G. A. Riley Preliminary Report on the Oceanography of Bikini Atoll. JTF=1 O13Blb il IIA il il A 037257 4 oO a =) Mm o oS PRELIMINARY REPORT ON THE OCEANOGRAPHY OF BIKINI ATOLL yTP 1 . Oceanographic Section Technical Staff 018B1b hee is ee 4 i: 1 ; 5 1 x jee f “a | 2 ' ‘ ’ , e ‘ bs bey Wy oak ‘ : i MTSE A Hf ig Che Sate a ey he kre ATC I fis is RESTRICTED 1 1. SUMMARY The atomic blast will contaminate directly a volume of water which is swall compared to the total volume of the la- goon. This small water mass will increase in size, with cor- responding decrease in the concentration of contaninant, by current transport and by the processes of horizontal and ver- tical diffusion. The water mass contaminated internally by the blast will be spread the full length of the lagoon within about two days. Radioactive materials deposited with the plume or by convective rains following the blast will be spread more widely and will reach the edges of the lagoon sooner, but their concen- tration probably will be relatively low. The current system is particularly important in predicting events subsequent to the blast. The description presented here is the result of a survey conducted during iilarch and April. fFur- ther studies will be made in June to determine whether conditions have changed significantly. The system consists primarily of a wind driven surface current flowing in a WSW direction with an average speed of 0.3 knot (varying slightly with wind velocity), extending to a depth of about 40 feet where it gives way to a thicker and slower (0.1 knot) ENE bottom current. These two currents form a continuous, rotary circulation, with bottom wa- ter upwelling at the eastern end of the lagoon to join the sur- face flow and surface water sinking at the western end. Oceanic water flows into the lagoon continuously over the eastern and northern reefs. The total volume of fiow is about three percent of the voluxe of the lagoon per day. Continuous outflow occurs through the western part of Enyu Channel. ilse- where, channels, passes, and the western reef, the current re- verses with the tide. The tidal flow is strongest through the southwestern passes, but the tidal interchange is relatively in- effective in flushing the lagoon. It is estimated that only 40% of the water leaving the lagoon on the ebb tide is true lagoon water. The remainder is oceanic water that has come into the lagoon on the preceding flood tide. Not much more than 10% of the water entering on the flood tide becomes thoroughly mixed with lagoon water and carried into the general lagoon circula- tion. By far the larger part of the water in the central part of the lagoon has therefore come in over the eastern and northern reefs. As this water flows in, it is absorbed into the rotary circulation of the lagoon, thus gradually renewing the lagoon water, while at the sane time the latter is being flushed out of the southwestern passes at a rate of 3.2% per day. At this rate of flushing, any given mass of water in the lagoon will on RUSTRICTZD 2 the average be reduced to one-half its original volume in 22 days and to one-tenth its volume in two and a half months. The rate of flushing will presumably be somewhat slower than average for water in the northwestern part of the lagoon, which has a relatively closed circulation, and faster in the eastern and southern portion, which is more exposed to tidal interchange. At the tine of test Able a patch of contaminated water will be formed at the surface in the target area. The contamination will ove with the surface current in a WSW direction at a speed of about 0.3 knot (assuring a 10 knot easterly wind), so that its center will have moved about 7 miles from the center of the tar- get area in the course of a day. At the same tine its concentra- tion will be reduced rapidly by vertical and horizontal diffusion. It is estimated that these processes will reduce the concentra- tion to 1% of the initial value in two hours and to 0.0003% in one day (not counting radioactive decay). In test Baker it is likely that the radioactive products will be uniformly distributed from surface to bottom. The patch of contaminated water, originally more or less circular, will be elongated rapidly by currents flowing west at the surface and east at the bottom. The contaninated water at the surface will be diluted by vertical mixing with underlying water at an esti- mated rate of 25% per hour. The reduction in the concentration of the bottom water moving eastward from the target area is ex- pected to be about 8% per hour, the difference being du to the fact that the bottom current is three times as thick as the sur- face current. Therefore part of the radioactive products will be carried away from the target area, but part will be transferred by ver- tical diffusion to the other current and will be ea~ried back again. Thus a strip of contaminated water is developed, which lengthens westward with the speed of the surface current flow put with rapidly diminishing concentration, and eastward with the speed of the bottom current. The maximun concentration ‘vill remain to the eastward of the target area. At the end of the first day the strin of contaminated water is expected to extend from Bikini Island to a point about 7 miles wSwW of the target. The concentration at the western end of the strip is expected to be about 0.01% of the initial value, taking into account vertical and horizontal diffusion but neglecting radioactive decay. At the eastern end of the lagoon the average concentration will be about 1% of the initial value, but there may be patches of upwelling bottom water with a concentration of 10% or wore. RESTRICTED 3 At the end of two days the strip of contaminated water will extend frou Bikini Island to the southwestern passes and will be about two miles wide. The maximum concentration in the eastern end will be less than 1% of the initial value and in the western end about 1 X 1075%, It is difficult to carry the analysis beyond this point, but it is certain that further dilution will take place at a much slower rate, and it will probably require a week or two to reduce the maximum concentration to 0.1% of its original value. If the winds were only 5 knots instead of 10, three or four days would be required to establish the condition de- scribed. If they were 20 knots, it would require about a day and a half. It is not possible to predict accurately what would happen in the case of flat calm or wind directicz other than easterly since neither has been observed. It is certain, however, that the above predictions would have to be modified. It is probable that with several days of calm weather, the internal circulation of the lagoon would be destroyed and that radio- active products would be spread only by tidal