-L. Even though the analysis given above is subject to considerable uncertainties, there is no question but thet resuspension is occurring from this aged source and at levels far in excess of what would occur if the decline in resuspended air activity indifinitely followed a 45-day half-time. Mass Loading Approach The other approximate prediction method is based upon measured or assumed levels of particulate matter in the atmosphere with the assumption that this material is derived from the contaminated soil. For fresh deposits this epproach is not a very good one because we can expect that the freshly deposited material is much more likely to be resuspended. After many years of weathering, however, one would anticipate that the material is sufficiently mixed with the ‘soil that the specific activity in airborne particulate matter should approximate that in the soil. A major difficulty could arise, however, if 239Pu and mass were distributed differently as a function of aerodynamically equivalent particle size of the soil material. The data derived from the Andersen cascade impactor study at NTS can be examined with this in mind. The mass collected during this experiment was also lognormally distributed with particle size with & geometric mean of 2.0 + 10 un. The specific activity values as a function of particle size were: I | * fpr Re . vote “ , mA, E 5 nn Size w 960 3.3 to 7 740 2.0 to 3.3 980 1.1 to 2.0 1200 ET hoot . es Ege , e 239p., (drm/<) >? 17 3, m ot pa fi + he PY Oe, ae ‘ ‘ wad -~ ‘ , e a4 totte ~ mee tae ohh Genk z= . : “a Cc ° ERE at ‘fr oc it 4 ” . a ve 2 at ty ‘ go? ir ; : y | an “ ms 3