Consideration on utilization of atmospheric dispersion models for a nuclear emergency preparedness and response

Togawa, Orihiko; Okura, Takehisa; Kimura, Masanori; Nagai, Haruyasu

JAEA-Review 2021-021, 61 Pages, 2021/11

JAEA-Review-2021-021.pdf:3.72MB

Triggered by the Fukushima Daiichi Nuclear Power Station accident, there have been a lot of arguments among various situations and levels about utilization of atmospheric dispersion models for a nuclear emergency preparedness and response. Most of these arguments, however, were alternative and extreme discussions on whether predictions by computational models could be applied or not for protective measures in a nuclear emergency, and it was hard to say that these arguments were politely conducted, based on scientific verification in an emergency response. It was known, on the other hand, that there were not a few potential users of atmospheric dispersion models and/or calculation results by the models within the Japan Atomic Energy Agency (JAEA) and outside. However, they seemed to have a lack of understanding and a misunderstanding on proper use of different kinds of atmospheric dispersion models. This report compares an outline of models and calculation method in atmospheric dispersion models for a nuclear emergency preparedness and response, with a central focus on the models which have been developed and used in the JAEA. Examples of calculations by these models are also described in the report. This report aims at contributing to future consideration and activities for potential users of atmospheric dispersion models within the JAEA and outside.

A Model intercomparison of atmospheric Cs concentrations from the Fukushima Daiichi Nuclear Power Plant accident, phase III; Simulation with an identical source term and meteorological field at 1-km resolution

Sato, Yosuke*; Sekiyama, Tsuyoshi*; Fang, S.*; Kajino, Mizuo*; Qurel, A.*; Qulo, D.*; Kondo, Hiroaki*; Terada, Hiroaki; Kadowaki, Masanao; Takigawa, Masayuki*; et al.

Atmospheric Environment; X (Internet), 7, p.100086_1 - 100086_12, 2020/10

https://doi.org/10.1016/j.aeaoa.2020.100086

The third model intercomparison project for investigating the atmospheric behavior of Cs emitted during the Fukushima Daiichi Nuclear Power Plant (FDNPP) accident (FDNPP-MIP) was conducted. A finer horizontal grid spacing (1 km) was used than in the previous FDNPP-MIP. Nine of the models used in the previous FDNPP-MIP were also used, and all models used identical source terms and meteorological fields. Our analyses indicated that most of the observed high atmospheric Cs concentrations were well simulated, and the good performance of some models improved the performance of the multi-model ensemble. The analyses also confirmed that the use of a finer grid resolution resulted in the meteorological field near FDNPP being better reproduced. The good representation of the wind field resulted in the reasonable simulation of the narrow distribution of high deposition amount to the northwest of FDNPP and the reduction of the overestimation over the area to the south of FDNPP. In contrast, the performance of the models in simulating plumes observed over the Nakadori area, the northern part of Gunma, and the Tokyo metropolitan area was slightly worse.

Development of dose estimation system integrating sediment model for recycling radiocesium-contaminated soil to coastal reclamation

Miwa, Kazuji; Takeda, Seiji; Iimoto, Takeshi*

Radiation Protection Dosimetry, 184(3-4), p.372 - 375, 2019/10

https://doi.org/10.1093/rpd/ncz128

The Ministry of the Environment has indicated the policy of recycling the contaminated soil generated by decontamination activity after the Fukushima accident. By recycling to coastal reclamation which is one of effective recycling application, dissolved radiocesium and absorbed radiocesium on soil particles will flow out to the ocean by construction, therefore evaluating radiocesium transition in ocean considering the both types of radiocesium is important for safety assessment. In this study, the radiocesium outflow during constructing and after constructing is modeled, and radiocesium transition in ocean is evaluated by Sediment model suggested in OECD/NEA. The adaptability of sediment model is confirmed by reproducing evaluation of the coastal area of Fukushima. We incorporate the sediment model to PASCLR2 code system to evaluate the doses from radiocesium in ocean.

