defined in a practical sense by investigating the fraction of the flux reaching the detector which originates from a given area on the ground. This can be obtained using Equation (3) combined with the relative angular response of the detector given in Figure 3-3. The results for 60 keV gamma rays are shown in Figure 3-5. It ean be seen that 95 percent of the total flux originates from a circle with a diameter of approximately 21 m, while 99 percent of the total originates from a circle having a diameter of approximately 25 m. Thus a 30 percent increase in area at the edge of the field-of-view only contributes an additional 4 percent to the total flux. In going from a circle 21 m in diameterto a circle 30 min diameter, the total area is doubled. However, the flux arriving at the detector from this additional area represents only 5 percent of the total. Due to the collimator, all 60 keV gammaraysoriginating beyond a circle of approximately 30 m in diameterare cut off. It can also be seen that minor variations in the detector angular response from system to system would not significantly affect the results of the in situ measurement. On most islands at Enewetak it was necessary to cut roads through the brush to survey in a grid and to allow the IMP access between locations. In many cases the method used to clear away the brush led to significant soil disturbance within the approximately 3 to 4-m wide area of the road. Figure 3-5 can be used to estimate the fraction of the total flux which originates from this disturbed area. The detector was routinely suspended directly over the center of the road. From Figure 3-5 it is seen that approximately 10 percent of the total flux originates from a circle with a diameter of 3.5 m directly under the detector. This entire area was normally within the road. The road also oceupies approximately 15 percent of the remaining area which contributes the other 90 percent of the total flux. Thus the disturbed area within the road contributed about 25 percent of the total flux reaching the detector. 3.2.9 Comparison with Soil Sample Data In order to obtain an independent measurement which could be used as a quality control check on the in situ measurements, a soil sampling program was established which attempted to obtain a sample which was representative of the average concentration within the area sampled by the IMP. A total of 109 locations on 17 different islands were compared using both techniques. Two soil sample composites, each comprised of 6 samples, were analyzed for each measurement location. (See Section 4.2 for details on the soil sampling program.) Results of the comparisons are summarized in Tech Note 8. Based on final IMP data (see Tech Note 23), the ratio of the mean of the soil sample results to the mean of the IMP results was 1.05. A difference of approximately 10 percent (based on laboratory soil moisture measurements) was expected since the soil sample results were expressed in terms of dry weight rather than in situ or wet weight as given by the IMP. After correcting for this difference in reporting methodology, the IMP mean value was approximately 5 percent greater than that given by the soil sample data. There are a number of factors which could account for the measurement difference. Probably the most important is the fact that the soil sample results, for each location, were based on a measurement of several thousand em? ofsoil compared to approximately 10 to 15 million em? of soil for the IMP measurement. This fact becomes more important when combined with data obtained on Tilda (See Tech Note 8) which showed that there could be a high degree of variability in 241am activity in both the horizontal and vertical directions within a single IMP measurement location. For many of the locations sampled, the two soil sample composites obtained within the same area gave significantly different results, in some cases by as much as a factor of 2 or 3. This again indicated that there could be a high degree of variability within a given measurement location. Because of this, one would not necessarily expect to achieve agreement at any given measurement location between soil sample analysis and an IMP measurement. This was indeed found to be the case. However, based on a large number of comparisons, the overall agreement was considered excellent. 3.2.10 Results for 13%7Cs and 59¢p9 Although the primary function of the in situ measurement system at Enewetak was to obtain surface (0-3 em) concentration values for 24. am, complete spectral data were measurement location for gamma-ray energies up to approximately 1500 keV. obtained at each The rather simple data reduction program used in the field, however, only processed these data for 241 am, 155py, IS%cs and 6%. The !55Eu data were used to correct the 98 24lam data due to interference from