An understanding of the environmental setting of the site is necessary for developing the CSM (refer section 4).
In general, the search radius should take into account the distance that contaminants could migrate to or from the site. A search radius of 500 m from the boundary of the site is suggested as a general guide for identification of potential ecological receptors such as surface water bodies, wetlands and areas of ecological significance.
If the site is located in low-lying land, consideration should also be given to whether the site is likely to be located in/affected by acid sulfate soils. Where there is the potential for acid sulfate soils to be present, this should be taken into account when preparing the sampling and analysis quality plan as appropriate procedures are required. WA DEC (2009) provides detailed information on the identification and sampling of acid sulfate soils.
The local and site-specific geological and hydrogeological settings influence the fate and transport of potential contaminants in the vicinity of and at the subject site.
The distribution of contaminants across a site is influenced by the local geology and natural or man-made/altered drainage features in the area or at the site. Their distribution within the sub-surface is influenced by geological structures, variations in the permeability of soil and rock (which may result in perched water tables), geochemical, biological and mineralogical variations and the presence of preferential pathways such as loose fill around services.
Certain sites may be located in areas that are naturally enriched with mineral resources and can appear to contain elevated levels of metals and metalloids in soil, surface water or groundwater. Consequently, it is essential to have an understanding of the background quality of these media and to evaluate potential contamination of this type of site in terms of the beneficial uses of the site and its water resources.
The geological/hydrogeological component of the desktop investigation may include review of the following types of published data:
- surface elevation
- regional and site-specific soil and geological records
- geophysical data
- drilling logs which clearly identify imported and locally derived fill (including refuse) and natural strata
- well logs including strata, casing or construction details, and water level, quality and pump/discharge rate information
- aquifer types (unconfined, semi-confined, confined) and aquitards/aquicludes present
- direction and rate of groundwater flow
- regional and site-specific hydrogeological information, including groundwater quality
- current usage/resource potential
- existing monitoring wells and records of registered production wells or survey of surrounding landholders to determine the existence of wells where the resource may potentially be used in the vicinity of the site.
For more comprehensive assessments, for example where groundwater fate and transport modelling is to be undertaken, desktop studies may also consider:
- values for soil bulk density and porosity
- aquifer storativity or storage
- soil organic matter content
- cation exchange capacity (CEC)
- soil pH and redox (Eh) potential measured in situ
- hydraulic and piezometric heads and hydraulic gradients
- hydraulic conductivity
- transmissivity
- other parameters as appropriate.
Appendix III of the Guidelines for groundwater protection in Australia (ARMCANZ & ANZECC 1995) gives helpful advice on hydrogeological desktop studies.
These data form the basis of an initial appraisal of the potential risk to a receptor. When the likelihood of an unacceptable groundwater impact is identified, Schedule B6 should be consulted.
A site inspection should be conducted by a professional who is suitably qualified and experienced in the assessment of contaminated sites. For further information on suitable qualifications and experience, refer to Schedule B9.
A comprehensive site inspection is a critical stage of the site assessment process. It validates anecdotal and historical information and can identify additional evidence of potential contamination.
The complexity and detail reported in a site inspection may vary depending on the level of historical information and anecdotal information relevant to the site and the complexity and detail of the site itself. The following features, among others, should be noted:
- current uses of the site and surrounding land
- disturbed, coloured or stained soil
- bare soil patches
- disturbed or distressed vegetation
- unusual odour
- quality of surface water
- sheens on water surfaces
- site topography and surface water drainage
- presence and type of groundwater bores on the site and adjacent landholdings
- condition of groundwater bore headworks
- measurement of groundwater (water table and/or piezometric) levels
- condition of buildings, concrete and bitumen floors and roads, etc.
- building construction (slab-on-ground or other, presence or absence of crawl spaces and basements)
- the means of heating (fuel type) and cooling buildings on the site
- presence or absence of bonded asbestos-containing materials (bonded ACM) on the ground surface
- presence of stockpiles, fill, containment areas, sumps, drains and waste disposal areas – operational and closed
- evidence of cut and fill activities
- presence of pits, ponds and lagoons
- presence and condition of chemical containers, holding tanks, bunds, etc.
- presence and condition of any underground storage tanks (USTs) and associated infrastructure
- underground structures that may be associated with sub-surface contamination
- condition of materials storage and handling facilities and any solid or liquid waste disposal areas
- any evidence of on-site spillage of dangerous goods and/or off-site migration.
For operating sites, an inventory of chemicals stored or used at the site and copies of Material Safety Data Sheets (MSDSs), dangerous good licences, operating licences, works approvals and notices, and results of environmental audits (e.g. audits conducted under ISO 14000) should be obtained where practicable.
A conceptual site model (CSM) is a representation of site-related information regarding contamination sources, receptors and exposure pathways between those sources and receptors. The development of a CSM is an essential part of all site assessments and provides the framework for identifying how the site became contaminated and how potential receptors may be exposed to contamination either in the present or the future.
Typically, the CSM should be presented in written format and illustrated with suitable graphics and flow diagrams. Example graphics can be found in Clements et al. (2009) and Davis et al. (2009a). An example CSM in the form of a flow diagram can be found in Schedule B4 (Figure 2).
The CSM can be a useful tool for informing discussions with stakeholders regarding the investigation and management of potential and known contamination impacts.
The complexity of the CSM should correspond to the scale and complexity of the known or potential contamination impacts.
Source: Clements et al. (2009); SA EPA (2009) and Davis et al. (2009a)
The development of a CSM is a dynamic process and it is important that all the information and data from each stage of an assessment are reviewed in an integrated manner (using a multiple-lines-of-evidence approach where appropriate) to refine the CSM and used to inform subsequent decisions on whether further investigation or management is necessary. Note changes to the CSM may also involve revision of the data quality objectives (DQOs)—see Section 5.
The initial CSM is constructed from the results of the PSI and is used to identify data gaps and inform a decision on whether detailed investigation is required. The CSM should be continually challenged and updated throughout the assessment process.
The sub-optimal performance of many remediation systems can be traced back to the failure to undertake adequate site characterisation and to fully integrate the information gained into the CSM. For large and complex sites, 3-D imaging (visualisation) software may be useful for displaying and interpreting the results of the investigations and to refine the CSM.
The CSM should identify complete and potential pathways between the known or potential source(s) and the receptor(s). Where the pathway between a source and a receptor is incomplete, the exposure to chemical substances via that pathway cannot occur but the potential for that pathway to be completed (for example, by abstraction of groundwater or a change in land use) should be considered in the assessment.
The essential elements of an initial CSM are:
- known and potential sources of contamination and contaminants of concern including the mechanism(s) of contamination (e.g. ‘top down’ spill or sub-surface release from corroded tank or pipe)
- potentially affected media (soil, sediment, groundwater, surface water, indoor and ambient air)
- human and ecological receptors
- potential and complete exposure pathways.
For the assessment of vapours, (refer Section 9.2.3) additional detail will be needed about preferential pathways for vapour migration and the design of buildings or planned buildings at the site — including the location of sub-surface utilities, foundation construction and condition, and ventilation and heating (Davis et al. 2009a).
Data gap identification and uncertainty assessment are key activities in developing and refining a CSM during site assessment. It is, therefore, important that the CSM addresses:
- how representative the available data is likely to be
- what the potential sources of variability and uncertainty are
- how important the identified gaps are to the objectives and reliability of the site assessment.
In developing the CSM, the assessor needs to distinguish between variability and uncertainty. Variability arises from true heterogeneity in the environment such as lateral variations in soil properties or lithology or changes in contaminant levels over time and space. Uncertainty represents lack of knowledge about factors, such as contaminant levels (which may be reduced with additional investigation).
The identification of data gaps should be carried out in a logical, structured manner, to facilitate the assessment of uncertainty and the significance of those data gaps to the assessment objectives. Subsequent investigative efforts should be focussed on addressing the critical data gaps in a manner that is proportional to the uncertainties identified and results in data which is representative of the assessment area.
A tool for assessing gaps and uncertainties in CSMs and assessing their level of significance can be found in Clements et al. (2009).
Further information about developing CSMs can be found in:
- API 2005, Collecting and interpreting soil gas samples from the vadose zone, API Publication no. 4741, American Petroleum Institute.
- ASTM E1689–95 (2008), Standard guide for developing conceptual site models for contaminated sites, ASTM International.
- ASTM E2531–06 (2006), Standard guide for development of conceptual site models and remediation strategies for light non-aqueous-phase liquids released to the subsurface, ASTM International.
- Clements et al. 2009, Characterisation of sites impacted by petroleum hydrocarbons: guideline document, CRC CARE Technical Report no. 11, CRC for Contamination Assessment and Remediation of the Environment, Adelaide, South Australia.
- Davis et al. 2009a, Field assessment of vapours, CRC CARE Technical Report no.13, CRC for Contamination Assessment and Remediation of the Environment, Adelaide, South Australia.
- EA 2000a, Guide to good practice for the development of conceptual models and the selection and application of mathematical models of contaminant transport processes in the subsurface, NC/99/38/3, Environment Agency, England and Wales.
- ITRC 2009, Evaluating natural source zone depletion at sites with LNAPL, LNAPL-1, LNAPL Team, Interstate Technology & Regulatory Council, Washington, DC.
- ITRC 2007a, Vapor intrusion pathway: a practical guideline, VI-1, ITRC Vapor Intrusion Team, Interstate Technology & Regulatory Council, Washington, DC.
- ITRC 2007b, ‘Vapor intrusion pathway: investigative approaches for typical scenarios’, a supplement to Vapor intrusion pathway: a practical guideline, Technical and regulatory guidance supplement prepared by the ITRC Vapor Intrusion Team, Interstate Technology & Regulatory Council, Washington, DC.
- NJDEP 2005b, Vapor intrusion guidance, New Jersey Department of Environmental Protection, (Available online at www.nj.gov/dep/srp/guidance/vaporintrusion/
vig.htm). - ODEQ 2010, Guidance for assessing and remediating vapor intrusion in buildings, Report no. 10-LQ-007, Oregon Department of Environmental Quality, Portland, USA.
- SA EPA 2009, Site contamination: guidelines for the assessment and remediation of groundwater contamination, Environment Protection Authority, Adelaide, South Australia.
It is recommended that a systematic planning process is used for defining the objectives of a site assessment and to develop a sampling plan for the collection and evaluation of representative data to achieve those objectives. Without systematic planning, the site assessment may be ambiguous or inconclusive, which may lead to additional sampling requirements, resulting in increased costs and project delays.
In its simplest form, the planning process should consider:
- the overall objective of the site assessment
- the decision(s) to be made on the basis of the site assessment findings
- the constraints on the assessment (financial, time and logistical) and
- the degree of flexibility to conduct follow-up investigations.
This project level information can then be used to identify the specific site information needed to address the assessment objectives. The next step is to develop a sampling and analysis quality plan (SAQP) to obtain the necessary representative data for the study area.
The US EPA seven-step Data Quality Objective (DQO) process is one example of a suitable systematic planning approach (US EPA 2000a, 2000b and 2006a). The DQO process is recommended when site contamination data is being relied on to make a risk-based decision as part of a detailed site investigation, though a simplified planning process may be appropriate for straightforward screening assessments.
The DQO process is applicable at both the project level (for example, is the site suitable for development?) and at the investigation level. Further information can be found in US EPA (2006a).
At the investigation level, DQOs are qualitative and quantitative statements, developed in the first six steps of the DQO process that define the purpose of the site assessment to be undertaken and the type, quantity and quality of data needed to inform decisions relating to the assessment of site contamination. In the seventh step of the DQO process, the SAQP is developed to generate data to meet the DQOs. The SAQP should document the criteria that a sample design should satisfy, including when, where and how to collect samples or measurements, acceptance (performance) criteria and the samples or measurements that should be collected.
The process includes development of the following:
- a statement of the DQOs
- the SAQP to achieve the DQOs
- procedures to follow if the data does not meet the specified DQOs.
The development of the DQOs should be guided by identification of critical data gaps in the CSM. The objectives for sampling may include:
- determining the nature and extent of contamination
- delineating the lateral and vertical extent of contamination
- developing an understanding of the geology and hydrogeology
- the identification of potential and actual contaminant migration routes
- determining whether relevant investigation and/or screening levels are exceeded
- determining whether further investigation or management is required.
Subsequent objectives may be to determine whether relevant investigation levels are exceeded and whether further action is required (additional investigation or management of some form). As understanding of the site will evolve over time, the iterative development of the CSM may also have implications for the DQOs and the SAQP. Data quality assessment (refer Sections 5.6 and 13.1) is also an important part of this iterative process.
A summary of the DQO process is included in Appendix B. More detailed information can be found in US EPA (2000a, 2000b and 2006a). ODEQ (2010) provides a detailed case study of the DQO process applied to a benzene and TCE spill.
A well-developed sampling and analysis quality plan (SAQP) has a critical role in ensuring that the data collected is representative and provides a robust basis for site assessment decisions. In order to meet this objective, an SAQP will generally include the following:
- site investigation objectives and a brief background including appropriate plans and diagrams
- a summary of the CSM
- a review of existing information indicating reliability and usability of any existing data (data gap analysis)
- DQOs including a quality assurance (QA) plan and details of quality control (QC) samples to be collected
- pre-mobilisation tasks (e.g. preparation of a site-specific health and safety plan)
- media to be sampled (soil, sediment, groundwater, vapour, NAPL (non-aqueous-phase liquids), biota, surface water, deposited dusts, indoor air, outdoor air)
- details of analytes and parameters to be monitored
- number, location (coordinates) and depth (elevation AHD) of sampling points
- frequency and pattern of sampling
- sampling methods and procedures
- field screening methods
- analysis methods
- the methods for analysing and interpreting field data obtained (for any dynamic or reactive sampling).
The scope and level of detail contained in the SAQP will vary according to the site-specific circumstances and the stage of the investigation.
Flexibility in the SAQP is advisable so that changes may be made during the course of the investigation in response to identified data gaps such as the specific location of sub-surface utilities (which can act as preferential pathways for volatile organic compounds and other gases or a physical hazard) or evidence of more widespread contamination than expected (for example, widespread distribution of contaminated fill).
Professional experience and judgement will be required to ensure that the SAQP contains adequate coverage (spatial and temporal) of all the relevant media to obtain representative samples capable of satisfying the DQOs. If the sampling pattern and density are adequate, a further increase in the density or frequency of sampling is unlikely to change the site-assessment outcomes.
Approaches and methods for assessing soils, groundwater, and vapours and gases are discussed throughout this Schedule.
Quality assurance (QA) and quality control (QC) are essential elements of the systematic planning process and should be documented in the SAQP. Field QA and QC procedures are discussed in this section. Laboratory QA and QC procedures are discussed in Schedule B3.
Further information is presented in Appendix C, including a QA and QC checklist.
Quality assurance involves all of the planned and systematic actions, procedures, checks and decisions undertaken to ensure the representativeness and integrity of samples collected for analysis, and the accuracy and reliability of the analytical results. In the field QA measures include:
- selection of appropriate sampling and preservation methods, sample containers and sample storage
- decontamination procedures such as cleaning of tools before sampling and between samples
- maintenance of the sample environment to minimise sample contamination and analyte losses
- delivery to the laboratory in good condition and within the timeframes required for the particular analytes.
Section 8 of the Standard guide to the investigation and sampling of sites with potentially contaminated soil (AS 4482.1-2005) provides a basis for developing a program of quality assurance. As many sites are small with limited sampling, the rate of blind replicates and split samples should be adjusted to an appropriate level to ensure sufficient quality assurance.
Quality control involves those parts of an investigation which serve to monitor and measure the effectiveness of the QA procedures by comparison with the relevant DQOs. In the field, this may include checking of sampling equipment cleanliness by keeping rinses for analysis, duplicate sampling and inclusion of ‘field blanks’ and ‘field spikes’.
Adequate QA is achieved when QC results demonstrate that agreed objectives such as freedom from contamination, method accuracy and precision can be reliably achieved. Selecting an appropriate level of QC is imperative to ensure that DQOs are met.
Standard AS 4482.1-2005 recommends the use of a variety of QC samples including blind replicate samples and rinsate blanks collected in the field which are sent to the primary laboratory to determine the precision of the field sampling and laboratory analytical program, and split samples (collected in the field) which should be submitted to the laboratory as two individual samples without any indication to the laboratory of their common source.
As a general rule, the level of QC required is that which adequately measures the effects of all possible influences upon sample integrity, accuracy and precision, and which is capable of predicting their variation with a high degree of confidence.
The integrity of all samples must be considered, particularly when dealing with VOCs and SVOCs. Reference should be made to Standards AS 4482.1-2005, and AS 4482.2-1999.
Weathering and biodegradation by soil microorganisms will result in a loss of volatile hydrocarbon components from the surface and near-surface of affected sites. An example situation would be an underground fuel storage site where the tanks have been removed and the excavation has been left exposed for several months. In these circumstances, collecting samples from sub-surface layers (at least 500 mm below the surface of the excavation) may provide a more accurate representation of contamination.
Samples should be placed in appropriate sample containers, preferably prepared by a laboratory, with gas-tight, non-absorptive seals, allowing no headspace, and kept cool, preferably with ice bricks or a refrigerated cooler, until arrival at the laboratory. Arrangements should be made to ensure delivery of chilled samples to the laboratory within the holding time of the specified analysis. Samples must remain preserved and be analysed within the time limitations that apply for the analyte and laboratory method. Additional information on sample integrity and appropriate procedures is available in Standard AS 4482.1-2005.
Site investigators must complete chain-of-custody documentation which details the following information:
- site identification
- the sampler
- nature of the sample
- collection time and date
- analyses to be performed
- sample preservation method
- departure time from site
- dispatch courier(s).
All parties in the chain (sampler, dispatcher, courier and laboratory) should complete the chain-of-custody documentation so that it gains the status of a valid record of sample transfer to the laboratory. An example of a chain-of-custody form can be found in NSW EPA (1994).
The assessment report should include a copy of the receiving laboratory’s advice with respect to:
- the condition in which the samples and chain-of-custody documentation were received and the container type
- cross-checking information on sample identification numbers and paperwork received
- confirmation of preservation method.
Analyte choice should be informed by the site history findings and data gaps identified in the development of the CSM and the DQO process. Depending on the available history, potentially contaminated fill may require a more extensive suite of analytes. The appearance and odour of soil and groundwater samples may influence the selection of analytes.
Appendix A provides a list of possible analytes by contaminant grouping. Specific information on the assessment of asbestos and dioxins can be found later in this Schedule.
Additional information on the selection of possible analytes is available in the Standard guide to the investigation and sampling of sites with potentially contaminated soil (AS 4482.1-2005), WA DoE (2004), and Turczynowicz (1991).
Checking the validity and usability of the data collected assists with ensuring that only representative and reliable data meeting the specified requirements is considered in the assessment. Activities include verification of sampling procedures, data verification and validation and determination of data usability. The principal assessment measures (also known as Data Quality Indicators or DQIs) are precision, accuracy or bias, representativeness, completeness and comparability.
Further information is provided in Section 13.1, Appendix C and US EPA (2006a).
The site assessor should exercise professional judgement to select and develop an appropriate sampling design, based on accurate and reliable site-specific information (as integrated in the CSM) as far as practicable to obtain sufficient representative data to address the DQOs. For example, if the objective is to establish whether a site is contaminated, a limited number of samples located in those areas most likely to be contaminated may be sufficient, however, a greater number of samples and effort would be required to delineate known contamination. An explanation of, and justification for, the sampling design selected should be provided in the assessment report.
Source: US EPA (2002)
There are two main categories of sampling design: judgemental and probability-based sampling programs. The advantages and disadvantages of judgemental and probability-based sampling are listed in Table 2.
Table 2. Advantages and disadvantages of probability-based and judgemental sampling
| Probability-based | Judgemental |
| - Designs are unbiased
- Provides ability to calculate uncertainty associated with estimates
- Provides reproducible results within uncertainty limits
- Provides ability to make statistical inferences
- Can handle decision error criteria
| - Can be less expensive than probabilistic designs
- Can be very efficient with a reliable and full site history
- Easy to implement
|
| - Random locations may be difficult to locate and implement on the ground
- An optimal design depends on an accurate CSM
| - Depends on expert knowledge
- Cannot reliably evaluate precision of estimates
- Depends on subjective judgement to interpret data relative to study objectives
- Designs are biased
|
Judgemental sampling designs involve selection of sampling locations based on expert knowledge or professional judgement. The value of judgemental sampling depends on the DQOs, the study size and scope, and the degree of professional judgement available to locate and interpret the data. When judgemental sampling is used in isolation, quantitative statements about the level of confidence in the results cannot be made.
Probability-based designs (such as random, systematic, grid, stratified, transect and composite sampling) apply statistical sampling theory and may involve random selection of sampling locations. An essential feature of this type of sampling is that each member of the population from which the sample is selected has a known probability of selection. When a probability-based design is used, quantitative conclusions (or statistical inferences) may be made about the sampled population from the analytical results. For example, the assessor may calculate a 95% upper confidence level (UCL) of the arithmetic mean for the parameter of interest, say, lead concentrations in soil. If comparing this with the relevant investigation levels, the assessor can state whether the data indicates that the concentration exceeds or is below the investigation levels with a certain level of confidence (in this example 95%). Expert judgement is then used to draw conclusions about the study area based on the results of the sample data. Data analysis is discussed further in Section 13.
In judgemental sampling, the selection of samples (number, location, timing, etc.) is based on knowledge of the site and professional judgement. Sampling is localised to known or potentially contaminated areas identified from knowledge of the site either from the site history or an earlier phase of site investigation. Judgemental sampling is commonly used to investigate sub-surface contamination issues in site assessment.
Although judgemental sampling can invalidate some statistical methods, particularly where the sampling size is small, alternative methods using non-parametric approaches can be used. Further information can be found in Gilbert (1987) and US EPA (2006b, 2007a). Judgemental sampling may be used in combination with other sampling designs to produce effective sampling for defensible decision-making.
In simple random sampling, the selection of samples (number, location, timing, etc.) is based on using random numbers, and all possible selections are equally likely. For example, a simple random sample of a set of drums containing soil for disposal can be taken by numbering all the drums and randomly selecting numbers from that list. Simple random sampling protects against bias (which may occur if sampling locations are subjective) providing that the sample size is not small (more than approximately 20 samples). Many commonly used statistical analysis methods assume that the data was obtained by using a simple random sampling design.
The method is most useful when the area of interest is relatively homogeneous and no major patterns or hotspots are expected. The main advantages of this design are:
- it provides statistically unbiased estimates of the mean and variability
- it is easy to understand and implement
- sample size calculations and data analysis are straightforward.
Information on implementing a simple random sampling approach may be found in US EPA (2006b).
As most site investigations deal with non-uniform distributions of contamination, simple random sampling is usually combined with a stratified approach.
In systematic and grid sampling, samples are taken at regularly spaced intervals over space or time. An initial location or time is chosen at random or based on a convenient site feature, and then the remaining sampling locations are defined so that all locations are at regular intervals over an area (e.g. grid intersections) or time (systematic). Examples of systematic grids include square, rectangular, triangular, herringbone and radial grids.
Systematic and grid sampling are used to search for hotspots and to infer means, percentiles or other parameters and are also useful for defining spatial patterns or trends over time. If the property/trend of interest is aligned with the grid, systematic/grid sampling has the potential to introduce bias (over or under representation) to the results.
Even though most contamination is not normally distributed, the data can often be transformed to be approximately normal. Also, if data sets are sufficiently large, statistical inferences can still be made, since in that case the sample mean is approximately normally distributed (Gilbert, 1987).
Where a reliable and full site history is available, judgemental sampling is generally preferred, however, grid sampling may be appropriate where there is an inadequate site history and there is reason to suspect contamination may be present or, there is a large area of contamination that requires characterisation. An example of the latter would be heterogeneous fill suspected or known to contain contaminated materials.
As grid spacing must be small to have a high probability of finding small hotspots, in practice professional judgement is used to locate areas of smaller grid size in areas most likely to contain hotspots and over areas where a higher degree of confidence is desirable. Information on implementing a systematic or grid sampling approach, including applications for soil and groundwater, can be found in US EPA (2006b).
Determining grid size/sampling density from mathematical formulae (for example, Appendix D of Standard AS 4482.1-2005) is not an acceptable approach without consideration of likely contaminant distribution and acceptable hotspot size.
In stratified sampling, the assessment area (generally the potentially contaminated area) is separated into non-overlapping sub-areas (or strata) which are known or expected to be more homogeneous than the whole assessment area. Different sampling patterns and densities may be used in the different sub-areas.
The strata may be chosen on the basis of spatial or temporal proximity, or on the basis of pre-existing knowledge (e.g. site history, soil type), or professional judgement. The main advantages of this design are:
- potential for achieving greater precision in estimates of the mean and variance where the measurement of interest is strongly correlated with the variable used to define the strata
- calculation of reliable estimates for subgroups of special interest.
Information on implementing stratified sampling approaches can be found in US EPA (2006b).
In transect sampling, the samples are collected along a vector (a line of specified bearing, commonly 90˚) across an assessment area. Transect sampling may be appropriate when specific spatial characteristics of the contamination are to be targeted, for example, where there is a predictable contaminant distribution downwind/downgradient from a point source of contamination.
In contrast to a discrete sample taken from a single location and analysed individually, a composite sample is taken by physically combining a number of subsamples, usually a maximum of four, into a single well-mixed sample for analysis. The subsamples should be preferably composited in the analytical laboratory.
Compositing can be cost-effective where the analysis costs are large relative to sampling costs. However, its use should be considered with caution because of the potential for individual high results to be masked by low results. Composite sampling is not recommended for site-specific health and ecological risk assessments. Its use is also dependent on there being no safety concerns or potential biases (for example, loss of volatile compounds) associated with the compositing process.
The SAQP should clearly state the qualifiers applying to selection of subsamples for a composite sample. Care should be taken to take the subsamples from the same soil horizon or stratum.
Composite sampling is not suitable for clay or fine-grained soils as subsamples are difficult to mix adequately. Consideration should be given to the moisture content of the soils to be sampled as subsamples are mixed without drying whereas laboratory results are reported in terms of dry weight.
Where non-volatile contaminants are present (for example, metals or heavy oils such as heating oils), composite sampling may be adopted as a cost-effective method for achieving low resolution data for screening purposes. Composites may also be useful in conjunction with other sampling designs or when the objective is to estimate the population mean and information on spatial or temporal variability is not needed (e.g. for characterisation of stockpiled materials).
Composite sampling is not suitable for the assessment of pH, volatile substances and semi-volatile substances such as OC/OP pesticides and lower molecular weight PAHs. A good understanding of the site history and the potential contaminants of concern are therefore a necessary precursor to adopting a composite sampling approach.
Where composite sampling has been used, the relevant assessment level should be divided by the number of subsamples in the composite and compared with the laboratory result. Further information may be found in AS 4482.1-2005 and SA EPA (2005).
Source: US EPA (2002)
The site should be subdivided into assessment areas based on the information collected in the preliminary site investigation (site history, local geology and hydrogeology and site conditions) and anticipated exposure areas (for example, size and location of proposed residential lots) and the sampling design selected according to the characteristics of the different sub-areas and the DQOs.
In general, when the source of contamination is known or is suspected to be limited to a specific area, sampling points are located relative to the suspected source(s) using judgemental sampling stepping out from the suspected source location, or systematic grid sampling centred on that location.
Specialised professional advice should be sought in developing sampling plans for rock soil mixtures at waste rock dumps, tailings dams, heap leach pads, and other artificial structures associated with mining site contamination.
Table 3 presents examples of example investigation scenarios that may be encountered and suggests sample designs that may be relevant. As indicated below, a more sophisticated sampling design may follow on from a preliminary (screening) investigation.
Table 3. Selecting an appropriate sampling design
If you are… | and there is… | consider using… | in order to… |
performing a relatively small scale screening investigation | limited budget and/or schedule | judgemental sampling | assess whether further investigation is warranted |
developing an understanding of where contamination is present | adequate budget for the number of samples needed | grid sampling | acquire coverage of the area of concern with a given level of confidence that a hotspot of a given size would be detected |
estimating a population mean | adequate budget for the number of samples needed | systematic or grid sampling | produce information on spatial or temporal patterns |
developing a detailed understanding of where contamination is present and/or estimating a population mean | spatial or temporal information on contaminant patterns | stratified sampling (includes judgemental and grid sampling) | increase the precision of the estimate in key areas of concern |
Adapted from US EPA 2002
The aims of an SAQP (refer Section 5.3) are to reduce the likelihood of under assessment (that could result in significant adverse effects from unidentified contamination) or over assessment (concluding that a site requires further investigation when in reality it does not) and to enable an appropriate level of remediation of contamination that is sufficient to protect human health and the environment.
