Since the 5 em averages are exponential, the underlying distribution must also be exponential. If so, the computed values of ¢ should be similar from one location to another (though k would certainly not be constant). It is easier to actually work with 1/¢e for comparison rather than ¢, since 1/c, commonly called the "relaxation length," has units of distance, in this case centimeters. Figures B-19-2 and 3 are histograms of the values of 1/c computed from the 0 to 5 and 5 to 10 em samples and the 5 to 10 and 10 to 15 ecm samples, respectively. While each set has some outliers, the bulk of the values lie between 1.5 and 3.5 em, and the two medians, at 2.51 and 3.09 cm, are quite close together. Since the only data not included in these figures are from disturbed locations or locations where all activity was low, the conclusion of an exponential activity distribution with depth seems well justified. In view of the foregoing, the activity in the 2.5 to 7.5 em interval was computed using the methods described here for each location on Belle and Daisy where this information was required. At disturbed locations and those with very low activity, a linear distribution was assumed; at all other locations, an exponential form was used. Average TRU activities over 1/16 areas were then computed by using the simple means of the 2.5 to 7.5 em estimates. TABLE B-19-1. TRU ACTIVITY IN TYPICAL SUBSURFACE SAMPLES FROM ISLAND BELLE ( MDA = Less than minimum detectable activity) 16-S-8* 16.125-S-7.875 15.875-S-7.875 16.125-S-8.125 15.875-5-8.125 16,.25-S-7.75 15.75-S-7.75 16.25-S-8.25 °15.75-S-8.25 16.5-S-8 16.5-S-7.5 16-S-7.5 15.5-S-7.5 16.25-S-7.25 15.795-S-7.25 14-8-2* 6-N-2* 5.25-N-1.75 6-N-1.5 5.25-N-1.25 Average TRU Activity in Interval, pCi/g 0-5cem 96 433 60 167 279 178 95 75 6 41 671 303 268 42 106 289 130 <MDA 6 6 5-10 em 10-15 em 178 52 10 6 5 26 40 5 8 5 31 34 24 5 32 181 224 <MDA 5 11 10 16 5 <MDA 7 7 17 3 <MDA 5 5 6 14 <MDA 6 32 26 <MDA <MDA <MDA Fon Location eon * Disturbed locations B-19-3