FOR OFFICIAL USE ONLY Report No. 4330568 wn 07 DOCUMENT } ASW SONAR COLLECTION J TECHNOLOGY REPORT THE MEASUREMENT OF SOUND SPEED IN THE SEA: VELOCIMETERS AND OTHER DEVICES ARTHUR D. LITTLE, INC. CAMBRIDGE, MASSACHUSETTS DEPARTMENT OF THE NAVY NAVAL SHIP SYSTEMS COMMAND NOOO24-68-C- 1137 ai Task 08648 ae MAY 1968 (150! iy dl | (U0: H33BO FOR OFFICIAL USE ONLY ==! CHAO A ig 1OHM/ Taw IN FOR OFFICIAL USE ONLY Report No. 4330568 ASW SONAR TECHNOLOGY REPORT THE MEASUREMENT OF SOUND SPEED IN THE SEA: VELOCIMETERS AND OTHER DEVICES by H. F. EDEN E. M. DUCHANE P. Von THUNA R. Van HAAGEN ARTHUR D. LITTLE, INC. CAMBRIDGE, MASSACHUSETTS DEPARTMENT OF THE NAVY NAVAL SHIP SYSTEMS COMMAND NOOO24-68-C- 1137 Task 08648 MAY 1968 FOR OFFICIAL USE ONLY AI 3HT We agage auvoe _epivia MAHTO. qua ea : vd Bis ae oa ar BATT) e SUNITA @TVIZUHOARZAM iain YVAU FHT IO THA GUAMMOD emaTeve _ XRt.D-6a-AS000K Re ie Bae YAM) WL. IIl. IV. TABLE OF CONTENTS SUMMARY THE MEASUREMENT OF SOUND SPEED IN THE OCEAN: REQUIREMENTS FOR THE NAVY A. Introduction B Requirements for Tactical Situations & Prediction C. Requirements for Oceanographic Survey D. Problem Areas 1. The Velocity of Sound in Ocean Patches 2. Interfaces 3. The Velocity in the Sea Bottom 4 Velocity Dispersion in Sea Water and the Pressure Coefficient of Velocity REVIEW OF INDIRECT METHODS OF VELOCITY MEASUREMENT A. Introduction B Thermistor Chain C Reversing Thermometer and Bathythermograph D. Expendable Bathythermograph 13 S.T.D. Instrument REVIEW OF VELOCIMETERS A. Introduction B. Tabular Listing of Velocimeters C. Discussion of Direct Methods COMPARISON OF METHODS AND RECOMMENDATIONS A. Comparison of Direct and Indirect Methods B. Problems and Recommendations 1. Depth Accuracy 2 Velocity Accuracy 3. Expendable Velocimeters 4. Possibilities of Hybrid Systems for Maximum Cost-Effectiveness and Recommendations CONCLUSIONS REFERENCES APPENDIX A - Ray-plots for Various Velocity Profiles ii Arthur D Hittle Ine. in a) i J vit ib a aaa Lite feed i Tose hG oe eon! A-3 TABLE II-1 ITI-1 II1I-2 IV-1 LV=2 LIST OF FIGURES AND TABLES Sound Velocity Profiles from Mechanical Bathythermograph Data and Nansen Bottle Casts Initially Positive Velocity Profile and Ray Plot Initially Isovelocity Profile and Ray Plot Initially Negative Velocity Profile and Ray Plot Ray Plot for Convergence Zone Propagation Ray Plot for Convergence Zone Propagation Comparison of Temperature Measuring Systems Comparison of Velocimeters Expendable Velocimeter Prototypes Comparison of Devices Navy Requirements for Velocity Data iii DL 29 33, 34 41 42 Arthur A AUittle, Inc. Wao ee See he yy ge SUMMARY Data on the speed of sound in the sea is a prime requirement in Navy operations. Only relative accuracy in the measurements of sound speed as a function of depth is needed in tactical situations because it is the shape of the velocity profile that is of major importance to hull mounted sonar. Absolute accuracy in the measurement of sound speed is desirable, however, in order to exploit longer range propagation in the deep sound channel. The sound speed can be measured either indirectly, by computation from temperature and salinity or density data, or directly, by using "velocimeters". In tactical situations, temperature data alone has sufficed for the indirect method except under certain near shore conditions, where skilled guesswork has been required. Present velocimeters based on the ''sing-around" principle are adequate for oceanographic survey purposes, from the points of view of accuracy and depth capability. However, their connecting cables render them unsuitable for tactical use; in general, the manufacturers of velocimeters are aware of and have been attempting solutions to the problems that arise from this application. The value of temperature data alone for defining the presence of a surface duct in shallow water where sharp salinity changes can occur should be studied. We believe that there exists a case for the develop- ment of an expendable instrument or of a synoptic technique for tactical use to supplement the now successful expendable bathythermograph. Arthur A Aittle Inc. I. THE MEASUREMENT OF SOUND SPEED IN THE OCEAN: REQUIREMENTS FOR THE NAVY A. Introduction Knowledge of the local sound velocity field in the ocean is one of the limiting factors in naval operations. Generally, this know- ledge involves the surface velocity (v) and the variation of the velocity with depth &. Various factors such as required accuracy, available time, convenience, and system cost can dictate the choice of measuring instruments used. All systems measure sound speed rather than the vector velocity. The speed of sound in sea water is affected by the temperature, the salinity, and the pressure (which depends on the depth) and increases with increase of any of these factors. The composite effect can be well represented by the empirical equation. v= 1449 + 4°06 — 0.055t- + (139 - 0.012t)(s - 35) + 0.017d (1) where v is the velocity in meters per second t is the water temperature in °C s is the salinity expressed in parts per thousand: d is the depth below the surface in meters This equation is a simplification of more complicated equations obtained by Wilson (Reference 1) and others as a "best fit" to experimentally measured data. Equation (1) is accurate to about one meter per second for those conditions commonly occurring in the various oceans. The Arthur A. ALittle, Inc. ty Wee ; os Wis iii Te yea) ft. Ne ik if sound speed in surface sea water at O0°C and with salinity of 35 parts per thousand is thus 1449 meters per second as contrasted with 1403 meters per second for fresh water at the same temperature and pressure. Equation (1) shows that change in temperature has the greatest effect on v. In the upper 10 meters of the ocean considerable tempera- ture variations may exist. Below 100 meters in deep water the temperature generally decreases more regularly. Often in the upper 100 meters a surface layer of water that is almost isothermal (the surface duct) overlies a thermocline of rapidly decreasing temperature. Areas where variations in salinity become important are: near the mouths of larger rivers or in estuaries and deltas, where appreciable amounts of fresh water run into the sea; in