4 Contamination investigation results

4.1 Dioxin data from this study

The dioxin data from the OCDD screen results for this study are summarised in Table 4.1 for the various scales of NaPCP use and use areas. (Note that more detailed results of this investigation (including PCP analyses) are shown in Appendix B.) The data are for WHO (1998) TEQ values only. It should be noted that, based on samples for which a full congener analysis was undertaken, the calculated values of WHO (1998) TEQ were approximately 80% of the corresponding I-TEQ values. The reduction occurs because the toxicity rating of the OCDD congener is an order of magnitude less in the WHO (1998) scheme.

As can be seen in Table 4.1, there is a wide range of concentrations found between sites (and often within sites). Therefore, when considering particular use areas, the data do not fit any regular distribution and caution is needed when using the data. Although they have limitations, the mean values given in the tables are the arithmetic means, except for situations where the arithmetic mean is distorted by one result. In these instances the value in the table is the average of the arithmetic and geometric means.

Table 4.1 Mean WHO (1998) TEQ (ng/g) concentrations in soil for various locations/ scale of use

View mean WHO (1998) TEQ (ng/g) concentrations in soil for various locations/ scale of use (large table).

The results in Table 4.1 are summarised in Table 4.3 and for risk assessment purposes can be compared with the interim soil acceptance criteria shown in Table 4.2 (see section 5 of this report).

Table 4.2 Interim soil acceptance criteria

Notes on Table 4.2:

1. I-TEQ criteria = dioxin soil acceptance criteria calculated in accordance with I-TEQ scheme and published as ‘interim criteria’ in Health and Environmental Guidelines for Selected Timber Treatment Chemicals, Ministry for the Environment and Ministry of Health, 1997. These criteria were first adopted for the study of the Waipa site (CMPS&F 1992).

2. WHO (1998) TEQ criteria = dioxin soil acceptance criteria calculated by adjusting the I-TEQ criteria in accordance with WHO 1998 TEQ scheme as per note 3 below.

3. The main difference between the two schemes is that the OCDD congener under the WHO (1998) TEQ scheme is assigned a TEF that is 1/10th that under the I-TEQ scheme. As the WHO (1998) TEQ is estimated to be 80% that of the I-TEQ, the equivalent WHO (1998) TEQ soil criterion is calculated as follows: eg, for agricultural landuse the WHO (1998) TEQ soil criterion is 0.01/0.80 = 0.012 µg/kg TEQ. The equivalent soil criterion in this table is of interest only to the New Zealand situation involving sawmills where a range of technical grade PCP formulations were used over time. As noted earlier (p11), these calculations do not take into account the WHO TEFs amended in 2005.

4.2 Discussion of results

Small PCP users

Data were collected from five sites classified as small-scale users. Although there is considerable variability in the results, they are all very low. At one site, soil from a former drum storage area was still accessible for sampling. Although mean dioxin concentrations are similar to the residential criterion, the concentrations at a few locations are above this, but well below any of the industrial criteria.

Medium PCP users

Data were collected from seven sites classified as medium users. At one of these sites antisapstain dipping had taken place at two locations. Again there is considerable variation in the results, both within and between sites. The one very high dioxin value (beneath the sorting table at site 10) is associated with a very high PCP concentration and is believed to be correct. The mean values are above the residential criterion and similar to the criterion for an unpaved industrial site or for maintenance work. The concentrations at some locations only meet the criteria for a paved industrial site with a management plan.

Large PCP users

Data were collected from three sites where former use areas were still accessible. There is uncertainty about the one high TEQ result (site 7 antisapstain dip tank), because there is a relatively low PCP concentration in this sample and the full congener analysis gave a considerably lower TEQ result. The mean values are above the residential criterion and similar to the criterion for an unpaved industrial site or for maintenance work. The concentrations at some locations only meet the criteria for a paved industrial site with a management plan.

Overall

In general, the lower the use category, the lower the level of contamination. Dioxin contamination is much lower for small-use sites compared with medium- and large-use sites, and for medium- and large-use sites the average dioxin concentration across all sites and depths is roughly comparable. However, the range of contamination concentrations is much greater at the large-use sites.

In general, both PCP and dioxin concentrations decrease with depth below the ground surface. It is well documented that dioxins, and particularly the more highly chlorinated heptachloro and octachloro congeners, are significantly less mobile in soils than PCP. The results are generally consistent with this, with a more rapid decrease (particularly for the lower-use categories) in dioxin concentrations with depth compared with PCP concentrations. Although there is often a correlation between PCP and dioxin contamination levels at a particular depth (see discussion below), this is not always the case.

