3.1.1 Data Reduction. A discussion of data reduction is presented here to indicate tne : tations and accuracy of the results. More detailed treatment, if required, will be found in particular section dealing with each aspect of the phenomenon. The main body of the Proje 2.3 results are dependent upon GITR measurements and subsequent readout by the GITOUT vice. Because of the nature of the two shots Studied, all other measurements were found t of secondary importance. Therefore, matters of principal interest are GITR response cor tions, errors and limitations due to the GITOUT, and possible errors due to time resolutio: plotting, and plot-reading procedures. AS described in detail in Section C.1 of this report, the directional response of the GITR detectors wasdetermined using a number of X-ray energies, a Cs!°7 source, and a Co®? sor (Figures 1.4, C.2, and C.3). Ali detectors were calibrated before and after each shot, usir 30°~beam, 120-curte Cs’! source, which was directed toward the top of the detector dome | direction of the calibrating source is designated as zero degrees in all response plots). Al components were precisely positioned; thus, the calibration procedure accurately reproducs detector response to a known source carefully alined with the vertical axis of the detecting chamber. The high-range chamber showed an appreciable increase in response when posit at right angles to the calibration beam; thus, it waS necessary to apply a correction factor } the case of a detector completely surrounded by a radiating source. An integrated detector response was determined by weighting each meaSured responsefc given 5° segment by the total solid angle subtended by that segment. The reciprocal of this integrated response thus represented a correction factor that normalized the total response unity. Since the maximum roll of a coracle is 45° and since the principal radiating source ° found to be airborne material, the total response for the vertically mounted standard GITR numerically integrated over a figure of revolution representing the measured directional re sponse from 0°to 135°. Thus, the factors employed in this report correct for radiation inc dent over 3.412 rather than 47 steradians (a response normalized to 47 steradians would be and 4 percent greater than those reported for the high- and low-range chambers, respectiv: Since the detectors for both the ASEL and UW-GITR’s were mounted with their axes of 5 metry in the horizontal plane, at least half the effects due to roll would cance! out; thus, co rection factors for these detectors were approximated by those for a 4m response. A weigh average of the detector response was determined both for the effects of roll and for attenua’ due to the coracle itself in these two latter cases. Both calculations resulted in only small deviations from the 4m response. Because the UW detector case was wrapped in the instrur control cable, a 2-percent attenuation factor was applied when it was used as a secondary C A simple 47 response was employed for the UW-GITR when used underwater. Correction factors were calculated for each energy for which directional response measi ments had been made and were then combined by weighting each factor in accordance with t. gamma energy groupings for instantaneous thermal neutron fission of U?™ (Reference 83). final correction factors, which were nearly equal to those determined for the Co® direction response, are presented in Table 3.2. These correction factors have been applied to all GI data presented in this report; the original gamma doSe rates may be obtained from tabulate or plotted data by applying the reciprocal of the appropriate factor. These correction factors are strictly applicable only to the case of a uniform radiating c completely surrounding the detector, a condition that is most closely approximated during 7 dose rates. The actual response of the GITR will vary as the radiating cloud approaches ar departs; however, for the conditions encountered during these events, this variation is alwe less than the stated limits of accuracy. Since neither the distribution of radioactive materi within the base surge, nor the velocity of approach is accurately known, no attempt has bee made to correct for changes in total response due to moving sources. Although no directional response corrections for Source movement were made, the diffe ences between ASEL and std-GITR dose rate records at early times can be partially explair by this means (Section 3.2). The differences in detector response during the approach of a finite source are shown in Figures 3.1 through 3.3. 79 These responses were determined as 5