currents and a very slow horizontal diffusion. With southerly winds, on the other hand, it is possible that the lagoon circulation would be accelerated by increased inflow through Enyu Channel. After the second day contaminated water will begin to leave the lagoon by a series of ebb tide pulses through the southwestern passes. The amount leaving the lagoon will be very small at first and will increase during the first week or so to a maximum value of about 3% of the total contaminant in a day's time. Thereafter the rate of loss will be about 3% of the reimaining contaminant per day. RESTRICTED 2e GENERAL DESCRIPTION OF THE AREA 2.1 ilorphometry fo) Bikini Atoll (see enclosed chart), centered at 11 35'N, 165922'E, is roughly oval in shape, 23 miles long on the east- west axis, 13 miles north and south. ieasurements of the ghart show that the total area of the lagoon is 641 kn* (192 mi~). About 40% of the lagoon has a depth of 25 to 30 fm. Bottom areas above and below these depths are progressively smaller except for the zone between O and 5 fm. which covers 8% of the lagoon area, twice that of the next deeper zone. The isolated coral heads, neglected in the measurements, would slightly increase the area of the shallowest zone. The total volume.of the lagoon below lowest low water is estimated to be 28 km’. Of this volume, about 72% lies below the general sill depth of 7 fi. in Enyu Channel. Only about 4% of the water lies below the 30 fm. bottom of the deepest sill, Eniirikku Pass. One-third of the circumference of the atoll is composed of islands. They constitute the only portion of the rim over which flow of water is completely prevented. Between the islands are long stretches of shallow reef which together make up about half the circumference of the atoll. They are exposed at lower low tides, and at high tides are covered by up to five feet of water. There are eight passes or channels, which constitute about 20% of the circumference of the atoll, all on the south and southwestern side. The largest one, Enyu Channel, amounts to three-fourths of the total width of the passes and two-thirds of the cross-section- al area. Its sill depth is about 3 to 10 fm., and the underlying reef is visible from the surface throughout its length. In the deeper southwestern passes the remains of the reef are visible at the edges, shelving steeply toward the center of the channels, which cut deeply into the reef. Table 1 summarizes the measurements of the periphery of the atoll and includes an estimate of the cross-sectional area of water in the passes and over the reefs. Table 1. Measurements of the periphery of the atoll [Periphery | Cross-section | Passes Total RESTRICTED 2e2 Oceanography and meteorology 2.21 The current system of the region Bikini Atoll lies in the North Equatorial Current, a wes- terly drift of water largely wind-driven by the NE Trades. The surface water flow is about half a knot. The velocity decreases with depth, but slight flow can be detected at depths of 200 fm. or more. The southerly limit of the current is believed to lie between 6° and 9°N at this longitude, but the available data are meager. It seems likely that the seasonal shift of the Trades and the North Equatorial Current is never large enough to place Bi- kini in the Doldrum Belt; however, it is mentioned in passing that if this should happen, the conclusions in this report about the current system inside and outside the lagoon would be invalid for the period in question. The simple picture of a westerly current is modified and - complicated by the presence of the atoll. Only the upper few feet of water can flow unimpeded into the lagoon. The remain- der splits and passes around the obstruction, giving rise to eddies and to variations in the direction of flow which extend soue dis- tance around the atoll. These currents will be described in more detail. in a subsequent report. The temperature of the surface water is about 80° to 82°F, There is a virtually mixed layer of water in the upper 300 to 400 feet, in which the decrease in temperature with depth is at most 29. Thus the water that enters the lagoon is relatively homogeneous, From the surface to the depth of the deepest sill, no observation has been obtained of a variation of as much as 1°, ah A tide station was operated inside the lagoon off Bikini Island at latitude 11937'N, longitude 165931'E, and another on the seaward side of the sand spit north of the island. The ob- served times of high and low water agreed with the predicted tides for Bikini (USCGS) within the limits of accuracy of the equipment. The observed tidal range averaged 87% of the pre- dicted range. Lowest low water corresponded to a reading of 1.19 feet on the tide staff. The station is believed to be representa- tive of conditions over the entire lagoon for the following rea- sons: (a) The size of the channels permits easy communication with the ocean. (b) No appreciable tidal lag has been found be- tween observations inside and outside the reef. (ec) Winds are steady and storm tides unlikely. The large Pacific tidal wave that caused extensive damage in Alaska and Hawaii on 2 April was recorded at Bikini at 1530 as a single wave raising the water level one and a half feet cy See And eer fie Srihe aat: 1G ROSE. VS? Be Ste as t ane SES: ae Re ae £8 eat MUA 3 8 2 Ne acre A RESTRICTED 6 above normal for a period oi twenty minutes. It was preceded and followed by twelve hours of unusually high sciches, many of them exceeding one foot, with periods of 13 to 15 minutes. Suall lagoon seiches of periods soiiewhat longer than one hour and heights up to 0.2 foot have been recorded frequently. They are of no iliportance in evaluating the circulation of the lagoon. According to HO Mise. 11275, the waves generated inside the lagoon should be 1.5 feet high at the anchorage area, 2.5 feet in the middle of the lagoon, and 3.5 feet at the western end with an 16 knot wind. With a 10 knot wind they should be re- spectively 1 foot, 1.5 teet, and 2 feet. The periods should be between 2 and 3 seconds. These values agree with observations. The trade winds give rise to large breakers on the exposed eastern and northeastern reefs. With an 16 knot wind, the break- ers were founda to be about 10 feet high. If the winds decrease during the sumer, these waves will becowe smaller, and should be about 4 feet high with a 12 knot wind. A swell recording unit has been in operation inside the lagoon near Bikini Island, which has shown the existence of swell about a. foot high and with a period of 9 seconds. Al- though generally too small to be noticeable from large ships at anchor, it breaks sharply against shore on the lagoon side of the reef. It is believed that this swell is not related to the waves generated by the