World-Wide Fallout from Operation Castle

Page 2

MAPS OF DAILY FALLOUT,................... 39 The work reportedon here was performedunder the directionof was establishd by the Healthand Safety Laboratory, New York Operations Office, Atotic EnergyCommission, Merril Eisenbud, DanielE. Lynch of the Healthand Safety Laboratoryservedas coordinatorof the program. performedthe laboriousand pemstax3ng plottingof data and preparation of the finishedmanuscript.

Page 3

Fallout Monitoring Network, Pacific Hemisphere ......2 Fallout Monitoring Netuork, Atlantic Hemisphere. .....3 Winds Aloft for Castle Events. ..............7 Meteorological Trajectories for Burst No. 5, Yankee. . . . 16 Total Fallout From Castle Series as of July 1, Total Fallout From Castle Series as of July 1, Total Fallout From Bravo, Pacific Hemisnhere . Total Fallout From Bravo, Atlantic Hemisphere. Total Fallout From Union, Pacific Hemisphere . Total Fallout From Union, Atlantic Hemisphere. Total Fallout From Yankee, Pacific Hemisphere. Total Fallout From Yankee, Atlantic Hemisphere &ximm Activity on Sampling Day at Individual ~aximm Activity on Sampling Day at Individual & j A.l@ - A.188 Average Daily Fallout During Month of June,

Page 4

A world-wide netvworkof gumned film stations xas established to monitor fallout followiq Operation Castle. Xlthoqh meteorological data were poor, a general connection of troyos?heric flow patterns with observed fallout was evident. There ws a tendency for debris to re.win in tro?ical latitudes, with incursions into tk teverate regions associated with mtecrological disturbances of t.he predo.tinantlyzonal flou. As tha season advanced, such incursions becam more evident. Outside of the tropics, tLhesouthwestern United States received the greatest total fallout, about five t_imest-hat-_ individual station in the It is concluded that the probability of early fallout in inhabited regiok would be reduced bp holding Pacific test series in the wkter months.

Page 7

L reassignedto the appropriateburst. In the case of observationsin the Pacificand adjoiningregions, it was usually possible to determine the burst respocsible for the activityfrom an exarcdnation ; .I of the trajectoriesof the debrisin conjunctionwith observed necessaryto use the arbitrarilyassignedburst. 4.1 maps of daily falloutvaluesindicatethe burst assignmentused in computingthe decay correction, Unlessotherwiseindicated,all radioactivityis reportedin unitsof disintegrationsper rrdnuteper squarefoot of greed film, decayed to 100 days after the dev of the burst. The twTo2_-- --- law for the decay of fission prodscmty has been used throughout. The maps of.dailyfalloutincludeonly the data from the land stations,since thereis considerableuncertaintyin the ship data. The locationsof the shipswere Imperfectlyknown and the procedures for avoidingcross-contatinationof samolesin handlingand mailing, particularly on ships exposed to heavy fallout at soT?etime d$.ng . their voyage, were not adequate. The SND data were utilized in the drawire of isol,ines of activity on the failout m~s and in the

Page 8

The bursta of the Castleseriesare given in Table 2.1. The first burst was detonatedon and the las; on Enik-etokAtoll. most of the radioactiveclouis created in tk Castle seriesextendedto very great heights, and the greaterpart of the cloud in levelsbeyond tk reach of routine meteorological observations. For this reason, it tis been Possible to prepare adequatemteorological ';rajectoriesto dete&e tte cath of the debrisat variouslevels. T&network of uper air obserm stationsin the tropicsis etireir.elysparseat best, and wind reports adjacentareas established-especiallyfor this test series; hen at these stations,the highestotiservationsrarely extend above

