Canada v. IPSCO Recycling Inc., 2003 FC 1518

Date: 20031223

Docket: T-2274-00

Citation: 2003 FC 1518

Ottawa, Ontario, Tuesday, the 23^rd day of December 2003

PRESENT:      The Honourable Madam Justice Dawson

BETWEEN:

                                                  HER MAJESTY THE QUEEN

                                                                                                                                            Applicant

                                                                         - and -

                             IPSCO RECYCLING INC. and GENERAL SCRAP &

CAR SHREDDER LTD., now known as JAMEL METALS INC.,

carrying on business as a partnership under the firm name and style of

GENERAL SCRAP PARTNERSHIP and XPOTENTIAL PRODUCTS INC.,

JACOB LAZARECK AND MELVIN LAZARECK

                                                                                                                                      Respondents

                                          REASONS FOR ORDER AND ORDER

DAWSON J.

[1]                 In this application Her Majesty The Queen, as represented by the Minister of the Environment ("applicant" or "Environment Canada"), seeks a permanent, mandatory injunction against the respondents pursuant to section 311 of the Canadian Environmental Protection Act, 1999, 46-47-48 Eliz. II, c. 33 ("Act"). The injunction sought is one which would require the respondents, and their agents and servants, to store all PCB material at the business premises of General Scrap & Car Shredder Ltd. and XPotential Products Inc. that is currently lying in open piles or otherwise improperly stored, in containers that provide sufficient durability and strength to prevent the PCB solids and PCB substances from being affected by the weather or released.

[2]                 These reasons are lengthy. In them I conclude that upon the totality of the evidence Environment Canada has failed to meet its burden to establish on a balance of probabilities that it appears that the respondents have committed an offence under the Act by improperly storing PCB material. In consequence, the application for injunctive relief is dismissed. For ease of reference, the following is an index of the headings and sub-headings pursuant to which these reasons are organized, and the paragraph numbers where each section begins.

Index

I.           Background Facts

(i)          The Parties [3]

(ii)         General Scrap and its Business [6]

(iii)        XPotential and its Business [13]

(iv)        PCB Materials [17]

(v)         The Dispute Between the Parties [18]

(vi)        The Current Status of the Material in Issue [41]

(vii)       The Specific Order Sought [42]

II.         The Issues [45]

III.        The Analysis

(i)          What are the requirements to be met in order to obtain injunctive relief under subsection 311(1) and what is the scope of subsection 311(1)? [46]

(a) Is subsection 311(1) to be limited to exceptional cases? [47]

(b) The constituent elements of subsection 311(1) [52]

1. The Text - Grammatical and Ordinary Sense [56]

-Text as it Speaks to the Standard of Proof [58]

-Text as it Speaks to the Prevention of Offences [69]

2. The Context

-The Scheme of the Act [80]

(c) Conclusion with respect to the scope of subsection 311(1) re existing and continuing situations [95]

(ii)         Is some or all of the material PCB Material? [98]

(a) The evidence in support of the application [100]

1. Affidavit of Mervin Fingas [101]

2. Affidavit of Shannon Kurbis [107]

(b) The evidence in opposition to the application [108]

1. Affidavit of Jacob (Jack) Lazareck [109]

2. Affidavit of David Clark [113]

3. Affidavit of Jan Merks [118]

4. Affidavit of Michael Bertram [126]

5. Affidavit of Detlef Birkholz [130]

6. Affidavit of Donald Davies [132]

(c) Analysis of the Evidence [133]

1. The Validity of the Environment Canada Data [137]

-Is the 1999 Environment Canada Report based on insufficient samples and an inappropriate sampling    system? [139]

-Did Environment Canada depart from its own sampling plan? [152]

-Did Environment Canada use an inferior method of sample selection? [155]

2. The use of individual test results [179]

3. The use of statistical analysis [194]

III.        If some or all of the material is PCB Material, which respondents are in breach of the Regulations? [233]

IV.        Conclusion, Order and Costs [237]

I. BACKGROUND FACTS

(i) The Parties

[3]                 The respondents General Scrap Partnership ("General Scrap") and XPotential Products Inc. ("XPotential") carry on recycling businesses.

[4]                 General Scrap is a partnership originally formed between IPSCO Recycling Inc. ("IPSCO") and Jamel Metals Inc. ("Jamel"). Jamel is the corporate successor of General Scrap & Car Shredder Ltd. It was the uncontradicted evidence of the only witness who has personal knowledge of the corporate structure of the respondents that currently IPSCO owns 100 percent of General Scrap. I accept this to be the case.

[5]                 Jacob and Melvin Lazareck are brothers who are the officers, directors and shareholders of Jamel. Jacob Lazareck testified, and I accept, that neither he nor his brother are now involved in the operation of the day-to-day business of General Scrap.    Jacob Lazareck is the sole owner, officer, director and operator of XPotential.

(ii) General Scrap and its business

[6]                 In 1967, General Scrap & Car Shredder Ltd. began operation of the first automobile shredder in Canada on Springfield Road, in Winnipeg ("General Scrap site"). General Scrap continues to operate the shredder at this site.

[7]                 An automobile shredder is a large, industrial machine powered by a high-horsepower motor. Scrap (for example, automobiles or large appliances) is fed into the shredder opening. The high-horsepower motor turns a series of rotating discs at great speed. Fitted on the discs are a series of large steel hammers which physically shred the scrap fed into the shredder into fist-sized pieces. The shredded material then passes through a magnetic drum which separates the ferrous metals from the remaining non-ferrous material. Non-ferrous metals (for example, copper and aluminum) are recovered during further downstream processing. The remaining non-metallic material is known in the industry as automobile shredder residue ("ASR").

[8]                 ASR is very heterogeneous material. It is solid and generally medium to dark brown in colour. It is largely made up of plastic and foam but also contains pieces of metal, rubber, fabric, carpet, wood, wire, glass, dirt and other materials. Individual components of ASR vary in size from fine dirt to pieces of foam, fabric and rubber over one foot in size. Individual objects are generally identifiable in the material.

[9]                 PCBs were banned from use in Canada in 1977. Prior to this, components containing quantities of PCBs were used in the manufacture of automobiles and household appliances. PCBs were used as additives in paints, rubber, foam, sealants and plastics. Quantities of PCBs can be found in automobile ignition systems, household appliance capacitors, wiring and transformer components, electric motors, air-conditioners and hydraulic devices. When older vehicles and appliances are shredded, the PCBs that they may contain remain in the ASR in low concentrations. A small electrical component containing only 500 grams of PCBs can contaminate up to 10 tonnes of ASR to a PCB concentration above the regulatory threshold.

[10]            Given the difference in materials that might comprise the shredder input from one load to another, the degree of contamination with PCBs varies in ASR. PCB contaminated ASR can be expected to be found in varying amounts and varying levels of contamination throughout ASR the storage piles. Mr. Lazareck acknowledged that there may be "hot spots" or small areas of concentration that could be over 50 ppm. As is discussed below, 50 parts per million ("ppm") is the regulatory threshold.

[11]            Approximately 20,000 metric tonnes of ASR is stored at the General Scrap site. The ASR is located in three areas which have been identified by Environment Canada as the "East Pile", "West Pile" and the "Central Pile". Much of the material is located below the natural grade level because it was used as fill in low-lying areas. The elevation above grade level of each of the East and West Piles is approximately 2 metres. The Central Pile is not the subject of any complaint in this proceeding.

[12]            The ASR in the three piles at the General Scrap site was generated from approximately 1969 to 1978. Since around 1994 no new ASR has been added to those piles. New ASR is transported directly to the XPotential site, as described below.

(iii) XPotential and its business

[13]            XPotential was incorporated in order to take advantage of a new recycling technology which permits the combination of ASR with post-consumer waste plastics in order to produce saleable products such as parking curbs, landscape crossings, fence posts, railroad ties and non-structural timbers.

[14]            XPotential operates a facility approximately one mile east of the General Scrap site ("XPotential site"). The site consists of a plant and a storage cell area. ASR supplied by General Scrap is stored in storage cells for later use as raw material. The Manitoba Department of Environment (now known as Manitoba Conservation) licensed the construction of the plant and the movement of ASR to the storage cells at the XPotential site. XPotential reports regularly to Manitoba Conservation with respect to groundwater monitoring wells maintained at the XPotential site. Manitoba Conservation has expressed no concern to XPotential about PCBs on its site, notwithstanding knowledge of this application brought by Environment Canada.

[15]            Since July of 1996, XPotential has received approximately 125,000 tonnes of freshly produced ASR from General Scrap. This is stored in the storage cell area at the XPotential site. Newly produced ASR from General Scrap continues to be transported to the XPotential site where it is used in its manufacturing process.

[16]            The storage cell area is essentially one large storage area divided into several compartments or cells. The cells are excavated to a depth of 2 metres below natural grade and native clay is compacted to construct 2 metre high berms between the cells and along the perimeter of the cells. The area of the cells is divided into 6 ASR storage cells, a drying cell (where ASR can be air-dried prior to processing) and a separate excess water retention cell. Environment Canada has advised that the ASR contained in cell 4, cell 5 and the drying cell are of concern. The ASR in each of these cells is comprised of ASR that had been stored at the General Scrap site for several years and newer material.

(iv) PCB Materials

[17]            The Storage of PCB Material Regulations, SOR/92-507 ("Regulations") define material containing over 50 ppm of PCBs to be "PCB material". PCB materials in an amount of 100 kg or more must be stored, reported, labelled and handled in accordance with the Regulations. Failure to comply with the Regulations is an offence under the Act.

(v) The Dispute Between the Parties

[18]            In September of 1997, Environment Canada informed the respondents that it was Environment Canada's intention to conduct inspections at the General Scrap and the XPotential sites and to take samples of ASR for analysis in order to determine the PCB content of the ASR.

[19]            As a result, General Scrap and XPotential instructed Wardrop Engineering Inc. ("Wardrop") to conduct an ASR sampling and analysis program. The purpose of the program was to determine whether the ASR at both the General Scrap and XPotential sites were PCB materials within the meaning of the Regulations.

[20]            In consequence of those instructions, Wardrop produced a report dated October 16, 1997 which described the collection and analysis of 24 samples of ASR that Wardrop collected at the XPotential and General Scrap sites. Wardrop concluded, among other things, that:

1.          PCB concentrations in the samples analyzed ranged from 8 to 49 ppm.

2.          At an 80% statistical confidence interval the PCB concentration of the ASR was between 20.9 and 27.5 ppm, which was below the regulatory threshold.

[21]            These results were communicated to Environment Canada October 17, 1997.

[22]            Environment Canada proceeded with its own investigation and collection of samples. Wardrop was engaged by General Scrap and XPotential to take duplicate samples from the locations at which Environment Canada took samples. Samples were taken from both sites over the period of October 20, 1997 to October 23, 1997.

[23]            In a report dated September-December 1997 ("1997 Environment Canada Report"), Environment Canada identified PCB at mean concentrations above the regulatory limit of 50 ppm at three locations. They were the West Pile at the General Scrap site (85.4 ppm) and storage cells 5 and 6 at the XPotential site (respectively, 61.3 and 96.7 ppm). Wardrop was engaged by General Scrap and XPotential to have those duplicate samples that were taken alongside the Environment Canada samples for each of the West Pile and cells 5 and 6 analyzed for PCB concentration.

[24]            The second report prepared by Wardrop as a result of this engagement concluded, among other things, that:

1.          The concentration of ASR in the duplicate samples ranged from 10 to 42 ppm;

2.          For the West Pile, the 95% statistical confidence interval placed the PCB concentration between 30.6 and 38.0 ppm;

3.          For cell 5, the sole sample showed a PCB concentration of 34 ppm; and

4.          For cell 6, the 95% statistical confidence interval placed the PCB concentration between 11.9 and 26.7 ppm.

[25]            Wardrop concluded that the ASR in question did not have a PCB concentration above the regulatory threshold.

[26]            In February of 1998, Environment Canada sent a letter to the respondents advising that on the basis of the 1997 Environment Canada Report it had determined that General Scrap and XPotential were storing PCB materials in violation of the Regulations. However, the analytical protocol used by Environment Canada which resulted in the 1997 Environment Canada Report was not the protocol referenced in the Regulations. Therefore, Environment Canada later concluded that it could not rely upon its 1997 report for enforcement purposes. Environment Canada therefore agreed to withdraw its determination that General Scrap and XPotential were storing PCB materials in violation of the Regulations.

[27]            In September of 1998, Environment Canada advised that it would be conducting another round of sampling and analysis. Representatives of Environment Canada attended at the General Scrap and XPotential sites to conduct a sampling program from September 21 to September 25, 1998. The activities of Environment Canada were monitored by an environmental engineer employed by Wardrop and by staff of General Scrap and XPotential.

[28]            By December of 1998, General Scrap and XPotential retained the services of Dillon Consulting Limited ("Dillon") in order to facilitate discussions with Environment Canada in an attempt to resolve the outstanding issues.

