USVI Public Records

Document text

AD-750 693 THE DEPTH PROFILE OF AMBIENT NOISE IN THE DEEP SEA NORTH OF ST. CROIX, VIRGIN ISLANDS R. J. Urick, et al Naval Ordnance Laboratory White Oak, Maryland 14 A ugust 1972 DISTRIBUTED BY: National Technical Information Service U. S. DEPARTMENT OF COMMERCE 5285 Port Royal Road, Springfield Va. 22151 NOLT9 '72-17 6 THE DEPTH PROFILE OF AMBIENT NOISE IN THE DEEP SEA NORTH OF ST. CROIX, ViRG "N 'SLANDS I By R. J. Urick G. R. Lund T.J. Tulko D D, 14 AUGUST 1972 94VMLC ORNANCE LABORATORY, WHITE OAK, SILVER SPRING, MARYLAND k, TA• QTf, I N\'CI AL AiJ '. \", I( ;N .f'J :[ ",I APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIMITED UNCLASSIFIED Si •Set Aiml Cla,i ,illcation DOCUMENT CONTROL DATA- R & D S' i -os,.0 "I titc, hod; of ibtract and Indexing annotniin ni be entered when the overall report is classified) I OtmGINA TING AC ' V TV (Cofporate atuthor) JZ. REPORT SECURITY CLASSIFICATION COMMANDERI UNCLASSIFIED NAVAL ORDNANCE LABORATORY 2b. GROUP WHITE OAK, SILVER SPRING, MARYLAND 3 RE RON tTI f THE DEPTH PROFILE OF AMBIENT NOISE IN THE DEEP SEA NORTH OF ST. CROIX, I'IRGIN ISLANDS. A OES•C 'T 1'.VE NOTES (T'ype of report and incluvie datet, NOLTR 72-176 14 August 1972 S AU 7 HORISP (First name, middle initial, last name) R.J. URICK G.R. LUND T.J. TULKO 6 REPORT OATL 7a. TOTAL NO Or PAGES Tb. NO. OF REFS 14 August 1972 __10 0 8i CON TRAC. Oil GRAN T NO 9i. ORIGINATOR'S REPORT NUMBERIS) b PRoLcro A370/370A/WFII-121-707 NOLTR 72-176 Prob. No. 202 C. 9b. OTHER REPORT NOIS) (Any other numbers that may be aselgned thli report) Al. is OISTNIOUT#ON STATE.MENT Approved for public release; distribution unlitgited iI SUPPLEMENTARY NOTrS 112. SPONSORING MILITARY ACTIVITY Naval Air Systems Command I Washington, D.C. 20360 13 AUSTRACI The deep sea appears to be quieter at mid-depth than at greater or lesser (depths. This was found in a series of measurements of the ambient noise background at a location north of St. Croix, V.I., using hydrophones suspended from a cable at intervals of 2000 feet down to 8100 feet. The noise profiles show a sharp decrease of noise from the surface down to 6100 ft and a reversal of the profile below. The measurement site ts an active, dynamic one, and shows a large variability in noise level from hour to hour. "Details of Illustrations in this document may be better studied on microfiche D DFORM (PAGE N , I NOV 473 UNCLASSIFIED S/N 0101. 807-6801 Security classification r UNCLASSITEEND Security Ciassi•ication 4 4 LINK A LINK U LINK C I•EY WOROS - I"OLE WT ROLE WT ROLE WT UNDERWATER SOUND AMBIENT NOISE PROPAGATION ACOUSTICS CARIBBEAN .4, S• - . .I l- I[ Il - -" DD f ORJ 1473 (BACK) UNCLASSIFIED ( NOVAGE 2 (PAGE' 2) ',. u ~.u* , a, 6d. 8 ••i |eL k'oI NOML 72-176 THE DEPTH PROFILE OF AMBIENT NOISE IN THE DEEP SEA NORTH OF ST. CROIX, VIRGIN ISLANDS Prepared by: R.J. Urick, G.R. Lund and T.J. Tulko ABSTRACT: The deep sea appears to be quieter at mid-depth than at greater or lesser depths. This was found in a series of measurements of the ambient noise background at a location north of St. Croix, V.I., using hydrophones suspended from a cable at intervals of 2000 ft down to 8100 ft. The noise profiles show a sharp decrease of noise from the surface down to 6100 ft and a reversal of the profile below. The measurement site is an active, dynamic one, and shows a large variability in noise level from hour to hour. Acoustics Division Physics Research Deparcm..n, NAVAL ORDNANCE LABORATORY White Oak, Silver Spring, Maryland 20910 iG NOLTR •'72-176 NOLTR 72-176 14 August 1972 THE DEPTH PROFILE 6F AMBIENT NOISL IN THE DEEP SEA NORTH OF ST. CROIX, VIRGIN ISLANDS This report describes how the natural underwater noise background at a deep water location varies with depth. It will be of interest to anyone contemplating the use of deep sound sources and receivers in sonar applications. The work was done as Project MARLIN for the