Wolfgang Raskob

Iterative ensemble Kalman filter for atmospheric dispersion in nuclear accidents: An application to Kincaid tracer experiment

Information about atmospheric dispersion of radionuclides is vitally important for planning effective countermeasures during nuclear accidents. Results of dispersion models have high spatial and temporal resolutions, but they are not accurate enough due to the uncertain source term and the errors in meteorological data. Environmental measurements are more reliable, but they are scarce and unable to give forecasts. In this study, our newly proposed iterative ensemble Kalman filter (EnKF) data assimilation scheme is used to combine model results and environmental measurements. The system is thoroughly validated against the observations in the Kincaid tracer experiment. The initial first-guess emissions are assumed to be six magnitudes underestimated. The iterative EnKF system rapidly corrects the errors in the emission rate and wind data, thereby significantly improving the model results (>80% reduction of the normalized mean square error, r=0.71). Sensitivity tests are conducted to investigate the influence of meteorological parameters. The results indicate that the system is sensitive to boundary layer height. When the heights from the numerical weather prediction model are used, only 62.5% of reconstructed emission rates are within a factor two of the actual emissions. This increases to 87.5% when the heights derived from the on-site observations are used.

Kernkraftwerke in Deutschland

Von den 436 KKWs, die im Marz 2011 weltweit betrieben wurden, befanden sich 17 KKWs in der BRD (Nuclear News, A publication of the American Nuclear Society, 2011). Dies waren seit Anfang der 1990er Jahre ausschlieslich DWRs und SWRs. Sie wurden in leicht unterschiedlichen technischen Ausfuhrungen gebaut und unterscheiden sich leicht in Leistungsgrose, Aufbau und Dimensionierung der Kuhlsysteme sowie ihrem auseren Sicherheitsbehalter (Containment). Acht dieser deutschen LWRs wurden nach dem Fukushima-Unfall in Japan im Marz 2011 abgeschaltet (Dreizehntes Gesetz zur Anderung des Atomgesetzes, 2011).

Sequential multi-nuclide emission rate estimation method based on gamma dose rate measurement for nuclear emergency management

In case of a nuclear accident, the source term is typically not known but extremely important for the assessment of the consequences to the affected population. Therefore the assessment of the potential source term is of uppermost importance for emergency response. A fully sequential method, derived from a regularized weighted least square problem, is proposed to reconstruct the emission and composition of a multiple-nuclide release using gamma dose rate measurement. The a priori nuclide ratios are incorporated into the background error covariance (BEC) matrix, which is dynamically augmented and sequentially updated. The negative estimations in the mathematical algorithm are suppressed by utilizing artificial zero-observations (with large uncertainties) to simultaneously update the state vector and BEC. The method is evaluated by twin experiments based on the JRodos system. The results indicate that the new method successfully reconstructs the emission and its uncertainties. Accurate a priori ratio accelerates the analysis process, which obtains satisfactory results with only limited number of measurements, otherwise it needs more measurements to generate reasonable estimations. The suppression of negative estimation effectively improves the performance, especially for the situation with poor a priori information, where it is more prone to the generation of negative values.

Measurements and modeling of tritium reemission rates after HTO depositions at sunrise and at sunset

In a series of small-scale field studies, two experiments were carried out on reemission after 1-h HTO deposition phases, one at sunrise (7.00 a.m.) and one at sunset (9.00 p.m.), on two subsequent days. In both cases, wind speeds were extremely low to absolutely calm during the observed 12-h reemission periods. Both sunrise and sunset experiments showed the same reemission rate of 28% h−1 during the first hour of reemission, in spite of the differences of the atmosphere/soil conditions. Time-of-day specific deviations of the reemission rates appeared only during the subsequent hours, leading to total losses of the deposited HTO of 84% during the 12-h day-phase after the sunrise deposition and of 57% during the 12-h nightphase after the sunset deposition. The experimental conditions were used to model the time courses of the reemission rates by the appropriate module of the UFOTRI code, which is characterized by a coupling of HTO reemission to H2O evaporation and a set of default values, where more detailed information is not available. The comparison shows a rather poor agreement, as neither the high first-hour values nor the subsequent strong reductions can be reproduced. Agreement is improved, however, after modification of the default values. The findings suggest that at least the initial reemission rate should be modeled independently from the H2O evaporation rate, and the spatial resolution of multilayer soil models should be of the same order of magnitude as the scaling length of the deposition profile, i.e. some 10−3 m, characterizing the penetration depth of HTO after 1h of deposition and approximating the thickness of the uppermost soil layer that releases tritium during the first hour of the reemission process.

