VaacraeeesAS om Dimensions of Atomic Clouds t End: of Three Hours The best understanding of how the utomic clouds were distersed in the atmosnhere is obtained from stulies of the effects of the upper vinds at all of the elevations at which cloud miterial occurred, As explained on pare 34 under the title Dispersions Acraent Atouic Clouds, an attonnot was mide to study the spreading of the crimary misses of the clouds by means of theadolite data for each side of the ocritury masses. ‘Acinuth ancle datu from thegdolites do not take into consideration the differences in rind directions and velocities which occurred at different levels within the primary masses, and such data are greatly affected by this variation of wind with altitude. Also, practical considerations of atomic clouds require thet the entire cloud columns be studied rather thon just the vrimry mgses. In these diayrans azimuth ancle inforwation has not been used. The shapes of the clouds have been determlued from the extimated upper wind date, torether with photographs and sketches. For a check, Scale models were mode from these diayrims and the models were comvared wilh the rhotogrephs. As far as can be determined the shapes and dimensions of the clouds os show in the djarroms are approximtely the gone as the The procedure used in the construction of these diaprrus was to drnn the wind vectors shorm on pupes 15, 16, and 17, to scale in order to represent a three hour movement on a chart of the Enivetok area. The ends of these vectors represent the positions of the etomic cloud mitericl at the respective elevations, The amounts of cloud mterinl at these nésitionsare simil:ted by the areas of the circles dram around the ends of the vectors. ‘The sises of the circles weredetermined by means of the photoprephs and sketches, The amounts of cloud miterial at various clevations rere estimated visually by conmring in a relative monner tha width and thickness of the mterjal with the overn]l size of the cloud. After the circles nere drawn they were connected by tangents common to any two consecutive circles, and the area enclosed ras shaded in. ‘here overlapping layers of cloud occur, additional shading has been added. At levels whare thqre is no cloud material visible in the photographs, or where none was seen, the lines representing edges have not becn removed, but there is no shading, . actual clouds, In order to verceive the effects of the wind it 1s necessary to study its effects for some period of time lone enough for easily recopnized changes Jn slupe to occur. Hire, a teriod of three hours has been used, After opproximitely three hours the sluipes of the clouds could not be determined by ground observers or by vhetogrephs made from surface lerations. Three hours is the longest time interval for which the celculited ahnpes can be conrared with the aoctunl ahopes, Uper wind data can be used to coleulute the oc proxiimtea dimensicns of the clouds at any tine after H-hour although no visual check of the shape of the clouds is possible, & calculztion besed just on upper wind data would necessarily assume thit cloud material was not changing elevation because of fall-out or convective activity. Also, the disper- sion at any particular elevation vould either be neglected or roughly estimated. Insmc~tion of photopraphs shows that fall-out of materiel from the upper ports of the atomic cloudy was not vercentible during the first three hours. On these horizotal vrojuctions of the clouds, dimensions have been added, These dimensions are thouvht to be close to those of the actual clouds as they are largely a Ametion of thea wind directions and velocities, Errors in aprroximating the area covered by material at any elevation would not, greatly affect the overall dimensions of a cloud, When dimensions are given ror tha length, width, or diameter of an atomic cloud, it should be retlizead that a very irremal:r shape is being deseribed, After producing the dingraums avrroximting the vievs which an observer would get by looking dovm an the clouds, the diagrams tere used to rroducec a projection in the wertical. Such vertical projections very mich resemble the views which were obtained from the ships or at Eninetok. The width of the cloud in the vertical nrojections is the sams as in the horizontal projections at the 5,000 foot intervals, but in between these elevations there there has been neo attemnt mide to esti ule the width. In other words, the ends of the lines rerresenting the widths at 5,000 foct interwi ls in the vertical projection have bean connected by straight lines, This givea the vertical projections a thinner apnearance than was observed in the actual clouds, ven so, if these vertical orojections are compared with the actual cloud sketches and photoeranhs, it will be secon that there is a very close resemblance. In comparing the vhotogravhy ond diagrams it should be remembered thatthere was a convergence of the lines which formed the shotograniiie images at Lnivetok, or aboard one of the observing ships; whereas maralle] lines in a north-south direction were used to produce the vertical projections. The dispersion at any particuler elevetion because of eddy notions in the atmosphere 4s more important then fall-out in a study of the visille cloud. The process uhich is resnonsible for the dispersing of air-borne mteris] because of circulir motions, or eddies, is frequently referred to a3 "eddy conductivity." It is tie factors ulich enter into considerations of eddy conductivity which determine the rate at which the cloud apreuds at any one level, Some of these factors ure the horizonte] area of the cloud miterial, the amount of mtericl] above and below the level in question, the wind shear at that level, and the temperature lapse irte. Such frctors determine the exse vith which cirewlar motions can cecur, The eddy conductivity processes in the free utvaosphure ara not understood sufficiently to m:ke spplieation to this study of the atomic clouds, Molecular diffusion plays a part in the svread of the cloud, but the spread of material in this way is not thought to be significant when commred to the much cprenter dispersing effects of the winds, Some of the differences between the diagrams and the nhotographs are accounted for by this chanre in rerspective. XRAY DAY~:s thio cloud semirated at aoroximately 20,000 feet, no shading is show near this altitude, also, since the shotorriehs de not shot the long aru of cloud seeping back between $0,000 and 55,000 feet, the shading has been elimineted from tit part of the verticnl crojections of the cloud, This rart of the cloud was studied in the model of the cloud which was constructed. Lither there wau not enough cloud material at these altitudes to form this arm or the winds were different from those estimeted, It 1s aven possible that the moisture in this top vart of the visible cloud evaporated so that it could no longer be seen after three hours, -38-