[Report here IAEA, tech Rep No. 486]
This publication addresses the sampling of soil and vegetation in terrestrial ecosystems, including agricultural, forest and urban environments, contaminated with radionuclides from events such as radiation accidents, radiological incidents and former nuclear activities.
A big issue with surveys of radioactivity in soil and vegetable following a major atmospheric release of radioisotopes is the complexity. We are 
generally interested in the dose implications of the release to people. There are various pathways to take into account; inhalation dose, cloud dose, ground dose, resuspension dose, direct contamination of foods and uptake into the food chain. You have the situation where a cloud with a range of radionuclides with a range of physical and chemical forms meets a highly complex and heterogeneous system i.e. the real world. How do you decide what to measure? How do you ensure that your measurements are representative? What are your monitoring objectives?
A monitoring programme might be required after a nuclear accident and should be implemented:
- To help safeguard the environment;
- To assess hazard, risk and effective response arrangements;
- To provide public reassurance;
- To assess the impact on wildlife;
- To assess the dose to a representative person;
- To generate data to serve as a reliable database, to establish a baseline, or to substantiate compliance with laws and regulations;
- To provide an independent check on the monitoring or modelling undertaken;
- To detect abnormal, fugitive and unauthorized releases;
- To support legal or regulatory action or to be used in ascertaining compensation and liability in case of spills or accidents;
- To delineate boundaries for clean areas or to establish priorities and thresholds for the cleanup of contaminated sites;
- To ascertain the type of treatment or disposal required for cleaning contaminated sites;
- To understand or assess the long term trends on the behaviour of radionuclides in the environment or the accumulated impact from licensed discharges.
Chapter 2 states that “soil and vegetation can become contaminated when radioactive solids, liquids or vapours are deposited on the surface, mixed with the soil or contaminated from a groundwater source”. It then goes on to discuss some of the factors that affect the fate of deposited radionuclides and the pathways by which people can be exposed to radiation dose (external radiation, skin contamination, inhalation and ingestion).
The IAEA report gives a number of lists of factors that should be considered when a sampling programme is designed. It warns that many sampling programmes fail to achieve their aims because of a failure to take account of the many variables that affect radionuclide movement in the environment.
Warnings such as “the inhomogeneous distribution of contaminants is often the largest contributor to uncertainty in the data and is usually not quantified”; “It is not uncommon for the concentrations of target analytics in soil samples collected within a short distance (e.g. 1 m) to have differences of 50–100%”, “Extreme spatial heterogeneity, such as the presence of ‘hot’ particles (particles of anomalously high activity) in samples can cause large errors in extrapolating the data”; “It is therefore necessary to know about: (i) the source of radioactive contamination; (ii) the physical and chemical characteristics of the radioactive material; and (iii) its depth migration into soils to obtain a representative sample from a field site” are sprinkled through the report.
The report identifies a number of different sampling strategies, judgemental sampling; simple random sampling; two stage sampling; stratified sampling; systematic grid sampling; systematic random sampling; cluster sampling; double sampling; search sampling; transect sampling. They define each of these strategies and have a useful table suggesting which might be appropriate for differ objectives (Table 2.1). This would be a very useful discussion to read prior to starting a major new sampling programme.
There is also a useful discussion about treating the variability of contamination within an area as being composed of three components; the trend across the site, localised variations (hot spots) and random variations. This analysis helps make sense of the distribution of results if applied appropriately.
Section 2.4.1 contains a discussion about the migration of radionuclides downwards into the ground. This depends on the chemical properties of the radionuclide and the soil in question. For example tritium can migrate with soil water and penetrate below root depth very quickly. On the other hand most plutonium is found in the surface few centimetres years after deposition.
An optimal sampling programme achieves the maximum number of objectives, is undertaken in accordance with appropriate quality standards. It is also fundamental that performance criteria (e.g. monitoring and sampling uncertainty, detection limits and confidence levels) are set to meet the objectives, while simultaneously ensuring proportionality and taking account of the urgency of the information required.
In-situ gamma spectroscopy can yield large area data more quickly. These can be undertaken from manned helicopter or fixed wing aircraft, drones at various heights and speeds, vehicles or carried by personnel. Different equipment used in different ways gives a different layer of information. The higher, faster processes cover more area at the expense of detail, good for finding major hot spots. Slower, lower surveys give more idea of local variation. All tend to be variants of a detector (these come with a range of resolutions and efficiencies) coupled to a GPS and a data logger. In practice a layered approach is often optimum.
An approach mentioned by the IAEA is to take a series of measurements, maybe over a number of evenly spaced grid points, and to declare an area as uniformly contaminated if the variation of each measurement from the mean is less than a pre-set value (30% is mentioned). If the variation is greater than this then the area should be subdivided.
The differences in sampling residential, agricultural and forested areas is discussed.
Chapter four discusses the tools and methodologies for sampling in useful detail. Chapter five is about sample preparation. Six quality assurance and seven safety.
Case studies, including in depth reviews of the sampling around Chernobyl and Fukushima, makes up the remainder of the report.
This report is an excellent introduction to environmental sampling. It explains much of the complexity, which many people would underestimate, in an easy to understand manner.