Paramount Resources Ltd v Chubb Insurance Company of Canada, 2023 ABKB 627

Court of King’s Bench of Alberta Citation: Paramount Resources Ltd v Chubb Insurance Company of Canada, 2023 ABKB 627 Date: 20231108 Docket: 1801 08113 Registry : Calgary Between: Paramount Resources Ltd Plaintiff - and - Chubb Insurance Company of Canada, Lloyd's Underwriters, and Royal and Sun Alliance Insurance Company of Canada Defendants _______________________________________________________ Reasons for Judgment of the Honourable Justice J.T. Eamon _______________________________________________________

Table of Contents I Introduction . 3 II Background of the loss . 5 III The insurance policies . 7 IV The CO&O and Conoco’s role as facility operator 8 V Regulatory regime . 13 VI The operation of the LVP pipeline and evidence of the release . 14 (

a) Date that the release commenced . 14 (

b) Date that Conoco concluded there was a release . 16 VII Whether the loss was detected within 720 hours . 19 (

a) Policy

interpretation principles . 19 (

b) Detection of the release . 21 (

i) Surrounding circumstances – insurance industry . 21 (ii) Surrounding circumstances – oil and gas industry . 24 (iii) Case law concerning similar provisions . 25 (iv) The meaning of detection . 27 (

c) The insurer’s argument that Paramount is bound by previous positions, representations and findings . 31 (

i) Paramount’s position in the arbitration with Conoco . 31 (ii) Insurers’ position that Paramount is bound by Conoco’s admissions . 33 (iii) AER findings . 38 (

d) Conclusion . 39 VIII The policy requirements that the insured is obligated to pay damages . 39 (

a) Introduction . 39 (

b) Background facts . 40 (

c) What Paramount must prove . 42 (

d) Whether settlement reasonable . 45 IX Whether Paramount failed to mitigate the loss . 50 X Other matters . 51 XI Decision . 52 I Introduction [ 1 ] The plaintiff Paramount, an oil and gas company, claims against the defendants, who are Paramount’s pollution liability or excess insurers, for coverage under the insurance policies arising from a release of pollutants from a LVP (low vapour pressure) pipeline. [ 2 ] The pipeline was part of the Resthaven facility near Grand Cache, Alberta, in which Paramount held a ½ interest.

The facility included the Conoco Resthaven gas plant (or the 01-36 plant), the LVP pipeline, which ran from the 01-36 plant to another gas plant (known as the 08-11 plant), and sales tanks at the 08-11 plant where the pipeline output was held for sale. [ 3 ] The other owner of the Resthaven facility, also holding a ½ interest, was ConocoPhillips (BRC) Partnership (“Conoco”).

Paramount and Conoco agreed to terms of ownership and operation of the Resthaven facility under the Construction, Ownership and Operating agreement for the Resthaven facility dated October 1, 2005 (“CO&O”). [ 4 ] Conoco operated the Resthaven facility pursuant to the CO&O. [ 5 ] The LVP pipeline was used to transport low vapour pressure condensate, a product derived from natural gas production in the Resthaven gas field that was received and partially treated at the Resthaven gas plant, to the sales tanks at the 08-11 plant. [ 6 ] The parties to this action agree that there was an unexpected and unintentional discharge, release or escape of pollutants into or upon land from the LVP pipeline (the "release") that commenced on or about April 21, 2016.

The pipeline was leaking condensate into the environment. [ 7 ] The volumes of condensate entering and exiting the LVP pipeline were metered by highly accurate meters (known as Coriolis meters), which had been recently installed and became operational on April 21, 2016. Conoco became aware of volume data anomalies showing significantly less condensate was exiting the LVP pipeline than entering it. The pipeline is buried in the ground within the pipeline right of way for most of its run, so it could not be visually inspected for leaks.

Conoco conducted a pressure test on the pipeline on May 6-7, 2016 to see if the LVP pipeline was leaking. The pipeline failed the test. [ 8 ] The volume and pressure data anomalies, separately or together, would have led a reasonable operator to shut in the pipeline unless and until the anomalies could be clearly and readily attributed to some reason other than a leak in the pipeline. Conoco believed other explanations might account for the anomalies and continued to operate the pipeline. Conoco visually monitored the pipeline right of way for the appearance of hydrocarbons on its surface.

Conoco continued in this fashion while discussions and tests continued in an attempt to explain the ongoing data anomalies and troubleshoot the Coriolis meters. [ 9 ] At 3:40 pm on June 9, 2016, Conoco personnel inspecting the pipeline right of way, saw a small patch of hydrocarbon staining thereon in a location where the pipeline right of way followed along a ridge or hilltop. Soon after, a Conoco supervisor who was called to the scene found extensive hydrocarbon contamination at the base of the slope and a nearby wetland off of the right of way.

Conoco notified the Alberta Energy regulator (“AER”) of the spill on June 9, 2016 and Paramount shortly thereafter. The parties agree that Conoco personnel observed the released substances on June 9, 2016. [ 10 ] The LVP pipeline had released hundreds of thousands of litres of condensate into the environment. AER ordered Conoco to clean it up. Conoco called on Paramount under the CO&O to pay ½ the cost of the clean up expenses. Paramount refused. [ 11 ] Conoco commenced arbitration proceedings under the CO&O to recover its claim against Paramount in May, 2017. Paramount defended.

As time passed, Paramount’s exposure to its alleged share of the clean up costs, with interest and Conoco’s legal costs of the arbitration, exceeded $30,000,000. A few months before the scheduled hearing date, Paramount and Conoco settled the claim for a lesser amount. [ 12 ] The defendants acknowledged that the commencement of the leak of the condensate from the LVP pipeline was an accidental release of contaminants, but denied coverage in December 2017 because the release was not “detected by any person” within 720 hours of the commencement of the release.

This detection was a required element of coverage under their respective insurance policies. Additionally, Chubb as the primary insurer refused to defend Paramount in the arbitration. [ 13 ] The insurers later refused to indemnify Paramount for the arbitration settlement.

Paramount seeks judgment against the insurers for the amount of the settlement up to the available coverage under the insurance policies. [ 14 ] The claim primarily turns on the meaning of “detected by any person” in the policies. [ 15 ] Paramount argues that Conoco detected the release within 720 hours of the commencement of the release on April 21, 2016, therefore Paramount is entitled to coverage. [ 16 ] The insurers respond that the release was not detected until Conoco personnel observed the hydrocarbon contamination and subjectively concluded the pipeline was leaking on June 9, 2016.

The insurers submit that before that date, Conoco personnel were “blind” to the warning signs that the pipeline was leaking and therefore had not detected the release. [ 17 ] The insurers submit the following additional arguments that they are not liable under the policies for the loss: (

a) They allege Paramount admitted in the arbitration with Conoco that Conoco had not detected the release until June 9, 2016 – long after the detection period expired. Paramount should be bound by that admission in this action. This argument primarily turns on the nature of the alleged admission and whether to give weight to it. (

b) They allege Conoco admitted on previous occasions in a regulatory investigation conducted by AER, that Conoco had not detected or discovered the leak until June 9, 2016. The insurers allege AER made a similar finding in the regulatory proceeding. They say these admissions and findings bind Paramount in this action, such that it cannot assert that Conoco detected the release at any time earlier than June 9, 2016. These arguments primarily turn on the insurers’ argument that Conoco acted as Paramount’s agent in the regulatory investigation and therefore Paramount is bound by representations by, and findings against, its agent. (

c) Paramount must prove that it would have been found liable for Conoco’s claim in the arbitration, not merely that it settled the claim. Paramount has not proved this required element. Instead, it is taking positions on the key issues in the present insurance action that are contrary to the defences it asserted in the Conoco arbitration. (

d) In any event, Paramount’s settlement of the claim was unreasonable. (

e) If Paramount is correct that Conoco detected the leak before June 9, 2016, then Conoco’s negligent delay in failing to

mitigate the loss should be attributed to Paramount, placing Paramount in breach of its obligations as an insured to mitigate the loss after it occurs. This argument also turns primarily on the agency argument that Conoco’s acts as facility operator should be attributed to Paramount as a facility owner. (

f) A number of other related submissions dealt with under the heading Other Matters in these reasons. [ 18 ] The parties agreed the claim was suitable for a

summary trial. [ 19 ] I am satisfied the issues are suitable for resolution in a

summary trial. [ 20 ] For the reasons set out herein, I allow Paramount’s claim. [ 21 ] These reasons comment adversely on Conoco’s operation of the LVP pipeline. Conoco is not a party to this action, has not had an opportunity to respond, and my findings pertain only to the dispute between Paramount and its insurers. II Background of the loss [ 22 ] The Resthaven production field and the Resthaven gas plant are located in a remote area of Alberta. The right of way in which the LVP pipeline runs from the 01-36 plant to the 08-11 plant is about 10 or 11 km in length.

