE ONTARIO ENVIRONMENTAL APPEAL BOARD ⁶

Pat Moran

Acme Widget has appealed the decision of the Ontario Ministry of the Environment Director to impose a more stringent condition for three reasons:

• The Director has no legal authority to impose conditions.

• Acme's proposal meets all legal requirements of the Act and Regulation 346.

• The Director does not have probable grounds for imposing the mercury emission limit to prevent or alleviate an adverse effect.

The rationale for the Director's decision is laid out below.

LEGAL AUTHORITY

Section 9 of the Ontario Environmental Protection Act states that no person, except in accordance with a Certificate of Approval, shall construct a plant from which may discharge a contaminant into any part of the natural environment other than water . Acme contends that the Director is dealing with a water discharge.

Clearly, the release is going into the air, but the Director has taken into consideration that, once the mercury is in the air, there are a number of potential effects, one of which is deposition into water. The mercury added to Lake Hypo will accumulate and eventually cause health problems.

The Ontario Water Resources Act, which predates the Ontario Environmental Protection Act, was enacted to deal with municipal and industrial discharges directly into the waters of lakes and rivers. The phrase "other than water" was added to Section 9 of the Ontario Environmental Protection Act to distinguish air approvals from all other approvals, thereby eliminating the double jeopardy of a proposed facility having to obtain two approvals for the same discharge. The wording of Section 9 was never intended to preclude or limit the Director's ability to deal with air discharges.

If Acme is correct, that the Director is trying to control a water discharge and therefore does not have authority, then one can conclude that there is no way for the province to control air emissions in order to prevent adverse effects, regardless of where they occur, including water, land, or uptake by animals and humans. The legislature did not intend to create a huge loophole in the Section 9 approval process.

COMPLIANCE WITH LEGAL REQUIREMENTS

The Director agrees that Acme will meet all legal requirements of the Act and Regulation 346. However, the Director contends that the requirement of Regulation 346 is a starting point, not the end point. In his decision to impose a more stringent condition, he took into consideration the total mercury loading to Lake Hypo. Regulation 346 considers only what one source will emit. It does not take into account the mercury already in the environment, nor other sources nor their plumes.

Guideline A-7, which establishes mercury emission rates for new and expanded municipal waste incinerators, is not a legal limit and does not apply to Acme. It is intended to assist the Director and the Board in reaching decisions. In reaching his decision regarding Acme, the Director went beyond the guideline in order to protect the environment, based on local circumstances.

PROBABLE GROUNDS

That mercury accumulates through the food web and poses a threat to human and ecosystem health is clear. Acme contends that its 9 kg contribution is small, compared to the 328 kg contribution via the Hypo River from upstream sources, and that Ontario should address sources contributing to that total. Ontario's MISA (Municipal-Industrial Strategy for Abatement) program is dealing with those sources so that the annual upstream loading will decrease with time. From another perspective, Acme's proposed annual mercury release will double the present atmospheric loading to Lake Hypo.

The Director agrees that Acme will apply best available technology to control stack emissions. The Director contends, however, that Acme should also review its production process and make improvements to reduce the amount of mercury to be dealt with through post-process stack controls. Acme knows its production process and is best suited to determine what improvements are most appropriate to meet the Director's 11 g/hour condition. The Board is in a position to uphold the limit imposed by the Director and continue to reinforce the proposition that mercury has to be phased out, striving toward a goal of zero discharge.

THE INTERVENER'S SUBMISSION

The arguments presented by Friends of Lake Hypo are similar to those of the Director, the difference being their desire for zero discharge. Since the Director is not opposed to Friends' arguments, no commentary is necessary.

TO THE ONTARIO ENVIRONMENTAL APPEAL BOARD ⁷

Paul Muldoon and Marcia Valiante

ROLE OF THE INTERVENER

The intervener represents a constituency. The intervener's role, especially in administrative tribunals, is to push the policy envelope, introduce terms, conditions, and concepts to the limits of the Board's tolerance. Some have accused the intervener of being naive for such suggestions, but this role is not trivial. A review of the history of Canadian tribunals and courts shows that many views, once considered very progressive, ambitious, if not naive are now part of accepted law. For example:

• The proposal to ban chlorofluorocarbons was initially outrageous, unacceptable, and radical, but culminated in the Montréal Protocol.

• The proposal to reduce organochlorines from pulp and paper operations by more than 90% was initially radical, naive, and wild-eyed, but was adopted by the Canadian federal government and led to regulations.