trade wind, since the period differs and since the bottom drops off too steeply off Enyu for the waves to be refracted inside. They are believed to have conuie through the channel from the south and to have been generated in the southern hemisphere. During July and August, the winter season in the southern hemisphere, they may be as high as five feet in the target area. 2.23 Meteorology Table 2 is a summary of meteorological observations obtained during the present investigation, and Figures 1 and 2 show daily wind averages and the diurnal variation in wind speed. It is Table 2. Weather observations froia Bikini lean air Amount |/Priliary Wind] Wind ee temperature | clouds] direction We © eO Oe Hi U [sae] ee ary ee x ; Page $36 at nina Rea bata a let mf ree seers < ts Ee ate ers = us nae G nse ad tay oc 9 WE Gea Se Vee eters a Sane > Fe a ‘ > ‘ 7 ah ey) J a SLONY) G3adS GNIM (S33Y9390) NOILDOAYIG GNIM FIGURE | AVERAGE WIND SPEED AND DIRECTION : » j ' . = t t La oD 22 O vs ec! O LJ lJ 20 Oo w QO 19 2 = 18 Tht 0000 0400 0800 1200 1600 2000 2400 LOCAL TIME FIGURE 2 DIURNAL CYCLE OF WIND SPEED af M = t i i -“ RESTRICTED ff by no means certain that the observed diurnal variation is typi- cal, for the May observations showed practically none. The wind observations are subject to a certain amount of error, first because the velocity was measured with an anemo- meter 90 feet above the water so that the recorded winds were stronger than surface values would be, second because readings were imade during only 10 minutes of each hour. The measurements are sufficiently accurate, however, to serve the purpose of cor= relation with oceanographic phenomena. bayneade erie Tats Sewinls BGs FAVES : 4 iy , é ere | as SORE) ae i ee a fi me hs oe) ere re sir nen che 3 ‘ : Ca) 7 . wi 1 / x i i 4 me 4 ¥ , ny RESTRICTED 3. OCEANOGRAPHY OF THE LAGOON 3.1 Currents Three methods were used to measure currents: (a) current meters, by which the velocity was determined at various depths from surface to bottom and the direction to a depth of about 100 feet, the limit of visibility; (b) current poles, which deter- mined the average drift of the upper fifteen feet of water over periods of from eight hours to a day and a half; (c) dye marker, which was used primarily in the channels and over the reefs, where other methods were impracticable. Figure 3 shows the general drift of the surface water of the lagoon as determined by current vole observations. Data obtained by all three methods are presented in Figures 4 to 7. The circulation of the lagoon as determined by the current measurements is as follows: (a) Over the eastern and northern reefs, continuous inflow results from the fact that outside currents and wave action maintain a gradient in water level between the outer reefs and the lagoon amounting to about 1.5 feet. (bo) Continuous outflow occurs through the western part of Enyu Channel. The volume of this flow is a little more than half the inflow over the reefs. (c) Elsewhere on the periphery of the lagoon the direction of flow changes with the tide. The ebb is stronger than the flood through the southwestern passes, (a) The dominant features of water movement inside the la- goon are a wind-driven surface current flowing in a generally WSW direction and a return current along the bottom. . (e) The surface current extends to a depth of 40 feet or more. Its velocity varies with the wind as shown in Figure &. Throughout the entire lagoon the current is influenced to some extent by the tide, decreasing on the flood and increasing on the ebb and with a more pronounced southerly component on the ebb. Near the southwestern passes the flood tide is strong enough to reverse the surface current. (f) Part of the surface current leaves the lagoon through the passes and channels with each ebb tide; however, the outflow accounts for only about 30% of the total transport into the western end of the lagoon. The remainder sinks and returns as an ENE bottom current, carrying with it some outside water that has come into the lagoon on the flood tide. The bottom current - (SLONW) G3adS GNV NOILOIYIA oe SE Nhelal cl lerep Sei ele lail@he us Scie uted ‘ ‘ ‘ ' ' ' i) 0 4 Uy U Wee 3] — Nw 1F)] ds IO TAHT ee (Z110HHOS3) TIOLV INIMTS Lu Vd NUAHLYON-SANVTSI TIVHSHVN NY¥3O0O Jftalovd ALHON \ y oy s “4 x x E he i ie ae ey |e : € r i ‘ + * i y (‘<---) WOLLOG LY GNV(<—) 3OV4AYNS IV-3GIL HOIH-(SLONY) ALIDONSA LN3YYND | v 3IYNOIS os! en ¥- SOV Sipeog SSN A ROY AY act Ber) — yh ant 771 '@ lg OT THE (Z110HHOS3) TIOLV INIMIG LYVd NYAHLYON-SANVISI TIVHSHVA NV@90 D1419Vd BLYON he, Oe ae, nega me me Li a iW ieey boy : SEA ot Cima ERB gape Seni ania ne np pee ‘ oe + Sea INNO IE Se ay 5 Sen ee SL tjea by arya a = = ee a eer reer’ Sch Ae AR eniccn ae ay gay p he Soe Sage pre Beis eeadaien tia LS Sl SRSA (Soh ie Seetet ae eee eee [oe eee eI a tnee a dar 2% ai ad ‘S 2 (<---) WOLLOS Lv GNV(<—) 3Ov4SYNS LV-3GIL, A@GI-(SLONY) ALIDOTAA LN3AYYND ¢ 3YNd!I4 Sa Pe) wat 197 de OT NHI (ZL10HHOS3) TIOLV INIMIE LUVd NUAHLUON-SANVISL TIVHSUVIN NV490 DI419¥d HIHON Sat aie | pie biteao nena a gor > oe —_ ‘af . ‘ N 7 D z . > = ms D. th 7 = _ b ae ; ‘ 8 ‘ yaa) ~, . af Wie ‘ ys = Fi 7 ; « t, hs fis A Rs “ 3 re * - 2 >| s ad q 3 i 4 = P “4 < r ‘4 ae Liisa wih ; “ies ‘ fr. be Bi = yA : at & = F o. or = j Z DL ai ny & pi yo r ‘1 eal i ‘a reas 4 ; vi tae : . See & t ci a z i n ae x < Y 5 é 7 ae vite ¥ . * * % ] < ’ ‘| ty t t 7 a id oh = ° = ts ‘3 * aan ¥, ‘ Party oh ca t ae * < y hen iw nh wt carne beng se = ost On y- SOV peg SS RO ORT Ry A zue! Bel NN ALI I] Beds HO | AMIE (ZL{0HHOS3) TYTOLY INIT LHVd NUSIMON-SUNWISE TIVHSUVIN N¥ doo Di4tovd HLHON a (<---) Wolice iv QNV(<—) 30v4yYNS Iv-30IL MOT 9 3YN9dI- oe I> ALIDOT3A 8] RAMU Ry IN3dYdND lat " oe 3 (<---) WOLLOG lv GNVW (<—) 39vV4YNS Iv-30!lL GOOTS- SLONY ALIDOISA LNAYYND Te! Fl — eee OO dg OT II (ZL10HHIS3) TIOLV INIMI4& LUYd NYUSJHLHON—-SONVISI TIVHSHYA NVZ90 JI41DVd BLUON -— NNN E nS aE Leasing or ane * ne x sy ah a Mae: oa a . Be $3 ics Ale ~ “ Pore Ries fetes whee, 2 a SACD tg Dw te Ee whe rpcneee =v des re S eines. Pepe se 4 , re “5 CR Ve | Bs: A ea 5 “tel Fishy) We Dinie Ros re SURFACE CURRENTS (KNOTS) © OPEN WATER STATIONS © INSHORE STATIONS 7 TIDE AIDING SURFACE CURRENTS 5 n fe) | WIND SPEED (KNOTS) FIGURE 8 EFFECT OF WIND ON CURRENT SPEED ye RESTRICTED 9 is thicker than the surface current but slower and more compressed laterally (occupying about the central one-half to two-thirds of the area of the lagoon). Its mass transport is probably between 70% and 90% that of the surface current. Like the latter, it is affected by the tide, the speed decreasing on the ebb and the direction changing near the passes. (g) From Bikini Island westward there is a current which runs at wid-depths just inside the northern reefs and