the vicinity of large ocean currents; in shallow water or water close to the surface (where rain and evaporation have greatest effect); and near fields of melting ice. Other serious possibilities about which little is known may occur on a smaller scale due to thermohaline convection or "salt fingers". Variation of the velocity with depth caused by the increasing " pressure is quite regular, i.e., 0.017 meters per second increase per meter increase in depth. This effect is masked by temperature effects in the upper region of the ocean except in the case of an isothermal surface layer. However, at depths greater than about 1000 meters, the temperature is very nearly constant (nearly the temperature of maximum density), and the velocity then increases at the pressure effect rate. Thermal anomalies may exist at abyssal depths and in any water column, from a few meters above the bottom to the bottom. For sonar operations a knowledge of the sound speed profile Arthur D.Little Ine. u] ae at least in the vertical direction is necessary for determining the path of the transmitted sound signal. The data should be sufficiently accurate to define “x . For sophisticated ray tracing techniques, knowledge Z of higher derivatives may be useful. More sophistication might require knowledge of horizontal variations in v due to ocean patchiness as well as measurements of the sound speed structure in the ocean bottom. The needs for sound speed data may be divided into "tactical" and "survey.'' Their requirements overlap in certain cases, e.g., the propagation of sound in deep sound channels, and where bottom—-bounce is utilized. The required absolute accuracy in the measurement of v(z) varies. For those active sonars where the upper portion of the water column is used, knowledge of the shape of the profile v(z) is sufficient, and only relative accuracy is required. For range estimation based on propagation in the deep sound channel absolute accuracy in the knowledge of v(z) is desirable. dv - The indirect dz There are two methods of determining v(z) and method consists of measuring the variation of temperature in the vertical direction and computing the sonic velocity from equations such as Equation (1). Generally, the salinity term is taken from reference tables but alternatively, the salinity is measured simultaneously either in situ or from collected water samples. The temperature profile is measured with bathythermographs, thermistor chains or reversing thermometers. The direct method uses sound velocimeters which measure the local sound speed at the instrument by transmitting an acoustic signal across a path determined by the instrument size which is generally of the order of a few centimeters. Because of this short path length the Arthur 7. ALittle, Inc. acoustic signal is usually of ultrasonic frequency, perhaps several megacycles. Combinations of these systems are sometimes used in experiments for comparison of the data measured by each. S.V.T.P. systems (e.g., Reference 2) measure salinity, sound speed velocity, temperature and pressure; and "birdcage" systems have been used consisting of Nansen bottles and velocimeters in a protective enclosure. B. Requirements for Tactical Situations and Prediction The problem of using active sonar to best advantage involves analysis of the ocean environment in order to decide the optimum operation modes and probable detection ranges. The shape of the velocity profile is of prime importance and so the requirement is principally for relative accuracy. A detailed discussion of accuracy requirements for range pre- diction is given in Reference 3. Several kinds of information are typically available to a sonar officer. Echo sounders give bottom depth and variability. In general, successful bottom bounce operation is not possible when the bottom slope exceeds a few degrees. Elementary geometry suffices to predict approxi- mate detection ranges for various beam depressions. Thus if On = sonar beam depression angle R = range z = depth of the bottom then R = 2z tan oe 5) Arthur OD AUittle, Inc. gaan : 4, pi aan on Av a 2 : ‘ 7 In order to predict the sound field for hull-mounted sonar operation, the sound speed must be known as a function of time and space. The speed of sound increases with temperature at roughly 3.0 meters per second per °C, with salinity at 1.3 meters per second per mille and with depth at approximately 1.7 meters per second per 100 meters. In general, the speed of sound in sea water varies from approximately 1460 to 1530 meters per second. A sonar officer generally has temperature data, (i.e., tempera- ture vs. depth data) for the upper part of the ocean. It is usual to assume that the speed of sound varies only in the vertical direction. Salinity measurements are not usually made in fleet operation, but charts of "average surface salinity" as a function of latitude and longitude are readily available to him. Pressure versus depth charts are also available. Usually pressure is assumed to be uniquely related to depth, although differences in latitude alone may introduce sound speed errors as high as one meter per second. The data are usually furnished in ft-lb-sec units. In order to synthesize velocity profiles then, a sonar operator depends on temperature data, using the empirical relationships previously mentioned which give the sound speed as a function of temperature, pressure and salinity. The Woods Hole Oceanographic Atlas of the Atlantic Ocean (Reference 4) contains both bathythermograph data and independent tempera- ture and salinity measurements made by sampling techniques for identical parts of the ocean. Some examples are shown in Figure 1. Nansen bottles were used to obtain samples at depths from which salinities were determined; reversing thermometers gave the temperatures. The estimated experimental errors involved in sampling were: temperature + 0.01°C depth + 5 meters, salinity + 0.01 parts per thousand. The mechanical bathythermograph used had Arthur DUittle Ine. iy h i i i i i i iy i aM it AY oF By rally wg Depth (Meters) Depth (Meters) Depth (Meters) 61 122 183 ATLANTIC OCEAN Station 233 BT 104 Crawford 16 43° 03'N 45° 39'w Bottom 4450 m —0O— BT Data —e— Sampled Data 1477 1483 1489 1495 1501 1507 1513 1519 