A comparison of the ellipsoidal and voxelized dosimetric methodologies for internal, heterogeneous radionuclide sources

Non-human biota dosimetry has historically relied on ellipsoidal dosimetric phantoms. In 2008, the International Commission on Radiological Protection (ICRP) presented a set of ellipsoidal models representative of wildlife, including dosimetric data for homogeneously distributed internal radionuclide sources. Such data makes it possible to quickly and easily estimate radiation dose rate. Voxelized modeling, first developed for use in human medical dosimetry, utilizes advanced imaging technologies to generate realistic and detailed dosimetric phantoms. Individual organs or tissues may be segmented and dosimetric data derived for each anatomic area of interest via Monte Carlo modeling. Recently, dosimetric data derived from voxelized models has become available for organisms similar to the ICRP's Reference Animals and Plants in 2008. However, if the existing ellipsoidal models are conservative, there may be little need to employ voxel models in regulatory assessments. At the same time, existing dosimetric techniques may be inadequate to resolve recent controversies surrounding the impact of ionizing radiation exposure on wildlife. This study quantifies the difference between voxel-calculated and ellipsoid-calculated dose rates for seven radionuclides assumed to be heterogeneously distributed: ¹⁴C, ³⁶Cl, ⁶⁰Co, ⁹⁰Sr, ¹³¹I, ¹³⁴Cs, ¹³⁷Cs, and ²¹⁰Po. Generally, the two methodologies agree within a factor of two to three. Finally, this paper compares the assumptions of each dosimetric system, the conditions under which each model best applies, and the implications that our results have for the ongoing dialog surrounding wildlife dosimetry. - Highlights: • Ellipsoidal models are fit-for-regulatory-purpose for wildlife investigated herein. • Use of realistic tissue versus 4-part tissue increases AF in an organ generally by less than 10%. • Ellipsoidal model estimates of organ dose rates are conservative by up to a factor of three