Naval Air Systems Command tinder Task No. A370-370A/WFl-121-707, Problem 202 ",OBERT WILLIAMSON II Captain, USN Commander Z.I. SLAWSKY By direction ii NOLTR 72-176 TABLE OF CONTENTS PAGE INTRODUCTION 1--------------------------- MEASUREMENT METHOD 2--------------------------- THIRD-OCTAVE SPECTRA - 3 STATISTICAL DATA - 3 DEPTH PROFILES - , 3 LEVELS AT THE 100-FT DEPTH - 4 SHIP SPOTTING 5--------------------------- VARIABILITY 5 COMPARISOM WITH A COMPUTER MODEL -------------------------------- 6 SUMMARY--- ----------------------------------------------------- 6 ACKNOWLEDGEMENTS - ---------------------------------------------- 7 REFERENCES - 8 iUL i UNCL.ASSIFIED I NOLTR 72-176 THE DEPTH PROFILE OF AMBIENT NOISE IN THE DEEP SEA NORTH OF ST. CROIX, VIRGIN ISLANDS INTRODUCTION 1. The ambient noise of the sea has been measured and reported for many of the world's ocean areas. A relatively abundant literature is extant, and the noise level to be exp,.cted under given conditions of wind, shipping and biological activit, is well defined (1). Yet, nearly all of this work has involved measurements made relatively near the sea surface. At depths in the body of the sea, the noise has not been greatly studied, partly because of experimental diffi- culty and partly because of a lack of practical application to which the information could be related. 2. The earliest attempt to determine the variation of noise with depth was made - as one of many "firsts" - by the Hudson Laboratories of Columbia University, using a slowly-sinking recording package aptly called the "Diving Duck" (2). Later, the bathyscaphe TRIESTE was employed for the purpose (3), and, more recently, the depth ,ariation of noise was reported by Perrone (4) from -measurements made with different bottom-mounted hydrophones at various depths in an area south of Bermuda. 3. Within the last few years an awakening of interest in this hitherto neglected subject has been brought about by the possibility of using deep propagation paths in long-range passive sonars. Recently, a number of novel instrumental approaches have been taken, and the results are beginning to appear in the literature. 4. However, the data obtained in these recent and current field studies is discordant, and no clear picture of the effect of depth on the noise background has been obtained. One reason for this lack of agreement is the difficulty of sampling a variable noise environ- ment. For if the noise samples at different depths are not obtained simultaneously, many samples are required to smooth out the constantly occurring variations of the background. Another problem is how to avoid, or allow for, changes in hydrophone calibration with depth and temperature. In short, the avera&e depth profile of the noise back- ground of the deep sea is difficult to determine, and requires special experimental measurement techniques. !:1 NOLTR 72-176 5. We have avoided the difficulties just mentioned by a method which permits taking data simultaneously at different depths, along with an on-the-spot calibration of the various hydrophones by means of a standard sound source located near the surface. The results show a distinct depth variation of the ambient background at the active, dynamic location where the observations were made. MEASUREMENT METHOD 6. A string of five hydrophones at depths of 100, 2100, 4100, 6100 and 8100 ft was used for recording the noise background at regular intervals over a period of several days. Fig. 1 shows the hydrophone string called MARLIN (MultLple Array Line) suspended from an anchored instrumentation barge maintained and operated for the Naval Air Systems Command by TRACOR/MAS, Inc. The barge vas located just north of the island of St. Croix, V.I., at the location of the cross in Fig. 2 on the side of a basin some 10 miles wide and 60 miles long. This area is one of frequent commercial ship traffic traveling between the Panama Canal and European ports. It is also popular with sports fishermen operating out of the near-by islands. Sources of ship noise are numerous and are relatively close-by the measurement site, and therefore may be expected to be highly variable in their contri- butions to the noise environment. Although a total length of 12,000 ft of cable was available, it could not all be deployed because of the limited water depth and an electrical fault discovered in the upper portion of the cable. 