Structuring and reusing knowledge from historical events for supporting nuclear emergency and remediation management

Abstract Disasters are characterized by severe disruptions in society׳s functionality and adverse impacts on humans, environment and economy. Decision-making in times of crisis is complex and usually accompanied by acute time pressure. Environment can change rapidly and decisions may have to be made based on uncertain information. IT-based decision support can systematically help to identify response and recovery measures, especially when time for decision-making is sparse, when numerous options exist, or when events are not completely anticipated. This paper proposes a case- and scenario-based approach to supporting the management of nuclear events in the early and later phases. Important information needed for decision-making as well as approaches to reusing experience from previous events are discussed. This work is embedded in a decision support method to be applied to nuclear emergencies. Suitable management options based on similar historical events and scenarios could possibly be identified to support disaster management.

Validation of models of radionuclide wash-off from contaminated watersheds using Chernobyl data

Abstract Based on data from the Chernobyl accident, the ‘Wash-off’ scenario was developed to provide an opportunity to test models intended to simulate the movement of trace contaminants from terrestrial sources to bodies of water. The specific objective of the test was to take into account chemical speciation, its effect on the transfer of contamination from soil to water, and the geochemical and geophysical processes that affect such transfer. Modellers were provided with descriptions of two experimental plots near the Chernobyl Nuclear Power Plant (NPP), one using simulated heavy rain (HR) and one using snowmelt (SM). They were requested to estimate the vertical distribution of total 137Cs and 90Sr and their specific forms in the soil prior to the experiments, concentrations of each radionuclide in surface runoff (separately for particulate and dissolved forms) and the total amount of each radionuclide lost from plot HR during the experiment. All predictions were to be provided as best estimates with 95% subjective confidence intervals about the best estimates. In this paper, a brief description of the modelling results is provided, together with discussions of the performance of individual models in comparison with the actual measurements and of the sources of uncertainty in the model predictions. Our conclusion is that the predictive accuracy of the mathematical models could be improved by (1) improvement of model structure to include all relevant mechanisms; and (2) further use and improvement of methods for estimation of parameter values for the situation being modelled.

The Decision Support System RODOS

The chapter begins with an outline of the historical development from the first UNIX-based RODOS system until the most recent Java-based version JRodos. This is followed by an overview of the models contained in RODOS, and a description of the RODOS Center in Germany, where RODOS operates since 2005 at a central location for use by the federal government and the federal states.

BOOSTER: Development of a toolbox for triage of large group of individuals exposed to radioactive material

The effective management of an event involving the exposure of a large number of people to radioactive material requires a mechanism for fast triage of exposed people. BOOSTER is a project founded by the European Union under the Seventh Framework Programme, addressing this requirement. It is a capability project designed to provide an integrated system which could easily be deployed and used. For this purpose, the BOOSTER consortium, relying on the expertise of seven members, researches and develops new approaches to allow an effective and fast management of most kind of nuclear threats. BOOSTER System was designed to help first responders mitigating the crisis by providing the necessary information to quickly assess the radiological situation, to support triage staff in performing an efficient and fast categorization of the potentially affected victims, and to give medical staff crucial information for further treatment at medium or long term post-accident.

Relevant Radiological Phenomena, Fundamentals of Radiological Emergency Management, Modeling of Radiological Situation

The chapter summarizes relevant radiological phenomena, the fundamentals of radiological emergency management, and the modeling of the radiological situation in computer programs. Topics as the data requirements of the models and the actual availability of data in the different phases of an accident and the respective uncertainties are also addressed. The chapter is mainly intended for readers without deeper familiarity with the respective scientific field.

Recent Developments in Nuclear and Radiological Emergency Management in Europe

The chapter outlines the Developments in Nuclear and Radiological Emergency Management from the 4th to the present 7th European Framework Program of the European Union.