s:ore or less parallel to trem. It increases in size and thickness as it progresses westward. The salinity of the water in this cur- rent (see section 3.21) indicates that it is reef water of fair- ly recent origin. Its final disposition in the western end of the lagoon has not been studied, but presumably some small part is lost over the western reefs or through the westernmost passes, while the remainder joins the bottom current. (h) Summarizing these observationst The lagoon derives its water by continuous inflow over the northern and eastern reefs and by tidal interchange along the rest of the periphery, of which the southwestern passes are the riost important. the la- goon has an active internal circulation which consists primarily of a westerly wind-driven surface current and a return flow along the bottom. The current measureiients presented in this section are the framework that will be used in fulfilling the practical reauire- ments of the report, namely the determination of the path of contaisination in the lagoon and the rate of flushing. However, before preosaring the final estimate it is necessary to examine the variations in temperature and salinity in the lagoon, which add to the gene’ °l1 knowledge of lagoon circulation and serve as an in- dependent check on the quantitative results. 3.2 ileasurewents of telperature and chemical constituents 3.21 Horizontal variations in temperature and chemical constituents Since the waters of Bikini lagoon are derived from the rela- tively homogeneous surface layer of the surrounding ocean and are subject to continual interchange with it, it is not to be ex- pected that a nigh degree of variability would occur in the la- goon. However, the small variations that have been observed are useful in analyzing the general system of circulation. Variability can arise in three ways: (a) In the surface water of the surrounding ocean there are slight north-south gradients in temperature and salinity, the temperature increas- on We RESTRICTUD 10 ing southward and the salinity decreasing. Thus the water enter- ing the lagoon from the north is about 0.1°9G colder and 0.39/00 more saline than that which enters the southern passes. (bo) Superimposed on this basic difference is a reef effect. During the short period of its passage over the reef, the water is subjected to an intensification of the surface processes of heating, cooling, and evaporation which in deeper water would be distributed downward by vertical mixing. The salinity of the water coming over the reef is constantly increased by evapora- tion. Assuming what appears to be a reasonable value for eva- poration of 0.5 cm. per day, the salinity will be increased 0.01 to 0.03°/oo, depending on the width of the reef and the strength of the current. The greatest flow over the reefs is on the nor- thern side of the lagoon. Therefore the effect of the reefs is to make the north-south gradient in salinity slightly stronger in the lagoon than it is in the oceanic water outside. The effect of the reefs on temperature appears to be important only locally. The water is heated one to two degrees as it cores over the reef during the day and is cooled at night. The ten- perature is therefore more variable than it is elsewhere in the lagoon, but the net effect on lagoon temperature appears to be negligible. Because of the effect of surf, and possibly by the photosynthesis of reef algae Guring the daytime, the oxygen con- tent of the water is higher near the reef than in the main body of the lagoon. These variations are used in a later section to analyze diffusion rates. (c) What has been said of surface exchanges over the reef applies to a lesser extent to the lagoon as a whole. In the open ocean, the effects of surface heating and evaporation are readily distributed through a mixed layer 300 to 400 feet deep. In the lagoon, with an average depth of 175 feet, these effects are more pronounced. it is estinated that evaporation will in- crease the salinity of the lagoon an average amount of 0¢01°/o00 in three days. The distribution of salinity in the lagoon, shown in Figures 9 to 12, is initially dependent on these three factors but is modified by the existing current pattern. . In general, the evi- dence gained from examination of the salinities corroborates the current data previously shown. ‘The current observations and the additional evidence derived from study of the salinities are com- bined to produce the current patterns shown in Figures 13 and 1k, which although somewhat idealized appear to be logical. The direction of the currents in the north and northwestern part of the lagoon indicates that‘it is an area of relatively closed circulation. The high salinity of the area is addition- al evidence. The presence of water with = salinity about 0.19/90 higher than any incoming water is indicative that some of it re- mains in the same general area a minimum time of 30 days. 30QI1L GOON4S - (°%) ALINITVS 3DvV4YNS 6 3YNOIS wy mins 85 nermuaey |! Daman Fe Nee) Bite QO PH (ZL 7OHHOS3) LIOLV INDI LUVd NUAN.LMON-SGNVISI TIVHSHVW N¥ 400 O1410¥d HLHON au ~ + he y aN 5 $ + ‘" 3 F 5 J j a ; 2 - D . . Bele - q % 1. q * wt - ; ‘ nx ; - in, : oo ar) = i M4 cs . ; $ _ : , Ns et iy) meewed ¢; ' E Pena & ates \ if 1 ect i] i Poa gi te ‘ Hoy F é i \ Ua ‘ / 3 en Pe) - hal 1 Ode RO HI , ou; i MS, 3 d os [ty _ (Z/10HHOS3) ; f TIOLV INIMIY LUuVd NYUFZHLYGN—-SANVISI TIVHSHUVYW NY¥Z9O JI4IDVA HLUON “, \S “hi, il \os 2 ne on gent eon y- Soy vapeg SS 4 ~~ ow pee 4 see) PN ae Beds yy Og (2) 10HH9S3) VIUVd NUAILEMON -SCANWVISEUINTISUVIN Neave year eve HEHE agli BER - (°%) ALININWS WOLLOS 2! aYNOI4 - et en “Ss <---77 = (<--)WOLLOG Lv GNV (<—) H1ld3d SLVIGAWYSLNI YO 35v4YNNS LV SLN3YYND 3AGIL GOON 40 NOILVLN3S3YdaY DILVW3HOS €l aYNDI4 \ ' ' ' 4 S ~. ~ =... 