The information presented in this section can be applied to both horizontal and vertical sampling.
Consideration of the CSM and DQOs should inform the requirements for sampling density and depth of sampling. The amount of sampling required will depend on an integrated appraisal of factors including:
- the size of contaminated areas to be detected
- the number of stages of sampling considered feasible
- the size of the site and final subdivided lots if the site is to be subdivided
- the distribution of uses on the site and the disposition of structures
- the site history (which may vary across the site).
When developing a sampling program, consideration should be given to numerous factors including, but not limited to:
- the likely heterogeneity of any surface fill and underlying geological units
- whether knowledge of background soil and groundwater quality is required
- the depth and thicknesses of soil/aquifer units
- soil properties that affect contamination migration (e.g. texture, moisture content, clay content)
- physical and chemical nature of the contaminant under investigation (e.g. solubility, volatility and density)
- the nature of the release (e.g. surface spill, leaking underground pipe, buried waste)
- the timing and duration of the release
- the amount of contaminant likely to have been released
- the possible effects of contaminant migration through the unsaturated zone and when and where the contaminant entered the saturated zone
- the effects of potential degradation processes
- the direction and rate of groundwater flow within each aquifer.
If a site is to be subdivided, the size of the subdivided lots should be taken into account when determining the sampling density. While predictions may be made on a ‘macro’ scale, residents or owners may seek information about their own particular area of land and the risks associated with this land, especially if the potential contamination on the original site was uneven in distribution and type.
The detection of hotspots is an important issue for sites to be used for residential purposes or other sensitive uses where children have regular access to soil or where there is potential groundwater contamination. A greater sampling density is usually required for these sites. The toxicity of the contaminant and the size and magnitude of the potential hotspot(s) needs to be considered in determining the sampling density.
The development of a suitably detailed CSM will inform decisions about the depth of sampling required. For health and ecological risk assessment, the soil strata to which people and other receptors could feasibly be exposed should be adequately sampled. This will result in a weighting towards near-surface sampling unless the history or the nature of the soil and the presence of groundwater suggests it should be otherwise. On residential sites, the maximum excavation depth (such as for a swimming pool) is unlikely to extend beyond three metres, but much deeper soil disturbance may occur on a commercial site.
If dealing with volatile contaminants such as light fraction petroleum hydrocarbons or chlorinated solvents, then vapour transport from depth and through a shallow soil zone may pose a risk. Deeper sampling to determine the nature and extent of the source of the vapours and the risk they represent may be required—refer Section 9.
The risk to groundwater needs to be assessed according to jurisdictional requirements, especially if receptors may be exposed by current or realistic future use of the groundwater resource—refer Schedule B6.
To delineate contamination laterally, typically samples should be taken until either no further contamination is detected or concentrations are below the relevant investigation levels or site-specific risk-based criteria.
The nature and appearance of drill cores will influence sampling at depth. It is essential that samples are taken from within a natural stratum or fill horizon and not across strata.
At the surface, samples at 0–100 mm or 0–150 mm should be taken unless there is evidence of a thin superficial layer of contamination. Where there is good evidence that contamination is restricted to a thin superficial layer, a shorter sampling interval may be appropriate, however, a subset of deeper samples should be analysed to inform/confirm the CSM. At greater depths, the sampled interval should be no more than 500 mm to avoid a compositing effect.
The selection of appropriate site investigation techniques depends on a number of factors including the stage of the investigation (for example, preliminary assessment or detailed delineation, the depth of investigation required, the contaminant type (volatile or non-volatile, bonded or unbonded asbestos-containing-material), the depth and nature of any fill, and whether an undisturbed sample is required.
The most commonly used investigation techniques are test pits, trenching and drilling of shallow boreholes. Samples from shallow depth are generally obtained from test pits and trenches or from augers. Samples from greater depths may be obtained by a range of drilling methods including direct push, hollow stem augers, split spoon, Shelby tube, mud rotary and sonic drilling. Methods capable of providing continuous or near-continuous soil cores, such as direct push, split spoon and sonic drilling, are preferred. Air drilling and solid flight augers provide highly disturbed samples and poor depth control which limits their value for site characterisation purposes. Further information on soil investigation methods can be found in Australian Standards AS 4482.1-2005, AS 4482.2-1999 and AS 1726-1993.
A number of screening tools are also available that can be used to rapidly and cost-effectively identify and delineate VOC and SVOC contamination in both the unsaturated (vadose) and saturated zones. These include soil vapour sampling, and the laser-induced fluorescence (LIF) and membrane interface probe (MIP) tools. LIF and MIP are real-time tools that can provide detailed logs of the sub-surface and can be used in a reactive or adaptive field sampling program, particularly for volatile substances where trial pitting and some coring methods are not as applicable.
Various geophysical techniques can be used for site characterisation purposes including determining depth to bedrock, delineation of groundwater contamination, location of voids, faults or fractures and the presence of buried items such as steel drums and tanks. The information gained can be used for selecting optimal locations for boreholes and test pits as well as to correlate geology between wells. The techniques available include metal detectors, magnetometers, electromagnetic conductivity surveys, electrical resistivity—or electrical impedance tomography—and ground-penetrating radar.
A detailed description of geophysical techniques is beyond the scope of this guideline, however, further information can be found in:
- ASTM D6432-99 (2005) Standard guide for the surface ground penetrating radar method
- ASTM D6429-99 (2006) Standard guide for selecting surface geophysical methods,
- ASTM D5753-05 (2010) Standard guide for planning and conducting borehole geophysical logging
- Clements, et al. 2009, Characterisation of sites impacted by petroleum hydrocarbons: guideline document, CRC CARE Technical Report no. 11, CRC for Contamination Assessment and Remediation of the Environment, Adelaide.
- NJDEP 2005a, Field Sampling Procedures Manual, New Jersey Department of Environmental Protection. (Available online at www.nj.gov/dep/srp/guidance/fspm).
Detailed information on site investigation techniques can be found on the US EPA CLU-IN characterisation and monitoring webpage at www.clu-in.org/characterization/. The advantages and disadvantages of various techniques applicable to petroleum hydrocarbons in soil, soil vapour and groundwater are presented in Clements et al. (2009). NJDEP (2005a) provides detailed field sampling procedures including for soil vapour surveys.
Test pits and trenches are generally excavated by hand using a shovel to shallow depths or by machine (backhoe or long-arm excavator) to greater depths. Soil samples may be collected from the walls of a test pit when they are shallow and it is safe to do so in accordance with the site health and safety plan. Only freshly exposed surfaces are suitable for sampling volatile and semi-volatile contaminants. Tests pits and trenches expose a large surface area for visual assessment of soil profiles and potential contamination and generally allow the investigator to gain a better appreciation of soil features and soil heterogeneity than that obtained with an individual borehole.
In general undisturbed samples obtained from near-continuous soil cores are preferred to grab samples for inspection and analysis. Intact soil coring is typically conducted by advancing a hollow rod or thin-walled metal tube into the sub-surface by direct push or other method such as sonic drilling. Direct push methods eliminate the need for a drilling fluid and avoid potential interferences from introduced fluids.
Auger and split-spoon samplers fitted with clear acetate sleeve liners may also be used to collect soil samples; however, sample quality is generally not as good as that obtained using direct push or sonic drilling methods.
Once soil cores have been obtained, samples from specific depth intervals can be taken and suitably preserved for laboratory analysis. Where an entire core is to be taken, the soil core tube should be quickly capped, labelled, wrapped and packed (and kept cool using ice bricks or refrigerated to keep the sample in a relatively undisturbed state) and dispatched to the laboratory for analysis.
Cone penetrometer testing (CPT) is an in situ form of direct push drilling where sensors are mounted in a cone at the tip of the drill rods. As the cone is advanced, the sensors measure the resistance of the soil to the force of the advancing cone and the data is relayed to an on-board computer which interprets the soil stratigraphy and other parameters. A range of additional sensors may be used with CPT for simultaneous measurement of multiple parameters. The range of available sensors includes pressure head transducers (allowing permeability and hydraulic conductivity assessment), conductivity probes (allowing soil types and saturation to be estimated) and nuclear and pH probes.
CPT is a useful tool for providing rapid, continuous profiles of sub-surface stratigraphy and can save considerable time and money, particularly at large sites with complex geology.
The membrane interface probe (MIP) tool consists of a heated probe equipped with a semi-permeable membrane mounted on a direct push or CPT drilling rig. VOCs diffuse across the membrane and enter a carrier gas within the probe. The carrier gas transports the contaminants to a gas chromatograph at the surface which can be equipped with various detectors for measurement of a wide range of VOCs: an electron capture detector (ECD) for chlorinated organics, a photo-ionisation detector (PID) for aromatic hydrocarbons, and a flame ionisation detector (FID) for straight-chained hydrocarbons.
For sites containing light non-aqueous phase liquids (LNAPL), MIP is typically used to locate and delineate dissolved-phase groundwater and soilvapour plumes, while laser induced fluorescence (LIF) (see below) is used to delineate the LNAPL source zone.
The MIP tool is usually equipped with an electrical conductivity sensor to interpret soil lithologies. The combination of sensors enables an increased understanding of contaminant distribution, particularly in heterogeneous lithologies.
One or more background MIP borings upgradient of each assessment area should be advanced in order to determine the background response. The MIP response can be used to determine concentrations of specific contaminants if it is calibrated with soil and groundwater samples from across the investigation area.
The MIP tool is typically used in the context of an adaptive sampling approach using a dynamic sampling plan and DQOs, that is, the investigation proceeds in a step-wise approach with the location and depth of each subsequent boring being determined in the field based on the results and interpretation of the preceding boreholes using a predetermined decision framework. QA/QC procedures should be developed for MIP surveys as part of the DQO process.
With multiple MIP locations and appropriate data interpolation and visualisation software, MIP data can enable a 3-D depiction of NAPL source zones in both the unsaturated (vadose) and saturated zones.
The laser-induced fluorescence (LIF) tool consists of an ultraviolet (UV) or visible wavelength laser connected to a sapphire window, mounted on the side of a direct push/CPT probe tip.
The LIF laser transmits light through the sapphire window, which is then absorbed by any PAHs in contact with the window causing the material to fluoresce at a characteristic wavelength. The fluorescence emission is recorded continuously by a detection system as the probe is advanced.
LIF tools are available which, depending on the wavelengths monitored, are capable of differentiating different types of product. UV LIF systems are appropriate for light fuels up to mid-range oils, but often fail to adequately respond to heavy fuel oil, heavy crudes, coal tars and creosotes. Visible wavelength systems detect heavy fuel oil, heavy crudes, coal tars, and creosotes but do not respond to light fuels such as petrol and kerosene. If possible, an appropriate NAPL sample should be tested to ensure the appropriate wavelength LIF is used.
One or more background LIF borings upgradient of each assessment area are recommended in order to determine the background LIF response. If NAPL is present at the site, a LIF borehole should be advanced adjacent to a well where NAPL has been measured to calibrate the LIF response to the specific NAPL contamination present at the site. The LIF data should also be validated with soil and groundwater sampling to determine concentrations of specific contaminants throughout the investigation area.
As for MIP, the LIF is combined with an electrical conductivity sensor to interpret lithology and is used in a similar reactive sampling approach. Similarly, with multiple LIF locations and the use of data interpolation and visualisation software, LIF data can enable a 3-D depiction of NAPL source areas in both the unsaturated (vadose) and saturated zones.
Source: NJDEP (2005a)
Soil vapour surveys may be used to screen sites for VOC and SVOC contamination source areas in the vadose zone and to delineate the extent of contamination. Soil vapour sampling, when applied appropriately, can be used as a screening procedure to assist in locating soil sampling and monitoring well locations.
There are two basic types of soil vapour surveys performed as part of site assessments. The first type is an active soil vapour survey where a volume of soil is pumped out of the vadose zone into a sample container or directly into an analyser. The second type is the passive soil vapour survey where a sorbent material is buried in the vadose zone so that contaminant vapours can be selectively absorbed over time using the ambient flow of vapours through the subsurface. The latter is particularly applicable to low permeability soils where active methods are less effective.
Further information on soil vapour sampling is presented in Section 9.
Ground penetrating radar is the most commonly used of the geophysical methods and is typically conducted by rolling a radar unit across the site in a grid pattern and recording and processing the data collected to provide a two-dimensional or three–dimensional image of the surveyed area. Metal objects or near–surface features (such as pipes or utilities) can cause noise on the measured signal; if the location of these features is known, their effect can be minimised in the data processing stage. In homogeneous soil profiles, ground penetrating radar surveys may assist in defining the lateral and vertical extent of NAPL plumes in shallow soil or groundwater—see Clements et al. (2009).
Accurate documentation and careful consideration of field observations is essential as this can greatly improve understanding of the variability of contaminant distribution across a site.
All boreholes (including groundwater monitoring wells) and test pits should be logged in accordance with AS 1726-1993 and the presence of strata, moisture, seeps or water-bearing zones, elevation of the water level/hydraulic head, imported fill and odorous or stained materials carefully noted. These logs are essential for interpretation of chemical data to establish the extent of contamination and to assist in the design of more detailed investigations. Example logs are included in Appendix D.
A photographic record that is well labelled for date, location and orientation is a valuable reference tool for documenting procedures and for understanding soil/aquifer heterogeneity and variability in laboratory results. Photographs are recommended to be taken of the strata present in test pits and soil cores and the appearance of split samples, particularly to illustrate visible heterogeneity in the field.
Field checklists to aid documentation of essential information are available for download from the EPHC website at www.ephc.gov.au.
A variety of field screening techniques may be used to provide immediate (real-time) information about the concentration and distribution of contaminants on contaminated sites. These tests, by their very nature, are less rigorous and reliable than analytical tests conducted in a laboratory, however, they provide cheaper and quicker results to guide the design of further sampling strategies for site assessment.
The most commonly used field tests include:
- gas detector tubes
- colorimetric test kits
- headspace testing using PIDs and FIDs
- field portable x-ray fluorescence spectrum analysers
- field gas chromatography
- immunoassay test kits.
These techniques can be used to gain a general understanding of the field conditions and the presence of possible contamination and may assist in the selection of samples for laboratory analysis. PID measurements, for example, may be useful as a field guide to indicate areas of volatile compounds. However, their role in providing real-time data needs to be augmented by laboratory chemical analysis.
Their use as the sole source of analytical data in the assessment of potentially contaminated sites is inappropriate as they may give falsely high or low results. For example, naphthalene is commonly reported in petroleum hydrocarbon-impacted soils and will evoke a response from a PID, in contrast to benzo(a)pyrene (a more significant PAH in terms of human health), which will not be detected by a PID. As these measurements do not always correlate well with laboratory results they are generally not suitable for validation sampling.
Prior to use of any field monitoring equipment there should be:
- a determination that they are capable of detecting relevant contaminants
- adequate understanding of the methods of use for the particular instrument, its limitations and site conditions that may affect the results
- appropriate calibration (and recording of the calibration data) for the substances being measured
- an appraisal of site conditions that may affect the results, e.g. high soil moisture may result in artificially high PID results for benzene.
Further information on field characterisation techniques may be found on the US EPA CLU-IN website at www.clu-in.org/characterization/.
Detector tubes have been developed that measure volatile gases including individual compounds, for example, hydrogen sulphide, or groups of compounds, for example, petroleum hydrocarbons. They can provide a direct measure of the analyte in ambient air or an indirect measurement of soil and groundwater contaminant concentration when used in field test kits for measurement of soil vapour and headspace for liquids. The reagents in the tubes may react with compounds of similar chemical properties; consequently, false positives and inaccurate results are possible and should be identified in the DQO process.
Colorimetric tests rely on the chemical reactions of indicator compounds with individual compounds or classes of compounds. Tests are generally performed by mixing reagents in specified amounts with the soil sample to be tested and comparing the resultant colour change with a colour chart or using a field colorimeter to determine concentration.
Colorimetric tests have been developed for a wide range of substances including BTEX, total PAHs, chlorinated hydrocarbons, PCBs and various individual pesticides and classes of pesticide. The detection limits in soil are generally in the low ppm range (lower detection limits are achievable in water as no extraction stage is necessary). Although these tests are relatively simple to perform, depending on the kit, they can suffer from interferences from other co-contaminants or naturally occurring materials or organic matter. Their usefulness for specific site-characterisation purposes can be evaluated by comparison of field colorimetric results with laboratory results over a range of analyte concentrations.
Field headspace testing is a commonly used method for screening soil samples for volatile and semi-volatile organic compounds. The procedure involves partially filling an airtight container with a fresh soil sample and then analysing the headspace vapour using an appropriately calibrated portable instrument, typically a PID or FID.
A FID uses a hydrogen flame to ionise the organic vapours whereas a PID uses an ultraviolet lamp to ionise the vapours. The instrument response is related to the electric current generated by the ionised compounds. FIDs are most sensitive to aliphatic hydrocarbons as these compounds burn more efficiently than aromatic compounds. PID instruments are most sensitive to aromatic hydrocarbons (for example, BTEX compounds) and can measure most VOCs in the range of C6 equivalent carbon atoms (for example, benzene) to C10 (for example, naphthalene). Neither instrument is effective for detecting non-volatile compounds such as highly weathered hydrocarbons. Care should be taken when using PIDs since a positive bias may result from water vapour or moist air and/or dust being drawn into the instrument. FIDs are not sensitive to water vapour.
A standardised field procedure for headspace testing should be followed and the details of the test method documented (size of jar, soil volume, equilibration time and ambient temperature) in the investigation report.
X-ray fluorescence (XRF) is a rapid screening tool that can be used to measure metal concentrations in soil. Performance is dependent on the metal, the soil matrix and soil moisture content. Although a range of heavy metals can be simultaneously detected, there are potential interferences that influence the method accuracy and precision. The US EPA has developed a methodology to guide XRF analysis (US EPA 2007b).
The advantages of XRF include real-time results, when used in scanning mode on surface soil, or near real-time results when deeper samples are collected and analysed in the field. The usefulness for specific site-characterisation purposes can be evaluated by comparison of results from split samples analysed by field XRF with laboratory results over a range of analyte concentrations.
Field gas chromatography (GC) may be used for the analysis of volatile and semi-volatile compounds in soil, soil vapour and water. The two main components of a GC are a column to separate the individual constituents and a detector (such as a PID or FID) to measure the signal response of the constituents. The analysis is compound-specific and potentially has the greatest accuracy of all the commonly used field analytical techniques.
Immunoassay test kits, using antibody-antigen reactions, can be used to measure petroleum hydrocarbons in soil and water. For most kits, the intensity of the colour development is inversely proportional to the amount of substance present. The concentration is determined by comparison with a reference standard or with a portable photometer.
An in situ soil sampling program informed by site history, inspection and contaminant form is the preferred approach for site assessment. On occasions it is necessary to stockpile soils that have not been assessed or only partially assessed in situ, and to devise a thorough stockpile sampling plan.
Excavation may result in mixing of low-level or uncontaminated soil with smaller quantities of contaminated soil, having the effect of diluting higher concentrations. It is preferable for assessors to supervise excavation and, as far as practicable, segregate stockpiles according to soil and contaminant types and to avoid dilution.
The process of excavating material often results in mixing of strata and different fill and soil types. Stockpiling may cause some segregation of grain sizes particularly on the exterior slopes. Specific grain sizes may contain the contaminant source and concentrate in some stockpile locations; for example, finer material may tend to accumulate at the toe of batters and coarser material towards the crest. Sticky clay material may be distributed into a different part of the stockpile than loose soils. The age and surface condition of the stockpile should be assessed, particularly if it has been weathered and subjected to leaching.
The composition of the stockpile should be documented by inspection of its external appearance and excavations into the stockpile by shovel (for small stockpiles) or excavator bucket where a shovel cannot reach the centre of the stockpile. The stockpile dimensions should be determined noting its regular or irregular shape and a 3-D plan prepared. The volume of material present should be estimated.
Table 4 below provides the minimum number of samples recommended for characterisation of stockpiles up to 200 m3 comprising similar materials. A greater number of samples may be required when there is a large range in contaminant concentrations or soil types. If only the minimum number of samples is collected and there is a large range in contaminant concentration, then either the maximum concentration should be assumed for disposal purposes or additional samples collected and analysed and the situation re-evaluated. In situ samples taken prior to excavation may be helpful for informing the decision on the number of samples required for adequate characterisation of stockpiles.
Table 4. Minimum number of samples recommended for initial assessment of stockpiles
Stockpile volume, (m3) | No. of samples |
<75 | 3 |
75 <100 | 4 |
100 <125 | 5 |
125 <150 | 6 |
150 <175 | 7 |
175 <200 | 8 |
The recommended sampling frequency (Table 4) applies to the characterisation of homogenous soils suspected of contamination. Lower sampling rates may be derived for soil quantities greater than 200 m3 by applying statistical analysis. Worked examples of applying 95% UCLave to characterise stockpiles are included in EPA Victoria (2010).
Jurisdictions may have specific requirements where materials are to be recycled, recovered and reused for beneficial purposes.
The stockpile should be sectioned into an appropriate distribution of sampling locations based on inspection, site history and other assessment data about the nature of contaminants present. If a section of the stockpile is known to have a higher level of heterogeneity and greater contamination risk and the balance of the stockpile is relatively homogenous with low-level contamination, sampling bias to the more contaminated section may be considered. If this information is not known, a uniform sample point distribution should be used. A plan should be developed of the stockpile sections and the corresponding sample locations that represent each section. This will allow physical separation of portions of the stockpile for further characterisation, if required, after receipt of the analytical results.
Collection of samples from the exterior 300 mm of the stockpile should be avoided due to the higher risk of weathering and grain size grading errors.
Samples for inorganic and non-volatile components should be taken at various depths towards the centre of the stockpile from 300 mm below the stockpile surface. Compositing may improve the reliability of samples for inorganic analysis. Composites should be based on equal quantities of material from 4 random locations and depths in the area of the stockpile allocated to the sample. The trowel should be cleaned after soil collection at each random location and the collected material thoroughly mixed on a clean surface, subsampled and preserved for chemical analysis.
Composites are not suitable for the assessment of pH, volatile substances and semi-volatile substances such as petroleum hydrocarbons, OC/OP pesticides and lower molecular weight PAHs. Samples for volatile and semi-volatile compounds should be taken without delay from a freshly excavated surface 500 mm or greater depth below the stockpile surface.
Systematic sampling directly from excavator buckets during the excavation and stockpile formation process or for appraisal of larger stockpiles using appropriate QA/QC processes is an acceptable strategy in site assessment. Further guidance on stockpile sampling may be obtained from EPA Victoria’s Industrial waste resource guidelines (2010).
Contaminants in soil can leach to groundwater under certain conditions. For inorganic substances, leachability is particularly affected by soil pH, contaminant solubility and redox (Eh) conditions.
The leachability characteristics of contaminated soil can be used to help assess:
- the impact of soluble soil contaminants on groundwater quality
- the impact of leaving contaminated soil materials on site.
Information on leachability tests applicable to the assessment of site contamination can be found in Sections 2.7 and 12 of Schedule B3.
Samples to be tested should be selected with reference to the CSM and be representative of the impacted materials. Analysis of appropriate background samples should be included for comparative purposes.
Methodologies are available which aim to predict the impact of leaching soil contaminants on groundwater quality and groundwater resources. An overview of the US EPA (1996) methodology is included in Schedule B5b.
Treatment and disposal of excavated contaminated soils should be in accordance with jurisdictional legislation or guidelines for re-use and/or disposal of contaminated soils.
The recommended risk–based approach to the assessment of groundwater contamination is outlined in Schedule B6.
The process involves a staged risk-based approach to delineation of contamination using guidelines such as the Australian and New Zealand guidelines for fresh and marine water quality (AWQG) (ANZECC & ARMCANZ 2000), the Australian drinking water guidelines (ADWG) (NHMRC 2011) and the Guidelines for managing risk in recreational water (GMRRW) (NHMRC 2008) as appropriate investigation and response levels. The process may include a detailed assessment of contaminant concentrations over time using fate and transport modelling to predict the current position and future movement of groundwater contaminants to assess potential risk to receptors.
This section deals with the basic requirements for groundwater investigation, including installation of monitoring wells, sampling of groundwater, presentation of data and delineation of groundwater contamination.
Fractured rock aquifers (for example, fractured basalts, bedrock aquifers and limestones) behave fundamentally differently from unconsolidated aquifers such as sands and gravels. Specialist advice should be sought from qualified contaminant hydrogeologists with experience in fractured rock aquifers. Further information on groundwater flow and groundwater sampling in fractured rock aquifers can be found in Cook (2003) and Nielsen (2006) and references therein. The US EPA CLU-IN website also provides useful information on characterising fractured rock aquifers and lists various related resources at: http://www.clu-in.org/contaminantfocus/default.focus/sec/Fractured_Rock/cat/Overview/.
Site assessors should be aware of (and comply with) relevant jurisdictional requirements such as groundwater protection policies and licensing requirements for the construction of monitoring bores and groundwater abstraction.
The collection and assessment of groundwater data and the selection and use of fate and transport models should be undertaken by appropriately qualified and experienced professionals. This is particularly important when applied to fractured and karstic rock environments.
There are several methods for collecting groundwater data. In general, these methods involve collection of:
- in situ measurements to calculate hydraulic head, groundwater flow direction and rate
- in situ measurements of apparent product thickness (NAPL, immiscible with water)
- in situ physical and/or chemical measurements of groundwater quality, e.g. redox potential, electrical conductivity, pH, and dissolved oxygen
- collection of groundwater samples for ex situ measurement/analysis.
The main issues that determine the selection of the appropriate method(s) are:
- the DQOs
- site-specific conditions such as depth to water table, soil/rock competency
- analyte-specific characteristics
- financial and logistical constraints.
Careful consideration and appropriate weighting of each of these issues will assist in determining the most appropriate method(s) of groundwater investigation.
The appropriate scope of the investigation is determined through the development of the site CSM and the DQOs. Generally this will include a preliminary site investigation comprising a desktop review of relevant background and historical information and a site visit followed by one or more intrusive field programs to update and refine the CSM until the objectives of the site assessment are met. The results of any earlier investigations, including soil and soil vapour investigations, should be used to refine the CSM and inform the scope of the groundwater investigation.
Typically this will include consideration of the following:
- the nature of the contaminant including its mobility and toxicity characteristics
- the type and location of known and potential contaminant source zones (including off-site) and associated contaminant plumes
- site geological and hydrogeological conditions (lithology, lateral and vertical extent of aquifers, perched water tables, confining layers, aquifer properties, etc.)
- depth to the water table and likely seasonal variation
- potentiometric surface(s)
- upgradient groundwater quality to assist in determining background groundwater quality
- hydrogeochemistry of relevant aquifer units
- direction of groundwater flow and hydraulic gradient
- location of recharge and discharge areas
- location of any abstraction wells
- current and future realistic use(s) of the groundwater resource and nearby surface water resources and water protection zones
- known and/or perceived risks to the environment and/or human health including the presence of potential pathways between contaminant source(s) and potential receptors.
Site-specific conditions that may limit or govern the choice of groundwater investigation techniques include:
- hydrogeological conditions including the depth to groundwater, soil/rock types and the presence of multiple aquifers
- potential risks to uncontaminated aquifers and/or surface water resources
- restrictions with regard to accessibility due to topography, ground bearing capability, site infrastructure or interference with site operations
- risks to the environment and/or public safety
- geotechnical limitations such as soft or saturated ground, cavernous or karstic terrains and stability
- natural events such as flooding and shifting sand dunes.
Any of these conditions may limit the applicability of certain methods of drilling, bore installation and groundwater sampling and make other methods more practical and cost-effective.
Appropriate measures should be taken to minimise the spread of contamination by not creating migration pathways from the surface to groundwater or between different aquifers. For example, where a monitoring well is targeting a deeper aquifer unit and contamination is present in a shallow aquifer unit or overlying fill horizon, this should be cased off so as not to permit cross-contamination between the two units.
The physical and chemical characteristics of contaminants have a profound effect on their sub-surface distribution and/or occurrence in groundwater at a given site. Physical and chemical characteristics that may have an effect on the distribution of contaminants include:
- contaminant solubility
- presence of NAPLs
- relative density (e.g. in the case of NAPLs, LNAPLs such as oils are less dense than water, whereas dense NAPLs (DNAPLs), such as some solvents, are denser than water; for aqueous liquids, relative salinities are important)
- stability (chemically and microbiologically)
- partitioning characteristics (e.g. sorption and volatility)
- aquifer redox conditions.
