the submerged instrument. Finally, some definite assumption must be made regarding the spectral distribution of energy existing in the sea at the time of the measurement. B.2 ESTIMATION OF THE RESPONSE TO UNDERWATER RADIATION The spectral character of the radiation arriving at the surfaces of a submerged gammadetector depends upon the character of the radiating sources and also upon the degree to which scattering degrades the radiation before it arrives at the detector. And since response of a gamma dosimeter is never completely independent of photon energy, consideration must be made both of initial photon energy and of scattering before a practical calibration of the instrument can be established. B.2.1 Estimates of the Source Spectra. Fortunately, estimates of the photon energy spectraof fallout material are available from other experiments. Estimaicd energy spectra supplied by Dr. Scoville of AFSWP were madeuseof in this report; Table B.1 lists these estimated spectra separately for each of the first four days. And in the right-hand column is to be found an averageof the four spectra. Numerical computations were carried out separately with each of the four spectra and the results were then averaged; however, it was later realized that the limited accuracy of the experimental measurements did not justify this detail and an average spectra might just as well have been assumedat the outset. Figure B.1 shows the four estimated spectra reduced to histograms. B.2.2 Calculations of theUnderwater Dose Spectra Corresponding to the Assumed Fallout Source Spectra. The amount by which the emitted radiation is degraded by scattering before reaching the submerged gamma detector can be determined approximately. Measurements at sea were made under circumstances approximating the mathematically simple case of a uniform distribution of activity in an infinite body of water. This scattering problem has been investigated with the ald of modern computers, and AFSWP Report 502A (1954) presents numerical solutions in graphical form. By use of these graphs, the spectrum of energy which arrives at any point inside the large scattering medium can be derived from the spectrum of the energy emitted from the sources. Figure B.2 shows the results when each of the four source spectra of Figure B.1 are degraded by scattering inside the large distributed source. These, therefore, must be taken to be the spectra of the gamma ray energy which the submerged instrument must measure. These degraded spectra are given again in tabular form in Columns 3, 5, 7, and 9 in Table B.2 where their ordinates are labeled D (Ej) consistent with the nomenclature of the AFSWP 502A Report. Table B.2 will be discussed further in the conclusion of this appendix. The intervals appearing in the abscissa of the “dose” spectra of Figure B.2 were chosen arbitrarily for convenience in the computations. B.2.3 Instrument Response Variation Due to Photon Energy Variation Alone. At the Bureau of Standards the instruments were exposed normal to their axes to several radiations; to X-rays correspondingto effective potentials of 58, 87, 132, 168, and 222 kev and also to radium and cobalt beams of known intensity. Only the results for the Mark If instrument will be considered here in any detail. The variation in its response to rays normal to its axis is summarized in Figure B.3. It will be noted that the photon energy has relatively small influence upon the response to rays norma! to the axis unless the photon energy happens to be less than about 0.080 Mev. B.2.4 Instrument Response Variation Due to Angle of Incidence Alone. The heavy-walled instruments, of course, responded differently when the angle of incidence of the rays differed from 90 degrees. Figure _ B.4 showsthe results of tests on Mark II in the Bureau when the incident angle was varied from 0 degrees to 180 degrees; the results (@) are given as response relative to that response at 90 degrees incidence. B.2.5 Estimates of 4 Pi Monochromatic Sensitivity. From the data in Figure B.4it can be determined what the effect would be if the radiating source were spread uniformly around the detector. It can be shown by use of the experimental values of ¢ (@) and by geometrical considerations that the ratio of response to a uniform distribution of sources to the response to a concentration of the sources at 90 degrees incidence will be: rs % f (6) sin @ de e Where: @ = angle of incidence in radians. ® 83