a ovata0 Although the quantity of debris may not be scalable with other detonation parameters, knowledge of its effect on the contour ratio and its frequency and conditions of occurrence is necessary in considering whether or not it is sufficiently important to warrant separate treat- - . ment and inclusion for consideration in decontamination investigations and operations. _ Of the several. measurement errors, the one least amenable to treatment or reduction by careful analytical techniques is that due to sampling bias. It will depend on type of sampler, sampling locetion, sample Bize, and many other factors. The parameters most seriously affected by this bias are m and a; the value of should not be very sensitive. For most collecting devices and sampling locations, the amount of fallout collected with respect to the local terrain > (average) will be low. However, this generalization is not valid for the island collecting stations at Operation CASTLE where the collectors were at grade level and were not recovered for several days after shot. In the meantime, both inert coral and fallout particles drifted into the collectors by action of the wind. , Combining the various correction factors which, if known, would provide a more reliable scaling function for each of the contour ratios than those for the idealized fallout model gives M(t) = K (XW) ba @ [Dent rpa(t de,(t) + EDje,,(t)| (24) and FD(t) . 6.89 x 10°"" “2h w -1 bq [Pop(t p(t p(t) + EDzc3t3(*)] pok/NY (25) For the idealized model function, K(X,W) is equal to 1.23 x 1072? w-038 for all values of x. are Tep(t) and Ty: The only terms in Eq. 25 that depend on distance b Ls