1525 1531 61 122 183 244 Calculated Velocity (Meters per Second) Station 333 BT 104 Crawford 17 17°30’ N 64°25’ W Bottom 3360 m —o— BT Data —+*— Sampled Data 1516 1519 1522 1525 1528 1531 1534 1537 1540 1543 61 122 183 244 Calculated Velocity (Meters per Second) Station 394 BT 45 Crawford 17 29°22'N 65° 28.8' W Bottom 3405 m Crawford from Haiti to Columbia —0o— BT Data —e— Sampled Data 1516 1519 1522 1525 1528 1531 1534 1537 1540 1543 Calculated Velocity (Meters per Second) FIGURE 1 COMPARISON OF SOUND VELOCITY PROFILES DETERMINED FROM MECHANICAL BATHYTHERMOGRAPH DATA AND REVERSING THERMOMETER, NANSEN BOTTLE CASTS (Reference 4) i ea Am il aT i an error of approximately + 0.2°C. It is immediately clear (Figure 1) that these profiles may be very complex. A few scattered points are not sufficient to give an accurate picture of the profile. The shape of the profiles obtained with a small error in temperature measurement or depth location may be indeter- minate. The absolute values of velocity may differ by only a fraction of a percent but the sign of the velocity gradient in the surface duct may differ. When the velocity profile is known, much can be easily predicted. The application of Snell's law allows one to trace sound rays through the ocean represented by the profile. Useful ranges and so-called ''shadow determines the zones" may be distinguished. The velocity gradient, x, dv : mae ag GN aeute positive, the rays curve upward; if —— is survature of rays. If ae negative, the rays curve downward. Since ranges and acoustic travel times for active sonar are relatively short, the prediction of target range does not require high accuracy in the absolute value of v(z), provided that the dv path taken by the sound is known approximately; the sign of aS Gisele; whether the gradient is positive or negative) is most important to deter- mine the probable path. In Appendix A, Figures A-1, A-2, and A.3 are rays plots produced by a computer which show the bending for positive, negative, and zero curvature conditions. Simple graphical techniques have been worked out and prepared for use by sonar operators. In order for convergence zone operation to be possible, what is known as a "'velocity excess" must exist. Velocity excess is the dif- ference between the velocity at the bottom of the ocean and the velocity at the bottom of the layer depth. A rule of thumb is that the velocity Arthur A. HLittle Inc. nt ta i DUS OTT l i PL, + Here i ; i My 5 ‘in Pa) SEA a ars sigs aN Mh L Yi ah LIC Tena MD} tee aad : beanie i Liat La i [thug Aas ny nh may a F wk ma wi pei i) Air Hie excess should be greater than or equal to 10 meters per second for success- ful convergence zone operation. This insures enough rays getting to the surface for a zone strong enough to be potentially useful. (A velocity excess identically equal to zero would allow the bottom grazing ray to pass; if slightly greater, the first convergence zone ray would pass.) Whether this condition exists or not is simply determined from the velocity profile. Figures A.4 and A.5 in Appendix A are ray plots illustrating convergence zone conditions. The SOFAR path is the propagation path which employs the deep sound channel whose axis is normally located at depths of 1000 to 1500 meters. Propagation conditions are excellent and signals can travel dis- tances of the order of a thousand kilometers with only cylindrical spreading loss. Although SOFAR propagation is a multipath phenomenon with different paths arriving at different times on a given signal, it has an abrupt ending corresponding to a the time to travel in a great circle along the channel axis where Ma is He velocity at the channel axis. The time of travel along longer paths is less than = since the fo) sound travels at a speed greater than Vise Because of this abrupt ending, accurate timing is possible, and if the time of arrival of a signal is measured at more than one location (say 3), the geographical position of the source can be determined in a manner similar to LORAN. The deep sound channel is characterized by a velocity profile which shows first a decrease in velocity with increasing depth below the surface or surface layer and then a more gradual increase in velocity with increasing depth. The depth of the minimum velocity occurs near 1300 meters in the mid-latitudes of the North Atlantic. Data from Arthur A Aittle, Inc. Reference 5 show remarkable spatial and temporal persistence of the channel in the mid-Atlantic. In a survey lasting twelve days and extending over some 1000 kilometers, the velocity minimum changed by less than 2 meters per second and remained at practically constant depth. At all the measurement stations the velocity below the minimum increased at the pressure rate. However, a different study (Reference 6) carried out south of Bermuda showed considerable fine structure in the sound channel region there, with sound velocity changes of about one meter per second in a few hours at the same depth, raising the question whether a single sound velocity profile is sufficient to describe the deep channel in a large area for any length of time. Probably extreme temporal variations in the deep sound channel are due to fronts and may be of a local and predictable nature. More data on SOFAR studies in the South Atlantic are given in Reference 7. Reference 8 discusses long range sound propagation in the Southern Ocean or Antarctic Ocean (Project Neptune). The Antarctic Ocean appears to have a different velocity structure from the Atlantic or Pacific. The depth of the SOFAR channel decreases rapidly between 40°S and 60°S and no true SOFAR channel exists further south. To make use of this propagation there is a requirement for absolute rather than relative accuracy since for accurate location an exact value of Ve is required instead of a knowledge of the velocity profile which suffices for predicting the ray paths in the channel. Possible errors in SOFAR location are similar to errors in location by LORAN due to timing errors. These are discussed in detail in Reference 9. The location of the source is determined from the intersections of hyperbolae which describe positions having fixed differences in distance from the listening stations. 