ANALYTICAL PROCEEDINGS. MAY 1989. VOL 26 161 Environmental Monitoring and Modelling The following are summaries of two of the papers presented at a Meeting of the Analytical Division held on October 25th, 1988, in the Scientific Societies' Lecture Theatre, London W.1. Assessment of Hazards From Nuclear Facilities G. C. Meggitt Safety and Reliability Directorate, United Kingdom Atomic Energy Authority, Wigshaw Lane, Culcheth, Warrington WA3 4NE The potential hazards to the public and to the environment from nuclear installations arise from radioactivity released in postulated accidents and as a result of normal operations. These hazards are routinel) assessed by using a varietq of modelling techniques, supplemented by information obtained by direct monitoring where appropriate, in order to demon- strate that the installation ma) be operated safely. This process involves the assessment of the amounts of radioactivity which might be released to the environment under normal and accident conditions.The environmental impacts of these releases are then estimated by using environmental modelling methods. This paper outlines the way in which the environmental impact of releases to the atmosphere is assessed as part of the demonstration that risks posed by facilities have been reduced to levels that are as low as is reasonably practicable. Accidental Releases to the Atmosphere Atmospheric Dispersion The atmosphere is an extremely complex sqstem and buoyant material released into it will not only drift downwind but will also be dispersed across the wind by the action of turbulence.There are two sources of turbulence: thermal instability and the action of wind. In the atmosphere the temperature generally falls with increasing height. If this happens in such a way that the decrease in pressure and density are matched by the fall in temperature, a parcel of air displaced vertically then experiences no buoyancy forces and the atmosphere is said to be "neutral." If the temperature gradient is greater than this adiabatic lapse rate (ALR) a parcel of air displaced upwards will be lighter than its surroundings and will continue to rise: the atmosphere will be unstable. If the temperature gradient is smaller than the ALR the atmosphere will be stable. In unstable atmospheres any turbulence will be enhanced; in stable atmospheres it will be damped.The action of the wind in passing over the ground and any obstacles causes turbulence within the boundary layer of the earth's atmosphere; the amount of turbulence generated depends upon the windspeed and the roughness of the surface over which it passes. The interaction between the thermal effects and mechanical ones are complex but it has been found that the conditions can be divided into six categories on the basis of solar insolation, cloud cover at night and windspeed. The categories (known as Pasquill categories) can be used to estimate the vertical and horizontal spread of a plume of activity. The crosswind spreads of activity are found to be quite well represented by Gaussian distributions. The half-widths in the vertical and horizontal crosswind directions at particular distances downwind can be obtained, using the categorisation, from a combination of experimental data and theoretical considerations.There are a number of extensions of this basic scheme to take account of different ground roughness (which affects vertical dispersion) and the time period of the release (which modifies the horizontal dispersion). It is also possible to allow for other atmospheric features and for such things as the behaviour of hot plumes and interactions with building wakes. As the plume moves downwind material will be deposited from it on to the ground: should it rain material will be washed out. Models of these processes are applied in computer codes designed for atmospheric dispersion calculations.Radiation Exposures There are a number of pathways of radiation exposure from the released activity. The cloud of radioactivity is a source of direct radiation to people nearby as it passes. The exposure is transient and is, generally speaking, more important for the less severe hypothetical accidents, where noble gases form the greater part of material released. The other transient pathway is inhalation of activity from the passing cloud. This results in activity being taken into the body and being distributed and retained for periods of time which depend on the element concerned and its body chemistry. This radioactive material continues to irradiate organs of the body until it is excreted or undergoes radioactive decay. It is usual to calculate the radiation doses resulting over an entire lifetime from material inhaled during the accident by people in a number of age groups.Other routes of exposure result from the deposition of activity on the ground. The two most important of these are direct external irradiation and the ingestion of foodstuffs contaminated by the deposited activity. The first pathway is relatively simple to calculate but ingestion calculations require the modelling of the transfer of radioactivity through the food pathways. Among the more significant is the ingestion of radioiodine through cows grazing on contaminated land, the transfer of the element to milk and the consumption of the milk or its products. Health Effects Radiation exposures may affect the person exposed or his descendants.The individual exposed may suffer two type. of radiation effects. The first of these, which could only occur in extremely severe accidents, are acute and would arise