2a 8) — 1 Oe. 1 Ode 0 1 (Z110HHOS3) TIOLV INIXMIY LuVd NHFHLYON—-SAGNV1SI TIVHSYVA MV¥ZOO0 DI41DVE BLUON | a pap ie a ss 7 sayin nears i pO rucLengh (<--)WOLLOY LY GNV (<—) HLddd 4LVIGSAWYSLNI YO 4DVaNNIS LV SLIN3YYND 3GIL BEI JO NOILVLN3S3IYd3ay DILVW3HDS vi SYNOl4 & orst Ew SOV Fupeog S'S me prety pe . wi Le be ° i mh My, e ww yy, 222.31 207] — NAPALI IPT BOS BQO “1 DAA 4 a bapa) 16; ON (ZLOHHOS3) TIOLV INIMITG LUVd NHAHLHYON—-SANVTSI TIVHSHUVN NV@a90 91419Vd HLYON Sec ai Ick la Shee BE y $f “8A8 5; : rs “a7 aS ‘ = 5 E 4 . E a ae a ist ay apace my hy pm rte Teme ° =e oe er te eel le pat . a. Se ra nes “ is ace, «oe Ce - = < j oo a ie j —- ho An o£. ee Brees 2 - . se a ee Oe teres Sn Ss alee oem Ome - ‘ a Pa, ; oe : oh ~¥ : ae : era ee ae 4 2 : ” es a3 e ond al + =i La a * = P = Sie my ‘ ~ is im , swe : at : eee: (a1 aye am ‘n ‘ : * a a ey way, — ch : ioe . e oe z cs ‘ : ae ia « = a es a ; 3 tg : are . =e = E sai A ‘ . . “oe F ok Be _ } : Pe = j ‘ 3 alten - sc balip aes a = cae me Gas = 4 . oe as “ —. a wml —— . moet ie Us Rees =. : r~ 7 < 7 Hae "x == » Mes, > * id “J ‘\. e t in 4 * 2 ; « > f ; 15. a . Se S| : : : aes ¥ bE, . ean ; se" A ; ¥ 7 ae ¢ Fr ae. = . £ ? ie fi Ne 4 cay : 4 = ts : oe << ee iy ~p eny “+ - wo Se ie" - ~ fre - aa ey Beat ‘ ¢ = ; - e ie poe 3 a ‘- ve —s =i rc + = Z ‘i ; te? ; : u } : ; : : FAR Ode } 3 - 5 = ; ‘ é } ‘ om 7% . rete J < t : : 2 w aes Meroe wr oe xe 4 LE oy 3 - 7 x ~ ~ Rey errs ren TLN fone eo ) )) = a a lJ a = uJ E 1200 1600 2000 2400 0400 0800 1200 LOCAL TIME FIGURE 17 DIURNAL CYCLE OF TEMPERATURE IN WEST-CENTRAL LAGOON AT SURFACE (o—)AND AT BOTTOM(e-—-) wy 9) Dy 6) WIND SPEED (KNOTS) 0 0.5 1.0 1.5 20 MAXIMUM TEMPERATURE DIFFERENCE(C) -SURFACE TO BOTTOM FIGURE [8 EFFECT OF WiND ON DIURNAL TEMPERATURE GRADIENTS ss RESTRICTED 13 (c) The volume of flow was obtained by multiplying velocities and cross-sectional areas. ‘wo correction factors were intro- duced for flow in the passes. The measurements were inade in the middle of the passes where tlow is at a maxinwi, by oceanographic theory the average flow in the pass at any given moment should be about three-fourths or the tmaxinium flow, and the figures are corrected accordingiy. In the second place, the rate of flow in the passes is not constant throughout ebb or flood, ‘she change in velocity with tinue is roughiy a sine curve, and the average rate of flow can be approximated by umltiplying the flow at mid- ebb and mid-flow by the factor 2/m. ‘able 3 shows the results of the calculations. Table 3. Calculated flow into and out of the lagoon S “Ss : x ei eee Velocity cm/sec.* , Volune (cnx 109/sec.) | | Ph i | ce a6 | aT ee stage Tide stage ° ° Bie ae vials | F H E Iie F Nan-Yuro Soh wr | fe | 30 Shs 2.70 0. BS. | 3.02 | |Amen=-Bikini 9-8 staag [50 ie eee iBikini-En 15 eb oF = | i 20) 30 Boby=West 8.5 |riac ) = On| 20 6.3! =2 5 5 5 jWest-Boro Channels [Rh reveals a FLO. | =o | 5 i =2.5 |-3.0 |-2.51-2.0 - -10 ° 5) ie) | oye) e =) ] 4ay The net exchange of water is determined from the duration of each stage of the tide and the volume of flow during that time, as shown in Table 4.2 = Positive Ee ee STL Ws BESET ees Ours + puko ji Fis a a aoe deep. pass “with Seiler reers on each ag) side. ‘The type of flow is difterent in the pass from / that over the reefs, and they are therefore listed separately. ©, se RESTRICTED lk Table 4. Net transvort (x 10)ein3) Tide stage | High | Ebb Low Flood | _Duration (hours) Oe AAS) O. o2 ' Reefs Ole °f5 0.93 Pye Channels ' 22560 0.12 | -2.1 -0.12] -0.71 Passes 4.629 -0.11 | -6.97| -0.1 Sum -0,0 = 20 ° 286 Net inflow 7.86 Net outflow 7.92 During a mean tide of 100 cm., the value used in all ots lations, the change in the volume of the lagoon is 6.4 x 10 em? or 2.3% of its total volume. ‘his value is about 20% lower than the estimate obtained in Table 5. ‘The difference, however, is not large enough to affect the essential validity of the results, The tables show that although the net transport is outward at high, ebb, and low tides, some water is brought in over the northern and eastern reefs. moreover, some water is lost through Enyu Channel at flood tide. ‘he total transport of water across the periphery of the lagoon is therefore larger than the net tidal transport. A budget of total transport can be obtained by calculating the sum of all positive values and the sum of all negative values as shown in ‘fable 5. ns wable 5. Total transport (x1014em3) Inf low The total transport into ae ouf of the lagoon is therefore estimated to be about 10.6 x 10°-* om’ per 12 hours, or 3.8% of ay. “ Z ‘tienen i i Be tke i sal her r és eM. Re een ta 9 hy verter Hee Fuchs Sh dasi i cuts Oy vr Np ke RESTRICTED the total lagoon volume. ‘he reservation is made, however, that these values may be as much as 20% too high. A final check on the validity of the data, particularly with respect to the relative proportion of water that comes in over the reefs as compared with that derived from the channels, can be obtained by calculating the salt budget. It has been ob- served that water entering the lagoon over the reefs has a salinity of about 34.80°/o0, along the southern passes and channels it is about 3/4.50°9/o0, and the average salinity of water flowing out during the ebb is 34.63-/oo. Combining these observations with estimated volumes of flow, the results are as shown in able 6. The error is about 1%, Table 6. Salt budget Channels, passes Outf Low Thus the total transport into and out of the lagoon is es- tablished with a fair degree of accuracy. There remains only the question of the dispersal of water inside the lagoon. ‘Yaking the figures from Table 6, the water coming in over the reef with a salinity of 34.80°/00 constitutes 35% of the total inflow, while 65% is channel water with a salinity of-34.50°/o00. If these proportions were mixed completely, .the resultant salinity of the lagoon would be about 34.61°/00 Instead, the average salinity of the lagoon is nearly 34.80 /oo. ‘he difference is too large to be accounted for by evaporation. ‘The most logical explanation of the apparent discrepancy is that rapid and relative- ly complete mixing occurs only in the immediate vicinity of the passes and channels, and part of the water entering this area on the flood tide is lost on the ebb without becoming incorporated in the main mass of central lagoon water. Again utilizing observed salinity values for purposes of calculation, the water in the passes on the ebb tide, with an average salinity of 34.63°/o0, : has a percentage composition of 43% central lagoon water (salinity 34.80) and 57% ocean water (34.50) which has been in the lagoon only a short time. ‘he results of this estimate are summarized as follows: (a) It is assumed that the Poe inflow and outflow are equal and have a value of 10.6 x 10-‘cm/’ per 12 hours. naan adn erage . i ) i, RESTRICTED 16 (b) 57% of the total outflow or 6.0 x 10!4em? is of recent oceanic origin by way of the southern passes and channels.’ it has been in the lagoon only during one or a few tidal cycles and has not had time to become thoroughly :uixed with the water in thg central part of the lagoon. ‘I'he remaining 43% or 4.6 x 1014 em’ is lagoon water. (c) The total inflow can ta allocated as follows: (1) 35% or 3.71 x 10**em? per 12 hours cores in over the northern reefs and joins the main mass of lagoon water, (2) 65% or 6.85 x 101 4em? comes in through the southern passes and channels, of which 6.0 is transient according to (b) above, and the reaining 0.25 is transported into the central part of the lagoon, (ad) Therefore of the total interchange of 10.6 x 10)4en3 per 12 hours, 4.6 x 10+4em3, or 1.6% of the total volwie of the lagoon, perform a slow flushing of the lagoon as a whole, while the remainder rapidly flushes a small area in the south and southwestern part of the lagoon, The estimates of total inflow and outflow, based on a ean tide of 100 cm., are adequate for deter:iining the average rate of flushing over a considerable period of time. wvuring shorter periods the rate will vary from about 50% of the calculated val- ues (neap tide) to 160% (spring tide). Still larger variations are obtained in the relative amounts of water passing over different parts of the periphery of the lagoon. 