11:00 〜 13:00
[MAG39-P04] 高解像度領域海洋モデルによる福島第一原子力発電所事故による137Csの海洋への直接放出率の検証
キーワード：福島第一原子力発電所事故、直接漏洩、放射性セシウム137、領域海洋モデル
The Fukushima Daiichi Nuclear Power Plant (1F NPP) accident caused by the Great East Japan Earthquake on March 11, 2011 resulted in the release of radioactive materials into the ocean. Since observed 137Cs activity concentration data in the ocean are sparse in both space and time, we simulated 137Cs activity concentration in order to investigate the environmental dynamics by the Regional Ocean Model System(ROMS) with meteorological reanalysis forcing data. It is necessary to set up an appropriate source terms for the simulation. Direct release rate of 137Cs, one of the major leakage pathways, was verified.
The direct release rate (Bq/day) can be estimated as the product of the seawater exchange volume rate (m3/day) and the observed 137Cs activity concentration (Bq/m3). The seawater exchange volume rate (m3/day) at the release site was estimated by simulation on one mesh (1/120 degree, 735m x 929m x 8m) of the model. 137Cs activity concentration was taken as the average value of the radioactivity observed at the 5, 6 and south discharge channels, which are located about 1 km away from each other adjacent to the 1F NPP site. The direct release rate from March 26 2011 until April 6 2011, when the visible direct release stopped, was estimated to be 2.2x1014 Bq/day. This result was consistent with both the direct release rate calculated from the seawater exchange volume rate and 137Cs activity concentration in the port (Kanda, 2013) and the direct release rate calculated from the flow rate of visually contaminated water and 137Cs activity in the contaminated water by the TEPCO. The seawater exchange volume rate in this estimation depends on the mesh size of the model. The source of the direct release, the Unit 2 intake, is far away from the observation points at the 5, 6, and south discharge ports. There are port structures between the source and observed points. It was difficult to properly set the volume for seawater exchange volume rate and the locations of the observed data. As a result, it would have been appropriate to use a mesh size of 735m x 929m for the seawater exchange volume rate to correspond to the average values of radioactivity observed at the 5, 6, and south discharge channels, which are about 1km apart from each other in the vicinity of the 1F NPP.
We estimated the direct release rate using a high-resolution model of 147m x 186m mesh, which is 1/5 of the original model. It is considered that the port structure was damaged immediately after the accident, and the release path from the port to the open ocean was not only at the port entrance but also other area. Therefore, it was difficult to set the releasae path in the high-resolution model. A high-resolution model simulation was performed using the port mouth as the main source of release. By using the seawater exchange volume rate of an area of about 1km2, we were able to estimate the same direct discharge as the previous model (735m×929m). In other words, the averaged 137Cs activity concentration of in about 1 km2 of sea area agrees with the averaged observed 137Cs activity concentration at 5, 6, and the south discharge channels. However, it was not possible to reproduce the respective concentrations at each observed points because the release route from the port to the open ocean was unknown and the reproducibility of the flow near the port was insufficient in this high-resolution model. The direct release of 137Cs from the 1F NPP site and its surrounding area continues after about 11 years from the accident. A high-resolution model simulation was performed using the estimated direct release rate to 2016 for verification with observed data. From the simulation, it was found that the distribution of 137Cs activity in the ocean is affected by coastal currents, meso-scale eddies and the Kuroshio Current, and has a large spatio-temporal variation. Therefore, the annual mean 137Cs activity distribution was calculated and compared with the annual mean observed ones, and good agreement was confirmed in the area considered to be dominated by direct release. The high-resolution model eliminates the underestimation of 137Cs at the Fukushima Daini NPP, 10 km south of the 1F NPP because the model more adequately reproduced the southward transport.
The direct release rate (Bq/day) can be estimated as the product of the seawater exchange volume rate (m3/day) and the observed 137Cs activity concentration (Bq/m3). The seawater exchange volume rate (m3/day) at the release site was estimated by simulation on one mesh (1/120 degree, 735m x 929m x 8m) of the model. 137Cs activity concentration was taken as the average value of the radioactivity observed at the 5, 6 and south discharge channels, which are located about 1 km away from each other adjacent to the 1F NPP site. The direct release rate from March 26 2011 until April 6 2011, when the visible direct release stopped, was estimated to be 2.2x1014 Bq/day. This result was consistent with both the direct release rate calculated from the seawater exchange volume rate and 137Cs activity concentration in the port (Kanda, 2013) and the direct release rate calculated from the flow rate of visually contaminated water and 137Cs activity in the contaminated water by the TEPCO. The seawater exchange volume rate in this estimation depends on the mesh size of the model. The source of the direct release, the Unit 2 intake, is far away from the observation points at the 5, 6, and south discharge ports. There are port structures between the source and observed points. It was difficult to properly set the volume for seawater exchange volume rate and the locations of the observed data. As a result, it would have been appropriate to use a mesh size of 735m x 929m for the seawater exchange volume rate to correspond to the average values of radioactivity observed at the 5, 6, and south discharge channels, which are about 1km apart from each other in the vicinity of the 1F NPP.
We estimated the direct release rate using a high-resolution model of 147m x 186m mesh, which is 1/5 of the original model. It is considered that the port structure was damaged immediately after the accident, and the release path from the port to the open ocean was not only at the port entrance but also other area. Therefore, it was difficult to set the releasae path in the high-resolution model. A high-resolution model simulation was performed using the port mouth as the main source of release. By using the seawater exchange volume rate of an area of about 1km2, we were able to estimate the same direct discharge as the previous model (735m×929m). In other words, the averaged 137Cs activity concentration of in about 1 km2 of sea area agrees with the averaged observed 137Cs activity concentration at 5, 6, and the south discharge channels. However, it was not possible to reproduce the respective concentrations at each observed points because the release route from the port to the open ocean was unknown and the reproducibility of the flow near the port was insufficient in this high-resolution model. The direct release of 137Cs from the 1F NPP site and its surrounding area continues after about 11 years from the accident. A high-resolution model simulation was performed using the estimated direct release rate to 2016 for verification with observed data. From the simulation, it was found that the distribution of 137Cs activity in the ocean is affected by coastal currents, meso-scale eddies and the Kuroshio Current, and has a large spatio-temporal variation. Therefore, the annual mean 137Cs activity distribution was calculated and compared with the annual mean observed ones, and good agreement was confirmed in the area considered to be dominated by direct release. The high-resolution model eliminates the underestimation of 137Cs at the Fukushima Daini NPP, 10 km south of the 1F NPP because the model more adequately reproduced the southward transport.