Page 9

even e&lower lemls, 811 trajectoriesgiven in this report were computedby personnel of the Air WeatherService-@uPA Branch)and The tenperaturesottndingsfor all of the Castleburstswere very similarin theirmajor feetures. There were no pronounced invwrsionsin the lowerlayers (exceptfor an inversionat about 7,000feet duringRomeo). The air w8s quite moist up to ebout ebove, The winds obtainedfrom observationsmade at or near esch of the shotsare shownin Figure 2.1. . ?’F The first burst of the Castle series,Bravo, MS detonatedfrom resultingcloud of radioactivedebrisreachedto feet Tha tropopause at this time ~8s at about s&,ooO feet, extendedto about 6,000 feet, with light westerlywinds incre8sing throughoutthe stratosphereto the highesteltitudereachedby the were eesterlyat about 50 knots. Trajectoriesof the lower parts bf tf-ecloud are shown in o servationsin the .hrstillsand at Guam) indicatesgeneralessterly winds in the lower stratosphere,so that this portionof the cloud moved toward the Phillipines.ho observationsto indicate the mov8ment of the cloud above lOO,OoO feet 8re availeble. However, it is likely thet essterly winds prewiled et these levels. The daily f8=0& maps for the period following th&Bram> test 8re particulerly interesting in that t%e background of fission product activity from previous tests Was negligible and the succeeding burst d%d not occur until 26 d8yS leter, so that the progressionof areas of falloutfrom dsy to day is more easily seen.

Page 12

The eiastuard mow debrisreachedthe Americason March 9 and 8, indicatingan average westi wind of -about40 knots, in good agreeaaantwith the few tid,observationsavailablein the upper troposphere.Althoughthe progressionof debris to the west appearsto be in good agreeraentwith tha 5,0&I-foottrajectory, indicatingthat the transportoccurredin the trade wind layers, it is entirelypossLblethat stratospheric debrismoving with the tpperlevel easterliescontributedto this falloutalso. The most stiking fact which emergesfrom a study of the falloutin the period followingthe Bravo test is the tendency far the debris to remain in the tropicallatitudes. By far the largestamountsof falloutoccurredin the latitudeband from 10'S to 20*N,+ith occasionalexcursionsinto the more temperate latitudesof each hemisphere,particularlyin the Americas. An exampleof this can be seen in tk southwesternUnitedStatesin the period beginningMarch 1s. At this time, a deco low pressure systemextendingthroughmost of the tropospherewas locatedjust ~8~8y. An examinationof the falloutmaps revealstihatfallout during this period was associatedwith t.kesoutkesterlywinds, which carrieddebris from the tropicalregions. It is signM.cant falloutValues occurredduring the first Uhree days of the period when there was no precipitation,and even on the 15th, when there ,in the region dominatedby the southwesterlywinds and uas not, closelyassociatedwith the existenceof precicitation;'A somewhat precipitationwas more uidespreacL% this case and may have had more influenceon the observedfalloutpatterns. . The secondburst of the Castle series??.o~eo? The Kind observationsassociatedwith this burstshowed light easterlywinds at virtuallyall levels increasingin 4 speedabove SO,w feet to a m&&&n of 92 knots from the SE at the

Page 14

and then eastward within a very short time. The lowest levels continued westward and the 18,000-foot trajectory appeared to curve back towards the United States on the 28th, although the meteorological data is uncertain. Winds in the stratosphere up to the level of the tOD of tha cloud were probably from the east, carrying most of the mushroom westward. A1thoug.h almost a month elapsed between the first and second bursts of the. Castle series, enough’debris from tha first burst was present to seriously interfere with attempts to trace the progress of the second cloud by an examination of fallout data. For example, an increase in deposited activity occurred at some stations in Central and South America on March 31 and -4pril 1, several days before the meteorological trajectories would indicate the arrival of debris, It is not certain if this is due to the complete lack of meteorological observations in the Eastern Pacific and the winds were really stronger than assumed, or that the debris was actually from t.he Bravo burst. (Note: Since all fallout data is extrapolated to 100 days after the assigned burst, values ssfiq?t?d to different bursts cannot be comzared d5rectly. The extraoola tion factor denends both on the day,of t.he burst and on the day the saqle *wascounted. For the areas mentioned in this oaragrarh, values assigned to burst 2 would have ‘XI be increased by about a factor of three if the debris By April 2nd and 3rd ,increases in a cti :<tv are evident along the Gulf Coast of the United States and certainly by the &th and 5th there is good evidence that debris from tMs burst has arrived over the United States. Again, as when fresh Bravo debris was present’,Sallout seemed. to occur irrespective 0f”‘ithe occlurrence of ‘.precipitation. The progression of debris we&ward from the. test--site appears tohave been more rapid than indicated by the low-level trajectories at 5,000 ad 10,000 feet, af least for t-he first few days follo*&ng t,k burst. !?hether the arrival of debris at YaD and Koror on Xarch 29 easterlies or in faster-t!-ar.-cbs-lrvsdlow-level trades is not certain, Again, as tith tiravodebris, there has a zar‘rc5dtendency for t.'?e fallout to occur in the tropical areas, -xithoccasional incursions into the United States. Koon, the third burst of the qies Bikini 3%13‘20 GCT, Aoril 6. 19%. bit. cloudy conditions prevented