The terrain is hilly or mountainous and forested. The leak in the pipeline occurred at a point where it runs along a ridge or hill top. An unnamed creek runs along the base of the slope in this area.

This creek is a tributary of Webb Creek, which is itself a tributary of the Simonette River. [ 23 ] The parties agree that in 2016 there was an unexpected and unintentional release of condensate from the LVP pipeline, that commenced “on or about” April 21, 2016. [ 24 ] From April 21, 2016 through April 30, 2016, the Coriolis meters on the LVP pipeline generated consistent data indicating a significant shortfall from the amounts of condensate pumped into the pipeline at the 01-36 plant compared to the amounts received at the outlet at the 08-11 plant about 11 km distant.

Conoco personnel became aware of these data anomalies when they performed a meter reconciliation of data up to and including April 30 th . Owing to concerns the pipeline could be leaking, on May 6, 2016 Conoco conducted a pressure test where the pipeline was pressured to 950 kPa and left overnight. On May 7, 2016 Conoco personnel observed that the pipeline pressure was 250 kPa, indicating a significant loss of pressure during the night. [ 25 ] Conoco’s personnel did not recognize the data anomalies as proof of a leaking pipeline.

The Coriolis meters had been recently installed (April 17 th ) and proved on April 20, 2016. They thought there must be a problem with this new equipment. They suggested other explanations for the anomalies and continued to use the LVP pipeline to ship the hazardous condensate. The field staff were nervous that the pipeline could be leaking. [ 26 ] Although Conoco did not shut in the pipeline, it stepped up the frequency of right of way inspections to monitor for leaks, consistent with Conoco’s historical method of leak monitoring for this pipeline.

Conoco believed that if the pipeline actually was leaking, hydrocarbons would appear on the right of way. There is no evidence in this

summary trial (to which Conoco is not a participant) that Conoco had or did not have information about the sub-surface soil or other conditions that may inform one of where the hydrocarbons likely would seep or appear once released from the confines of the pipeline. [ 27 ] The Coriolis meters continued to record anomalies between the inlet and outlet volumes of condensate in the LVP pipeline. [ 28 ] During one of these right of way inspections, on June 9, 2016 at around 4 pm, Conoco employees observed a hydrocarbon stain on a small patch of the LVP pipeline right of way.

Shortly after, a senior employee walked off the right of way and saw hydrocarbons at the base of the slope and adjacent waterway. Conoco commenced an emergency response plan and reported the matter to AER. [ 29 ] Conoco did not tell AER or Paramount about the data anomalies or the issue whether the LVP pipeline was leaking before June 9, 2016. [ 30 ] Post-leak investigations showed that the leak occurred in a portion of the pipeline running along the top of a ridge.

Some of the escaping hydrocarbons migrated east from the point of the leak through the subsurface downhill about 90m to the unnamed creek at the base of the slope and eventually into Webb Creek. An entire low-lying area of Webb Creek dammed off by beavers was saturated with condensate, with heavy staining and pooling in the water and on the bed and shore. There was some evidence that small amounts of the condensate eventually entered the Simonette River. Software analysis indicated the affected area covered 2.8 million square metres.

The release caused significant loss and damage to vegetation, wildlife, waterways, and public land. [ 31 ] Conoco later determined, using various business records, that the released volume of condensate was approximately 379,400 litres. The cost to remediate the environment was in the tens of millions of dollars. [ 32 ] Analysis of records and the failed portion of the pipeline indicated the failure was caused by gradual bacterial corrosion, which had slowly occurred over a period of many years.

It was theorized that the bacterial contamination was introduced when the pipeline was pressured tested using untreated water in 2006 or 2007 and exacerbated by periods of inactivity in using the pipeline. III The insurance policies [ 33 ] The defendants each issued an insurance policy to the Plaintiff.

[ 34 ] Chubb's policy was the primary policy. The Lloyd’s policy was excess to the Chubb policy, and RSA’s policy was excess to the Chubb policy and Lloyd’s policy. [ 35 ] Pursuant to the Chubb policy, Paramount was covered for those sums that the insured becomes legally obligated to pay as compensatory damages because of "property damage" caused by a "pollution incident" (and subject to the other terms, conditions and exclusions set out in the policies).

The conditions included that the insured’s responsibility to pay damages must be determined in a “suit” (including an arbitration proceeding to which the insured must submit or submits with the insurer’s consent) on the merits in the “coverage territory” (Canada) or in a settlement that the insurer agrees to. By endorsement, coverage was extended to clean up costs on property owned, rented or occupied by, loaned to or used by, or in the care, custody or control of the insured. [ 36 ] The Chubb policy further undertook to defend any suit seeking those compensatory damages.

Defence costs were included in the policy limits. Chubb could investigate and settle any claim or suit in its discretion.

Chubb’s obligation to defend ended when it used up the applicable limit of insurance. [ 37 ] The Chubb policy definition of a “pollution incident” included a requirement that the "pollution incident" be detected within 720 hours after the commencement of the emission, discharge, release or escape of pollutants, as follows (the “Endorsement"): " Pollution Incident " means an unexpected and unintentional emission, discharge, release or escape of pollutants into or upon land, the atmosphere, or any watercourse or body of water, provided: (1) that such emission, discharge, release or escape results in " environmental damage "; (2) that such emission, discharge, release or escape is detected by any person within 720 hours after commencement of such emission, discharge, release or escape; (3) that the insured mails or delivers to us notice, in writing, or such emission, discharge, release or escape not later than 2160 hours following the discovery of such emission, discharge, release or escape as described in paragraph (2) above.

However, if the insured is a non-operator, such notice must be delivered to us not later than 2160 hours following notification to the Insured by the operator of such emission, discharge, release or escape; and (4) that such emission, discharge, release or escape does not occur in a quantity or with a quality that is routine or usual to the lnsured's operation.

The entirety of any such emission, discharge, release, or escape shall be deemed to be one " pollution incident ". [...] Pollutants mean any solid, liquid, gaseous or thermal irritant or contaminant, including smoke, vapour, soot, fumes, acids, alkalis, chemicals and waste.

Waste includes materials to be recycled, reconditioned or reclaimed. [...] As a condition precedent to our obligation under this policy, there must be strict conformance with the requirements specified above, regardless of whether we are prejudiced by the failure of these requirements to be met. (Emphasis in original). [ 38 ] With respect to the definition of “pollution incident” the phrase “environmental damage” is defined by the policies as: “Environmental damage” means the injurious presence in or upon land, the atmosphere, or any watercourse or body of water of solid, liquid, gaseous, or thermal contaminants, irritants, or pollutants. [ 39 ] The insurers’ amended statement of defence and the parties’ Agreed Statement of Facts provides: The Lloyd’s Policy and the RSA Policy were at all times subject to the same terms,

definitions, exclusions and conditions (except as to the amount of the premium, the amount of coverage, and the limits of liability) contained in the Chubb Policy. [ 40 ] There are wording differences among the Chubb, Lloyd’s and RSA policies, some of which might be significant. For example, the RSA policy contains two pollution liability exclusions that might or might not be important.

