Before beginning a sampling program is important to first identify the goals of the program (Correll 2004). This can be done by asking the questions: Is the data going to support a preliminary risk assessment or something more detailed? Are we interested in protecting ecological organisms, or just humans? Are we interested in the effects caused by one chemical group or a wide range of possible contaminants? Once the questions have been answered it is important to determine what information already exists, the pathways of exposure and available budget for the program.
What information already exists (or is easily attained) for the site?
Are there existing reports addressing contamination issues at the site such as concentration, type and extent? Do people or animals live or work on or near the site? A good tool for gathering useful information is a site questionnaire. (see the Hotspot identification questionnaire in example field data sheets)
What are the likely human exposure pathways which need to be assessed?
Although it may not be possible to complete a problem formulation before going to collect samples it is useful to consider the components of a problem formulation while developing the sampling plan (USEPA 2008). Specifically:
- What are the sources of contaminants (e.g. soil, water, food etc.);
- Who is possibly exposed (workers, local residents, children etc.); and
- What are the pathways linking the contaminants to the people (diet, inhalation of dust or smoke, direct contact via skin etc.).
What is the budget available?
Unless a risk assessment has already been performed at the site, the first sampling program for the risk assessment will be relatively simple and use conservative assumptions. The reason for starting simply is to avoid doing more work than is necessary because if it can be shown that risks are negligible using simple, conservative assumptions, then there is not need to proceed. One technique for keeping costs low is to use composite samples, to learn about when this is appropriate and how to conduct this type of sampling visit the USEPA website.
It is recommended that the data quality objectives (DQO’s) are considered during the sampling program design (USEPA 2008). This would include the number and types of quality assurance quality control (QA/QC) samples, analytical detection limits, as well as statistical power. More information on DQOs is provided in the USEPA’s document on DQO’s (PDF file) .
For more information
The following web resources are helpful for designing a sampling program:
- http://www.ccme.ca/assets/pdf/pn_1101_e.pdf (PDF file)
- http://www.ccme.ca/assets/pdf/pn_1274_e.pdf (PDF file)
learn about Quality Assurance and Quality Control (QA/QC)
References: Correll R and M Barnes. 2004 Sampling for Risk Assessment – What Works and What Can Go Wrong (PDF file), presentation in: Workshop on Contaminants and Ecological Risk Assessment, 5-7 April 2004, Adelaide, Australia. United States Environmental Protection Agency. 2008. Site Conceptual Model
Correll R and M Barnes. 2004 Sampling for Risk Assessment – What Works and What Can Go Wrong (PDF file), presentation in: Workshop on Contaminants and Ecological Risk Assessment, 5-7 April 2004, Adelaide, Australia.
United States Environmental Protection Agency. 2008. Site Conceptual Model