3D simulations for evaluation of location factors in an urban environment: application of a novel methodology to calculate external exposure doses for evacuees from Pripyat.

This article presents a methodology for assessing the radiation doses in an urban environment due to external irradiation from radionuclides deposited on the ground and other surfaces as well as from a passing radioactive cloud. The approach was developed and applied to assess individual doses of residents of the town of Pripyat who were evacuated shortly after the Chernobyl accident. Typically, the so-called location factor is defined as the ratio of the dose rate at a point of exposure and the dose rate at an undisturbed lawn far from any buildings. The present study used a new definition of the location factor as a regular four-dimensional grid of ratios of air kerma rates indoors and outdoors distributed in space and time. The location factors were calculated for two scenarios: outdoor and indoor values for typical apartments and buildings in Pripyat. Indoor location factors varied within two orders of magnitude depending on the floor of residence and place of staying inside the apartment. Values of the indoor location factor differed during the daytime and night by a factor of 30-40 depending on the behaviour of an individual within the apartment. Both, outdoor and indoor location factors decreased with decreasing distances between buildings. It was shown that during the first 4 days after the accident, air kerma rates in Pripyat were governed by the radionuclides deposited on the ground surface, and not by radionuclides in the cloud. Specifically, the contribution of the radioactive cloud to air kerma rate was maximal (i.e., 2.3%) on the morning of 28 April 1986. The methodology and results of this study are currently being used to reconstruct the radiation gonadal dose for the subjects of the American-Ukrainian study of parental irradiation in Chernobyl cleanup workers and evacuees for investigating germline mutations in their offspring.