The RSA policy also does not incorporate any obligation to defend from the Lloyd’s policy, and contains a specific definition of loss slightly different than the Chubb policy. [ 41 ] It is important to understand, that the defendants are sophisticated and competently represented, and there may be agreements or other circumstances among the defendants of which the Court is not aware that have led them to make the foregoing admissions in their pleadings and the Agreed Statement of Facts. I have proceeded to address the issues raised by the parties in this

summary trial on the specific basis and finding in accordance with the Agreed Statement of Facts, that the terms of the Chubb policy (with the limited exceptions pled in the amended statement of defence and Agreed Statement of Facts) reflect the terms of the Lloyd’s policy and the RSA policy. IV The CO&O and Conoco’s role as facility operator [ 42 ] Conoco was the designated operator under the CO&O and was responsible for managing and operating the facility. The costs to remediate and repair were for the Joint Account. The contract obliged Paramount to bear ½ of these expenses. These expenses were

not allocated under the Joint Account to the working interest owners in the event they were a direct result of or directly attributable to the gross negligence of the operator, or its affiliates, directors, officers, consultants, agents, contractors or employees. Gross negligence was a defined term under the CO&O.

I set out greater detail of the CO&O’s provisions supporting these conclusions, and pertaining to the parties’ relationship, in the following paragraphs. [ 43 ] The CO&O incorporates the template 1999 Petroleum Joint Venture Association Model Construction, Ownership and Operating Agreement including Operating Procedure, as modified by certain express elections and modifications made by Paramount and Conoco to the Operating Procedure. [ 44 ] Pursuant to clause 601 of the CO&O, the Joint Owners designated Conoco as the Operator and Conoco accepted such designation.

Under clause 603, the interests of the owners in the facility or its separate components are held in trust by the operator for the owners subject to the provisions of the CO&O. [ 45 ] Pursuant to clause 201 of the Operating Procedure, the Joint Owners would form an Operating Committee composed of their duly appointed representatives.

Under the CO&O modifications, the parties had equal votes on the committee and disputes would be resolved under the mediation and arbitration provisions of the CO&O. [ 46 ] Pursuant to clause 203 of the Operating Procedure, the Operating Committee shall, in accordance with the terms of the CO&O, exercise overall supervision and control of and shall determine all matters of importance relating to Joint Operations (as defined in the Operating Procedure), except for those matters: (

i) designated in the Operating Procedure to be within the exclusive jurisdiction and control of the Operator; or (ii) excluded in the Operating Procedure from the jurisdiction and control of the Operating Committee. [ 47 ] Pursuant to clause 401 of the Operating Procedure: 401. Control and Management of Joint Operations Operator shall consult with the Operating Committee from time to time with respect to decisions to be made for the conduct of Joint Operations, and Operator shall keep the Owners informed in a timely manner with respect to important or significant Joint Operations.

Operator is hereby delegated the management of the Facility on behalf of the Owners and shall, subject to the direction of the Operating Committee, conduct or cause to be conducted all Joint Operations diligently, in a good and workmanlike manner, in accordance with good oil field and environmental practice, the Regulations and the terms of this Agreement. In the absence of specific instructions from the Operating Committee, Operator shall conduct or cause to be conducted, all Joint Operations, as would a prudent operator under the same or similar circumstances.

Without limiting the generality of any of the foregoing provisions of this Clause, Operator shall conduct and oversee all Joint Operations, and in particular shall: (

a) make and file all reports as required by governmental authorities relating to Joint Operations; (

b) maintain in the Province of Alberta complete and accurate accounts, books, records and documents in relation to the Facility and Joint Operations and provide each Owner with reasonable access thereto; (

c) provide Owners with reports as required and on a frequency and containing the information about Joint Operations as directed by the Operating Committee; (

d) on behalf of the Owners, complete all applications and obtain all licenses and approvals required by Regulations to conduct Joint Operations; (

e) promptly pay and discharge all expenses and taxes (other than income taxes) incurred in connection with Joint Operations and keep the Facility free and clear from all adverse claims and liens occasioned by Joint Operations, except claims or liens created under or pursuant to this Agreement or being contested in good faith; (

f) acquire and maintain all necessary surface rights, Material and services required to conduct Joint Operations and where Operator deems appropriate, use Its own equipment and facilities to serve such operations, subject to the Accounting Procedure; (

g) procure and maintain for the Joint Account the insurance set forth in the Appendix titled "INSURANCE" and use reasonable efforts to require contractors and subcontractors to procure and maintain such insurance as Operator deems necessary; (

h) comply with and, where applicable, require its agents, contractors and their sub contractors to comply with Regulations governing Joint Operations; (

i) subject to Clause 402, subcontract such portion of Joint Operations as Operator deems appropriate; (

j) furnish each Owner as soon as practicable with written notice of: (

i) physical damage to the Facility in excess of Operator's expenditure limit as provided in the Accounting Procedure; and (ii) any environmental, health, safety or other occurrence which is required to be reported under any Regulation and which either requires remediation costs exceeding the single expenditure limit set forth in the Accounting Procedure or could result in a punishable offence under the Regulations; (

k) extend to each Owner, at that Owner's sole risk and expense, the right to examine and inspect the Facility at all reasonable times in the presence of a representative of Operator and after giving Operator reasonable notice, except for portions of the Facility which are proprietary to a licensor to the extent that such licensor expressly prohibits examinations and inspection by such Owner; and

(

l) prepare and submit to the Operating Committee for approval the Forecasts provided for in Clause 605; provided further that, during Initial Construction of the Facility or any Enlargement or Modification, Operator shall also: (

m) carry out or cause the construction of the Facility and any Enlargement or Modification; (

n) contract with such Persons as Operator may deem appropriate for the performance of such work or undertaking, or any portion thereof; (

o) supervise all work related to such construction; (

p) acquire all Material required for such construction and the commencement and continuation of Joint Operations; (

q) supervise and have direct charge of all matters regarding design, construction and installation of the Facility and any Enlargement or Modification; and (

r) provide Owners with reports as required and on a frequency, and containing the information about construction, a Modification or an Enlargement as directed by the Operating Committee; [ 48 ] Clause 404 of the Operating Procedure states that the Operator is an independent contractor in conducting the Joint Operations (as defined in the Operating Procedure). The Operator shall determine the number of employees and contractors respecting its operations, their selection, their hours of labour and their compensation thereunder.

All employees and contractors used in the operations shall be the employees and contractors of the Operator. [ 49 ] Clause 305 of the Operating Procedure contemplates the operator is in possession and control of the facility, funds, records and other materials and substances. [ 50 ] Pursuant to clause 505 of the Operating Procedure, each of Paramount and Conoco are responsible for their proportionate share of environmental liabilities arising in relation to the Joint Operations. [ 51 ] Pursuant to clause 601 of the Operating Procedure, Conoco would set up a joint account for administering costs and expenses incurred by Conoco in connection with the Joint Operations (as defined in the CO&

O) of the Facility.

Pursuant to clause 602 of the Operating Procedure, if a Joint Owner failed to pay any of the costs and expenses incurred for the joint account, Conoco may charge "compound interest, as computed monthly... at the rate of two percent (2%) per annum higher than the rate designated as the prevailing prime rate for Canadian commercial loans by the principal Canadian charted bank used by the Operator" on the unpaid amount. [ 52 ] Pursuant to clause 503 of the Operating Procedure, except as set out in the CO&O, all liabilities and indemnities arising from Joint Operations would be for the joint account and would be borne by the Joint Owners in the proportion of their interest in the Facility (which was 50% for each of Conoco and Paramount). [ 53 ] Pursuant to clause 501 of the Operating Procedure, Conoco would not be liable to the Joint Owners for any loss, expense, injury or damage except when and to the extent that such loss was "a direct result of or is directly attributable to the Gross Negligence of Operator or its Affiliates, directors, officers, consultants, agents, contractors or employees". [ 54 ] Pursuant to clause 101(

y) of the Operating Procedure, "Gross Negligence" is defined as follows: (

i) a marked and flagrant departure from the standard of conduct of a reasonable person acting in the circumstances at the time of the alleged misconduct; or (ii) such wanton and reckless conduct or omissions as constitutes in effect an utter disregard for harmful, foreseeable and avoidable consequences. [ 55 ] Pursuant to clause 1102 of the Operating Procedure nothing therein shall be read or construed as creating a partnership between Paramount and Conoco and each their liabilities under the CO&O and Operating Procedure shall be several and not joint or joint and several. [ 56 ] Insurer’s counsel read in from the discovery of Paramount’s former corporate operating officer, that Paramount expected Conoco, as a major oil and gas producer, to operate the facility consistent with the operating agreement and in accordance with good oilfield practice.

