187 ZERO-LEVEL LOCATION, CHANNELS TOTAL-ABSORPTION GAMMA-RAY SPECTROSCOPY 180 181 182 183 184 185 186 187 188 189 190 2.615=MEV PHOTOPEAK LOCATION, CHANNELS/ 200 Fig. 4—Bismuth-214 matrix output error as a function of zero level and “TI! photopeak location. tion of the °T1 peak at channel 184.5 and the zero-energy position at channel O for perfect calibration and the errors created by improper calibration of the average field spectrum for the matrix amounts of 7453, Figure 5 shows the errors introduced in the 407a amounts present in May 1963. The radionuclides initially detected to be present on Dec. 22, 1962, were “871, “4ce—™4pr, “Ba !9L ag MK MZ r_*Nb, ics “mn, '°Sb, ‘1 and an unknown cluster providing lines of about 0.5 Mev. A recent study has attempted to determine the sources that create the energy line at 0.5 Mev. Calibrations were obtained for each noncoincident gamma source in close geometry with the use ofa large tank of sand contaminated with the desired radioactive source. For those sources providing the coincident gamma radiation, a solution containing the source was spread and watered into the ground over a large detector solid angle on the ground surface. With all input calibrations then known, the amounts of each input relative to the calibration amounts were determined by matrix-reduction techniques. These amounts were determined for each gamma-ray-producing radionuclide for each 40-min spectrum and plotted as a function of time in units of months. A typical period is shown in Fig. 6, which shows ®Zr—*°Nb amounts. Rainfall modulates the measured amounts of all inputs to a high degree. Figure 7 shows the measured amounts of *!*Bi found to be pres-