Calibration of Safecast dose rate measurements

Highlights: • A new calibration methodology is presented for Safecast dose rate data using U.S. Department of Energy (DOE) data as a baseline. • Missing data, normal background and proper decay rates are considered during the analysis. • A model is found to define the ratio of ¹³⁴Cs and ¹³⁷Cs as function of time over the domain. • A Gaussian plume model is also run to further validate the calibration model and minimize errors. • Using the proposed calibration methodology, Safecast data can be used during emergencies. - Abstract: A methodology is presented to calibrate contributed Safecast dose rate measurements acquired between 2011 and 2016 in the Fukushima prefecture of Japan. The Safecast data are calibrated using observations acquired by the U.S. Department of Energy at the time of the 2011 Fukushima Daiichi power plant nuclear accident. The methodology performs a series of interpolations between the U.S. government and contributed datasets at specific temporal windows and at corresponding spatial locations. The coefficients found for all the different temporal windows are aggregated and interpolated using quadratic regressions to generate a time dependent calibration function. Normal background radiation, decay rates, and missing values are taken into account during the analysis. Results show that the standard Safecast static transformation function overestimates the official measurements because it fails to capture the presence of two different Cesium isotopes and their changing magnitudes with time. A model is created to predict the ratio of the isotopes from the time of the accident through 2020. The proposed time dependent calibration takes into account this Cesium isotopes ratio, and it is shown to reduce the error between U.S. government and contributed data. The proposed calibration is needed through 2020, after which date the errors introduced by ignoring the presence of different isotopes will become negligible.