The average americium measurementis 8.1 pCi/g. The sample standard deviation is 0.41 pCi/g (5.0 percent). The average for cesium is 9.92 pCi/g, with a sample standard deviation of 0.32 pCi/g (3.2 percent). Figure B-21-2 shows that, as one might anticipate, there is no apparent correlation between the individual americium and cesium concentration measurements. Linear regression analysis indicates that R42 = 0.03, which supports this assumption. It should be noted that the IMP was not moved during the course of the day. Hence, the above values do not include any error associated with repositioning the detector. It is likely that there was some drying of the soil during the progress of the experiment since it did not rain during the day. The results show no obvious change which might be associated with time of day. Some informal reproducibility studies have been conducted of IMP remeasurements at the same location which involved repositioning the IMP on different days. Data from three comparisons are shown in Table B-21-2: TABLE B-21-2. 24l4m MEASUREMENTS REPEATED ON DIFFERENT DAYS Janet 6-NW-4 Date 241Am (pCi/g) Pearl 4-N-1 Pearl 1-N-1] Date 241Am (pCi/g) Date 241Am(pCi/g) 09/22/77 21.3 + 3.0 10/20/77 19.5 + 2.7 10/28/77 35.2 + 4.7 10/03/77 19.5 + 2.8 10/27/77 18.0 + 2.5 10/28/77 36.7 + 5.8 10/05/77 20.3 + 2.9 11/18/77 18.2 + 2.5 11/18/77 32.2 + 4.4 10/10/77 18.5 + 2,7 11/15/77 17.4 + 2.6 Mean 19.4 + 1.52 18.6 + 0.81 7.8% 4.4% Std. Deviation 34.7 + 2.29 , 6.6% For several islands, reproducibility has been studied by comparison of IMP readings taken several months apart. Different detectors were used for these comparison pairs. Usually, the IMP vehicle and electronics and the operating technician were different. Sometimes the measurement points had been restaked. Comparisons for two islands are given in Table B-21-3. The ratio of old/new americium values is 1.11 + 0.10 for Pearl and 0.97 + 0.12 for Lucy, and for both sets of data combined the ratio is 1.03 + 0.13. Originally, a complete remeasurement of Lucy was planned but the plan was changed due to equipment failure after five locations had been remeasured. A comparison of these five new measurements with five previous measurements was close enough that -ERSP management cancelled the balance of the remeasurements. A set of IMP vs IMP measurements was obtained at the Tilda test plot, and was presented in Table B-8-2 of Tech Note 8 The ratio of IMP I/IMP II] measurements is 1.03 + 0.13 for four pairs of comparisons. Each point compared was itself the average of two measurements. The counting error for each single measurement was 5 to 6 pereent. Tech Note 8 calls "effective area factor" the "detector sensitivity correction factor," and assigns the then used value of 1.1 to it for detector 496. Later investigation showed the proper effective area factor for detector 496 at that time was 1.28 instead of 1.1, as discussed in Tech Note 5.2. 1.00 for detector 513. Area Exp. Control Data given below uses 1.28 for detector 496, and Detector Height (em) Ratio 740 460 1.17 1.08 460 0.86 740 1.03 Mean B-21-3 1.03 + 0.13