Paramount did not, through the operating committee or otherwise, inquire into the maintenance or monitoring systems for the LVP pipeline. I accept this evidence. [ 57 ] In submissions, counsel for the insurers were critical of Paramount’s conduct as a member of the operating committee, effectively asserting it was complacent with respect to the operation of the LVP pipeline. To the extent they suggested Paramount itself was negligent in the operation of the pipeline for failing to oversee or supervise its operation, I do not accept their position. [ 58 ] Paramount’s reliance on Conoco was reasonable.

Paramount did not know, and there was no circumstance by which it ought to have known, of the various operating deficiencies that came to light after June 9, 2016 or that Paramount later alleged in the arbitration proceedings against Conoco: (

a) Conoco is a major producer with extensive oil and gas operations. Conoco’s personnel supervising the operation of the pipeline were long time and experienced employees. Paramount generally expected that Conoco would operate the facility with the necessary degree of prudence, as required under the CO&O and the extensive regulation of pipeline operations. (

b) There was no evidence of any corrosion issues in the Resthaven area generally or relating to the LVP pipeline. There was no evidence suggesting Paramount ought to have been aware of the deficiencies in Conoco’s corrosion mitigation or monitoring practices.

(

c) There was no evidence of any historical issues with the maintenance or operation of the LVP pipeline of which Paramount ought to have been aware, except that Paramount learned in December 2015 though receiving a mail in ballot to approve an expenditure on the LVP pipeline, that Conoco was proposing to install metering equipment to bring the pipeline into compliance with CSA Standard Z662 as required under Alberta law. Paramount agreed to the necessary expenditure to do so. (

d) Prior to April 2021, the pipeline was equipped with sufficient metering equipment to permit manual mass balancing (as defined in

Part V below) to monitor for leaks. There is nothing that would have caused Paramount to inquire into whether Conoco was actually performing manual mass balancing in any structured way or at all. (

e) Although Paramount was surprised when it received the mail in ballot to learn that the LVP pipeline required additional metering equipment to bring it into compliance with the recently revised CSA Standard Z662, the solution proposed by Conoco appeared normal and would not have provided any reason to inquire further into the leak detection systems or protocols. (

f) There is no evidence Paramount was aware or ought to have been aware of the various operating deficiencies which Paramount later alleged against Conoco or of any events that ought to have put it on inquiry into Conoco’s operating or maintenance practices relating to the LVP pipeline. V Regulatory regime [ 59 ] The LVP pipeline was a liquid hydrocarbon pipeline regulated under the Pipeline Act , RSA 2000, c P-15 . Pursuant to

section 9 of the Pipeline Regulation , AR 91/2005 (the “ Regulation ”), the latest published edition of several codes or standards issued by the Canadian Standards Association (CSA) apply to pipelines. [ 60 ] These codes include CSA Standard Z662, Oil and Gas Pipeline Systems , which sets out the minimum requirements for the design, construction, testing, operation, maintenance, repair and leak detection of pipelines ( Regulation , section 9(3) ).

The leak detection requirements contained in Annex E of CSA Standard Z662 are mandatory for liquid hydrocarbon pipelines ( Regulation , section 9(4) ). [ 61 ] Annex E requires the operator to develop, implement and periodically evaluate a leak detection strategy for its pipeline. The purpose of the strategy is to ensure methods are in place that will contribute to the certain and timely detection of a service fluid release in order to support and inform appropriate pipeline control and emergency response actions.

The strategy “should” include a pipeline leak detection system with a continuous monitoring capability. The leak detection strategy and systems “should” be integrated into pipeline control and emergency response procedures (Annex E, para E.1.2). [ 62 ] A leak detection system “shall” be implemented using one or a combination of various methodologies. An operating company “shall” evaluate applicable leak detection methodologies to determine their effectiveness for the pipeline under consideration and how various methodologies can complement each other.

This evaluation “shall” be documented. (Annex E, para E.1.3). [ 63 ] Where direct leak detection methods (a leak detection approach that uses one or more methods that directly sense leaked or leaking hydrocarbons) are used, the assessment “shall” include any factors that might impact the performance of the system, which “can” include the probable path of hydrocarbons, soil type, water content, depth of cover of the pipe, and type of product (Annex E, para E.3.2).

Examples of direct methods are liquid sensing, vapour sensing, acoustic emissions sensing, and “visual methods” (Commentary to para E.3.2). [ 64 ] In respect of computational leak detection methods (a method that relies on measurement of process variables from which an inference of a leak is drawn), the operator must establish leak detection thresholds based on expected hydraulic conditions that meet the sensitivity and reliability targets of the internal leak detection system (Annex E, para E.3.3.1).

Leak detection thresholds “shall” be set to the lowest practical values having regard to the considerations outlined in Annex E, para E.3.3.2. Leak detection system performance and alarm limits must be appropriate for the characteristics of the individual pipeline or particular segment thereof (Annex E, para E.3.4.1). [ 65 ] The leak detection system must provide clear alarms to alert the operator of a possible release. Further: A leak detection alarm shall result in initiation of a procedure to evaluate the leak condition and to determine the cause of the alarm.

Leak alarm evaluation shall be integrated into pipeline control procedures. The leak alarm evaluation shall lead to control action to mitigate the leak (such as pipeline shutdown) unless such deviations can be clearly and readily explained . (Annex E, para E.3.4.2. underlining added). [ 66 ] If the volume of product coming out of a hydrocarbon pipeline is less than the volume going in, the anomaly might indicate a leak. A small imbalance for a short period of time might not be material.

I accept the Plaintiff’s expert evidence (from Mr Scott) that Annex E requires operators to establish acceptable material balances based on normal operating conditions, and an imbalance above the threshold value should result in a shutdown unless the deviation can be readily and clearly explained. According to Annex E, para E.2 (Specific

definitions), material balance is a mathematical procedure based on the laws of conservation of mass and fluid mechanics, which is used to determine if a release of service fluid has occurred on a pipeline system. This is also referred to as a mass balance. VI The operation of the LVP pipeline and evidence of the release [ 67 ] The main coverage issue in this case is whether the release was detected within 720 hours of its commencement.

[ 68 ] It is necessary to determine when the release commenced to ascertain the time frame in which it must have been detected, then to decide what Conoco knew or believed and when, in order to decide whether the insurance coverages apply. (

a) Date that the release commenced [ 69 ] The insurers’ counsel asserts in their brief, that the parties agree the release started “April 21, 2016, if not earlier”. There was no elaboration of what was meant by “if not earlier”. [ 70 ] The insurers’ evidence includes a report of a consulting engineer opining that the release began on April 9, 2016 and that the leak volume was about 1/3 greater than Conoco’s historical estimates on which the parties and AER relied.

Conversely, there is also evidence that Paramount had previously retained an expert who had calculated that the leak commenced on or about May 16, 2016. [ 71 ] Paramount’s counsel accurately described the parties’ agreement in its brief, that the release of pollutants from the LVP pipeline commenced “on or about April 21, 2016”. [ 72 ] The parties did not address the duration contemplated by their agreement to a date “on or about” April 21, 2016.

The usual, objective meaning of this phrase is that the date is approximate and usually suggests that the potential variance between the approximate date and actual date is immaterial. [ 73 ] The pipeline was not active during March 2016 and through to either April 1, 2016 or April 9, 2016. The pipeline then became active until April 14, when the installation of the Coriolis meters commenced. Conoco again began using the pipeline for its ordinary purpose on April 21, 2016.

The insurers’ engineering consultant prepared a report acknowledging a possibility that the corrosion developed into a leak during the period of inactivity and that up to about 4 M 3 could have leaked out during the period of inactivity. However, he discounted the reliability of such estimate.

I do not accept this as evidence on the balance of probabilities that the pipeline began to leak during the period of inactivity. [ 74 ] The same report opines that a steady imbalance of condensate - which is indicative of a leak – was first apparent (in hindsight) around April 9, 2016. [ 75 ] There is contradictory evidence in the record. In addition to the competing expert reports that were prepared for the insurers or Paramount during the events leading up to the denial of coverage, Conoco had conducted a review in connection with AER’s regulatory investigation.