7. A frequency-modulated multiplexing-demultiplexing system was employed to transmit the output of the five hydrophones up the common coaxial cable. Each hydrophone modulated its own carrier frequency between 120 and 480 kHz; demodulation took place just before analog recording on a Precision Instrument 7-channel FM magnetic tape recorder. Noise samples 1 1/2 minutes long were recorded at hourly intervals during two 2 1/2 day periods separated by an interval of a few days (for battery charging) during March, 1972. A total of 114 hourly samples were recorded. 8t Calibration was accomplished by transmitting to the hydrophones the CW output of a standard Navy J-9 sound source placed at a depth of 50 ft. The response of tnis projector was checked at the NOL Brighton Dam calibration facility. Tones of known intensity were tape-recorded over each hydrophone channel and were used to convert the recorded data to acoustic levels. 9. Wind speeds during the data-taking period ranged from 0 to 17 knots. A sound velocity profile was not obtained. However, a histori- cal sound velocity profile for the same month of the year at a location a few miles away is given in Fig. 3; a 900-ft bathythermograt taken during the data period showed a 400 foot layer abole a thermocline. 2 NOLTR 72-176 10. Analysis of the field tapes was made with a General Radio Model 1921 Real Time Analyzer yielding one-third octave band levels. Four one-third octave bands (50, 160, 500, 3150 Hz) were selected for playback of the hourly recordings. The over-all analysis band exteacd- ed from 12 to 8,000 Hz; the upper frequency limit was imposed bv the tape speed (30 ips) used in recording. T;!IRD-OCTAVE SPECTRA 11. Fig. 4 shows three analy~es, over the above frequency range, of individual noise samples taken under different conditions of wind and shipping. The upper group of spectra (Fig. 4a), one for each of the five hydrophones depths, was obcained for a single 4-second noise sample recorded during an exceptionally quiet period of rearly calm wind and inaudible ship noise. Two additional curves are superimposed on the measured spectra: The solid curve shows the Knudsen-Wenz spc.,trum (5) for conditions of light shipping and a 4 to 6 knot wind: the dashed curve is the system electronic noise level recorded when one of the hydrophones was replaced by an equi.- valent capacitor. The other two groups of spectra (Figs. 4b, 4c) were obtained on samples taken under other combinations of ship traffic and wind speed. 12. We observe from these spectra that, except for the 100 ft P hydrophone depth, the measured levels agree well with the Knudsen- Wenz levels for the same noise conditions. Alsc, the measured levels are well above system noise, except for the 8100 foot hydrophone at high and low frequencLes nder quiet conditions. The noise level apparently falls off oith depth at all frequencies; Lhe deepest (8100 ft) hydrophone is appreciably quieter than the shallowest (100 ft). STATISTICAL DATA 13. Fig. 5 gives the cumulative distribution of the levels of the 114 hourly samples in the four one-third octave bands centered at 50, 160, 500, and 3150 1z. These plots show the spread of the 1 1/2- minute average levels at each hydrophone. The standard deviation (o) of the samples, based on straight lines drawn by eye through the points, ranges from 3.8 to 7.6 db, with art average i of 6.0 d1b. A sinmilar statistical spread was found by Wenz (6) in the levels of Pacific Ocean ambient noise over a t.o-year period. 