4.0 Audit Process

4.1 Introduction

1. confirm the remediation was carried out as intended by the Remediation Action Plan (RAP);

2. confirm that soil in the various areas complied with relevant SAC;

3. confirm that groundwater complies with relevant SAC;

4. confirm that the site (including the tidal areas of the Waimea Inlet and Channel) have been remediated to the extent practical and/or are fit for their future purposes; and

5. confirm that the reporting (including the Validation Report and site management plan) contain adequate information for record and future management purposes.

In order to do this, the auditors must gain sufficient confidence that the actual condition of the site is essentially as claimed by the various reports and records. In order to gain such confidence, sufficient records must be examined to determine that:

• the remediation work was actually carried out as proposed and/or reported;

• validation sampling and testing was carried out in accordance with appropriate protocols and at an appropriate frequency (density) of testing for the various parts of the site. This is to determine both that the sample result can be trusted and that the sample is sufficiently representative of what is being tested;

• the results have been appropriately interpreted, including the use of statistical analyses;

• systems were in place to detect non-compliant results, that appropriate remedial action was taken in the event of non-compliance and that such action was recorded; and finally

• that proposed future management is appropriate to the site’s condition and intended use

In the event that confidence was not gained on any aspects, the audit then sought to identify any deficiencies or gaps in information, assess the effect of those deficiencies or gaps on the final outcome and make recommendations for additional work to fill the gaps if appropriate.

The primary sources of information to guide the audit process were expected to be:

• the resource consent conditions relevant to the soil and water quality;

• the RAP³, which should reflect the resource consent conditions and set out the detail of the remediation to ensure the remediation will achieve its intent. This would typically include monitoring of the remediation, validation testing and means of interpreting the testing;

• the Validation Report (a consent requirement) which should describe the actual remediation (or at least reference other documents that do this), summarise the results of testing, interpret the results relative to the requirements set out in the consent and RAP and determine whether the requirements were achieved, or not. The Validation Report should also record any variations from the RAP and non-compliance with mandatory requirements;

• a random sample of primary records such as daily site records, laboratory results and weekly or monthly reports;

• additional reports and other information to assess the condition of the groundwater and consider the impact of the various reagents used in the MCD process, as these were beyond the scope of the Validation Report; and

• various guideline documents, including MfE’s Contaminated Land Management Guidelines No.1 – Reporting (which sets out recommendations for the contents of a validation report) and No.5 - Site Investigation.

4.2 Audit Progression

The audit commenced with an initial review of the following documents to gain a preliminary understanding of the remediation, and to determine the adequacy of the information presented for audit:

• the Assessment of Environmental Effects (AEE) for the remediation (T&T, 2003);

• the various resource consents covering the remediation works;

• the two basic versions of the RAP (Thiess, 2004 and MfE, 2005a) and a summary of changes to the RAP (MfE 2007). The 2005 version of the RAP contained various work plans, which themselves had also been revised several times in some cases; and

• the Validation Report (SKM, 2008)

It was immediately apparent that there was not a TDC (as the consent authority) approved RAP against which to judge the remediation. There were conflicts between the RAPs, and between different versions of work plans appended to the MfE RAP (particularly different versions of Work Plan 13 – Validation Sampling and Analysis, which set out the SACs and where and how samples were to be taken and the results interpreted). It was not clear which RAP or work plan should be referred to. This makes it difficult for the Validation Report to do one of its jobs, which is to compare actual methods employed with the intention, and report any variation that might have significance for the success of the remediation.

In addition, it became apparent that the Validation Report did not contain a sufficiently complete description of the methods employed for the various aspects of the remediation to gain confidence that what was supposed to have been done had in fact been done. If it is not clear how sampling was carried out or whether the proposed remediation method was actually followed, there is a loss of confidence in the subsequent validation of the remediation. That was the case on first examination of the above reports.

It therefore became necessary to review a large volume of site records and consult individuals involved in various parts of the remediation. The additional information was sought to: clarify what was actually done; to determine whether what was done complied with the intentions as defined by the RAPs or, in the absence of an approved RAP, some other agreed document (e.g. record of acceptance of an approach recommended by the Site Auditor or Peer Review Panel); and to arrive at a judgement that what was actually done was satisfactory.

