CSA N288.2-14

Preface

This is the second edition of CSA N288.2, Guidelines for calculating the radiological consequences to the public of a release of airborne radioactive material for nuclear reactor accidents. It supersedes the previous edition published in 1991 under the title Guidelines for calculating radiation doses to the public from a release of airborne radioactive material under hypothetical accident conditions in nuclear reactors.

This Standard is part of a series of Standards on environmental management for nuclear facilities. This Standard describes acceptable methods for modelling the consequences of accidents at nuclear reactors for safety assessment and real-time emergency response. This Standard also identifies acceptable data sources and acceptable methodologies to account for specific effects, and recommends standardized end points for the calculations.

This edition has been updated to reflect current industry practice and new research and analysis methods. Major changes to this edition include

a) updating definitions and terminology in accordance with current usage;

b) incorporation of new guidance from relevant national and international publications that address doses from accidental releases (e.g., new ICRP guidance on dose coefficients);

c) broadening the applicability to include assessments that are conducted for licensing, emergency planning, or environmental assessment purposes;

d) provision of guidance on consequence assessments for emergency response during a real event;

e) inclusion of all radionuclides that could be released to the atmosphere in a postulated or real accident;

f) allowance for a stochastic treatment of meteorological data in which doses are calculated for many records in the meteorological archives at a given site;

g) discussion of the uncertainty in the dose estimates;

h) provision of guidance on how to obtain the meteorological information required by the models (e.g., stability class) and performance requirements for data measurement;

i) inclusion of health risks resulting from the predicted doses (including organ doses for deterministic effects);

j) consideration of approaches to account for time-dependent releases to the environment;

k) provision of guidance on the location and age of the representative person for whom doses are calculated;

l) advanced methods for treating the release of tritium;

m) provision of guidance on how to determine individual doses from stochastic results consistent with regulatory expectations for conservative analysis; and

n) provision of guidance on the attributes that atmospheric dispersion computer codes should consider for use in the Canadian regulatory context.

Users of this Standard are reminded that the site selection, design, manufacture, construction, installation, commissioning, operation, and decommissioning of nuclear facilities in Canada are subject to the Nuclear Safety and Control Act and its Regulations. The Canadian Nuclear Safety Commission might impose additional requirements to those specified in this Standard.

The CSA N-Series Standards provide an interlinked set of requirements for the management of nuclear facilities and activities. CSA N286 provides overall direction to management to develop and implement sound management practices and controls, while the other CSA Group nuclear Standards provide technical requirements and guidance that support the management system. This Standard works in harmony with CSA N286 and does not duplicate the generic requirements of CSA N286; however, it may provide more specific direction for those requirements.

Introduction

0.1 Previous edition

The previous edition of this Standard provided guidance on modelling postulated accidents at nuclear power plants. It was used to demonstrate compliance of the design with licensing requirements that were expressed in terms of dose to a critical individual located at the site boundary. The previous edition used a straight-line Gaussian dispersion model that calculated concentration in the air and on the ground surface. The previous edition also calculated effective dose and equivalent dose to thyroid for the estimation of stochastic health effects. This effectively limited the previous edition to design basis accidents at nuclear power plants, which required demonstration that the effective dose was less than 250 mSv.

The previous edition of the Standard described equations and parameters that were to be used to calculate the doses to an individual. The selected equations and parameters were closely associated with an implementation of the Standard in the code PEAR (Public Exposures from Accidental Releases).

Since then, the PEAR code has been superseded by a new software called ADDAM (Atmospheric Dispersion and Dose Analysis Method) and the Canadian nuclear industry has adopted other code packages created by the U.S. government [such as MACCS2 (Melcor Accident Consequence Code Systems)] and the European Union [COSYMA (Code System from MARIA - Methods for Assessing the Radiological Impact of Accidents)] for estimating the consequences of postulated accidents. For emergency response, the ERP (Emergency Response Plan) series of codes designed by Ontario Power Generation (OPG) and Emergency Management Ontario (EMO) have been used alongside the RODOS (Real-time On-line Decision Support) and ARGOS (Accident Reporting and Guidance Operational System) codes from the European Union.

0.2 Evolution of methodology

Since the publication of the previous edition, the range of accidents that are routinely modeled includes beyond-design-basis-accidents and severe accidents. The radioactivity released during these postulated accidents can potentially expose individuals to doses that exceed the threshold for deterministic health effects, including early death from radiation syndrome. In this context, calculating the committed effective dose is not sufficient to assess the health effects to individuals. Another important factor concerns the methodology for the analysis of severe accidents, which uses probabilistic safety assessment (PSA) methods. These methods proceed by identifying initiating events and the frequency of postulated end-states (such as core melt) using fault-tree methods. The thermohydraulic behaviour of the system during accident progression is then modelled to arrive at a source term, the activity released to the atmosphere. The last step, which is the focus of this edition, consists hourly meteorological data.

Furthermore, there has been broad interest in modelling real-time accidents for emergency response purposes. In this last context, the models use current or forecast meteorological data and measured or estimated source terms to calculate the consequences for individuals living near a nuclear power plant.

0.3 Industry review

Several industry reviews of the previous edition of this Standard identified gaps and areas for improvement. The reviews recognized the need to broaden the type of consequence assessments to include those conducted for licensing, emergency planning, or environmental assessment purposes. There was also a need for guidance on consequence assessments for emergency response during a real event. It was felt that the theoretical models themselves needed to be reviewed to better reflect current knowledge in the field.

Finally, numerical methods and data have evolved over the last twenty years. The ICRP has published new guidance on dose coefficients and the reviews recommended that these changes be incorporated into the new edition.
<