3 Methodology

3.1 Timber industry consultation

3.2 Data collection and compilation

Over the last 10-15 years a number of site investigations have been undertaken, the majority of which have been PCP rather than dioxin investigations. In a number of regions investigations have been carried out by site owners, principally to establish appropriate remedial works. Although the results of these investigations have not been available for this study, Tonkin & Taylor and SPHERE were aware of many of the results, and this information, combined with that derived from investigations carried out by regional councils/unitary authorities, formed the basis of this study.

In October 2000, each local government authority provided Occupational Safety and Health with information on timber treatment sites in their region where it was believed that PCP or NaPCP had been used. Relevant staff at regional councils and unitary authorities were then contacted at the beginning of this project to provide existing information on sites believed to have used PCP or NaPCP, estimates of the quantities used, results of investigations for PCP or dioxin contamination and any other site information. The information varied from simple lists of sites, to comprehensive reports detailing site investigations. Reports for some regions appeared to be incomplete. Only a few councils provided an estimate of quantities of PCP used on the sites. Council staff were supplied with regular updates of the database as information was gathered from other sources, and provided clarification/confirmation or additional input.

This information was supplemented by reviewing the former Timber Preservation Authority files, which contain data on PCP as a preservative use and NaPCP as a fungicide in boron diffusion processes (during the period when recording the fungicide was required), and some references to NaPCP use for antisapstain.

Based on the available information, the number and geographical distribution of the 255 sites where PCP and/or NaPCP is believed to have been used is as follows. (For a regional breakdown of the results of earlier PCP and dioxin investigations, see Appendix A.)

Although no complete records were ever kept, it is estimated that commencing around 1950 and over the next 40 years, about 5,500 tonnes of PCP were imported into New Zealand for use in the timber industry (Ministry for the Environment, 2000). This present study indicates the use may have been approximately 6,000 tonnes and therefore accords reasonably well with the previous estimate. At the height of its use during the 1970s, approximately 200 tonnes per year were used for antisapstain treatment (Shaw, 1990) and 100 tonnes per year for preservative treatment (Bingham, 1992).

Table 3.1 Number and geographical distribution of sites where NaPCP and/or PCP are believed to have been used

The available information has been used to classify the sites based on the quantity of PCP estimated to have been used over the duration of its use on a particular site. This gave the following figures (Table 3.2).

Table 3.2 Estimated number of sites where NaPCP and/or PCP were used, by size

Although these estimates are similar to earlier estimates of site numbers (Ministry for the Environment, 2000) the following points need to be emphasised.

• The classification into small, medium, large and very large users is based on the best information available; in most cases definitive quantities of PCP use are not available.

• There is a significant degree of uncertainty associated with classifying some sites as small or medium users.

• There may be some small and medium sites included that should not be there, and some sites missing that should be included.

• The very large and large users are probably well identified and none have been overlooked.

• The distribution of sites within the various use categories is consistent with survey returns from the timber industry.

• Information was not available for approximately 40% of the sites. The lack of information is considered to indicate that these sites were small PCP users and/or operated for only a brief period. These sites have been allocated to the medium and small categories in the same ratio as sites for which information is available.

3.3 Contamination investigation design

Two of the tasks for this study were to obtain more definitive and representative information on site contamination and to extrapolate from the results to a national figure for the dioxin reservoir at sawmill and timber treatment sites. The previous estimated reservoir from antisapstain treatment reported in the dioxin inventory (80 g I-TEQ, Ministry for the Environment, 2000) was based on data from one very large user and preliminary data from three small users. The previous calculation assumed a linear relationship between these two data sets and derived values for medium- and large-scale users. Therefore a primary objective of this study was to obtain information on medium- and large-scale users and to supplement that already available on small users (recognising that the majority of the sites in the country were small users).

Within the available budget it was considered that approximately 12 sites could be investigated and that there should be a minimum of three sites per use category. It was also necessary to ensure that there was a good spread of sites investigated throughout the country so that a range of soil types and climatic conditions was covered. A general approach to the industry was made through the New Zealand Timber Industry Federation, with follow-up communication on a one-on-one basis. A key consideration in companies offering sites for investigation was an assurance of anonymity. Seventeen sites (6.7% of the total) were offered, with the following geographical spread. Based on knowledge of the industry, Tonkin & Taylor Ltd and SPHERE consider the sites offered were representative of the levels of contamination most likely to be encountered.

Table 3.3 Geographical spread of the study sites offered

Of these sites, seven were classified as small users, seven medium and three large.³ Although this did not match the ideal distribution, it was decided that all should be investigated. In practice, on one small site the relevant areas had been extensively modified (and concreted) so that sampling was not possible, and on a second only limited sampling was possible. The one PCP result for this particular site was low, and consequently no dioxin analyses were undertaken.

At the outset it was agreed that individual site contamination information would be treated in confidence. Individual sites are therefore not described in the study.

For each site it was considered that the minimum number of samples to provide a realistic measure of the dioxin contamination on a site was 10 (two depths at five separate locations). The study allowed for 10% of the samples to be subject to a full dioxin congener analysis and the remaining 90% an OCDD screen analysis.⁴ Soon after the start of site investigations it was decided to also analyse the samples for PCP content to determine if a PCP/dioxin relationship could be established to guide future investigation/remediation work.

3.4 Sampling methodology

At each former NaPCP treatment facility the following was undertaken before collecting the soil samples.

• A meeting was arranged with the site contact and a site-specific health and safety induction was undertaken, as required.

• Long-serving staff members were interviewed (if available) regarding the use of NaPCP on the site, the method of application, the approximate volumes, and whether access to the locations where NaPCP was used was possible or practical.

• A walkover inspection of the general facility was made to develop and record an understanding of the environmental context (including site paving, drainage, etc.).

• A walkover inspection of target NaPCP use areas was made with site personnel to identify possible sampling locations and sampling methodology issues with each location (given that the sampling method was to be manual).

Samples were then collected from each selected sample location (as described below). Soil sampling was carried out in accordance with the Health and Environmental Guidelines for Selected Timber Treatment Chemicals (Ministry for the Environment and Ministry of Health, 1997). The sampling protocols detailed in Section 3-8 of the Guidelines were followed.

Samples were collected from spade-excavated pits or hand-augered boreholes at the near surface (0.1 m depth), 0.3 m and, when possible, 0.5 m depth. The samples were taken from approximately 50 mm above and below the specified depth. The soils were collected into glass sample jars, placed in iced chilly bins and transported to the AgriQuality laboratory in Lower Hutt under chain of custody documentation. AgriQuality sub-divided the samples and transferred representative sub-samples to Hill Laboratories (also in iced chilly bins with chain of custody documentation) for PCP analysis.

3.5 Analytical methodology

Dioxin analyses (full congener and OCDD screen) were carried out by the AgriQuality UltraTrace Laboratory using USEPA Method 1613B. (The ‘OCDD screen’ was based on USEPA Method 8290, as reproduced in Appendix J in CMPS&F, 1992. However, the ‘OCDD screen’ now used by AgriQuality is consistent with USEPA Method 1613.) The soils were spiked with a range of istopically labelled standards and extracted with ethanol/toluene, followed by solvent evaporation and exchange into hexane. The hexane extract was purified by column chromatography, reduced in volume and ultimately analysed by high-resolution gas chromatography/mass spectrometry.

PCP (and TCP) analyses were undertaken by Hill Laboratories using an IANZ accredited in-house test. The soils were extracted with acetone / hexane followed by gas chromatography – electron capture GC-ECD analysis.