Critical Remarks to Early Phase Austrian Decision Support Tools: determination of Source Terms and Nuclide Dispersion Modelling

Beschreibung

1. INTRODUCTION In order to adopt potential countermeasures to protect the population during the course of a severe accident in a nuclear power plant a fast prediction of the radiation exposure is necessary. In most of these severe accidents several release periods will be possible. The start of the first releases to the environment can occur several hours after the beginning of the accident. In addition the transport of the released radioactivity to a specific area takes time. Both effects lead to the situation that in the very first phase of the accident no measured data will be available. For an efficient emergency management it is necessary to estimate the radiation exposure as early as possible. In Austria several tools are available to estimate this radiation exposure. The goal of this paper is a critical review of two of these tools in the context of an international harmonization. 2. SOURCE TERM DETERMINATION The first step in a nuclear emergency is the determination of the source term. To bridge the gap from the time when first informations on the accident arrive to the time when measured data are available the code ASTEC (Austrian Source Term Estimation Code) was developed at ARC Seibersdorf. For each of the European countries generating nuclear power, a databank with the basic informations of the nuclear units of the country is necessary. For each power plant a set of precalculated time dependent source terms (release fractions and different inventories) is needed. Using the first informations and selecting the path "plant - inventory - accident sequence" an input file for a dispersion code (in our case TAMOS) will be generated automatically. Due to the fact that only very few source terms are published in the open literature most of the source terms have to be calculated for the data bank. This can be achieved best with specific reactor data and not with typical data. In addition a source term calculation including sequence modelling and data compilation is very time- and manpower-consuming. Furthermore different results from different countries can result in inconsistent decisions. Therefore it is recommended to generate a harmonized European source term data bank for emergency preparedness. In addition the development of a more user-friendly emergency form (starting from IAEA EMERCON form) is proposed. 3. NUCLIDE DISPERSION MODELLING In case of a nuclear emergency the Central Institute for Meteorology and Geodynamics is in charge of giving advice to the decision maker. For this purpose a real-time long range atmospheric dispersion modelling system was developed. The modelling system consists of several modules such as input data pre-processing, trajectory model, diffusion model, deposition model, results post-processing. The system was built with existing scientific codes (trajectory- concentration- and deposition modules), which had to be modified to match the local hardware and data. Pre-processing and post-processing modules were separately developed. The construction of the interfaces to match all the modules in an operational model with high availability of the service is a major undertaking (several man months to years). With the implementation of such emergency response modelling systems at all the different European emergency response centres much of the effort was reduplicated at local level and the interchange of model components is difficult. International emergency response model intercomparison efforts (ATMES, ETEX, RTMOD, ENSEMBLE) have evaluated model results. The option of flexible model module exchange would offer more insight into the question on the source of the observed result differences. Discrepancies could then be tracked down to the specific processes such as the influence of input data, diffusion model, deposition model. This knowledge would support better decisions on the selection and combination of the modules within a specific modelling system and such lead to more reliable results. The development of conventions and standards for the design of the different interfaces between the modules would enable the exchange of once developed tools between different institutions and thus significantly reduce the enormous amount of international redundancy in emergency response modelling. For meteorological purposes such standards were developed by WMO for the exchange of meteorological model data and measurements. The paper will present preliminary ideas on the definition of conventions and standards for emergency dispersion modelling systems with application examples. This comprises the exchange of data, coding conventions for software, notes to coordinate systems as well as the treatment of non-standard data and fields. The conventions for the exchange of measured and modelled data need to be defined based on scientific data formats, which are independent of the underlying hardware and self-describing such as GRIB, BUFR, netCDF and HDF. Information on map projection and vertical coordinates needs to be defined. For the exchange of software modules, coding conventions can be adopted in analogy to those developed at large weather forecasting modelling centres like ECMWF and NCAR. Standard computer programs and software libraries for converting coordinate systems need to be developed. Generally available software needs also to be developed for the processing of specific non-standard data such as vertical velocities, instant boundary layer fluxes, mixing heights, precipitation rates. The foundation of a working group is suggested, with the task to develop such conventions and standards. In a next step participating groups will adjust their modelling modules according to the defined conventions and standards e.g. within an EC -project. A web based platform is envisioned for the exchange of these adjusted modules. This would enable a flexible exchange of components between different modelling systems and foster model improvement. 4. SUMMARY This paper describes two examples of developing decision support tools in Austria: The code ASTEC to determine source terms and the modelling of a real-time long range atmospheric dispersion. Each tool is reviewed critically in the context of an international harmonization.

Zeitraum	29 Sept. 2003 → 3 Okt. 2003
Ereignistitel	International Symposium on Off-site Nuclear Emergency Management
Veranstaltungstyp	Sonstiges
Bekanntheitsgrad	International