The Ontario Environmental Appeal Board has a legal responsibility and duty to decide on the law. However, the present case is more than the consideration of jurisdictional authority under Section 9 of the Ontario Environmental Protection Act and application of strict legal principles. The case raises vital questions about gaps and challenges facing the Canadian environmental law and policy regime. The Board must also consider the goal of virtual elimination and ensure that the regulatory strategies advanced are in the philosophy of zero discharge. Friends of Lake Hypo suggests that application of the precautionary principle is now part of the legal and policy framework of the province of Ontario. To protect its clients, it must ask, how does Ontario's regulatory system:

• accommodate scientific uncertainty?

• account for cumulative effects?

• seek to protect the most sensitive populations?

• seek to advance technology and innovation?

• seek to deal with toxic substances which have no safe threshold?

FRIENDS' POSITION ZERO DISCHARGE

Acme Widget requested approval to release mercury at a rate of 21.2 grams/hour, but the Ontario Ministry of the Environment (MOE) Director issued a Certificate of Approval with a condition of 11 grams/hour, which Acme contends is too onerous. Friends contends that the condition is not strong

• Mercury is a persistent toxic substance that bioaccumulates in the food chain. Certain populations such as aboriginal people and other subsistence fishers are at higher risk because of their heavy reliance on fish as a source of food.

• Friends' clients, the Hypo First Nation, an aboriginal community located on the delta of Hypo River where it enters Lake Hypo, is heavily dependent on area wildlife and fish in Lake Hypo as food. Unless a zero discharge standard is imposed, the Board is penalizing the Hypo First Nation for living off the land.

• Mercury adversely affects the nervous system, sensory motor function, and human and wildlife reproduction. Even at very low doses where the mother shows no or only mild symptoms, adverse impacts are manifested in offspring. The offspring of mothers who regularly consume wildlife predators at the top of the food chain are the most vulnerable. Because of the particular sensitivity of the human fetus to mercury, Canada and the United States have severely restricted the consumption of fish containing mercury by pregnant women, women of childbearing age, and children.

• There is no evidence that human residues of mercury are declining; therefore, all exposure pathways should be addressed.

• During the public notice period for this permit, 450 responses were submitted. The issue is controversial, emotional, and media-heightened, factors that have not been captured in the Board's deliberations.

DOES THE DIRECTOR HAVE AUTHORITY TO IMPOSE A CONDITION?

Friends adopts the Director's submission that the Board has jurisdiction under Section 9, and that there is no impediment to impose a condition on an air permit that affects water, wildlife, and public health. This decision is well founded in case law, logic, and statutory interpretation. Acme's position regarding Section 9 ignores the fact that all ecosystem components are interconnected and would not allow the Director to protect one component of the environment.

DOES THE BOARD HAVE AUTHORITY TO IMPOSE A MORE STRINGENT CONDITION?

Acme contends that its requested condition is as stringent or better than what the law requires. Friends contends that the regulations are a floor and not a barrier against a more stringent condition. Friends contends that the Board and the Director clearly have authority, as shown in prior cases, to impose a more stringent condition, and the ceiling is at the Board's discretion.

Acme challenges whether the models used will provide answers. Friends contends that the Regulation 346 model, which provides the basis for the point-of-impingement regulatory standard for mercury, is seriously inadequate and its nuances should not be debated. An Ontario MOE critique of Regulation 346 models states:

"Dispersion models are outdated and do not consider all adverse meteorological conditions or lakeshore effects. ... The regulation does not directly limit the persistent toxic emission loadings to the environment. ... The combined effects of multiple emission sources are not addressed."

The model does not accommodate what is, in effect, an attempt to allocate the capacity of Lake Hypo to accept mercury, a contaminant for which there is no safe threshold. The model deals with only one source Acme and fails when there are several facilities all emitting the accepted amount of mercury.

PROBABLE GROUNDS TO PREVENT OR ALLEVIATE AN ADVERSE EFFECT

Friends contends that there are probable grounds to enact a more stringent standard and that the probable grounds will meet the standard of proof of Section 9. Friends' interpretation of Section 9 is that one has to be convinced that there are probable grounds for an adverse effect to happen. Given its inherent characteristics, mercury will, over time, invariably cause some adverse effect. The question is when and to whom.

In a previous case, the Board indicated that it is open to evidence that emissions, at a level that complies with numerical standards, are capable of causing adverse effects. In other words, Friends submits that it need only demonstrate that mercury emissions from Acme are capable of causing an adverse effect. Friends believes that the evidence before the Board shows that mercury can cause harm even at the regulatory level that is set.

