View all publications

Appendix 8: Comparing Different Sheep-dip Sampling Strategies

The following section summarises the outcomes of a study that was carried out by a team comprising HortResearch, Waikato University, Waikato Pesticide Awareness Committee (WaiPAC) and Environment Waikato, including the research of two Masters students. Three different sampling strategies were compared at four historical sheep-dip sites to determine which sampling approach is the most effective at detecting contaminant hot spots and contamination distribution.

Judgemental sampling

This strategy includes a visual assessment of the site to determine the most likely areas of contamination. This process was carried out with due consideration of the type of dip structure, the location of the entry point and drainage pens, likely position of entry and exit paths to each dip, and likely route(s) for off-site migration of dip contamination. Ten sample locations were selected judgementally at each site and samples were taken to a depth of 7.5 cm using a foot corer. To accommodate quality control of the sampling and subsequent analytical procedures, one blind replicate sample was taken from two single-sample locations at each of the four sheep-dip sites. The samples were analysed for the chemicals of concern (see Table 1) at an accredited laboratory for trace metal residues and organochlorine pesticide residues. Preliminary sampling was used to determine which contaminants were present and to provide an indication of the range of potential contaminant concentrations.

Judgemental sampling proved to be a good method for assessing the contaminants present at each site. It generally provided a reasonable indication of the contaminants present and their concentrations. However, this type of sampling relies on the experience of personnel, and by its very nature, contains inherent bias due to the way sample locations are selected. The results of the study show that it is unlikely to detect contamination that has migrated some distance from the immediate dip site. If information is available on the use of chemicals at a dip site, an experienced operator can collect judgemental samples to confirm which contaminants are present and provide a reasonable indication of contaminant levels. Judgemental sampling is often the least expensive sampling regime because it generally involves taking fewer samples.

Statistically designed systematic sampling

A common systematic sampling strategy is to place a sampling grid over the primary sheep-dip area. Contamination at historical sheep-dip sites is generally localised to the immediate dip area. An off-site zone can be included to assess the extent of offsite movement of dip contaminants (eg, along a sloping profile away from the dip structure).

A statistical software programme was used to help determine the grid spacing and number of samples to be taken to detect the presence of a single hot spot of a specified size and shape, with a specified probability of missing the hot spot. Critical input parameters for this process included:

  • the shape of the grid (eg, triangle, square or rectangle)
  • the size and shape of the hot spot (eg, circle, ellipse or long ellipse)
  • the acceptable probability of missing the hot spot (eg, 10 percent, 20 percent, etc)
  • the size of the area to be sampled (eg, 100 m2, 2 m2, etc).

The systematic, statistically designed grid-sampling exercise was completed at one site only, to determine whether this would provide a fuller picture of the distribution and range of contamination. Samples were taken from each grid square with a stainless steel foot corer using a five-point dice pattern. This provided five individual sub-samples from each grid square, which were combined to form a single composite sample for each individual grid square.

Results from systematic sampling provided high-quality data, and the spatial extent of contamination was best assessed with this approach. It also detected significant contaminated areas that were not identified by either judgemental or sniffer dog sampling. However, due to the high number of samples taken (137 samples over a 378 m2 area), the cost associated with systematic sampling is high when using a small-diametre hot spot and high probability factor. If a smaller confidence interval and larger grid spacing are acceptable, sampling costs are significantly reduced and a larger area can be assessed.

The systematic sampling strategy employing data quality objectives proved to be the most effective method for characterising contamination at historical sheep-dip sites and provides current best practice for assessing contamination at sheep-dip sites.

Sniffer dog sampling

Dogs have been used for many years to detect trace odours at levels below the human sensitivity limit (eg, in border control operations to detect food products and narcotic drugs in the luggage of travellers). The two Australian sniffer dogs used in this trial have been trained to be sensitive to about 0.5 pbb of organochlorines (including aldrin, dieldrin and DDT isomers), all of which emit a characteristic odour. They are not able to detect arsenic contamination in soil, but sheep dips that were operational in New Zealand over the 1950s and 1960s are invariably co-contaminated with arsenic and organochlorine pesticides and so the presence of organochlorines provides a suitable tracer for corresponding arsenic contamination. This strategy has been used successfully in Australia to rapidly detect organochlorine pesticide contamination on farm properties where sheep and/or cattle dipping was either carried out, or suspected of being carried out.

The New Zealand study showed, however, that the use of sniffer dogs was not sufficiently effective as a tool for identifying areas of contamination at the dip sites. While many of the samples identified by the dogs provided measurable levels of organochlorine pesticide residues, they were generally at lower concentrations than those obtained by the preliminary judgemental sampling exercise. More importantly, the sniffer dogs did not detect previously identified hot spots of contamination where dieldrin was measured in excess of 100 mg/kg.