These characteristics will determine if contaminants are:
- capable of leaching through a soil profile and/or are soluble in the groundwater
- more or less dense than the groundwater, such that there is a likelihood for them to be present close to the water table (e.g. LNAPLs or where low salinity water infiltrates into more saline groundwater) or more extensively throughout the aquifer (e.g. with DNAPLs or where saline water infiltrates through fresh groundwater)
- relatively susceptible to effects of volatilisation, reaction with other chemicals/substances in the sub-surface, biodegradation, or attenuation.
Where there is a potential for contaminants to be present in an aquifer it is important to understand and predict where they are most likely to be concentrated prior to selecting the appropriate groundwater investigation method. Without this consideration, there is the potential for errors, some of which may result in:
- cross-contamination within and/or between aquifers
- non-detection of groundwater contamination
- inaccurate or misleading data
- expenditure of excessive resources where more simple and cost-effective methods could have been used.
In general, most groundwater investigations in Australia are conducted using information obtained from cased, semi-permanent or temporary groundwater monitoring wells. Monitoring wells are used for a range of applications including:
- groundwater sampling for ex situ analysis
- monitoring and/or profiling in situ groundwater parameters
- monitoring of groundwater level fluctuations
- aquifer testing.
Wells retained as part of a monitoring network should be properly maintained to ensure the integrity of the sample data collected. Well lifespan will depend on the materials used, the standard of installation and whether aggressive ground conditions are present. The monitoring plan should include provisions to inspect and assess monitoring wells for their suitability for monitoring purposes. Damaged or abandoned bores may provide conduits for future contamination unless properly decommissioned. Monitoring wells which are no longer required or are unsuitable for continued monitoring should be decommissioned in accordance with jurisdictional requirements.
The following overview of drilling methods is largely based on information in EPA Victoria (2000) and SA EPA (2007), EA (2006) and Standard practice for design and installation of groundwater monitoring wells, ASTM D5092 (2004). Additional information can be found in Aller et al. (1989), Driscoll (1986) and the Manual of methods, applications and management produced by the Australian Drilling Industry Training Committee (ADITC 1997).
A careful record of the geology encountered during drilling should be described and classified in accordance with Geotechnical site investigations, AS 1726-1993. Example logs are included in Appendix D.
Field check lists are available in the Toolbox at www.scew.gov.au/nepms/assessment-of-site-contamination.html to aid documentation of essential information including the soil profile and well construction.
Monitoring wells may also be logged using various geophysical techniques, for example, to determine aquifer characteristics in more detail and to supplement other methods (for example, geological logging, core analysis, aquifer tests, water sampling and analysis). Further information can be found in ASTM D5753-05 Guide for planning and conducting borehole geophysical logging (and related standards) and at the United States Geological Survey website: http://ny.water.usgs.gov/projects/bgag/intro.text.html.
Monitoring wells should be constructed to an appropriate standard and from suitable materials to ensure that high quality samples can be collected over the projected lifetime of the well. The assessor should ensure that the drilling technique, depth and diameter of the borehole, screen length, well construction materials (screen, casing, filter pack, seals and grout) and headworks design selected are compatible with the monitoring objectives.
For general guidance on monitoring well installation procedures see ASTM D5092-04.
A decision on the appropriate means of constructing monitoring wells involves consideration of a number of factors including the hydrology, geology and geochemistry of the formation, the nature of the contamination, the chemical resistance and leaching properties of the construction materials, the cost and the necessity to maintain the integrity of samples. Further information may be found in EPA Victoria (2006), SA EPA (2007), EA (2006) and ASTM D 5092-04.
There are several standard drilling methods available including hollow-stem auger, air and mud rotary, cable tool, sonic and direct push. The general suitability of these techniques for a range of ground conditions is discussed in EA (2006). The assessor should consider drilling methods that minimise the introduction of drilling additives wherever possible.
Certain drilling techniques can cause smearing (for example, rotary auger) or compaction (cable tool) of borehole walls and may also promote transport of geological formation materials and drilling fluids into different aquifer zones. In a worst-case scenario, this can result in almost complete blockage of the well screen resulting in non-representative groundwater samples when the boreholes are monitored.
Drilling fluids are used to clean and lubricate the drill bit, to remove rock cuttings from the borehole and to keep the borehole open during drilling. These may include air, water and specific drilling mud formulations or native clay slurries. Drilling fluids can have a range of effects on groundwater quality such as the following:
- air may severely disturb hydrochemical profiles through oxidation processes, e.g. oxidation of ferrous Fe2+ to ferric Fe3+
- water may dilute or flush groundwater near the borehole and cause precipitation of minerals, thereby blocking or obstructing groundwater pathways
- mud may invade the permeable formations and block pathways to the well screen.
Care should be taken to avoid contamination of the borehole and surrounding geology during drilling and construction of the well through the inappropriate use of lubricants, oils, grease, solvents, or materials with incompatible coatings. If the groundwater quality is altered, the samples obtained may not be representative, leading to uncertainties and potential errors in the assessment. Some considerations for material selection (EA 2006) include:
- the chemical environment in which the installation is placed – aggressive environments (saline, free-phase, low or high pH) will rapidly degrade or corrode some materials
- effect of contaminants on materials – corrosion, solution, strength, leaching
- effect of materials on groundwater – leaching, oxidation, pH.
As some drilling-related effects are frequently long-lived or even permanent, it is important to record drilling method, materials used and details of bore development on the well logs.
Groundwater investigations should be designed to target the part, or parts, of the aquifer most likely to be affected by contamination.
Under laminar flow conditions, contaminated groundwater flows in discrete zones controlled by the physical properties of the aquifer and the presence of any preferential pathways such as higher permeability units (such as the cleanest sands in an interbedded sand and silt sequence) and fractures. The location and length of the well screen is therefore critical to obtaining a representative sample of contaminated groundwater.
The selection of screen length depends on the objectives for the monitoring well; however, in general, well screens should be kept as short as possible to avoid potential dilution effects. The interval of aquifer potentially contributing to flow includes the filter pack either side of the well screen as well as the screened interval itself. To minimise the potential for vertical flow between aquifers via the well bore, screens should not be installed across different geological units or water-bearing zones.
Screen design should consider the likely fluctuation in the water table and the well screen should be located such that at least part of the screen remains within the saturated zone throughout the year. Where extreme variations are likely to be present, (e.g. drought and non-drought periods) consideration may need to be given to installing additional monitoring wells.
In the initial phases of investigation, well screen lengths of 3 m or more are common. However, once contamination is suspected or confirmed, shorter screens of the order of 1 m long located specifically within the zone of interest are recommended since small-scale heterogeneities are important in controlling contaminant flowpaths. Where the geological unit of interest exceeds 1–2 m in thickness, multiple wells completed in well nests or vertical groundwater profiles are recommended to evaluate and define the contamination. In thick homogeneous granular aquifers, the benefits of short well screens are more limited, given that mixing and contaminant dilution will occur within the aquifer itself (EA 2006). However, consideration should still be given to the potential for vertical gradients within thick aquifer systems.
Monitoring dissolved contaminants in plumes requires consideration of the likely plume characteristics and its behaviour in the aquifer. Plumes are typically elongated in the direction of groundwater flow and will undergo longitudinal, lateral and vertical dispersion. Plumes will also tend to sink as additional recharge is added to the aquifer downgradient of the plume source area (EA 2006).
General guidance on the selection of appropriate screen length is given in Table 5 below.
Table 5. General guide to selection of an appropriate monitoring screen length
Aquifer conditions/monitoring objectives | Screen length |
Multi-level | Very short (<1 m) | Short (12 m) | Long (3+ m) |
Monitor general water quality (thick aquifer) | | X | | |
Monitor general water quality in thin or heterogeneous aquifer | X | | | |
Monitor LNAPL (fluctuating water table aquifer) | X | X | | |
Monitor DNAPL | X | | | |
Detailed delineation of contamination | | | | X |
Key: | |
X Not appropriate | Appropriate |
Appropriate but not ideal | Most appropriate |
Adapted from EA (2006) | |
| | | | | |
Correct slot size and location of well screens is particularly important when dealing with NAPLs. Representative samples of the dissolved phase can only be obtained if the screened interval is outside the influence of any mobile or residual NAPL. Interface meters and tapes with oil-indicating pastes can be used to confirm the presence and thickness of NAPLs.
Further information on sampling LNAPLs can be found in Clements et al. (2009). Further information on sampling and identification of DNAPLs can be found in Keuper and Davies (2009) and EA (2003).
A filter or gravel pack is used to minimise the entry of fine-grained material into the well screen. The filter pack should be chemically inert and matched to the aquifer particle size and to the screen slot size. For further information see ASTM D5092-04.
In general, the filter pack should extend no more than 1 m above and below the well screen in the well annulus after settling, taking care not to extend the filter pack across geological units or water-bearing zones.
Geotextile wraps (filter socks) are not recommended for use in bores intended for monitoring groundwater quality. The redox conditions, and therefore biological activity, within the mesh can be different from that present in the aquifer, which can produce misleading sample results. Where LNAPLs are present, a greater thickness of NAPL would be required to overcome the increased surface tension forces present in the fine mesh compared with the well screen.
The annular space from the top of the filter pack to ground level (or next monitoring screen in multi-level monitoring wells) should be backfilled with bentonite or a non-shrinking bentonite-based grout (cement grout or a cement/bentonite mix). A seal should always be placed on top of the filter pack to prevent these materials from entering the well screen.
Incorrect installation of wells can result in costly cross-contamination of aquifers. In installations above the water table, the use of bentonite pellets in isolation to form a seal is discouraged as the bentonite pellets can set dry and crack, resulting in an ineffective seal between the aquifer and contamination near the ground surface. In these circumstances it is preferable to hydrate the bentonite at surface and then install as a slurry. For guidance on appropriate installation procedures see EA (2006) and ASTM D5092-04.
Drill cuttings should be collected in suitable containers and disposed of appropriately. It is not acceptable practice in site assessment to use drill cuttings as backfill in boreholes.
It is essential to correctly finish all monitoring wells at the surface such as with a suitable bentonite plug and cement seal, to ensure that surface water runoff does not collect at the wellhead and leak down the outside of the casing. The borehole headworks form the interface between the borehole and the surface environment.
In designing headworks a number of issues should be considered (EA, 2006):
- security—to prevent vandalism or malicious actions and to prevent access by animals
- protection—from entry of surface water or other foreign material and from activities at surface (such as vehicle movements)
- accommodation of equipment—storage of equipment such as data loggers and dedicated sampling devices
- visibility—designed to be clearly visible or non-obtrusive depending on location.
In general, an above-ground completion is preferred as this type of design is less likely to suffer from inundation, is easier to find in the field and is more easily secured.
Casing materials such as PVC, ABS, Teflon, etc. which project from the ground can easily be damaged and should be protected by a steel or similar outer protective collar.
All bores intended for monitoring water quality should be developed after drilling to remove fine sand, silt, clay and any drilling mud residues from around the well screen to ensure the hydraulic functioning of the well. Development should be carried out as soon as possible after drilling and installation, however, a minimum of 24 hours should be allowed for bentonite seals to fully hydrate and grout to cure (harden and set). A detailed record should be kept of well development activities and reported in the relevant site assessment report.
Development usually involves agitating the water column in the well bore and pumping the water out until it runs clear. During development, bore yield should be estimated by monitoring the rate of recovery of water in the bore after pumping. This information can then be used to select suitable methods for subsequent purging and sampling (SA EPA 2007).
Development should continue until a defined endpoint has been reached (EA, 2006), such as:
- chemical indicator stability– using field measuring techniques for pH, EC and dissolved oxygen, development is continued until these parameters stabilise in abstracted water, or
- reduced turbidity – development is continued until the abstracted water is reasonably clear and free of suspended solids.
After development, bores should be left for a period until borewater chemistry can be demonstrated to have stabilised (generally between 24 hours and seven days) before samples are collected. Longer periods are applicable to low permeability aquifers and to reduced groundwater conditions where it may take days to weeks to fully equilibrate, depending on the aquifer properties.
Care should be taken to dispose of any contaminated water responsibly and not to allow it to enter the stormwater drainage network or to impact uncontaminated soils at the site.
It is essential that groundwater sampling methods result in the collection of samples that are representative of aquifer conditions. Management decisions that may involve considerable expenditure and potential inconvenience to the public will be based on these results. In many circumstances, budgeting for additional sampling and analysis costs for site characterisation for definition of groundwater contamination problems could save further assessment expenditure and costly delays to property transactions and site development.
Where possible, established ‘standard methods’ from recognised sources such as Standards Australia, the United States Environmental Protection Agency (US EPA), the American Public Health Association (APHA) and International Standards Organisation (ISO) should be used for the analysis of groundwaters. The general reference used by laboratories is Standard methods for the examination of water and wastewater (APHA et al. 2005).
Overviews of groundwater sampling procedures are readily available; for example, SA EPA (2007) includes information about:
- development of monitoring plans
- pumping and sampling equipment
- sampling methods
- groundwater level measurement
- purging
- sample collection methods
- filtration
- NAPL sampling
- decontamination
- sample identification, transport and storage
- labelling and identification
- preservation techniques
- QA/QC.
Detailed information can be found in Standard AS/NZS 5667.11-1998; MDBC (1997); EPA Victoria (2000); Nielsen (2006) and Nielsen & Nielsen (2005).
An appropriate method of groundwater sampling should be selected in relation to the nature of the target analytes and the hydraulic characteristics of the monitoring well. In general, the use of low-flow submersible pumps or positive-displacement pumps capable of controlling flow rates and minimising purging requirements are the preferred methods of groundwater sampling for site characterisation purposes. A discussion of the benefits and limitations of low-flow purging and sampling can be found in ASTM D6771-02.
No-purge sampling techniques (see below) may also be appropriate, particularly for long-term monitoring applications. A discussion of the applications and the benefits and limitations of passive sampling can be found in ITRC (2005).
Purging and sampling methods using bailers or high speed pumps are not recommended due to the difficulty of obtaining a representative groundwater sample. These methods result in degassing of samples and can also introduce high levels of turbidity. Sampling-induced turbidity may be mitigated by using low-flow purging and sampling techniques (Puls & Barcelona 1996).
Generally, the same methods should be used each time the wells are purged and sampled to avoid introducing sampling method-related uncertainties to the analytical data (SA EPA 2007). Where an improved technique becomes available, it is recommended that it is trialled in combination with the existing sampling method to establish the nature and magnitude of any changes in analytical results as a result of the new sampling method.
Passive sampling devices (for example, passive diffusion bags for VOCs) do not require pumping or purging of groundwater to acquire a sample. These sampling devices are placed at a selected depth in the well and rely on ambient flow through the well screen for sampling.
Three types of passive sampling technologies are available:
- devices that recover a grab sample of groundwater (producing an equilibrated ‘snapshot’ of groundwater quality)
- devices that rely on diffusion of the analytes for the sample to reach and maintain equilibrium with the sampled medium
- devices that rely on diffusion and sorption to accumulate analytes in the sampler.
Some of these passive sampling devices are applicable to the sampling of surface waters and vapour as well as groundwater. Further information can be found in ITRC (2005) and ITRC (2007a).
The selection of the appropriate equipment for a groundwater investigation should be based on careful consideration of the attributes of the target analytes, the likely contaminant distribution, cost and logistical issues, field filtration requirements, and decontamination requirements.
Certain analytes are prone to effects of aeration and agitation and sampling equipment should be selected to cause minimal agitation and chemical alteration of the sample, for example, low-flow techniques are recommended for quantitative assessment of VOCs and SVOCs; bailers are not appropriate.
Information on monitoring and sampling LNAPLs using oil-water interface probes, oil indicator pastes and special bailers can be found in Clements et al. (2009).
Sampling equipment should also have negligible capacity for sorption, precipitation and oxidation of analytes of interest.
Due to a range of chemical and/or physical characteristics, contaminants may be concentrated in certain parts of the aquifer under investigation. The sampling equipment should be capable of targeting the depth interval most likely to contain the target analytes. For example, special bailers are available for the sampling of NAPLs, while bottom-loading bailers are available for investigating DNAPLs.
All equipment used in the sampling procedure which either enters the well bore or holds the groundwater sample should be decontaminated before and after each sample is collected. Samples of the rinsate should be included in the QA/QC program. Depending on the potential for cross-contamination between wells or within the profile of a single well, certain equipment may be relatively difficult to decontaminate and it may be necessary to opt for more simple sampling systems or to dedicate sampling equipment to a particular well or interval.
In addition to the above decontamination procedures, it is good practice to sample wells with no/minimal contamination first to minimise potential cross-contamination of samples.
In surface water bodies, a substantial amount of metals can be transported adsorbed to suspended particles and filtering needs to be undertaken to identify the dissolved component if quantification of dissolved metals is required.
This is much less the case in groundwater systems where particles cannot easily pass through the porous aquifer matrix. Typically filtration with a 0.45 μm filter will remove the majority of suspended particulates, however, it may be necessary to filter samples with a 0.1 μm filter to remove all suspended particulates.
For dissolved metals, in-line disposable filters (or micro-filtration syringes) are recommended to ensure that groundwater samples have minimum exposure to the atmosphere. Micro-filtration syringes are now widely available and present a viable option to filter in the field even for silty aquifers.
Filtered samples should be collected in pre-prepared bottles containing sufficient acid to maintain the pH of the sample to < pH 2.
Filtration to remove sediment from groundwaters upon receipt in the laboratory is not recommended for analysis of dissolved metals unless it has been demonstrated that the analystical results are consistent regardless of whether filtering is carried out in the field or the laboratory.
Field filtration is not required for total metal analysis.
Sample filtration devices should be decontaminated between uses or discarded to prevent cross-contamination and to ensure continued effectiveness. Further information may be found in SA EPA (2007) and EPA Victoria (2000).
Some physicochemical parameters cannot be reliably measured in the laboratory as their characteristics change over a very short timescale. Parameters that should be measured in the field include pH, electrical conductivity (EC), temperature, dissolved oxygen (DO) and redox potential (Eh). If ferrous iron is one of the selected analytes, it also is best analysed in the field.
It is recommended that field parameters are measured in a flow-through cell to avoid contact between the groundwater and the atmosphere. A flow-through cell can also enable continuous measurement and monitoring of key parameters during purging to identify when a representative sample may be obtained.
There is a wide range of equipment available for the measurement and logging of these parameters. It is important that quality assurance protocols are developed and implemented. The procedures should include the use of suitable calibration standards, where the calibration spans the anticipated range of results, and accuracy checks. Where measurements are made over a number of hours, periodic readings of appropriate reference solutions should be incorporated to ensure that the calibration is stable. Calibration procedures vary between meters and between manufacturers so it is important to follow the manufacturer’s instructions for correct and accurate operation of each piece of equipment. Further information can be found in SA EPA (2007) and Sundaram et al. (2009).
Groundwater level measurements are essential to determine groundwater and contaminant flow directions within aquifers and interaction with surface water bodies. These measurements can provide information on lateral and vertical head distribution and hydraulic gradients within individual aquifers and between aquifers in layered aquifer systems (EPA Victoria 2000). Long-term groundwater monitoring data provides information on temporal trends in groundwater levels (and hence flow directions and rates) due to seasonal, climatic and groundwater pumping effects (EPA Victoria 2000).
The groundwater elevation (standing water level) in a monitoring well is an expression of the hydraulic head of the aquifer unit in which the well has been screened. The standing water level should be measured relative to a permanent surveyed reference point (such as the top of the casing) before any purging or sampling takes place using a calibrated pressure transducer and/or purpose-built tape or meter. The data should be reported relative to a common datum, preferably Australian height datum. Bores installed at multiple depths within an aquifer are required to assess vertical groundwater flow direction(s).
Relative groundwater elevations within the same aquifer unit indicate the hydraulic gradient between wells and, given at least three wells spaced roughly equilaterally, a groundwater flow direction may be calculated. Where the wells are completed with long screens and/or at different relative depths within the aquifer, inconsistencies may arise if there are vertical groundwater gradients present. Groundwater flows may vary significantly at a site so it is recommended that groundwater contour maps are based on several bores monitored over a period of time to determine groundwater flow directions and variability across the site over time.
Water level measurements for a given study area preferably should be taken on the same day. High frequency monitoring may be required to quantify groundwater pumping (abstraction) and/or tidal effects. Consideration should be given to use of data loggers to identify fluctuations in groundwater levels depending on the uncertainties identified in the CSM and the assessment objectives.
The use of hydraulic head measurements in groundwater of variable density is more complicated than is the case for constant-density groundwater. Density variations can result from differences in temperature or pressure but more commonly in site assessments, these effects are caused by differences in solute concentration. Variable density is particularly relevant for sites in coastal areas where deeper wells may be screened within a saltwater wedge and shallow wells within freshwater. Water-level data obtained from wells screened within saltwater must be converted to an equivalent freshwater head to enable correct calculation of vertical and horizontal gradients and to interpret groundwater flow. Further information may be found in Post et al (2007) and Serfes (1991).
Where LNAPL is present, it will affect the groundwater elevation measured at a groundwater monitoring well. If significant amounts of LNAPL are present, groundwater level corrections are necessary and are based on the measured thickness and relative density of the product. However, due to the uncertainties involved, corrected groundwater elevations from wells affected by LNAPL should not be used to definitively determine groundwater flow direction.
The hydraulic heads measured in wells screened in different aquifers should not be used to infer lateral groundwater flow direction at a site; however, they may be used to determine the relative hydraulic head, or potential for vertical flow between aquifers.
If vertical (downward) hydraulic gradients are present, there is the potential for a dissolved-phase contaminant plume to migrate downwards along the flow path resulting in uncontaminated water overlying sections of the contaminant plume (API 2006).
Knowledge of aquifer hydraulic (hydrogeological) properties is important for:
- the assessment of potential migration of contaminants in groundwater
- calibration and development of numerical models
- determination of applicable groundwater remediation methods.
In particular, knowledge of the rate of groundwater flow or groundwater velocity is essential for determining the timescale in which contamination may migrate off-site or threaten a receptor. Where the nearest receptor lies some distance from the site, screening level estimates may suffice; however, where greater certainty is required (for example, presence of nearby and/or sensitive receptors) then a more precise estimation method will be required.
Groundwater velocity in a porous medium aquifer can be estimated using a modified version of the Darcy equation:
v=Ki/n
where v is the advective groundwater velocity, K is the hydraulic conductivity, i is the hydraulic gradient and n is the effective porosity of the aquifer unit. The groundwater velocity calculated by this method assumes plug flow of contaminants and ignores dispersion. In reality a proportion of contaminant mass may arrive at a monitoring point (or receptor) much more quickly than is predicted by this method.
Hydraulic gradient is generally calculated based on groundwater elevation data (groundwater flow maps). Effective porosity (the percentage of interconnected pore space) is rarely measured in site contamination assessments and typically falls in a relatively narrow range for defined lithology types, for example, 2040% for sandstones. Hydraulic conductivity, K, may be estimated with varying accuracy by a variety of methods depending on the level of acceptable uncertainty. Commonly used methods include:
- literature approaches (screening level data only)
- literature values based on grain size/lithology descriptions
- hazen formula with grain size analysis
- slug tests that provide an indication of local hydraulic conductivity at the well bore
- pumping tests that provide information on a much larger volume of aquifer compared with slug tests
- tracer tests (in which the travel time of a conservative anion such as chloride is monitored between two points over time to directly estimate velocity).
More information on literature-based and aquifer test methods can be found in Fetter (2001) and other standard hydrogeology textbooks.
In general, aquifer testing involves the determination of a range of hydraulic properties within an aquifer. This is accomplished by stressing the aquifer at a test well, either by the addition or removal of water (or an equivalent volume of water using a weight or ’slug’) and measurement of the hydraulic response at one or more observation wells within the test area. Depending on the type of aquifer testing carried out, it is possible that groundwater monitoring wells could be used either as test wells and/or observation wells.
However, most aquifer test methods require specific well construction procedures such as screening of the full aquifer thickness. Further information on aquifer testing can be found in Standard AS 2368 (1990) and Kruseman and de Ridder (1994).
Using pumping tests to determine average hydraulic conductivity in an area of severe groundwater contamination can be undesirable (for example, where there is a risk of exacerbating DNAPL contamination) and in these circumstances less intrusive methods should be considered. Alternatively, where aquifer properties do not vary significantly, it may be possible to perform aquifer testing outside the impacted area.
The groundwater monitoring bore network should cover an appropriate study area to delineate the lateral extent of the contamination; define background groundwater quality, the groundwater flow system for the geological units of interest; and to assess the risk to relevant receptors. Generally the number of monitoring wells should be sufficient to define, at an appropriate scale, the lateral and vertical extent of the plume exceeding relevant assessment levels (for example, GILs, HILs, HSLs and/or site-specific risk-based criteria) and to understand any seasonal or longer-term variation in groundwater flow direction and rate of plume advance or retreat.
For large and/or complex sites with VOCs and/or SVOCs, consideration may be given to reducing uncertainty in lateral and vertical contaminant distribution by using various screening tools to identify and delineate contamination in both the vadose and saturated zones. These tools include soil vapour sampling (refer Section 9) and the membrane interface probe (MIP) and laser-induced fluorescence (LIF) tools (refer Sections 7.2.4 and 7.2.5 respectively). These tools can also be used in a reactive or adaptive field sampling program.
Although the number of bores, locations, depths and screen intervals are site-specific, groundwater site investigations require as a minimum:
- one upgradient bore to establish the quality of groundwater entering the site (one for each aquifer or geological unit of interest)
- two or three bores to monitor groundwater quality immediately downgradient and also lateral to each contaminant source (for each aquifer or geological unit of interest).
Sites with significant contamination and/or complex hydrogeology will require numerous bores at various depths to assess the lateral and vertical extent of contamination and the nature of any temporal variation.
The initial investigation bores should be:
- close to each potential contamination source
- installed with similar construction techniques to minimise sources of variation and uncertainty in the data
and, where appropriate,
- screened across the water table to locate any LNAPL and to identify contamination derived primarily from surface spills and leaching.
The number, spacing and depth of follow-up wells are site-specific considerations that should be informed by the CSM. The installation of bores without consideration of hydraulic gradient and conductivity values may result in bores being sited at improper spacings.
However, as a general guide for plumes estimated to be <200 m long, well spacing should be of the order of 2050 m in the direction of groundwater flow and 1020 m perpendicular to flow. Well spacing should generally be less than 10 m for the delineation of source zones.
Consideration should be given to installing one or more ‘sentinel’ wells to monitor the migration of an expanding or detached plume or to provide confirmation of the continued absence of contamination at a particular location. For example, in some situations it may not be possible to delineate the position of the contaminant plume front due to logistical constraints such as the presence of buildings. However, in this case an acceptable approach would be to install sentinel wells upgradient of the relevant receptors to provide an early warning of any significant plume advance. The location of the sentinel wells would ideally allow the implementation of management actions to protect the receptor if the plume were to advance significantly.
Source: Clements et al. (2009)
Delineation of vertical variability in groundwater chemistry is critical for risk assessment and remediation planning and reliance on too few monitoring points can lead to inaccurate estimation of contaminant distribution and behaviour.
Multiple wells may be required to adequately characterise the vertical groundwater profile and contaminant distribution. Samples obtained from short, targeted, multiple screens are more likely to be representative of the maximum concentrations present in the aquifer as they are less likely to be affected by the dilution that may occur with a longer well screen. Multiple monitoring wells should be considered where contaminant distribution is likely to be complex (for example, presence of numerous migration pathways or presence of pooled and residual NAPLs).
There are several methods available for screening multiple depths, including installing multiple wells in a small area, nesting multiple wells in the same borehole, and using a pre-fabricated bundle of multi-level wells. There are cost and technical considerations with each approach. Nested wells are cheaper to install; however, if poorly installed, cross-contamination may occur between screens. Bundled multi-level wells (consisting of multiple small diameter tubes in a bundle) can provide confidence in samples at relatively low cost. Multiple wells are typically more expensive, but provide greater confidence in monitoring results.
Consideration should be given to the potential for a ’diving plume’ to develop under the influence of natural or anthropogenic recharge or in response to large scale groundwater abstraction (for example, public supply or industrial process water). The depth to which a plume will downwardly migrate in an unconfined aquifer is dependent on the recharge rate and the groundwater seepage velocity.
Generally, greater recharge rates will result in a greater magnitude of downward migration but the recharge effects will be less at higher seepage velocities (API 2006).The US EPA provides an online tool which can be used to estimate plume diving caused by recharge and assuming simplified flow in a water table aquifer.
See www.epa.gov/athens/learn2model/part-two/onsite/index.html.