[29]            In January of 1999, Environment Canada's final sampling report ("1999 Environment Canada Report") was issued. This report indicates that two ASR piles at the General Scrap site and three of the four storage areas at XPotential had average PCB concentrations that exceeded 50 ppm. The average PCB concentrations for the PCB contaminated piles, along with the 95% confidence (probability) ranges were as follows:

Average                                    95% Probability

(Absolute minimum)                               (lower value to upper value)

__________________________________________________________________

General Scrap

East pile 59.7 ppm                                                    67.0 to 82.8 ppm

Central pile             41.2 ppm                                                    46.3 to 57.1 ppm

West pile                 54.6 ppm                                                    61.3 to 75.7 ppm

XPotential

Cell 4                        65.3 ppm                                                    73.3 to 90.6 ppm

Cell 5                        50.7 ppm                                                    56.9 to 70.3 ppm

Cell 6                        42.0 ppm                                                    47.2 to 58.3 ppm

Drying Cell             50.8 ppm                                                    57.0 to 70.5 ppm

[30]            On February 8, 1999, the 1999 Environment Canada Report was forwarded by Environment Canada to General Scrap.

[31]            In April of 1999, General Scrap and XPotential advised Environment Canada that:

1.          General Scrap and XPotential would undertake a further sampling and characterization of the ASR identified by Environment Canada as PCB material;

2.          General Scrap would implement a procedure to identify common sources of PCBs in its scrap and to take steps to control PCB imports by requiring its suppliers first to remove potential PCB sources from household appliances analytical equipment before delivery to General Scrap.

3.          Soil and groundwater data would be assembled from the General Scrap and XPotential sites contiguous to ASR storage and would be provided to Environment Canada.

4.          General Scrap committed to continue to monitor new and emerging practical, economically achievable technologies for low-level PCB removal from ASR.

[32]            Environment Canada responded by letter dated April 30, 1999. In this correspondence Environment Canada took the position that General Scrap had failed to provide a detailed compliance plan with specific steps and time frames for putting this plan into effect. This was taken by Environment Canada to evidence a lack of commitment towards coming into compliance with the Regulations.

[33]            Jan Merks of Matrix Consulting Limited was engaged by General Scrap and XPotential as an expert consultant to assist Dillon in the design of the sampling program and in the interpretation of the data generated from the sampling and analysis.

[34]            In September of 1999, General Scrap and XPotential informed Environment Canada that Dillon had prepared a sampling program which it intended to initiate commencing on September 27, 1999. This program had been completed with the assistance of Mr. Merks.

[35]            Dillon carried out the sampling plan between September 27 and October 8, 1999 and began by sampling cell 5 at the XPotential site. Dillon applied three different sampling methods with a view to determining which method would provide the most accurate results. The first method replicated Environment Canada's method of direct sample selection (for comparison purposes), the second method used a mechanically split discrete sample, and the third method used mechanically split interleaved, composite samples. On the basis of this sampling and the analysis of the resulting data Dillon subsequently recommended a protocol to be followed in the final sampling and analysis of ASR at the General Scrap and XPotential sites.

[36]            On May 17, 2000, representatives of General Scrap, Dillon and XPotential met with an Environment Canada enforcement officer to review the results of the sampling and analysis of cell 5 and to advise that the remaining sampling analysis would proceed in accordance with the plan developed by Dillon.

[37]            On June 14, 2000, Environment Canada advised that it was unwilling to provide any comments on the proposed ASR sampling and analysis procedures proposed by Dillon.

[38]            General Scrap and XPotential instructed Dillon to proceed with the sampling and analytical protocol that Dillon proposed. In July of 2000, Environment Canada was advised that the remaining sampling and analysis was proceeding.

[39]            Environment Canada commenced these proceedings on December 6, 2000. Prior to seeking injunctive relief no other enforcement options available under the Act were pursued by Environment Canada.

[40]            The results of Dillon's sampling and analytical program are contained in a report prepared by Dillon dated January 2001. Dillon concluded that the ASR of concern to Environment Canada stored within the General Scrap - East Pile, General Scrap - West Pile, XPotential cells 4 and 5 and the XPotential drying cell, do not contain mean concentrations of PCBs above the federal or provincial regulatory threshold of 50 ppm.

(vi) Current Status of the ASR in Issue

[41]            In response to an application for an interim injunction made by Environment Canada regarding the storage and handling of the ASR at issue pending the determination of this application, the respondents agreed to the terms of a consent order setting out the manner in which the ASR at issue would be dealt with. The consent order was entered into with no admission by the respondents that the ASR at issue is PCB material under the Regulations.

(vi) The Specific Order Sought by the Applicant

[42]            The applicant says that it is not possible to determine visually which location in the problematic piles and cells have concentrations of PCB over 50 ppm. However, the applicant also says that regulatory compliance requires the PCB material to be identified and then stored, labelled, reported and handled in accordance with the Regulations.

[43]            Accordingly, in its written materials the applicant seeks an injunction on the following terms:

a)              All of the ASR in the East and West Pile at General Scrap, and all of the ASR in Cells 4, 5 and the Drying Cell at XPotential must be considered to be PCB Material, unless and until those areas that are under regulatory threshold are identified and separated from the Piles and Cells;

b)             Any protocol for the sampling, analysis and identification of non-PCB Material must be carried out by a third party under the direction and supervision of Environment Canada and at the expense of the Respondents;

c)              The remaining PCB Material must be stored or disposed of in accordance with the PCB Regulations.

[44]            Alternatively, the applicant seeks an injunction on such other or further terms as the Court considers just. No such terms were suggested. The notice of application seeks an order compelling the respondents to store all PCB material as required by law.

II. THE ISSUES

[45]            In order to determine whether the injunction requested should issue, the following issues must be considered:

i)           What are the requirements which must be met in order to be entitled to obtain an injunction pursuant to subsection 311(1) of the Act and what is the scope of subsection 311(1)?

ii)          Is some or all of the ASR PCB material as defined in the Regulations so that some or all of the respondents are in breach of the Regulations?

iii)          If so, which respondents are in breach of the Regulations? and

iv)         Should the Court grant the requested injunction?

III. ANALYSIS

(i) What are the requirements to be met in order to obtain injunctive relief under subsection 311(1) of the Act and what is the scope of subsection 311(1)?

[46]            Counsel advise that this is the first occasion on which a court has considered a request for injunctive relief under subsection 311(1) of the Act. Section 311 of the Act is as follows:

(a) Is subsection 311(1) to be limited to exceptional cases?

[47]            The respondents' fundamental argument with respect to the application of section 311 is that the Act in its entirety is enacted pursuant to Parliament's jurisdiction over criminal law matters. It follows, they submit, that the jurisdiction of the court to grant an injunction in aid of criminal law is a jurisdiction to be used with caution and only in the most exceptional of cases. Reliance is placed upon authorities such as Gouriet v. Union of Post Office Workers, [1978] A.C. 435 (H.L.), Ontario (Attorney General) v. Ontario Teachers' Federation (1997) 36 O.R. (3d) 367 (H.C.) and Attorney General for the Province of Ontario v. Hale (c.o.b. Hale Sand and Gravel), (1983) 38 C.P.C. 292 (On. H.C.). The respondents say that the exercise of this exceptional jurisdiction has been confined to cases where a law has been repeatedly flouted, the alleged breach of law is clear, and the enforcement provisions of the statute in question have proven ineffective.

[48]            With respect, I find the authorities relied upon by the respondents to be distinguishable. In the cases cited by the respondents there was either no specific legislative provision which authorized injunctive relief, or, as in Hale, supra, the statutory provision which provided for injunctive relief was not applicable. Therefore, in all of the cases what was in issue was the right of an Attorney General to sue at common law in order to attempt to enforce a law by way of injunction.

[49]            The nature of an injunction available at common law to an Attorney General in order to enforce public rights is well described by Justice MacPherson in Ontario Teachers' Federation, supra. This remedy reflects the role of the Attorney General in securing compliance with the laws of the land. Courts have held this to be a remedy granted in exceptional cases.

[50]            There is, however, a significant distinction between an injunction authorized by statute and an injunction available to the Attorney General at common law. This distinction is aptly illustrated in Ontario (Minister of the Environment) v. National Hard Chrome Plating Co. (1993), 11 C.E.L.R. (N.S.) 73 (Ont. C. of J.). There, the statutory provision with respect to the granting of an injunction contemplated an injunction to "restrain" contravention of the statute. The Court concluded that because the statute only provided a basis for the issuance of a prohibitory injunction, a mandatory injunction was only available at common law at the request of the Attorney General suing in the public interest. Such common law relief was available only where the law was being flouted and the legislation was inadequate to protect the public interest.

[51]            On the basis of the authorities cited by the parties I am satisfied that where a statute provides a remedy by way of injunction, different considerations govern the exercise of the court's discretion than apply when an Attorney General sues at common law to enforce public rights. The following general principles apply when an injunction is authorized by statute:

i)           The court's discretion is more fettered. The factors considered by a court when considering equitable relief will have a more limited application. See: Prince Edward Island (Minister of Communications and Cultural Affairs) v. Island Farm and Fish Meat Ltd., [1989] P.E.I.J. No. 32 (P.E.I.S.C.); Maple Ridge (District) v. Thornhill Aggregates Ltd. (1998), 162 D.L.R. (4^th) 203 (B.C.C.A.).

ii)          Specifically, an applicant will not have to prove that damages are inadequate or that irreparable harm will result if the injunction is refused. See: Shaughnessy Heights Property Owners' Association v. North Up (1958) 12 D.L.R. (7d) 760 (B.C.S.C.); Manitoba Dental Association v. Byman and Halstead (1967) 34 D.L.R. (2d) 602 (Man. C.A.); Canada (Canadian Transportation Accident Investigation and Safety Board) v. Canadian Press, [2000] N.S.J. No. 139 (N.S.S.C.).

iii)          There is no need for other enforcement remedies to have been pursued. See: Saskatchewan (Minister for Environmental Assessment Act) v. Redberry Development Corporation, [1987] 4 W.W.R. 654 (Sask. Q.B.).

iv)         The Court retains a discretion as to whether to grant injunctive relief. Hardship from the imposition and enforcement of an injunction will generally not outweigh the public interest in having the law obeyed. However, an injunction will not issue where it would be of questionable utility or inequitable. See: Saskatchewan (Minister of Environmental Assessment Act) v. Redberry, supra; Maple Ridge (District) v. Thornhill Aggregates Ltd., supra; Capital Regional District v. Smith (1998), 168 D.L.R. (4^th) 52 B.C.C.A.

v)          It remains more difficult to obtain a mandatory injunction. See: Canada (Canadian Transportation Accident Investigation and Safety Board) v. Canadian Press, supra.

(b) The constituent elements of subsection 311(1)

[52]            Having rejected the respondents' argument that relief pursuant to section 311 of the Act is only available upon proof that the law has been repeatedly flouted and that other enforcement provisions of the statute have proven ineffective, I turn to consider what must be established in order to permit the Court to issue a mandatory injunction.

[53]            The starting point for the interpretation of subsection 311(1) of the Act is the following well-known and accepted statement of principle:

Today there is only one principle or approach, namely, the words of an Act are to be read in their entire context and in their grammatical and ordinary sense harmoniously with the scheme of the Act, the object of the Act, and the intention of Parliament.

See: E.A. Driedger in Construction of Statutes (2^nd ed. 1983) at page 87 as cited in Chieu v. Canada (Minister of Citizenship and Immigration), [2002] 1 S.C.R. 84 at paragraph 27.

[54]            This approach requires a court to attribute to a legislative provision the meaning that best accords with both the text and the context of the provision. While neither can be ignored, as the Federal Court of Appeal observed in Bristol-Myers Squibb Co. v. Canada (Attorney General), 2003 FCA 180 at paragraph 13, the clearer the ordinary meaning of the provision, the more compelling the contextual considerations must be in order to warrant a different reading.

[55]            Before beginning this analysis it is convenient to again set out the text of subsection 311(1) of the Act:

1. The Text - Grammatical and Ordinary Sense

[56]            An ordinary reading of subsection 311(1) leads to the interpretation that for an injunction to issue it must appear to a court that either:

i)           the respondent has done any act or thing constituting an offence under the Act, or done any act or thing that is directed toward the commission of an offence; or

ii)          the respondent is about to do, or is likely to do, any act or thing that constitutes an offence under the Act or is directed toward the commission of an offence.

[57]            If so satisfied, the court may:

i)           issue a prohibitory injunction restraining the respondent from doing any act or thing that it appears to the court may constitute or be directed toward the commission of an offence; or

ii)          issue a mandatory injunction requiring the respondent to do any act or thing that it appears to the court may prevent the commission of an offence.