It reviewed data from various sources (such as facility condensate production, meters and tank levels, tank volumes, condensate transfer records, and trucking tickets), and concluded the leak commenced April 21, 2016. [ 76 ] The phrase “on or about” can include periods before and after the specified date. Depending what the parties meant by using the phrase “on or about”, the question could be whether the leak commenced: (

a) April 9, 2016 (therefore, outside detection date is May 8, 2016). (

b) April 21, 2016 (therefore, outside detection date is May 20, 2016). (

c) A few days before April 21, 2016. (

d) A few days after April 21, 2016. [ 77 ] It is unlikely the parties objectively contemplated their agreement for litigation purposes to include April 9 th , given the structure of their submissions and the various concerns expressed in the records whether a reliable inference could be drawn from the various meters and gauges installed prior to the provision of the Coriolis gauges.

Objectively speaking, they probably agreed to April 21, 2016 to avoid the debate whether to rely on Conoco’s estimate, the insurers’ estimate, or Paramount’s estimate. [ 78 ] Consequently, I find that “on or about April 21, 2016” means April 21, 2016 and any variation therefrom is immaterial. [ 79 ] In any case, if the parties did contemplate that the leak started as early as April 9 th , the possible contradiction in the evidence would not hinder the

summary trial. [ 80 ] Paramount submits that detection requires only that a person has observed or is aware of information indicative of a release. The Insurers submit detection requires that a person subjectively concludes or is actually aware that a release occurred or is occurring.

Another possible alternative, suggested by Paramount’s counsel during oral submissions, might be a person ought to know that the information of which they are aware is indicative of a release. [ 81 ] Given the information of volume discrepancies and pressure loss known to Conoco personnel by the end of May 7, 2016 and the lack of evidence that Conoco personnel had reason to believe before April 30, 2016 that a release was occurring, the possible variation in the exact date the leak commenced is immaterial for the purpose of establishing when Conoco detected the release or the nature of Conoco’s negligence during the events in question. (

b) Date that Conoco concluded there was a release [ 82 ] The LVP pipeline was constructed in 2006 by Conoco’s predecessor. It was equipped with pressure instruments and flow meters, and was connected to tanks with level gauges, that would permit Conoco as an operator to perform a manual mass balance comparing the quantities of product entering and exiting the pipeline. [ 83 ] Although Conoco could perform such calculations as part of a leak detection strategy, it did not regularly or frequently did so.

Instead, Conoco was (in its words) “flying the pipeline every two weeks to ensure that there are not any spills”. In other words, the historical leak detection strategy consisted of right of way inspections every two weeks. [ 84 ] In late 2015, Conoco issued a mail in ballot to Paramount seeking approval under the CO&O to install measurement of the liquids leaving the Resthaven gas plant through the LVP pipeline and arriving at the riser to the sales tank at the 08-11 plant. Conoco stated the purpose was to bring the pipeline into compliance with CSA Standard Z662 Annex E.

To accomplish the required leak detection, measuring for the liquids would be installed so that calculations could be done continuously to verify there were no pipeline leaks while pumping. [ 85 ] The proposed measurement equipment consisted of devices called Coriolis meters and associated equipment (for example, remote data acquisition equipment). The installation of these meters at each end of the pipeline allowed any flow imbalance to be directly measured with a high level of accuracy.

Coriolis meters are renowned for their accuracy and reliability. [ 86 ] Apart from Conoco’s acknowledgment that using visual right of way inspection as the leak detection method was not compliant with CSA Z662, Mr Scott opined during cross-examination that a visual right of way inspection cannot be the primary means of leak detection in Alberta pipeline operations.

I accept his evidence that applicable operating standards in Alberta did not, by early 2016 at the latest, permit visual inspections as the primary means of leak detection in Alberta. [ 87 ] The pipeline went out of service on April 14, 2016 to install the new Coriolis metering equipment. The installation was completed on April 17, 2016 (except for some remote data acquisition equipment).

The meters were calibrated and proven on April 20, 2016 and recording accurate information as of April 21, 2016. [ 88 ] At that time, the outlet meter was not yet connected to an automatic data acquisition system, so imbalances had to be calculated manually. There is no evidence that Conoco had set threshold values for data anomalies in the condensate volumes entering and exiting the pipeline, or if it had, whether it applied them. If there were such values, the evidence suggests they were not applied. [ 89 ] The first attempt by Conoco personnel to reconcile the new Coriolis meters was on April 30, 2016.

Conoco operators noted a discrepancy between the condensate shipped and received through the LVP pipeline on that day of 7.82 m 3 .

This discrepancy was 33% of the product shipped in a 24-hour period. [ 90 ] One of the Plaintiff’s experts, Mr Scott, is a professional engineer with over 23 years experience in liquid hydrocarbon pipelines in both technical and management positions and specializes in pipeline control, pipeline hydraulics, hydraulic simulation systems, and software based pipeline leak detection systems. [ 91 ] He opined that the discrepancy was material and indicated, or constituted an easily recognizable and clear alarm of, a large leak that under prudent operating practices required the operators to shut down the pipeline for intense investigation following logical step by step procedures.

The investigation should have continued until the reasons for the loss could be fully explained. In his cross- examination, he likened this information to a “deafening” alarm. [ 92 ] Conoco’s operators did not provide evidence in the

summary trial. Their log notes and interview statements with AER investigators were in evidence in the

summary trial. These record that the operators held various theories that could explain the measured shortfall and disbelieved the data provided by the meters. These meters were newly installed, and Conoco believed there was an issue with the new equipment. [ 93 ] However, the evidence does not indicate the operators had eliminated the concern that the pipeline could be leaking. They recognized there was an issue. They arranged to pressure test the pipeline. On May 6, 2016, Conoco pressured up the pipeline to 950 kPa and shut it in for the night.

This pressure was at or a little less than the normal operating pressure of the pipeline. Conoco recorded that on May 7, 2016, the line had “bled down” to 250 kPa. [ 94 ] Mr. Scott observed that the pressure test was an independent means of checking pipeline integrity. Although the records he reviewed did not provide the rate of pressure decrease, the low line pressure at the end of the test would have required, under good practices, that operator shut down the line and investigate to explain the reasons for the failed test. [ 95 ] Conoco continued to use the pipeline to transport condensate.

The operators continued to question the data (now both metered volumes and pressure data from the test) and suggest alternative explanations that could (not would) explain why the meters were out of balance and why the pressure had dropped. [ 96 ] On May 9, 2016, Conoco’s operators recorded further failed meter reconciliations. The meters remained about 30% out of balance. The reason for the imbalances were apparently not understood by the operators, as clearly demonstrated by an email of that day from the area foreman to an operations leader: … Let me know what you would like to try next and we will proceed.

This system has not balanced from Day 1 when we started it.” [ 97 ] The recipient responded the same day: The 7 day average is showing 29.7% variance. When the pump was started today it took 26 min and 3.9m 3 at 01-36 before we could see any flow at 08-11. [ 98 ] Mr. Scott characterized the anomaly at this point as overwhelming evidence of a leak. [ 99 ] On May 19 th , Conoco operators recorded a similar anomaly. On this occasion, the operators also recorded and entered in their log, the increase in the sales tank volume at the 08-11 plant recorded by the tank level gauge.

This value was very similar to that provided by the Coriolis meter at the outlet end of the LVP pipeline, providing strong evidence that the Coriolis meter at the outlet was accurate.

[ 100 ] Mr. Scott opined that each of these occurrences were strong leak “triggers”. The May 19 th result, taken with the previous anomalies and observations, was (in his opinion) conclusive evidence of a leak. [ 101 ] Conoco’s operators did not see the situation that way. Discussions continued. One note in evidence records that even as late as May 25 th , in a Conoco meeting: May 25 th leadership meeting working with all parties on the LVP balance issue. Felt confident the line was not leaking.

Has been thoroughly inspected. [ 102 ] During this period, Conoco continued to inspect the pipeline right of way and did not observe indications of hydrocarbons on the surface.