In addition, the courts have said that impairment can be taken to include cumulative impact on the environment, such as when the addition of a substance to a lake would not cause immediate harm but would delay recovery from prior pollution.

THE BOARD'S OBLIGATIONS THE BROADER POLICY AND LEGAL CONTEXT

Because of its devastating human and environmental effects, a large number of international and governmental institutions have targeted mercury for drastic reduction and elimination. Therefore, the Board has a responsibility not only to apply Section 9 of the Ontario Environmental Protection Act but also to respond in the larger policy and legal context. Among the institutions promoting virtual elimination are:

• The federal Canadian Environmental Protection Act sets national policy for pollution prevention and virtual elimination of designated substances.

• Canada-wide standards for mercury are being developed under the auspices of the Canadian Council of Ministers of the Environment.

• The Commission for Environmental Cooperation has a goal to prevent or minimize anthropogenic inputs of mercury to the environment, specifically, that ambient mercury levels should approach natural background levels and human releases should be reduced until naturally occurring levels are achieved.

• The Great Lakes Water Quality Agreement established a goal of virtual elimination for inputs of persistent toxic substances. Annex 12 requires the design of regulatory strategies in the philosophy of zero discharge. Unless these commitments are translated to this hearing, the Agreement has no meaning.

• Through the Great Lakes Binational Toxics Strategy, the governments of Canada and the United States have committed to finish the business of virtual elimination.

• The Canada-Ontario Agreement seeks to implement the international Agreement and to buttress the Binational Strategy goal to reduce the use, generation, and release of mercury by 90% by the year 2000. The Minister of the Environment has confirmed Ontario's commitment to this goal.

• The International Joint Commission has recognized and applied the concept of zero discharge and virtual elimination. The report of the Commission's Virtual Elimination Task Force provides enormous insight on how to apply the vague concept of virtual elimination to the specific issue of mercury.

UNCERTAINTY AND PRECAUTION

Friends represents Needyville, whose citizens are concerned about the quality of their environment. Their government, the province of Ontario, has committed to a policy of virtual elimination. This policy is implemented through the Board, the Director, and other environmental decision-makers in Ontario. However, that commitment is not reflected in the Certificate of Approval issued to Acme Widget.

Friends submits that the Board should take a quite different view from what Ontario MOE and Acme suggested. In Friends' view, if the science were certain, that there is no threshold limit for mercury and the models are accurate, the Board and the Director would, without question, have imposed virtual elimination as a condition in the Certificate. However, there are uncertainties in all parameters the quantity of mercury emitted, dispersion, deposition, food web accumulation, metabolism. The issue is too complex to trace a given mercury molecule into a particular fisher. There will never be 100% certainty about the impacts of mercury, and the Board will have no more certainty than it has now.

The traditional tort model of environmental regulation in Canada is that a source operates under the presumption that it is not causing harm until proven otherwise. Acme argues that it is the responsibility of governments to develop certainty and that it should be allowed to operate until government decision-makers say stop. Acme argues that it is better than the standard and should not bear the burden of uncertainty, for economic reasons. Acme offers the benefit of jobs. It does not have to be certain, but relies on uncertainty and reaps the advantage.

Friends submits that, in the context of a persistent, bioaccumulative toxic substance like mercury, this decision-making model is outmoded. It is unfair and unjust to make aboriginal people, women and their babies, children, fish, and wildlife bear the burden of scientific uncertainty. Lake Hypo is already burdened, as evidenced by fish-consumption advisories on both the U.S. and Canadian sides of the lake. Acme is asking for dispensation, the right to add to that burden. The effect of that mercury will be dramatic and irreversible.

The Board must not choose between jobs and the environment. A productive economy and a clean environment go hand in hand. If the environment is impaired, society will not have a quality economy, either. When faced with serious or irreversible environmental degradation, uncertainty should be resolved in favor of the environment. Friends submits that the Board is faced with serious and irreversible environmental degradation by Acme's application. Some suggest that the hourly input of mercury to Lake Hypo is minuscule, but 9 kg per year for 25 - 30 years is a lot to add to an already burdened ecosystem.

Friends asks the Board to take a precautionary approach. The precautionary principle has been developed in international environmental law and adopted in numerous international instruments, most notably the 1992 Rio Declaration of the United Nations Conference on Environment and Development. Principle 15 states:

"In order to protect the environment, the precautionary approach shall be widely applied by States according to their capabilities. Where there are threats of serious or irreversible damage, lack of full scientific certainty shall not be used as a reason for postponing cost-effective measures to prevent environmental degradation."