Source: Keuper and Davis (2009) and EA (2003)
DNAPLs are only slightly soluble in water and therefore exist in the sub-surface as a separate phase immiscible with both water and air. Common types of DNAPLs include timber treating oils such as creosote, transformer and insulating oils containing PCBs, coal tar, and a variety of chlorinated solvents such as trichloroethene (TCE) and tetrachloroethene/perchloroethylene (PCE). DNAPLs have the potential to migrate to significant depth below the water table through unconsolidated and consolidated materials and fractured bedrock, where they slowly dissolve into flowing groundwater and give rise to aqueous phase plumes.
Due to their physicochemical properties, DNAPLs migrate through the sub-surface in a very selective and tortuous manner and as a result can be challenging to investigate with traditional drilling techniques. Upon release, DNAPL will move and distribute itself into disconnected blobs and ganglia of liquid (residual DNAPL) and in connected distributions (pooled DNAPL). Residual DNAPL is found both above and below the water table within the migration pathways and typically occupies between 5% and 30% of pore space in porous media and rock fractures. Residual DNAPL is trapped by capillary forces and typically will not enter an adjacent monitoring well, even under the influence of aggressive pumping.
Pooling of DNAPL occurs above capillary barriers, typically layers, and lenses of slightly less permeable materials. Penetration through silts and clays may occur if windows are present within the layers or if the layers are penetrated by preferential pathways, for example, tree roots. The presence of dipping fractures, bedding planes, joints and faults may enable a DNAPL to continue to migrate downwards. Downward migration of chlorinated solvents may cease within a few months to a few years of release in relatively permeable media, compared with many decades for high viscosity DNAPLs such as creosote and coal tar to cease migration.
The DNAPL source zone comprises the overall rock volume of the sub-surface containing residual and/or pooled DNAPL. In addition to the DNAPL, there may be significant amounts of contaminant mass that has diffused into low permeability zones. Back diffusion of sorbed contaminant mass from the aquifer to groundwater may sustain dissolved-phase plumes for significant periods of time (decades to hundreds of years).
Above the water table, volatile DNAPLs can vaporise into air-filled pore spaces and for DNAPLs with significant vapour pressure such as chlorinated solvents, this can lead to expanded vapour-phase plumes in the unsaturated zone. Passive soil vapour surveys may be useful for delineation of DNAPL source zones particularly in situations where groundwater has been impacted by VOCs but the source has not been identified. In warm dry conditions, the persistence of some DNAPLs such as chlorinated solvents can be relatively short (months to a few years) in unsaturated media. The absence of residual and pooled NAPL in the unsaturated zone may not, therefore, be sufficient evidence to conclude that DNAPL has not migrated below the water table at the site of interest.
Determining the presence or absence of a DNAPL is an important consideration for the development of the CSM. If the presence of DNAPL is suspected, care should be taken to avoid drilling through the DNAPL and dragging or spreading pooled DNAPL beyond the current location or creating a pathway for the DNAPL.
It is now commonly accepted that direct visual observation of DNAPL does not occur at most DNAPL sites and instead, the presence of DNAPL is usually inferred from converging lines of evidence. Site-specific considerations will dictate which lines of evidence (see below) should be pursued. Care, however, should be taken to ensure that a negative response to one or more lines of evidence is not simply attributable to inadequate characterisation and an insufficient amount of data.
The site investigation methods and related interpretation techniques (lines of evidence) which can be useful for characterising DNAPL source zones include:
- visual observation in groundwater samples or drill core
- chemical saturations in soil above threshold DNAPL saturation
- chemical concentrations in soil above equilibrium partitioning threshold
- mapping of a vapour-phase plume (based on shallow soil vapour measurements) if present
- hydrophobic dye testing of DNAPL in soil or water samples or using a down-hole ribbon sampler impregnated with dye
- interpretation of groundwater concentration data from locations immediately downgradient of the suspected source zone and trends with depth and over time.
As a general ‘rule of thumb’, groundwater concentrations in excess of 1% effective solubility may indicate that the groundwater has come into contact with DNAPL. Values of 1% solubility concentration for various chlorinated solvents can be found in Appendix B of US EPA (2009).
Further information may be found in Keuper and Davis (2009) and references therein. A tabulation of parameters and other information that may be needed at various stages of site investigation, risk assessment and selection of management options can be found in EA (2003).
Source: EA (2000b)
Assessors should be aware that dissolved contaminants may move at different rates not only as a result of physical processes, but also because of chemical interactions with soil and aquifer components. Attenuation processes include advection, dilution, dispersion, diffusion, sorption, degradation (biotic and abiotic) and volatilisation.
Consideration should be given to the fate of the contaminant(s) as it moves along the migration pathways. This requires that chemical, physical and biological interactions between sources and sub-surface materials are taken into account. The CSM should describe the processes that control the movement of contaminants in soil and the unsaturated and saturated zones.
Degradation can be a significant process in decreasing contaminant mass. The actual rate of biodegradation varies according to a range of factors including contaminant type, microbial populations, redox conditions, temperature and the chemical composition of aquifer materials and groundwater. Evidence for the occurrence and efficiency of degradation processes should be considered in the development of the CSM. Where modelling is undertaken, care should be taken to ensure that the biodegradation process(es) being modelled is appropriate and that realistic reaction rate constants are used.
Further information on attenuation processes and their effects can be found in EA (2000a, 2000b), ITRC (1999), ITRC (2010) and Beck and Mann (2010).
This section provides an assessment framework for vapour intrusion (migration of vapours into a building) and basic requirements for measurement of volatile organic compounds (VOCs) in soil vapour, indoor air and outdoor (ambient) air. Primarily, the methods included are applicable to chronic low levels of vapour concentrations as are typically encountered in contaminated site assessments.
Vapours may be generated by biological, chemical and physical decomposition of spilled or dumped wastes. Assessment of ground gases associated with operating or closed landfills (‘landfill gas’) or buried putrescible wastes is beyond the scope of this guidance. Information on these applications can be found in NJDEP (2005a) and Wilson et al. (2007).
Soil vapour surveys have a wide application in the assessment of volatile contaminants, for example, they may be used when:
- assessing the presence or absence of VOC contamination
- delineating VOC contamination in soil and groundwater
- characterising VOC contamination
- identifying/differentiating between sources of VOC contamination
- assessing VOC migration pathways in groundwater
- monitoring biodegradation of contaminants
- assessing vapour intrusion risk.
An overview of vapour fate and behaviour processes relevant to VOCs can be found in Davis et al. (2004, 2009a and b). Additional information on assessing vapours in the context of human health risk assessment is provided in Schedule B4.
The assessment of vapours should be undertaken by appropriately qualified and experienced professionals.
If vapour intrusion is suspected of posing an existing or imminent threat to human health, including from inhalation exposure or risk of explosion, then immediate mitigation or management strategies should be implemented.
Source: API (2005), ITRC (2007a), Davis et al. (2009a) and ODEQ (2010)
For the vapour intrusion pathway to be complete, there must be three components present – a source of sub-surface vapours (in soil and/or groundwater), occupied buildings or the potential for occupied buildings, and a migration route to connect them. Once the pathway is identified as being complete or potentially complete, a staged approach to assessment informed by the iterative development of a CSM is recommended—refer Section 4. For smaller sites, a single phase of work may be adequate to determine vapour intrusion potential, while larger sites can require multiple phases of vapour sampling to fully define the area of concern and accurately characterise the risks.
As a preliminary screening measure, the potential for a vapour intrusion risk should be considered where the Henry’s law constant for a substance is greater than 10-5 atm/m3/mol and its vapour pressure is > 1 mm Hg at room temperature. In addition to these measures, a substance should be assessed as volatile if its saturated vapour concentration results in exposure concentrations that are a risk to the exposed population. Some chemicals with low Henry’s law constants, or low vapour pressures, are so toxic that even a small amount that moves into the vapour phase could be enough to contribute to a risk. Hence both measures of volatility and toxicity need to be considered (Refer Schedule B4). This includes substances such as petrol, diesel, solvents and certain pesticides and PAHs.
In addition, some sites may be screened out of the assessment by the use of a lateral exclusion distance of 30 m from the sub-surface extent of the vapour source̶further information on the rationale for this criterion may be found in Davis et al. (2009a). A shorter exclusion distance may be considered for petroleum hydrocarbons where there are no other volatile contaminants of concern.
To apply this criterion to a groundwater source, there should be a high degree of confidence based on field data that the dissolved phase plume is stable or shrinking in lateral extent and is not continuing to expand.
This exclusion distance is not applicable to soil or groundwater sources where:
- the source is intersected by utilities or other potential preferential pathways
- continuous low permeability cover (for example concrete) is present between the source and the nearest buildings or enclosed spaces which impedes the diffusion of oxygen into the subsurface
- conditions are present that could promote lateral migration (e.g. landfill gas production, highly layered soils).
As an investigation progresses, soil vapour sampling results should be used to inform and establish the site-specific boundaries for the area of potential vapour intrusion concern.
A well-developed CSM incorporating vapour risk is essential for understanding current site conditions, determining potential vapour behaviour (including possible variation in soil vapour concentration) and, as part of the DQO process, identifying data gaps and uncertainties and priorities for investigation. The general requirements for the development of CSMs and DQOs are discussed in Sections 4 and 5.
Site-specific data which may be needed for vapour intrusion pathway risk assessment includes measurement of:
- VOCs in soil vapour within the fill and/or native soils below/adjacent to existing buildings
- VOCs in groundwater beneath or adjacent to potentially affected buildings or future buildings
- VOCs in indoor air, outdoor (ambient) air, or soil
- ambient VOCs that may contribute to VOCs measured at the site
- VOCs in preferential migration pathways such as service trenches for utilities
- physical properties, such as soil moisture content, saturation porosity and grain size distribution, relevant to vapour intrusion.
Consideration of preferential vapour migration pathways is an essential part of the development of the CSM. These may intersect vapour sources or soil vapour migration routes, for example, building sumps, drains, or utility and service connections to any buildings. Natural preferential pathways may also occur, for example, tree roots or fractured bedrock where the fractures are interconnected and in direct contact (including connection by permeable fill) with the building foundation and vapour source.
For the assessor to conclude that the vapour intrusion/emission pathways are unlikely to be active or to present a significant risk, multiple lines of evidence are required. This requires the assessor to present several reasoned lines of evidence as to why the pathway is considered inactive/unlikely to present a significant risk.
The following are some possible lines of evidence which may be considered (listed in no particular order):
- soil vapour spatial concentrations sub-slab, near-slab (or crawl space) with some level of vertical profiling if appropriate
- groundwater spatial data with vertical soil vapour profiling if appropriate
- information on background outdoor and indoor sources
- building construction and operating conditions
- indoor air data and concurrent ambient air data
- comparison of vapour constituent ratios in soil vapour with crawl space/indoor air
- biodegradability of vapours and availability of oxygen.
Measurement of indoor or ambient air is the most direct approach to assessment of vapour exposure. However, indoor air sampling can be expensive if many samples over a reasonably long period are needed to obtain representative results. In homes and workplaces, gaining access can be difficult and may lead to unnecessary concern on the part of the occupants. Depending on the volatile compounds considered, ambient and indoor air results may be difficult to interpret since confounding sources of contamination (refer Section 9.3.1) may be present. Where affected by background sources, the collection of indoor or ambient air measurements may not be considered the most appropriate approach.
Soil vapour measurement is the preferred route in most situations where a vapour issue (from a subsurface source) is considered likely to exist.
In the absence of measured soil vapour concentrations, it is also possible to model the generation of vapour from soil, groundwater and non-aqueous phase liquids. This procedure adds another level of uncertainty to the process. The uncertainties associated with the use of a model should be well understood and discussed in relation to the nature of the volatile contaminants assessed. Where unresolved uncertainties or unacceptable risks are predicted by modelling vapour concentrations, direct measurement of soil vapour and/or indoor and ambient air should be obtained.
Source: API (2005), Davis et al. (2009a), ITRC (2007b), ODEQ (2010) and US EPA (2012a)
When designing an SAQP, consideration should be given to the following:
- confounding sources of VOCs and SVOCs
- degradability of vapours and potential presence of daughter compounds
- land use
- environmental factors including spatial and temporal variability issues
- reliability, representativeness, precision and accuracy of available measurement techniques
- potential for preferential migration pathways.
Indoor air sampling is the most direct method of measuring VOC exposures where the CSM has identified that vapour intrusion is a potentially complete pathway. In circumstances where very high levels of contamination are present or the contamination has a unique character, the data can provide relatively quick confirmation of vapour intrusion impacts. However, for most sites, simply detecting VOCs inside a building is not definitive evidence of vapour intrusion.
Outdoor or ambient air commonly has detectable levels of VOCs, sometimes exceeding ambient air guideline values. The largest sources of these contaminants include vehicle emissions, fuel storage facilities and emissions from commercial/industrial activities (including service stations). As outdoor air typically makes up 99% to 99.99% of indoor air, ambient VOC levels tend to represent the minimum concentrations in indoor air. Buildings can also contain interior sources of VOCs, which include building materials, paints, dry-cleaned clothes and some commercial and household cleaning products. It is therefore advisable to conduct a survey of the building interior in advance of any indoor sampling to identify potential confounding sources and eliminate them as far as practical prior to sampling and to obtain concurrent ambient air samples.
As it is often not possible to remove all interior sources of VOCs prior to sampling, indoor air results should only be used in the context of a multiple-lines-of-evidence approach. To reduce the frequency of false positives, indoor air sampling is not recommended until other information (lines of evidence) indicates a potential vapour intrusion risk.
Further information including detailed protocols for the collection of indoor air data can be found in ITRC (2007b and 2007c), NYSDOH (2006), and NJDEP (2005b).
The concentration of petroleum hydrocarbon (such as TPH and BTEX) vapours in well-oxygenated, generally near-surface soil can be significantly reduced by biodegradation (Davis et al. 2009a, 2009c). However, this is generally not the case in less oxygenated soil such as under large areas of impermeable hardstanding or building foundations.
For petroleum hydrocarbons, the fundamentals of an approach to include an exposure reduction factor due to aerobic biodegradation are discussed in Davis et al. 2009c and included in Schedule B1. The approach is applicable to vapour sources at depths of 2 m or greater and requires the recovery of a soil vapour sample from a depth of at least 1 m below ground in close proximity to the building (or in a similar nearby soil, soil moisture and soil coverage environment). Where the building slab penetrates the ground by more than 0.3 m, then the additional depth of penetration of the slab below 0.3 m should be added to the depth at which the soil vapour sample is recovered for oxygen analysis. It is noted that the measurement of oxygen in the soil profile can be difficult and care should be taken when using this data to support biodegradation.
Halogenated hydrocarbons can also undergo biodegradation, though the process for most halogenated compounds occurs in anaerobic conditions via a number of steps that can be much slower than for the aerobic degradation of petroleum hydrocarbons. The SAQP should address potential degradation products as appropriate.
Assessing the potential for vapour intrusion to a future building on vacant land poses unique challenges. Some of the investigative tools of the vapour intrusion pathway (for example, indoor air and sub-slab sampling) are not possible when there is no slab or structure present, though others (soil, soil vapour and groundwater sampling) may be able to be used with appropriate precautions or adjustments.
As for existing buildings, a multiplication factor (x10 or x100 as appropriate) due to biodegradation may be able to be applied to relevant HSLs if the proposed maximum building size can be determined with a high degree of certainty, and the exclusion/inclusion criteria listed in Davis et al. (2009c) can be fulfilled (refer Schedule B4).
If there is significant vapour migration via preferential pathways that connect a contaminant source to a building, then the measurement of contaminant concentrations in soil vapour may not be representative of vapour concentrations that would migrate into the indoor environment. Other investigative techniques (for example, vapour measurements in utilities or indoor air measurements) may provide more representative data for the evaluation of the inhalation exposure pathway in these circumstances (API, 2005).
VOC concentrations in the environment are highly variable, and collecting sufficient data to thoroughly understand and predict their temporal and spatial distribution can be time-consuming and costly.
VOC levels in ambient air can vary greatly over time. Diurnal fluctuations occur due to changes in vehicle traffic (for example, rush-hour effects), commercial activity, and as a result of atmospheric heating and cooling cycles, air pressure changes and wind speed. These fluctuations and their impact on the data analysis can be reduced by collecting time-integrated samples. Additional information on environmental factors and their effects can be found in Davis et al. (2009a).
To compensate for these inherent uncertainties, consideration should be given to identifying and characterising the main factors that may lead to a reasonable worst-case exposure scenario. The time period selected for sample collection should be appropriate to characterise the site-specific exposure scenario.
Rates of vapour intrusion are affected by both short-term and seasonal changes in weather conditions. Changes in barometric pressure associated with the arrival of weather fronts can move gases into or out of the vadose zone. This phenomenon, known as barometric pumping, increases the rate of vapour emission as low pressure systems arrive and decreases rates when transiting to higher pressure. This effect is only of importance for soil vapour where sampling is shallow (less than 12 m). Wind can also enhance vapour intrusion rates by depressurising a building relative to the underlying soil, causing more vapours to enter the building from the sub-surface. Similarly, high volume air conditioning systems in buildings may affect vapour intrusion.
To account for wind and barometric pressure effects, consideration should be given to sampling during stable weather conditions and recording local barometric pressure and wind-speed data over the three days before and during the sampling event.
Variations in soil temperature result in the expansion and contraction of soil air, leading to partial exchange with the atmosphere. Hence vapour measurements may change daily and from season to season. However, temperature effects decrease with depth below ground and typically show minimal variation much below 1 m below ground. Temperature variations are not expected to have a large influence on soil vapour or indoor air concentrations unless the source is very close to the surface.
Soil moisture increases due to rainfall infiltration may inhibit gas exchange processes and, in particular, vapour movement towards the ground surface, and oxygen ingress from the atmosphere. An increase in moisture content decreases the air-filled porosity and results in lower vapour and gas diffusion rates in the vadose zone. This is likely to be particularly the case for heavier textured (clay) soils (Davis et al. 2009a).
Sampling of soil vapour (particularly from depths shallower than 1 m to 1.5 m) directly after significant rainfall events (greater than 25 mm) should generally be avoided, unless the rainfall is representative of normal conditions. Soil vapour samples collected from depths greater than 1.5 m are unlikely to be significantly affected by rainfall events. No specific guidance on how long to wait before sampling shallow soil vapour (shallower than 1 m to 1.5 m) after a rainfall event is given as it is dependent on the soil type and other climatic conditions.
If uncertainty remains as to the potential for a rainfall event to affect the outcome of a vapour assessment, then consideration should be given to repeat sampling and measurement of soil moisture at the time of vapour sampling.
This section provides a summary of commonly used methods for sampling and characterising soil vapour at a site and largely has been adapted from information provided in Davis et al. (2009a) and API (2005). The decision on which methods to use will be informed by consideration of the investigation objectives and analytical requirements as documented in the SAQP.
More detailed information can be found in Davis et al. (2009a), Baker et al. (2009), API (2005), NJDEP (2005a and 2005b), NYSDOH (2006), and ITRC (2007b, 2007c). Baker et al. (2009) contains discussion on the advantages and disadvantages of various sampling techniques and sampling equipment.
Sufficient sampling should be carried out to ensure that the results are representative of the site conditions and appropriate for assessing the risk to identified receptors. The following factors should be considered in the design and implementation of a soil vapour sampling program:
- Location and number of sampling points— the number of locations will depend on the CSM and access considerations. As a minimum for vapour intrusion assessment, samples should be collected above the maximum source concentration near or under a building (located within the 30 m screening distance – refer Section 9.2.1) and at each corner or along each side of the building (if practical).
- Depths — the depth of samples should be based on the CSM and take into consideration the depth of sub-surface sources, the nature of the contamination and the likely migration pathway(s). In most cases sampling should be undertaken at depths >1 m to avoid transient effects. Where shallow sources are present or where deep samples cannot be obtained, the collection of soil vapour from shallow depths (<1 m) may be appropriate; however, sampling from these depths requires justification. When installing sample equipment at (or using data from) shallow depths, the potential for aerobic degradation and potential transient influences should be considered.
- Frequency —Multiple sampling events are generally required to characterise and assess vapour risk particularly where (i) the data is close to (compared with the likely variation in soil vapour concentration and the precision and accuracy of the data) or above guideline values, (ii) if samples were collected from shallow depths (<1 m), and/or (iii) seasonal variations in temperature and/or soil moisture and the effects on soil vapour concentrations are not fully understood.
Active soil vapour samples may be taken from probes installed in open ground or recovered via access holes drilled through sealed surfaces such as a driveway or parking area (’near-slab‘) or beneath building foundations (‘sub-slab‘). Sampling installations may be permanent, semi-permanent or temporary depending on access and the need to re-sample. The basic sampling approaches for soil vapour sampling include:
- point samples at specific depths in one or more lateral locations and
- vertical profiles of samples at two or more depths at one or more lateral locations (‘transects’).
Additional factors for consideration when designing and implementing an active soil vapour sampling program are discussed in Section 9.4.2.4.
Spear probing (driven soil vapour probes) of soil involves driving a spear/rod into the ground to a shallow depth (for example, 1.5 m–2.0 m below ground surface), extracting a soil vapour sample for analysis of the vapours of concern and/or major gases (for example, oxygen), and withdrawal of the spear probe. The reliability of the results may be improved by using bentonite slurry to seal the area around the drive point and conducting a leak test prior to sampling. Samples should be recovered below the zone influenced by transient effects, which is likely to extend to 1 m or greater below the surface. The probe should be decontaminated before using at the next location.
Spear probing is generally used as a screening tool (as it permits a large number of locations to be sampled in a cost-effective manner) to inform a more detailed investigation of identified areas of interest.
The method can be used to collect samples from a vertical profile (from as shallow as 0.3 m) to assist in the identification of various vapour zones and to define the potential aerobic reaction zone. As a quantitative technique, spear probes can be installed and sampled in the same manner as permanent probes/samplers, however, it has the disadvantage that the results are ‘snapshots’ which cannot be later repeated.
Additional considerations associated with the sampling of soil vapour are noted in Section 9.4.4 in this Schedule.
The installation of permanent multi-level probes/samplers for soil vapour measurement permits:
- depth profiling of vapour concentrations through the soil profile from near source to near the ground surface
- repeat sampling and monitoring over time at fixed locations.
Single depth permanent probes can either be installed at depth (close to the source) or in the shallow sub-surface (particularly where the source is shallow). Multiple depth (or multi-level) gas sampling installations may be undertaken by installing multiple sample ports at different depths (separated by a bentonite seal) within the one sampling well (API 2005; Hartman 2002), or installing separate soil vapour probes at different depths (separated by at least 0.6 m) (API 2005; NYSDOH 2006).
There is a range of methods available for installing permanent probes. The probe installation method used should be determined based on site-specific factors such as access and environmental conditions (for example, soil texture or moisture conditions that may limit the use of very narrow tubing).
A log of soil types encountered during drilling should be documented. To assist in the assessment, a soil core may also be recovered and subsampled to determine organic carbon and soil parameters (for example, bulk density, porosity and soil moisture content) at the depths of the sampling ports, and for analysis of the soil for the chemicals of concern.
Correct sealing of the installations is essential, especially in low permeability soils. Separate installations rather than multi-level samplers may be necessary in low permeability soils to ensure a good seal is able to be achieved or where the upper sample is less than 1 m below the surface.
Sampling of permanent probes and multi-level samplers can be carried out using a range of sampling methods. Typically, permanent probes should be left for a minimum of 24 to 48 hours to equilibrate prior to sampling (DTSC 2009; NYSDOH 2006), depending on the installation method and the site conditions.
Near-continuous measurement of total vapour (or VOCs) and oxygen concentrations is possible using online VOC and oxygen probes (Patterson & Davis 2008; Patterson et al. 1999, 2000). These can be buried at multiple depths to give near-continuous measurements of total vapour and oxygen concentrations for extended periods (months to years). Apart from providing vapour and oxygen depth profiles, the detailed information derived from these probes allows seasonal trends in vapour fluxes and other parameters such as degradation rates to be assessed.
Online VOC probes (at the time of drafting), do not directly monitor individual compounds such as benzene, but can be subsampled to obtain a gas sample which can then be analysed by conventional means for component VOC and major gas concentrations.
The following factors require consideration in the design and implementation of an active soil vapour sampling program:
- Probe integrity/seal — soil vapour probes (temporary and permanent) should be installed in a manner that ensures that ambient air is not drawn into the sampling system and that a representative soil vapour sample can be collected. This may require an additional seal around the probe using bentonite slurry and leak testing, even for temporary installations. An effective seal is particularly important in low permeability soils and for shallow probes or sub-slab probes. A number of tracer methods are available to test seal integrity see API (2005) and ITRC (2007b).
- Tubing type — the tubing type should be selected to minimise false positives due to outgassing from the tubing materials. Low sorbent materials such as HDPE and nylon are generally preferred. Further information on material properties may be obtained from suppliers or manufacturers.
- Sample volume — sample volumes should be minimised as far as practicable to meet the requirements of the sampling/analytical method selected. A review of available studies on sample volumes by Hartman (2006) suggested that the sample volume is less important for coarse-grained soil, but in finer grained soils large volumes may be difficult to collect due to the creation of a vacuum during sampling. Large sample volumes increase the likelihood that the sample may originate from different depths and locations, hence sample volumes collected should be minimised. Near ground surface, recovering large sample volumes may result in ambient air being drawn from outside the annulus of the shaft of the probe.
- Purge volumes — the sample probe, tubing and equipment have an internal volume that must be purged prior to sampling to ensure that only soil vapour is sampled and that the data obtained is representative. Generally, three to four system volumes should be purged where flow rates allow and as long as the purge volume is not large. Whatever calculation is used to estimate the volume purged, this should be consistent for all sample locations. As large purge volumes can result in low pressure/vacuum conditions which may cause contaminant partitioning from the soil to soil vapour, the purge volume should be minimised as far as practicable to ensure that the sample collected is representative. Real-time gas monitoring (using a landfill gas meter to measure oxygen, carbon dioxide and methane or a PID to measure total VOCs) can be useful for assessing the effectiveness of purging (and decrease the purge volume to less than three to four system volumes).
- Sample flow rates — to minimise the potential for desorption of contaminants from soil to soil vapour in the sampling zone, the assessor should select a sample flow rate appropriate for the soil type. A low sample flow rate (<0.2 L/min) is important where soil vapour is collected from low permeability soil (McAlary et al. 2009); however, higher flow rates may be applicable for coarse-grained soils. Low permeability or high moisture content can induce greater suction pressures when sampling, which can make samples difficult to recover.
- Equilibration time after installation — the equilibration times for soil vapour sampling is highly dependent on the drilling method. Direct push methods cause minimal disturbance to soil vapour profiles and sampling may be carried out after 30 minutes, whereas 48 hours is recommended for augered installations (API, 2005).
Source: Davis et al. (2009a)
‘Passive’ soil vapour sampling or passive implant sampling refers to the burial or placement of an adsorbent or other material in the ground, which is recovered for analysis after an appropriate period of time (hours to days). It is termed passive because no gas sample is actively recovered from the soil profile. The adsorbed mass cannot be equated to a concentration because the volume of air associated with the adsorbed mass is largely unknown.
The method enables a screening level assessment of the presence of vapours in the vadose zone to identify if the vapour pathway is complete and to identify hotspot areas for further sampling using more quantitative methods. Passive samplers may be of benefit in areas where soil vapour probes cannot be installed, in areas where preferential pathways are suspected (or need to be assessed) or where very low permeability soils limit the practicality and integrity of sampling from soil vapour probes (API, 2005).
Since the sorbent can be deployed for long periods (typically 3 to 14 days), this concentrates the mass of contaminants absorbed to the sampler and enhances sensitivity. Longer exposure time does not improve sensitivity except during prolonged rain events which cause soil saturation and interrupt vapour migration in the subsurface (NJDEP, 2005a).
Passive samplers may desorb soil vapours from fine-grained layers that are otherwise not mobile, thus overestimating the amount of soil vapours that are capable of being transported into overlying zones.
More detailed information can be found in Davis et al. (2009a), Baker et al. (2009), ITRC (2007a and 2007b) and NJDEP (2005a and 2005b).
Source: Davis et al. (2009a)
A flux chamber (or a flux hood) is a device that is placed on a surface to measure vapour/gas flux (or emission rate) discharging through that surface. The surface may be open ground or be part of a building foundation such as a concrete slab.
Flux chamber methods have generally not been widely used in site assessment or considered a primary vapour intrusion assessment method due to a number of limitations and disadvantages, which are discussed in Davis et al. (2009a) and Baker et al. (2009). However, flux chamber methods may be applicable when a direct measurement of vapour flux is required and as an additional line of evidence in combination with other methods.