14. The levels at the 50Z !,oints on these figures are the median noise levels equalled or exceeded by half of the noise samples. For three hydrophones depths, these2 median levels are compared with the Knudser.-Wenz curves in Fig. 6. '4e observe that, again with the exception of the 100 foot (eoth, the measured median levels agree with the curves for the conditions of moderate-to-heavy shipping and a 9-knot wind. The median wind speed dur!-ig tho two 2 1/2 day data per ods ias 7 knots. NOLTR 72-176 DEPTH PROFILES 15. 1he measured median levels are plotted against depth in Fig. 7. These noise profiles show a rapid decrease of noise with depth from 100 to 2100 ft, followed by a slower decrease down to 6100 ft and a suggestion of a reversal below. The noise profile is roughly the same for all four frequency bands. This result is contradictory to the findings of Perrone (4) in an area south of Bermuda, where the noise at four hydrophones located on the bottom in different water depths down to 2500 fathoms was found to decrease with depth at high frequencies and to increase with depth at low frequencies. However, in the pres-nt data the similarity of the depth profile at all freauencies is both reasonable and expectable, since the noise sources at bith high and low frequencies, whether due to shipping or the wind, are iumerous and are located near the sea surface. LEVELS AT THE 100-FT DEPTh 16. The profiles show levels at 100 feet to be some 5 to 10 db higher than those at greater depths. These high levels at 100 ft will at once be suspected to be the result of proximity to the measurement barge, where a generator was running to provide 60 cycle a.c. power and where wave slap and human activity aboard the barge might be susi :ted to overwhelm the ambient sea background only 100 ft away. In, -- d, narrow-band sound spectrograms of the noise picked up by the S... _oot hydrophones did show the presence of line components at -3, 90, and 120 Hz originating in the barge power supply. 17. However, various pieces of evidence indicate that the recorded noise at this depth is in large part valid ambient sea noise in nearly all third-octave bands. First, the levels of the power supply line components just mentioned were not high enough to affect the third-octave band levels, except under very quiet noisa conditions. Second, the levels are high in all bands, from the lowest to the highest; a barge-proximity effect would be expected to occur princi- pally at low-frequencies. Third, distant shipping (faint screw beats) could be heard much better on the 100-foot hydrophone than on the deeper hydrophones. Fourth, a computed noise-depth profil.- (Fig. 16) shows a strong near-surface decrease of noise. Finally, |,igher noise levels at a shallow depth were observcd 6 months earlier at the saile location with another experimental arrangement: two hydrophones at depths of 55 and 305 ft floated 2000 ft away from the barge (7). Typical one-third octave spectra under these conditions are shown in Fig. 8. A difference of 5 db or more between the two depths appears to exist at Crequencies above about 800 liz. This frequency is about equal to the lowest frequency trapped in the mixed-layer duct 120 ft thick that existed during the earlier data period; the corresponding frequency for the 400-ft duct existing for the present MARLIN data is 120 !1,,. 