10 Arthur D.HLittle Inc. Habe len ay, Vive ee es | Me iahast sees) ek ae er i ae hoe Hanis viet ey ite May od Pai Tui ‘isa uso, sob | bib was | a untintad ih Ney a sein ne at He & 1 vi } on : - ‘i Prien Cea ofchiy ahha Hi hla! ete Apa Aa Gece A An error in travel time or knowledge of Va causes the points of inter- section of the hyperbolae to become areas. The size of the area depends on the bearing angles i> bo of the listening stations. Let us consider a simple example. The travel time in the sound channel for an acoustic pulse over a distance of 3000 km to each of two stations at 90° to each other would be approximately 2 x 10° seconds. If there is an error in the value of Ve of + 2 meters per second, then the source of sound may lie anywhere in an area of 64 square km. If the bearing between the stations is much less than 90°, this area can become a long, thin band so that rapid location of a moving source is difficult. The persistence of the sound channel indicates that accurate measurements in the channel can possibly define the channel over considerable distances for periods of time as long as a day. Probably a single profile is not sufficient to fully define the channel over a large area, however. For optimum ranging an accurate value of Nis is required. It is worth noting in our example that + 2 meters per second corresponds to a temperature error no greater than about + 1°F, or in considering the pressure correction to a depth error of 110 meters at a depth of 1500 meters coupled with a possible temperature error of + 0.2°F. From the point of view of current tech- nology we should note that velocimeters are quoted by the manufacturers to be accurate to 0.15 meters per second. For field use we might safely double this. The temperature error in XBT measurements is + 0.2°F and the error in depth location (from relatively shallow tests) is currently quoted as 2 - 3% of depth implying an error in Ve of + 0.7 meters (1) per second at 1500 meters. Obvious questions with regard to tactical situations are: 1. Would present systems benefit if either deeper data or data with higher absolute accuracy were available. (1) See Conclusions - Section V. 11 Arthur J Hittle, Inc. i ; WW Pai if Mi at VO Mite Voit a a eh iin ‘f f Vi i i Baran hs Patt dia ale g t) ere LSet nie aanllbl Wy ¥ i vanes Sits Mise i ie: {" Piva’ ied in KS i ; 2. Will future systems benefit from or require such data? The first question should be considered from the point of view of both deep and shallow water conditions. In some shallow water condi- tions such as deltas where there are strong salinity changes and the possibility of fresh water overlying sea water, BT data may be inadequate to predict sound velocity. Here a need for reasonably accurate data implies direct measurement of sound speed. It is possible that there may be horizontal changes in velocity in shallow water and that ranging will be in error. Nevertheless, the shape and variability of the vertical profile is required to determine operating mode. In particular in tactical situations a knowledge of the existence of a surface duct is the most important requirement. In deep water, the knowledge of velocity to greater depth would be of value now if it indicated the occurrence of a convergence zone which current sonars could exploit. Generally we can state that the accuracy required for current active sonar operation is rather in the profile shape than in absolute value of the velocity. The accuracy to which the profile should be known depends entirely on the sophistication which is employed for shipboard prediction in tactical situations. The exception to this requirement for only relative accuracy is in the use of the deep sound channel for passive location. C. Requirements for Oceanographic Survey There is considerable value to the Navy in surveys of large areas of the oceans. The ideal from the point of view of naval operations is long term mapping of velocity profiles and surface duct depths and deep sound channels as regards occurrence and short-time and seasonal persistence. aU. Arthur DUittle Inc. Neue i a, Lape * dig : a Py uepe: i ee ia es , ry . Bi 4 ih, 7 4 an , i i : in ie ee qi i oh ive os | ie j hone 7 oat ne an i ry Sat i cae . ier) | | ibs iW Tay Cha ear ate f TS MATRA MISUSE Fath TNS iN ela Din ini | x ee athatnis Pub Pie bay the ov dee iy " be awr nll wie Ven Some work on persistence of surface channels has been carried out from BI measurements (Reference 10) and several surveys of the deep sound channel have been referenced. The main requirement for such surveys is that the instruments used are accurate and have a good depth capability, to at least 5000 meters. This accuracy must be good in both the relative sense, to ensure that “x is known, and the absolute sense. In such survey work much valuable information could be obtained on various problems such as horizontal variations, interfaces, internal waves and the sea bottom. For surveys, temperature and salinity should be measured as well as sound velocity. SVIP systems are commonly used. D. Problem Areas 1. The velocity of Sound in Ocean Patches It is well known that while the temperature variations in the vertical direction tend to exhibit a layer structure, the horizontal variations, due to the turbulence in the upper layer, are more in the nature of random patches. The patch size spectrum ranges from diameters of the order of the layer depth (i.e., about 50 meters) to small diameters which correspond to the dissipation of the thermal structure by molecular heat conduction. Details of the temperature variations are given in References 10, 11; generally a dominant patch diameter of the order of twice the measurement depth is noted until the thermocline depth is reached. The temperature fluctuations are generally less than 1°C but may be sufficiently large to be of the order of 1 - 2°C. The spectrum of patch size apart from the largest might be expected to follow the well known Kolmogorov distribution law for turbulent eddies. 