Model intercomparison of atmospheric Cs from the Fukushima Daiichi Nuclear Power Plant accident; Simulations based on identical input data

Sato, Yosuke*; Takigawa, Masayuki*; Sekiyama, Tsuyoshi*; Kajino, Mizuo*; Terada, Hiroaki; Nagai, Haruyasu; Kondo, Hiroaki*; Uchida, Junya*; Goto, Daisuke*; Qulo, D.*; et al.

Journal of Geophysical Research; Atmospheres, 123(20), p.11748 - 11765, 2018/10

https://doi.org/10.1029/2018JD029144

A model intercomparison of the atmospheric dispersion of Cs emitted following the Fukushima Daiichi Nuclear Power Plant accident was conducted by 12 models to understand the behavior of Cs in the atmosphere. The same meteorological data, horizontal grid resolution, and an emission inventory were applied to all the models to focus on the model variability originating from the processes included in each model. The multi-model ensemble captured 40% of the observed Cs events, and the figure-of-merit in space for the total deposition of Cs exceeded 80. Our analyses indicated that the meteorological data were most critical for reproducing the Cs events. The results also revealed that the differences among the models were originated from the deposition and diffusion processes when the meteorological field was simulated well. However, the models with strong diffusion tended to overestimate the Cs concentrations.

Oceanic dispersion of Fukushima-derived Cs-137 simulated by multiple oceanic general circulation models

Kawamura, Hideyuki; Furuno, Akiko; Kobayashi, Takuya; In, Teiji*; Nakayama, Tomoharu*; Ishikawa, Yoichi*; Miyazawa, Yasumasa*; Usui, Norihisa*

Journal of Environmental Radioactivity, 180, p.36 - 58, 2017/12

https://doi.org/10.1016/j.jenvrad.2017.09.020

This study simulates the oceanic dispersion of Fukushima-derived Cs-137 by an oceanic dispersion model and multiple oceanic general circulation models. The models relatively well reproduced the observed Cs-137 concentrations in the coastal, offshore, and open oceans. Multiple simulations in the coastal, offshore, and open oceans consistently suggested that Cs-137 dispersed along the coast in the north-south direction during the first few months post-disaster, and were subsequently dispersed offshore by the Kuroshio Current and Kuroshio Extension. Quantification of the Cs-137 amounts suggested that Cs-137 actively dispersed from the coastal and offshore oceans to the open ocean, and from the surface layer to the deeper layers in the North Pacific.

Development of a short-term emergency assessment system of the marine environmental radioactivity around Japan

Kobayashi, Takuya; Kawamura, Hideyuki; Fujii, Katsuji*; Kamidaira, Yuki

Journal of Nuclear Science and Technology, 54(5), p.609 - 616, 2017/05

https://doi.org/10.1080/00223131.2017.1286272

The Japan Atomic Energy Agency has, for many years, been developing a radionuclide dispersion model for the ocean, and has validated the model through application in many sea areas using oceanic flow fields calculated by the ocean model. The Fukushima Dai-ichi Nuclear Power Station accident caused marine pollution by artificial radioactive materials to the North Pacific, especially to coastal waters northeast of mainland Japan. In order to investigate the migration of radionuclides in the ocean caused by this severe accident, studies using marine dispersion simulations have been carried out by JAEA. Based on these as well as the previous studies, JAEA has developed the Short-Term Emergency Assessment system of Marine Environmental Radioactivity (STEAMER) to immediately predict the radionuclide concentration around Japan in case of a nuclear accident.