Page 15

The winds were easterly to 5,000 feet, light southerly above to about 30,000 feet, becoming westerly about 30-40 knots to the tropopause. tiecauseof large amountscf f&Lout from the second burst, which occurred eleven days earlier, $t was impossgble to trace the history of the debris from Koon. ,lowest layers moved uestvard, the mid-tropospheric portion milled about to the north of the Marshalls for many days 811dthe upper portion moved eastward, remaining south of the'Hawaiian Islands, reaching the southwestern states on April 13. No fallout station reported debris which can be definitely assigned to this burst, elthough it is likely that some of the activity assigned to Romeo is a ndxture of debris from the two bursts. No fallout has been assigned to Koon in this report. The fourth test of the series. Union'iK.deton8tedat 3ikini at becoming westerly near the tropopause, and strong easterlies above 70,000 feet. Trajectories of this burst are shxounin Figure 2.5. If the 30,000- and 40,000-foot trajector,,ias are correct, very little f8llOUt 118sevident from these levels, since no debris MS detected days is in good agreement with t& U&000-foot meteorological trajectory. It is very possible t-hat'thelack-of mateorological data resulted in erroneous trajectories at 30,000 and 40,000 feet, since debris arriving in Central and South America on May 5 MS most likely transported at these levels. Fallout to,t.hewest of Bikini seemed to be in good agr,,.'oavnt kiltht:beirajectcries. It should be noted that even though a month had elapsed since the last tropic_a,qZ5i% is b;rno means certain that the debris assigned to &&n is not from an earlier burst, or that some of the activity Bssumed to be from Romeo is not actually from Union. Yankee, the fifth burst of th2jXER5es, w8sAe_$ona_tedfrom

Page 18

Although the winds, In general, were similar to those of the previous bursts, th3 westerly winds just belo-wthe -tropopauseattained Mgher speeds t than had occurred during the previous tests, 55 to 65 knots at this burst cmtinuad over t& w&tern &If of tti Mted States for 8 period of more than 8 week. The falloutfrom Yankee in this xegian exceeded, by almost an order of magnittlde,the fallout from =~,_~~,the other tests of the series. The westward moving debris appeared to proceed faster than indicated by the low-level trades, reaching Koror by.May 6 and Singapore by my 9. Again, it is Very possible that high-level easterlies carried the debris, aince the. '2$3O knot uirds required are somewhat fester than expected in the trades of t& Western Pacific. The last test of the series, Nectar, was tiaa only burst detonated from Fsriwetok. It occurred at 18x) GCT, ?by 13, 19%, to the b8se of the stratosphere. The trajectories from this burst (Figure2.7) began with 8 slightly greater component towards the north than for the previous bursta. Slice Yankee and tictar &re separated *by only Nne days, it is virtually impossible to distlngui~h between debris from the two bursts. An attempt to separate the two sources of debris was made for tha first mek follo>@r~ Nactar,. but. MS not attempted bayond this time. Da-fly fallout maps for the remainder of the month, &y 22-31, are given with all dab extrapolated to 100 deps after &actiar because of the arbitrary system of burst ass;_gnment used. However, it is likely ttit the major portion of the fallout reported on these days origin8ted from Yankee. To convert the reported activity to 100 days after Yankee,assumi~ the debris originating from Yankee, the ~811~s given on the maps should be increased bs

