Radiation protection: site assessment

The trouble with scattered sources

3 December 2012

Radiation in contaminated land is often non-uniform, and may vary in radionuclides present and activity concentrations; it may be associated with discrete objects. In these cases, although generalisations are difficult, estimates of risk to health need to take into consideration land use (and plans for such), the population of sources, exposure pathways, and the probability of receiving a serious deterministic effect.

Land may become contaminated with radioactive material as a result of a number of past uses [1], current uses, or accidents. Examples include radionuclides in the ground as a result of nuclear accidents such as Chernobyl or accidental releases such as leaking pipes on nuclear or non-nuclear sites. Due to the various mechanisms that can result in radioactivity being present in land the level of radioactivity, the radionuclides, and the area where radioactivity is present, could vary considerably.

Radioactivity may also be associated with discrete objects, which contain levels of radioactivity per unit mass many orders of magnitude higher than the ambient level. Although the term ‘hot particle’ is commonly used to describe radioactivity associated with a discrete object, such objects can range in size from less than a millimetre to many centimetres. Typical examples include flakes of metal, scale from pipes, wires and radium dials. Such objects may therefore not resemble what may commonly be thought of as a particle. The term ‘discrete radioactive object’ is therefore preferred, recognising that it is important that a clear description of the form of the contamination is given in any assessment to help avoid misinterpretation by stakeholders.

The term ‘heterogeneous’ contamination is used to encompass land that is contaminated in a non-uniform way. This includes land where there are patches of contamination, land contaminated by discrete radioactive objects, or land that is contaminated with a mix of patches and objects.

A radiological assessment of contaminated land may be performed for a number of different purposes. These include assessing land as part of the delicensing process, where the aim is to demonstrate that there is ‘no danger’ after controls associated with a site license have been removed [2]; as part of the investigation into whether land should be determined as contaminated land under the UK’s Part 2A regime [3-7], or as input to a remediation strategy for land contaminated by historical events [8-10].

When assessing the radiological consequences of contamination by discrete radioactive objects it is important that both the dose and the probability of receiving that dose are taken into account, and presented separately. This aids interpretation of the results, particularly in situations where deterministic effects could result but where the probability of the dose being received is low. Where exposure is limited to the risk of stochastic effects, the risk to health can be considered to be proportional to the magnitude of the dose and the overall risk to health can be calculated by multiplying the probability of receiving the dose, the effective dose and the health risk factor.

Proposed guidance

Assessment of the radiological implications of heterogeneous contamination is not straightforward [11] and in 2009 the UK Health Protection Agency (HPA) committed to providing guidance in this area [12]. This paper presents an overview of a document being prepared by HPA that fulfils that commitment.

Details of the assessment calculations, such as the selection of appropriate land uses, the area to include in the assessment, the radioactivity level and the appropriate radiological criteria need to be determined for each site. The radiological assessment of land containing discrete radioactive objects poses particular difficulties and specific guidance is being developed for this situation. In all cases, care should be taken to avoid the use of overly cautious assumptions to compensate for the level of uncertainty.

The guidance will consider situations where land is being redeveloped for future use (planned exposure situations) and situations where people are already being exposed (existing exposure situations). It does not address the situation where land has become contaminated as a result of an accident and is still considered as an emergency exposure situation. The guidance will also not consider in any detail management options that are available for remediating contaminated land, or how suitable options should be selected; other guidance is available on this [13-15].

Outline of the proposed guidance

The guidance will be divided into a number of sections that cover the main aspects of a radiological assessment. Within each section a description of the aspects that need to be taken into account will be discussed, together with approaches that can be used to address these. Examples of aspects of a radiological assessment where HPA will offer guidance include the following:

  • Performing a tiered assessment, including guidance on performing a scoping assessment using generic data and how this can be refined where necessary in order to improve the level of confidence in the results (see also Table 1)
  • Defining suitable assessment areas, including guidance on how an area can be divided up into separate areas for assessment based on contamination location, radioactivity levels, radionuclide composition and land uses
  • Identifying a suitable source. For land contaminated by patches of radioactivity, guidance will be given on how to determine a representative activity concentration. For discrete radioactive objects, where the radioactivity content of the objects may cover many orders of magnitude, guidance will be given on how to select appropriate source terms.
  • Selection of appropriate exposure scenarios and pathways with respect to land use, exposed population groups and contaminated object size
  • Assessing the probability of encountering discrete radioactively-contaminated objects, including guidance on estimating object populations and accounting for dynamic environmental processes (see also Table 2)
  • Interpreting estimated doses and health risks including defining when the health risk could be considered to be ‘acceptable’ or ‘negligible’ (see also Table 3)
  • The use of sensitivity and uncertainty analysis within an assessment.