The key sources of additional information utilised were:

• a site visit and meeting with Jenny Easton of Tasman District Council (TDC);

• a meeting with Paul Russell and Juliet Westbury of MWH (the Site Engineer). Additional information was subsequently supplied by MWH as a result of the initial meeting;

• a meeting with Chris Purchas, Bruce Clarke and Philip Outram of SKM (the authors of the Validation Report). Additional information was subsequently supplied by SKM as a result of the initial meeting;

• a review of correspondence between the remediation Site Auditor (Peter Nadebaum of GHD, Melbourne) and MfE during the remediation works

• telephone contact with the Site Auditor;

• a review of a partially completed QA / QC document by EMS which was discovered in MfE records (EMS, 2007). This document sets out most of the actual methods employed during the remediation and provides discussion where the methods vary from the original Thiess RAP;

• a review of all monthly remediation summary reports prepared by EMS for MfE;

• a review of selected site records by EMS; and

• a review of additional monitoring of sediment, groundwater and biota undertaken by or for TDC at various times.

The audit conclusions have been arrived at by assessing the entire set of information reviewed. Where relevant, the specific source of information relating to a particular aspect of the audit is cited if the information is not contained in the Validation Report.

Key components for which extra detail was sought included:

• the adequacy of the sampling and analytical suites;

• the various validation sampling methods used;

• the tracking and control of the various material categories during the remediation;

• the quality assurance and quality control methods employed;

• the marine sediment remediation methods; and

• the validity of the various datasets and the statistical methods used to assess them in the validation analysis.

Instances where a perceived information gap could not be addressed are discussed in the section of the report relating to the missing information. Overall, the level of information finally reviewed was sufficient to support the audit conclusions. A list of documents reviewed is provided in .

4.3 RAP Status

As noted above, it would appear that the RAP (and associated management plans) was not approved by the consenting authority, as required by Condition 9 of the various consents, due to parts of the RAP never being finalised. However, Condition 12 states:

To ensure flexibility in achieving the results for avoiding remedying or mitigating any adverse effects the provisions of the plans and work procedures submitted in accordance with Condition 9 shall not be conditions of this Consent provided however the Consent Holder shall not depart from the provisions of any plans submitted and approved in accordance with Condition 9 in a manner that any adverse effect caused by departure will be more than minor.

In addition, Condition 13 states with respect to reporting to the TDC’s Compliance Coordinator, the Site Auditor and the Peer Review Panel:

Where necessary, each of these parties can request a meeting of all the parties to discuss the activities of the remedial works and their compliance with consent conditions and to make recommendations for amendments of the Plans cited in Condition 9 to the Consent Holder.

It would appear that the RAP was a “living document” that was updated from time to time. However, EMS (2007) in managing the site stated that MfE indicated in November 2006 that, until such time that a new RAP was approved, EMS was to follow the Thiess RAP dated July 2004 for validation. As noted below, this was not always the case as subsequent variations to the work plans were implemented by EMS on-site. SKM appears to have followed the Thiess RAP in preparing the Validation Report.

It is clear, however, that TDC approved variations from the original SACs, which were reflected in later versions of the RAP, including changes to the SACs for manganese and nickel. There was also discussion between the Site Auditor and MfE on acceptance or otherwise of treated soil where the test result was within 2.5 times the SAC (as set out in the Thiess RAP). The conclusion was that it was appropriate to accept results up to 1.5 times the SAC for the treated soil, rather then 2.5 times the SAC, to ensure the average (calculated as the 95% upper confidence limit of the mean, or 95% UCL) remained within the SAC.

Given the changes throughout the remediation, for the purpose of this audit it has been assumed that the latest version of the RAP (MfE, 2005 and Work Plan 13, v3 dated 30 April 2007) provides the relevant criteria against which to judge the remediation. It is apparent from reviewing the various documents on the progress of the remediation, including the Site Auditor’s recommendations, that at least with respect to most issues of importance to this audit (but not necessarily all issues), there was essential agreement by all parties.

There is an important exception; that of the requirement on the original RAP to split a percentage of samples for analysis in both the primary laboratory and an independent second laboratory. This was never carried out. The last version of the Work Plan 13 appears to have been written as a record of what occurred, i.e. post remediation sampling of FCC East, rather than what should have been a pre-determined programme of checks. This is considered in greater detail in Section 1.1.3.