Page 21

To conserve space, daily.fa.Ilautlnapsfor the month of June for the month is given, together-withthe number of days for whichdata was available-atesch station. Again, the extrapolat#ar fsbased o~%.Nectar,and activityia shown at 100 days after burst. It 18 also likely that the Ilvljorportionof the fallout ul.Jme originatedfrom Yankeeand all 'vales shouldbe increased by rbout 258 to give -valuesat 100 days after Yankee. b Althoughths discussionof the transportof debrisin the atmospherehas been confinedto essentiallyhorizontaltmjectcries, the actualpaths of individualradioactiveparticlesare complex, three-dimensionalphenomena,influencedby the fall velocitiesof.

Page 22

Tha total world-wide fallout from ea.chof the Castle tests (eXceDt Roan) and from the whole series has been comuuted on the basis of results from the monitoring network. Since none of the stations were located imediately downwind of the test area so as to emerience fallout in the first day or two following a detonation, it is apparent that by far the largest fraction of the fallout, the “close-in” fallout, has not been measured. A comDosite mar,for the comnlete series. showing the total of all fallout occurring through June 30, 19j&, 3rd decayed to July 1, 195&,,is shown in Figures 3.1 and 3.2. These maps contain the cumulative to+21 of all debris 3eoosited on t:y network from February 2& through June 30, 19%. T5e debris has extrapolated ta July 1 on the basis of the burst assignments indicated in Annendix A (except for fallout occurring after ?ay 21, w.hichwas reextranolated to Ya;lk.ee,see Section 2.6). Isolines of activity were interpolated between stations and the average fallout for the wo5 Id was computed, by numerical integration, to be 9101: d/m/f\ for a. total of 2.2.73 megacuries. To obtain t.hetotal fallout due to each of the individual tests, the following crocedure, was zsed. At each station, all fallout assigned to t:~ given b>urst,as indicated on the mars of Ancendix A, =s summed, and titetotal fallout values, in d/m/ft2 at 100 days after burst, were entered on a man. (For th?se comcutations, fallout occurrixq after May 21 was not considered, since there was some doubt as to burst assignment.) In the event that data were missing for an occasional day at a given station, the --missing values were estirratedby interpolation. If data were missing for a number of days, the sum ijas entered in oarentheses and indicated as a .lower limit of ‘activity. Isolines of activity were drawn and th? total fallout commuted by numerical integration. -; :

Page 25

For example,falloutfrom Bravo was not masked by later debris for about 8 month in the region of tests, and could be identified for an even longerperiod in regionsremotefrom the test site, On the other hand, Union debriswas quicklyovershadowedby fallout from Yankee.,which occurrednine days later. the periodfrom February28 to April 5, 19%. with the exception of debrisin a limitedarea which was determinedto be from Romeo t0 100 days after burst,was 1937 d/m/ft2, for a total fallout of h. 79 megacuries, or 3.74 megacuries as of July 1,'1954. Figures3.5 and 3.6 show the total Romeo fallout from the time of the bursts through ~%y 3, 1954. T:ke2world-wideaverage activity at 100 days after burst was 1445 d/m/ft for a total of 3.57 megacuries, or 3.71 megacuries on July 1. tiedebris was assigned to the third burst, K’oon. Fallout from Union (Figures 3.7 and 3.8) covers the period through May 12, a somew.%t shorter Teriod than the first two bursts, since Yankee was detonated only nine days after Union. The world-wide averagefalloutwas 28h d/m/ft at ,100days after burst for a total of 0.70 megacuries,or 1.13 megacurieson July 1. Yankee cumulativeresults are given in Figures 3.9 and 3.10.. : However, much of the fallout u!!ch occurred beyond …