The Text as it Speaks to the Requisite Standard of Proof

[58]            The respondents argue that the Act requires proof beyond a reasonable doubt of the facts giving rise to the commission of an offence. In my view, the language used in subsection 311(1), read in its grammatical and ordinary sense, does not support this conclusion. I so conclude because the provision speaks to the situation where "it appears to a court" that an act or thing has occurred or is about to occur or is likely to occur, and that act or thing constitutes or is directed toward the commission of an offence. If proof beyond a reasonable doubt of the commission, or likely commission, of an offence was required it is reasonable to infer that Parliament would have used more specific language in the nature of "where it is established that a person has done or is about to do or is likely to do any act or thing constituting an offence".

[59]            Further, the court may restrain any act or thing that "appears to the court may constitute or be directed toward the commission of an offence". The court may order anything to be done that "may prevent the commission of an offence". This wording again falls short of requiring proof beyond a reasonable doubt that an offence has occurred or is about to occur or is likely to occur.

[60]            Moreover, the wording used in subsection 311(1) is to be contrasted with that found in section 39 of the Act. Section 39 permits a person who suffers, or is about to suffer, loss or damage "as a result of conduct that contravenes any provision of this Act" to apply to a court for injunctive relief. Section 39 therefore requires that the court be satisfied that loss or damage results from conduct that "contravenes the Act" in order to grant injunctive relief. The use of wording in subsection 311(1) which only requires that it "appears" that an offence has occurred, or is about to or likely to occur, must be taken to reflect Parliament's intent that a lower degree of proof is required under section 311 than is required under section 39. That lower degree of proof would not equate to proof beyond a reasonable doubt, or even proof at the high end of the civil standard.

[61]            In so concluding, I have considered the respondents' argument that section 29 of the Act supports the conclusion that section 311 requires proof to the criminal standard. Section 29 provides:

[62]            The respondents argue that because section 311 and related provisions contain no similar provision invoking the civil standard, the standard of proof must be intended to be the criminal standard.

[63]            However, it is significant, in my view, that an environmental protection action referenced in section 29 may only be brought by a person who has applied to the Minister for an investigation of an offence and the Minister has either failed to investigate and report as required or has responded unreasonably to the investigation. An environmental protection action is therefore a form of substitution for a proper investigation of an alleged offence. The gravamen of the action is proof of an offence. In that circumstance, the need for clarification of the standard of proof is apparent. Viewed in this context I am not prepared to infer from the absence of a similar provision applicable to section 311 that the criminal standard of proof was intended to apply to section 311.

[64]            To conclude on this point, I also observe that nothing in section 311 indicates that the application commenced by the Minister is criminal in nature so as to attract the criminal standard of proof.

[65]            On the other hand, Environment Canada asserts that it need only establish that there is reason to believe that a violation of the Act is occurring. It is said by Environment Canada that in the absence of an express statutory requirement it is not necessary for it to prove reasonable and probable grounds for that belief. In the words used in Environment Canada's written submission:

31.            Environment Canada need only prove that they [sic] have reason to believe a violation of the Act is occurring. Unless the statute indicates otherwise, reasonable and probable grounds for such belief or actual proof of the violation is not required.

Prince Edward Island (Minister of Community and Cultural Affairs) v. Island Farm and Fish Meal Ltd. (1989), 79 Nfld. & P.E.I.R. 228 (P.E.I.C.A.)

[66]            While I reject the respondents' submission that what is required is proof beyond a reasonable doubt of facts giving rise to the commission of an offence, I also reject the submission of Environment Canada that it is not necessary for it to establish reasonable and probable grounds upon which to base a belief that a violation of the Act has occurred, or will occur, or will likely occur.

[67]            The ordinary meaning of the words used in subsection 311(1) places the onus on the Minister, as moving party, to satisfy a court of competent jurisdiction that it appears that an act or thing constituting or directed toward the commission of an offence has occurred or is about to occur or is likely to occur. If so satisfied the court may enjoin any act or thing that it appears may constitute or be directed toward the commission of an offence. Alternatively, the court may mandate any act or thing that it appears may prevent the commission of an offence. While the language used falls short of requiring proof that an offence has occurred or will occur, it is necessary for the court at least to come to a bona fide belief, on a balance of probabilities, that a serious possibility exists that an offence has been committed, or is likely to be committed, or conduct directed toward the commission of an offence has occurred or will likely occur unless an injunction is issued. Unless the court is so satisfied, the Minister will have failed to establish the existence of facts that make the commission of an offence or conduct furthering an offence appear likely. The court's belief must be based on credible evidence, and any inferences that such evidence properly supports. The onus is upon the Minister to meet that burden.

[68]            I do not find the Island Farm and Fish Meal case relied upon by Environment Canada to assist its position. This case turned upon the wording of the specific provincial legislation which expressly allowed a ministerial order to be issued simply where the Minister had reason to believe that a violation had occurred. The case is not authority for any broader proposition applicable to subsection 311(1) because of differences in the language used in each statute.

The Text as it Speaks to the Prevention of Offences

[69]            Environment Canada submits that the language of subsection 311(1) when read in its entirety is directed toward the prevention of offences under the Act. I agree. It does so by allowing a court to prohibit acts or things that may constitute or be directed to the commission of an offence and by allowing a court to order that any act or thing be done where it appears that the resulting effect of the order may prevent the commission of an offence.

[70]            For example, in the case of a single discrete act that constitutes or may constitute an offence under the Act, on proper evidence the court could enjoin the act or could order that any act or thing be done so as to prevent the occurrence of the offence. However, if that single, discrete act had already taken place, there would be no scope for the application of subsection 311(1) because there would be no act to restrain and no way to prevent the commission of an offence after the fact.

[71]            What then of the case where a past or present act has a present and ongoing effect? For example, if the ASR at issue now constitutes, in whole or in part, PCB material can prohibitory or mandatory injunctive relief be granted?

[72]            Environment Canada argues that in such a situation a court may (pursuant to paragraph 311(1)(a) of the Act) by injunction require a respondent to discontinue acting in a manner that constitutes or is directed to the commission of an offence. The ordinary meaning of the words used in paragraph 311(1)(a) support that submission in the following manner.

[73]            The offence at issue in this case is the alleged failure to store PCB material in conformance with the Regulations. This is an offence pursuant to paragraph 272(1)(a) of the Act, which makes it an offence to contravene a provision of a regulation to the Act. Paragraph 311(1)(a) would therefore apply where, to paraphrase the language of the provision, it appears to the court that a person has stored PCB material in contravention of the Regulations. In that circumstance, the court could order that the respondents refrain from storing PCB material in contravention of the Regulations because this non-compliant form of storage would constitute the commission of an offence. While the English version is not explicit with respect to an ongoing situation, the French version of paragraph 311(1)(a) expressly provides that one may be ordered to abstain from all acts capable of continuing or perpetuating an offence. Thus, the use of the phrase "de s'abstenir de tout acte susceptible [...] de perpétuer le fait [...]".

[74]            In this type of situation, the court could as well order pursuant to paragraph 311(1)(b) that the material be stored in accordance with the Regulations if satisfied that this would "prevent the commission of an offence".

[75]            In this connection, Environment Canada argues that the offence of improper storage of PCB materials is an offence of a continuous nature as contemplated by section 276 of the Act. Section 276 provides:

[76]            It follows in the present case, in the submission of Environment Canada, that while any offence of failing to comply with the Regulations occurred before the proceeding was commenced, it continues to occur. So long as PCB material remains stored in a manner inconsistent with the Regulations, Environmental Canada argues that the offence continues and a fresh, separate and discrete offence occurs each day. Thus, a mandatory injunction may issue requiring PCB material to be stored as stipulated in the Regulations in order to prevent the commission of a fresh, separate offence.

[77]            In response, the respondents argue that section 276 of the Act does not in its language provide that the continuation of an offence is itself a separate offence. Rather, the section provides that "the person who committed the offence is liable to be convicted for a separate offence for each day on which it is committed or continued". This is said to allow for the multiplication of penalties and for the continuation of any limitation period. The wording used in section 276 is contrasted by the respondents with section 78.1 of the Fisheries Act, R.S.C. 1985, c. F-14. The environmental provisions contained in the Fisheries Act are noted by the respondents to be administered by Environment Canada. Section 78.1 of the Fisheries Act provides:

[78]            The Fisheries Act pre-dates the Act and therefore could provide support for the respondents' submission that the wording used in section 276 means something other than the continuation of a contravention of the Act constitutes a separate offence for each day it persists. Notwithstanding that submission, section 276 of the Act does expressly state that the continuation of an offence over more than one day renders the offender liable to be convicted for separate offences.

[79]            It is at this point helpful to turn to the contextual factors that aid in the interpretation of subsection 311(1) and that specifically should aid in interpreting whether subsection 311(1) permits the issuance of a mandatory injunction in order to remedy an existing situation.

2. The Context

The Scheme of the Act

[80]            The Act may be described as being a comprehensive code respecting pollution prevention and the protection of the environment.

[81]            Part 2 of the Act deals with public participation in the administration and enforcement of the Act. An individual who is at least 18 years of age and a resident of Canada may request an investigation of an alleged offence under the Act (section 17). As referred to above, should the responsible Minister fail to conduct an investigation, or respond unfavourably, and if there has been significant harm to the environment, then the individual who requested the investigation may proceed with an environmental protection action (section 22). Subsection 22(3) of the Act provides that in such an action the individual may claim any or all of the following relief:

(a) a declaratory order;

(b) an order, including an interlocutory order, requiring the defendant to refrain from doing anything that, in the opinion of the court, may constitute an offence under this Act;

(c) an order, including an interlocutory order, requiring the defendant to do anything that, in the opinion of the court, may prevent the continuation of an offence under this Act;

(d) an order to the parties to negotiate a plan to correct or mitigate the harm to the environment or to human, animal or plant life or health, and to report to the court on the negotiations within a time set by the court; and

(e) any other appropriate relief, including the costs of the action, but not including damages.

[82]            Of significance is the explicit reference in paragraph 22(3)(c) to an order "requiring the defendant to do anything that [...] may prevent the continuation of an offence" under the Act. [underlining added]

[83]            Section 39 of the Act, also previously referred to, allows a person who suffers, or is about to suffer, loss or damage as a result of conduct contravening the Act to seek an injunction. Such an injunction may require the contravenor to refrain from doing anything that it appears causes or will cause the loss or damage, or may require the contravenor to do anything that it appears prevents or will prevent the loss or damage.

[84]            Part 10 of the Act deals with the enforcement of the Act and provides extensive enforcement powers. The responsible Minister may designate enforcement officers who have peace officer powers (section 217). An enforcement officer may enter and inspect any place where there are reasonable grounds to believe there might be a substance or activity regulated under the Act (section 218). During the course of an investigation or search, an enforcement officer may issue an Environmental Protection Compliance Order ("EPCO") where there are reasonable grounds to believe that any provision of the Act or Regulations has been contravened by a person who is continuing the commission of the offence. Section 235 deals with EPCOs and is, in material part, as follows:

[85]            An EPCO may therefore direct that a person who is continuing the commission of an offence to "do anything to comply with this Act or the regulations".

[86]            A person to whom an EPCO is directed shall immediately on receipt of the order comply with it (subsection 238(1)). An EPCO is valid for up to 180 days (subsection 235(7)). The affected person may make representations to the enforcement officer before the order is issued, or seek review of the order by an independant review officer (sections 243-268). The order remains in effect until the review officer otherwise rules. An appeal from that decision lies to the Federal Court. Where a person fails to take any measures specified in an EPCO an enforcement officer may cause those measures to be taken (subsection 239(1)).

[87]            The maximum penalty under the Act for an offence is a fine of up to $1 million dollars per day or up to 5 years imprisonment (subsections 273(2) and 274(1)). A court can also levy a fine equal to any profits earned as a result of the offence. (section 290) Subsection 291(1) of the Act provides that upon conviction for an offence under the Act the court may make additional orders. Subsection 291(1) is as follows:

[88]            Relevant are paragraphs 291(1)(a), 291(1)(b) and 291(1)(f). Paragraph 291(1)(a) allows a court to prohibit any act that may result in "the continuation" of the offence, paragraph 291(1)(b) allows the court to order that an offender take any action considered by the court appropriate to remedy any harm to the environment resulting from the offence and paragraph 291(1)(f) allows the court to direct an offender to have an environmental audit conducted as prescribed by the court and to further direct the offender to remedy any deficiencies revealed during the audit.

[89]            The respondents argue that a review of the legislative scheme contained in the Act reveals that where Parliament intends the Act to be directed to continuing offences it uses express language in that regard. Thus, paragraph 22(3)(c) of the Act authorizes a court to require a defendant to an action to do anything to "prevent the continuation of an offence", subsection 235(1) enables an EPCO to be issued where there are grounds to believe the Act or its Regulations have been breached "by a person who is continuing the commission of the offence" and paragraph 291(1)(a) allows a court to prohibit an offender from doing any act that may "result in the continuation [...] of the offence". The absence of explicit reference in the English version of section 311 of the Act to prevention of the "continuation of an offence" is said to reflect Parliament's intent that section 311 not apply in that circumstance.

