A Review of the Radiological Assessments Corporation's Fernald Dose Reconstruction Report (1997)

Doses to real or hypothetical people hinge not only on the magnitude of a source, but also on how the radionuclides released are transported in the environment. Transport varies with the nuclide and the medium through which transport occurs, that is, surface water, groundwater, or soil. Primary sources of exposure assignable to the Fernald facility are uranium dust and radon gas and its progeny.

In 1994, the committee reviewed a draft report of task 4 of the Fernald Dosimetry Reconstruction Project, Environmental Pathways—Models and Validation. Most of the committee's comments (NRC 1994) have been taken into account, but some seem to have been disregarded.

The GARDEN model, which deals with the intake of radionuclides via ingestion, seems not to have included the contamination of fresh fruit and bakery products, which, according to National Council on Radiation Protection and Measurements report 77 (NCRP 1984), are among the 5 food categories that contribute the most to intake of background uranium in the United States. The GARDEN model also estimates radionuclide concentrations of raw agricultural products, but people consume food after some handling and preparation, such as washing and removal of the outer leaves of green vegetables, peeling and cooking of potatoes, and cooking of meat products. The report does not discuss the methods or parameter values needed to estimate radionuclide concentrations of prepared foodstuffs.

It is not clear how the soluble and insoluble forms of uranium are characterized and treated in the environmental-transport model. In volume II, appendix C, it is assumed that 35% of the uranium deposited on the ground is in a soluble form that migrates rapidly through the upper layer of soil. Presumably, the soluble and insoluble fractions of uranium will lead to widely different concentrations in some foodstuffs.

Large amounts of pulverized uranium were released from several buildings during the early years, as shown in the RAC report in table 5 of volume I, for a total of about 276,000 kg

in elemental or oxide form. Most of this material was relatively coarse or associated with mist droplets (see figure 24), settled close to the point of release, and might have been removed in surface runoff. The meteorologic model described seems to account adequately for the fraction of this airborne release that moved beyond the fence line (see figure 22). The other uranium releases to the environment were contained in liquid effluents consisting of 3 components: monitored plant drain discharges to the Great Miami River; surface runoff, mainly to Paddy's Run Creek; and soil seepage followed by leakage into the subsurface aquifer. In general, these pathways have been treated adequately in the report. The only exception is the assumed seepage of uranium into the subsurface aquifer from Paddy's Run. To the committee, that does not seem to be a credible pathway. Vertical migration of uranium from contaminated soil into the aquifer seems to be a more credible pathway to give rise to the slowly moving groundwater plume described in figure F-1.

The committee has questioned the particle size distribution of the airborne uranium particles given in the report because it affects the transport characteristics and the predicted health effects of inhalation. However, in the overall perspective, any divergences in the assumed size distribution are probably of only secondary importance, and the committee accepts the uranium transport estimates as they stand.

Radon is a noble gas, and its dispersion in the atmosphere is governed by a simple diffusion model, as described in the report. However, any biologic effects are due to the decay products of the disintegration of radon into a series of short-lived polonium, bismuth, and lead isotopes, collectively referred to as radon progeny. These isotopes rapidly attach themselves to aerosol particles in the atmosphere, and both transport characteristics and inhalation dynamics are affected by the proportion of attached and unattached radon-progeny atoms. The model assumes equal atmospheric transport for both categories, which might be reasonable under the circumstances. However, plant operations before 1970 undoubtedly resulted in a dustier atmosphere with a higher attachment ratio, and a proportion attached to coarser particles was removed rapidly, on site, by local fallout. Indoor exposure off site resulted almost entirely from radon transport alone.