Predictability of a short-term emergency assessment system of the marine environmental radioactivity

A Short-Term Emergency Assessment system of the Marine Environmental Radioactivity (STEAMER) was developed at the Japan Atomic Energy Agency (JAEA) to forecast or reanalyze the oceanic dispersion of radionuclides that are released into the ocean around Japan from nuclear facilities during routine operation or in an emergency. STEAMER is currently in daily operation at JAEA to conduct single forecast simulation. The predictability of STEAMER is validated by utilizing oceanographic forecast and reanalysis data in this study. The oceanic dispersion simulations that use oceanographic reanalysis data as input data are assumed to have true solutions. Reanalysis data that has been optimized by data assimilation is the most reliable input for post-analysis. Rigorous oceanic dispersion simulations are conducted for the hypothetical release of Cs-137 from the Fukushima Daiichi Nuclear Power Plant. The predictability of the Cs-137 oceanic dispersion is quantitatively estimated over a forecast period. Moreover, ensemble forecast simulations are also performed applying the Lagged Average Forecast methodology and they successfully improve the predictability of the Cs-137 oceanic dispersion over that obtained using single forecast simulation. The ensemble forecast simulations need to be installed in STEAMER in the future.