13 Arthur OD. ALittle Inc. Tear ead, v3 HSER hot 1 WEY © Veen PEI ILIT cao THONTL tn Tn { it Cana ee ai 1 Gea The investigation of these patches has always been with the thermal transducers, although several attempts have been made to correlate this data with observed fluctuations in acoustic transmissions due to scattering by the patches (e.g., References 11, 12). The question arises as to whether a knowledge of the distribution of sound speed in these patches would be of value and whether current velocimeters are suitable instruments to investigate them. The patches cause a transmitted acoustic signal to fluctuate in intensity by + 6 db due to scattering. However, the distribution and size of the patches are functions of both space and time. Thus, it seems unlikely that any finer data would be of value in a tactical situation. For scientific or survey purposes, two problems arise in measuring the sound velocity in patches. These are 1. Most currently used instruments measure during vertical transects. How could they be adapted to measure the horizontal changes? 2. How can the spatial and temporal behavior of the patches be distinguished and recorded. It seems quite feasible to adapt current instruments to such studies by towing at fixed depths. One variation would be to use a velocimeter together with an isotherm follower system and measure changes in velocity along an isotherm. The ship velocity would necessarily be quite low. Another possibility would be to use free-swimming vehicles such as the University of Washington Applied Physics Laboratory Self Propelled Underwater Research Vehicles, already equipped to perform this task. The time response of the instruments appears perfectly adequate for all likely spatial or temporal rates of change of temperature or salinity. 14 Arthur D Little Inc. ' atl iW . vin A) A : pe r on i rai 1 ns hin Similar remarks might be expected to apply to the case of velocity changes caused by internal waves. There has been considerable interest in this subject(e.g. Reference 13). The space-time aspects can only be separated by a stationary experiment holding velocimeter arrays at various fixed depths, an expensive experiment. 2. Interfaces Anomalous effects may occur at both the sea water-sea bottom interface, which depends on bottom type; and at the water-air interface, which depends on sea state. Sound reflected at anything beyond normal incidence is affected in both cases. a. Bottom Interface Sometimes an increase in temperature of deep water close to the ocean floor has been observed. This is attributed to the geothermal flux through the floor, a heat flux which amounts to some 1x TON cals per ema per second for the Atlantic (e.g. Reference 14). The depths concerned are approximately 4000 meters (see e.g., Gerard et al. Reference 15). Such an increase in temperature would cause an increase in sound velocity. Descents with a deep submersible should indicate any such effect. Any small changes in velocity at the lower interface would only affect grazing angle of the sound travelling in a deep channel. Although this would not be a tactical consideration to surface ship sonars, present instruments usually have sufficient depth capability to investigate this effect where it occurs. b. Air-water Interface The behavior of the upper interface is tactically more interesting because sound transmission in the duct aan ILS) Arthur D.Little, Inc. ii i nee mr uit! Hi ii " Bs 7 et | a aaa re i ied iH eae eth Hae feet Prete tit ig Ay en. i Aye " ee se nen ine a ot yi : : ual involve reflection at the surface at grazing incidence and because the sound might travel long distances near the surface, in an obviously disturbed layer. Measurements of the depth of this layer as a function of sea state would be interesting from the point of view of sound velocity. In this case it seems unlikely that temperature or temperature and salinity measurements would be at all meaningful by themselves, the effects of surface geometry and entrained air being so much stronger. 3. Velocity in the Sea-bottom we A knowledge of the density and the velocity in sea-bottom sediments in specific locations would be of interest from the points of view of sonar reflectivity, seismic reflection and refractive profiling. Reflectivity is currently measured by large pulsed sources or explosives. The velocity of sound in sediments is a function of several variables and is strongly dependent on any trapped gas. The dependence of sound speed on pressure in marine sediments is unknown. Measurements at ultrasonic frequencies may be of little value to the Navy if dispersion frequencies occur, since extrapolations to low frequencies might allow gross errors. Many of these factors could be determined by laboratory experiments. However, velocity measurements in areas known to have bottoms which are suitable for sonar operation would be of both short and long term value. An instrument for making velocity measurements in situ in ocean sediments (Reference 16) has been under development by Hudson Laboratories and the Benthos Company. It is a combination of a sound velocimeter and a free-falling boomerang corer. The transmission frequency is 0.5 MHZ; the path length is 15 cm. The data is recorded on magnetic tape in the buoyant section which is separated from the free-fall vehicle by a timer. 16 Arthur DUittle Inc. Se ayia The buoyant section containing the electronics withdraws the transducers and a core sample and returns to the surface. 4, Velocity Dispersion in Sea Water and the Pressure Coefficient of Velocity The possibility of variations in the speed of sound in sea water with frequency is a basic question in the use of velocimeters to measure the speed. Sonar frequencies are typically less than 10 KHZ. Velocimeters work at ultrasonic frequencies. The frequency in the pulse is perhaps 3 MHZ. Velocity dispersion in sea water is always assumed absent in this frequency range. No dispersion has been reported in ultrasonic measurements and no dispersion has been observed in pure water. There are theoretical reasons why dispersion in pure liquids can only occur at very high frequencies (say greater than 50 MHZ). It seems then that the assumption is reasonable that the velocity measured by the velocimeter is the same as for sonar frequencies or that the discrepancy is extremely small. However, this is a vexed question which has not been totally answered, primarily because of the difficulties of making velocity measurements at long wave lengths. The indirect method of sound speed measurement using bathythermo- graphs relates the pressure dependence of the speed with depth in the ocean. The pressure dependence of the speed is determined in general from labora- tory experiments at ultrasonic frequencies. There is some detailed discussion in the literature (Reference 17) that the values of the depth term may be in error and there are certainly discrepancies up to 3 meters per second between laboratory measurements by Wilson (Reference 18), the tables by Kuwahava (Reference 19), and in situ measurements (Reference 17). 