Improving the estimation of direct release rates and transport processes from the Fukushima Daiichi Nuclear Power Plant accident using higher-resolution oceanic dispersion model

A series of accidents at the Fukushima Daiichi Nuclear Power Station (F1NPS), following the 11 March 2011 earthquake and tsunami, resulted in the release of radioactive substances into the ocean. In particular, the release of large amounts of radioactive caesium has damaged the fishing industry, leading to voluntary restrictions on fishing and shipping. Oceanic dispersion simulations based on estimates of the pathways and fluxes of radioactive materials provide useful information for assessing the environmental impacts and formulating measures to mitigate the effects of the accident. For the direct release rate from the F1NPS site, an estimation method was developed using the results from nearby monitoring, and the seawater exchange rate was estimated in target volume using a numerical simulation. However, the influence of volume on the seawater exchange rate was not considered. Appropriate volumes must be considered when estimating the effects of future accidents. In addition, the directional coastal transport was underestimated in the simulations of the F1NPS accident because of the low resolution. To estimate the pathways and fluxes of radioactive material to the ocean and understand the distribution of the concentration of radioactive material based on ocean dispersion simulations, a study was conducted using a higher-resolution model. The horizontal resolution of the conventional ocean dispersion model Regional Ocean Modelling System (ROMS), was increased from 1 km to 200 m. The optimal settings of the seawater exchange rate were investigated, and the radioactive caesium transport process in the coastal direction was more accurately reflected. We found that the conventional volume for determining the seawater exchange rate, including the locations of release sources and observation points, is optimal. The validity of this finding was confirmed using experimental equations from previous oceanic tracer release experiments. To estimate future release rates, it an appropriate volume must be defined, for example, depending on the distance between the locations of the release sources and the observation point. In addition, improvements in the accuracy of the simulation of the coastal transport process were observed owing to the higher resolution, which increased reproducibility. However, with a horizontal resolution of 200 m, problems with repeatability near the harbours arose. A higher resolution, achieved using nesting or other methods, would be desirable to deal with releases smaller than those in the F1NPS accident.