The technique enables direct measurement of vapour flux from the surface of the ground or building foundation, thus providing a direct estimate of the parameter of interest (rather than calculating it from sub-surface vapour distributions). Flux methods effectively integrate all sub-surface processes (for example, phase partitioning, biodegradation, preferential pathways, advective and diffusive transport), often close to the point of potential exposure.
There are two primary types of flux chamber methods: a static (closed) chamber method, and a dynamic chamber method.
The static chamber method requires the placement of the flux chamber on the surface of the ground or building foundation, excluding passage of air through the chamber. This allows vapours to be trapped and the stagnant chamber vapour concentration to build up over time. Active samples can be collected at discrete intervals through a time period and at the end of a time period.
The dynamic chamber method involves the use of an inert sweep gas which is continually introduced into the chamber with an equivalent amount of gas allowed to escape. The system is allowed to reach steady-state, (assumed to be four or five chamber volumes) before the chamber is sampled. The sample can be a discrete sample or monitored continuously.
When designing a flux chamber sampling program the following should be considered:
- Coverage of the area of concern —adequate coverage of possible vapour conduits, areas of maximum source concentrations and consideration of other site-specific building features as required
- Deployment period — this should be adequate to address the issues of concern and, where possible, enable temporal variability to be assessed.
- Basements — flux chambers may not be suitable for dwellings with basements because of additional potential fluxes from the basement walls to the interior of the dwelling.
- Sub-surface conditions — flux monitoring provides little information about the processes that may be occurring within the vadose zone such as oxygen penetration and hydrocarbon degradation. Longer-term controls on emissions and hence potential changes in sub-surface conditions may not be detected with such a device, unless long-term near-continuous emission monitoring is undertaken.
- Buildings — because of the usually limited surface area of coverage, flux chambers may not measure the actual flux into a built structure, especially if there is preferential access to the structure. Also, air-movement conditions within the chamber may not reflect natural room conditions in a structure – leading to overestimation or underestimation of fluxes depending on relative pressure differentials inside and outside a chamber.
More detailed information can be found in Davis et al. (2009a), Baker et al. (2009) and Hartman (2003).
Source: Davis et al. (2009a)
Samples for analysis may be collected using a range of media which include sorbent tubes (charcoal or multisorbent), Summa canisters, Tedlar® bags, glass vials, and syringes. Standard operating procedures should be developed for sample collection and any variations to the procedure (for site conditions or equipment limitations) fully documented.
It may be possible to carry out a field screening assessment of the contaminants present using a PID, FID or other handheld detector, providing the instrument detection limits are sensitive enough to measure concentrations at levels relevant to health risk assessments. For field screening, soil vapour samples may be collected via a vacuum (evacuation) chamber into a Tedlar® bag or similar and the meter connected directly to the Tedlar® bag.
Commonly used active and passive collection methods are discussed in the following sections, while more detailed information can be found in Davis et al. (2009a).
Sorbent materials, packed into tubes, typically comprise activated carbon and/or a range of multi-sorbent materials (one or more different sorbent media may be present in each sample tube). Vapour samples are collected by drawing air (using pumps) at a calibrated rate through the tube over a specified period of time. The flow rate and sampling volume are dependent on the sorbent media used, the range of target chemicals and the required limit of reporting. The reporting limit is determined by the volume of air drawn through the sample tube, the adsorbent and analytical method used, and the potential for high concentrations (requiring dilution of the sample during analysis).
Sorbent tubes have a maximum capacity which may be exceeded in circumstances where the source concentrations are high and/or the sample volume drawn through the tube is large. A control section of the tube, analysed separately to the sample section, indicates whether breakthrough has occurred (i.e. whether the capacity of the tube has been exceeded). If the tube capacity is exceeded, the reported concentration will under-represent the actual site conditions.
Whole air samples can be collected using specially prepared canisters (Summa canisters) which are sent to the field under vacuum and certified clean and leak-free. The canister is fitted with a calibrated regulator that, when opened, allows air to be drawn into the canister over a pre-set time period at a constant flow rate. Initial and final vacuums are recorded for each canister, as well as the vacuum when received at the laboratory.
Whole air samples can also be collected using Tedlar® bags or syringes and glass vials.
Passive methods can involve the use of a wide range of sorbent materials. These materials are available in a range of forms (badges, canisters, tubes, strips) where the collection of compounds is based on the diffusion of the compound to the surface of the sorbent material. Other samplers/systems are also available and can be used depending on the target analytes, required use and reporting limits.
The range of compounds that are commonly analysed with passive sorbents include petroleum hydrocarbons, chlorinated hydrocarbons, ammonia, aldehydes, phenols and creosols, hydrogen chloride, hydrogen fluoride and hydrogen sulphide. Other compounds can be targeted using passive methods depending on the sorbent materials/housings used, the ability to assess uptake (diffusion) rates, and analysis methods. The limit of reporting varies depending on the sampler (sorbent material) used, the analysis method and the sample time.
Passive methods are generally considered to provide a qualitative measure of concentration; however, quantitative results may be obtained under certain conditions. This is dependent on the concentration present in air, the time sampled and, for some samplers, the movement of air past the sampler. The concentration is calculated based on diffusion principles (uptake rates). The reliability of the results should be assessed as part of the DQO process.
The use of a passive sampling system, selection of appropriate sampler (to adequately address the range of compounds required), sampling time and analysis method should be considered in the design of the sampling plan.
The analytical method(s) selected should be considered with respect to the target compounds, DQOs, the availability of analysis, and the advantages/disadvantages of each method.
Ambient air and soil vapour samples are generally analysed using methods sourced from the US EPA’s Compendium of methods for the determination of toxic organic compounds in ambient air (TO-methods). Site assessors may wish to use alternative methods, in which case the alternative method should be at least as rigorous and reliable as the TO-methods. For further information on reference methods and alternative methods, see Schedule B3.
Source: EA (2000a)
Risk assessments undertaken when groundwater or soil vapour contamination is present may involve the use of quantitative contaminant fate and transport models. Specific expertise and experience are required to carry out this type of modelling because of the highly complex nature of most contaminant fate and transport problems.
In the context of this guidance, a model is defined as a mathematical representation of reality in the form of equations or computer code and values of parameters. Output from this type of modelling may include travel times to receptors and concentrations of contaminants likely to reach receptors.
A model should only be used when it is clear how and why it is to be used. In deciding whether a modelling approach is appropriate, some of the questions that need to be considered are:
- What is the objective for modelling and what are its benefits? For example, a model may help in the decision-making process by quantifying the potential impact on a receptor and therefore the need to take action to protect the receptor.
- Can a model provide reliable answers? For example, the hydrogeological system may be too complex to be adequately represented by the available modelling resources, in which case the application of a model would serve no purpose.
- Is the hydrogeological system sufficiently understood to warrant the use of a model? A model should not be used as an alternative to collecting further site-specific information; however, it may be used to guide further data collection.
If the decision is taken to use a model, then the limitations and assumptions of the model selected should be assessed to determine whether it is fit for the selected purpose.
Modelling is unlikely to be appropriate where preferential migration pathways are present. These pathways may be natural features; for example, solution channels associated with karst development in limestones, weathered shear zones, and permeable geological faults, or anthropogenic in origin; backfill around foundations, backfill in trenching for buried utilities such as sewer, water, gas and electricity lines, and backfill around buried tanks and associated piping.
The key stages in developing a contaminant fate and transport model are:
- scoping study, comprising a review of existing information and consultation with relevant stakeholders to define the objectives of the study and the scope of work
- development of a CSM of the saturated and unsaturated zones and consideration of how the contaminant fate and transport processes can be represented in a model
- selection of an appropriate model based on the objectives of the study, the CSM and data availability
- construction/application of the model and comparison of model results with field data to assess model validity
- sensitivity analysis to determine which parameters have the most significant influences on the model results
- uncertainty analysis to take account of uncertainty in the conceptual model, parameter measurement and natural variability of parameters
- assessment of results and reporting, including assumptions and limitations.
The development of the CSM is a critical step and should identify and consider the relevant aspects of the flow system and the contaminant transport processes likely to be operating. In constructing the CSM, a number of assumptions regarding the system behaviour will need to be made. The assessment should consider whether the assumptions and uncertainties are important, that is, whether it is possible to adopt a relatively simple mathematical model of contaminant transport or, alternatively, whether understanding and definition of the system behaviour is so poor that development/use of a mathematical model is inappropriate, and that the first priority should be to obtain further site-specific information.
A phased approach to using mathematical models is recommended, moving from simple calculations to analytical models and, finally, to numerical models if appropriate. The quality and quantity of the data available should be taken into account when selecting the mathematical model. Where data is limited, complex models are generally not appropriate. In each case the selection of the modelling approach should be justified and appropriate to the available data and understanding of the system behaviour.
Data collection should be an iterative process and linked to the development and refinement of the CSM and the mathematical model. Site-specific data should be obtained whenever possible and, for certain parameters, site-specific data is essential. Literature values may need to be used for some parameters, and the values selected will need to be justified.
Construction/application of a model using parameter data will generally involve a calibration step whereby the model parameters are adjusted within a credible range to achieve the best fit between model results and field data. If an acceptable fit cannot be obtained in this calibration step, the appropriateness of the model, the need for further site-specific data, and the CSM should be reviewed.
When completed, the model should be fully documented, including the objectives of the model, the model code used and its limitations, description of the conceptual model including all parameters used and any assumptions made, how the model was constructed and calibrated, and information on the accuracy of its predictions.
The quality and reliability of contaminant transport model results are dependent on the data that has been used to develop the conceptual model and to construct and refine the mathematical model. If the data is inadequate, the model results will be unreliable.
Data requirements vary at different stages in the modelling process but are dependent on the objectives, the complexity of the problem and the sophistication of the analysis. The assessor will need to determine the key parameters for which site-specific data is required and those parameters for which literature values will be acceptable. The ease of collection and relative cost of obtaining site-specific values for flow and transport parameters are summarised in Table 6 below.
In some instances, the collection of site-specific hydrogeological data may not be possible; in which case, reasonable default values should be selected based on geological records.
Rigorous scrutiny should be applied to ensure that input parameters are consistent with the geology, hydrogeology and geochemistry of the site or region modelled. Special care should be taken to ensure that values for hydraulic conductivity, contaminant load and degradation rates (if applicable) are appropriate, and that conclusions drawn on the basis of fate and transport modelling are supported by the available monitoring data.
Problems can arise at different stages of contaminant fate and transport modelling due to:
- poor sampling and analysis
- inadequate CSM
- inappropriate model selection
- use of inappropriate data sources (literature)
- (mis)interpretation/use of results.
Further information on generic good practice to avoid these problems can be found in EA (2000a), Middlemis (2000) and Barnett et al. (2012).
The two main types of mathematical model are analytical models and numerical models.
Analytical models use exact solutions to equations that describe the migration of contaminants. In order to produce these exact solutions, the flow/transport equations have to be considerably simplified such that they are typically only applicable to simple flow and contaminant transport systems. Analytical models can be simple formulae, spreadsheets or sequences of calculations packaged up in a piece of software, for example, BIOSCREEN and BIOCHLOR from US EPA.
Numerical models use approximate numerical solutions to the governing equations of groundwater flow and transport. Parameter values are specified at certain points in space and time and provide a more realistic representation of the variation of parameters than is possible with analytical models. Numerical models range from relatively simple one-dimensional steady-state transport models to three-dimensional time-variant models, for example, MODFLOW from the US Geological Survey (USGS), and may consider any or all of advection, dispersion and retardation, biodegradation, multiphase flow and density-driven flow.
A summary of 3-D groundwater modelling codes can be found in Middlemis (2000). When considering using models, advice should be sought from suitably experienced persons in hydrogeology and geochemistry and the application of such models. Comprehensive information and software is available from:
- US EPA Centre for Subsurface Modelling Support
- www.epa.gov/ada/csmos/index.html#download
- USGS water.usgs.gov/software/lists/groundwater/.
A comprehensive software catalogue of a wide range of models, which includes information on the advantages and disadvantages of each type, is maintained by the
International Groundwater Modelling Centre at the Colorado School of Mines (www.mines.edu/igwmc/).
Table 6. Summary of site-specific data requirements for contaminant fate and transport modelling
Parameter | Site-specific data essential | Site-specific data useful | Comments on ease and cost of obtaining site-specific data |
Aquifer depth/geology | | | Easy to obtain but data quality, reliability and cost depends on site-investigation techniques used. |
Hydraulic conductivity | | | Relatively easy to obtain but data quality depends on method used. Pump tests provide the best data but can be expensive, particularly where contaminated water needs to be managed. |
Hydraulic gradient and direction of groundwater flow/ seasonal variability | | | Relatively easy to obtain but data quality, reliability and cost depend on number and construction of boreholes and frequency of measurement. |
Porosity | | | Intergranular porosity is inexpensive and easy to measure. Generally difficult to measure in fractured aquifers due to factors such as the presence of preferential flow paths e.g. fractures, joints, faults or caverns. |
Transport porosity | | | Difficult to measure – requires tracer test. |
Bulk density | | | Inexpensive and easy to measure. |
Partition coefficient (Kd) | for inorganics | | Generally inexpensive and easy to measure but data quality, reliability and cost will depend on methods used. |
Cation exchange capacity (CEC) | | for in-organics | Inexpensive and easy to measure. |
Moisture content of unsaturated zone | | | Inexpensive and easy to measure. |
Total organic carbon (TOC) | | for organics | Inexpensive and easy to measure. Representative data can be difficult to obtain in low organic carbon aquifers, in which case, reasonable default values should be selected based on geological records. |
Infiltration | | | Meteorological data is easy and relatively inexpensive to obtain. |
Degradation | (not for metals) | | Relatively difficult and expensive to measure and requires long-term monitoring but is essential to provide confidence in outcomes. |
Contaminant concentrations | | | Cost dependent on analytical suite and number of samples. |
Redox conditions | | | Cost dependent on analytical suite and number of samples (DO, pH and redox inexpensive and easy to measure). |
Adapted from EA (2000a)
The recommended general process for assessment of site contamination, including for assessment of asbestos, is shown in Schedule A to the NEPM. The process starts with a Preliminary Site Investigation (PSI), which may lead to a Detailed Site Investigation (DSI) and/or an appropriate management strategy if required. Where remediation is required, appropriate validation should be carried out to verify the effectiveness of the measures undertaken. All soil asbestos investigation and management work should be conducted by a competent person.
A competent person in the context of asbestos and the NEPM is a person who has acquired, through training or experience and qualification, the knowledge and skills to identify, investigate and assess asbestos in the context of an environmental site assessment. This includes identifying the potential for asbestos contamination from site history information.
The site-specific assessment of sites contaminated by asbestos in soil should be aimed at describing the nature and quantity of asbestos present in sufficient detail to enable a site management plan to be developed for the current and/or proposed land use as relevant for the site. The site management plan should consider what action would be necessary in circumstances where asbestos fibres could become airborne and pose a human health risk.
This guidance is designed to be used in combination with the guidance on asbestos in Schedule B1 and with reference to Guidelines for the Assessment, Remediation and Management of Asbestos-Contaminated Sites in Western Australia published by the Western Australia Departments of Health in 2009 (WA DoH 2009a). The latter and related publications on asbestos, including a summary of the guidelines, which is updated annually, may be downloaded from
http://www.public.health.wa.gov.au/3/1144/2/contaminated_sites.pm.
The types of asbestos referred to in this guidance include:
- bonded ACM
- fibrous asbestos, FA
- asbestos fines, AF.
These terms are defined in Schedule B1 Section 4.4.
As for all site assessments, the PSI should include a desktop study (including assessment of site history) and a site inspection. This should be carried out by a qualified and experienced assessor/competent person.
As noted in WA DoH (2009a), asbestos contamination needs to be identified early in the assessment process and properly handled to ensure that disturbance does not result in dissemination of asbestos contamination and hence delays and additional investigation effort.
The site history investigation should follow the process outlined in Section 3 and include the following asbestos-specific considerations:
- an evaluation of information, including inspection of aerial photographs, to determine the likely presence of asbestos associated with
any remaining or demolished structures and buildings (including footprints) particularly of pre-1990 construction
possible disposal, burial and dumping activities
- an evaluation of information relating to fill materials on-site, particularly if they may incorporate demolition waste.
The site inspection should include a comprehensive assessment based on a grid-based walkover by a qualified and experienced assessor/competent person to determine whether visual indications of asbestos contamination are present. The assessment report should specifically comment on the presence or absence of asbestos material and the inspection method employed.
The identified areas should be surveyed in more detail (noting condition and distribution) together with any suspect locations identified as a result of the site history investigations. After noting the size and condition of fragments, all visible asbestos should be removed.
Where the site is thickly vegetated, then confidence in the visual inspection results will be lower. Where appropriate, some careful vegetation clearance may clarify the situation.
The default assumption by the assessor should be that any suspect material contains asbestos and further investigation/appropriate management action initiated.
Where confirmation is required regarding the nature of suspect material, laboratory analysis is required. This should be undertaken by a National Association of Testing Authorities Australia (NATA) (or its mutual recognition agreement partners)-accredited laboratory in accordance with Australian Standard AS 4964 – 2004: Method for the qualitative identification of asbestos in bulk samples.
Soil contamination by free asbestos fibres may be deduced from the site history and, if suspect material is identified in the site walkover, confirmation of the presence or absence of fibres may be determined according to AS 4964-2004. Where significant amounts of free asbestos fibres may have been exposed over time, the immediate surrounding area should also be considered contaminated. If free fibres are detected, the focus should then be on management, as there is yet no validated method of reliably estimating the concentration of free asbestos fibres in soil.
A video and/or photographic log may assist with site documentation.
Sampling during a PSI is generally not recommended, since either an asbestos management strategy may be adequately defined without it or because it is evident that a DSI will be necessary. Limited sampling during the PSI may be appropriate, however, in the following circumstances:
- to confirm that suspect material contains asbestos
- to roughly delineate the extent of bonded asbestos-containing-material (bonded ACM) contamination in surface soil or fill
- to inform the sampling and analysis plan for a DSI
- to inform consideration of appropriate management options.
Issues that should be considered during preliminary assessment are described below.
Bonded ACM fragments are often present as surface deposits on sites due to poor demolition and building practices. While isolated fragments in good condition across the surface of a site are usually of low concern, surface material may present an exposure risk to airborne fibres over time from deterioration of the bonding compound through corrosive weathering, abrasion or crushing by vehicle traffic and other activities.
Bonded ACM may be able to be easily broken by hand force and be more readily crumbled when water-saturated or corroded. In a partially crumbled state, bonded ACM may be of greater concern, particularly if it is exposed at the surface and susceptible to abrasion during land use. In particular, roofing material containing asbestos may be heavily weathered, which can corrode the cement matrix and expose fibrous asbestos to the atmosphere.
Bonded ACM that can be easily crushed by hand should be considered friable and assessed for management actions accordingly. Similarly, unbonded asbestos or fibrous asbestos (FA), including loose material such as insulation products and damaged low density board (up to 70% asbestos in calcium silicate), are considered friable.
The condition of asbestos materials should be considered equivalent to the most degraded samples found in the relevant assessment area.
Generally accepted guidance for considerations of site setting and characteristics should be applied when developing the scope of an investigation and when developing management strategies and cleanup methods. Any potential for exposure of bonded ACM to an acid generating environment may be a factor that will increase the potential for release of fibres from the bonded matrix. Many Australian soils are weakly acidic, however, some sites may contain acid sulfate soils or other acidic soil conditions that can lead to faster rates of degradation. The clay and moisture content of soils is also a consideration, as these factors tend to inhibit the release of fibres by binding and damping mechanisms.
A DSI may not be necessary although this will depend on the site-specific circumstances and the proposed remediation approach. Conservative management of presumed asbestos contamination may avoid the need for a DSI. The circumstances where a DSI would be necessary include when:
- the remediation or management approach requires asbestos contamination to be fully delineated and assessed (e.g. asbestos contamination is to be relocated and contained on-site)
- land uses are to be determined and delineated according to the extent and nature of the asbestos contamination.
A DSI may also resolve uncertain findings from the PSI, or assist in assessing the likely effectiveness of alternative remediation and management strategies.
A DSI is not necessary where there is a high degree of confidence that the asbestos contamination is confined to bonded ACM in superficial soil, i.e. the site history can be established with confidence and this clearly indicates that there is no reason to suspect buried asbestos materials and the site inspection confirms that any bonded ACM is in sound condition and only present on the surface/near surface of the site. In these circumstances the assessment can proceed directly to remediation (removal of bonded ACM fragments and ensuring that the soil surface is free of visible asbestos) and validation. However, investigation will be required if the soils at the site have been disturbed and potential asbestos-contaminated-material moved around the site or incorporated into sub-surface soils.
Unnecessary investigations should be avoided, for example, investigation for bonded ACM is not recommended below the proposed deepest excavation level during construction or likely maximum depth of disturbance for the proposed/current land use.
If a DSI is undertaken, a sampling and analysis quality plan (SAQP) is required to support and inform the investigations. A site management plan, including dust management and airborne fibre monitoring, may also be required to protect the public and workers during investigation works and earthmoving/development works. Asbestos fibre and dust (as a surrogate for asbestos fibre) are of particular interest. Dust management measures should be adopted to ensure that airborne fibres remain below 0.01f/ml, which is the practical lower detection limit of the membrane filter method (enHealth 2005). Any dust-related asbestos fibre analysis should be undertaken by a NATA-accredited laboratory. For further information and guidance on dust and airborne fibre monitoring refer to WA DoH (2009a) and enHealth (2005) and relevant work, health and safety guidance such as Section 3.11 of Safe Work Australia (2011b).
The SAQP should include an appropriate CSM and DQOs based on knowledge of the site history and the continuing use and/or future use of the site as relevant. Sampling may include both large area (hand-picking/raking, mechanical screening and tilling) and localised methods (test pits, trenches and boreholes) to delineate lateral and vertical extent (refer Table 7). All methods usually start with handpicking to ensure that the site surface is free from visible asbestos material.
With regards to reliability of findings, test pits and trenches are preferred to boreholes to determine the presence or extent of any asbestos contamination, because a larger area of the subsurface is exposed during assessment and is available for visual inspection. It is therefore recommended that the SAQP places greater reliance on judgmental sampling involving test pits and trenches based on a thorough site history, rather than boreholes. Appropriately designed judgmental sampling plans can help avoid unnecessary broad area sampling. Grid sampling, however, is appropriate when asbestos contamination is widespread or is of unknown extent/location(s) at a site.
The sampling density and field procedures should be sufficient to characterise the nature and extent of contamination and to enable an appropriate management plan to be developed.
Table 7 Sampling methods for evaluating asbestos contamination
Sampling method | Suitable for | Limitations |
Hand-picking (emu-bob)/raking - can use rake to sample down to about 10 cm
- at least two passes with 90º direction change
- hand-picking can be used with care to remove surface FA material (assessment of likely free fibre release associated with it required)
- % contamination calculated using 1 cm as soil depth (for hand-picking surveys) or rake teeth length (for raking) as appropriate
- Final visual inspection should not detect visible asbestos
| - bonded ACM and low levels of FA
- surface or near-surface contamination
- characterising the extent and level of contamination while reducing bonded ACM impact
| - raking may only be effective in sandy soils
- reduced confidence for vegetated or debris-covered areas
- not suitable for deeper contamination (>10 cm)
|
Tilling (mechanical turning over of soil) with manual collection - pre-wet soils to control dust
- at least two passes with 90º direction change
- material should not be further damaged or buried by the process
- rotor blade speed should be controlled to allow spotters to hand-pick revealed fragments
- conducted across the entire area of suspected impact
- % contamination calculated using an estimate of the average tilled depth per grid square
- final visual inspection should not detect visible asbestos
| - bonded ACM only
- contamination to about 30 cm depth depending on rotor blade size
- characterising the extent and level of contamination while reducing bonded ACM impact
| - not for fibre-generating materials
- limited application for deeper contamination (> 30 cm) or areas obscured by surface vegetation or debris
- evaluated areas cannot usually be considered representative of other locations
|
Mechanical screening - conducted across the entire area of suspected impact, large areas should be sub-divided for assessment
- large pieces may be removed by larger mesh sizes prior to final 7 mm effective mesh size
- % contamination calculated using weight of bonded ACM found per given volume of soil screened from each strata/location
- if percentage of small fragments is high, sampling of resulting screened stockpiles may be appropriate to ensure effectiveness of screening procedure
- location of screened material should be carefully documented to permit follow-up sampling or segregation if required
- final visual inspection of soil surface and segregated stockpiles should not detect visible asbestos
| - minor bonded ACM
- larger volumes of reasonably accessible and delineated soil contamination
- characterising the extent and level of contamination while reducing bonded ACM impact
| - not for fibre-generating materials or high levels of contamination
- may not be suitable for compacted soils or soils with high clay content
- evaluated areas cannot usually be considered representative of other locations
- requires extensive management procedures to monitor and control dust and fibres
|
Test pits and trenches - sampling to 30 cm below likely limit of contamination or to likely maximum depth of soil disturbance
- suspect asbestos materials and construction debris should be targeted
| - all types of asbestos contamination but particularly bonded ACM, and FA if fibre disturbance is manageable
- asbestos contamination extends below surface soils (about 30 cm)
- if contamination is buried and of unknown location and depth
| - shoring of exposed faces may be required to protect workers and the public from wall collapse or open excavation hazards and potential fibre release during excavation/sampling
|
Boreholes - sampling to 30 cm below likely limit of contamination or to likely maximum depth of soil disturbance
- suspect asbestos materials and construction debris should be targeted
- core diameter should be at least 15 cm
| - all types of asbestos contamination
- if contamination is buried and of unknown location and depth
- fibre control may be more successful than for test pits and trenches
| - smaller area/volume available for visual examination compared with test pits and trenches
- measures may be required to protect workers and the public from potential fibre release during drilling/sampling
|
Adapted from WA DoH (2009)
Careful documentation of the sampling process and rationale is essential to the assessment of the findings. A summary of the findings annotated on a suitable site inspection plan can be helpful. Documentation should include:
- nature and condition of bonded ACM fragments and whether associated with potential free fibres
- range and average size of fragments in each affected unit
- description of affected fill/soil unit(s).
- location and depth of samples taken for analysis
- location/direction of photographs/videos
- relevant details of equipment/machinery used (e.g. rake teeth length, rotor blade size, screen/sieve size(s).
Bonded ACM is the most common and the most readily quantifiable form of asbestos soil contamination due to its ease of visual detection. Gravimetric assessment of bonded ACM is the recommended measure for total asbestos contamination where FA and AF (derived from bonded ACM only) are not likely to be significant as established by the PSI including the site inspection (as a guide, this may be taken to be where FA and AF together are likely to make up less than 10% of the total amount of asbestos present).
For sites contaminated with bonded ACM only (i.e. no insulation materials or other non-bonded asbestos products), assessment for free fibres is only warranted where greater than 10% of the total bonded ACM is significantly damaged i.e. present as small pieces less than 7 mm x 7 mm or can be crushed/crumbled with hand pressure.
Guidance on calculating asbestos concentration in soil is provided in Section 4.10 in Schedule B1.
For large area methods (hand-picking/raking, tilling and mechanical screening), the weight of bonded ACM and its condition should be recorded for each grid area or location evaluated. A grid area of up to 10 m x 10 m is generally reasonable when large surface areas are impacted; however, non-impacted soils should be excluded from calculations to avoid dilution effects.
For localised methods (test pits, trenches and boreholes) samples should be taken from each relevant stratum in one wall of test pits and trenches or each relevant stratum for boreholes at each sampling location, and additional samples from suspect spots.
Bonded ACM and FA samples from test pits, trenches and boreholes
- sampling should be conducted to 30 cm below the likely limit of potential contamination or to the likely maximum depth of disturbance (large sites may be split into sub-areas for sampling purposes, the rationale should be included in the assessment report)
- at least one minimum 10 L sample from each relevant stratum (or per 1 m depth for thick units) for test pits, trenches and core from boreholes and additional samples from any suspect spots
- individual samples (minimum of 10 L) should be manually screened on-site through a 7 mm sieve and the material retained on the sieve examined for any bonded ACM and/or suspect material
- for heavy soils (e.g. clay soils), the samples may need to be gently disaggregated by washing and spread out on a suitable contrasting colour material to determine the amount of bonded ACM present
- if visible FA material is present or suspected, the soil should be wetted to minimise the release of fibres and the sample spread out for inspection on a contrasting colour material, to identify suspect material
- identified bonded ACM and FA (and suspect materials assumed to contain asbestos) should be weighed for each sample and documented to assist with calculating asbestos soil concentration as described in Schedule B1
- if suspect materials are found (which are suspected of containing asbestos), representative samples (e.g. 1 in 10 of similar materials, the number analysed should take into account the variation in appearance and form) should be forwarded for laboratory analysis in accordance with AS 4964-2004. Alternatively the suspect materials may be assumed to contain asbestos.