The precautionary principle is also associated with other principles of international law, such as sustainable development and the concept of inter-generational equity, where one does not compromise the ability of future generations to live in a quality environment. The common theme is justice and equity, both within and between generations.

Canada champions the precautionary principle at the international level, and it is becoming a common part of Canadian domestic law. The essence of the principle, as articulated in the Rio Declaration, was included in the Canadian Environmental Protection Act. The precautionary principle is also a fundamental tenet of the federal-provincial Harmonization Accord, signed by Ontario, which provides binding legal commitments for cooperative implementation.

Friends cites a number of cases in its brief from tribunals across Canada that have adopted the precautionary approach in decisions. This Board, in an earlier decision in the Uniroyal case, made choices based on precaution.

More specific to Ontario MOE, the Environmental Bill of Rights provides statutory grounding to the effect that the Minister and presumably his delegates shall take every reasonable step to ensure that Ontario MOE's Statement of Environmental Values is considered whenever they take decisions that might significantly affect the environment. Friends contends that the Director did not consider the Statement of Environmental Values in his decision regarding Acme. The Board, with the authority to exercise any power that the Director has, should now apply the Statement, with consideration to the following.

• Application of the ecosystem approach. The Statement of Environmental Values indicates that, when making decisions, Ontario MOE will consider cumulative effects on the environment, the inter-dependence of air, land, water, and living organisms and the inter-relationships among the environment, the economy, and society. Acme singled out the economy as most important, but Friends suggests that the factors are a seamless web and the balance among them is most important, and that the Board consider all.

• Application of the precautionary principle. The Statement says that Ontario MOE will exercise a precautionary approach in its decision-making, especially when there is uncertainty about the risk presented by particular pollutants. Friends suggests that persistent toxic substances are at the top of list. To exercise caution in favor of the environment conveys that environmental protection is more important than the economy and Acme turning a profit on the back of the community.

• Impact on First Nations. The Statement says that Ontario MOE will promote and implement the principles of a Statement of Political Relationship, which deals with the political status of First Nations. Within that context, Ontario MOE will evaluate the impact of proposed decisions on First Nations and aboriginal communities. Friends suggests that, in evaluating Acme's application, this was not done. In determining the numerical limit as the condition in the Certificate of Approval, the Director took an across-the-board approach but did not account for the particular vulnerability of the community that will be directly impacted.

• Priority. The Statement says that Ontario MOE's environmental protection strategy will place priority first on preventing and, second, on minimizing the creation of pollutants that can damage the environment and that, when the creation of pollutants cannot be avoided, the priority will be, first, to prevent their release and, second, to minimize their release. Friends suggests that the Director did not consider how to prevent or minimize the creation of mercury in Acme's process and how to prevent its release.

Failure of the Board to consider the precautionary principle in imposing a condition on Acme would constitute a failure to fulfill the requirements of the Environmental Bill of Rights, as well as a failure to fulfill the purpose of the Ontario Environmental Protection Act itself, to protect and conserve the natural environment of Ontario.

The precautionary principle applies in this case. There are uncertainties about mercury but enough is known about its impact on human and ecosystem health. Every government on this continent and most around the world have said mercury should approach naturally occurring levels and that all anthropo

In its Statement of Environmental Values, Ontario MOE has already identified the appropriate response for this Board place pollution prevention as the first priority. Friends submits that the Director did not ask Acme why mercury was in its process or how the process or product could be changed to prevent its presence at all or, if mercury must be created, given the importance of the widget being manufactured, how its release could be prevented entirely. Acme presented no evidence to suggest that pollution prevention was considered, or that pollution prevention would be cost prohibitive. In fact, evidence indicates that introduction of pollution prevention generally makes companies more competitive and improves their bottom line.

Ontario MOE promised the citizens of Ontario a clean air program almost 15 years ago that, like the MISA (Municipal-Industrial Strategy for Abatement) program for point source discharges to water, would have adopted virtual elimination of persistent toxic substances in its regulations. The failure to do so does not preclude this Board from going beyond the basic requirements of Regulation 346.

SUMMARY

Friends submits that:

• The Board has jurisdiction to impose a zero discharge condition, when necessary, to prevent a source from causing an adverse effect.

• There are sufficient grounds in this case to do so.

• Ontario has committed to the principles of precaution, pollution prevention, and virtual elimination through signed agreements and the incorporation of the principles into its policy and legal instruments.

• The Ontario Environmental Protection Act is sufficiently flexible to allow the Director and the Board to require implementation of these policies in individual applications, such as Acme's.

