11:00 AM - 11:15 AM
[HCG22-08] An investigation into the simulation of "black rain" in Hiroshima and Nagasaki using the numerical mesoscale weather prediction model WRF
Keywords:The "Black Rain" caused by atomic bomb explosion, numerical weather prediction model, WRF, historical weather reanalysis data
The "black rain," which was caused by the atomic bombings of Hiroshima and Nagasaki during the end of World War II, serves as a typical case of an artificial heat source resulting from a large explosion of an atomic bomb exerting its influence on the local meteorological conditions. The updraft driven by the heat source led to the formation of cumulonimbus clouds and subsequent precipitation, thereby facilitating the diffusion and deposition of harmful substances. The estimation of the area of precipitation of black rain has been the subject of numerous studies, which have utilized various approaches. For instance, Masuda (1989) conducted a series of interviews as the basis for their research, while Yamamoto et al. (2010) measured the radioisotope Cs-137 in soil samples taken from houses constructed after 1945. Maruyama and Yoshikawa (2005) tried to simulate the precipitation using a numerical forecast model. However, the model was based on hydrostatic equations, which does not take into account a vertical advection term, the development of cumulus clouds and the precipitation process were not sufficiently calculated. As a result, an appropriate rainfall area could not be estimated. In this presentation, we report an attempt to simulate the precipitation of black rain using the WRF model version 4.6.1 based on non-hydrostatic equations.
First, downscaled experiments were conducted to reproduce the meteorological fields during the period including at the time of the atomic bombings, from August 3, 1945, 2100JST to August 9, 2100JST. The outer domain encompasses 220 x 220 grid points in the north-south and east-west directions, respectively, with a horizontal grid spacing of 8 km, thereby mainly covering the western region of Japan. The inner domain comprises 120 × 120 grid points, with a spatial resolution of 1 km, and is centered over Hiroshima. An additional inner domain is centered over Nagasaki. The model employs a total of 50 vertical levels, with a maximum height of 10 hPa, to ensure comprehensive representation of atmospheric dynamics. The lateral boundary data for the outer domain were obtained from the ERA5 (ECMWF Reanalysis v5), 20CRv3(NOAA-CIRES-DOE 20th Century Reanalysis V3) and OCADA(JMA-MRI). The horizontal distribution of sea level pressure at the time of the bombings, as calculated by the WRF model using any of the lateral boundary data, exhibited good agreement with the corresponding weather maps provided by the JMA. The accuracy of surface wind speed, wind direction, temperature, and vapor mixing ratio calculated by the WRF model was validated using in situ measurements acquired at the observation station in Hiroshima and Nagasaki.
Subsequently, an ideal experiment was conducted to solve the time evolution of a heat source simulating a nuclear explosion by placing it in the atmosphere. The horizontal and vertical grid resolutions in the idealized experiment are set at 400 m and 250 m, respectively, with a total of 200 x 200 x 80 grid points. The experiment utilized a total of six profiles: three vertical profiles of wind direction and speed, temperature, and water vapor mixing ratio obtained directly from the reanalysis data used, and three vertical profiles obtained from the weather field calculated in the downscaling experiment. The heat source is defined at a grid point located at the center of the domain, with an arbitrary horizontal extent, h, and vertical extent, v, characterized by a temperature differential, dT, relative to the surrounding atmosphere. Parameter sweeps for h, v, and dT were conducted to ensure that the heat content is equivalent to that of the nuclear explosions in Hiroshima and Nagasaki. The results indicated that, across all vertical profiles, precipitation areas were calculated to extend up to ten several km north of ground zero in Hiroshima and up to ten several km east or south in Nagasaki within two hours following the detonation. The amount of precipitation at the ground zero of the explosion was a few mm in Hiroshima and a few dozen mm in Nagasaki, which depend on the amount of water vapor in the input vertical profiles. A comprehensive analysis is planned to investigate the influence of these heat sources on wind field as well as the temporal evolution of cloud microphysics throughout the precipitation process.
Acknowledgment: This work was supported by JSPS KAKENHI Grant Number JP24H00367.
First, downscaled experiments were conducted to reproduce the meteorological fields during the period including at the time of the atomic bombings, from August 3, 1945, 2100JST to August 9, 2100JST. The outer domain encompasses 220 x 220 grid points in the north-south and east-west directions, respectively, with a horizontal grid spacing of 8 km, thereby mainly covering the western region of Japan. The inner domain comprises 120 × 120 grid points, with a spatial resolution of 1 km, and is centered over Hiroshima. An additional inner domain is centered over Nagasaki. The model employs a total of 50 vertical levels, with a maximum height of 10 hPa, to ensure comprehensive representation of atmospheric dynamics. The lateral boundary data for the outer domain were obtained from the ERA5 (ECMWF Reanalysis v5), 20CRv3(NOAA-CIRES-DOE 20th Century Reanalysis V3) and OCADA(JMA-MRI). The horizontal distribution of sea level pressure at the time of the bombings, as calculated by the WRF model using any of the lateral boundary data, exhibited good agreement with the corresponding weather maps provided by the JMA. The accuracy of surface wind speed, wind direction, temperature, and vapor mixing ratio calculated by the WRF model was validated using in situ measurements acquired at the observation station in Hiroshima and Nagasaki.
Subsequently, an ideal experiment was conducted to solve the time evolution of a heat source simulating a nuclear explosion by placing it in the atmosphere. The horizontal and vertical grid resolutions in the idealized experiment are set at 400 m and 250 m, respectively, with a total of 200 x 200 x 80 grid points. The experiment utilized a total of six profiles: three vertical profiles of wind direction and speed, temperature, and water vapor mixing ratio obtained directly from the reanalysis data used, and three vertical profiles obtained from the weather field calculated in the downscaling experiment. The heat source is defined at a grid point located at the center of the domain, with an arbitrary horizontal extent, h, and vertical extent, v, characterized by a temperature differential, dT, relative to the surrounding atmosphere. Parameter sweeps for h, v, and dT were conducted to ensure that the heat content is equivalent to that of the nuclear explosions in Hiroshima and Nagasaki. The results indicated that, across all vertical profiles, precipitation areas were calculated to extend up to ten several km north of ground zero in Hiroshima and up to ten several km east or south in Nagasaki within two hours following the detonation. The amount of precipitation at the ground zero of the explosion was a few mm in Hiroshima and a few dozen mm in Nagasaki, which depend on the amount of water vapor in the input vertical profiles. A comprehensive analysis is planned to investigate the influence of these heat sources on wind field as well as the temporal evolution of cloud microphysics throughout the precipitation process.
Acknowledgment: This work was supported by JSPS KAKENHI Grant Number JP24H00367.