3.4 Vertical diffusion In section 3.22 it was shown that vertical mixing is effect- ive in maintaining a relatively uniform tesuperature in the la- goon. Diurnal heating in the absence of wind would increase the temperature about 2°C at the surface during the day. ‘fhe increase at the bottom would be less than 0.019. Actually, however, the surface increase was never more than about 0.2°. ‘he rest of the heat was transferred downward by vertical diffusion, and the temperature change at lower levels was correspondingly increased. The rate of vertical transfer is readily determined for any par- ticular temperature distribution: the constant in the equation, known in oceanographic literature as the coefticient of eddy diffusivity and designated by the symbol A,, can then be used to determine the rate of transfer of any property of the water with any assumed initial distribution. ‘his method is essential in determining the rate of dilution of contaminated water after the blasts. RESTRICT2D 17 There are two inherent difficulties in applying these methods in bikini lagoon. First, the observed difrerences are small, so that the errors of measureiient in any particular set of observa- tions may be as much as 40% of the total difference. Second, it was not possible to. sample the same body of water at successive intervals, which introduces ranaoim oceanographic variations. whe errors are to a considerable extent elininatea by smooth- ing the curves and by using several indevendent methods of corz- putation. ‘hese methods were: (a) The relationship between wind velocity and surface cur- rent gives values of eddy viscosity, which in uniform water should be numerically equal to Aggy « (b) The difrusion coefiicient was computed froi the diurnal temperature variations previously aescribed. (c) At noon, and especially during low tide, the water flow- ing in over the eastern and northern reefs is appreciably warmed during its passage, its salinity is raised, and it becouwes rich in oxygen. ihe mixing of this characteristic water trou a "line source" with the rest of the water in the lagoon provides another method of computing verticai ditfusion which is of particular interest because of its analogy with the surface contauination expected in test Able. Method (a) and method (b),for the eastern station gave val- ues of around 200 to 250 cm* per second. liethod (b) for the western station gave values much higher than are reasonable, indi- cating that processes other than vertical diffusion were active. whe result is believed to be due to sinking of surface water, which effectively brings the surface temperature fluctuations to greater depths by other means than turbulence. ‘he sinking is the result of the gradual slowing dovm of water which is being driven against the western reefs, and seems to be distributed over a large part of the western lagoon. Method (c) leads to somewhat smaller values of eddy diffu- sivity near the reefs. ‘his may be partly due to upwelling, partly to the fact that the "line source" is located near the surface, where the scale of turbulence must be suppressed by the existing boundary. However, theoretical considerations indicate that Ay will increase to a value of about 250 at a depth of 2.5 m. and probably changes little if any from that depth down to very near the botton. These figures will be used in the section that follows to determine the rate of dilution of radioactive products by ver- tical diffusion. até ae = Me o 18 4. DECONTALINATION “STLuATo The factors exclusive of radioactive cecay that ust be considered in calcuiating the rate of deconta:iination after the blasts are horizontal and vertical diffusion and current direction and velocity. Since all these factors are consid- erably affected by weather, it is impossible to make a precise prediction that will fit all cases. But barring a radical change such as reversal of wind direction, the results should be of the right order of magnitude. ‘the present section is based on the assumption of a 10 knot uNE wind at the time of the tests. A further section will attempt to describe what would be likely to happen with certain other wind conditions, Assune that the explosion produces a volune of uniformly contaminated water with a radius of about 400 m As horizon-= tal diffusion begins to operate, the size of the patch of con- taminated water will increase, a gradient in concentration will develop from the center of the patch toward its periphery, and the concentration in the center will decrease gradually. An oceanographic theory developed by G. F. iicEwen predicts that the effect of diffusion will be as shown in Table 7. Yable 7. Recuction of contamination by horizontal diffusion (% of initial concentration, redioactive decay not included) Distence in meters from the center 1600 Further dilution will take place by vertical diffusion, and this effect can be determined from the measurements described in the previous section. At the time of test Able, the contamina- tion will be largely confined to the immediate surface layer. Assuming for purposes of calculation that at the end of three minutes the radioactive products will have become uniformly distributed through the upper 2.5 m., then further dilution of this surface layer is expected to take place according to the figures in Table 8. The patch of contaminated water will be carried WSW from the target area at a rate of about 0.3 knot. ‘he contaminated water that is diffused down into the bottom layer will be car- ried back toward the target area. This removal of contaminated RESTRICTED 19 Table 8. Reduction of surface contamination by vertical diffusion in test Able (% of concentration 3 minutes after blast) water by the counter current is important in maintaining a high and relatively uniform rate of reduction of concentration in the surface layer, a reduction that is estimated to continue at a rate of about 25% to 30% per hour after the first four hours. In test Baker it is assumed that the contamination is initially distributed in a cylinder of 400 m. radius extending from the surface to the bottom. ‘the part of the cylinder in the surface current, namely the upper one-fourth, will move in a WSW direction at a speed of 0.3 knot. The lower three-fourths will move ENE with the bottom current at 0.1 