LEGAL AUTHORITY

1. Section 9 of the Ontario Environmental Protection Act refers to discharge into any part of the natural environment other than water . This raises a fundamental question about the jurisdiction of Section 9. How realistic is the argument that a Certificate of Approval cannot address any problem in water as a result of a release to the air?

This potential issue has not been challenged in a time span of more than two decades. If a challenge were raised and the case ended up in court, the Ontario Ministry of the Environment (MOE) has a high level of confidence that the challenge would fail. The court tends to take a purposeful, rather than a technical approach to interpretation. The aim is to reinforce delivery of such a social welfare measure rather than to erect barriers. Nonetheless, many people would feel more comfortable if there were language in Section 9 to clarify that point.

This scenario was highly structured, with one contaminant from one source to one water body. A real case would have been much broader. For instance, the Ontario MOE Director's condition would include rationale related to air and land in addition to water, for example, land deposition of contaminants followed by transport to the lake.

REGULATIONS

2. Do Ontario's air regulations contain specific limits for specific contaminants?

Yes and no. The air regulation numbers date from 1965 and originally covered 100 contaminants. That number is now 87. The other 13 were removed from the list primarily because the limits were above what was considered safe, as determined from more recent science. However, agreement has not been reached on new numbers, in part because ambient concentrations are above levels known to cause adverse effects. The need in these cases is not so much for a number than for elimination.

Through Ontario MOE's Clean Air Program, new lists of contaminants with lower numbers were proposed in 1986-87. These are still under discussion, reflecting the challenge to transform policy decisions into law. An August 1998 list covers 200+ contaminants, and numbers are incorporated into Certificates of Approval on an ad hoc basis.

GEOGRAPHIC LOCATION

3. Is plant location a consideration? For example, if a determination were made that a plant would have a large impact on a lake if located on the shoreline but a small impact if inland, would a decision to locate the plant inland be construed as pollution prevention?

Location is not a consideration under the Ontario Environmental Protection Act. Ontario MOE

In designing the scenario, the organizers rejected placing Acme under the Ontario Environmental Assessment Act. The myriad additional variables would have diverted focus from the original question.

MODELS

4. What are the capacities of Ontario's Regulation 346 model in determining ambient air concentrations or deposition?

This dispersion model has been part of Regulation 346 for almost 30 years. More advanced models have subsequently emerged and are typically used by Ontario MOE and others to quantitatively estimate the impact of particular pollutant emissions. The Regulation 346 model is unsuited to this application, given Acme's location near the shoreline and with other mercury sources in the vicinity. Ontario MOE has indicated its intent to revise portions of this regulation, including the model, but has not yet done so.

BALANCE

5. The point was made that the Ontario MOE Director and the Ontario Environmental Appeal Board should consider both environmental and economic factors in making decisions. Would an already existing source, that is, one contributing to the economy, be treated differently from a proposed new source?

Ontario legislation makes no distinction and Ontario MOE has actively addressed existing sources with a strong presence in the economy. For instance, Ontario MOE issued a generic order requiring all kraft pulp mills to reduce aqueous contaminant discharges down to specified levels. Also, the MISA (Municipal-Industrial Strategy for Abatement) program was developed and implemented to deal with a range of municipal and industrial sources. Further, under the Environmental Bill of Rights, the Ontario Minister of the Environment is obligated, upon written request from any two Ontario citizens, to review the permit for an existing facility. When an existing air source requests a major amendment to its permit, Ontario MOE routinely requires the applicant to provide emissions inventory information for use in decision-making.

6. What are the respective roles of law, policy, and principle, and which prevails in achieving a balance?

All sources in Ontario have to meet regulations, which are grounded in law. There are specific rules to interpret the language in the regulations, and one can also refer to previous cases for interpretation.

Ontario is also committed to general principles, such as the Statement of Environmental Values. The challenge is how to apply the principles. The law requires the government to consider or factor these principles into its decisions. Interpretation can be either broad or narrow.

Policies have been adopted into various legal documents such as the Canada-Ontario Agreement and the Great Lakes Water Quality Agreement. As a signatory to the former, Ontario is legally bound to implement its provisions.

7. Why would the Ontario MOE Director not impose a condition of zero discharge?

Zero discharge hinges very much on the availability of appropriate production and control technology. The prevailing view with Ontario's approvals process is that, if affordable technology is available to meet a limit, a company should use it, especially for a contaminant like mercury. Generally, technology questions would be discussed between the applicant and Ontario MOE, and possibly with the intervener, and those questions largely settled before the case came to the Ontario Environmental Appeal Board.