only if a high threshold level of dose were exceeded. However, if this level were exceeded the effect would become progressively more severe as the dose increased until death became very likely. The other type of effects include cancer. The current hypothesis, based on extrapolation of the results of exposure to very high doses, is that any radiation dose, however small, may trigger the development of a cancer and the severity of the162 disease will be independent of the amount of radiation. The probability of the cancer being triggered will, however, depend on the radiation dose. It is considered that a radiation dose of 1 mSv will lead to a probability of about 0.001% of a fatal cancer of some kind developing. The induction of cancer is a stochastic process and it is believed that hereditary effects have a similar nature.Countermeasures If there were a release of radioactivity to the atmosphere then, if considered necessary, a number of countermeasures would be implemented. In the short term these might be, depending upon the severity of the accident, sheltering, the issue and consumption of stable iodate tablets (to prevent radioiodine accumulating in the thyroid gland), evacuation or the banning of certain contaminated foodstuffs. In the longer term, for the very severe accidents, it would be necessary to consider relocation of people.All of these measures would reduce the number of people who might suffer the acute effects of radiation or might have an enhanced risk of developing cancer later in their lives. Calculation of Impact The different stages of the calculation (dispersion, doses, countermeasures and health effects) are combined by com- puter programmes for each of the hypothetical accidents. For each defined accident, even if the amount of activity released is fixed, there will be a distribution of the number of casualties because of: firstly, the various atmospheric stability categories which might be present; and secondly, the different possible wind directions, when combined with a non-uniform popula- tion distribution around the site, leading to exposure of different numbers of people.The results of the calculations are typically presented in the form shown in Fig. 1, where the probabilities per unit time of various accidents occurring has been folded in. Routine Discharges to the Environment In order to obtain Authorisations for discharges of radioactiv- ity to the atmosphere, nuclear operators are required to assess the radiation doses which might arise from these discharges. The calculation proceeds using similar methods to those outlined above for accidents. There are, however, a few differences. While accidental discharges are generally assumed to take place over short periods, those from routine discharges are taken to be continuous and uniform. As a result the activity is spread in different directions according to the wind rose (rather than having a probability of being blown in a particular direction). The longer-term releases experience turbulence over much longer periods. The pathways of exposure are generally similar to those for the accidental case but there is, of course, no question of ANALYTICAL PROCEEDINGS, MAY 1989, VOL 26 countermeasures because doses are very much lower.Because the doses are low only the health effects which are postulated to be linearly dependent on dose (cancer and hereditary effects) need be considered. The annual limit on dose to members of the public employed by authorities is 1 mSv, corresponding to a fatal risk of the order of 10-5 per year. The dose to most exposed individuals is generally, in practice, a small fraction of this limit. 10-8 r 8 I 1 3 Y I" ' - 1 10' 102 103 Fig.1. Frequency distributions of early deaths from degraded core accidents per reactor year of operation. Curve numbers identify particular releases Number of early deaths These levels should be compared with those to which the public are exposed from natural background. Cosmic rays, radiation from activity in rocks and soils and that incorporated from natural sources into our bodies leads to average radiation doses of, on average, about 2 mSv per year. Almost one half of this comes from exposure indoors to the decay products of the naturally occurring radioactive gas radon. Conclusions The modelling procedures used in the assessment of the potential hazards from accident conditions and normal opera- tions at nuclear facilities have been outlined. The assessment methods are applied, in varying forms, to the whole range of nuclear facilities routinely in the course of demonstration that they can be operated safely.Modelling of Radioactivity in the Environment - From Now to Eternity? David R. Williams School of Chemistry and Applied Chemistry, University of Wales, Cardiff CF? 3XF Over and above natural background radiation, man, as part of there is the accidental incident from rather unpredictable the environment, is faced with two types of release of sources, such as the recent Chernobyl radioactive plume which radioactivity; first, there is the intentional slow release of very crossed Europe. small amounts of radioactivity over a large time scale from a In the former case, it is possible to set up many experiments specially constructed radwaste disposal vault, and secondly, to assess the exact material from which the packaging andANALYTICAL PROCEEDINGS, MAY 1989, VOL 26 163 disposal vault will be constructed, to do exploratory drilling and analyses to assess the geosphere and the groundwater flows, and to set up both probabilistic and deterministic