knot. As the patch of contaminated surface water moves away from the target area, vertical mixing with uncontaminated water underneath will reduce the surface concentration about 25% per hour. ‘he material lost from the surface layer will be carried back toward the target area by the counter current. Reduction of concentration will also occur in the patch of contaminated bottom water moving eastward from the target area, but the rate of reduction will be only about one-third as high, or 8% per hour, because of the greater thickness of the bottom layer. From this figure, the amount of contaminant moving into the surface layer at any time is readily determined. Since a part of the radioactive material diffuses out of the patches of contaminated water and into the opposite current moving back past the target area, the net result of the current system and vertical diffusion will be to produce a long, narrow strip of contaminated water passing through the target area along a WSW-ENE axis. ‘The strip will be gradually broadened by horizontal diffusion, but this effect is of relatively minor importance. Table 9 shows the estimated reduction in concentra- tion of contaminant by horizontal and vertical diffusion. At the end of the first day the eastern end of the strip of contaminated water is expected to reach Bikini Island. The 20 Table 9. Calculated maxinum concentration of radio- active materials in the surface water after test Baker (% of initial concentration, radioactive decay not included) Downwind from target: Distance travelled miles Horizontal diftusion Vertical diffusion Combined effect Upwind from target: bistance travelled miles Horizontal diffusion vertical diffusion Combined effect analysis cannot be carried beyond this point with any degree of accuracy since both the mathematical theory and the oceanographic factors involved are very complex. Considerable upwelling of bottom water occurs near Bikini island. Since this water will have a higher concentration of radioactive material than the surface water, the area between the target and Bikini is more likely to be dangerous than any other part of the lagoon. ‘he maximum concentration in this area probably will not be more than 13.5% of the initial value. ‘'his figure is based on verti- cal diffusion alone, since horizontal diffusion is expected to be much reduced in the bottom layer. ‘there is considerable likelihood that upwelling will occur over a broad enough area to dilute the contaminant considerably; however, it is better to be conservative and consider that there may be patches of water with 10% or more of the original concentration of con- taminant. At the end of two days most of the bottom water at the eastern end of the lagoon will have upwelled to the surface and will have been diluted by vertical diffusion as it moves westward with the surface current. The rate of diffusion will be ‘decreased, since it will be mixing with bottom water that is already slightly contaminated. However, rough estimates indi- cate that none of the water in the target area will contain as much as 1% of the original concentration of contaminant, although higher concentrations may persist near the reef for another day. the other end of the strip will have reached the south- western passes at.the end of the second day, and henceforth - a small amount of radioactive material will be discharged from the lagoon on each ebb tide. RESTRICTED Pal The nature of the diftusion process is such that the rates will decline rapidly after the first two days except insofar as random variations in current direction carry patches of con- taninated water into uncontaminated areas, leading to rapid dilution. If the radioactive material were spread uniformly through the southern half of the lagoon, the concentration would be reduced to 0.1% of the initial value, but it seems un- likely that this could occur in less than one to two weeks, It is clear that the natural processes of current flow and vertical diffusion act together to maintain a gradient in concentration with the greatest amount in or uear the target area, and the gradient will be destroyed only very slowly by horizontal dif- fusion and random variations in currents. After the first few weeks, further dilution will take place only by tidal interchange. Since the major path of contaminated water lies south of the area or most complete stagnation, it is expected that the radioactive materials will be removed at least as rapidly as the average lagoon flushing rate of about 3% per day. This will require two and a half months to reduce the con- centration by a factor of 10. An unpreaictable but probably small quantity of radioactive material will become attached to bottom sediments and sinking organic matter, from which it will be liberated gradually over a long period of time and will be a minor source of contamina- tion, particularly in the eastern part of the lagoon where up- welling of bottom water is most pronounced. it is not expected to be of any practical significance at the surface but might be hazardous to diving operations. RESTRICTED 22 5.EFFECT OF WEATHER ON LAGOON CIRCULATION AND DECONTAIIINATION The quantitative estimate of the decontamination rate presented in the previous section required certain basic assump- tions about the weather, varticularly as regards wind direction and speed. ‘These factors are of the greatest importance in de- termining the circulation of the lagoon in general and the dissi- pation of radioactive products in particular. ‘he calculation was based on what appeared likely to be the prevailing conditions at the time of the blast, namely an ENE wind with a speed of about 10 knots. Different conditions would require modification of the predictions, and over a limited range of variations the modifications can be made accurately. ‘the effect on the lagoon of winds between 10 and 20 knots is well known. ‘the curves can be extrapolated to 5 knots with no great error. ‘hese are sin- ply questions as to rates of diffusion and water transport. "he curves previously shown in Figure 8 are a fair indica- tion of the effect of wind on current transport. With winds of 5 knots, the rate of flow of the surface current would be reduced to 0.1 to 0.2 knot, and it would require three or four days for the contamination to spread the full length of the lagoon. whHori- zontal and vertical diffusion would also be reduced. The east- west gradient in radioactive products would be less pronounced. Upwelling of bottom water in the eastern end of the lagoon would be likely to produce patches of water with high concentrations of contaminant for three days or more. With a 20 knot wind on the other hand, the rate of flow would be increased to 0.5 knot, and the contaminant would spread across