The province's strategy to reach the zero discharge goal is to ratchet limits down bit by bit, availing itself of opportunities such as when a Certificate holder seeks Ontario MOE approval to add new equipment. By tightening up on limits and challenging sources to look at production processes rather than just end-of-pipe controls, Ontario MOE leverages the process toward zero discharge. With most process improvements, the company's economic bottom line is also improved.

8. Acme's proposal would double the atmospheric deposition of mercury to Lake Hypo. The Ontario MOE Director's condition would increase the loading by 50%. Is this not contrary to a goal of zero discharge?

Ontario MOE favors a staged approach to reducing overall discharges. This includes ensuring that the newest sources come on line with best available technology (BAT). This process may result in a temporary increase in overall loadings, but it does not punish newer companies that want to introduce BAT.

Existing sources likely are not using BAT. Ontario MOE can review their Certificates and ask them to incorporate new technology, thereby reducing their loadings. Ontario MOE can also issue an order and unilaterally amend a company's Certificate. Rather than a systematic review of all Certificates for all contaminants, Ontario MOE follows a case-by-case approach, exploring ways for broad application of new technology as that technology becomes available.

It is easier to persuade older sources to adopt BAT that is proven and in use by newer sources. There is no set time line. Each Certificate is handled individually, based on review of an application and consultation with the company. This approach facilitates consideration of such factors as the cost of the technology, challenges of installation, and testing to ensure full equipment operation.

9. In advocating zero discharge, is the intervener arguing that the facility should not be built, regardless of the product manufactured?

The debate centres on pollution prevention. The rationale behind the Ontario MOE Director's condition was to encourage the company to look at production processes in addition to end-of-pipe control. The intervener could strengthen that position by introducing evidence that other processes could manufacture the product and avoid the use or creation of mercury. Setting the condition below what control technology can deliver encourages innovation. The company can still operate, but with different processes and materials, and the environment benefits. This also provides a lever, through the Certificate of Approval process, to apply tighter limits on other existing sources.

VIRTUAL ELIMINATION

10. A virtual elimination policy was incorporated into the Great Lakes Water Quality Agreement more than 20 years ago. How realistic is this policy in light of a large number of facilities seeking allowable releases?

The need for virtual elimination or other limits, such as the assimilation capacity of a particular water body for a particular contaminant, may come to be recognized only after an extended period of time. A case-by-case approach functions in a changing landscape and may be unfair to applicants, who have business decisions to make. They need to know up front what the requirements are and not have

The Canadian Council of Ministers of the Environment are addressing a number of contaminants, taking into account technological, environmental, and economic factors. If they develop a new mercury limit that is less than the number presently in Regulation 346, and that limit is, in turn, adopted by Ontario MOE can lower that number and review existing Certificates to determine whether they are in compliance. Ontario MOE and the sources could cooperatively work out a schedule to phase in changes over a period of time. This keeps the pressure on to improve both control technology and apply alternative production processes.

11. Consider the following situation. An existing source has a Certificate of Approval to manufacture X widgets per day but, despite capacity, produces only ½X. The economy improves, so the company increases production to X. As a result, emissions of a persistent toxic substance are increased. The source is still within the limits of its Certificate, but the emissions increase is contrary to a policy of virtual elimination. What are Ontario MOE's options?

Any process or production rate change requires a new Certificate of Approval. This requirement would apply, for instance, if the company received a Certificate for a plant 2Y in size, to be built in stages. Initially, only Y is built, so normally, Ontario MOE would approve only Y. The company would come back to Ontario MOE when it wants to expand beyond Y. This would also apply if the company received approval for Z but, after start-up, put out only ½Z and later wanted to expand and still stay under Z.

In its application for a Certificate, the company does not request a number for a contaminant. Rather, it provides information on how the facility will be run, for example, the volume of air exiting the stack and the concentration of contaminants in that air. Ontario MOE approves a package, not just a contaminant load. The company cannot increase its contaminant load without a change to its Certificate. Ontario MOE thus keeps control to incorporate new or improved technology, revised standards, or policy changes. Ontario MOE can also amend an existing Certificate at any time.

DISCHARGE AND EMISSION DATA AND ITS AVAILABILITY

12. Discharge and emission data for all sources would benefit scientists, for instance, to estimate the relative contributions from different sources to a given water body. This, in turn, would benefit policy- and decision-makers. The availability of such information would also benefit others who wish to influence policy and decisions. What inventories are available and can the data be used for such purposes?