Mr Scott stated that this approach was mistaken, because it is generally known that when a pipeline failure occurs the product might travel underground and surface at the bottom a slope. [ 103 ] As noted, until June 9, 2016 Conoco personnel had not conducted an inspection of the lands or waterbodies adjacent to the right of way. [ 104 ] Conoco’s operations leader in the area, with years of experience, stated to AER in a post-loss interview shortly after the event, that they had never observed product travelling underground in this manner without coming to the surface. [ 105 ] Under Annex E, a proper external inspection plan must assess the probable path of the hydrocarbons and subsoil conditions (Annex E, para E.3.3).

There is little evidence in the record of the

summary trial that Conoco personnel did or did not have information of the subsurface conditions. [ 106 ] In post-leak proceedings against Conoco, AER found that Conoco did not have a leak detection manual, fully in effect and implemented, for the LVP pipeline and did not have an established and effective leak detection programme for the LVP pipeline. The parties included these findings in their agreed statement of facts.

In the absence of a contrary agreement, I can (and do) take these statements as evidence of the truth of their contents. [ 107 ] In doing so, I am mindful that Conoco’s post-incident report indicated it lacked an effective leak detection manual for the LVP pipeline. Further, it noted conflicting information existed as to the intent and status of the manual that was located; some personnel regarded it as in effect and others regarded it as a draft. Conoco observed there was no control of the document. [ 108 ] Conoco’s report is hearsay. It appears in the records for the

summary trial and neither side objected to its admissibility. Given that Conoco could not demonstrate to the regulator that it had an effective manual, I have no difficulty in accepting AER’s findings on this point set out in the parties’ Agreed Statement of Facts. [ 109 ] In short, the Conoco operators did not have an organized, documented approach to assessing evidence of leaks for the LVP pipeline.

They did not have or express a decisive explanation for the volume anomalies detected and observed commencing April 30 th and in early May, the pressure loss observed in the test of May 6-7, 2016, or known anomalies thereafter. Conoco personnel questioned the metering and pressure test data and believed the pipeline was not leaking.

In the meantime, they continued to operate the pipeline and thereby transport hazardous materials through and adjacent to a remote and environmentally sensitive landscape at great risk of significant environmental damage. [ 110 ] By doing so, the evidence in this trial (to which Conoco is not a party) demonstrates that Conoco did not comply with the mandatory requirements of Annex E for leak detection in the period following the commencement of the leak through June 9, 2016. It had not established threshold values for the LVP pipeline.

It relied on visual inspection of the pipeline right of way as the primary method of monitoring for leaks. It did not shut in the pipeline or bring it to a safe state when faced with mounting data indicating a leak, which it could not explain (to the required standard) was due to other causes. [ 111 ] It is implausible that Conoco would have continued to operate the pipeline if its personnel subjectively believed it was leaking. The only plausible explanation for Conoco’s actions was as they stated: they remained in disbelief that the Coriolis meters, which had been recently installed, were accurately set up.

VII Whether the loss was detected within 720 hours [ 112 ] The primary issue is: when did Conoco personnel detect the release? Was it when the known data indicated the possibility of a leak? The date when the operator was legally required to behave as if the pipeline were leaking, or when Conoco ought to have known or had reasonable grounds to believe the pipeline was leaking having regard to the known data? When Conoco personnel subjectively concluded the pipeline was leaking? When the evidence shows on the balance of probabilities that the pipeline was leaking?

This requires the Court to discern the meaning of “is detected by any person” as used in the insurance policies. [ 113 ] In addition, the insurers argue that Paramount previously admitted in the arbitration that Conoco did not detect the release until June 9, 2016 and is bound by that admission.

The issues here are the nature of the admission and whether Paramount is bound by it. [ 114 ] The insurers further argue that to the extent Conoco previously asserted or represented that Conoco was not aware of the leak until long after its commencement, Paramount is bound by such positions and cannot take a contrary position in the present litigation. The insurers advance a similar argument arising from AER’s regulatory findings.

The issues here are mainly whether Conoco acted as Paramount’s agent so as to bind Paramount to these admissions and findings, and the nature of the admissions and findings and the context in which they were made. (

a) Policy

interpretation principles

[ 115 ] The insured bears the onus of first establishing that the loss falls within the coverage grant of the policies ( Ledcor Construction Ltd v Northbridge Indemnity Insurance Co , 2016 SCC 37 , [2016] 2 SCR 23 at para 52 ). I agree with the insurers that the coverage conditions are not interpreted as if exclusion clauses ( Bassett & Walker International Incorporated v Export Development Canada , 2017 ONSC 618 at para 54 ). [ 116 ] The parties are in agreement about the applicable principles of contract

interpretation. [ 117 ] The goal is to “ascertain the objective intent of the parties through the application of legal principles of

interpretation” ( Sattva Capital Corp v Creston Moly Corp , 2014 SCC 53 , [2014] 2 SCR 633 at para 49 ). To this end, “the exercise is not to determine what the parties subjectively intended but what a reasonable person would objectively have understood from the words of the document read as a whole and from the factual matrix” ( IFP Technologies (Canada) Inc v EnCana Midstream and Marketing , 2017 ABCA 157 at para 79 ).

Determining the intention of the parties is a “fact-specific goal” that requires a trial court to “read the contract as a whole, giving the words used their ordinary and grammatical meaning, consistent with the surrounding circumstances known to the parties at the time of formation of the contract” ( Sattva at para 47; Ledcor at para 27). [ 118 ] The primary interpretive principle is that “where the language of the insurance policy is unambiguous, effect should be given to that clear language, reading the contract as a whole” ( Ledcor at para 49; Sabean v Portage La Prairie Mutual Insurance Co , 2017 SCC 7 at para 12 ). [ 119 ] Where, however, ... the policy’s language is ambiguous, general rules of contract construction must be employed to resolve that ambiguity.

These rules include that the

interpretation should be consistent with the reasonable expectations of the parties, as long as that

interpretation is supported by the language of the policy; it should not give rise to results that are unrealistic or that the parties would not have contemplated in the commercial atmosphere in which the insurance policy was contracted, and it should be consistent with the

interpretations of similar insurance policies. ( Ledcor at para 50; also see Sabean at para 12 ). [ 120 ] Only if ambiguity still remains after the above principles are applied can the contra proferentem rule be employed to construe the policy against the insurer ( Ledcor at para 51; Sabean at para 12 ). A “corollary of this rule is that coverage provisions in insurance policies are interpreted broadly, and exclusion clauses narrowly” ( ibid ).

Also, where an insurance policy is ambiguous, courts “strive to ensure that similar insurance policies are construed consistently” ( Ledcor at para 40). [ 121 ] The courts should be “loath to support a construction which would either enable the insurer to pocket the premium without risk or the insured to achieve a recovery which could neither be sensibly sought nor anticipated at the time of the contract” ( Ledcor at para 79). [ 122 ] Surrounding circumstances (or the factual matrix) in which the contract was made are important considerations ( Ledcor at para 27, 31).

These are the facts that were known or ought to have been known by the parties at the time of contracting ( Sattva at paras 58, 60; IFP at para 83). [ 123 ] Surrounding circumstances include the purpose of the contract, the nature of the relationship it creates, and the market or industry in which it operates ( Ledcor at para 31; IFP at para 83).

Surrounding circumstances may include the customary practices in the industry in which the insured operates ( Nexxtep Resources Ltd v Talisman Energy Inc , 2013 ABCA 40 at para 35 ) and regulatory regime ( Nexxtep at para 33; Nodel v Stewart Title Guaranty Company , 2018 ONCA 341 at para 55 - 56 ). [ 124 ] While the surrounding circumstances are relied upon in the interpretive process, “courts cannot use them to deviate from the text such that the court effectively creates a new agreement” ( Sattva at para 57). [ 125 ] When does an ambiguity arise?

In IFP the Court stated at para 87: Mere difficulty in interpreting a contract is not the same as ambiguity: Paddon Hughes , supra at para 29. A contract is ambiguous when the words are “reasonably susceptible of more than one meaning”: Hi-Tech , supra at para 18. An ambiguity in the contract also allows courts to consider evidence of the parties’ subsequent conduct post-contract: Shewchuk v Blackmont Capital Inc ., 2016 ONCA 912 at paras 46 , 56, 404 DLR (4th) 512; Hall , supra at 83-85.

