Although the magnitudes of the uncertainties in the mean unweathered activity per unit area were not well defined, we thought that the estimates for Rongelap Island had a standard deviation based partly on the original Bikini Ash activity measurements (e.g., 9577 §.D. = £20% (1856)) and partly on the variation in exposure-rate measurements |e.g., radsafe team survey S.D. = £202 (oc6s) }. A measurement of activity per unit area based on a single soil sample would be_highly uncertain, because of random fallout deposition and be-~ cause of physical mechanisms which move deposited fallout. This variation was dampened considerably by our use of exposure-rate survey results to estimate mean surface activity rather than use of a few gross beta measurements on soil. Our estimate of the standard deviation for the few soil samples collected by the radsafe team was +140% of the mean value, based om the surface activity measurements reported by O'Conner (0C68) for surface samples taken shortly after detonation from one island in the Northern Marshall Islands. If soil sample results were used in the final estimate of dose, this large standard deviation would propagate through the calculations. greater certainty in the result. Thus, we chose a method which offered Qur estimate of each nuclide'’s mean unweathered activity per unit area of Rongelap Island was extrapolated back to 0.5 day post~-detonation. Results are listed in Table 13. The 0.5-day post-detonation time was chosen as the time at which the fallout at Rongelap Island had effectively ceased (Sh57). We used first-order linear kinetics for serially related nuclide species (Bal0) and decay schemes from the Table of the Isotopes (Le78) in order to calculate the 0.5-day activity from the day-26 activity. iii. Areal Activity of Nuclides Without Descendants in Bikini Ash. Many short-lived nuclides did not have daughter radionuclides present on day 26. We based the activity of these short-lived nuclides on the activity of a reference nuclide. Equation (3) was used by us to relate the unknown activity of the short-lived nuclide with no daughters present on day 26 to the known activity of a nuclide which had fractionated in the same fashion as the unknown. Thus, if no isobar was present on day 26, an isotope or an isotope of an isobar of the un- known was chosen to represent the fractionation behavior and be the reference nu- clide for the estimate of activity per unit area. The equation used to relate activity of a short-lived nuclide to a reference nuclide was A= BA, An where Ay Bn (3) , A = activity of nuclide A per unit area at time t post~-detonation, B = activity of nuclide B per unit area at time t post-detonation, Ag = decay constant of nuclide A, Ap = decay constant of nuclide B, A, = number of A atoms per unit fission at time t, B, = number of B atoms per unit fission at time t. The quantity A, or B, was calculated using 1) first-order linear kinetics equations, 2) fission yields for 14-MeV fission of 238y obtained from the evaluated nuclear data files of the National Nuclear Data Center - 32 -