Emission information is reported to Ontario MOE and is accessible under the Freedom of Information and Protection of Privacy Act. The provincial program is presently being brought on line, so the amount of emission inventory information presently available is limited. Ontario will concentrate on obtaining an inventory of 35 major air sources. Other sources are required to provide this information when applying for a new Certificate. Occasionally a company will request confidentiality. There are legislative provisions in place to challenge the request and also to protect the company.

Inventories assembled by the Canadian federal government are also available and routinely used.

13. The Freedom of Information and Protection of Privacy Act places the onus on the requester to identify the source and the type of information desired. Would it not be more logical to place that information onto a publicly available register?

A request can be worded as broadly or narrowly as the requester wishes.

14. The intervener identified a number of commitments to virtual elimination. How strong is the linkage between national and international policy declarations by the Canadian federal government and legal mandates to deal with sources like Acme?

The linkages need greater emphasis and profile. The International Joint Commission may want to consider how to better develop the path to move from policy declarations to the permitting process and other individual functions of environmental regulation.

Other declarations:

• Canada has signed onto the UNECE (United Nations Economic Commission for Europe) heavy metals protocol through the LRTAP (long-range transport of air pollutants) convention.

• The New England Governors - Canadian Premiers Mercury Action Plan is presently under development.

• The Northern Contaminants Program deals with mercury and persistent organic pollutants in traditional foods.

the Canadian and United States Legal Landscapes

1. Please elaborate on the respective roles and responsibilities of the Canadian federal and provincial governments.

The British North America Act, which created Canada as a nation in 1867, did not specifically address the environment. Through interpretation and application, the responsibility for environmental protection is seen as shared, but there are no strict lines of division between the federal and provincial levels of government. The federal authorities have clear responsibility to deal with certain issues, and provincial authorities with others, with occasional overlap which is open to discussion. The federal government deals with environmental matters, for example, under its commerce and trade powers and in treaties.

The federal-provincial process is dynamic and continues to evolve. Governments sort things out as they go along. It would be unusual to engage in a direct jurisdictional fight, which would lead to a constitutional argument about authority. The federal Fisheries Act is an example of a shared authority, with both the federal and provincial governments availing themselves of certain provisions. In a similar vein, Ontario can lay charges against a source under the Ontario Water Resources Act, and the federal government can under the Fisheries Act.

Generally, the Canadian federal government role in the provincial approvals or permitting processes is confined to reviewing and commenting on applications that raise issues falling within the mandate of federal departments.

The Canadian approach to environmental protection is primarily based on the principle of cooperative federalism. Consider, for example, the Harmonization Accord signed by the federal, provincial, and territorial governments in January 1998. That Accord sorts out roles and responsibilities, and priorities have been set to deal with a range of environmental issues. In the context of air, this includes smog, SO ₂ , ozone, NO _x , and mercury. Through the Accord, a broadly-agreed-to national action plan and standards will be created, but responsibility for implementation would go to the most appropriate level of government, be it federal, provincial, or other. The Canadian Environmental Law Association brought a lawsuit against the federal government challenging the legality of the Accord because it alters the way the federal and provincial governments interact. The merits of the case demonstrate that the dynamic can be controversial. The issue is more political than legal.

One could contend that the Canadian approach is long and frustrating, and a downward spiral to the lowest common denominator but, in the Canadian context, it is an efficient way to operate. The issue is not how the systems operate but, rather, can environmental problems be resolved. For example, under its trade and commerce authority, the federal government has set automobile emission limits applicable at the time of sale. The provinces have, from time to time, made the emission limits more stringent. To ensure compliance with all, a company applies the most stringent limit.

The federal Canadian Environmental Protection Act and the U.S. Clean Air Act contain reciprocity provisions which would apply in the event of real or potential transboundary issues. There have been a number of cases where governments have intervened across the border. The U.S. federal or a state government would have to seek leave if it wished to intervene in an Ontario case.

3. Please contrast the U.S. and Ontario approaches to dealing with emission sources.

U.S. legislation has established technological requirements for air sources, such as MACT (maximum achievable control technology) standards, along with a mechanism for mandatory review. In Ontario, there is no specific point where the Ontario Ministry of the Environment could review all existing Certificates that involve, for instance, mercury. Rather, sources are handled on a case-by-case basis. This introduces greater uncertainty, but allows greater flexibility and opportunity to implement technological advances as they become available. Because of this approach, the newest plants are always the cleanest and the older ones not considered until other action is taken.

4. The TMDL (total maximum daily load) approach in the United States systematically looks at all sources that contribute a contaminant to an impaired water body, then regulates those sources to remove the impairment. Is there a legally grounded parallel in Canada?