the lagoon in about one and a half days, but the east- west gradient would be stronger and more persistant. Since the available weather data indicate a decrease in wind velocity during the summer and a shift in the average di- rection toward the east or southeast, it is barely possible that the test might come at a time when these changes are extreme, namely no wind or a southerly wind. ‘there has been no opportun- ity to determine what would happen in such cases. Any predic- tions are largely speculatory, but a few general comments can be made. It is believed to require twelve to twenty-four hours for a wind driven current to be generated or for it to cease when the wind stops. ‘Yherefore in a prolonged period of calm weather the rotary circulation of the lagoon would soon be destroyed. {here would remain only the slow movements generated by tidal interchange. Vertical and horizontal diffusion would be greatly reduced. It seems likely therefore that a large concentration of He <5 oe a iy eT : Be i cod PPE 3 Rt ge ine ‘ 5 ab ex nes ee Be rater 4 cargrit egesed {ity eantule ance Lie POR 2 Sagi ode edt % ahi ae ¥ i Beis oO 3 FHOD* pe fa Sw Seles 3 : Devote fide Puce aul i; wee Se ben (RAN Re aa nee. | pol RESTRICTED contaminant would reiain in the target area and the latter would be unsafe for re-entry until wind currents were again generated. A southerly or southeasterly wind would change the direc- tion of the lagoon circulation but otherwise would not change the previous description of the spread of radioactive products. It seems possible although by no means certain that the rate or flushing of the lagoon would be drastically altered if southerly winds persisted long enough to alter the direction of the oceanic current outside. The most likely guess is that water would then flow in constantly along the full length of Enyu Channel at a rate of about 10% of the lagoon volume per day and would be flushed out by tidal interchange across the northern and western reefs and through the southwestern passes. This would rapidly clear the target area and triple the observed rate of flushing. Veering winds during or after the tests would alter the dir- ection of flow of the currents so that the path of contaminated water would no longer be a straight line across the lagoon and back. This would increase the horizontal spread of radioactive INMIaterials, and the rate of vertical diffusion would be maintained at the initial high level for a longer period of time. The net result probably would be that the rate of dilution during the first few hours or the first day would not be greatly altered but that subsequent dilution would be greatly accelerated. Sa [Aare i eee ge Srer ea Ocean ie yee am Bo Ga ne fe oF i. eet oy RESTRICTED 2k 6. APPENDIX ON OCEANOGRAPHIC METHODS 6.1 Current measurements Measurements of surface currents were obtained by the cur- rent pole method, ‘he poles were 12 to 16 feet long. Some of them were four inches square; others were made of smaller strips of wood with aluminum fins inserted to increase the cross-section- al area. They were weighted so as to hang vertically with about one foot exposed above the surface of the water. A light alun- inua staff was rigged to the upper end of each vole, bearing some device to aid in sighting the pole. Various methods used at one time or another or in combination were pennants, life jacket dye marker (bags of fluorescein, which left a trail of green dye in the water), lights, and radar reflectors. ‘Three or four poles were used simultaneously. They were set out one to two miles apart, and their position was determined every few hours by con- ing alongside and taking bearings on beacons or other landmarks. At night, and at considerable distances from land, radar ranges were used for obtaining fixes. Along the reefs and in the channels, the currents were ‘studied by dropping dye bombs from an airplane and photogre ph- ing the dye patches at frequent intervals over a period of 15 to 30 minutes. It was largely the surface currents that were measured by this method, but packages of dye lashed to the bombs left a trail in the water as the bombs sank, permitting some con- clusions as to subsurface currents. ‘The dye bomb method proved to be particularly useful on the reefs and in any circumstance in which a high degree of variability required a large number of nearly simultaneous observations. Vertical profiles of current velocity were obtained with a Von Arx current meter. It consisted of a propellor mounted in a tube oriented to the current by means of a vane on one end. Each turn of the propellor induced a small electrical potential that was used to determine the number of revolutions, from which the current velocity was computed. ‘Yhe current direction was determined by observation of the vane through a water glass. The current was measured at depth intervals or 5 to 20 feet fron the surface to bottom. The direction could be determined to a depth of about 150 feet, the limit of visibility in the lagoon waters. An underwater floodlight was used for night stations. It was necessary to exercise considerable care in the current meter work since swinging ot the ship at anchor introduced an error. Dye marker and the cable angle were used to determine the times when valid measurements could be made. noe. RESTRICTED 25 6.2 Measurement of temperature and chemical constituents The distribution of temperature and chemical constituents has provided information on diffusion rates, lateral mixing, and currents. Reversing thermometers and Nansen bottles were used to measure the temperature and collect a water sample at any desired depth. ‘The Nansen bottle is a tubular instrument with a valve at each end. The bottie is lowered with the valves open so that the water passes ireeiy through the bottle. A "messenger" sent down the wire trips a mechanism wnich releases the upper end of the bottle so that it turns upside down, clos- ing the valves and entrapping a water sample, The thermometer mounted on the side of the Nansen bottle is designed so that turning it upside down breaks the thread of mercury, and the thermometer records the temperature at the tine of reversal. The salinity of the water samples was determined by the Knudsen method based on a chloride titration with silver nitrate. The Winkler method was used for oxygen analyses, and the Atkins- Deniges method for phosphate. 6.3 Tides and swell Tides were measured by two standard tide gauges. Une was located in shoal water in the eastern part of the lagoon near Bikini Island, the other on the outer reef. Portable 24-hour gauges were also used occasionally in various places. swell was computed from an instrument which measured short period pressure fluctuations on the bottom in shoal water. ite beat