AF-related sampling
Quantification of bonded ACM may be used as a surrogate measure for AF in certain circumstances, refer to the discussion at the start of this Section.
If sampling for AF is necessary:
- sampling should be conducted to 30 cm below the likely limit of potential contamination or to the likely maximum depth of disturbance (large sites may be split into sub-areas for sampling purposes, the rationale should be included in the assessment report)
- at least one wetted 500 ml sample from each relevant stratum (or per 1 m depth for thick units) for test pits, trenches and core from boreholes and additional samples from suspect spots should be submitted for laboratory analysis. The rationale for this sample size is discussed in Section 4.10 of Schedule B1.
Additional information on recommended practice for carrying out gravimetric analysis can be found in WA DoH (2009a) and the annual summary/update of the guidance document (WA DoH 2012) available from the WA DoH website:
http://www.public.health.wa.gov.au/3/1144/2/contaminated_sites.pm.
Laboratory analysis for dioxins is only recommended when the site history clearly indicates that dioxins are likely to be present as a by-product resulting from specific manufacturing and industrial activities, or from waste disposal. Dioxin contamination may be present following long-term and large-scale use of a site for the following activities:
- manufacture and waste disposal associated with certain chlorinated compounds, for example, PCBs, phenoxy herbicides, organochlorine pesticides, chlorinated benzenes, chlorinated aliphatic compounds, chlorinated catalysts, and halogenated diphenyl ethers
- bleach pulp and paper mill processes known to produce dioxin
- incineration of substantial chlorinated compounds
- former municipal solid waste incinerators
- hospital waste incinerators
- extensive use of pentachlorophenol (PCP) in timber treatment.
Where dioxins are detected at levels significantly above background, a site-specific assessment will be required to determine the appropriate action (refer to Schedule B4 for further information).
Further background information on dioxins is provided in Appendix E.
Prior to carrying out any processing or statistical analysis of the data set, an evaluation of the data quality should be carried out. As a minimum, this should include:
- checks on the completeness of the data as specified in the DQOs (all sample locations, sample depths etc. reported)
- checks on the accuracy of the reported data (all samples are correctly identified by location, depth, type etc.)
- identification of any obviously anomalous results such as elevated levels that are unexpected given the CSM and field notes on sampling (indicating a possible labelling or laboratory error)
- identification of invalid data (for example where the field or laboratory record indicates that sample integrity may have been compromised).
The possible reasons for anomalous data results (also see section 13.2.2) should be investigated and sampling and analysis repeated if appropriate.
Further information is provided in Appendix C and US EPA (2006a).
Detailed guidance on statistical procedures is beyond the scope of this guidance but some general considerations are outlined below. It is the responsibility of the site assessor to ensure that appropriate statistical procedures are followed when comparing site data with the investigation and screening levels listed in Schedule B1 and any site-specific assessment levels.
Evaluation of appropriate summary statistics and graphical displays of the sample data set are recommended for developing an improved understanding of contaminant distribution(s) and to determine whether any investigation and/or screening values have been exceeded.
Many spreadsheet and statistical software packages provide graphical methods, for example boxplots and histograms/frequency distributions, which are suitable for displaying site data. These displays can provide insight into the distribution of the data such as multi-modal, normal, log normal or exponential, which is a necessary precursor for selecting an appropriate statistical approach.
Evaluation of graphical displays such as frequency distributions can also assist the assessor in determining whether the data set should be split up into ‘domains of interest’ in which there is confidence that homogenous populations of data exist (that is, uni-modal and not bi-or multi-modal distribution) and for which sufficient data for meaningful statistical analysis is available.
Given that much of the sampling in contaminated site assessments is judgemental rather than random, caution needs to be taken when applying conventional (parametric) statistical methods which assume a normal (including log normal) distribution. Non-parametric methods (which do not make the assumption that data is normally distributed) provide an alternative approach for data assessment and may be useful in the early stages of a site assessment when typically there is little data available.
Non-parametric methods rely on ‘rank’ or ‘order’ statistics that are simply the percentiles of a distribution. Rather than using the mean to describe the centre of the distribution, non-parametric approaches more commonly use the median or 50th percentile. The difference between the upper quartile (75th percentile) and the lower quartile (25th percentile) is called the ‘interquartile range’ and is the non-parametric equivalent to the standard deviation for describing the spread of the data about the mean. Boxplots display these key non-parametric statistics and the ‘whiskers’ show the minimum and maximum of the data. Some software applications also show the position of the arithmetic mean (although this is not a percentile based statistic).
For multiple analytes, the range of concentrations and statistical distribution of results for each assessment area can be presented as in Table 8. Summary statistics should be provided for each soil unit/stratum tested and according to assessment sub-areas or domains of interest, if applicable and where sample size permits.
Table 8: Summary statistics for multiple analytes and assessment areas
Chemical name | XXX |
Investigation Level: Number of samples: | |
Minimum: Maximum: Inter-quartile (25th – 75th percentile) range: | |
Median (50th percentile): | |
Arithmetic mean: | |
Arithmetic standard deviation: | |
Geometric mean: | |
Geometric standard deviation: | |
95% Upper Confidence level (UCL) | |
Frequency distributiona | Number | % |
Less than investigation level: | | |
> 1 and < 2 times investigation level: | | |
>2 and <5 times investigation level: | | |
>5 and < 10 times investigation level: | | |
>10 times investigation level: | | |
a: An arbitrary method used to categorise data.
Maximum observed contaminant concentration—This generally provides a conservative assessment of exposure because if estimated risks from the maximum concentrations are not of concern, then the site should be suitable for the land use scenario(s) considered. However, a maximum concentration may not be representative of the source as a whole and may result in an overestimation or underestimation of risk if the data is extremely limited.
Mean concentration — The mean contaminant concentration can be a suitable metric provided that it can be shown that it adequately represents the source being considered. It is important that small areas of high concentrations or hotspots are not ignored by averaging with lower values from other parts of the site. The mean value may be more representative of the source as a whole than the maximum, and may provide a better estimation of the actual concentration that a population would be exposed to over a period of time.
The 95% upper confidence limit (UCL) of the arithmetic mean contaminant concentration provides a 95% confidence level that the true population mean will be less than, or equal to this value. The 95% UCL is a useful mechanism to account for uncertainty in whether the data set is large enough for the mean to provide a reliable measure of central tendency. Note that small data sets result in higher 95% UCLs.
The procedure for calculating the 95% UCL of the arithmetic mean and also of the mean for a log normal distribution is provided in NSW EPA (1995).
Further information on understanding data distributions and statistical procedures can be found in Gilbert (1987), US EPA (2006b) and in the ProUCL user guide US EPA (2007a).
Source US EPA (2006b)
Data generated from chemical analysis may fall below the limit of detection (LOD) or limit of reporting, (LOR) of the analytical procedure. These measurement data are generally described as ‘non-detects’ rather than ‘zero’ or ‘not present’ and the appropriate limit of detection for the analytical procedure should be reported. Data that includes both numerical data and ‘non-detect’ results is referred to as censored data in statistical literature.
Where the approximate percentage of non-detects is less than 15% of the relevant data set, then substitution of the LOD/2 or the LOD may be satisfactory, depending on the purpose of the analysis (US EPA 2006b). More detailed adjustments may be appropriate for where more than 15% of the relevant data set is below detection limits. In addition, sample size influences which procedure should be used to evaluate data. For example, the case where 1 sample out of 4 is not detected should be treated differently from the case where 25 samples out of 100 are non-detects. Further information can be found in US EPA (2006b) and US EPA (2007a).
Although substitution methods (such as replacing with the LOD/2 or the LOD) are reported widely in the literature for analysing data with non-detects, these approaches result in bias of the summary statistics calculated from the adjusted data set. The US EPA ProUCL software package provides alternative methods for calculating summary statistics such as the mean which do not rely on substitution methods (US EPA 2007a).
Adapted from US EPA (2006b) and BC Environment (2001)
Potential outliers are measurements that are extremely large or small relative to the rest of the data and therefore are suspected of misrepresenting the population from which they were collected (US EPA 2006b). Outliers may result from:
- transcription errors
- data-coding errors
- measurement problems
- true extreme values (hotspots).
Graphical displays of data, for example probability plots (concentration plotted against cumulative frequency), and x-y scatter plots (for example, ratios of contaminants expected to be associated with each other), can assist with identifying outliers. Evaluation of a combination of graphical displays with reference to relevant site layout diagrams is recommended.
It can be tempting to dismiss unexpectedly high values as ‘outliers’; however, this is not good practice, as a more thorough examination of the reasons for these unexpected values may lead to new insights into the data (such as the presence of an unsuspected hotspot of contamination) or to reconsideration of underlying assumptions about the data and its distribution.
Potential outliers should be checked for human error due to transcription/data-coding errors and invalid measurements from malfunctioning equipment. The former may be corrected whereas the latter can properly be discarded. Following the procedure outlined in Section 13.1 should minimise the impact of outliers from these causes.
If an outlier is not due to human error, then consider the available qualitative information regarding the data provenance and the site history and discard the outlier only if there is documentation to support the belief that the outlier is not part of the population under study. In all such cases, describe the population that the outlier belongs to and justify why this population is not considered relevant to the study objectives (e.g. elevated PAH due to presence of road bitumen fragments as opposed to contamination in soil derived from fuel leaking from an above-ground storage tank).
Discarding an outlier from a data set should be done with extreme caution as environmental datasets often include legitimate extreme values (US EPA 2006b). The decision taken should be based on scientific reasoning and be fully documented. Repeat sampling close (<1 m) to the original location may provide greater certainty in the decision process.
US EPA (2006b) describes several statistical tests for determining whether or not one or more observations are statistical outliers.
An efficient and accurate appraisal of a site requires that the data be collated in a form, or 'model' that facilitates understanding of the location, extent, trends, and likely 'behaviour' of any contamination.
An adequate understanding of what is occurring on a site is almost impossible to achieve from pages of raw data, especially where there are abnormal results or more than a handful of results. At its worst, sample identification numbers, sampling points, geotechnical logs, and results for each analyte will be on separate pages.
A uniform approach to the location and presentation of data makes for more rapid and accurate assessments of reports.
The major problems that can occur with data sets and assessments are:
- a failure to collate data and to condense it into logical and comprehensible tables
- cluttered data sets, tables and graphs
- treating the sum of the data as somewhat greater than the sum of its parts.
This is exemplified by:
- over-elaborate contour maps (some can be useful) based on a very limited number of data points which are not annotated on the map
- providing definitive conclusions unsupported by the data
- considering the numbers in isolation from other data important to interpretation, for example, site history and soil characteristics.
Reports should preferably be printed on A4 size paper, with durable covers and binding which allows for easy opening. Photographs and figures should be of high quality and adequately display the points of interest. Tables and figures should be formatted to enable easy reading (font size can be a particular issue when displaying large amounts of data) and printed as foldouts or enclosures where appropriate.
Where there is a series of site reports, each succeeding report should summarise the important and relevant portions from the preceding reports. This will assist in the rapid comprehension of new material by all parties involved.
Reports should follow appropriate subject headings and be structured in a logical way.
To support the site history investigation, copies of all current and old site layout plans, diagrams, correspondence, photographs, permits, etc. should be included in appendices. Where the site history is complicated because of numerous past uses and/or occupiers, information may be more effectively presented as a table or time line. An example is provided in Section 3.
A discussion of assumptions made in relation to the assessment, including those related to sampling density, sample locations, choice of analytes, off-site impacts and potential groundwater contamination, should be made.
Reports should also include the assessor’s opinion and conclusions relating to the environmental condition of the site, as well as recommendations for any further assessment of site contamination or site work the assessor considers necessary.
For all but the most simple of sites, some form of graphical representation is imperative for the assessor and other relevant parties to accurately visualise the site. Without such representations, inaccurate (and probably costly) decisions may be made. For large and complex sites, 3-D visualisation software may also be useful to illustrate the distribution of contamination etc.
Graphics should be well designed to promote understanding of the data. Some basic principles of graphic representation are given in Table 9.
Example graphics can be found in Appendix D.
Table 9. Helpful vs unhelpful graphics
Helpful | Unhelpful |
No cryptic abbreviations
No elaborate encoding | Numerous abbreviations requiring searching the text for explanation |
Words run in natural left to right direction | Words run vertically or in several directions |
Brief text messages explain data | Understanding graphic requires repeated references to text |
No elaborate shading, cross hatching or overpowering colouring | Elaborate or obscurely coded patterns requiring continual return to legend or key |
Simple, upper-and-lower case font | Multiple overbearing fonts |
Clearly printed | Murky and clotted printing |
Enlightens and arouses curiosity | Graphic repels interest and obscures meaning |
(adapted from Langley 1993 and Tufte 1983)
Site plans should be drawn to a scale appropriate to the size of the project and the level of detail required. Drawings on A3 or larger paper as foldouts or enclosures may be necessary. Plans should show:
- a north-facing arrow
- scale
- lot boundaries
- location of present and former infrastructure and site activities
- distribution of fill types
- locations of affected vegetation, stains, odours, chemical containers, etc.
- direction of surface run-off and drainage
- presence of above and below ground services
- areas covered by an impermeable seal (e.g. concrete, bitumen and buildings).
In some situations, it may be necessary to show previous site layouts as overlays over the current layout and perhaps have another overlay of sample locations or show sample excavation boundaries (see Appendix D).
Figures showing topographical contours in relation to site features and sample locations can assist with the assessment of sites with varied topography/changes of level.
Sample locations, identification numbers, results and depths should be plotted on one or more site layout figures. Sites with a large number of sample locations and numerous elevated results can be difficult to fully comprehend and time-consuming to assess. Therefore, to minimise assessment times and to allow, at a glance, a clear representation of contamination issues associated with the site, site plans should be used to display sample results. For large and complex sites, 3-D visualisation software may also be useful.
Contoured figures and/or maps can be useful for illustrating the distribution and trends of contamination, however, the interpolation methods used, for example, kriging, regression, minimum curvature, etc. can influence the results. For this reason, contours should be interpreted with caution and figures should include labelled data points for clarity.
If there is ‘too much’ data available, this may be addressed by displaying only significant results on the map. However, this should be done cautiously as censoring some of the data can obscure trends. ‘Normal’ results can be important if elevated results were anticipated and may need to be displayed.
An alternative method is to display a subset of the data e.g. separate figures for metals and petroleum hydrocarbons or provide some form of surrogate measure of where contamination may occur on a site. A series of figures, each with a different analyte, can be useful in this situation.
The following techniques may be useful to clearly display results:
- a separate site plan for each elevated analyte, which displays sample locations, sample identification numbers and depths, and shows different concentration ranges in different colours
- a separate site plan displaying analyte results (including locations, identification numbers and depths) for each elevated analyte, highlighting any exceedences of the guidelines by concentration range.
- a site plan displaying all analytes tested at each depth at each location and highlighting all results above environmental investigation thresholds in one colour and all results above health investigation thresholds in another colour (same colour regardless of analyte)
- a site plan displaying all results at each depth at each location in a specific colour for each analyte
- concentration contours, for each specific sample depth, to show plumes from a point source. Care should be taken when using this technique because inferred areas may be misleading if only a small number of sample locations are used
- cross-sections (noting vertical exaggeration) to illustrate the distribution and concentrations of contaminants and to display complex local geology
- statistical diagrams such as histograms and side-by-side boxplots.
It may be necessary to provide separate site plans for various depth ranges if plots are cluttered.
A particular technique will not be suitable in every situation. For example, choosing the third point above would not be useful if the majority of sampling results were above investigation levels. In this situation, a technique which showed concentration ranges in different colours would be more applicable.
To assist report assessors to review a site, a blank site plan which shows only sample locations, identification numbers and depths should also be provided. Examples of appropriate data presentation on site drawings are shown in Appendix D.
A separate site plan for validation samples should always be provided which clearly displays locations, depths and results of all relevant samples, including samples from Stage 1 and 2 reports.
Summary tables should show at least the essential details of sample locations and depths against the laboratory results. Results exceeding investigation threshold levels should be highlighted. For ease of reference, the addition of information such as date sampled, date received at laboratory, date analysed, and soil profile data to the summary table can expedite assessments by reducing cross-referencing. Examples are shown in Table 10 to Table 14.
Copies of the analytical results as originally received from the laboratory should be included as an appendix to the report together with details of relevant QA protocols, and QC results and chain-of-custody documentation. Further information is provided in Appendix C.
Bore logs and test pit logs are necessary to provide accurate descriptions of soil types encountered throughout the profile and should clearly distinguish natural soils from fill. Sample locations and perched water and groundwater levels should be shown. If rubble or rubbish is encountered, the percentages of each type of foreign matter should be estimated. Soil profile information may be presented as an appendix or used to construct cross-sectional drawings of the site. Presentation of the locations of odours, stains and field test measurements on the logs would assist with the site assessment. Bore logs are also to be used to represent the construction of monitoring wells. Refer also to Section 7.3 and Section 8.2.2.
Examples of bore, monitoring well and test pit logs are shown in Appendix D.
A photographic record that is well labelled for date, location and orientation is a valuable reference tool for items such as the site inspection (for example, topography, soil staining, state of underground storage tanks when removed, visual signs of plant toxicity), and the strata present in test pits and soil cores. It may also be useful for recording the appearance of split samples, particularly any visible heterogeneity in the field.
The following QA/QC documentation should be included (but not limited to):
- disposal dockets and receipts issued when contaminated soil and fuel tanks or other structures are removed from the site
- validation of any ’clean fill‘ used at the site
- certificates of clearance for asbestos removal or remediation clearance
- QA/QC protocols for field and laboratory work
- calibration reports for all field monitoring equipment
- chain-of-custody documents for all soil, vapour, groundwater and surface water samples and laboratory receipt notices.
Further information is provided in Appendix C.
Consultants, assessors and government agencies should have access to electronic data (such as site data in spreadsheet form) as it avoids a further source of transcription error and facilitates the further analysis of data using other software packages.
Users of data should be aware of copyright, data protection and data integrity issues.
Table 10. Example report structure for soil analytical results
Soil Bore | Depth | As | Cd | Co | Cu | Hg (inorganic) | Ni | Pb | Zn |
| mm | mg/kg | mg/kg | mg/kg | mg/kg | mg/kg | mg/kg | mg/kg | mg/kg |
A7 | 0-50 | 3 | <0.05 | 4 | 28 | 0.25 | 14 | 200 | 210 |
A7 | 150-300 | 3 | <0.05 | 6 | 100 | 0.25 | 15 | 170 | 220 |
A7 | 300-450 | <0.05 | <0.05 | 8 | 20 | <0.05 | 20 | 10 | 34 |
A8 | 0-50 | 2 | <0.05 | 18 | 8.0 | 0.50 | 75 | 36 | 24 |
A8 | 150-300 | 2 | <0.05 | 12 | 28 | 0.05 | 28 | <0.05 | 46 |
A9 | 0-50 | 3 | <0.05 | 4 | 50 | 0.55 | 15 | 250 | 310 |
A9 | 150-300 | 2 | <0.05 | 5 | 60 | 0.40 | 13 | 160 | 240 |
A10 | 0-50 | <0.05 | <0.05 | <0.05 | 1.0 | <0.05 | 8 | 10 | 16 |
A10 | 150-300 | 5 | 1 | 5 | 1.8 | 0.05 | 13 | 24 | 34 |
A10 | 300-450 | 3 | 1 | 7 | 1.8 | <0.05 | 15 | 12 | 30 |
A10 | 750-900 | <0.05 | 1 | 6 | 1.5 | <0.05 | 14 | 4 | 22 |
A11 | 0-50 | 5 | <0.05 | 4 | 24 | 0.10 | 11 | 290 | 540 |
A11 | 150-300 | 10 | <0.05 | 5 | 1750 | 0.70 | 15 | 450 | 760 |
A11 | 300-450 | 5 | <0.05 | 9 | 1.9 | 0.05 | 17 | 90 | 30 |
A12 | 0-50 | 3 | 2 | 6 | 28 | 0.25 | 15 | 100 | 80 |
A12 | 150-300 | 5 | <0.05 | 7 | 60 | 2.70 | 18 | 940 | 190 |
A12 | 300-450 | 1 | <0.05 | 12 | 26 | 0.20 | 24 | 46 | 46 |
| HIL A | 100 | 20 | 100 | 6000 | 40 | 400 | 300 | 7400 |
HIL A = Health investigation level for standard residential use
BOLD font indicates result exceeds the relevant HIL
Table 13. Example statistical analysis of results for a particular sampling event
Sample no. | Arsenic | Cadmium | Cobalt | Chromium | Copper | Nickel | Lead | Zinc |
A1 | 12 | 1 | 27 | 256 | 51 | 69 | 116 | 398 |
A2 | 9 | 3 | 12 | 316 | 131 | 36 | 47 | 105 |
A3 | 8 | 1 | 26 | 294 | 236 | 82 | 25 | 73 |
A4 | 7 | 1 | 5 | 15 | 1290 | 19 | 154 | 1660 |
A5 | 8 | 1 | 34 | 132 | 403 | 166 | 99 | 105 |
A6 | 4 | 1 | 20 | 39 | 333 | 130 | 11 | 64 |
A7 | 12 | 1 | 43 | 300 | 546 | 84 | 58 | 128 |
A8 | 10 | 2 | 11 | 231 | 766 | 45 | 117 | 159 |
A9 | 6 | 1 | 52 | 304 | 642 | 62 | 57 | 131 |
A10 | 36 | 1 | 7 | 254 | 836 | 34 | 95 | 571 |
A11 | 8 | 1 | 22 | 255 | 33 | 92 | 19 | 46 |
A12 | 7 | 5 | 27 | 225 | 541 | 63 | 140 | 1380 |
A13 | 4 | 1 | 24 | 365 | 321 | 87 | 42 | 150 |
A14 | 3 | 0.5 | 83 | 257 | 453 | 71 | 22 | 30 |
A15 | 4 | 4 | 57 | 235 | 678 | 84 | 111 | 261 |
A16 | 3 | 1 | 22 | 223 | 165 | 59 | 385 | 584 |
A17 | 5 | 2 | 58 | 277 | 207 | 92 | 840 | 1740 |
A18 | 7 | 2 | 45 | 330 | 105 | 86 | 1870 | 649 |
A19 | 5 | 0.5 | 62 | 503 | 26 | 65 | 80 | 94 |
A20 | 6 | 1 | 46 | 400 | 345 | 65 | 217 | 4310 |
A21 | 12 | 1 | 30 | 273 | 16 | 81 | 180 | 458 |
A22 | 12 | 1 | 27 | 256 | 789 | 69 | 116 | 398 |
A23 | 15 | 1 | 15 | 254 | 345 | 44 | 117 | 218 |
A24 | 9 | 3 | 12 | 316 | 16 | 36 | 47 | 105 |
A25 | 34 | 1 | 29 | 169 | 342 | 100 | 43 | 135 |
A26 | 8 | 1 | 26 | 294 | 132 | 82 | 25 | 73 |
A27 | 12 | 1 | 32 | 215 | 107 | 104 | 272 | 360 |
A28 | 7 | 1 | 5 | 15 | 1290 | 19 | 154 | 1660 |
A29 | 14 | 2 | 51 | 266 | 119 | 112 | 383 | 852 |
A30 | 6 | 1 | 77 | 365 | 74 | 91 | 23 | 64 |
A31 | 14 | 1 | 53 | 205 | 33 | 101 | 34 | 39 |
A32 | 8 | 1 | 34 | 132 | 40 | 166 | 99 | 105 |
A33 | 17 | 1 | 43 | 291 | 32 | 74 | 58 | 112 |
A34 | 4 | 1 | 20 | 39 | 357 | 130 | 11 | 64 |
A35 | 12 | 1 | 31 | 285 | 1260 | 79 | 66 | 139 |
A36 | 12 | 1 | 43 | 300 | 345 | 84 | 58 | 128 |
A37 | 8 | 2 | 121 | 236 | 156 | 148 | 32 | 94 |
A38 | 9 | 2 | 53 | 454 | 435 | 79 | 10 | 19 |
A39 | 6 | 1 | 32 | 207 | 534 | 81 | 15 | 37 |
A40 | 8 | 1 | 46 | 240 | 39 | 102 | 84 | 165 |
A41 | 8 | 1 | 15 | 269 | 30 | 48 | 59 | 88 |
A42 | 10 | 2 | 11 | 231 | 66 | 45 | 117 | 159 |
A43 | 9 | 2 | 44 | 250 | 42 | 88 | 92 | 155 |
A44 | 6 | 1 | 52 | 304 | 42 | 62 | 57 | 131 |
A45 | 5 | 2 | 35 | 412 | 615 | 62 | 25 | 982 |
A46 | 36 | 1 | 7 | 254 | 55 | 34 | 95 | 571 |
A47 | 6 | 1 | 39 | 221 | 453 | 59 | 11 | 30 |
A48 | 8 | 1 | 22 | 255 | 65 | 92 | 19 | 46 |
A49 | 7 | 1 | 55 | 278 | 34 | 87 | 28 | 64 |
A50 | 5 | 1 | 34 | 239 | 66 | 87 | 21 | 67 |
A51 | 9 | 1 | 79 | 300 | 75 | 103 | 57 | 142 |
A52 | 8 | 2 | 29 | 188 | 67 | 83 | 312 | 643 |
A53 | 9 | 2 | 34 | 227 | 34 | 72 | 86 | 164 |
A54 | 4 | 1 | 57 | 153 | 42 | 204 | 33 | 80 |
A55 | 7 | 1 | 48 | 259 | 50 | 101 | 204 | 251 |
A56 | 16 | 4 | 24 | 143 | 169 | 79 | 1310 | 10 900 |
A57 | 8 | 1 | 45 | 207 | 36 | 191 | 30 | 122 |
A58 | 5 | 1 | 34 | 239 | 1185 | 87 | 21 | 67 |
A59 | 8 | 2 | 29 | 188 | 1034 | 83 | 312 | 643 |
A60 | 4 | 1 | 57 | 153 | 442 | 204 | 33 | 80 |
A61 | 16 | 4 | 24 | 143 | 116 | 79 | 1310 | 10 900 |
A62 | 5 | 1 | 40 | 147 | 47 | 199 | 10 | 100 |
A63 | 6 | 1 | 28 | 177 | 231 | 106 | 54 | 110 |
A64 | 2 | 1 | 16 | 107 | 184 | 35 | 79 | 366 |
A65 | 9 | 1 | 48 | 206 | 395 | 98 | 33 | 166 |
A66 | 11 | 1 | 26 | 156 | 845 | 54 | 216 | 251 |
A67 | 6 | 1 | 13 | 287 | 25 | 70 | 46 | 71 |
| | | | | | | | |
Arithmetic mean | 9 | 1 | 36 | 239 | 314 | 86 | 164 | 675 |
Standard deviation | 7 | 1 | 21 | 92 | 346 | 41 | 322 | 1913 |
Geometric mean | 8 | 1 | 30 | 210 | 158 | 77 | 70 | 193 |
Minimum | 2 | 0.5 | 5 | 15 | 16 | 19 | 10 | 19 |
Maximum | 36 | 5 | 121 | 503 | 1290 | 204 | 1870 | 10900 |
Median | 8 | 1 | 32 | 250 | 165 | 82 | 58 | 135 |
90 percentile | 14 | 2 | 57 | 322 | 808 | 137 | 312 | 1141 |
95 percentile | 17 | 4 | 72 | 390 | 1140 | 184 | 703 | 1716 |
Number of data points | 67 | 67 | 67 | 67 | 67 | 67 | 67 | 67 |
Adapted from Queensland Department of Environment 1998
Table 14. Example frequency distribution data for copper data listed in Table 13
Concentration range (ppm) | Frequency | Cumulative % |
060 | 20 | 30 |
60200 | 16 | 54 |
200400 | 11 | 70 |
400600 | 8 | 82 |
600800 | 5 | 90 |
8001000 | 2 | 93 |
10001200 | 2 | 96 |
12001400 | 3 | 100 |
Adapted from Queensland Department of Environment 1998
Assessment of site contamination, or potential contamination, can present risks to the environment as well as to site personnel and local residents. This guidance provides the minimum measures that should be adopted to ensure protection of the environment during site assessment. Site-specific environmental management measures must ensure compliance with environmental management and protection legislation applying in each jurisdiction.
All states and territories have work health and safety legislative requirements. Plans developed under such legislation should address all relevant exposure pathways for site-specific contaminants of concern. Site assessment activities should comply with relevant work health and safety guidance and legislation applying in each jurisdiction.
