ls 0021620 Real differences between observed values of the contour ratios and those predicted from the idealized contour ratio functions are expected to oceur. The major causes of variation in the functions, including those that cause variance from the idealized function, are: 0 ON ON 1. 2. 3. ~ . - . Weapon type and yield Fractionation Effect of terrain roughness on fallout deposition patterns and on the radiations delivered at a point from a given radiation source ' Instrument response to the radiations Depth or height of detonation Activity and mass particle size relations Type of environmental material at shot point r 7 Degree of mixing of crater material with the radioactive nuclides Meteorological factors 10. Nonscalable or extraneous debris. In the measurement of the observed values, there will be discrepan- cies due to sampling bias, recovery losses, analytical error, and instrument error. The weapon type will mainly influence the values of the fraction of fission yield, b, and the values of the neutron capture ratios, C4; it may indirectly influence other factors such as fractionation. The idealized mass contour ratio functions suggest that the yield itself should not influence the value of the mass contour ratio as much as other factors. DoF jNM The absence of the more volatile radionuclides in fallout particles results in fractionation. When certain of the fission product tracer nuclide or nuclides are used in determining the value for the number of fissions, and other radionuclides are not present in the proper amount, the true values of ipp and igp are lower than given in the idealized scaling functions forthe unfractionated fission products and the observed value of the contour ratios will be larger. If the reduction of a given radionuclide from its normal percentage (say, for U23> fis- sion products) is given by the radiochemical "R" value, rj, for the jth radionuclide, then the gross reduction in the value of tepl*) may simi- larly be defined by the gross fission product "R" value, rrp(t), from gross ionization-rate measurements or from knowledge of the r, of all the important radionuclides. values Since "R" values for a given radio- nuclide may vary with particle size, rp,(t) may vary with distance (i.e. be a point function). The contour ratid scaling parameter sensitive to fractionation is G@ (t); in terms of rpp(t) and rj, it is given by