the importance of different sources of overestimation or underestimation of inhalation dose depends on the conditions of exposure. Therefore, estimates of inhalation dose obtained in dose reconstructions, which are intended to be upper bounds, probably provide credible upper bounds in some cases; they almost certainly do not in other cases, and it is difficult to determine one way or the other in the rest.
Another complicating factor is that the committee could not fully evaluate methods used by the NTPR program to estimate concentrations of radionuclides in fallout deposited on the ground or other surfaces on the basis of assumptions about the composition of fallout and external photon exposures measured with film badges or field instruments. Those methods are important because estimated concentrations in fallout are used to calculate inhalation dose in most cases. Therefore, although some assumptions embodied in the methods of estimating concentrations of radionuclides in fallout are likely to be overpredictive or underpredictive, it often is difficult to judge whether the net effect of all such assumptions is that estimated inhalation doses from exposure to descending or resuspended fallout tend to be overestimates or underestimates and by how much.
An example of a scenario in which credible upper bounds of inhalation dose probably are obtained in dose reconstructions involves exposure to descending fallout throughout the period of descent. Such exposures occurred, for example, on residence islands in the Pacific (see cases #3, 5, 8, 16, 22, 32, 38, 43, 47, 58, 60, 63, 78, and 96). Suppose that a participant who was exposed mainly to descending fallout filed a claim for compensation for a cancer in an internal organ other than the lung or an organ in the GI tract (such as the kidney). Several assumptions used in such a case should result in substantial overestimates of inhalation dose. Assumed dose coefficients for those organs usually apply to a particle size (AMAD) of 1 μm, even though most particles in descending fallout presumably were large and essentially nonrespirable, and dose coefficients for inhalation of respirable particles used by the NTPR program often are at least a factor of 2 higher than values for inhalation of large particles based on current ICRP recommendations. Because fractionation of radionuclides in fallout is ignored in all dose reconstructions, doses due to inhalation of volatile radionuclides attached to particle surfaces should be overestimated by a factor of about 3 or more. Doses due to inhalation of refractory radionuclides may also be overestimated substantially, even though their amounts in fallout probably are underestimated by a factor of about 3 or more because of neglect of fractionation; refractory radionuclides are dispersed mainly throughout the volume of large and highly insoluble fallout particles and therefore may be absorbed from the respiratory and GI tracts into blood to only a small extent before the particles are eliminated from the body. If the radionuclide composition in an atmospheric cloud is reasonably well characterized on the basis of cloud sampling data, the several factors that should tend to result in overestimates of inhalation dose probably are sufficient to compensate for uncertainties in all dose coefficients amounting to a factor of