Environmental protection plans should address the following issues:
- management of dust emissions and on-site and off-site odours
- protection of groundwater resources
- prevention of migration of contamination to adjacent sites or uncontaminated areas within the site
- prevention of contaminated run-off water reaching stormwater systems or local surface water environments
- prevention of initiation or spread of fire, either underground or above ground
- collection and disposal of excavation spoil
- collection and disposal of contaminated groundwater.
Less obvious assessment issues that need to be addressed include:
- extending contamination or assisting contaminant migration during site investigation works by, for example, drilling through a contaminated aquifer into an uncontaminated lower aquifer thereby creating a conduit through which contamination may migrate
- introducing contamination to an otherwise clean soil stratum by backfilling a test pit found to be contaminated at surface level but clean at depth using the contaminated soil. It is always preferable to temporarily stockpile test pit spoil in excavation sequence so that it may be returned to the pit to roughly the same depth from which it was excavated
- initiating or extending underground fire by the introduction of oxygen
- enhancing acid run-off by enabling oxidation of in situ materials through exposure to atmosphere
- destabilising an otherwise stable embankment by introducing water.
The following elements of environmental protection should be considered prior to site assessment and be incorporated into the site assessment plan for each site. In particular, site contamination that is likely to cause public concern by the scale of operations, the nature of the site contamination or the potential for emission of noxious or offensive odours should indicate the commencement of public consultation and community engagement (refer Schedule B8) well before the commencement of site assessment works.
Environmental concerns regularly encountered on site assessments are dust and odour emissions which may be wind-blown and aggravated by the actions of trucks or other plant on the site. When warranted by the scale of site assessment and specific site conditions, area/boundary monitoring for dust deposition, inspirable and respirable dust and respective contaminants should be undertaken. Protection measures are important to ensure that dust inhalation or noxious or offensive odours do not pose a health risk for site operatives, nor a health risk or nuisance to local residents or passers-by and that concentrations of chemical substances do not exceed any relevant state or territory guidelines.
The traditional methods of dust and odour control include:
- application of a water spray with the objective to dampen the soil and not to saturate it, as potentially contaminated run-off from saturated soils entering adjacent sites, stormwater systems, or local waterways must be avoided (note: care should be taken when applying water onto soil that has recently been contaminated with volatiles or semi-volatiles, as this can result in a large increase in contaminant emissions from the soil)
- covering exposed faces with barriers (e.g. synthetic barriers, mulch) to prevent the emission of odours and dust
- minimising traffic and its speed on exposed contaminated soils
- the use of ground covers
- installation of screens to act as windbreaks.
Many sites, particularly those with petroleum hydrocarbons, organic contamination or putrescible wastes, may generate offensive odours or noxious vapours. In such cases, intensive odour control measures should be considered including minimising the exposed surface of the odorous materials at all times, timing excavation activities to minimise off-site nuisance, and by re-covering exposed faces overnight or during periods of low excavation activity. Such odorous materials should not be stockpiled unless closely contained or covered.
When dealing with volatile pollutants an assessment should be made of the need for the regular analysis of atmospheric levels of pollutants on site and at site boundaries to ensure that workers and residents are not being exposed to unacceptable levels of substances (for example, benzene) that may give rise to adverse health effects.
In addition, site boundary and competent community monitoring of offensive odours should be regularly undertaken during assessment of problematic sites. Site work practices relating to odour-generating activities should be promptly amended or stopped and reassessed in response to the results of boundary and community monitoring.
The social impact from the excavation of odorous or noxious materials can often be mitigated by excavating only when the wind direction is such that there will be the minimum possible effect upon neighbouring populations.
Where excavation of odorous or noxious material is expected or planned as part of an assessment process, the local population and other stakeholders should:
- be advised of the expected duration of the operation
- be advised that the operation will last for a limited time only
- be advised whether or not the odours may pose any potential health risk
- be given reassurance with regard to mitigation measures being undertaken.
An effective risk communication and community engagement program is an essential consideration for sites that pose a risk of offensive or noxious odours. Information on development and implementation of community engagement and risk communication programs is provided in Schedule B8.
Before commencement of any drilling work, sufficient research should be undertaken to establish how much information is available regarding the geology and hydrogeology of the area to be investigated. If groundwater contamination is suspected there should be an audit of local bores. If more than one aquifer is expected, care should be taken to ensure that the potential for cross-contamination is minimised. Bores should be constructed so that different aquifers are isolated.
Licensing of monitoring bores may be a statutory requirement in some states and territories. There may also be state or territory guidelines that apply to minimum bore requirements and their decommissioning.
Care must be taken to avoid surface run-off from assessment activity impacting on adjacent sites, wetlands, water courses or stormwater drainage systems. The site assessor should be aware of the topography and geology of the site under assessment, and the possibility of migration of contaminants within the site or to adjacent sites, whether wind-blown, adhering to vehicles, plant and equipment, as free-flowing liquids, as surface run-off, or in groundwater flow. Stockpiled, excavated materials awaiting removal from site may create a particular risk to the environment.
Mitigation measures may include the use of temporary (waterproof) covers, excavation of drainage or run-off water diversion trenches, collection or absorption pits, or installation of temporary barriers in the form of hay bales, geofabrics or similar materials. Temporary bunding around stockpiles, or location of stockpiles on waterproof surfaces such as asphalt or concrete, or under cover where available, should be considered. Designation of an area within which all run-off and infiltration is to be controlled in accordance with strict performance objectives (for example, zero uncontrolled run-off) should also be considered. Disposal of any run-off should be carried out in accordance with relevant state or territory legislation.
Following rainfall it may be necessary to retrieve any sediment which has been carried in run-off or drainage water and manage this material appropriately. Respraying contaminated water onto stockpiles of contaminated soil as a means of effectively managing the water is also a possibility depending on jurisdictional guidelines and the nature of the contamination.
Treatment and disposal of collected contaminated run-off water should be appropriate to the contamination expected. If water treatment facilities are not immediately available, following consultation with local waste water authorities, diversion to sewer should be considered. Removal to landfill (not permitted in certain states) or treatment facility by means of road tanker is an expensive final option.
Potential carry-over of contamination to public roads and highways is an issue where excavation plant is operating on a site. Care must be taken to ensure that potentially contaminated material is not transported off site. Vehicle washing systems with facilities for handling the wash water and the installation of ‘rumble strips’ to help dislodge dust and mud, should be considered for installation at exits from sites where potential carry-over is perceived to be a problem. Procedures should be set in place for the handling and disposal of potentially contaminated water arising from wheel-wash operations.
Sample pits should be backfilled soon after sampling and sampling should not take place during rain. Contaminated water may be encountered where sample pits have been left open, and in boreholes. Care should be taken in disposing of contaminated flush water from borehole purging to ensure that contamination is not spread on the site. Gross contamination from borehole purging should be collected in drums or other suitable container for approved off-site disposal.
After excavation test pits may fill with rain or groundwater. Care should be taken to ensure that backfilling of the test pit does not rapidly displace this water, causing it to flow over the site. If necessary, the test pit should be part-backfilled and then bailed out to a suitable storage to enable full backfilling with spoil. Contaminated water should be disposed as appropriate.
All containers remaining temporarily on-site, and containing potentially contaminated materials, should be labelled with appropriate hazard warnings and waste producer contact details.
It is normal practice to return excavation spoil from test pits to the excavation from which it came. However, care should be taken to ensure that materials are replaced in soil horizon order and that contaminated materials are not returned to a pit where they could contaminate unaffected strata or groundwater. Due to practical difficulties in compaction of excavation spoil there will inevitably be excess spoil after backfilling of a test pit. Care should be taken to ensure that contaminated spoil does not become spread across an otherwise uncontaminated surface. Drilling cuttings should not be returned to a bore.
Excess spoil should be stored in a lined skip or lined drums brought to site or placed on an impermeable surface such as concrete, asphalt, polyethylene sheeting or similar until analytical results can be assessed to enable cost-effective and safe methods of disposal. Where excess spoil is stored on site, and is not stored within a container, bunding should occur around the area to contain potential run-off. If contaminated materials are to be drummed for disposal or for treatment, the contents should be analysed, and management decisions made, based on the analytical results. All containers remaining temporarily on-site, and containing potentially contaminated materials, should be labelled with appropriate hazard warnings and waste producer contact details.
Allowances should be made within site assessment budgets for any necessary safe removal of a quantity of soil/fill from the site to an appropriate waste disposal or treatment facility. Transport and disposal of contaminated soil should be carried out in accordance with relevant state or territory legislation.
Noise can be a health risk to workers and is often a nuisance to those in the vicinity of a site. The potential for noise arising from site assessment activities should be evaluated and appropriate control measures put in place to reduce unacceptable noise (for example, by installing screens or noise baffles). Noise should not be a nuisance to people living or working around the site. Activities with potential for noise generation should be carried out in accordance with relevant state or territory legislation.
Similarly, vibration from excavation and drilling, from plant, or from the movement of heavily laden trucks can sometimes result in damage to foundations of adjacent structures or to underground services or utilities. This possibility should be addressed and any risks assessed prior to choice of excavation or drilling method.
Acid sulfate soils (ASS) are naturally occurring soils, sediments and peats which contain iron sulfides. In an anoxic state, these materials are benign and do not pose a significant risk to human health or the environment. However, the disturbance of ASS and exposure to oxygen has the potential to cause significant environmental and economic impacts including fish kills and loss of biodiversity in wetlands and waterways and contamination of groundwater resources by acid and metals (WA DEC 2009). Activities that have the potential to disturb ASS, either directly or by affecting the elevation of the water table, need to be managed appropriately.
Where ASS is identified as a potential hazard, investigation and management of ASS should be carried out in accordance with relevant state or territory requirements. Jurisdictions should be consulted for advice on appropriate control measures to apply on the management of ASS prior to any dewatering or excavation activities taking place. Further technical information is provided in WA DEC (2011).
Special care should be taken to ensure that any assessment works or activities on or adjacent to sites of cultural or natural heritage significance will not have an adverse impact. Heritage places may include buildings, structures, archaeological remains, or landscaped or natural areas of aesthetic, historic, scientific or social value. Where appropriate, advice should be sought from the local representatives of the National Congress of Australia’s First Peoples, the Australian Heritage Council, and state or territory heritage bodies and local councils.
Special care should be taken to ensure that any assessment works or activities will not impact upon rare natural habitats or any endangered species. Advice may be sought from the relevant jurisdiction to ensure that site environmental protection plans are sufficiently protective.
Aller, L, Bennett, TW, Hackett, G, Petty, RJ, Lehr, JH, Sedoris, H, Nielsen, DM & Denne, JE 1989, Handbook of suggested practices for the design and installation of groundwater monitoring wells, Report EPA 600/4-89/034, 398, United States Environmental Protection Agency.
ANZECC & ARMCANZ 2000, National water quality management strategy. Australian and New Zealand guidelines for fresh and marine water quality, Australian and New Zealand Environment and Conservation Council & Agriculture and Resource Management Council of Australia and New Zealand.
APHA, AWWA & WEF 2005, Standard methods for examination of water and wastewater, 21st edn, American Public Health Association, the American Water Works Association & the Water Environment Federation, Washington, DC. (Available online at http://www.standardmethods.org)
API 2005, ‘Collecting and interpreting soil gas samples from the vadose zone. A practical strategy for assessing the subsurface vapour-to-indoor air migration pathway at petroleum hydrocarbon sites‘, Regulatory Analysis and Scientific Affairs, publication no. 4741, American Petroleum Institute.
API 2006, Downward solute plume migration: assessment, significance, and implications for characterisation and monitoring of 'diving plumes', API Bulletin 24.
ARMCANZ & ANZECC 1995, Guidelines for groundwater protection in Australia, Agriculture and Resource Management Council of Australia and New Zealand & Australian and New Zealand Environment and Conservation Council, Canberra, Australia.
ADITC 1997, The manual of methods, applications and management, Australian Drilling Industry Training Committee Ltd, CRC Press, Florida.
AS 1726—1993, Geotechnical site investigations, Standards Australia.
AS 2368—1990, Test pumping of water wells, Standards Australia.
AS 4482.1—2005, Guide to the investigation and sampling of sites with potentially contaminated soil, part 1: non-volatile and semi-volatile compounds, Standards Australia.
AS 4482.2—1999, Guide to the sampling and investigation of potentially contaminated soil, part 2: volatile substances, Standards Australia.
AS 4964—2004, Method for the qualitative identification of asbestos in bulk, Standards Australia.
AS/NZS 5667.11—1998, Water quality – sampling, part 11: Guidance on sampling of groundwater, Standards Australia.
ASTM D5092 (2004e1), Standard practice for design and installation of groundwater monitoring wells, ASTM International.
ASTM D5753–05 (2010), Guide for planning and conducting borehole geophysical logging, ASTM International.
ASTM D6429–99 (2006), Standard guide for selecting surface geophysical methods, ASTM International.
ASTM D6432-99 (2005), Standard guide for the surface ground penetrating radar method, ASTM International.
ASTM D6771–02, Standard practice for low-flow purging and sampling for wells and devices used for groundwater quality investigations, ASTM International.
ASTM E1527–05, Standard practice for environmental site assessments: Phase 1 Environmental Site Assessment Process, ASTM International.
ASTM E1689–95 (2008), Standard guide for developing conceptual site models for contaminated sites, ASTM International.
ASTM E2531 (2006), Guide for development of conceptual site models and remediation strategies for light non-aqueous-phase liquids released to the subsurface, ASTM International.
Baker, K, Hayward, H, Potter, L, Bradley, D & MacLeod, C 2009, The VOCs handbook. Investigating, assessing and managing risks from inhalation of volatile organic compounds (VOCs) at land affected by contamination, CIRIA C682, CIRIA, London.
Barber, C & Briegel, D 1987, 'A method for the in situ determination of dissolved methane and other volatiles in groundwater in shallow aquifers', Journal of Contaminant Geology, vol. 12, pp. 5160.
Barber, C & Davis, GB 1987, 'Representative sampling of groundwater from short-screened boreholes', Ground Water vol. 25, no. 5, pp. 581587.
Barber, C & Davis, GB 1994, 'Representative sampling of groundwater from boreholes.', in Adriano, DC, Iskandar, AK & Murarka, IP (eds), Advances in Environmental Science: Contamination of Groundwaters — Methodology and Modelling, Science Reviews, UK, pp. 141160.
Barcelona, MJ, Gibb, JP, Helfrich, JA & Garske, EE 1985, Practical guide for groundwater sampling, Illinois State Water Survey Report 374, US National Groundwater Association, p.94.
Barnett, B, Townley, LR, Post, V, Evans, RE, Hunt, RJ, Peeters, L, Richardson, S, Werner, AD, Knapton, A & Boronkay, A 2012, Australian groundwater modelling guidelines, Waterlines report, National Water Commission, Canberra, Australia.
BC Environment, 2001, Outliers: a guide for data analysts and interpreters on how to evaluate unexpected high values, British Colombia Environment Contaminated Sites Statistical Applications Document No. 12-8.
Beck, P & Mann, B 2010, A technical guide for demonstrating monitored natural attenuation of petroleum hydrocarbons in groundwater, CRC CARE Technical report no. 15, CRC for Contamination Assessment and Remediation of the Environment, Adelaide, Australia.
Clements, L, Palaia, T & Davis, J 2009, Characterisation of sites impacted by petroleum hydrocarbons: guideline document, CRC CARE Technical report no. 11, CRC for Contamination Assessment and Remediation of the Environment, Adelaide, Australia.
Cook, RG 2003, A guide to regional groundwater flow in fractured rock aquifers, CSIRO Land and Water, Glen Osmond, South Australia.
Davis, GB, Trefry, MG & Patterson, BM 2004, Petroleum and Solvent Vapours: Quantifying their Behaviour, Assessment and Exposure, A report to the Western Australian Department of Environment, CSIRO Land and Water, Perth, Western Australia.
Davis, GB, Merrick, NP & McLaughlan, RG 2006, Protocols and techniques for characterising sites with subsurface petroleum hydrocarbons – a review, CRC CARE Technical report no. 2, CRC for Contamination Assessment and Remediation of the Environment, Adelaide, Australia.
Davis, GB, Wright, J & Patterson, BM 2009a, Field assessment of vapours, CRC CARE Technical Report no. 13, CRC for Contamination Assessment and Remediation of the Environment, Adelaide, Australia.
Davis, GB, Trefry, MG & Patterson, BM 2009b, Petroleum vapour model comparison, CRC CARE Technical Report no. 9, CRC for Contamination Assessment and Remediation of the Environment, Adelaide, Australia.
Davis, GB, Patterson, BM & Trefry, MG 2009c, Biodegradation of petroleum hydrocarbon vapours, CRC CARE Technical Report no. 12, CRC for Contamination Assessment and Remediation of the Environment, Adelaide, Australia.
Driscoll, FG 1986, Groundwater and wells, 2nd edn, Johnsons Filtration Systems Inc, St Paul, Minnesota.
DTSC 2009, Vapour intrusion mitigation advisory, Department of Toxic Substances Control, California Environmental Protection Agency, April.
EA 2000a, Guide to good practice for the development of conceptual models and the selection and application of mathematical models of contaminant transport processes in the subsurface, NC/99/38/3, Environment Agency, England and Wales.
EA 2000b, Guidance on the assessment and monitoring of natural attenuation of contaminants in groundwater, R&D publication 95, prepared by Aspinwall & Company/NGC, Environment Agency, England and Wales.
EA 2003, An illustrated handbook of DNAPL transport and fate in the subsurface, R&D publication 133, Environment Agency, England and Wales.
EA 2006, Guidance on the design and installation of groundwater quality monitoring points, Science Report SC-020093, Environment Agency, England and Wales.
Edwards, JW, van Alphen, M & Langley, A 1994, Identification and assessment of contaminated land. Improving site history appraisal, Contaminated sites monograph series, no. 3, South Australian Health Commission, Adelaide, Australia.
enHealth 2005, Management of asbestos in the non-occupational environment, Department of Health and Ageing, Canberra, Australia.
enHealth 2012, Australian exposure factor guidance. Environmental Health Subcommittee (enHealth) of the Australian Health Protection Principal Committee, Canberra, Australia.
EPA Victoria 1991, Guidelines for a site assessment report. Guidelines for Environmental Auditors, Environment Protection Authority, Melbourne.
EPA Victoria 2000, Groundwater sampling guidelines, Publication 669, Environment Protection Authority, Melbourne.
EPA Victoria 2006, Hydrogeological assessment (groundwater quality) guidelines, Publication 668, Environment Protection Authority, Melbourne.
EPA Victoria 2010, Industrial Waste Resource Guidelines (7), Sampling and Analysis: Soil Sampling, Environment Protection Authority, Melbourne.
Ferguson, CC 1992, 'The statistical basis for spatial sampling of contaminated land', Ground Engineering, June, pp. 3438.
Fetter, CW 2001, Applied hydrogeology, 4th edn, Prentice Hall Inc.
Gilbert, RO 1987, Statistical methods for environmental pollution monitoring, Van Nostrand Reinhold Company, New York.
Hartman, B 2002, ‘How to collect reliable soil gas data for upward risk assessments, part 1: active soil gas method’, LUSTLine Bulletin, no. 42, pp. 1722.
Hartman, B 2003, ‘How to collect reliable soil gas data for upward risk assessments, part 2: surface flux-chamber method’, LUSTLine Bulletin, no. 44, pp. 1418; 34.
Hartman, B 2006, ‘How to collect reliable soil gas data for risk-based applications — specifically vapor intrusion, part 4: updates on soil gas collection and analytical procedures’, LUSTLine Bulletin, no. 53, pp. 1419.
Helsel, DR 1990, 'Less than obvious, statistical treatment of data below the detection limit', Environmental Science and Technology, vol. 24, no. 12, pp. 17671774.
Heyworth, J 1991, 'Sampling and statistical analysis for assessing contaminated sites', in El Saadi, O & Langley, AJ (eds), The health risk assessment and management of contaminated sites, South Australian Health Commission, Adelaide, Australia.
IGMC 2010, Software catalogue, International Groundwater Modelling Centre, Colorado School of Mines, USA, available online at http://www.mines.edu/igwmc/.
ITRC 1999, Natural attenuation of chlorinated solvents in groundwater: principles and practices, ISB-3, ITRC In situ Bioremediation Team, Interstate Technology & Regulatory Council, Washington, DC. Also see www.itrcweb.org.
ITRC 2005, Technology overview of passive sampler technologies, DSP-4, ITRC Diffusion/Passive Sampler Team, Interstate Technology & Regulatory Council, Washington, DC.
ITRC 2007a, Vapor intrusion pathway: a practical guideline, VI-1, ITRC Vapor Intrusion Team, Interstate Technology & Regulatory Council, Washington, DC.
ITRC 2007b, Vapor intrusion pathway: investigative approaches for typical scenarios, a supplement to Vapor intrusion pathway: a practical guideline, technical and regulatory guidance supplement prepared by the ITRC Vapor Intrusion Team, Interstate Technology & Regulatory Council, Washington, DC.
ITRC 2007c, Protocol for use of five passive samplers to sample for a variety of contaminants in groundwater, DSP-5, ITRC Diffusion/Passive Sampler Team, Interstate Technology & Regulatory Council, Washington, DC.
ITRC 2009, Evaluating natural source zone depletion at sites with LNAPL, LNAPL-1, LNAPL Team, Interstate Technology & Regulatory Council, Washington, DC.
ITRC 2010, A decision framework for applying monitored attenuation processes to metals and radionuclides in groundwater, APMR-1, Attenuation Processes for Metals and Radionuclides Team, Interstate Technology & Regulatory Council, Washington, DC.
Keuper, BH & Davies, KL 2009, Assessment and delineation of DNAPL source zones at hazardous waste sites, EPA/600/R-09/119, US EPA National Risk Management Research Laboratory, Cincinnati, Ohio.
Kruseman, GP & de Ridder, NA 1994, Analysis and evaluation of pumping test data, International Institute for Land Reclamation and Improvement, Wageningen, The, Netherlands.
Langley, AJ 1993, ‘The representation of data and the appraisal of contaminated land’, in The health risk assessment and management of contaminated sites, proceedings of the second national workshop, by Langley, AJ & van Alphen, M, South Australian Health Commission, Adelaide, Australia.
Lock, WH 1996, Composite sampling, National environmental health monographs, Soil series no. 3, South Australian Health Commission, Adelaide, Australia.
McAlary, TA, Nicholson, P, Groenevelt, H & Bertrand, B 2009, ‘A case study of soil-gas sampling in silt and clay-rich (low permeability) soils’, Ground Water Monitoring & Remediation, vol. 29, no. 1, pp 144152.
MDBC 1997, Murray-Darling Basin groundwater quality sampling guidelines, Technical report no. 3, Murray-Darling Basin Commission, Australia.
Middlemis, H 2000, Groundwater flow modelling guideline, Murray-Darling Basin Commission, Australia.
NHMRC 2008, National water quality management strategy. Guidelines for managing risk in recreational water, National Health and Medical Research Council, Australia.
NHMRC, 2011, Australian drinking water guidelines, National Health and Medical Research Council, Australia.
Nielsen, DM 2006, The Practical Handbook of Environmental Site Characterisation and Ground-Water Monitoring, Second Edition, CRC Press, Taylor and Francis Group.
Neilsen, DM and Neilsen, G 2005, The Essential Handbook of Groundwater Sampling, CRC Press, Taylor and Francis Group.
NJDEP 2005a, Field Sampling Procedures Manual, New Jersey Department of Environmental Protection. (Available online at www.nj.gov/dep/srp/guidance/fspm)
NJDEP 2005b, Vapor intrusion guidance, New Jersey Department of Environmental Protection. (Available online at www.nj.gov/dep/srp/guidance/vaporintrusion/vig.htm)
NSW DECC 2006, Guidelines for the NSW site auditor system, 2nd edn, NSW Department of Environment & Climate Change, Sydney, Australia.
NSW EPA 1994, Guidelines for assessment of service station sites, NSW Environment Protection Authority, Sydney, Australia.
NSW EPA 1995, Sampling Design Guidelines, NSW Environment Protection Authority, Sydney, Australia.
NSW EPA 2011, Contaminated sites. Guidelines for consultants reporting on contaminated sites, NSW Environment Protection Authority, Sydney, Australia.
NYSDOH 2006, Guidance for evaluating soil vapor intrusion in the State of New York, New York State Department of Health.
ODEQ 2010, Guidance for assessing and remediating vapor intrusion in buildings, Report no. 10-LQ-007, Oregon Department of Environmental Quality, Portland, Oregon, USA.
Pagano, RR 1986, Understanding statistics in the behavioural sciences, West Publishing House, St Paul, Minnesota, USA.
Patterson, BM & Davis, GB 2008, ‘An in situ device to measure oxygen in the vadose zone and in groundwater: laboratory testing and field evaluation’, Ground Water Monitoring & Remediation, vol. 28, no. 2, pp. 6874.
Patterson, BM & Davis, GB 2009, ‘Quantification of vapour intrusion pathways into a slab-on-ground building under varying environmental conditions’, Environmental Science and Technology, vol. 43, no. 3, pp. 650656.
Patterson, BM, Davis, GB & Johnston, CD 1999, ‘Automated in situ devices for monitoring of VOCs and oxygen in water and soil environments’, in Johnston, CD (ed.), Contaminated site remediation: challenges posed by urban and industrial contaminants; proceedings of the Contaminated site remediation conference, Fremantle, Western Australia, pp. 227234.
Patterson, BM, Davis, GB & McKinley, AJ 2000, ‘Volatile organic compounds in groundwater, probes for the analysis of’ In Mayers, RA (ed.), Encyclopedia of analytical chemistry: instrumentation and application, John Wiley and Sons Ltd, pp. 35153526.
Post, V, Kooi, H & Simmons, C 2007, ‘Using hydraulic head measurements on variable-density ground water flow analyses’, Ground Water, NovDec 2007, pp. 664671.
Puls, RW & Barcelona, MJ 1996, Low-flow (minimal drawdown) groundwater sampling procedure, Report EPA/540/S-95/504, US Environmental Protection Agency, pp. 12.
Safe Work Australia 2011a, How to manage and control asbestos in the workplace Code of Practice.
Safe Work Australia 2011b, How to safely remove asbestos Code of Practice.
Scalf, MR, McNabb, JF, Dunlap, WJ, Cosby, RL & Fryberger, J 1992, Manual of groundwater sampling procedures, Joint report of the US National Groundwater Association and the US EPA, Robert S. Kerr Laboratory.
SA EPA 2005, Composite soil sampling in site contamination assessment and management, Environment Protection Agency, Adelaide, South Australia.
SA EPA 2007, Guidelines: regulatory monitoring and testing groundwater sampling, Environment Protection Agency, Adelaide, South Australia.
SA EPA 2009, Site contamination: guidelines for the assessment and remediation of groundwater contamination, Environment Protection Agency, Adelaide, South Australia.
Serfes, ME 1991, ‘Determining the mean hydraulic gradient of groundwater affected by tidal fluctuations’, Ground Water, 29, number 4, pp. 549555.
Sundaram, B, Feitz, A, Caritat, P, de Plazinska, A, Brodie, R, Coram, J & Ransley, T 2009, Groundwater sampling and analysis – a field guide, Record 2009/27 95, Geoscience Australia.
Tufte, ER 1983, The visual display of quantitative information, Graphics Press, Cheshire, Connecticut, USA.
Turczynowicz, L 1991, ‘Approaches to analyte selection’, in El Saadi, O & Langley, AJ (eds), The health risk assessment and management of contaminated sites, South Australian Health Commission, Adelaide, Australia.
US EPA 1990, Sub-surface contamination reference guide, EPA/540/2-90/011, US Environmental Protection Agency, Washington, DC.
US EPA 1996, Soil screening guidance: technical background, EPA/540/R-95/128, US Environmental Protection Agency, Virginia, USA.
US EPA 2000a, Guidance for the data quality objective process, EPA/QA/G-4, US Environmental Protection Agency, Washington, DC.
US EPA 2000b, Data quality objectives process for hazardous waste site investigations, EPA/QA/G-4HW, US Environmental Protection Agency, Washington, DC.
US EPA 2002, Guidance on choosing a sampling design for environmental data collection, EPA/QA/G-5S, US Environmental Protection Agency, Office of Environmental Information, Washington, DC.
US EPA 2006a, Guidance on systematic planning using the data quality objectives process, EPA QA/G-4, US Environmental Protection Agency, Office of Environmental Information, Washington DC.
US EPA 2006b, Data quality assessment: statistical methods for practitioners, EPA/QA/G-9S, US Environmental Protection Agency, Office of Environmental Information, Washington, DC.
US EPA 2006c, Data quality assessment: A Reviewer’s Guide, EPA/QA/G-9, US Environmental Protection Agency, Office of Environmental Information, Washington, DC.