Where a full congener analysis has been undertaken, the hepta and octa congeners make up approximately 95% by weight, 56% of the I-TEQ values and 46% of the WHO (1998) TEQ values.

PCP–dioxin relationship

The results of PCP and TEQ (OCDD screen) analyses have been compared (Figures 3.1 and 3.2), and these show a reasonable correlation between the concentrations of the two contaminants at the two different depths. This indicates that, in most instances, PCP could be used as an indicator of likely dioxin contamination. However, there are a number of results with poor correlation between PCP and dioxin levels (and a significant change in the relationship of contamination with depth, due to the different mobility of PCP and dioxins in soil), so it will be necessary to perform some dioxin analyses at both the contamination characterisation stage and for validation of remedial work where this is undertaken.

Figure 3.1 Logarithmic plot of soil PCP vs I-TEQ and WHO (1998) TEQ upper depth

18-09-2013 10-21-26 a.m..jpg

• Edit
• Delete

Figure 3.2 Logarithmic plot of soil PCP vs I-TEQ and WHO (1998) TEQ, lower depth

18-09-2013 10-22-45 a.m..jpg

• Edit
• Delete

4.3 Revised estimate of the dioxin reservoir

Antisapstain use of NaPCP

The above dioxin data, summarised in Table 4.3, have been used to calculate the contaminant burden in various areas at the different scale-of-use sites. The areas and depths, estimated to be affected at the sites are also provided in Table 4.3.

Based on a count of 220 small-scale users, 28 medium-scale users, six large-scale users and one very large-scale user, the total dioxin reservoir in soils at sawmill sites from antisapstain use of NaPCP is estimated to be in the range 80-250 g WHO (1998) TEQ. This is summarised in Table 4.4 and may be compared with the previous estimate from the existing database of 80 g I-TEQ.

The figures given in Table 4.4 may be an overestimate, because the compiled information indicates that remedial work (excavation and storage/disposal) has taken place on a number of sites (estimated to be up to 20% of the total). In nearly all cases the remediation has been directed at PCP clean-up, but will have resulted in significant dioxin removal. However, the study has not generally investigated low-level dioxin contamination that may exist in areas where treated timber has been stored, nor waste disposal sites. Therefore the estimated reservoir total is considered to be reasonable.

Table 4.3 Parameters used to estimate the total TEQ reservoir from NaPCP use at sawmill sites

* An arbitrary value, as no former boron dip areas were available for sampling at the large sites.

Table 4.4 Estimated dioxin reservoir in soils from NaPCP antisapstain use

Notes:

The number of chemical stores is unknown, so the value in the table is an arbitrary estimate, as is the value for the boric dip for large sites. A mean soil bulk density of 1.2 tonne m^-3 has been assumed.

Preservative use of PCP in diesel

PCP use as a preservative in diesel oil was undertaken at four sites in New Zealand: the Waipa mill (where antisapstain treatment also occurred), Hanmer Springs, Christchurch and Waikoau (which was a comparatively small user). No information is available on the last of these, and only a limited amount of PCP contamination information is available for the site in Christchurch, which indicated a very low level of contamination. Therefore the estimate of the dioxin reservoir is based on the previous estimate for Waipa and Hanmer Springs, adjusted for WHO (1998) TEQ rather than the previous I-TEQ. The estimate is detailed in Table 4.5.

Table 4.5 Dioxin reservoir in soils from PCP preservative treatment (1 site each)

There has been no further investigation at the very large-scale user site and the large PCP-in-oil site, and the level of confidence for these sites remains high.

The level of confidence in the contamination characterisation for NaPCP use is medium to high. Data for the very large user are good, because there has been extensive investigation. The level of confidence for large users is medium because although the majority of sites with large-scale use have been investigated, no data are available for certain use areas, either because they are no longer in existence or are not available for sampling. There is a good data set for about 25% of the medium-scale use sites and similarly a good data set for small-scale use sites.

The level of confidence in the reservoir estimate for NaPCP use is medium, with decreasing confidence in the values for decreasing scale of use. The quantity of NaPCP use at many of the medium- and small-scale use sites is uncertain. The soil dioxin concentrations and areas of contamination are based on a limited amount of site-specific data, which are then extrapolated over the whole sector.