[90]            The respondents also argue, inferentially, that orders issued pursuant to sections 22 and 291 ordering a defendant to do anything that may prevent the continuation of an offence or refrain from any activity that may result in the continuation of an offence are issued following judicial determination that an offence under the Act has occurred. As noted above, on a proceeding under subsection 311(1) of the Act it is not necessary for the Minister to prove that an offence has occurred, but only that it appears that a person has done, is about to do, or is likely to do an act or thing constituting or directed toward the commission of an offence under the Act. The respondents argue, and I accept, that it is to some degree anomalous for the same relief to be available pursuant to subsection 311(1) when it "appears" that an offence has occurred or may occur as is available after conviction of an offence under section 291 of the Act. However, an EPCO may issue ordering someone to refrain from contravening the Act or to comply with the Act simply where an enforcement officer has reasonable grounds to believe that the Act or its Regulations have been contravened. While an EPCO is of limited duration, an EPCO may nonetheless direct a person to comply with the Regulations and an enforcement officer may cause the measure to be taken. This reflects a legislative intent that sweeping remedial measures may be ordered on less than proof that an offence has occurred.

[91]            The anomaly asserted by the respondents does result in the situation that the procedural safeguards which would be available on a prosecution for an offence under the Act can be avoided where the Minister elects to bring proceedings for injunctive relief under subsection 311(1) of the Act. The safeguards available to a person prosecuted for an offence include the right to full and proper disclosure from the Crown, the presumption of innocence and the right to be proven guilty beyond a reasonable doubt. Proceedings under subsection 311(1) of the Act, commenced in this Court by application, carry with them no right to discovery of the Crown, no presumption of innocence and, I have found, a civil standard of proof.

[92]            This anomaly and the potential for abuse would be avoided by interpreting section 311 to be prospective and pro-active in nature, operating to maintain the status quo by restraining future acts or directing future acts so as to prevent the commission of an offence, other than one that has occurred and is ongoing. Under this interpretation section 311 would not be available to prevent an offence which has already been committed whether or not it is a continuing offence. In that circumstance, the issuance of an EPCO or the commencement of a prosecution would be appropriate enforcement options.

[93]            Such an interpretation would not, the respondents argue, be inconsistent with the objects and intent of the Act.

[94]            The objects of the Act may be taken from the preamble to the Act. There, Parliament expressed the commitment of the federal government to, among other things, "pollution prevention as a national goal" and to "implementing the precautionary principle that, 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". Those principles are said to be consistent with interpreting subsection 311(1) as applying to the prevention of offences and not to the remediation of already existing situations. The contrary interpretation is not necessary in order for there to be effective remedial tools. The provisions of the Act with respect to environmental protection actions, EPCOs and the scope of an order that may be made on conviction for an offence would provide full enforcement options and remedies.

(c) Conclusion with respect to the scope of subsection 311(1) of the Act

[95]            The use of inconsistent language in the Act makes it, in my respectful view, more difficult to attribute meaning to section 311 in the case of an existing and continuing situation. The failure to consistently and expressly refer to the prevention of the continuation of an offence and the use of language in section 276 which differs from that found in section 78.1 of the Fisheries Act do provide a basis for the interpretation urged by the respondents.

[96]            I am concerned at the potential to use section 311 as an alternate to a prosecution under section 291 and so deprive a respondent of his or her rights to disclosure of the Crown's case, the presumption of innocence, and to proof of the Crown's case beyond a reasonable doubt.

[97]            However ultimately, in light of my assessment of the evidence, it is not necessary to reach a final conclusion as to whether a mandatory injunction may issue to prevent the ongoing continuation of an offence where what is sought is the remediation of an already existing, static situation. The language used in paragraph 311(1)(b), when read in conjunction with section 276, is capable of supporting the interpretation that each day the offence continues is a new offence that may be prevented by the issuance of a mandatory injunction. My analysis will proceed on this basis.

(ii) Is some or all of the ASR PCB Material?

[98]            I now turn to review the evidence before the Court and to state my findings with respect to that evidence.

[99]            The evidence before the Court is contained in the affidavits filed by the applicant and the respondents, together with the cross-examination of some, but not all, of the deponents. Pursuant to an order of the Court, all cross-examinations were conducted viva voce before me at the hearing of the application.

(a) The evidence in support of the application.

[100]        Environment Canada relies upon the affidavits of Mervin Fingas and Shannon Kurbis. Both were cross-examined.

1. Affidavit of Mervin Fingas

[101]        Dr. Fingas is a scientist and the Chief of the Environmental Emergencies Science Division of Environment Canada. He holds a doctorate degree in Environmental Sciences as well as an M.B.A. in general business and managerial sciences and a Master of Science degree in Chemistry. Since 1974 he has worked with Environment Canada as a scientist in spill technology. As such he has experience and training in the sampling, analysis and impact of environmental pollutants, including PCBs. He was the author of the 1999 Environment Canada Report.

[102]        Dr. Fingas designed and implemented Environment Canada's sampling and analysis program at the General Scrap and XPotential sites. After researching the scientific literature related to ASR sampling and analysis, and visiting an Ottawa area scrap shredding company, Dr. Fingas determined there were a number of specific and important considerations to be addressed in the sampling program. They were set out in his affidavit. Dr. Fingas concluded that samples from a sufficient number of locations within each ASR cell or pile would be required in order to achieve results that were statistically reliable and representative of each cell or pile.

[103]        Once sampling locations were selected an excavating device was used to access the ASR at all depths and a number of 2 litre samples were taken. The actual sampling and the analysis of the sampled ASR were both carried out under his supervision.

[104]        Dr. Fingas concluded that, at a minimum, the East and West ASR cells at the General Scrap site and storage cells 4 and 5 as well as the drying cell at the XPotential site have average PCB concentrations in excess of 50 ppm, and are therefore properly considered to be PCB Material. A more detailed statement of the results of his study is set out at paragraph 29 above. Dr. Fingas concluded that the ASR is fairly uniformly contaminated with PCBs and his conservative estimate was that there are 1,300 kg of PCBs in the ASR.

[105]        In light of this, it was Dr. Fingas' opinion that the improper storage of PCB material poses a risk to human, plant and animal health and safety.

[106]        With respect to the sampling program developed by Dillon in consultation with Jan Merks, Dr. Fingas disagrees that the methodology used by Dillon produced a more correct result. It is Dr. Fingas' opinion that the fact that the results produced using this methodology give rise to a lower standard deviation simply reflects that standard deviation can be expected to be lower in a process that mixes many samples before taking a sub-sample.

2. Affidavit of Shannon Kurbis

[107]        Ms. Kurbis is an enforcement officer employed by Environment Canada. She detailed the chronology of events. Ms. Kurbis provided no independent scientific evidence with respect to the levels of PCB contamination, and relied upon the advice and opinion of Dr. Fingas with respect to the preparation of the sampling plan, and also with respect to the analysis, reporting and interpretation of the data collected by Environment Canada.

(b) The evidence in opposition to the application.

[108]        The respondents rely upon the affidavits of Jacob Lazareck, David Clark, Jan Merks, Michael Bertram, Detlef Birkholz and Donald Davies. Only Messrs. Lazareck, Clark and Bertram were cross-examined on their affidavits.

1. Affidavit of Jacob Lazareck

[109]        Mr. Lazareck is the president of XPotential and formerly was the vice-president of operations of General Scrap. Mr. Lazareck is a professional engineer who has been involved in the scrap recycling industry since 1970. He has been active in the Canadian Association of Recycling Industries, which is an association whose membership includes most scrap shredders, serving at one point as its first vice-president. Mr. Lazareck has also participated in the Institute for Scrap Recycling Industries, which represents scrap recycling industries in the United States.

[110]        Mr. Lazareck described the business carried out by General Scrap and XPotential, described the sites on which each stored ASR, and detailed the chronology of events.

[111]        With respect to steel recycling and ASR, to Mr. Lazareck's knowledge Environment Canada has never conducted inspections or sampling of ASR at other scrap shredding operations in Canada.

[112]        Mr. Lazareck provided no specific evidence with respect to the levels of PCB contamination in the ASR at either the General Scrap or the Xpotential sites.

2. Affidavit of David Clark

[113]        Mr. Clark is a professional engineer, and is the Regional Managing Partner of Dillon for the North/West region, and its senior environmental advisor. He holds a Bachelor of Science degree in engineering and a Master's degree in environmental engineering. He has over 20 years of experience in the fields of human health/environmental risk assessment; environmental impact investigations, and site assessments and remediation. Prior to entering private consulting with Dillon, Mr. Clark was the Director of the Land Protection Branch with the Saskatchewan Department of Environment. In that position, he was responsible for solid and hazardous waste management, chemical management, spill response and remediation programs. He served for 5 years as the Saskatchewan representative on the Canadian Council of Ministers of the Environment Waste Management Steering Committee. This committee was responsible for the development of protocols and codes of good practice for waste stream characterization, site decommissioning, PCB management, and the management of fuel storage and distribution systems.

[114]        Dillon was retained to characterize the ASR contained in the East and West Piles at the General Scrap site and in cells 4 and 5 as well as the drying cell at the XPotential site. This required Dillon to design a sampling program, collect samples, provide the samples to Enviro-Test Laboratories for analysis and report and interpret the sampling and analysis results. Jan Merks was engaged as an expert consultant in the design of the sampling program and in the interpretation of the sampling and analysis data.

[115]        Mr. Clark concluded that the ASR currently stored at the 5 piles/cells in question do not contain mean concentration of PCBs above the regulatory threshold of 50 ppm.

[116]        Dillon initially took three sets of ASR samples from cell 5 at the XPotential site in order to test the precision of three protocols for field sampling and sample preparation. The three protocols tested were:

(i)          the protocol used by Environment Canada;

(ii)         discrete sampling using mechanically sample splitting; and

(iii)        a mechanically split then interleaved protocol prescribed by Mr. Merks.

[117]        The Environment Canada sampling protocol was found to have the highest variance and therefore, in Mr. Clark's opinion, was the least appropriate sampling plan. In his further opinion, the conclusions in the Dillon Report ought to be accepted over the conclusions in the 1999 Environment Canada Report for reasons including:

(i)          Dillon's field sampling and sample preparation generated representative samples of ASR material in a random and totally objective fashion while Environment Canada's field sampling and sample preparation was subjective; and

(ii)         Dillon's sampling plan (which employed stratified systematic sampling) is preferable to Environment Canada's sampling plan (which employed stratified random sampling) for the purpose of characterizing highly heterogeneous waste material such as ASR.

3. Affidavit of Jan Merks

[118]        Mr. Merks is a metrologist with extensive experience in the design and evaluation of sampling plans and in the statistical analysis of data in relation to a wide range of the industries and circumstances. He has been accepted as qualified to give expert testimony on the subjects of statistical analysis and sampling design in a number of court and administrative proceedings. He has written a text on the sampling and analysis of bulk solids. He has participated in the activities of a number of Canadian advisory committees to the international standards organization in relation to the sampling of bulk materials and the application of statistical methods. He has experience as an analytical chemist, a chief chemist and a laboratory manager.

[119]        Mr. Merks was initially engaged for the purposes of providing advice to Dillon on the design of sampling and sampling preparation procedures to be used by Dillon in the determination of the concentration of PCBs in the piles and cells as issue in this proceeding. Mr. Merks was also retained for the purposes of reviewing and commenting upon the sampling methodology and statistical methods employed by Environment Canada as described in the Fingas and Kurbis affidavits and he reviewed the results of the sampling and analysis of the ASR undertaken by Dillon.

[120]        It is Mr. Merks' opinion that statistical analysis of the data contained in the 1997 Environment Canada Report and in the 1999 Environment Canada Report does not provide proof that any cell or pile of ASR has a mean PCB concentration that statistically differs from the regulatory threshold of 50 ppm. Conversely, he opines that the statistical analysis of Dillon's results establish that the mean PCB concentrations of each pile or cell tested by Dillon are either significantly less than or statistically identical to the regulatory threshold.

[121]        It is Mr. Merks' opinion that the sampling program implemented by Dillon is to be preferred over the sampling program employed by Environment Canada for the following reasons:

1.          Dillon applied stratified systematic sampling which is the methodology that is most appropriate when sampling materials such as the ASR stored at the General Scrap and XPotential sites. The stratified random sampling used by Environment Canada is a methodology more appropriately used for applications such as the statistical quality control of consumer products.

2.          The stratified systematic sampling protocol applied by Dillon allows for a more rigorous statistical analysis of the analytical data generated from the sampling program.

3.          Dillon carefully homogenized its primary samples in order to minimize the variance of the selection stage. In contrast, Environment Canada homogenized its samples in a haphazard manner.

4.          Dillon's procedure not only reduces the variance of the sample preparation stage but also gives an unbiased estimate for the measurement variance.

[122]        Mr. Merks compared the test programs employed by Dillon and Environment Canada and his comparison showed that:

1.          Dillon's sampling procedure is proven to be significantly more precise than that applied by Environment Canada.

2.          A statistical comparison of the variances between and within the piles or cells for each of the Dillon Report, the 1997 Environment Canada Report and the 1999 Environment Canada Report proves that Dillon's variances are homogeneous, but the Environment Canada's variances are not.