18. From all of these bits of evidence it appears likely treat, for frequencies above 100 lHz at least, the high noise levels at 100 ft are those of the natural awbient background at the measurement site. 4 NOLTR 72-176 One )ossible cause of high noise levels at shallow depths is pro- pagation in the surface duct of the noise from distan. sources. Still and all, no such strong layer effect appears to have been reported in the past. Indeed, it is contrary to the notion that at frequencies above a few hundred cycles, the background of surface noise originates in the vicinity of the receiving hydrophone, rather than at ranges great enough for ducting to be effective. Further investigation is required. SHIP SPOTTING 19. A Navy aircraft, together with radar services of the Atlantic Fleet Weapons Range, enabled ship spotting to be done on two consecu- tive days. During the exercise, the locations of major ships in an area around the measuremenc site, together with the type, speed, and course of eac"., ship, was determined. The results of this exercise are shown in Fig. 9. Fig. 10 shows the one-third octave spectra of noise samples recorded at the times the shipping was identified. It will be noted that higher noise levels at low frequencies occurred on the second day (open circles) than on the first (closed circles), in keep- ig with the larger number of ships observed on the second day. VARIABILITY 20. It has already been noted that the 1 1/2 minute noise samples are variable in level from hour-to-hour, and have a fluctuation expressed by a standard deviation of about 6 db. Since in Fig. 5 the data points fall on straight lines, the levels in db are roughly normally distributed. 21. This variability is demonstrated in Fig. 11. This is a series of 1 1/2 minute playouts recorded hourly during an overnight period between 1.530 hours local time and 1730 hours the next morning. The levels at all hydrophones in the one-third octave band at 160 Hz are seen on close inspection to change in the same way from hour to hour, though not by the same amount. It follows that the depth profile of t.he noise level must be variable, and so change its shape from houc to hour. 22. This variability in the depth profile is illustrated by the consecutive hourly profiles at two frequencies plotted Fig. 12. Here we see that although the general shape of the profile is pre- served fro r hour to hour to hour, there are cons!.derable variations in detail; for example, at 3150 liz, where the noise tends to be dominated by the wind, the difference in level at 0630 hours between the 100 ft and 8100 ft hydrophones is 20 db, but is only 6 db an hour later. Similar, though smaller, differences occur regularly throughout Sthe sequence of profiles. In other words, wiile the . evel variations of the hourly samples tend to be correlated bet\'een hydrophones, the correlation is not perfect. 23. B•y contrast, shott-term, short-period variations do not correlate at all. This is shown by Fig. 13, where, over a total record length 5 NOLTR 72-176 of 19 seconds, we observe noise transient! on individual hydrophones that do not appear on others. In short, long period changes tend to occur at all depths in a generally similar way, while short period changes do not. 24. At low freque:-.cies, changes in the location andtype of ship pattern are doubtless the cause of much of the variability of the background. This is illustrated by the sound spectrograms of Fig. 14 corresponding to the same series of hourly samples as Fig. 11. Here we note the variability in the line component structure of the low frequency noise, both from hour-to-hour at a single hydrophone, and in the relative strength of the lines from hydrophone to hydrophone at any one time. Although the changing distribution of shipping produces changes in the pattern of the tonal structure from hour to hour, any one instant of time interference effects produced by prop- agation cause differences to appear in the tonal components from hydrophone to hydrophone. 