US EPA 2007a, ProUCL version 4.00.04 user guide, EPA/600/R-07/038, US Environmental Protection Agency, Washington, DC. (also see
www.epa.gov/OSP/hstl/tsc/proucl-4-0-02-user.pdf.
US EPA 2007b, SW-846, Method 6200, Field Portable X-Ray Fluorescence Spectrometry for the determination of elemental concentrations in soil and sediment, Revision 0, February 2007.
US EPA 2009, DNAPL remediation: selected projects where regulatory closure goals have been achieved, EPA/542/R-09/008, US Environmental Protection Agency, Washington, DC.
US EPA 2012a, Petroleum hydrocarbons and chlorinated hydrocarbons differ in their potential for vapour intrusion, Office of Underground Storage Tanks, Washington, DC, March 2012.
US EPA 2012b, EPA’s Vapor Intrusion Database: Evaluation and Characterisation of Attenuation Factors for Chlorinated Volatile Organic Compounds and Residential Buildings, EPA/530/R-10/002, March 16, 2012.
van Alphen, M 1993, 'A format for site history reports' in The health risk assessment and management of contaminated sites; proceedings of the second national workshop, by Langley, AJ & van Alphen, M, South Australian Health Commission, Adelaide, Australia.
Van den Berg, M, Birnbaum, LS, Denison, M, De Vito, M, Farland, W, Feeley, M, Fiedler, H, Hakansson, H, Hanberg, A, Haws, L, Rose, M, Safe, S, Schrenk, D, Tohyama, C, Tritscher, A, Tuomisto, J, Tysklind, M, Walker, N & Peterson, RE 2006, ‘The 2005 World Health Organisation re-evaluation of human and mammalian toxic equivalency factors for dioxins and dioxin-like compounds’, Toxicological Sciences, vol. 93, pp. 223241.
WA DEC 2009, Identification and investigation of acid sulfate soils and acidic landscapes, Acid Sulfate Soils Guideline Series, Western Australian Department of Environment and Conservation, Perth, Western Australia.
WA DEC 2011, Treatment and management of soils and water in acid sulfate soil landscapes. Acid Sulfate Soils Guideline Series, Western Australian Department of Environment and Conservation, Perth, Western Australia.
WA DoE 2004, Potentially contaminating activities, industries and land uses, Contaminated sites management series, Western Australian Department of Environment, Perth, Western Australia.
WA DoH 2009a, Guidelines for the assessment, remediation and management of asbestos-contaminated sites in Western Australia, Western Australian Department of Health and Western Australian Department of Environment and Conservation, Perth, Australia.
WA DoH 2009b, Management of small-scale low-risk soil asbestos contamination, Western Australian Department of Health, Perth, Australia.
WA DoH 2012, Summary of the Guidelines for the assessment, remediation and management of asbestos-contaminated sites in Western Australia—May 2009, Western Australian Department of Health, Perth, Australia.
Weaver, J 1992, Groundwater sampling: a comprehensive guide for sampling methods, Report 339, South African Water Research Commission.
Wilson, S, Oliver, S, Mallett, H, Hutchings, H & Card, G 2007, Assessing risks posed by ground gases to buildings, CIRIA C665, Construction Industry Research and Information Association, London, UK.
WHO 1989, Polychlorinated dibenzo-p-dioxins and dibenzofurans, Environmental Health Criteria #88, International Program on Chemical Safety, World Health Organization.
17 Appendix A: Possible analytes for soil contamination
This list is indicative only and analytes for analysis should be selected based on site history.
Inorganic contaminants
Analysis name | |
Metals | Where a general purpose screen for metal contamination in soils is indicated, it may include: Arsenic Cadmium Chromium Copper Lead Manganese Mercury Nickel Zinc If more detailed investigation is indicated, soil may be examined for: Aluminium Antimony Barium Beryllium Boron Calcium Cobalt Iron Magnesium Molybdenum Potassium Selenium Silver Strontium Thallium Tin Vanadium |
Anions | Where a general purpose screen for anion contamination in soils is undertaken, it may include: Bromide Iodide Sulfate Chloride Nitrate and Nitrite Sulfide Cyanide Phosphate Fluoride |
Organic contaminants
Analysis name | |
Monocyclic aromatic hydrocarbons (MAHs) | Where a general purpose screen for MAH contamination in soils is undertaken, it may include: Benzene Toluene ortho-Xylene meta- Xylene (para- Xylene) Ethyl benzene Styrene (vinyl benzene) Cumene (isopropylbenzene) 1,3,5 Trimethylbenzene 1,2,4-Trimethylbenzene 1-Methyl-4-isopropylbenzene n-Propylbenzene n-Butylbenzene iso-Butylbenzene tert-Butylbenzene sec-Butylbenzene If more detailed investigation is indicated, the following analytes may be included: Chlorobenzene 1,2-Dichlorobenzene 1,3-Dichlorobenzene 1,4-Dichlorobenzene Nitrobenzene Dinitrobenzenes Nitrotoluene Dinitrotoluenes Trinitrotoluenes |
Polycyclic aromatic hydrocarbons (PAHs) | Where a general purpose screen for PAH contamination in soils is undertaken, it may include: Naphthalene Benzo(a) anthracene Acenaphthylene Chrysene Acenaphthene Benzo(b) fluoranthene Fluorene Benzo(k) fluoranthene Phenanthrene Benzo(a) pyrene Anthracene Dibenz (a,h)anthracene Fluoranthene Benzo(ghi) perylene Pyrene Indeno(123-cd) pyrene |
Phenols | Where a general purpose screen for phenols contamination in soils is undertaken, it may include: Phenol o-Cresol p-Cresol 2,3-Dimethylphenol 2,4-Dimethylphenol 2,5-Dimethylphenol 2,6-Dimethylphenol 3,4-Dimethylphenol 3,5-Dimethylphenol 2,3,5-Trimethylphenol 2,3,6-Trimethylphenol 2,4,6-Trimethylphenol If more detailed investigation is indicated, the following may be included: 2-Nitrophenol 4-Nitrophenol 2,4-Dinitrophenol |
Chlorinated phenols | Where a general purpose screen for chlorinated phenols contamination in soils is undertaken, it may always include: 2-Chlorophenol 3-Chlorophenol 4-Chlorophenol 2,4 –Dichlorophenol 2,6 –Dichlorophenol 2,4,5 –Trichlorphenol 2,4,6 –Trichlorphenol 2,3,4,5-Tetrachlororphenol 2,3,4,6-Tetrachlororphenol 2,3,5,6 –Tetrachlororphenol Pentachlorophenol |
Chlorinated benzenes | Where a general purpose screen for chlorinated benzenes contamination in soils is undertaken, it may include: Chlorobenzene 1,2-Dichlorobenzene 1,3-Dichlorobenzene 1,4-Dichlorobenzene 1,2,3-Trichlorobenzene 1,2,4-Trichlorobenzene 1,2,4,5 Tetrachlorobenzene Pentachlorobenzene Hexachlorobenzene |
Organochlorines (OCs) | Where a general purpose screen for OCs contamination in soils is undertaken it may include: Aldrin HCB alpha-HCH, beta-HCH gamma-HCH (lindane), delta-HCH Chlordane DDD, DDE, DDT Dieldrin Endrin Endosulfan (alpha-, beta- and sulfate) Heptachlor, Heptachlor epoxide Methoxychlor Toxaphene (chlorcam, campheclor) Where site history indicates possible PCB contamination, the following may be included: PCB (Aroclors 1016, 1221, 1232, 1242, 1248, 1254 and 1260) |
Organophosphorus insecticides (OPs) | Where a general purpose screen for OP contamination in soils is undertaken it may include: Chlorpyrifos Coumaphos Diazinon Dichlorvos Dimethoate Ethion Fenthion Malathion Parathion methyl Parathion ethyl If more detailed investigation is indicated, the following may be included: Azinphos methyl Sulprofos Demeton-s-methyl Disulfoton Ethoprophos Mevinphos Monocrotophos Naled Phorate Prothiophos Tetrachlorvinphos A Nitrogen/Phosporus Detector (NPD) or flame photometric detector (FPD) or GC/MS should be employed for screening purposes. |
Acid/phenoxy herbicides | Where a general purpose screen for acid herbicides contamination in soils is undertaken, it may include: 2,4-D 2,4-DB 2,4,5-T 2,4,5-TP (Silvex) Dicamba and 5-Hydroxydicamba MCPA MCPP 4-Nitrophenol If more detailed investigation is indicated, the following may be included: Acifluoren Bentazon Dichlorprop Dalapon Picloram |
Triazine herbicides | Where a general purpose screen for triazine herbicide contamination in soils is undertaken, it may include: Atrazine Ametryn Prometryn Simazine Hexazinone |
Phthalate esters | Where a general purpose screen for phthalate contamination in soils is undertaken, it may include: Bis (2-ethylhexyl) phthalate Butyl benzyl phthalate Di-n-butyl phthalate Dicyclohexyl phthalate Diethyl phthalate Dihexyl phthalate Diisobutyl phthalate Dimethyl phthalate Dinonyl phthalate Di-n-octyl phthalate If more detailed investigation is indicated, the following may be included: Bis (2-n-butoxyethyl) phthalate Bis (2-ethoxyethyl) phthalate Bis (2-methoxyethyl) phthalate Bis (4-methyl-2-pentyl) phthalate Diamyl phthalate Hexyl 2-ethylhexyl phthalate |
The DQO process is a seven-step iterative planning approach that is used to define the type, quantity and quality of data needed to inform decisions relating to the environmental condition of a site. The summary of the process below is adapted from US EPA (2006a) and NSW DEC (2006).
The DQO process should commence before any investigative work starts, with the timing for various stages of the project being clearly understood by all parties. It is useful to apply the process initially at a project level to determine the overall project requirements and then modified as required for specific investigation activities.
The seven steps in the DQO process are:
Step 1: State the problem
Step 2: Identify the decision/goal of the study
Step 3: Identify the information inputs
Step 4: Define the boundaries of the study
Step 5: Develop the analytical approach
Step 6: Specify performance or acceptance criteria
Step 7: Develop the plan for obtaining data
The first step involves summarising the contamination problem that will require new environmental data and identifying the resources available to resolve the problem. A preliminary CSM will be required to complete this step. The matters to consider at this stage include:
- the objective of the proposed investigation, noting that the ability to meet objectives may be limited by constraints such as time, resources, climatic conditions and access restrictions
- the possible content of a problem statement that gives a brief summary of the contamination issue(s) at the site that is to be addressed in the project
- the reason the project is being undertaken
- identification of the project team and technical support experts, such as field manager/site supervisor, field personnel, toxicologists, risk assessors and statisticians
- budget and community concern issues that may also be factors in designing and carrying out the environmental assessment
- identification of the regulatory authority(ies) and the local government area.
Step 1 of the DQO process should assist in developing the following:
- a concise description of the problem
- a list of the planning team members and identification of decision-maker
- a summary of available resources and relevant deadlines for the study
- a preliminary conceptual model of the site, based on available information prior to the commencement of the site investigation, covering:
- previous investigations
- present and historical use(s) of the site and adjacent sites
- geology, hydrogeology
- potential contaminants of concern
- potential contaminant migration pathways both to and from the site (such as waterways, drains, service conduits)
- areas of environmental concern (drawings showing chemical storage, use, disposal)
- media in which potential contaminants of concern may be present and through which they may migrate (habitat(s) of contamination, lateral and depth extent, temporal and climatic variability)
- potential exposure pathways to human and/or environmental receptors
- future land uses.
The conceptual model of contamination of the site that is produced at this early point can be progressively refined through subsequent stages of the assessment.
The second step involves identifying the decisions that need to be made about the contamination problem and the new environmental data required to make them.
The objective(s) of the data collection part of the investigation is project-specific and may be identified by:
- referring to the history of use of the site, chemicals of concern and likely concentration range(s), media that may be impacted and likely migration routes, such as groundwater, surface water flow, wind, and service trenches
- considering relevant site criteria for each medium (fill, soil, sediment, groundwater, surface water, air)
- making a series of decision statements that need to be addressed (e.g. a decision statement could consider whether parts of the site would be suitable for a proposed use if the 95% UCL on the mean concentrations for all chemicals of potential concern were less than the appropriate site criteria).
Step 2 of the DQO process should assist in developing a decision statement linking the principal project objective(s) to the possible actions that will address the problem.
The existing conceptual model can then be reviewed to determine whether existing data is satisfactory to complete the investigation or whether data gaps or an unacceptable level of uncertainty exists.
The third step involves identifying the information needed to support any decision and whether new environmental data will be needed.
Decisions made during this step are of a draft or preliminary nature and are reviewed in Step 7 to develop the sampling analytical and quality plan (SAQP).
Step 3 of the DQO process should assist decision-makers to resolve decision statements and make informed, defensible decisions by identifying:
- the media that needs to be collected, such as fill, soil, groundwater, sediments, surface water and air
- the environmental parameters that will be measured for each media
- site criteria for each medium of concern
- analytical methods that are required for chemicals of potential concern so that assessment can be made relative to the site criteria
- the basis for any decisions that are to be made from field screening, such as from PID data, and what action is to be taken if a defined concentration is attained
- any additional information required to make the required decisions.
The fourth step involves specifying the spatial and temporal aspects of the environmental media that the data must represent to support decision(s). The matters to consider at this stage include:
- the geographical extent of the proposed investigation
- time and budget constraints
- spatial extent (property boundaries, accessibility constraints to parts of the site, potential exposure areas)
- temporal boundaries (the time frame of the investigation, taking into account seasonal conditions, presence of near-surface groundwater or surface water and discharges, access restrictions, availability of key personnel)
- for large sites, the boundaries of each segment to be investigated (based on proposed use of each area of the site, which will influence the required sample density, appropriate regulatory guidance)
- the lateral and vertical intervals in which contamination distribution is believed to be uniformly distributed
- the scale of decisions required: site-wide, each residential lot, etc.
- the presence of any heterogeneous materials that may require specific sampling methods
- potential constraints to carrying out the investigation, such as access, presence of infrastructure, health and safety issues.
Step 4 of the DQO process should assist in developing:
- a detailed description of the spatial and temporal boundaries of the problem
- an understanding of any practical constraints that may interfere with the assessment.
The fifth step involves defining the parameter of interest, specifying the action level, and integrating information from Steps 1–4 into a single statement that gives a logical basis for choosing between alternative actions. Acceptable limits should be defined for the following:
- chemicals of concern detected in field blanks, rinsate blanks, volatile-spiked trip samples, laboratory method blanks
- recovery of matrix spike additions, surrogate spike additions, laboratory control samples
- relative percent differences (RPDs) of matrix spike and matrix spike duplicates.
Step 5 of the DQO process should assist in producing:
- the statistical parameter (the parameter of interest) that characterises the population
- confirmation that the action level exceeds measurement detection limits
- an ‘if …, then …’ statement that defines the conditions that would cause a decision-maker to choose from alternative actions.
The sixth step involves specifying the decision-maker’s acceptable limits on decision errors, which are used to establish performance goals for limiting uncertainties in the data. (For more information about decision errors and decision-making, see notes at the end of this Appendix). Some of the matters to consider at this stage include:
- determination of the possible range of the parameter of interest
- identification of decision errors and formulation of the null hypothesis
- specification of a range of possible parameter values where the consequences of decision errors are relatively minor (grey region)
- assignation of probability values to points above and below the action level that reflect the tolerable probability for the occurrence of decision errors.
Step 6 of the DQO process should assist in calculating the decision-maker’s tolerable decision error rates based on a consideration of the consequences of making an incorrect decision.
The seventh step involves identifying the most resource-effective sampling and analysis design for generating the data that is required to satisfy the DQOs.
Step 7 of the DQO process should assist in developing:
- the most resource-effective design for the study that is expected to achieve the DQOs
- the optimum manner in which to collect the data required to meet the objectives for the assessment and which will meet the project DQOs
- the SAQP.
Decision errors are incorrect decisions caused by using data that is not representative of site conditions due to sampling or analytical error. As a result, a decision may be made that site clean-up is not needed when really it is, or vice versa.
There are two types of decision error:
- sampling errors occur when the sampling program does not adequately detect the variability of a contaminant from point to point across the site. That is, the samples collected are not representative of the site conditions (e.g. an appropriate number of representative samples have not been collected from each stratum to account for estimated variability)
- measurement errors occur during sample collection, handling, preparation, analysis and data reduction.
The combination of the above errors is referred to as ‘total study error’. This directly affects the probability of making decision errors. Study error is managed through the correct choice of sample design and measurement systems. Note that the attainment of a nominated probability generally requires use of a statistically based sampling plan.
The possibility of making a decision error, although small, is undesirable because of the adverse consequences arising from that incorrect decision. Decision error can be controlled through the use of hypothesis testing. This test can be used to show either that the baseline condition is false (and therefore the alternative condition is true) or that there is insufficient evidence to indicate that the baseline condition is false (and therefore the site assessor decides by default that the baseline condition is true).
The burden of proof is placed on rejecting the baseline condition, because the test hypothesis structure maintains the baseline condition as being true until overwhelming evidence is presented to indicate that the baseline condition is not true.
The null hypothesis is an assumption assumed to be true in the absence of contrary evidence, for example, that the site is contaminated unless proved to be clean.
If we reject a hypothesis when it should be accepted, we say that a type I error has been made. If, on the other hand, we accept a hypothesis when it should be rejected, we say that a type II error has been made. In either case, a wrong decision or error in judgment has occurred:
- type I error (false positive decision error) — rejecting the hypothesis as false when it is really true
- type II error (false negative decision error) — accepting the hypothesis as true when it is really false.
In order for decision rules (or tests of hypotheses) to be sound, they must be designed to minimise decision errors. This is not always simple, as for any given sample size, an attempt to decrease one type of error is generally accompanied by an increase in the other type of error. The only way to reduce both types of error is to increase the sample size, which may or may not be always possible.
In testing a given hypothesis, the maximum probability with which we would be willing to accept a type I error is referred to as the ‘level of significance’ or significance level of the test. A significance level of 0.05 or 0.01 is commonly adopted, although other values are used.
If for example the 0.05 (or 5%) significance level is selected for a decision rule, then we are accepting that there is a 1 in 20 (that is, 5 chances in 100) chance that we would reject the hypothesis when it should be accepted; that is, we are about 95% confident that we have made the right decision. In this case we say that the hypothesis has been rejected at the 0.05 significance level, which means that the hypothesis has a 0.05 probability of being wrong.
Source: NSW DEC, 2006.
Contaminated site practitioners should undertake an assessment of the reliability of field procedures and analytical results using the data quality indicators (DQI) of precision, accuracy, representativeness, completeness and comparability. DQI should be used to document and quantify compliance or otherwise with the requirements of the project SAQP.
The DQI for chemical data will differ depending on which analytical methods have been used in a site assessment. These fall into three main categories:
- field methods
- laboratory screening methods
- methods specific for contaminants that are known or expected to be present at a site.
The following issues should be documented and discussed in assessment reports:
- the applicability and limitations of field methodologies where used
- instrument calibration and validation of field measurements, and comparison with laboratory results
- the significance of the results of field screening methods compared with the results of laboratory analyses, for example, that the results reported for field screening using a photo-ionisation detector are compatible with the results reported by the laboratory for volatile organic compounds. Where not compatible, an adequate explanation should be provided.
Laboratory screening methods are used to determine the type of contamination present and the constituents of a sample that might cause interferences in specific methods. Assessment reports should include appropriate discussion of the applicability and limitations of any screening methodologies used.
DQI for screening methods may be less rigorous than for specific analytical methods. Nevertheless, screening method performance should be known and should be expressed as a multiple of specific analytical method performance.
Site assessors should ensure that appropriate discussion and documentation about the following issues is included in the assessment report:
- that the analytical methods used for site validation are of appropriate precision and accuracy, and that the sensitivity and selectivity of the analytical methods are appropriate for the assessment of the risk
- that the precision and accuracy criteria set out in the QA/QC plan, for a given method and matrix, meet the performance expected of the reference method
- that the quality of data supplied by the analytical laboratory meets the objectives of the testing laboratory’s quality plan for at least 95% of test results for blanks, spikes, control samples, duplicates and holding times. (Note that these DQOs do not refer to field duplicate reproducibility or other measures of sampling variance. Sampling variance should be addressed in the choice of sampling method.)
Contaminated site practitioners should undertake an assessment of the DQIs that relate to both field and laboratory procedures, and provide appropriate documentation in the assessment report.
Completeness
A measure of the amount of usable data (expressed as %) from a data collection activity |
Field considerations | Laboratory considerations | Comments |
All critical locations sampled All samples collected (from grid and at depth) Standard operating practices (SOPs) appropriate and complied with Experienced sampler Documentation correct | All critical samples analysed according to SAQP All analytes analysed according to SAQP Appropriate methods and PQLs Sample documentation complete Sample holding times complied with | The required percentage completeness should be specified in the SAQP All required data must be obtained for critical samples and chemicals of concern Incompleteness is influenced by: - field performance problems (access problems, difficulties on site, damage…)
- laboratory performance problems (matrix interference, invalid holding times…)
- matrix problems
|
Comparability
The confidence (expressed qualitatively) that data may be considered to be equivalent for each sampling and analytical event |
Field considerations | Laboratory considerations | Comments |
Same SOPs used on each occasion Experienced sampler Climatic conditions (temperature, rainfall, wind…) Same types of samples collected (filtered, size fractions...) | Sample analytical methods used (including clean-up) Sample PQLs (justify/quantify if different) Same laboratories (justify/quantify if different) Same units (justify/quantify if different) | Same approach to sampling (SOPs, holding times…) Quantify influence from climatic or physical conditions Samples collected, preserved, handled in same manner (filtered, same containers) |
Representativeness
The confidence (expressed qualitatively) that data is representative of each medium present on the site |
Field considerations | Laboratory considerations | Comments |
Appropriate media sampled according to SAQP All media identified in SAQP sampled | All samples analysed according to SAQP | Samples must be collected to reflect the characteristics of each medium Sample analyses must reflect properties of field samples Homogeneity of the samples Appropriate collection, handling, storage and preservation Detection of laboratory artefacts, e.g. contamination blanks |
Precision
A quantitative measure of the variability (or reproducibility) of data |
Field considerations | Laboratory considerations | Comments |
SOPs appropriate and complied with | Analysis of: - laboratory and inter-laboratory duplicates
- field duplicates
- laboratory-prepared volatile trip spikes
| Measured by the coefficient of variance or standard deviation of the mean or by RPDs Field duplicates measure field and laboratory precision Laboratory duplicates measure analytical precision* |
Accuracy (bias)
A quantitative measure of the closeness of reported data to the true value |
Field considerations | Laboratory considerations | Comments |
SOP appropriate and complied with | Analysis of: - field blanks
- rinsate blank
- reagent blank
- method blank
- matrix spike
- surrogate spike
- reference material
- laboratory control sample
- laboratory-prepared spikes
| Bias introduced: - by chemicals during handling or transport
- from contaminated equipment
- from contaminated reagent
- during laboratory analysis
- during laboratory preparation and analysis (may be high or low)
- precision of preparation and analytical method
- during laboratory analysis
- during collection/
transport (may be high or low)
|
* Laboratory duplicates measure analytical precision when the sample is totally homogenous. When sample heterogeneity exists, laboratory duplicates (and intralaboratory splits) measure the sum of laboratory precision plus sample heterogeneity. High sample heterogeneity impacts confidence in data and may warrant additional sampling to increase confidence or detect hotspots.
Environmental practitioners should ensure that the following issues are addressed in the field QA/QC program and that appropriate documentation is included in the assessment report:
- replicate samples are split in the field and submitted to two separate laboratories in accordance with the requirements of Schedule B3
- the sampling program includes assessment of all relevant environmental media, including soil, dust, surface water, groundwater, air, sediments and biota as appropriate
- the sampling strategy is appropriate for the conditions at the site and the nature of the contamination, with the rationale for the strategy described in the assessment report and the sampling locations shown on a scaled site sampling plan
- sample collection, handling and transportation procedures are documented and appropriate to meet the project DQOs
- sampling is representative of site conditions, based on the selection of appropriate numbers of sampling points and of samples from each relevant strata and material types stated in a site sampling plan to meet the project DQOs
- the field QA/QC plan includes details of:
- the sampling team
- sampling method(s), including the actual methods employed for obtaining samples, type(s) of sample containers, order and degree of filling, preservation, labelling, logging, custody
- evidence of appropriate decontamination procedures carried out between sampling events
- completed logs for each sample collected, showing time, location, initials of sampler, duplicate locations, duplicate type, chemical analyses to be performed, site observations and weather conditions
- completed chain-of-custody documentation, identifying for each sample the name of the sampler, the nature of the sample, collection date, analyses to be performed, sample preservation method, departure time from the site and dispatch courier(s) and condition of samples at dispatch
- sample splitting techniques
- a statement of duplicate frequency for intra-laboratory and inter-laboratory duplicate samples and duplicate sample results
- field blank results
- background sample results
- rinsate sample results
- laboratory-prepared trip spike results for volatile analytes
- trip blank results
- field instrument calibration for instruments used on site.
Environmental practitioners should ensure that the following issues are addressed in the laboratory QA/QC program and that appropriate documentation is included in the assessment report:
- sample analyses use appropriate methodologies for each potential contaminant in the matrix in laboratories accredited for those analyses by the National Association of Testing Authorities (NATA) or an equivalent government-endorsed provider of accreditation for laboratories
- appropriate practical quantitation limits (PQLs) for the chemicals of concern for use in the assessment of risk
- a laboratory QA/QC plan with the following information:
- a copy of signed chain-of-custody forms acknowledging receipt date and time, conditions of samples on receipt and identity of samples included in shipments
- record of holding times and a comparison with method specifications
- analytical methods used
- laboratory accreditation for analytical methods used
- laboratory performance in inter-laboratory trials for the analytical methods used, where available
- the results for blind duplicate samples collected from the field.
The site assessment reports should include documentation of QA/QC procedures including all information relevant to the site assessment:
- the QA/QC checklist items (see Section 19.10), related to field quality assurance and quality control, laboratory QA/QC and data evaluation QA/QC
- the names of the accredited laboratories used and relevant details of their accreditation for each analytical method
- the limits of reporting (ensuring that appropriate assessment can be made according to site criteria as stated in the DQOs for relevant media)
- the acceptance limit(s) for each QC test, such as duplicate RPDs and recoveries for laboratory quality control analyses
- where used, the origin of certified reference material (CRM), its batch number and the concentrations of the chemicals of potential concern
- the QC results relevant to the sample analysis
- for each sample, the highest measurement result wherever replicate measurements are taken (or all measurement results for each sample)
- results for all data tabulated separately according to each type of soil, fill, groundwaters, surface waters and sediments, with appropriate statistical analysis
- the laboratory specifying compliance with the requirements of Schedule B3 and equivalence with the reference method or non-standard methods.
Field quality assurance and quality control
details of sampling team
decontamination procedures carried out between sampling events
field logs for samples collected — including time, location, initials of sampler, duplicate locations, duplicate type, chemical analyses to be performed, site observations and weather conditions
chain-of-custody fully identifying (for each sample) the sampler, nature of the sample, collection date, analyses to be performed, sample preservation method, departure time from the site and dispatch courier(s)
sample splitting techniques
statement of duplicate frequency
field blank results
background sample results
rinsate sample results
laboratory-prepared trip spike results for volatile analytes
trip blank results
field instrument calibrations (when used)
Laboratory QA/QC
a copy of the signed chain-of-custody forms acknowledging receipt date and time, and identity of samples included in shipments
record of holding times and a comparison with method specifications
analytical methods used
laboratory accreditation for analytical methods used
laboratory performance in inter-laboratory trials for the analytical methods used, where available
description of surrogates and spikes used
percent recoveries of spikes and surrogates
instrument detection limit
method detection limits
matrix or practical quantification limits
standard solution results
reference sample results
reference check sample results
daily check sample results
laboratory duplicate results
laboratory blank results
laboratory standard charts
QA/QC data evaluation
evaluation of all QA/QC information listed above against the stated DQOs including a discussion of:
- documentation completeness
- data completeness
- data comparability
- data representativeness
- precision and accuracy for both sampling and analysis for each analyte in each environmental matrix informing data users of the level of reliability or qualitative value of the data
results of data comparability checks to assess bias that may arise from various sources, including:
- collection and analysis of samples by different personnel
- use of different methodologies
- collection and analysis by the same personnel using the same methods but at different times
- spatial and temporal changes (because of environmental dynamics)
relative percent differences for intra- and inter-laboratory duplicates.
Figure 2. Example site layout overlay