Table 1: Examples of initial questions to consider when performing an assessment
Table 2: Examples of questions relating to the probability of encountering a discrete radioactive object
Table 3: Examples of key considerations in assessing the radiological implications of contaminated land in existing exposure | situations


Since 2006 an intensive programme of monitoring for radioactive objects has been carried out on beaches in the vicinity of the Sellafield site in West Cumbria and a large number of radioactive objects have been identified and removed. These objects comprise particles with sizes smaller than or similar to grains of sand (less than 2 mm) and contaminated pebbles and stones. HPA carried out an assessment of the health risks to people using the beaches along the Cumbrian coast from contaminated objects on the beaches on behalf of the Environment Agency [16].

Due to the unique nature of the area around Sellafield, and also due to public perception, the researchers, Brown and Etherington, decided that their assessment would be as realistic as possible and would therefore use a considerable amount of site specific information, recognising that this would take time and effort to collect. As a consequence, specific surveys were carried out to determine the habits of users of the beaches in that area. In addition, laboratory investigations were carried out to determine the physical and chemical characteristics of the objects, including derivation of realistic gut uptake fractions. However, the assessment did make use of some generic parameter values, for example inadvertent ingestion rates, as no site-specific information was available.

As suggested in the proposed guidance, Brown and Etherington assessed the potential risk to health of exposure to radioactive objects by dividing the population of objects into groups that share similar characteristics, according to:

Object size
Objects were labelled as stones or particles depending on their size. Different pathways and parameter values were defined for these groups of sizes.
Emitted radiation
Objects were labelled as either ‘alpha-rich’, ‘beta-rich’ or ‘cobalt- rich’ objects. This grouping allowed the health effects to be determined for objects that led to similar doses and health risks.
Radioactivity level
Objects were placed within order-of-magnitude bands based on measured radioactivity content. This allowed more realistic estimates of the health risks from coming into contact with objects containing different levels of radioactivity, as the object population distribution was skewed towards objects with relatively low activity.

In deciding whether it was necessary to divide the beaches into smaller areas for an assessment (which is also suggested by the proposed guidance), Brown and Etherington took into account the available monitoring and habit information. In particular, it was noted that monitoring has not occurred across the whole beach due to limitations in using the vehicle-based monitoring system on non-sand areas and also that monitoring has been targeted to beaches with high occupancy and where the highest numbers of objects have been found. Habit surveys of beach use were undertaken along the affected coastline with most effort targeted to those beaches with suspected high occupancy. Based on the information gathered, Brown and Etherington divided the beach area into five separate beaches. These areas were selected to be those where sufficient objects had been found to allow an estimate of the instantaneous object population to be made, and also where sufficient habit data existed to define with confidence the characteristics of a set of representative persons. Although objects had been found in other areas it was decided that no quantitative assessment could be made due to the lack of suitable information and the resulting very large uncertainties. In those areas, only a subjective evaluation of the health risk was made.

Within Brown and Etherington, exposure scenarios were defined using information gathered by the habit surveys. In order not to require excessive amounts of calculation, beach use was grouped into broad categories of walking, bait-digging and leisure. These groups covered several land uses identified within the habit survey. For example, the walking group were intended to include those beach users who walked dogs, combed the beach for artefacts and general walkers. Account of different object size was made when defining appropriate exposure pathways within each scenario. For example, objects available for inhalation were limited to those that were relatively small, whilst larger objects were considered available for inadvertent ingestion.

In order to estimate the probability of encountering an object, the number of objects present should be used. Within Brown and Etherington, the population of objects on the beaches, within each of the groups listed above and for each pathway, where this was dependent on object size, was estimated. This calculation used the estimated detection probability of the monitoring system and information about the number of objects found, including the depth at which they were found. When estimating the object population, account was made of any repeat monitoring.

Brown and Etherington concluded that the overall health risks to beach users are very low and significantly lower than other risks that people accept when using the beaches. The highest calculated lifetime risks of radiation-induced fatal cancer are of the order of one hundred thousand times smaller than the level of risk that the Health and Safety Executive considers to be the upper limit for an acceptable level of risk (1 in a million) for members of the public and workers [17]. It is also very unlikely that deterministic effects, such as skin ulceration, could occur from encountering an object. As a result of the assessment, HPA recommended that regular monitoring of Sellafield beach (where the majority of the objects have been found) and monitoring at one or two other beaches with high public occupancy will provide regulators and the public with continued reassurance that the risks associated with radioactive objects remain very low.


It is important to recognise that every site has unique features that need to be explicitly considered when performing a radiological assessment. It is therefore not possible to specify an assessment methodology that could be used for all sites and contamination types. However, some over-arching guidance can be given reflecting the general approach that could be followed in most situations.