Numerical simulations of short-term migration processes of dissolved Cesium-137 due to a hypothetical accident of a nuclear submarine in the Japan Sea

Kobayashi, Takuya; Chino, Masamichi; Togawa, Orihiko

Journal of Nuclear Science and Technology, 43(5), p.569 - 575, 2006/05

https://doi.org/10.3327/jnst.43.569

A dissolved radionuclide migration code system that consists of a ocean circulation model, Princeton Ocean Model, and a particle random-walk model, SEA-GEARN, has been developed. The oceanic migration of Cs discharged from a nuclear submarine in a hypothetical accident at the Tsushima Strait was calculated in the southwestern area of the Japan Sea as a model application. The calculations for instantaneous releases every 10 days were carried out for one year to study the seasonal differences of migration process of the dissolved radionuclides. The migration tendencies of dissolved radionuclides were divided into two patterns. For the releases started from January to September, all of the high concentration areas migrated to the northeast along the coastline of the Main Island of Japan from the release point. As for the releases from October to December, some high concentrations areas migrated to the west from the release point and the concentrations of Cs along the coastline of the Main Island of Japan were comparatively low.

Improvement of Worldwide version of System for Prediction of Environmental Emergency Dose Information (WSPEEDI), 2; Evaluation of numerical models by Cs deposition due to the Chernobyl nuclear accident

Terada, Hiroaki; Chino, Masamichi

Journal of Nuclear Science and Technology, 42(7), p.651 - 660, 2005/07

https://doi.org/10.3327/jnst.42.651

The prediction performance of WSPEEDI (Worldwide version of System for Prediction of Environmental Emergency Dose Information), which consists of the atmospheric dynamic model MM5 and the Lagrangian particle dispersion model GEARN-new, is evaluated by measurements of precipitation and surface deposition of Cs over Europe during the Chernobyl accident. It is concluded that MM5/GEARN-new can predict Cs deposition distribution with good accuracy when accurate precipitation is predicted by using a explicit scheme on cloud microphysics with ice phase processes. High-resolutional calculation is also conducted for the area surrounding Chernobyl by a nesting method. MM5/GEARN-new can predict quite a realistic distribution of Cs deposition around Chernobyl which was not calculated by the previous version.

Real-time environmental contamination assessment system during nuclear emergency

Terada, Hiroaki; Furuno, Akiko; Chino, Masamichi

Proceedings from the International Conference on Radioactivity in the Environment (CD-ROM), 4 Pages, 2002/09

The present study aims to expanding the capability of WSPEEDI, so that it can be applied to atmospheric releases of radionuclides in the world for terrorist attack as well as nuclear accident. A terrorist attack would be possible in any time, any place and any scale. Altough WSPEEDI has already had a function to acquire global meteorological forecasts to generate a geographical map at arbitrary region in the world, it had no capability for simultaneous multi-scale predictions. Thus, the combination of models, non-hydrostatic meteorological model MM5 and atmospheric dispersion model GEARN, is introduced to WSPEEDI. MM5 can forecast local and regional meteorological condition simultaneously by domain nesting calculations. By the input of meteorological condition generated by MM5, GEARN can forecast multi-scale environmental contaminations considering detailed boundary layer and precipitation processes. Using this improved WSPEEDI, we made test calculations assuming a nuclear accident or terrorist attack in Asia.

Development of ocean pollution prediction system for Shimokita region; Model development and verification

Kobayashi, Takuya; Lee, S.; Chino, Masamichi

Journal of Nuclear Science and Technology, 39(2), p.171 - 179, 2002/02

https://doi.org/10.3327/jnst.39.171

A three-dimensional model system was developed to predict oceanic dispersions of radionuclides released into the eastern area of Shimokita Peninsula. This system is a combination of the Princeton Ocean Model (POM) for predicting ocean currents and a particle random walk model for oceanic dispersion of radionuclides. The model was verified by using measured currents, temperature and salinity at the coastal area of Shimokita, Aomori-ken, Japan, where a nuclear fuel reprocessing plant is under construction. The results obtained from simulations area as follows; (1) Wind and the Tsugaru Warm Current entering into the objective region through the Tsugaru Strait significantly affect the structure of current over the region. (2) POM can represent seasonal variations of the Tsugaru Warm Current well with hypothetical oceanographic data. The calculation succeeded to reproduce the coastal mode from winter to spring and the gyre mode from summer to autumn.