But it must be understood that even under this ambiguity exception to the parol evidence rule, there are limitations as to what parol evidence is admissible. In this regard, evidence as to the parties’ subjective intentions is generally inadmissible. [ 126 ] Similarly, Strekaf JA dissenting in 2102908 Alberta Ltd v Intact Insurance Company , 2023 ABCA 34 described the jurisprudence as follows: [44] “An ambiguity exists where there are ‘two reasonable but differing

interpretations of the policy”: Barbara Billingsley, General Principles of Canadian Insurance Law , 3rd ed (Toronto: LexisNexis, 2020) at 139-140, citing Sabean at para 42 . “[M]ere articulation of a differing

interpretation does not always establish the reasonableness of that

interpretation and does not necessarily create ambiguity”: Sabean at para 42 . “Where more than one

interpretation is supported by the text of a policy, the court is directed to consider the reasonable expectations of the parties, and to avoid an

interpretation that would give rise to an unrealistic result or that would not have been in the contemplation of the parties”: Tien Lung Taekwon-Do Club v Lloyd’s Underwriters , 2015 ABCA 46 at para 25 . [45] An ambiguity must be ‘real’. “That is, the words of the provision must be reasonably capable of more than one meaning having regard to the entire context of the provision”: Cardinal v Alberta Motor Association Insurance Company , 2018 ABCA 69 at para 11 , citing Bell ExpressVu Limited Partnership v Rex , 2002 SCC 42 at para 29 . [46] “Whether or not a word is ambiguous involves the consideration of its use in its place and context. It is only when two or

more different meanings are equally, reasonably and sensibly applicable that it can be said to be ambiguous”: Pentagon Construction(1969) Co Ltd v United States Fidelity & Guaranty Co, 77 DLR (3d) 189 at 192, (BC CA), cited in Gordon GHilliker, Liability Insurance Law in Canada, 7th ed (Toronto: LexisNexis, 2010) at 44, § 2.77 [Hilliker]. In other words, “[i]f the wordsof the exclusion are not reasonably capable of more than one meaning having regard to the entire context of the Policy, there is noambiguity”: Condominium Corporation No 9312374 v Aviva Insurance Company of Canada, 2020 ABCA 166 at para 14. (

b) Detection of the release (

i) Surrounding circumstances – insurance industry [127] The insurers provided a report from a prominent and experienced insurance consultant. This report describes the history andevolution of pollution liability and insurance coverage in Canada and the United States.

The consultant was asked to address in his reportthe purpose of detection and reporting cover in a pollution liability policy; the purpose of using the words “detected by any person” insuch coverage; and the steps reasonably expected of an insured after a loss involving the escape of pollutants is identified. [128] I agree with Paramount that the consultant’s opinions of the meaning of “detect” or “discover” and whether Conoco’srepresentatives detected the leak within 720 hours of its commencement, are not admissible. (

a) To the extent he provided the meaning of “detected” or “discovered”, that is an insurer’s subjective perspective. Throughoutthe discussion, the consultant does not ascribe any customary or accepted industry meaning to, or widespread understanding of, the word“detect”. (

b) The objective meaning of the detection clause is a question of mixed law and fact for determination by the Court, and hisopinions on this point do not meet the necessity requirement for admitting expert opinion evidence. (

c) His opinions that Conoco did not meet the requirement are applications of the meaning of “detected” to assumed facts. Thisis not a case where expert expertise is necessary to determine historical facts concerning Conoco’s information and beliefs, or to applythe meaning of detection to those facts. This aspect of his opinion does not meet the necessity requirement. [129] I pause to note that the insurers’ engineering consultant also expresses opinions about the meaning of detection and discovery,Conoco’s state of mind and beliefs, and whether Conoco detected or discovered the release.

These aspects of his report are similarlyinadmissible because they are beyond the scope of his expertise as an engineer and the necessity requirement for admission of expertevidence is not satisfied.

I have not considered similar opinions in Paramount’s affidavit evidence. [130] The insurance consultant’s description of the general historical evolution of pollution coverage is admissible as surroundingcircumstances that were known or ought to have been known to the parties, including Paramount as a sophisticated insured assisted by anexperienced insurance broker. [131] According to the consultant, coverages written in the 1960’s to cover accidents and occurrences became inadequate with theonset of increasingly expensive and unpredictable claims for contamination or environmental damage.

Some insurers responded bycurtailing the writing of coverage for pollution losses by attempting to exclude all cover for pollution liabilities. However, competitivepressures and demands from insureds for coverage led to evolution of coverage language so that pollution coverage could be madeavailable.

In this movement to curtail liability, the objectives were to remove ambiguity over the scope of coverage; differentiate betweenthe environmental impact of catastrophic events as opposed to pollution arising from the ongoing daily operations of an insured; excludethe reckless and intentional polluters who failed to take reasonable steps to prevent pollution; and curtail liability which the industry feltwas “excessive”. [132] This history mainly focussed on various attempts to re-write coverage and exclusion provisions.

The consultant acknowledgedthat little was written about discovery clauses, but the general principles behind pollution coverage are helpful “in determining why adiscovery clause is the critical “gatekeeper” to determine whether the policy will respond or not”. [133] The consultant summarized that insurers sought to limit their obligations so that they did not cover liabilities arising fromescapes, spills or discharges which were “continual in the work product of the insured” and ensure that deliberate discharges and smallleaks or escapes that were considered part of carrying on business of operating a pipeline or oil production facility were not covered. [134] The consultant stated that the discovery and detection provisions were part of the industry’s response to this problem ofexcessive liability and to clarify the intent to cover “catastrophic events, not ongoing continual pollution as a result of industrialoperations”.

Further: ... While the commencement of the escape is a precise point in time, insurers except insureds to manage their operations in a manner thatfacilitates early detection of leaks and environmental threats. Put another way, the insurers assume the pollution risk only if the operationuses procedures, processes and machinery to monitor the flow so that any leakage resulting in environmental damage could be quicklydetected, and preventative action taken.

The reporting provisions try and bring the insurer into the event quickly enough so that the insurers can assist in monitoring andmitigating the remediation costs. [135] Throughout his report, the consultant did not ascribe any customary or accepted industry meaning or widespread understandingto the word “detected” or the concept of detection generally.

At one point, the consultant states: The insurance industry developed a response that, rather than focus on exclusions to the existing coverage, would define the risknarrowly. “Pollution incident” was the peril but to fall within the definition to claim under the peril, several conditions had to be met. Notonly did the loss have to be unintended and unexpected from the standpoint of the insured, but the “discovery” and “detection” of the

loss had to be within a certain number of days of escape. The insured must have become aware of the commencement of the escape within a certain number of days. A footnote to the end of this passage states: This was intended to remove the “new” discovery of pollution that had been ongoing for years. For example, the recent discovery of open sewage going into Hamilton Harbour determined it had been ongoing for decades. The commencement of the spill was in the 1920s. [ 136 ] This passage ascribes an intention to a collective of pollution liability insurers.

However, surrounding circumstances do not include the subjective intentions of contracting parties. [ 137 ] I take from the consultant’s report only that the detection clause is part of the policy structure intended to limit exposure to accidents that are ongoing and unaddressed for a defined time period. Pollution cover was liable to result in exposure to expensive losses that accumulated over long periods of time without any reason to suspect their existence.

The detection requirement was one means of controlling that risk. [ 138 ] However, the history of these policies as related by the consultant does not assist in determining in the context of specific policy wording whether a loss is detected by a person (

a) if they become aware of material or important evidence of a release or (

b) only where a person subjectively believes or becomes aware that a release has occurred. [ 139 ] The consultant noted in his report that the polices were drafted in the context of the insurance industry expectation that insureds would conduct their operations in a responsible way, in the words of the consultant “to manage their operations in a manner that facilitates early detection of leaks or environmental threats” or use “procedures, processes and machinery to monitor the flow so that any leakage resulting in environmental damage could be quickly detected, and preventative action taken.” [ 140 ] This expectation is probably true of most regulated industries handling dangerous substances, and in my opinion would obviously have been in the minds of both parties.