3.          All of these inferences establish that Dillon's sampling program is significantly more precise than Environment Canada's.

[123]        While the 1999 Environment Canada Report states that Environment Canada applied procedures that were analogous to those recommended by the United States Environmental Protection Agency ("EPA"), in Mr. Merks' opinion Environment Canada implemented procedures that were fundamentally dissimilar to those recommended by the EPA. Environment Canada's departures from suggested EPA procedures were:

1.          Environment Canada failed to apply stratified systematic sampling.

2.          Environment Canada failed to apply, or improperly applied, the formula for determining the minimum number of samples required from each sampling unit suggested by the EPA in order to be able to draw statistically valid conclusions about PCB concentrations.

3.          Environment Canada failed to correctly calculate 95% confidence limits for the arithmetic mean concentration of PCBs.

4.          Environment Canada failed to apply the recommended EPA criteria that would have proved the test program used to produce the 1999 Environment Canada Report to be inconclusive.

[124]        Mr. Merks observes that the results of the 1997 Environment Canada Report should have given Environment Canada an indication of the extreme variability in its data.

[125]        Apart from these concerns, in Mr. Merks' view Environment Canada's conclusions are entirely invalidated as a result of Environment Canada's grossly erroneous "total error calculation". He views the principal deficiencies in Environment Canada's methods to be:

1.          Environment Canada erroneously used the certified value of the standard reference materials as its true PCB concentration.

2.          Environment Canada's attempt to use the analysis of the standard reference material as a calibration procedure was incorrect.

3.          Environment Canada erroneously applied a correction for percentage variances in violation of the fundamental properties of variances as defined in probability theory and applied statistics.

4. Affidavit of Michael Bertram

[126]        Mr. Bertram is a professional geoenvironmental engineer who was employed by Dillon at the material time. He has 9 years experience in environmental site assessment and remediation and contaminant characterization. He was present when samples were collected and he supervised the sample collection.

[127]        Mr. Bertram states that the sampling protocols as described in the 1999 Environment Canada Report differed from those employed by Dillon in 4 respects:

1.          Environment Canada, in collecting primary samples with a back-hoe, scraped the side of each test pit from top to bottom and mixed the scraped ASR in the bottom of the test pit with the back-hoe. The bucket of the back-hoe was then used to bring up the primary sample. In comparison, the procedure used by Dillon in all but 10 of the 127 test pits was to start the back-hoe bucket at the bottom of the test pit and to scrape the ASR into the bucket from the bottom of the test pit up along the wall of the test pit to the top to collect approximately 20 litres of ASR in the bucket.

2.          Environment Canada selected sub-samples by using a hand-held scoop to manually fill a 2 litre sample jar from the material in the back-hoe bucket. Dillon used a riffle splitter to randomly select sub-samples from each of the 20 litre primary samples to be placed into a 2 litre sample jar. The use of the riffle splitter substantially eliminated the potential for subjectivity and human bias in the sample selection process.

3.          Environment Canada collected up to 8 samples at each pile or cell. Dillon collected between 20 and 42 primary samples at each pile or cell.

4.          Environment Canada submitted discrete samples from each test pit for laboratory analysis. Dillon interleaved samples from even and odd numbered test pits for each sample section to generate a representative composite sample which was then submitted for laboratory analysis.

[128]        It is Mr. Bertram's opinion that the Environment Canada procedure of scraping down the test pit is a less reliable sampling method because finer material will accumulate at the bottom of the test pit from digging the test pit. Fine materials may contain PCB concentrations materially different from PCB concentrations in coarse materials. Further, because the test pit's width was usually the same as the width of the back-hoe bucket, adequate mixing would be improbable.

[129]        In Mr. Bertram's opinion Environment Canada's methodology of collecting its 2 litre samples by using a metal scoop would result in a non-representative volume of fine materials which would bias the samples in favour of fine materials. The samples would not therefore be representative of the ASR in each sampling unit. In addition, Mr. Bertram observes that manually sampling using a scoop unavoidably introduces human bias in selecting what material is to be placed in the sample jars.

5. Affidavit of Detlef Birkholz

[130]        Dr. Birkholz is the Executive Vice-President of Enviro-Test Laboratories. He holds a PhD in environmental toxicology and has 28 years practical experience in analytical chemistry and 13 years research and practical experience in environmental toxicology. He has been qualified as an expert witness on the subject of analytical chemistry, laboratory procedure and environmental testing in numerous court proceedings.

[131]        Enviro-Test Laboratories was retained to provide analytical services in respect of the sampling and analysis program undertaken by Dillon with respect to ASR samples collected from the General Scrap and XPotential sites. Dr. Birkholz reviewed the Enviro-Test analytical data for the quantification of PCB concentrations in the samples of ASR collected by Dillon and also reviewed the related quality assurance/quality control data. Upon his review of the data and quality control results it is Dr. Birkholz' opinion that the analytical results provided by Enviro-Test to Dillon are sound, reliable and accurate.

6. Affidavit of Donald Davies

[132]        Dr. Davies is the Vice-President, Scientific Programs with Cantox Environmental Inc. He holds a doctorate degree in nutrition and toxicology and has extensive experience in the preparation of health risk assessments in respect of a variety of industrial sites and in relation to a variety of substances. He has in the past appeared and been qualified as an expert witness on the subject of toxicology and health risk assessment in numerous court proceedings and in quasi-judicial hearings. He is certified by the American Board of Toxicology. On December 15, 2000, he attended at the General Scrap site and the XPotential site for the purpose of preparing a preliminary health and environmental risk assessment. It is his opinion that the PCBs contained in the ASR stored on the General Scrap and XPotential properties do not constitute a risk to the environment, the general public, employees and visitors at the sites, or to wildlife in the area. This opinion holds regardless of whether the level of PCB concentrations measured in the stored ASR is as reported by Dillon or by Environment Canada. This is, in his opinion, because the reported differences in the concentrations are of no significance from a health or environmental hazard perspective.

(c) Analysis of the Evidence

[133]        In oral argument counsel for the applicant observed that having read the materials filed by the parties it appears that there are two different cases before the Court, reflecting two completely different views on a common set of facts. This is very much the case.

[134]        In order for the applicant to succeed in this application Environment Canada must establish, on a balance of probabilities, that it appears that the ASR contains PCB material in a sufficient quantity that the Regulations require it to be labelled and stored in a particular fashion.

[135]        The analysis of the evidence logically begins with what is not in dispute. That is:

1.          For the Regulations to apply the ASR must be "PCB Material". This requires that the ASR be a PCB solid or PCB substance, as those terms are defined in the Regulations. This in turn requires that the ASR contain more than 50 mg of PCBs per kilogram (or more than 50 ppm), and there must be 100 kg or more of the PCB solid or substance.

2.          Neither the Act nor the Regulations prescribe any particular sampling strategy or methodology.

3.          In the present context a sampling strategy is the process which determines how many individual samples will be taken from each ASR cell or pile, and the locations at which they will be taken. Based on the analysis of those samples, inferences can be drawn about the ASR cell or pile as a whole, or about the other non-sampled portions of the ASR cell or pile. What is at issue is what is contained in the ASR which has not been analyzed.

4.          Environment Canada does not take issue with the individual analytical results which were obtained as a result of the respondents' sampling program, as reported by Dillon. Environment Canada takes issue with the respondents' interpretation and analysis of those results.

5.          The respondents take issue with the individual analytical results obtained by Environment Canada and with the analysis and interpretation of those results by Environment Canada.

6.          The central dispute between the parties is whether the average of an entire cell or pile, or smaller individual areas within the pile or cell referred to as hot spots should be used to determine whether there is 100 kg of ASR with a PCB concentration above 50 ppm.

[136]        In light of the submissions advanced by the parties, the evidence before the Court will be considered under the following headings:

1.          The validity of the Environment Canada Data.

2.          The use of individual results.

3.          The use of statistics.

1. The Validity of the Environment Canada Data

[137]        The respondents assert that the data contained in the 1997 Environment Canada Report and in the 1999 Environment Canada Report ought to be rejected, or that the data contained in the Dillon Report ought to be preferred, for the following reasons:

1.          The Environment Canada reports are based upon an insufficient number of samples and an inappropriate sampling system.

2.          Environment Canada departed from its own sampling plan, at least with respect to the West Pile at General Scrap.

3.          Environment Canada used an inferior method of sample selection.

[138]        Each assertion will be considered in turn.

Is the 1999 Environment Canada Report based upon an insufficient number of samples and an inappropriate sampling system?

[139]        Dr. Fingas swore in his affidavit and confirmed in cross-examination that he was required to "design and implement a sampling methodology using standard and proven procedures that would allow the sampling team to select samples from each of the large piles ("cells") of ASR at General Scrap and XPotential which, when analysed, would provide results that would be representative of each entire cell". Dr. Fingas determined that "samples from sufficient locations within each cell of ASR would be required to achieve results that were statistically reliable and representative of [...] each cell".

[140]        With respect to the number of samples, Dr. Fingas' report stated that he applied a statistical formula in order to calculate the minimum number of samples, and Dr. Fingas referenced two EPA studies in support of this approach: "Sampling Guidance for Scrap Metal Shredders: Field Manual, Environmental Protection Agency, 1992" ("1992 EPA Report") and "Test Methods for Evaluating Solid Waste, Chapter 9, Sampling Plan, Environmental Protection Agency, 1993" ("1993 EPA Report"). Both of these documents were cited as references in Dr. Fingas' report, and he confirmed on cross-examination that he listed as references whatever reports he used "very heavily" in the preparation of his report. Of the two documents, Dr. Fingas stated that he relied more upon the 1992 EPA Report as a result of advice received from the EPA.

[141]        As to the sampling system, at each site the pile or cell at issue was divided into units of 3 metre "cells". Using a random number selection process certain cells were selected for sampling and analysis. This was said to be a procedure analogous to that recommended by the Environmental Protection Agency 1992 and 1993 Reports.

[142]        Jan Merks, in his affidavit, swore with respect to the number of samples taken that Environment Canada failed to apply or inappropriately applied the formula recommended by the EPA for determining the minimum number of samples required by each pile or cell.

[143]        With respect to the sampling program, Mr. Merks swore that the sampling program implemented by Dillon is to be preferred over the Environment Canada sampling program because Dillon applied stratified systematic sampling which is a methodology most appropriate when sampling materials in bulk. In contrast, Environment Canada applied stratified random sampling which is a methodology more appropriate, in Mr. Merks' view, for applications such as the statistical quality control of consumer products. It was also Mr. Merks' view that the stratified systematic sampling protocol applied by Dillon is more precise for bulk materials and that the EPA suggested stratified systematic sampling.

[144]        The Dillon Report also expressed the view that stratified systematic sampling is the most suitable protocol, although in substance Dillon looked to Mr. Merks to design the investigation and sample preparation protocols.

[145]        Dr. Fingas was cross-examined on these issues. Mr. Merks was not. It was Dr. Fingas' evidence on cross-examination that in addition to relying upon the references cited in his report he met separately with representatives of the EPA who provided him with other reports. Dr. Fingas also testified that he sent his sampling plan to the EPA which probably provided verbal comments about the plan which he did not document. No testimony was given as to the nature of the comments provided.

[146]        With respect to the number of samples taken, Dr. Fingas confirmed that the 1993 EPA Report on sampling guidance for scrap metal shredders recommended that 20 samples be taken from each pile and that he took far fewer samples than that. He testified that "[w]e follow largely the official sampling manual [the 1992 EPA Report] [...] which is at odds to this [1993 EPA] manual". The 1992 EPA Report does not deal directly with ASR. Rather, the 1992 EPA Report sets out a statistical equation to be used for calculating the appropriate number of samples to collect from solid waste. Dr. Fingas set out that equation in his report. Dr. Fingas then used the examples from the literature to calculate the minimum number of samples and then doubled the resultant answer in order to determine the number of samples to be taken.

[147]        I am satisfied that the development of an appropriate sampling plan is of crucial importance in order to evaluate the PCB content of ASR. I reach that conclusion on the basis that the 1992 EPA Report, which Dr. Fingas relied heavily upon, states at page NINE-1:

The initial - and perhaps most critical - element in a program designed to evaluate the physical and chemical properties of a solid waste is the plan for sampling the waste. It is understandable that analytical studies, with their sophisticated instrumentation and high cost, are often perceived as the dominant element in a waste characterization program. Yet, despite that sophistication and high cost, analytical data generated by a scientifically defective sampling plan have limited utility, particularly in the case of regulatory proceedings. [underlining added]

[148]        Dr. Fingas agreed on cross-examination that analytical data generated by a scientifically defective sampling plan has limited utility.