The sniffer dog handler/trainer believed that with further training under New Zealand conditions the dogs could adapt and learn to distinguish areas of higher organochlorine pesticide contamination at historical sheep dips. While this is an interesting strategy to provide a quick and cost-effective means of characterising contamination at sheep-dip sites, at the moment it is not suitable to detect organochlorine pesticide residues in New Zealand.

On-site characterisation methods/technologies

The costs associated with contaminated site assessment can be significantly reduced by using on-site characterisation technologies. Field analytical and site characterisation techniques offer the advantage of rapidly establishing the boundaries of contamination, providing targeted sampling and chemical analysis with significant savings in time and costs. Some of these technologies have the added advantage of on-site chemical analysis, providing almost real-time measurement of contaminants. In these situations, the costs of site assessment are largely associated with manual field sampling.

Many factors affect the technical feasibility and cost of field analytical and site characterisation technologies. These include physical constraints, site layout, data quality requirements, matrix interferences, and the expected level of contamination at a site (US EPA 1997).

There are a limited number of chemical measurement technologies suitable for on-site measurement of contaminants. These currently include the following.

  • An immunoassay is a biochemical test that measures the level of bodily reaction to a foreign object in order to detect the presence of certain substances in a sample. Immunoassays can be divided into two groups: enzyme immunoassay (EIA) and radioimmunoassay (RIA). The former is also called enzyme-linked immunosorbent assay (ELISA) and utilises antibodies specific to the substance; these antibodies are linked to an enzyme which causes a chromogenic or fluorogenic substrate to produce a signal. The RIA test, in contrast, uses a radioisotope, which is bound to the antibody or antigen.
  • A portable gas chromatograph is a chemical analysis instrument for identifying chemicals in a sample. A gas chromatograph uses a thin capillary fibre, known as the column, through which different chemicals pass at different rates depending on various chemical and physical properties. When the chemicals exit the end of the column, they are detected and identified electronically. The function of the column is to separate and concentrate different components in order to maximise the detection signal.
  • Infrared (IR) spectroscopy is a type of absorption spectroscopy that uses the infrared portion of the electromagnetic spectrum to investigate the composition of a sample. IR spectroscopy works because chemical bonds vibrate at specific frequencies. In order to measure a sample, a beam of monochromatic infrared light is passed through the sample, and the amount of energy absorbed is recorded. By repeating this operation, a chart can be built up, and an experienced user can identify the substance from the information on the chart. Fourier transform spectrometers are common laboratory instruments used for spectroscopy in many diverse disciplines.
  • One of the technologies currently best suited for on-site field assessments of historically contaminated sheep dips in New Zealand is X-ray fluorescence (XRF). In this technique, a material under investigation is exposed to X-rays. These photons with a relatively high energy are capable of exciting (ejecting) the electrons in the core levels of the material. The induced excited state relaxes under emission of an X-ray photon with a smaller energy. This results in emitted light, which is analysed in a spectrometer. Because the core levels have very different energies for different elements, the XRF spectrum contains information on the elemental composition of the sample under investigation. XRF qualitatively and quantitatively measures metals and can be optimised to selectively detect targeted metalloid contaminants, including arsenic residues in soil.

XRF can greatly reduce sampling and analysis costs for indicating the extent of contamination. It is important to note, however, that XRF is restricted to elements, so it does not analyse for dieldrin. Soil samples can be analysed with a good deal of accuracy, quickly and on-site. The XRF technique can be influenced by a number of factors, including the sensitivity of the instrument model, sieving, and rapid drying of soil samples as opposed to using field-moist cores.

Environment Waikato funded AgResearch to conduct a field investigation at one dip site of the previous study using a field-portable X-ray fluorescence (XRF) instrument. The purpose of this investigation was to determine whether in situ XRF could be an effective technique for determining arsenic concentrations in soil at historical sheep-dip sites.

The comparison with previously obtained results from that dip site by traditional analysis proved problematic due to differences in the sampling procedures. However, XRF proved to be a viable method for relatively rapid on-site determination of arsenic concentrations at historical sheep-dip sites on field-moist soils. The study concluded that:

Particle size created small differences between samples, but did not appear to cause substantial differences. High concentrations of lead in soils can create errors in arsenic concentrations. However, there were low concentrations of lead at the study site. XRF can greatly reduce sampling and analysis costs for indicating the extent of contamination. Samples can be determined accurately, quickly and on site. The XRF can be used on site in conjunction with a car battery and an inverter. (Dewar and Rajendram 2005, p 12)

However, arsenic is generally not the only contaminant present at historical sheep-dip sites. Although arsenic can be used as a tracer contaminant to find the extent of contamination, further soil sampling and analyses are recommended to determine other contaminants present at the sheep-dip site. For validation of a remediated site, a council would still be required to process the samples through an IANZ accredited laboratory.

Although on-site XRF assessment may not be as accurate as traditional chemical analysis, such as acid digestions, the speed and lack of sample preparation when using field-moist samples is a significant advantage. To determine the full extent of contamination, further soil sampling and analyses are recommended to determine other contaminants present at the site.