Both sides’ proposed definition of the concept of detection is consistent with this expectation, because responsible and prudent operators will take credible evidence of a release as seriously as a belief that there is a release and commence the necessary further investigations and mitigating actions. (ii) Surrounding circumstances – oil and gas industry [ 141 ] The regulatory requirements for the safe and efficient operation of pipelines including the obvious need for effective leak detection requirements, were part of the context or surrounding circumstances in which the policies were made. [ 142 ] Both insurers and insured ought to have known that the insured was operating in a highly regulated industry, where participants are expected to observe safe and efficient practices in the construction, operation, discontinuation and abandonment of pipelines including the control of pollution and conservation of the environment ( Pipelines Act , RSA2000, c P-15,

section 4 ), and required to adhere to detailed requirements for the design, construction and operation of pipelines ( Pipeline Regulation , Alta Reg 91/2005 ). These standards include CSA Standard Z662, which had been modified shortly before the policies were issued to enhance detection and monitoring standards for hydrocarbon pipelines.

Both knew or ought to have known that pipeline integrity and leak detection are extremely important in the oil and gas industry given the potential consequences of a leak of hydrocarbon substances into the environment. [ 143 ] It is notable that according to mandatory industry standards under Alberta law in place when the policies were issued, a leak may be detected through direct (or external) methods or computational (or indirect methods) (CSA Standard Z662, Annex E, para E.2, E.3.1, E.3.2, and E.3.3).

A leak detection system: ... shall provide clear alarms to alert the pipeline controller of a possible release. A leak detection alarm shall result in initiation of a procedure to evaluate the leak condition and to determine the cause of the alarm. Leak alarm evaluation shall be integrated into pipeline control procedures. The leak alarm evaluation shall lead to control action to mitigate the leak (such as pipeline shutdown) unless such deviations can be clearly and readily explained . (Underlining added). [ 144 ] Further, the Z662 Standard requires an operator to presume a leak until the alarm is cleared.

Para E.4.3.2 provides: Analysis of leak alarms shall determine the cause of the alarm. The leak alarm shall not be discounted and declared invalid without such analysis; all alarms shall be assumed to have a cause . Methods to determine the cause of the alarm shall be developed. The leak detection system analysis procedure shall state a maximum analysis period.

If the cause of the leak alarm has not been declared within this period, the pipeline shall be brought to a safe state until the leak alarm cause shall be determined . (Underlining added). [ 145 ] The Z662 Standard also addresses the converse situation, where critical components are inoperative. A “critical process” is one upon which the leak detection methodology relies and which is essential for the operation of the leak detection system, and “critical data” is any data that drives the leak detection system application or is fundamental to the calculations (Annex E, para E.2,

definitions of “Critical process” and “Leak detection system”). If critical data is missing or a critical process is inoperative, the pipeline controller shall determine whether the leak detection system is considered to be ineffective and the pipeline shall be shutdown, or an alternative leak detection method may be used, allowing the pipeline to remain in service (Annex E, para E.4.2.3). This regulatory context lends some support to Paramount’s definition of detected – a leak is detected where there is some evidence of a leak and that evidence has not been clearly and readily explained by some other cause.

(iii) Case law concerning similar provisions [146] Both sides stated they could not find precedent case law interpreting the specific coverage grant in question. [147] Paramount’s counsel cites case law imposing an objective standard (whether a person of ordinary prudence would foresee that aclaim would arise) for determining whether an insured had sufficient information to trigger its obligation to give notice of claim under anauto policy (Hogan v Kolisnyk, (AB KB) at paras 59 – 62 citing Marcoux v Halifax Fire Ins Co, (SCC), [1948] SCR 278). [148] Paramount’s counsel also provided a lengthy

schedule of cases using the words “detect” and “discover” in various contexts.None address a similar case. [149] The insurer’s side cites three cases. However, none are directly on point. [150] In Irving Oil Ltd v Institute of London Maritime Insurance Co, 2000 NBCA 23, the policy required that the accident beidentified as first commencing at a specific point in time during the term of the policy and “became known” to the insured within 180days. The issue was whether coverage arose upon knowledge of the damage resulting from a leak of a pollutant or upon thecommencement of the leak.

In that case the Court observed: [15] We appreciate that sub-surface migrating petroleum is not often capable of being detected at the commencement of the originalleak or discharge, or even at the moment of first escape from the property of the insured. However, the terms of the policy before us dofocus on the initial commencement of pollution and it is that escape of pollutant that triggers liability coverage. [151] The insured could not prove the leak commenced during the policy period, therefore its claim failed. [152] In Compagnie d'assurance du Québec c.

Groupe pétrolier Nirom inc, (QC CA) the Court considered apolicy condition that the pollution incident begin or commence during the policy period and be discovered within 120 hours. The Courtobserved: If this were a pollution accident or disaster where the moment of occurrence could be established with accuracy, there would be nodifficulty in calculating the period of 120 hours prior to discovery. If a fuel truck overturned spilling its contents into a field, forexample, it would normally present no problem in fixing the time of the accident and determining whether it was discovered within 120hours.

But where the pollution is caused by a leaking underground tank with a defective solder, as was the case here, the exact moment offailure of the defective soldering work cannot be fixed with any precision. Nor should the insured be held to a delay when it wasunaware of the occurrence of failure or the commencement of the leak.

The moment for the commencement of the period betweenoccurrence and discovery must be the moment when the insured can reasonably have acquired knowledge that there was a leak or failure.The failure causing the leak may well have occurred at a precise moment in January or February 1992, but it was not apparent at once. Quite obviously, it manifested itself gradually, probably over a period of several days. ...

In the context of the present case, what this principle would mean is that the delay of 120 hours required under the insurance policy forthe discovery of the pollution accident would only begin to run when the insured had knowledge, or should have had knowledge, thatthere was a leak or a failure in the underground fuel storage system. Commencing on the date of that knowledge, or presumedknowledge, the insured was then bound by a delay of 120 hours to discover the cause of the problem and the actual condition of theunderground fuel storage equipment.

Obviously that discovery required an excavation and physical examination of the fuel tanks, anoperation that no insured would want to undertake unless there were reasonable grounds to believe a leak or failure had occurred. [153] The policy in that case was differently worded than the present case because it did not specify the starting point of the 120 hourperiod. The question was, within 120 hours of what event?

Further, it did not consider the meaning of “detected”. [154] Finally, the insurers cite Harvey's Oil v Lombard General Ins, 2003 NLSCTD 158, where the Court quoted a passage fromLichty & Snowden, Annotated Commercial General Liability Policy on which the defendant insurer relied, as follows: [31] Lombard submits that the Pollution Liability Coverage Extension Endorsement, although commencing with language similar toan Absolute Pollution Exclusion, is not in fact an exclusion but rather is an endorsement extending the coverage of the CGL policy tocertain specific circumstances.

Lombard argues that the Endorsement accordingly should be interpreted in a manner consistent with theintention of the parties, that intention being to exclude some pollution incidents and at the same time to grant coverage in other specificcircumstances. Lombard refers to Lichty & Snowden, Annotated Commercial General Liability Policy, at p. 31-25: The insurance industry, while recognizing that it does not wish to assume the full brunt of a pollution risk, has come around to the viewthat it can underwrite limited exposures where the policyholder has sufficient monitoring and safety measures in place.

The result is anendorsement which, although still excluding most risks, gives back certain time-limited coverage in the form of an exception clause orclauses.

Known to some as '120 hour' detection and reporting cover, this endorsement is intended to pick up the truly accidental incident in anindustrial setting, where the risk of loss and contamination is likely to be contained and dealt with prudently as part of an overall riskmanagement program. [155] The Court did not accept or reject this information, and instead decided the claim on a different coverage provision to which the"120 hour detection and reporting cover" did not apply.

[ 156 ] A more recent edition of the same publication, Lichty & Snowden, Annotated Commercial General Liability Policy (Thomson Reuters, looseleaf) at p 31-56, observes: ... Certain defined and limited coverage is available, by way of endorsement, added to the CGL policy. The limited coverage provided through such endorsement varies from insurer to insurer. Generally the endorsement, while continuing to exclude coverage for most loss arising from polluting events, provides certain time-limited coverage. The endorsement is sometimes referred to as a “120-hour detection and re

[…]