[149]        To the extent that the evidence conflicts between that of Dr. Fingas and Mr. Merks with respect to the numbers of samples to be taken and the use of a stratified systematic sampling plan, I prefer the evidence of Mr. Merks. I prefer Mr. Merks' evidence because of his recognized expertise specifically with respect to the design and evaluation of sampling plans. In this connection he has been qualified as an expert witness on the subject of sampling design in numerous court and administrative proceedings and has written a text on the sampling and analysis of bulk solids. Mr. Clark testified that Mr. Merks' work with respect to waste characterization had been accepted by the British Columbia Ministry of Environment, Lands and Parks Branch.

[150]        While Dr. Fingas has expertise in sampling matters, and has written about 10 peer reviewed publications that "deal with sampling, with PCBs with issues of that type" his curriculum vitae describes his specialties to include "spill dynamics and behaviour, spill treating agent studies, in-situ burning and the technology of personal protection equipment". I am not satisfied that Dr. Fingas' knowledge and experience with respect to the design and evaluation of sampling plans is as extensive as that of Mr. Merks. I place little reliance on whatever verbal information or comments Dr. Fingas received from the EPA with respect to his sampling plan. This was input not cited in his report or affidavit, and apparently was not of sufficient importance to warrant being noted or recorded on Dr. Fingas' file. The specifics of any input he received were not put in evidence.

[151]        Moreover, the applicant did not cross-examine Mr. Merks upon his evidence that Environment Canada erred in calculating the minimum number of samples required and that the sampling program implemented by Dillon is to be preferred over the sampling plan implemented by Environment Canada. Nor was Mr. Merks cross-examined upon his evidence that Environment Canada did not use procedures analogous to those recommended by the EPA, but rather implemented procedures that were fundamentally dissimilar to those recommended by the EPA with respect to stratified systematic sampling. Mr. Merks' testimony on these matters is, therefore, unimpeached.

Did Environment Canada depart from its own sampling plan with respect to the West Pile at General Scrap?

[152]        The applicant concedes that the Environment Canada sampling plan was departed from with respect to the West Pile in that only surface samples were taken with a shovel from a small portion of the West Pile. The sampling plan contemplated using a back-hoe to take samples at randomly selected locations. Generally each hole sampled was of a depth varying from 2 to 5 metres so that each sample would be, in Dr. Fingas' words, "representative of a slice through that entire pile". By taking 5 slices through 5 different locations in a pile Dr. Fingas believed this "characterizes whether or not there is PCB material".

[153]        The departure from the sampling plan resulted from the fact that at the time Environment Canada took its samples, scrap metal had been placed on top of the pile making parts of it inaccessible.

[154]        The consequence of this departure would be to reduce the extent to which the "slices" taken from the West Pile were representative of the pile at that location. This is because only the surface of the slice was sampled. This would provide a basis for discounting the extent to which Environment Canada's test results properly represent the West Pile.

Did Environment Canada use an Inferior Method of Sample Selection?

[155]        To obtain its primary increments Environment Canada used a back-hoe to dig a hole through the ASR. Having exposed a vertical sidewall of the cell, the back-hoe scraped material down the side of the hole to the bottom of the hole where the scrapings were then mixed. The back-hoe operator then removed a bucket of material from the bottom of the hole. Workers inspected the bucket load and if pieces too big to fit into the sampling jar were found they were cut with snips. The sampling crew then mixed the back-hoe bucket sample with a shovel, and then selected random samples with a hand scoop and placed them in 2 litre sample bottles.

[156]        The respondents, on the other hand, dug with a back-hoe a hole through the ASR to expose the cell. The back-hoe then scraped the side of the hole upward from the bottom of the hole. The material contained in the back-hoe bucket was then inspected and large pieces were cut. Next, the material in the back-hoe bucket was mixed with a fork or shovel and then passed through a machine called a riffle splitter. The riffle splitter passed the sample material through it, randomly splitting the sample into two piles. Sample material was passed through the riffle splitter and spewed out at two ends of the splitter, designated as being odd and even. A random number generator was used to generate a list of random numbers. ASR was placed in the riffle splitter and, depending upon the next available random number, the even or odd pile was placed into the sample container. The non-chosen split pile was collected and then put through the splitter again. The process continued until the required 20 litre sample, the primary increment, was collected. The riffle splitter was used to maximize the degree of homogeneity of the sample.

[157]        As explained in Mr. Clark's affidavit, Dillon did a test in cell 5 in which the Environment Canada sample selection method was followed whereby samples were manually taken from the back-hoe bucket in the manner done by Environment Canada. The one difference with this test is that the back-hoe bucket had been filled by scraping up, not down, the side of the selected cell. The samples were analyzed and compared with the analysis of samples which had been taken and then mechanically split with the riffle splitter. The results of the test were as follows:

PCB ppm                      Standard

(mean)              Deviation

Mechanically

Split Sample                                           50.4                                18.6

Environment Canada method                  68.3                                59.4

[158]        In considering these respective methods, the parties agree that the purpose of mixing the ASR in the pit (by Environment Canada) in the back-hoe bucket, and with the riffle splitter (by the respondents) was to homogenize the ASR at the sampling location. The parties further agree that it is important that the primary increment be homogenized to the greatest extent possible.

[159]        Dr. Fingas swore that PCBs primarily reside in the oils within ASR and are concentrated in the fine granular ASR material and within the permeable upholstery material component of ASR. It follows, in Dr. Fingas' view, that subsampling which selects larger or more dense components of ASR lowers the apparent PCB content and taking only the finer components raises it.

[160]        It was the evidence of Mr. Merks that the riffle splitter carefully homogenized the primary samples while Environment Canada homogenized its samples in a haphazard manner. It was Mr. Clark's opinion that Environment Canada's field sampling was subjective while Dillon's was random and objective. Mr. Bertram, the professional engineer who supervised the sample selection, was of the opinion that:

1.          Fine material would accumulate at the bottom of the test pit as a result of digging the pit. As well, the test pit width would usually be the same width as the back-hoe bucket making adequate mixing improbable. For these reasons, he concluded that Environment Canada's sampling procedure was less reliable.

2.          Environment Canada's manual scooping of ASR samples into the sample jar would result in a non-representative volume of fine materials so that the samples would not be representative of the ASR in each sampling container.

[161]        Mr. Merks was not cross-examined on his opinion. Both Mr. Clark and Mr. Bertram were cross-examined, but I do not find their evidence to have been impugned on these issues.

[162]        Environment Canada argues that there is no analytical evidence that scooping down or up the pit is preferable, and no analytical evidence that the riffle splitter is preferable. That is correct, because the Dillon test of the Environment Canada method, which produced higher levels of PCB contamination, was based on a primary sample obtained by scraping up the side of the pit.

[163]        However, the 1992 EPA Report, which Dr. Fingas endorsed and followed, notes at page NINE-8 that:

It is important to emphasize that a haphazardly selected sample is not a suitable substitute for a randomly selected sample. That is because there is no assurance that a person performing undisciplined sampling will not consciously or subconsciously favor the selection of certain units of the population, thus causing the sample to be unrepresentative of the population.

[164]        Dr. Fingas agreed that PCBs are concentrated in the fine granular ASR material. Mr. Bertram's evidence, which I accept, is that Environment Canada's sampling method results in finer material accumulating at the bottom of the test pit, and that manual scooping of the ASR samples results in a non-representative volume of fine materials.

[165]        Taken together I am left with real concern that the Environment Canada method will result in higher levels of PCBs being reported, as was observed by the Dillon tests results set out above.

[166]        I place more reliance, therefore, on Dillon's method of sample selection.

[167]        One further issue may be considered under this heading and that is the propriety of the respondents' method of selecting analytical samples using the Merks' riffle split and the interleaving protocol. In order to submit a representative sample of collected ASR for laboratory analysis, 2 litre samples were prepared from the primary 20 litre samples by riffle splitting and interleaving composite samples.

[168]        It is Mr. Merks' opinion that compositing sets of primary increments into pairs of interleaved primary samples and preparing primary samples with the riffle splitter not only reduces the variance of the sample preparation stage, but also gives an unbiased estimate for the measurement variance.

[169]        Dr. Fingas disagrees and in his affidavit he swore that it is incorrect to conclude that the Dillon results are more correct on the basis that the standard deviation is lower. In Dr. Fingas' view the standard deviation can be expected to be lower in a process that mixes many samples before taking a sub-sample. Mixing the material reduces the variability and the standard deviation.

[170]        Environment Canada submits that the method used by Dillon serves to hide the true character of the material contained in a cell and that the only explanation for using the Merks interleaving riffle split protocol is because it artificially produces the lowest possible result.

[171]        Returning to the tests conducted by Dillon, referred to above, Dillon reported a mean PCB value of 50.4 ppm with a standard deviation of 18.6 for the mechanically split sample. When testing the mechanically split and interleaved samples a mean PCB value of 39.6 ppm was reported with a standard deviation of 10.8. This difference is significant.

[172]        In choosing whether to prefer the opinion of Dr. Fingas over that of Mr. Merks, I am left with the difficulty that it was Mr. Merks who prescribed the interleaving protocol (see: paragraph 23(c)(ii) of Mr. Clark's affidavit) but the applicant chose not to cross-examine Mr. Merks on this, or any part of his opinion.

[173]        What is also relevant, in my view, are the comments found in the 1992 EPA Report about composite sampling. The 1992 EPA Report was the report which Dr. Fingas relied upon in preference to the 1993 EPA Report on the basis of verbal advice he received from the EPA.

[174]        At page NINE-21 the following appears in the 1992 EPA Report:

In composite sampling, a number of random samples are initially collected from a waste and combined into a single sample, which is then analyzed for the chemical contaminants of concern. The major disadvantage of composite sampling, as compared with noncomposite sampling, is that information concerning the chemical contaminants is lost, i.e., each initial set of samples generates only a single estimate of the concentration of each contaminant. Consequently, because the number of analytical measurements (n) is small, s_x and t.₂₀ are large, thus decreasing the likelihood that a contaminant will be judged to occur in the waste at a nonhazardous level (refer to appropriate equations in Table 9-1 and to    Table 9-2). A remedy to that situation is to collect and analyze a relatively large number of composite samples, thereby offsetting the savings in analytical costs that are often associated with composite sampling, but achieving better representation of the waste than would occur with noncomposite sampling.

The appropriate number of composite samples to be collected from a solid waste is estimated by use of Equation 8 (Table 9-1), as previously described for the three basic sampling strategies. In comparison with noncomposite sampling, composite sampling may have the effect of minimizing between-sample variation (the same phenomenon that occurs when the physical size of a sample is maximized), thereby reducing somewhat the number of samples that must be collected from the waste. [underlining added]

The reference in the above quotation to s_x is to standard error. The reference to t.₂₀ is to tabulate values set out in Table 9-2 of the EPA Report.

[175]        No evidence was lead that Dillon did not collect and analyze a sufficiently large number of composite samples.

[176]        In the result, on the basis of the reference to composite sampling in the 1992 EPA Report in the context of achieving better representation of the waste, and on the basis of the failure to cross-examine Mr. Merks, I prefer the evidence of Mr. Merks that compositing samples through interleaving was an appropriate procedure. Environment Canada has failed to satisfy me on the evidence that the interleaving protocol was used because it artificially produces the lowest possible results.

[177]        To summarize my conclusions with respect to the validity of the Environment Canada data, I conclude that:

1.          The sampling plan implemented by Dillon is preferable to that implemented by Environment Canada because of Dillon's use of stratified systematic sampling and because of the greater number of samples it took.

2.          Environment Canada departed from its sampling plan with respect to the West Pile.

3.          Dillon's method of sample selection was superior in that samples were carefully homogenized with the riffle splitter and were randomly and objectively selected.

4.          It was appropriate to composite samples for analysis.

[178]        I turn now to consider the use which can be made of the individual test results, particularly those obtained by Dillon.

2. The Individual Test Results

[179]        As noted above, it was Dr. Fingas' task to scientifically determine whether the ASR at the General Scrap and XPotential sites was PCB material. In order to do so he was required, in his words, to design and implement a sampling methodology "that would allow the sampling team to select samples from each of the large piles ("cells") of ASR at General Scrap and XPotential which, when analysed, would provide results that would be representative of each entire cell" [underlining added]. Dr. Fingas' work resulted in a report that expressed the PCB content on the basis of an average for each pile, with upper and lower probability limits expressed.

[180]        On cross-examination, in addition to confirming that he took a sufficient number of samples, in his view, to be representative of the whole piles Dr. Fingas stated that:

i)           the individual hole results were important;

ii)          his first objective was to characterize the PCB content of the material in the sites;

iii)          because of the size of each hole, on the basis of the result of one hole alone they could conclude under the Act that there was PCB material in that hole;

iv)         he believes that individual sample results, in addition to stating what is in the 2 litre sample bottle, are representative of what is in the hole in the ground from which the sample was taken.

[181]        Environment Canada now argues that the individual results of all of the samples are determinative of the application. Asserting that "[t]here can be absolutely no dispute that the analytical results tell us what was in each two litre sample before PCB extraction", Environment Canada argues that the results from each 2 litre sample must represent a larger area. At a minimum the samples are said to represent the ASR in the 20 litre primary increment. Taking each primary increment to weigh 6 kg, Environment Canada then calculates the weight of all primary increments from which samples returned values of more than 50 ppm in order to establish that there is more than 100 kg of PCB material. Environment Canada through this exercise calculates there to be 396 kg of PCB material.