Under Ontario's MISA (Municipal-Industrial Strategy for Abatement) program, initial regulations required all municipal and industrial sectors to inventory their direct discharges to water. Based on that inventory, a second set of regulations was developed to control and reduce those releases over time. The intent was to also consider air emissions under the Clear Air Program, but that has never been done. The number of air sources is ten-fold greater than water sources, resulting in concerns about cost and management problems.

Through the acid rain program, Canada, and Ontario in particular, identified significant sources and developed specific regulations to deal with them.

Releases of persistent, bioaccumulative toxic substances occur in the Great Lakes, water quality standards are exceeded, and fish consumption advisories are in place. The Acme Widget / Lake Hypo scenario was constructed to explore whether legal authorities exist to deal with deposition of contaminants from the air and the extent to which governments are using those authorities. The specific question was:

What legal tools are available, under current Canadian and United States law, to control atmospheric deposition of persistent toxic substances to the Great Lakes originating from sources within and outside the basin?

In light of the information and arguments presented, there is some debate whether governments are doing everything they can. To present its perspective on the adequacy of the legal frameworks in Canada and the United States, the panel answered two specific questions.

Canada

1. Does the Canadian legislative framework allow Canada to effectively address the problem of water pollution via the atmosphere?

A qualified yes. The combination of federal and provincial frameworks is patchy, not comprehensive, and tends to be ad hoc and subjective.

2. Does Canada's domestic legislative framework provide for effective fulfillment of its international obligations?

A qualified no. The framework is not sufficiently prescriptive.

United States

1. Does the United States legislative framework allow the United States to effectively address the problem of water pollution via the atmosphere?

A qualified yes. Section 112 of the Clean Air Act is adequate. However, a more aggressive and creative approach needs to be taken for application of that framework.

2. Does the United States' domestic legislative framework provide for effective fulfillment of its international obligations?

There is a disconnect between the U.S. legislative framework and the international commitment to virtual elimination, specifically in the definition of the goal. If virtual elimination is defined as absence of harm, that is, removal of fish-consumption advisories, the answer is a qualified yes. If virtual elimination is defined as zero discharge, the answer is a qualified no.

The Great Lakes Water Quality Agreement calls for restoration of beneficial uses, one of which is fish edibility. Restoration would be reflected in the lifting of fish-consumption advisories. Rather than exploring ways to ratchet releases down to a point where advisories could be lifted, the tenor of the presentations and discussion was oriented toward not increasing releases.

An initiative such as development of MACT (maximum achievable control technology) standards combined with residual risk analysis, as presently envisioned, might be sufficient to lead to removal of some fish-consumption advisories. However, to eliminate all advisories, additional work will be required, for example, with contaminated sediment, sources such as the utility sector, and sources on a global scale.

The Agreement's Annex 15 (Airborne Toxic Substances) commits both countries to reduce air emissions that contribute to deterioration of Great Lakes water quality. To ascertain whether programs are on track, the underlying science must be improved and monitoring programs expanded, for example, in terms of emission inventories, contributions from different geographic regions, and trends.

VIRTUAL ELIMINATION AND PRECAUTION

Both Canada and the United States have adopted the concept of virtual elimination for persistent toxic substances. Their commitment is reflected in domestic legislation and as signatories to the Agreement. Through various declarations, protocols, and other instruments, both countries have also ascribed to the precautionary principle at the international level. However, both need to more explicitly incorporate virtual elimination and precaution into specific domestic programs and decision-making processes. Further, to help meet their international obligations, both need to better align their existing legislative

and institutional frameworks and better use their authorities.

THE SCIENCE

Clearly, there are technical challenges to the linking of loadings to a particular water body back to specific pollutant sources. The Lake Michigan mass balance study and associated modelling work have helped determine contaminant sources within and outside the basin, and the Devil's Lake and Everglades studies are exploring relevant scientific questions and identifying individual sources. The Lake Michigan work has provided a basis to explore jurisdictional issues but, for the most part, science has not yet been linked to the institutional framework. The scientific community must be better engaged so their work informs policy making. The science to link atmospheric deposition back to individual sources, where reductions are required, can proceed in parallel with work on the institutional framework to achieve

those reductions.

THE U.S. LEGAL FRAMEWORK

The NPDES Program

There are two significant drawbacks to addressing the air deposition issue through the NPDES (National Pollutant Discharge Elimination System) water permit program under the Clean Water Act the length

of time that would be required to establish a solid legal foundation through the judicial system, from district courts, through courts of appeal and ultimately the Supreme Court; and case-by-case application of requirements to deal with a broad, widespread area-type of problem. A more comprehensive approach is required.