~~. $3.3 Special Problems and Solutions. Models can be checked most readily by means of ' taliout-buildup data, because this depends only on the aerodynamic properties of the particles, their initial distribution in the cloud, and intervening meteorological conditions. The construction of land-equivalent radiation contours, on the other hand, requires characterization of the composition and radiations of the fallout in addition to information on the total amount deposited. ,. 1.3.4 Radionuclide Composition and Radiation Characteristics. In the present case, for ex- : ample, expioratory attempts to resolve beta-decay curves into major components failed, because “at the latest times measured, the gross activity was generally still not decaying in accordance 2 with the computed fission-product disintegration rate. It was known that, at certain times, induced activities in the actinides alone could upset the decay constant attributed to fission products, and that the salting agents present in some of the devices could be expected to influence _the gross decay rate to a greater or lesser extent depending on the amounts, half lives, and ‘decay schemes of the activated products. The extent to which the properties of the actual fission products resembled those of thermally fissioned U*™ and fast fission of U*™ was not known, nor -were the effects of radionuclide fractionation. In order to establish the photon-emission characteristics of the source, a reliable method of calculating the gamma-ray properties of a defined ‘quantity and distribution of nuclear-detonation products had to be developed. Without such infor- mation, measurements of gamma-ionization rate and sample activity, made at a variety of times, ‘could not be compared, nor the results applied in biological-hazardstudies. Fission-product, induced-product, and fractionation corrections can be madeonthebasis of “radiochemical analyses of samples for important nuclides. This leads to an average radionuclide ‘ composition from which the emission rate and energy distribution of gamma photons can be computed for various times. A photon-decay curve can then be prepared for any counter with known response characteristics and, by calculating ionization rates at the same times, a corresponding fonization-decay curve. These curves can in turn be compared with experimental curves to check the basic composition and used to reduce counter and survey-meter readings. 1.3.5 Sampling Bias. Because the presence of the collection system itself usually distorts the local air stream, corrections for sample bias are also required before the total fallout de“Posited at a point may be determined. To make such corrections, the sampling arraysat all stations must be geometrically identical, so that their collections may be compared when corrected for wind velocity, and an independent and absolute measure of the total fallout deposited at one or more of the stations must be obtained. The latter is often difficult, if not impossible, to do and for this reason it is desirable to express radiological effects, such as dose rate, in terms of a reference fission density. Insertion of the best estimate of the actual fission density then leads to the computed infinite-plane ionization rate for that case. In principle, on the deck of a ship large enoughto simulate an infinite plane, the same falloutradiation measurements can be made as on a land mass. In actual fact, however, there are imPortant differences: an additional deposition bias exists because of the distortion of the airflow * around the ship; the collecting surfaces on the ship are less retentive than a land plane, and their geometric configuration is different; a partial washdown must be used if the ship is manned, and this requires headway into the surface wind in order to maintain position and avoid sample Contamination in the unwashed area. For these reasons, the bias problem is even more severe ' aboard ship than on land. The preceding considerations were applied in the development of the present experiment and will be reflected in the treatment of the data. All major sampling stations were constructed alike and included an instrument for measuring wind velocity. The buried-tray array surround- ing the major station on Site How was intended to provide one calibration point, and it was hoped that another could be derived from the water-sampling measurements. In the analysis which follows, fractionation corrections will be made and radiological quantities expressed in terms of 104 fissions wherever possible. Relative-bias corrections will be included for each major Station, and an attempt will also be made to assess absolute bias for these stations. 17