[182]        Environment Canada further argues, however, that the most reasonable inference is that each 20 litre primary increment represents the entire mass of the material excavated from the sample hole. Calculating what is said to be the weight of the material removed from the test pits from which samples returned values over 50 ppm, Environment Canada calculates that there is over 73,080 kg of PCB material. None of these calculations were contained in Dr. Fingas' report, nor did he testify as to those calculations.

[183]        The respondents reject this argument and state that individual sample results cannot be used by themselves to draw conclusions about the characteristics of a given volume of ASR.

[184]        In support of this submission, the respondents rely upon the testimony of Mr. Bertram and Mr. Clark.

[185]        On cross-examination Mr. Bertram testified as follows:

2816.        Q              Would you agree then that viewing the sampling results on an individual test pit basis that to the extent that individual test pits showed returns over 50 parts per million, that you can infer from that that you are identifying areas of hotspots or higher concentrations?

A              I can't view this as an individual test pit. These are samples taken from one media. They are taken from different locations. They are taken with very heterogeneous material that is representative of the whole cell, as we have been calling it. To say that one test pit is higher than another is common sense. We could have done one test pit one foot away from another test pit and got different results. There is going to be variability in such material. I can't view this as a single sample or a single entity per - - this is a collection of a whole.

[...]

BY MR. KATZ:

2817.        Q              Would you agree with me, sir, that it is possible to utilize these test results or fresh rather test results if they were taken, utilize them to zero in on areas of high concentration and then step out from those areas to determine the extent of those areas of high concentration?

A              No, you could not.

2818.        Q              You could not?

A              You could not, unless you went to a particle by particle basis you could not, and even at that point, since there is not a homogenous mixture of those individual particles that you are analyzing, you cannot determine a plume from one, from one source point. [underlining added]

[186]        It was Mr. Clark's evidence that:

i)           The individual test results did not show anything about the quantity of PCB material in total in piles or cells and that there was a danger in over-interpreting a singular laboratory result.

ii)          One cannot on the basis of the PCB extraction protocol assume that PCBs are uniformly distributed within each 2 litre sample. The PCBs could be contained in a very small portion of the 2 litre sample.

iii)          Different samples taken from the same primary samples will not provide the same analytical results. Rather, it will be a population of data with a distribution of its own. This highlights sampling error. Sampling error is a statistical term which reflects the natural variation that exists from one sample to another. It arises due to the fact that not all of the material in the primary sample is analyzed and thus there is variability in the results from one sample to another.

iv)         With respect to cell 5, Dillon caused different samples from the same primary increment to be taken and analyzed. Samples were analysed in January and November of 2000 and the results differed. This showed the inherent variability within the primary samples, which variability arises through sample development. The results of the analysis of different samples from the same primary increment were:

[187]        To the extent that the evidence of Dr. Fingas with respect to the use which may be made of individual sample results conflicts with the evidence of Mr. Clark and Mr. Bertram, I prefer the evidence of Messrs. Clark and Bertram for the following reasons.

[188]        First, I reject the premise of the applicant's argument that there can be absolutely no dispute that the analytical results tell us what is in each 2 litre sample. This is demonstrated to be untrue by the Dillon sampling results set out above where significantly different analytical results were obtained from analysis of different samples taken from the same primary sample. Dr. Fingas admitted on cross-examination that you can get different results from the same primary sample taken at different times.

[189]        Indeed, one of the references Dr. Fingas relied upon and cited in his 1999 Report was the Westat Inc. PCB Lead and Cadmium Levels in Shredder Waste Materials study, which was an EPA pilot study done in 1991. Mr. Clark testified that in such study they observed and reported variability within their primary sample.

[190]        Environment Canada did not test to see whether there was variability between one test result and another in the same location. If samples taken from the same primary sample can produce different analytical results, one result from a primary sample cannot logically be extended to apply to a larger area such as the material in a 20 litre back-hoe bucket, or the entire mass of a hole.

[191]        Second, Dr. Fingas admitted that sampling and measurement errors are inherent when one samples ASR, and that because ASR is heterogeneous and because of sampling and analytical errors different results may be obtained between samples taken 2 feet apart. This would seem to make it very difficult to draw inferences from a single result.

[192]        Finally, both parties acknowledged the expertise of the EPA and both the 1992 and 1993 EPA Reports contain relevant instruction on the point. The 1992 Report, which according to Dr. Fingas was more authoritative, states the following at page NINE-30:

During the implementation of a waste sampling plan or a statistical experiment, an effort is made to minimize the possibility of drawing incorrect inferences by obtaining samples that are representative of a population. In fact, the term "representative sample" is commonly used to denote a sample that (1) has the properties and chemical composition of the population from which it was collected, and (2) has them in the same average proportions as are found in the population.

In regard to waste sampling, the term "representative sample" can be misleading unless one is dealing with a homogeneous waste from which one sample can represent the whole population. In most cases, it would be best to consider a "representative data base" generated by the collection and analysis of more than one sample that defines the average properties or composition of the waste. A "representative data base" is a more realistic term because the evaluation of most wastes requires numerous samples to determine the average properties or concentrations of parameters in a waste. (The additional samples needed to generate a representative data base can also be used to determine the variability of these properties or concentrations throughout the waste population.) [underlining added]

[193]        The parties agree that ASR is not homogenous and the EPA 1992 Report is express that one sample can only represent the whole population if the waste is homogenous. In other cases, the EPA concludes that it is best to consider a representative data base. This underscores the danger in relying upon one sample to represent the contents of one sample hole or even one back-hoe bucket.

3. The Use of Statistical Analysis

[194]        The nub of the dispute between the parties is characterized by Environment Canada as being whether the average of an entire cell or pile or of smaller individual areas within the pile or cell (known as hot spots) should be used to determine whether there is 100 kg of ASR with a PCB concentration of 50 ppm or more.

[195]        Environment Canada takes the position that statistical inferences (the applicability of the mean, standard deviation and confidence intervals for entire cells and piles) are irrelevant to the issue of compliance with the Regulations.

[196]        The respondents, however, take the position that the relevant question is whether there is PCB material. This is stated to be essentially "an exercise in hypothesis testing". In order to determine whether particular ASR is PCB material the respondents' experts developed a sampling and analytical program and determined that the appropriate approach in order to characterize the ASR was by identifying each entire pile and cell at issue as a sampling unit. This approach is said to be consistent with the 1992 EPA Report.

[197]        Environment Canada relies upon the evidence of Dr. Fingas. He testified on cross-examination that:

i)           If one looks at the literature, the distribution of the PCBs in the ASR is not in a normal or logged normal or other known distribution. It is a very random distribution. Therefore, one should not by any form of science use statistics on that type of data.

ii)          It is well-known scientifically that it is inappropriate to apply statistical methods to non normal data.

iii)          Notwithstanding the reference to statistical applications in the 1992 and 1993 EPA Reports in order to analyze sample data, those statistical applications could not be made to ASR.

iv)         Dr. Fingas also took issue with the following statement from the 1993 EPA Report (the field manual to provide sampling guidance for scrap metal shredders):

Because of the sampling error and laboratory error, it is not possible to determine exactly the concentration of toxic substances. However, by using the methods in this section, you will be able to make statements such as, "As a result of our study, we are 95% certain that the concentration of PCBs in this pile of stored fluff is between 40 and 100 ppm".

The basis for Dr. Fingas' disagreement with this statement was his view that statistics could not be applied to material that is not normally distributed.

[198]        In addition to relying upon the evidence of Dr. Fingas, Environment Canada argues that the language of the Regulations does not support the use of averages. In material part, the Regulations provide:

Environment Canada argues that the respondents' reliance upon an "average" as the basis for determining whether the PCB Regulations apply, contradicts the word "aggregate" used in the Regulations. This is said to be so because the word "aggregate" in the context of the PCB Regulations is effectively the opposite of the word "average". To aggregate is to collect only similar units together, where as to average is to collect all dissimilar units together.

[199]        The respondents rely upon the affidavit and expert report of Mr. Merks whose report was based upon statistical analysis of the test results. The respondents argue that statistical methods and applications are crucial to proper interpretation of the sampling data.

[200]        Dealing first with the applicant's argument flowing from the use of the word "aggregate" in the Regulations, I disagree, respectfully, with the submission that the use of the word "aggregate" is inconsistent with the use of statistical averages. I accept the submission of the respondents that what is first required is that the suspect material be categorized. If some or all is categorized to be PCB material, then one must consider if the amount of PCB material when added up equals 100 kg or more.

[201]        As to the exercise of characterizing suspect material, the Act is express in paragraph 2(1)(i) that in the administration of the Act the Government of Canada is to apply science and technology to identify and resolve environmental problems. Dr. Fingas was express in his affidavit that his task was to implement a program so "the results would scientifically determine" whether the ASR in question was PCB material. The proper characterization of ASR is therefore a question to be answered by science.

[202]        Before one considers the conflicting evidence of Dr. Fingas and Mr. Merks it is necessary to deal with the argument that the opinion of Mr. Merks is irrelevant because his evidence relates to the calculation of the average concentration of PCBs in each cell or pile. I am unable to conclude that when dealing with a heterogeneous substance such as ASR the mean or average PCB concentration in each pile is irrelevant for the following reasons.

[203]        First, Dr. Fingas swore that he was required to design and implement a methodology that would provide results that would be "representative of each entire cell". This is inconsistent with the position that Environment Canada now takes that what is relevant is smaller individual areas within each pile or cell.

[204]        Second, in the 1997 and 1999 Environment Canada Reports, Dr. Fingas' results were expressed as an average for each entire cell or pile. Thus, for the 1997 and 1999 reports, the results were:

     Average                                                                    Average

(1997 Report)                                                             (1999 Report)

__________________________________________________________________

General Scrap

East pile 33.4 ppm                                                    59.7 ppm

Central pile             35.9 ppm                                                    41.2 ppm

West pile                 85.4 ppm                                                    54.6 ppm

XPotential

Cell 5                        61.3 ppm                                                    50.7 ppm

Cell 6                        96.7 ppm                                                    42.0 ppm

On the basis of the 1999 Environment Canada Report, Dr. Fingas swore that "at the minimum, the east and west ASR cells at General Scrap, and both storage cells 4 and 5 and the drying cell at XPotential have average PCB concentrations in excess of 50 ppm, and are therefore properly considered PCB Material". [underlining added]

[205]        In addition to Dr. Fingas' evidence, the affidavit of Shannon Kurbis, the Environment Canada enforcement officer, speaks to the use of average concentrations. At paragraphs 74 and 76 she states:

74.            The Emergencies Science 1999 Report indicates that two ASR piles at General Scrap and three of the four storage areas at XPotential had average PCB concentrations that exceeded 50 ppm.

[...]

76.            The average results taken from the Emergencies Science 1999 Report reflect the minimum PCB concentration. The letter accompanying the Emergencies Science 1999 Report states that General Scrap should take immediate steps to ensure that all PCB Material is stored in compliance with the PCB Regulations.

[206]        This use by Environment Canada's own expert and enforcement officer of the average concentration of each entire cell or pile in order to characterize whether the cell or pile is PCB material belies this argument now advanced by Environment Canada that the average concentration is irrelevant.

[207]        Finally, as Ms. Kurbis' affidavit shows, Environment Canada determined whether or not to bring enforcement proceedings on the basis of the average pile and cell concentrations set out above. Thus in February of 1998, Environment Canada advised the respondents that the General Scrap West Pile, and cells 5 and 6 at the XPotential site contained quantities of PCB in excess of the regulatory limit. On the basis of the 1999 Environment Canada Report the applicant asserts that both the East and West Pile at General Scrap and the drying cell and cells 4 and 5 at the XPotential site contain PCB material.

[208]        In sum, Environment Canada (through Dr. Fingas) designed a sampling plan to provide results representative of the whole of each pile or cell and calculated the average concentration per pile and cell, then used the average concentrations as a basis for enforcement proceedings, and adduced evidence in this proceeding in terms of these average concentrations. I therefore do not now accept Environment Canada's submission that the average concentration per cell or pile is irrelevant.

[209]        Having determined that average concentrations are relevant, what then of the conflicting opinion of Dr. Fingas and Mr. Merks as to the appropriateness of applying a statistical analysis?

[210]        Mr. Merks was not cross-examined on his opinion, Dr. Fingas was. The following extract from Dr. Fingas' cross-examination is, in my view, telling:

393.          Q              And you keep coming back to that and we've had a lot of questions around that. So you're saying the application of statistics just doesn't apply with this type of material.

A              That's correct.

394.          Q              And what's your authoritative source for that, other than yourself? Do you have a source that you can point me to that says statistical applications don't apply with ASR?

A              Yes. Every textbook says one should not use normal statistics or statistics of that type on heterogenous non normal distributions.