17 Arthur D Little, Inc. ii oath, De 1 Pin Vee othe j be Tee tn fs 1 ’ Fie Petes Ey ol ane i ! ¥ x ! A ave ] f a Ry : i crys Py beDiy i ae Peri VINA a d i oh Aes Maal aa at em Oke i) my 1 i! eee ee ee OT see ee a Bean ly "a he he ‘ 1 e ay; 1 Ning it ah i Pattee Thy in : Pate Murase if ilaesA Wh Vay Tite ‘gates II. REVIEW OF INDIRECT METHODS FOR THE MEASUREMENT OF SOUND SPEED A. Introduction As indicated in section I, the indirect method of measuring the sound speed profile consists of measuring the vertical temperature profile and computing the sound speed from this. At present, there are five types of instruments in use to determine the vertical temperature profile. These are the thermistor chain, the reversing thermometer, the mechanical bathythermograph, the expendable bathythermograph, or XBT, and the electronic S.T.D. B. Thermistor Chain This instrument is intended entirely for survey purposes and would never be used in a tactical situation. The thermistor chain used by USNEL is a string of 34 temperature sensors that operates in essen- tially a subsurface vertical line as the ship moves on the surface (Reference 20). The signals from the sensors are scanned electronically and interpolated for all whole degree centigrade isotherms which are printed on a continuous chart. In addition to this analogue output of two dimen- sional temperature distribution, the water temperatures at the 34 levels are planned to be recorded on punched tapes. The depth of the array is monitored by a pressure transducer at the submerged end, and ducted current meters monitor shear currents. An accelerometer is being applied for stability monitoring. 18 Arthur D.Little, Inc. wiv of fi f nw He nes a ee i (ag: “shi ae? i aay Ho a i) iba Aphis ry katy " ay ‘ pas wha ha ‘Mes ut ag faba braver ‘te ikea ae [bles tii Vor eae ivi) we ee bi cunt ci A ‘Nerd ye co . is Aue mi im . sue ; The chain hoist is a self-powered winch designed for work re- quiring measurements to 250 meters depth at slow ship speeds, and to about 160 meters depth at a speed of 13 knots. The 34 thermal sensing heads are type GB 32 P168 thermistors placed at intervals of 9 meters. The beads are matched to one part in two thousand which is 0.02° at the matching temperature. The chain is designed so that a defective thermistor can easily be changed by unplugging a small unit and inserting a new one. The contouring temperature recorder plots the vertical distri- bution of temperature as a continuous record. Each isotherm is displayed as a depth profile similar to that given by an echo sounder. The recorder uses a helical drum and blade principle and writes on electrosensitive paper. The measuring circuits cover a range from -20°C to 32°C and 10 isotherms can be plotted for this entire range. When gradients allow finer plotting, 0.1° or 0.05° isotherms can be recorded. The rate of scan can be varied, but 12 seconds is the optimum time for taking the temperature information for the near vertical column of water, a surface bead and the pressure sensor. The isotherms are obtained in 8 seconds; the remaining 4 seconds are used for the depth and surface measurements. The depth sensor is a Bourdon tube driving a potentiometer. The d-c output signal from this potentiometer is balanced against the d-c feed back from the feed back potentiometer in the depth servo-mechanism. The balanced signal is then amplified by a conventional 400 cycle servo- amplifier. The depth is recorded on the lower portion of the paper record. Wy) Arthur D Little, Inc. . i 1 Q Although the thermistor chain is admirably suited to survey purposes in the upper 300 meters of the ocean, the time taken for reeling out and in, and the fact that the ship's velocity must be low to ensure that the chain is quasi-vertical, preclude the use of such devices in tactical situations. There seems to be little reason why, considering current ability in towing technology, a similar system could not be adapted to considerably greater depths and be used to survey the semi-permanent deep sound channels, but it might not be competitive with a system using fixed vertical stations and research submersibles investigating the hetero- geneity between the stations. It also seems possible that the use of suitable powered depressors on the fish could make it possible to hold the chain stable at steep angles at higher ship velocities. C. Reversing Thermometers and Bathythermographs Reversing thermometers and mechanical bathythermographs are the usual methods of obtaining temperature profiles in the ocean and our present records of subsurface temperature structure consist almost entirely of data by these two instruments. Both systems are well estab- lished and are described in many references (e.g., Reference 21). Table II-1, taken from that reference, compares the two systems and also the author's suggestion of an "ideal'' general purpose instrument. In fact the numbers quoted there for the mechanical bathythermography are generally too optimistic. D. Expendable Bathythermograph The need for apparatus improved for tactical situations from 20 Arthur D. Little Inc. ix i ei! stsATeue ngTSs ut o[duts 1o uotqdeT uoTIIeTTOO oTd uOT}DeTTOO oTd -[oo otTduwes yAtM oTqrzeduo) —wes YUI—IM etTqrzeduo) -—wes YIM oTqrzeduo) Sut dues (eqoid eTqepuedxe i107) euSsTS IT}SNODe AO ‘Si0}ONpuUOD aTdrjap[nu to eTSuTs YyATM eTqe9 Tessa Jesuessoul pue o7qe P Tqd9 QUSsIeg OF UOTJOSOUUOD yndjno [Te IT3tp JO Japzodea1 SAa}eW FO Saat] peiz0qs pue peyorj,ep eq ued ‘plover OF SuTpeoi ~Tensta seitnb —PuteITe O47 YIM |, yoop uCd,, |pzA0,9y “JUaeUNAQSUT UT -ay ‘JuewnAAsUT ut uot Je UeseIg einjesoduse ainjzeirodus3 DTWUATAeZSOT AO APSUTT quejsuod 3e AesUTT