Geospatial Analysis of 137Cs in Hiroshima Soil Cores Collected in 1976 and 1978

Harry M. Cullings
Statistics Department, Radiation Effects Research Foundation

 

Abstract

In 1976 and 1978 soil samples were collected at a large number of locations in the area of Hiroshima at distances up to 30 km from the hypocenter, to measure radioisotopes such as 137Cs for the purpose of detecting local radioactive fallout from the Hiroshima atomic bomb. The lack of any obvious pattern in these measurements that might correspond to fallout from the Hiroshima bomb has been a source of scientific curiosity over the years. Much has been learned in the past 35 years about the behavior of such radioactive materials in the environment and the interferences caused by worldwide ("global") deposition of radioactive fallout from testing of nuclear weapons in the 1950s and 1960s. However, the analysis of the data from these soil samples, to detect any pattern from the Hiroshima bomb, remains a statistically and analytically challenging problem. We know that the 137Cs in soil from global fallout is large, and it is more recent than any that could remain from the Hiroshima bomb in 1945. Because there are so many variables that affect the retention and migration of the deposited radioactivity in soil, which we presently cannot include in a statistical model, there is a large variation in the measurements from place to place, and it has a spatial structure. To be valid a statistical analysis must consider the resulting "spatial autocorrelation" in the data.
Although an analysis of the data including a careful estimation of the variable effect of long-term rainfall on local deposition of global fallout does not suggest a pattern consistent with local radioactive fallout from the Hiroshima bomb, an important question that remains is how large such a deposition from the Hiroshima bomb would have to have been to be evident in these data. Before an analysis could begin to address the question of how large a deposition of local fallout in 1945 would be detectable in these samples with the interference of global fallout, there is a need to better understand the variables related to soil, terrain and weather that affected the deposition and migration in soil of any radioactivity deposited on the soil surface, and the extent, if any, to which their effect over time on the fallout from the Hiroshima atomic bomb may have been different from their effect on global fallout.

 

 


Copyright (c) Harry M. Cullings. All rights reserved.