395.          Q              ASR, do you have - - in any publications, because I've only got one and the one I've got says you should do it. So you show me the publication that says you can't do statistical application with ASR to determine whether it meets a regulatory threshold, because that's the issue we have before the Court today. Where's your authority? I'd like to see it.

A              Well I have the authority in the sense that it's authority well known scientific principal one does not do normal statistics on non normal samples. This is you know, the first page of statistic textbooks.

[211]        Unfortunately no specific texts or articles were identified by Dr. Fingas nor did he demonstrate that these were non normal samples. The reference by counsel in the extract quoted above to the publication dealing with ASR was to the 1992 EPA Report which specifically dealt with ASR and which stated that because of sampling and analytical errors "we must use statistical analysis" to obtain confidence intervals and levels of confidence.

[212]        As noted, the 1992 and 1993 EPA Reports deal extensively with the use of statistics. Environment Canada, however, submits that there is no evidence before the Court that the EPA Reports are appropriately considered "best practices" for ASR sampling. Therefore, Environment Canada says that they ought not to be used to support or challenge the appropriateness of any sampling plan.

[213]        My difficulty with this submission stems from the fact that the 1999 Environment Canada Report specifically lists both the 1992 and 1993 reports as references. Dr. Fingas agreed in cross-examination that the EPA is a recognized, reputable environmental regulatory authority, and stated that he specifically referenced reports in his 1999 Environment Canada Report that he used very heavily in his preparation. Dr. Fingas sent his sampling plan to the EPA for review where a statistician looked at it. This was presumably done because Dr. Fingas had no prior experience with the sampling and characterization of ASR.

[214]        Further, by letter dated September 11,1997 Environment Canada responded to questions put forward on the respondents' behalf and stated that "[i]n response to your questions regarding sampling and analysis methodology, Environment Canada will be following standard sampling procedures as outlined in Test Methods for Evaluating Solid Waste, Chapter 9, Sampling Plan, Environmental Protection Agency, 1992. Analysis of all samples will be carried out according to the test methods specified in the Storage of PCB Material Regulations," as more specifically defined in the letter. The document referenced as being "Test Methods for Evaluating Solid Waste, Chapter 9" is the document which is described in these reasons as the 1992 EPA Report.

[215]        In light of this use of the 1992 EPA Report and Dr. Fingas' reliance, in part, upon the EPA I am persuaded that it is relevant to consider what the 1992 EPA Report says about the use of statistics.

[216]        The following extracts from that report are relevant:

This section of the manual addresses the development and implementation of a scientifically credible sampling plan for a solid waste and the documentation of the chain of custody for such a plan. The information presented in this section is relevant to the sampling of any solid waste, which has been defined by the EPA in its regulations for the identification and listing of hazardous wastes to include solid, semisolid, liquid, and contained gaseous materials. However, the physical and chemical diversity of those materials, as well as the dissimilarity of storage facilities (lagoons, open piles, tanks, drums, etc.) and sampling equipment associated with them, preclude a detailed consideration of any specific sampling plan. Consequently, because the burden of responsibility for developing a technically sound sampling plan rests with the waste producer, it is advisable that he/she seek competent advice before designing a plan. This is particularly true in the early developmental stages of a sampling plan, at which time at least a basic understanding of applied statistics is required. Applied statistics is the science of employing techniques that allow the uncertainty of inductive inferences (general conclusions based on partial knowledge) to be evaluated.

9.1.1    Development of Appropriate Sampling Plans

An appropriate sampling plan for a solid waste must be responsive to both regulatory and scientific objectives. Once those objectives have been clearly identified, a suitable sampling strategy, predicated upon fundamental statistical concepts, can be developed. The statistical terminology associated with those concepts is reviewed in Table 9-1; Student's "t" values for use in the statistics of Table 9-1 appear in Table 9-2.

[...]

9.2.2.1    Statistics

A discussion of waste sampling often leads to a discussion of statistics. The goals of waste sampling and statistics are identical, i.e., to make inferences about a parent population based upon the information contained in a sample.

Thus it is not surprising that waste sampling relies heavily upon the highly developed science of statistics and that a sampling/analytical effort usually contains the same elements as does a statistical experiment. [underlining added]

[217]        The evidence of Dr. Fingas and Mr. Merks as to the appropriateness of statistical analysis are opposed. Mr. Merks was not cross-examined on his opinion, which opinion (for the reasons set out above) I have found to be relevant. The 1992 EPA Report which Dr. Fingas states he relied upon contemplates the use of statistics for the purpose of making inferences about a parent population based upon the information generated from a sample. Shannon Kurbis advised the respondents that Environment Canada would be following the standard sampling procedures outlined in the 1992 EPA Report. Dr. Fingas did not point to any specific article or book that would provide support for his view that statistics could not be applied to ASR, nor did he give evidence that he did any calculation to determine whether there was in fact a normal distribution.

[218]        For these reasons, I find the opinion of Dr. Fingas to be less persuasive on this point than the opinion of Mr. Merks. I therefore accept the opinion of Mr. Merks that a statistical analysis is relevant, and for the reasons given by the respondents' witnesses and in the 1992 EPA Report I also accept that a statistical analysis is required in order to interpret properly the data.

[219]        Having concluded on the evidence adduced in this application that a statistical analysis is required in order to interpret properly the data, it follows that I accept the opinion of Mr. Merks that the mean PCB concentration in each of the sampling units is lower that the regulatory threshold and that the ASR does not constitute PCB material. In accepting Mr. Merks' opinion I have regard to the fact that Environment Canada did not challenge any of the primary data he relied upon, and did not challenge the propriety of his actual statistical analysis. Rather, Environment Canada challenged the appropriateness of the use at all of statistical applications.

[220]        In so concluding, I have considered the submission of counsel for the applicant that the statistical analysis is an exercise performed by "mathemagicians" and its "poof and all the high numbers are gone". While an evocative image, logically all of the low numbers vanish as well. I accept on the evidence that recourse to the mean concentrations is appropriate in view of the error that is inherent in individual sample results.

[221]        In this regard, I think that it is of some assistance to compare the results Dr. Fingas obtained in his 1997 and 1999 reports as set out above at paragraph 204. The sampling plan and methodology did not differ significantly between the two reports. While the analytical chemical protocols differed, (in that 1/RM/3 was used in 1999 and 1/RM/31 was used in the 1997 report) Dr. Fingas testified that some studies showed that there was no difference in qualification between using the two different analytical protocols. The results therefore may properly be compared.

[222]        Comparing the results, in 1997 Environment Canada determined the areas containing PCB material to be the West Pile at General Scrap and cells 5 and 6 at XPotential. In 1999, the East Pile at General Scrap was said to contain PCB materials, but cell 6 at XPotential did not. In 1997 Environment Canada reported that the East Pile had an average PCB concentration of 33.4 ppm compared with 59.7 ppm in 1999. The West Pile was reported to have a PCB concentration of 85.4 ppm in 1997 and 54.6 ppm in 1999. No new ASR had been added to either pile after around 1994. An Environment Canada publication in evidence states that PCBs do not decompose or biodegrade significantly in the natural environment.

[223]        On the basis of logic and common sense, it would seem that a scientifically valid and reliable methodology ought not to produce such disparate results for the East and West Piles.

[224]        I have also considered the submissions of the applicant that the EPA Final Rule (1998) ("Final Rule") provides support for the Environment Canada approach.

[225]        The Final Rule is a statutory instrument, an extract of which was put in evidence, by consent, as a result of questions addressed to Mr. Clark by the Court. No other witness or affidavit referred to the Final Rule.

[226]        Mr. Clark advised that under sub-part R the Final Rule prescribes procedures for developing representative samples. The 1993 EPA Report is specifically referenced in the Final Rule as one of the methods to be used and followed for determining the PCB concentration of samples. Mr. Clark relied upon the Final Rule with respect to sample size, sub-sampling and the concept of compositing samples.

[227]        Environment Canada argues that there is no reference in the Final Rule to any statistical analysis and so it is persuasive evidence of how analytical results ought not to be applied to PCB contaminated ASR. Environment Canada also asserts that it is informative that the Final Rule makes no mention of the 1993 EPA Report and only mentions the 1992 EPA Report in addressing laboratory protocols. Environment Canada also relies upon portions of sub-part N of the Final Rule.

[228]        Given the limited use of the Final Rule by Mr. Clark, and the fact that no other witness referred to the document I am not prepared to give any significant weight to the document.

[229]        No evidence was adduced as to the applicability of sub-part N to the situation at issue and it was not suggested to Mr. Clark that sub-part N was applicable. No evidence was adduced as to the interrelationship, if any, between the 1992 and 1993 EPA Reports and what is apparently a statutory instrument.

[230]        Further, if in 1998 the EPA departed in a significant manner from views or positions expressed in the 1992 and 1993 EPA Reports it is reasonable to infer that Dr. Fingas would have referenced this either in his report of January 1999, or in his affidavit sworn in 2000 in this proceeding, or in his oral evidence. He did not refer at all to the Final Rule, but specifically referenced in his report and his testimony the 1992 and 1993 EPA Reports.

[231]        One final argument advanced by Environment Canada must be considered. That is its submission that if the appropriate method for determining compliance is the statistical calculation of the average concentration and the confidence intervals for the cell or pile as a whole, the respondents' data establishes that the drying cell and cell 5 at XPotential and the West Pile at General Scrap are PCB materials. This is said to be so because these cells and this pile returned averages which, combined with the statistical confidence intervals, have the upper limit of the confidence interval over 50 ppm. Environment Canada relies upon the following extract from the 1992 EPA Report:

If the upper limit is less than the threshold, the chemical contaminant is not considered to be present in the waste at a hazardous levels; otherwise, the opposite conclusion is drawn.

[232]        However, close examination of the relevant portion of the 1992 EPA Report shows that in that report the applicable equation for determining the confidence interval determined an 80% confidence interval. The numbers relied upon by Environment Canada in the respondents' data were determined at an 95% confidence level. If the respondents' data is re-calculated at an 80% confidence level (which is the level contemplated by the EPA Report, and which calculation was done in Court) the upper limit of the confidence interval is less than the regulatory threshold. Therefore, the data does not support the contention that the two cells and pile contain or constitute PCB material.

III.        IF SOME OR ALL OF THE ASR IS PCB MATERIAL, WHICH RESPONDENTS ARE IN BREACH OF THE REGULATIONS?

[233]        My finding that Environment Canada has failed to establish that some of the ASR is PCB material makes it unnecessary for me to deal with this. However, even if some ASR at the General Scrap site was PCB material, I see no basis in the evidence for any liability on the part of either individual respondent or on the part of Jamel in respect of the General Scrap site.

[234]        Mr. Lazareck's evidence was clear and unchallenged that IPSCO now owns 100% of General Scrap, that Jamel provides employee services to General Scrap, but Jacob and Melvin Lazareck are no longer involved in the day-to-day operations of General Scrap. The only evidence of the nature of the employment services provided by Jamel was that the payment of the General Scrap payroll is handled through Jamel.

[235]        The obligation imposed by the Regulations with respect to PCB material is placed upon a person who owns, controls or possesses PCB material.

[236]        On the evidence none of Jamel, Melvin Lazareck or Jacob Lazareck own, control or possess PCB material at the General Scrap site.

IV.        CONCLUSION, ORDER AND COSTS

[237]        In view of my findings on the evidence, I see no reason to consider whether in the ultimate exercise of the Court's discretion an injunction should issue.

[238]        For the reasons given, the application for injunctive relief is dismissed. It follows, and the parties agree, that the interim consent order should be set aside.

[239]        As requested by counsel the issue of costs is reserved. Before January 31, 2004 counsel for the respondents should contact the Registry to request a teleconference for the purpose of discussing how the outstanding issue of costs is to be dealt with.

ORDER

[240]        IT IS HEREBY ORDERED THAT:

1.          The application for injunctive relief is dismissed.

2.          The interim consent order issued in this proceeding by Madam Justice Heneghan is set aside.

3.          The issue of costs is reserved for further submissions.

"Eleanor R. Dawson"

                                                                                                                                                    Judge                        

                                                            FEDERAL COURT

                        NAMES OF COUNSEL AND SOLICITORS OF RECORD

DOCKET:                                T-2274-00

STYLE OF CAUSE: The Queen v. IPSCO Recycling Inc. et al.

PLACE OF HEARING:         Winnipeg, Manitoba

DATE OF HEARING:           April 3, 4, 7, 8 and 9, 2003

June 23 and 24, 2003

REASONS FOR ORDER:

AND ORDER:                         Hon. Madam Justice Dawson

DATED:                                   December 23, 2003

APPEARANCES:

Duncan A. Fraser                       FOR THE APPLICANT

Joel I. Katz

James G. Edmond                      FOR THE RESPONDENTS

John D. Stefaniuk

SOLICITORS OF RECORD:

Morris Rosenberg

Deputy Attorney General                                     FOR THE APPLICANT

of Canada

Thompson Dorfman Sweatman              FOR THE RESPONDENTS

Winnipeg, Manitoba