queqjsuod 4e 1e8UTT : Jnopesy Des/u 4 JO 9]eI xeu Je puodses T['gQ puodes [°Q> spuodes Qz7v :qUeJSUOD OUT] qJUSUSAOWM TeUOT aaoqge sy |-deATptTun JoJ Jojow TF aaoge sy ; UOTJEUTUTIISTG STu} se poos se jou ATtensn fyjdep auq jo yadep ey JO ¥T°Q + WO CTF Ssiajow ¢F %47°O + Steqow yz 07 dQ > Aoeinodoy wy ¢ 02 Q slojew QGZ 01 Q wy G 020 :o8uey WOT JET OIAOD yadeq wy/28S G*Q + puodes [°Q> puooses 7°Q 07 uMmog spuoves Q7u :}ueJSsUOD UT] 0,£00°0 9.S0°0% 9.S00°0 ® T0°O UeeMqeg ;UOTIEUTUTIOSTG 0,10°0 Do1°0 8 Z°O UseM}0g 9.10°0 ® €0°O UeeMjeg :Aoeinod.y D.0€ 09 Z= DaOE C2 C= || DoOr O23 Go tO D0 OB Go :osuey einjeiedusy, ydersouzey zAyieg TeoTueyooy TR2Pl JajouourZsYy],, BUTSIDADY b= il eIienyaL 21 Arthur D.ULittle, Inc. ni Tye I i ft the points of view of cost, reliability, precision and time available for measurements has led to recent interest in and development of expendable devices. These consist of a free falling transducer which returns its output to the ship by a cable link. XBT is a generic name for expendable bathythermographs. The design of expendable instruments illustrates nearly all the problems and advantages which can occur with particular oceanographic instrumentation. Thus expendable instrumentation offers the advantage of speed for tactical situations because the ship does not have to slow down to take the measurement. However, the device must be inexpensive and, further, problems may occur in the telemetry of the information and the depth location of the device. For details of XBT design see Reference 22. The use of a pressure sensor to indicate depth with the required accuracy and uniformity would add greatly to the cost. Consequently, depth is computed by relating elapsed time after launching to the sink rate. This method has received wide application with expendable devices and is discussed in the references. The current XBT is wire connected to readout and recording equipment on board ship. In December 1964, it was announced that the Navy Bureau of Ships was evaluating and testing expendable BT's from three instrument companies, Bissett-Berman, G.M. Defense Research Laboratories, and Francis Associates. The basic designs of these XBT's are similar. The instruments consist of a thermistor probe in a housing containing a spool of wire which connects to a second spool of wire in a launching unit. When the probe is launched the wire from both spools is free to unwind allowing the probe to fall at a precalculated rate. The dual 22 Arthur D.Uittle Inc. et tee eae pal GMail PNT LNT ats ane it a) thee fit pin) ue het spooling technique ensures that no drag is exerted on the wire because of the ship's travel or the falling probe and the probe falls vertically. Thus a low strength, small diameter wire may be used. Changes in the resistance of the thermistor, due to temperature changes in the water are recorded on shipboard via the wire link. The rate of descent is precalculated and so depth can be recorded. At some depth, say 500 meters, the wire supply is exhausted and the probe is expended. The advantages of these expendable systems may be summarized as follows: 1. They can be used from a vessel travelling at high speed. 2. A suspect record can be easily checked by a second sensor. 3. They can be employed in weather conditions which would forbid the use of conventional equipment. 4. Since the vessel need not slow to take a measurement there will be economical and tactical advantages. 5. They exhibit greater accuracy and precision than the mechanical BT. 6. The cost per observation is low. Considerable detail on the mechanical and electrical design of these instruments is given in References 3, 22, 23. A detailed descrip- tion of the results of an experimental evaluation of the accuracy, precision and reliability of the three instruments, which was carried out under the management of the U.S. Navy Underwater Sound Laboratory, is given in Reference 24. A detailed discussion of the economic and tactical advan- 2) Arthur D.Little, Inc. ce A: tages of XBT's versus mechanical BT's, as suggested in point (4) above, is given in Reference 23. As a result of the evaluation, the Navy chose the XBT system of Francis Associates. The Francis Associates" probe is capable of obtaining water temperatures to 500 meters with an accuracy of +0.1°C over a range of -2°C to +30°C. Depth accuracy is said to be +3% over the range. Thus at a depth of about 600 meters the error in velocity due to the error in the pressure correction term approximately equals that due to temperature error. The recorder is a modified strip chart null balance positioning system. Two compensating lead wires and a sea water path are used to the probe. Measurements can be made with a ship underway at speeds up to 30 knots as demonstrated in sea tests carried out by the Navy. Sea tests have been performed to test XBT capability to 1500 meters and the results are currently being analyzed. A detailed discussion of the adequacy of the XBT to satisfy current accuracy requirements for technical active sonar is given in Reference 3. E. S.1T.D. Instrument Twenty years ago Jacobson (Reference 25) described an S.T.D. developed at W.H.O.I., which included a conductivity cell, a resistance thermometer, and a pressure transducer connected by multiple conductor cable to a deckside unit containing amplifiers, salinity computing cir- cuits, and a three-channel recorder. Salinity was +0.3%, over the range of 20 to 40%Z,; temperature covered a range of 28° to 90°F., and depth to 1200 feet. This became a "production" instrument and several were built. 24 Arthur D.Uittle, Inc. aie 0 & F iy ial it me nt iv wo Equivalent instruments are now being utilized in our national oceano- graphic programs with resolutions, accuracies (and price) from ten to 100 times as high, but about the same size. Expendable S.T.D.'s are being developed by several groups, but no obviously superior technique Or apparatus has emerged. 