25. At t1'e higher frequencies, changes in wind speed are a cause of noise variability. This is illustrated by Fig. 14, where the levels of the hourly samples are plotted against the wind speed at the tii,.a the sample was taken. it is evident that there is an absence of dependence on wind speed, along with considerable scatter of the data points, at 50 and 160 H1z; on the other hand, there is a strong dependence on wind speed, with little scatter, at 3150 Hz. This behavior with frequency is caused by the increasing dominance of wind noise with incre- tng frequency, and has been repeatedly observed in the past, in both deep (1), (8), and shallow (9) waLer. COMPARISON WITH A COMPUTER MODEL 26. An ambient noise computer model has been developed by the Naval Air Development Center (10). This model assumes an infinite uniform distribution of equal noise sources over the sea surface, and sums up the contributions of sources in different range increments by the use of a ray trace intensity program based on a given velocity Sprofile. 27. Fig. 16 is a comparison of the computed noise profile, using the k velocity profile at the right, with the median profile observed at 3150 Hz (Fig. 7). There is reasonable agreement between model and data; the principal features of the noise profile occur similarly in the model and the observations. SUMMARY 28. We ray summarize the results of this field study of thu noise background throughout the water coluimi at the St. Croix site by the following staLt'ments: 1. It is feasible to observe the awrbient noise b,.ckground at a depth by means of hydrophones suspended from an armored cable pro- vided vibration isolation is used between hydrophones and cable. 6 NOLTR 72-176 2. At St. Croix, the sea becomes quieter with increasing depth down to about 6000 ft, with a suggestion of a reversal between 6100 and 8100 ft. 3. The noise background is variable over hourly intervals. One-minute noise samples have a standard deviation of about 6 db. 4. Although long-period changes in level tend to occur at all depths in a similar way, short period changes do not, and remain uncorrelated between depths 2000 ft apart. 5. The noise near the surface (100 ft) is higher at all frequencies than at greater depths. This effect appears to be inherent in the noise background at the measurenent site, but is difficult to explain. 6. A computer model assuming a uniform distribution of ship sources over an ocean of constant depth gives a depth profile reason- ably close to that observe.7. ACKNOW LEDGEMENTS 29. The authors wish to acknowledge their indebtedness to the personnel of TRACOR/MAS , Inc. for their fine assistance in all phases of the field work, and to W .A. Fisher of the Atlantic Fleet Weapons Range for arranging aircraft and range services during the ship spotting exercise. 7 NOLTR 72-176 REFERENCES I. G.M. Wenz, Acoustic Ambient Noise in the Ocean: Spectra and Sources, J. Acoust. Soc. Am, 34, 1936, 1962. 2. A. Berman and A.J. Saur, Ambient Noise as a Function of Depth J. Acoust. Soc. Am., 32, 915, 1960 (A). 3. M. Lomask and R. FrasetLo, Acoustic Measurementq in Deep Water Using the Bathyscaph, J. Acoust. Soc. Am., 32, 1028, 1960. 4. A.J. Perrone, Ambient Noise Spectrum Levels is a Function of Water Depth, J. Acoust. Soc. Am., 48, 362, 19Y'0. 5. R.J. Urick, Principles of Underwater Sound for Engineers, McGraw-Hill, New York, 1967, Fig. 7.5. 6. G.M. Wenz, Low Frequency Deep Water Ambient Noise Southwest of Pt. Sur, California, NEL Report 1960, 1967, Unpublished. 7. R.J. Urick, The Noise Signature of an Aircraft in Level Flight over a Hydrophone in the Sea, J. Acoust. Soc. Am. in publication. 8. A.J. Perrone, Deep Ocean Ambient Noise Spectra in the Northwest Atlantic, J. Acoust. Soc. Am. 46, 762, 1969. 9. R.J. Urick, The Underwater Acoustic Environment at Two Contrasting Shallow Water Locations, NOL report (unpublished). 10. J. Keene, Some Causes and Properties of Ambient Noise with Applications to Airborne ASW, NAOC-72013-AE, 1972, Unpublished. 8 NOLTR 72-176 ANCHORED BARGE 6100 FT 2100 FT 4100 FT 6100 FT S8100 FT BOTTOM 9900 FT FIG. 1 MARLIN HYDROPHONE STRING .1k NOLUi 72-176 24t 22 ~ S 2 24 .- 2r2,, 4n...,..2....2.4,-2 - . -I -~I 2.-n.222.4 2-4 ~4 " &2 2- 2.i f 4 2 r; .i .2 7- 00 ' .0V) -/~ I; * I , 2E* .00 2 C4 1 *' 0 2~~ 0 22* 1- 10.: 7n 0 2 IIA '2* ~ 2 4,2 2 k.0 0O * C . 