摘要
Short-term predictions of potential impacts from accidental release of various radionuclides at nuclear power plants are acutely needed, especially after the Fukushima accident in Japan. An integrated modeling system that provides expert services to assess the consequences of accidental or intentional releases of radioactive materials to the atmosphere has received wide attention. These scenarios can be initiated either by accident due to human, software, or mechanical failures, or from intentional acts such as sabotage and radioIogicaI dispersal devices. Stringent action might be required just minutes after the occurrence of accidental or intentional release. To fulfill the basic functions of emergency preparedness and response systems, previous studies seldom consider the suitability of air pollutant dispersion models or the connectivity between source term, disper- sion, and exposure assessment models in a holistic context for decision support. Therefore, the Gaussian plume and puff models, which are only suitable for illustrating neutral air pollutants in fiat terrain conditional to limited meteorological situations, are frequently used to predict the impact from accidental release of industrial sources. In situations with complex terrain or special meteorological conditions, the proposing emergency response actions might be questionable and even intractable to decision- makers responsible for maintaining public health and environmental quality. This study is a preliminary effort to integrate the source term, dispersion, and exposure assessment models into a Spatial Decision Support System （SDSS） to tackle the complex issues for short-term emergency response planning and risk assessment at nuclear power plants. Through a series model screening procedures, we found that the diagnostic （objective） wind field model with the aid of sufficient on-site meteorological monitoring data was the most applicable model to promptly address the trend of local wind field patterns. However, most of the hazardous materials being released into the environment from nuclear power plants are not neutral pollutants, so the particle and multi-segment puff models can be regarded as the most suitable models to incorporate into the output of the diagnostic wind field model in a modern emergency preparedness and response system. The proposed SDSS illustrates the state-of-the-art system design based on the situation of complex terrain in South Taiwan. This system design of SDSS with 3- dimensional animation capability using a tailored source term model in connection with ArcView~ Geographical Information System map layers and remote sensing images is useful for meeting the design goal of nuclear power plants located in complex terrain.
Short-term predictions of potential impacts from accidental release of various radionuclides at nuclear power plants are acutely needed, especially after the Fukushima accident in Japan. An integrated modeling system that provides expert services to assess the consequences of accidental or intentional releases of radioactive materials to the atmosphere has received wide attention. These scenarios can be initiated either by accident due to human, software, or mechanical failures, or from intentional acts such as sabotage and radioIogicaI dispersal devices. Stringent action might be required just minutes after the occurrence of accidental or intentional release. To fulfill the basic functions of emergency preparedness and response systems, previous studies seldom consider the suitability of air pollutant dispersion models or the connectivity between source term, disper- sion, and exposure assessment models in a holistic context for decision support. Therefore, the Gaussian plume and puff models, which are only suitable for illustrating neutral air pollutants in fiat terrain conditional to limited meteorological situations, are frequently used to predict the impact from accidental release of industrial sources. In situations with complex terrain or special meteorological conditions, the proposing emergency response actions might be questionable and even intractable to decision- makers responsible for maintaining public health and environmental quality. This study is a preliminary effort to integrate the source term, dispersion, and exposure assessment models into a Spatial Decision Support System （SDSS） to tackle the complex issues for short-term emergency response planning and risk assessment at nuclear power plants. Through a series model screening procedures, we found that the diagnostic （objective） wind field model with the aid of sufficient on-site meteorological monitoring data was the most applicable model to promptly address the trend of local wind field patterns. However, most of the hazardous materials being released into the environment from nuclear power plants are not neutral pollutants, so the particle and multi-segment puff models can be regarded as the most suitable models to incorporate into the output of the diagnostic wind field model in a modern emergency preparedness and response system. The proposed SDSS illustrates the state-of-the-art system design based on the situation of complex terrain in South Taiwan. This system design of SDSS with 3- dimensional animation capability using a tailored source term model in connection with ArcView~ Geographical Information System map layers and remote sensing images is useful for meeting the design goal of nuclear power plants located in complex terrain.