HPA aims to provide guidance on the radiological protection aspects of interpreting the results of a radiological assessment. An important point is that decisions made in relation to remediation of land where there is radioactivity present are not based solely on numerical estimates; account must also be made of stakeholder views, especially with regards to perceived levels of risk.


This article is based on a paper presented at the IRPA 13th international congress 13-18 May 2012, Glasgow, Scotland. The presenting authors at IRPA 13 were Wayne Oatway (wayne.oatway@hpa.org.uk) and Joanne Brown, Health Protection Agency, Centre for Radiation, Chemical and Environmental Hazards, Chilton, Didcot, Oxon, OX11 0RQ, UK. This article first appeared in the November 2012 issue of Nuclear Engineering International magazine.



[1] Defra (2006). Industry Profile: Industrial activities which have used materials containing radioactivity. http://tinyurl.com/8fx247y Viewed March 2011.

[2] HSE (2005). HSE criterion for delicensing nuclear sites. Health and Safety Executive. Available from the HSE website (July 2011) at: http://www.hse.gov.uk/nuclear/delicensing.pdf

[3] GB Parliament (1990). Environment Protection Act. 1990. London. HMSO.

[4] GB Parliament (2007a). The Radioactive Contaminated Land (Modification of Enactments) (England) (Amendment) Regulations 2007. Statutory Instruments no. 3245

[5] GB Parliament (2007b). The Radioactive Contaminated Land (Modification of Enactments) (Wales) (Amendment) Regulations 2007. Statutory Instruments no. 3250

[6] GB Parliament (2007c). The Radioactive Contaminated Land Regulations (Northern Ireland) (amended) Regulations 2007. Statutory Instruments no. 3236

[7] Scottish Government (2007). The Radioactive Contaminated Land (Scotland) Regulations 2007. Scottish Statutory Instruments no. 179

[8] Haywood SM and Smith J (1990). Assessment of the potential radiological impact of residual contamination in the Maralinga and Emu areas. NRPB-R237. Chilton.

[9] SAFEGROUNDS (2011a). Remediation of a radioactively and chemically contaminated site at Harwell. Available from the SAFEGROUNDS Learning network website: http://www.safegrounds.com/pdfs/remediation_at_harwell_ciria.pdf

[10] SAFEGROUNDS (2011b). Characterisation of the Dounreay Castle site using the Groundhog system. Available from the SAFEGROUNDS Learning network website: http://www.safegrounds.com/pdfs/groundhog_case_study.pdf

[11] Dale P, Robertson I and Toner M (2008). Radioactive particles in dose assessments. Journal of Environmental Radioactivity 99 (1589-1595). Elsevier.

[12] HPA (2009). Application of the 2007 Recommendations of the ICRP to the UK. Doc HPA, RCE-12

[13] EA (2004). Model Procedures for the Management of Land Contamination. Contaminated Land Report 11. Defra and the Environment Agency.

[14] Towler P, Rankine A, Kruse P at el (2009). Good practice guidance for the management of contaminated land on nuclear-licensed and defence sites. Version 2. CIRIA W29. Safegrounds learning network.

[15] SAFEGROUNDS (2011c). SAFEGROUNDS Learning network website http://www.safegrounds.com/ [16] Brown J and Etherington G (2011). Health risks from radioactive objects on beaches in the vicinity of the Sellafield site. HPA-CRCE-018.

[17] HPA (2006). Dose Criteria for the Designation of Radioactively Contaminated Land. Doc HPA, RCE-2


Other references:

DPAG (2006). Dounreay Particles Advisory Group, Third Report. Scottish Environmental Protection Agency (SEPA) September 2006. ISBN: 1-901322-64-5.

DPAG (2008). Dounreay Particles Advisory Group, Fourth Report. Scottish Environmental Protection Agency (SEPA) November 2008. ISBN: 1-901322-69-6.

HSE (2001). Reducing risks, protecting people. HSE’s decision-making process. HSE Books. Her Majesty’s Stationery Office, Norwich, UK. ISBN 0 7176 2151

ICRP (2007). The 2007 Recommendations of the International Commission on Radiological Protection. ICRP publication 103. Ann ICRP, 37 (2-4).

NRPB (1998). Radiological protection objectives for land contaminated with radionuclides. Doc NRPB 9(2)

Wilkins BT, Harrison JD, Smith KR, Phipps AW, Bedwell P, Etherington G, Youngman M, Fell TP, Charles MW, Darley PJ and Aydarous A SH (2006). Health implications of fragments of irradiated fuel ay the beach at Sandside Bay. Module 6: Overall results. Chilton. RPD-EA-03-2006

Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.