SE 01620 The data for thejee shots were fitted to equations of the forms const/X, and const/x1/2 as suggested by the form of Eq. 19 and 20 since no other form was apparent from the plots of the data (exception is for Shot Tewa). The variation of K(X,W)/q with distance for the water shots was indeterminate for the Romeo data, did not occur for the Navajo data, and was determined from an empirical fit of the data to an equation of the form aX" for the Flathead data. The close-in samples from the two JANGLE Shots were known to contain a large amount of inert crater material, a surface desert sand raised by the blast wave. This extraneous debris was grtater on the "gs" Shot for some locations than for the "U" Shot. It may be noted that the higher values of K(X,W)/q from the "U" Shot data between the 1 MP scaled distances of 10 x 103 and 40 x 103 are for stations on the left side of the pattern (with respect to the downwind direction) where a high ridge of activity in the fallout pattern occurred. The excess soil must have originated from material blown out asymmetrically from the crater in that direction. The difference in the fit of the Zuni and Tewa data to the assumed functions is due either to large errors in sampling and analysis of the Tewa data (large scatter) or to the presence of a larger amount of water in the Tewa fireball and cloud (it was detonated over water 25 ft deep). The equation constants for the assumed dependence of K/q on X are summarized in Table 15. The best fit of the data, where values occurred over a range of X, is for the PLUMBBOB Shasta Shot and K/q inversely proportional to xl/2, The JANGLE "U" Shot data was best fitted by the equation of K/q proportional to X-L, No explanation is available for the high values of K/q for CASTLE Koon. Since the most reliable sampling data are, in order of reliability, from Shot Shasta, Shot Zuni, and the two JANGLE Shots, and since preference of the two equations by measure of the percent standard deviation in the product KX or Kx1/2 is for the latter, it was retained for use in the mass contour scaling function. However, on the basis of the JANGLE "U" Shot results, the variation of K/q with X as X-1 may be considered for use with the fallout from underground shots. DOE/NV. The data for the water shots do not show a consistent trend in K/q with distance for all weapon yields. Within the large percentage standard deviations indicated, a constant value independent of X appeared to be the appropriate selection. 61 Ato ¢oT