4 2~~~"ce * 22- 4 *~ r ~ .-.. *~ . 2 * 2 2 2'~ , 22 2 02 2*. .~'II - ~. ~* .:. A4. ~5. 2 224 ('v0 1.2 U., #0 . NOLTR 72-176 "-- lO0FT 2- 4 OBSERVED 6-- 2100 FT 8 10 12 • ~4100 FT 14 . DEPTH (HUNDREDS 16 OF METERS) \ EXTRAPOLATED 18 -_ 6100 FT \ 0.017 SEC- 1 20- 22 24 -- 8100 FT 26- 28- 30 -- BOTTOM I I I I I I 1480 1490 1500 1510 1520 1530 1540 VELOCITY, METERS/SEC. FIG. 3 VELOCITY PROFILE, NODC REF. 31-302 AT CONSES. STA. 460, PROVIDED BY D.F. FENNER AND W. RANDLETT, NAVOCEANO. EXTRAPOLATED BELOW 1637 METERS AT THE RATE OF 0.017 SEC-]. LAT 170 55' N, LONG 640 31'W, 12 MARCH, YEAR UNKNOWN. 4i- ýig NOLTR 72-176 Rý0425 -28 MAR '72 o 100 F~l -I -x 2100) / *a--ýo_ + 1(9 -2 a 0.0 -. 6100 SPECc olM . a'0-o -0--o' A -1 0- D -70' .1q0 01?0 27 MAR '72 -10- 00 -JrL - +,A$0 0 '0 ~-0 CM 2 H, -0 _0% -40 4t;f __________________________ SHIPPING ~ ~ ~ ~ ~ 10 (c(ID1cKOS)OEAT HPIG -I- ........ NOLTR 72-176 0 b ~z0 4 ' 0 00 0 b 0 4Z ::1 000 o 00 .4 - ~ L _00 17, ,* ]j V). Z~4 4'~ D , oa -4 L 0 - I 0 0.. Z SbA -004 .. 01 b. > Z3 .0~~ XxXt ~U, 0 00 L)-,)e ' U> 0 ::EI(.) b -- 0 LL. NOLTR 72-176 0.. 0 CL( LU. 10 (N 0 1'- 0-U- 0 _ w 2j 00 LU ( 0> -, 0 (A U <LU 0n LU n (N LUZ C + LUZ (ALU o L -ý <0 <LU . C14 I -j LU L LUj 04 L 3-u 4,,0 NOLTR 72-176 r0 2000 1 4000 DEPTH (FEET) 6000 8000 3150 Hz 500 Hz 160 Hz 50 Hz -60 -50 -40 -30 -20 -10 0 SPECTRUM LEVEL (db re 1 DYNE/CM2 FIG. 7 MEDIAN NOISE PROFILES, USING THE 50% LEVELS OF FIG. 5 NOLTR 72-176 $ -20 5FTLAYER DEPTH 120 FT -30 THEORETICAL 305 FT LOWEST FREQUENCY SPECTRUM TRAPPED IN A 120 FT DUCT LEVELI dBRE 2 I DYNE/CM2 -40 55 FT •. ••::: • •HYDROPHONE -50 HDOHN 100 200 5,0 1000 2000 5000 FREQUENCY (Hz) FIG. 8 NOISE SPECTRA OBSERVED Al DEPTHS OF 55 AND 305 FEET SIX MONTHS EARLIER AT THE SAME LOCATION. 'S, NOLTR 72-176 LI-. U.. Lo <. tftn C14* NOLTR 72-176 U- - U- U- 0 0- 0 0 100 III 0 w co OWLLJCNZ 0 L- I- z 0 Z (Ni Ln 0 U.i oi 00 ((n lyo NC)LTR 72-176 -®rr v h.- -- 0 44 * -<7 T -, 4T 77 IT L- N 14 ~ ~ ~ 1 i- . 42 _____ LU -Zi- a-j 'p II FF-f LU - rl' -7*- -4 4, U'o NOLTR 72-176 -15 103 -15 -12 -14 -18 -21 -21 --7' -7* -9* -14'* -17 -14 -14 2100 •a ~2100 4100 "DEPTH, FT 6100 8100 . 1730 0030 72 O 1 0730 20 MAR ,72 21 MAR ,72 Hz 21 MAR ,72 SSHIP SMALL CRAFT -42 -43 -46 -44 -46 -47 -48 -48 -47' -38' -45' -36' -46 -46 -45 S,~~~oo.o,, 2100 - DEPTH, FT S~~~~6100 o f i I I Ie 8100 1730 0030 0o 0730 20 MAR ,72 21 MAR ,72 21 MAR ,72 FIG. 12 CONSECUTIVE HOURLY PROFILES AT TWO FREQUENCIES, SHOWING THE VARIABILITY Of THE PROFILE FROM HOUR-TO-HOUR NOLTR 72-176 cr .. T1 I .I- 1$ r *>t- 4 .;KT 'ii; 1 1/4 itg: +H +t -~~ -. 0 0 r ~ :::x ~ ~ ' L LU -t 0, ,*. ,'+,'-i*&2 0 * :,'![ 1 ' '7' ''.4 A~' 0. u1l02 4- ,I~tfz 4, * ' ' ,L4 k , - U-U- C: 'U U-~ u L ý11 C) U-I.-cU NOLTR 72-176 5-~ 0 ~ -~ -v .1 0 -~ r ,.. , -- uj M i. . - A,." 0- .1 -j - . 0 0 ~c C* .27 - IC)t j 'n 44 0 Clj V);-*~U 41 L o o00000 C C Q0 ('0 C it ' ) e' ' ~ 'C tt~(jN OO'0~'~C)0~('I") ..- '. 0 LL Rleproue 0 rom best :, eale b IoIpY •.,4 NOLTR 72-176 0 -10 U -I00 0 00 0 0 -20 - 8-?--2 8 0 0 50 Hz 0 •0 0 0~ o m 0 SPECTRUM 0 0 0 00 _ LEVEL -30 -- 160-Hz (db re I DYNE/CM2 ) t x _x ý,x x 0 x 3150 Hz -50 0 9 , I I I IllI I jI........,I . 2 1 4 5 6 7 8 9 10 15 20 25 WIND SPEED (KNOTS) FIG. 15 NOISE LEVEL !N FOUR BANDS PLOTTED AGAINST WIND SPEED FOR A HYDRCDHONE DEPTH OF 2100 FEET. NOLTR 72-176 $0 / I- -2 0 COMPUTED -~ / /1 (NADC) -3 COMPUTER PROFILE -4 0 1 0 OBS ERVE D -5 I3150 H z -6 ; OBSERVED PROFILE -7 DEPTH -8 (KILOFEET) _9-9 -10 C,' -11 -12 -13 -14 -15 -16- SI I I I I -12 -8 -4 0 4900 4950 5000 5050 LEVEL (db ARB. REF.) FT/SEC AMBIENT NOISE SOUND VELOCITY FIG. 16 COMPARISON OF THE OBSERVED MEDIAN PROFILE AT 3150 Hz WITH A PROFILE COMPUTED BY THE NAVAL AIR DEVELOPMENT CENTER. AT THE RIGHT IS SHOWN THE VELOCITY PROFILE USED IN THE COMPUTER PROGRAM ALONG WITH THE PROFILE OF FIG. 3. THE WATER DEPTH ASSUMED IN THE MODEL WAS 16,500 FEET. COMPUTED AND OBSERVED PROFILES ARE MATCHED AT THE SURFACE.