models in order to estimate the doses and risks to man over many thousands of years.For the latter sudden release of radiation, the best we can achieve is to set up a series of monitoring stations interlinked such that early warnings are obtained and planned evasive measures can be taken as appropriate. These two releases are summarised in the Fig. 1, indicating that modelling is not only a most useful asset to decision making but, in some instances, it is the only way forward.The disposal of radioactive waste is considered on time scales of thousands of years and involving concentrations of species way below those analysable by available techniques, both factors which cannot be subjected to life-size blank experiments involving non-radioactive species. Secondly, although much can be established from a retrospective study of Nagasaki, Chernobyl, etc., the uncertainty about the location, underlying cause and environmental circumstances associated with a future incident cannot be predicted and so, once again, a monitoring network that will continuously collect data for consequence modelling is essential. This network, known as RIMNET (Radioactive Incident Monitoring Network) has now been established in the UK.1 I Models 1 Slow release - Deterministic Models Data I I 4 Burst m Proba bi I istic Monitor, Model and Manage Programs Validation Fig.1. Two different aspects of radioactivity modelling: of planned protracted time scale radwaste disposal and of sudden incident monitoring and dose prediction Natural radioactivity has been well described in a range of publications and will not be discussed here, although it must be remembered that all the foregoing radiation considerations must be viewed against this natural background and the considerable variation that exists from place to place.2 The environmental consequences of radwaste disposal are quantifiable by a series of large computer models, which have been reviewed recently.3.4 Essentially, it is necessary to know, or to compute, the species present at each stage between disposal vault, waste packaging, backfill material and geo- sphere through to the biosphere.Solid, liquid and gaseous wastes are involved and are expressed in terms of physical chemistry parameters, such as solubility products, absorption coefficients, formation constants, etc. This data, extracted from databanks such as the one housed in Cardiff (-14 000 items), is combined with groundwater flow characteristics and site-specific details, such as the prevailing geology, the presence of humic and fulvic acids, etc. The output is a list of the radioactive species reaching man, albeit of future genera- tions, in 50, 500 or even 5000 years time and the probable risk (or dose) experienced assuming a “worst possible case” scenario.The main codes to be used in this assessment in the UK are based upon NAMMU, to be used by NIREX, and SYVAC, to be used by the DOE to assess the NIREX proposals.5-7 Recent versions of the latter (called TIME2 and VANDAL) incorpor- ate simultaneous isotope pathways and the cycle of ice ages.7 It is reassuring to note the large amount of effort that has gone into verifying the codes, into database comparisons, and into validating the models. An outstanding example of European Community collaboration in this respect is the CHEMVAL project which compares speciation databases present in the research laboratories of eight countries.8 Recognising that there are many conflicts of interests between waste producers, central government, local govern- ment and opponents to radwaste disposal sites, the Govern- ment has implemented the Best Practicable Environmental Option (BPEO) approach .9Jo Decisions are considered, taken and recorded (in the format of an audit trail) against the checklist shown in Table 1.10 Table 1.Check list for selecting a BPEO 1. Define the objective 2. Generate options 3. Evaluate the options 4. 5 . Select the preferred option 6. Review the preferred option 7 . Implement and monitor Throughout steps 1 to 7: maintain an audit trail Summarise and present the valuation Matching “hard” science to the “soft” politics (at govern- ment and opponent levels) is always exceedingly difficult but can be considerably assisted by the BPEO concept. Technol- ogy transfer from the waste disposal field is now occurring.Just as the space race had its high-tech spin-offs, so too radwaste disposal modelling is benefiting other aspects of civilisation. For example, non-radioactive waste is also being subjected to BPEO scrutiny and the models and databases are being employed for studies and predictions of non-radioactive toxic and hazardous wastes in the environment. A recent recommen- dation places such speciation-based decisions into the hands of local and regional authorities.11 3 t r e Fig. 2. Pattern of UK primary sites forming the skeleton monitoring under the RIMNET scheme for164 ANALYTICAL PROCEEDINGS, MAY 1989, VOL 26 Recognising that the public has great difficulty in under- standing radiation, its units of measurement and dose-effect relationships and that, in the eyes of the public, waste and radioactivity are definitely NIMBY (Not In My Back Yard!) but, as far as monitoring is concerned, the same public ask WNIMBY (Why Not In My Back Yard?),’* the Government has set up a framework for monitoring radiation across the UK; this is called RIMNET.We recently undertook a survey of radioanalytical facilities in Wales on behalf of the Welsh Office.13 All instruments located in the Principality were listed. These involved hospi- tals, health authorities, industry (CEGB and Amersham plc), local authorities and educational institutes. There was a range of expertise on both the technical (equipment and sample handling) and scientific (survey planning and data interpreta- tion) fronts.Clearly, in the event of an incident such as Chernobyl, all of these facilities could be generating data in overdrive gear! Scaling-up this data to National level could lead to Government, press and public being swamped with data and interpretations of a wide range of qualities. Such consideration has led the Government to produce RIMNET, a national network of environmental radiation monitoring facilities based upon background monitoring sites on the skeleton of the locations shown in Fig. 2. The equipment and staff in these sites would be accredited, by training as necessary, to feed background levels by electronic mail into a central database facility (CDF). In the event of an incident being announced or detected by RIMNET, this skeleton could be fleshed out by expanding the amount of monitoring data collected by the locations indicated, and by bringing in supplementary accredited monitoring sites and facilities.The information received at the CDF would be assessed, bulletins prepared and regional and local authorities and the media informed. In addition, advice would be provided along the lines of restricting the consumption of certain foods, medical treatment, general advice, etc., as appropriate. It is pleasing to note that the system has already been set up and that the first dry run is scheduled for this autumn. It was Leonard0 da Vinci who remarked that “no human investigation can be called real science until it can be demonstrated mathematically.” This is never more true than in the field of radiation protection, where the long time scales, the extremely low speciation concentrations and the general complexities of the system render progress without a computer impossible.From experiences on the Government advisory committees on which I am privileged to serve,14 the impression is gained that the scientific knowledge of waste disposal and environ- mental monitoring is as good as that of any other country: the political will to implement varies, being somewhat tardy on the radwaste disposal front but enthusiastic on the monitoring network scene; the public perception of risk is sadly lacking and it is now the responsibility of all scientists in the field to liaise with interested persons in plain layman’s language. I am pleased to thank my colleagues in the applied chemistry speciation research group in Cardiff and also Government departments and industry for collaborating with our re- searches.The views expressed, however, are my own. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. References “Nuclear Accidents Overseas-the National Response Plan and Radioactive Incident Monitoring Network (RIMNET) ,” Department of the Environment, London, 1988. “Living with Radiation,” Third Edition, National Radiological Protection Board, Didcot, 1986, pp. 1-53. Duffield, J. R., and Williams, D. R., Chem. SOC. Reviews, 1986, 15, 291. Duffield, J. R., and Williams, D. R., Chem. Br., 1989, in the press. “NIREX Disposal Safety Programme,” NIREX, Didcot, 1988, “SYVAC ‘NC’ User’s Guide,” DOE SYVAC User Document, UD-SCI-2, First Edition, Department of the Environment, London.1985. Frizelle, C., New Scientist, 1988, October 15, 45. Read, D., and Broyd, T. W., Radiochim. Acra, 1989, in the press. Royal Commission on Environmental Pollution, “Best Practic- able Environmental Option,” 12th report, Cm 310, HM Stationery Office, London, 1988. Miles, J., Chem. Znd., 1988, 480. Health and Safety Executive’s Local Authority Enforcement Committee (HELA), cited in ref. 8. Page, R., Municipal Journal, 1987, 1493. Duffield, J. R., Griffiths, P. J. F., Robbins, D., and Williams, D. R., Report to Welsh Office, September, 1988, Ref. WEP5717713. Committee on Medical Aspects of Radiation in the Environ- ment, Department of Health; Radioactive Waste Management Advisory Committee, Department of the Environment; Radio- active Research and Environmental Monitoring Committee, Department of the Environment. pp. 1-24. SPECTROSCOPIC PROPERTIES OF INORGANIC AND ORGANOMETALLIC COMPOUNDS - VOLUME 21 Senior Reporters: G. Davidson, University ofNottingharn, and E.A.V. Ebsworth, University of Edinburgh This book reviews the recent literature published up to late 1987. Its Brief Contents are: Nuclear Magnetic Resonance Spectroscopy; Nuclear Quadruple Resonance Spectroscopy; Rotational Spectroscopy; Characteristic Vibrations of Compounds of Main-group Elements; Vibrational Spectra of Transition-element Compounds; Vibrational Spectra of Some Co-ordinated Ligands; Mossbauer Spectroscopy; Gas-phase Molecular Structures Determined by Electron Diffraction. ISBN 0 85186 193 8 Hardcover 525pp. Specialist Periodical Report (1988). Price E120.00 ($240.00) For further information, please write to: Royal Society of Chemistry, Sales and Promotion department, Thomas Graham House, Science Park, Milton Road, Cambridge CB4 4WF, UK. To Order, please write to Royal Society of Chemistry, Distribution Centre,Blackhorse Road, Letchworth, Herb SG6 IHN, UK. or telephone: (0462) 672555 quoting your credit card details. We cannow accept Access, Visa, Mastercard & Eurocard. RSC Members are entitled to a discount on most RSC publications and should write to: The Membership Manager, Royal Society of Chemistry, Thomas Graham House, Science Park, Milton Road, Cambridne CB4 4WF, UK. ROYAL SOCIETYOF CHEMISTRY Information Services