25 Arthur OD. ULittle Inc. y b ean tt oa Cans 2) 1 a IIIT. REVIEW OF VELOCIMETERS A. Introduction The existing instruments belong essentially to two classes: laboratory apparatus and instruments for oceanographic survey work. There have been prototype expendable instruments constructed. The measurement in the laboratory of the speed, possible dispersion, and absorption of sound waves does not concern us here directly, even though such results are of course the basis for the design and in particular for the calibra- tion of all field instruments. Equation (1) describing the speed of sound in sea water as a function of temperature, salinity, and pressure, is the result of laboratory velocimetry and the relative sizes of its terms indicate the precision required in the measurement of the respective parameters. Velocimeters for oceanographic survey work, Table III-1, (Ref- erence 26, 27, 28) require almost the same accuracy as the laboratory apparatus, because the results obtained in the ocean are of course to be compared with laboratory data and with other oceanographic data obtained at different locations and times. In addition, scientific velocimetry in situ will in general require the simultaneous measurement of other parameters in equations of the type of (1). When the speed is being measured and the velocity profile is therefore known, the true depth can of course be ob- tained very accurately from sonic measurement, as for example by inverted echosounder. The instruments must be capable of maximum depths, they must be repairable in the field, and their power supply must be capable of pro- longed operation. On the other hand, the price for such oceanographic 26 Arthur D.ULittle Inc. ty by ay my my 1 \ at i instruments can be high enough to cover all these requirements. Repre- sentative devices, based on the "sing around" circuit, are sold for $1700 to $3000, to which one must add about $2000 for readout electronics or about $5000 for printout electronics, in addition to a cable and winch. The third group, the expendable velocimeters, Table III-2, (Ref- erence 29, 30, 31) are intended mainly for taking the velocity profile in tactical situations. The instrument must have fine resolution and fast response, while absolute accuracy is of secondary importance. As Tables III-1 and III-2 show, several instruments for oceano- graphic work are available commercially and at least two manufacturers have designed and demonstrated expendable velocimeters for naval tactical applications (Reference 29, 30), and one system designed along the lines of, and of little more complexity than, the present expendable bathythermograph has been proposed. (Reference 31). We do not intend to judge in this report the performance and quality of instruments without having actually performed tests ourselves. We do, however, quote specifications and other information on design that is available from the manufacturers so that we may discuss this material critically. The US Naval Oceanographic Instrumentation Center has a test and evaluation program for velocimeters and the results, issued as "fact- sheets" (Reference 32), were also used in the preparation of this report. In addition, users of velocimeters were interviewed and the large body of literature on the measurement of the speed of sound was studied. A number of basic engineering problems emerged, all apparently recognized by the manufacturers, and in addition we heard opinions about the performance of specific instruments. The section on the comparison of instruments and 27 Arthur D. Little Inc. i “i AS i the potential for improvement is partly based on this material, with the exception of opinions which may have pertained to one particular specimen of unknown history and which are as likely to reflect a user's skills and prejudices as the quality of an instrument. Many of the data shown in Table III-1 and III-2, mostly in the manufacturer's own terms, are very much defined ad hoc, and they are dif- ficult to translate into absolute terms. The "accuracy" of velocimeters is usually defined as the maximum deviation of measured velocity values, which have been derived according to a calibration formula similar to Equation (1), from "true" values. The so-called "true" velocity in the calibration tank of distilled water is calculated (Reference 1, 17, 33) from pressure and temperature. This method of calibration is, however, difficult due to the strong dependence of velocity on temperature and due to the high demands on purity of the water. The temperature must be known, for example, within 0.02° at 0°C to avoid velocity errors greater than 0.1 meters per second from this source alone (see Equation 1). B. Tabular Listing of Velocimeters The following, Tables III-1 and III-2, give a tabular listing of velocimeters followed by explanatory notes. 28 Arthur D. Little Inc. f i; . See yo x ro a i ‘ ¥ oe i] Uf ; im b | F ) ‘ i we one (OZ) 00SZ$ Inoge (61) 472 l ZH 000ST/000S= Cals) Ole O= ST‘0 0002 punoie 3uts (91) TE€/O€OW (ST) (AMSSH1d) HOULAH (seseyjquere (VT) %ZT (YT) OF $/33 OOT/AT € c €T 48°T4S (1) 1s (ZT) Cie) 220) (72) €°0 Cue) £20 (ZT) 3893 ISdO00T (€T) eoueqsTp-ouTy, (ZT) T-HOd/NV (11) HaldWa-VNAG ur Sa}0U004) 8¢ O89TS L Saas) ZH OOSZL= (8) €°0 000ZT punoie 3uts vv /00604 (L) @aHHADOT dHaNLOVANNVW sjUeunijsuy eTqepuedxeuou pue AsAAIng otydeasouea|g 0S6z$ LT 8¢ ZH OOO€= 7 c él Hd ¥-ZLT 38°T4S OLT (OTe (9) €0°0 T0°0 (S) T°0 000cT punoie 3uts A-XN (7) AVSWVY T-III Advi ce 0S9zs GT/6 G°9T/S°6T ZH OO8€= 9°E c zeauy /3S*TIS OVT eaOla (9) 0/sT0°0O T0°0 (Ss) SECO 0002T punoie 3uts (7) €-AL/y-UL (1) doV (81) e9UeteFey “ITT Ajddns zomod IO jnopesi io 2T98) 3NO/Mm 9soTIg (yout) Wetq °xeW (qT) 243TOM gndjng AizjzeweTey, (ZHN) *beag SUT3UEY AIoONpsuerzy, sieonpsuerz3 Jo -on siojstsueiq Jo °on yaed Tevfasnooy SUTUTJOq TeTI97eW (sit) potieg surytdues (S/33/T) 2044q MOTA (9, s/w) quaeToOTFJe09 6° duay, (S/W+) kat TTqQeqeedey (s/w) AITTEWQeIS (s/uz) Adeandoy (w) yjdeq *xeW eTdtoutig “on odd, sowey Aueduog OILSTaaLOVaVHO Arthur D. Little Inc. 29 ye DO Lt, 0h A fon as i ; Pin yd ie ie: Cu) Be Cli a dae ie (1) (2) (3) (4) (5) FOOTNOTES TO TABLE III-1 The former "ACF" instrument is now made by: NUS Corporation, Underwater Systems Division, Hackensack, New Jersey. Tel: 201-265-2400. NUS ("ACF") make two types, TR-3 and TR-4. The TR-3 is shorter but its sound path is determined by the dimensions of a bulkhead of stainless steel and a temperature and pressure correction is hence required. The TR-4, the later model, defines the path by three freestanding Invar rods and therefore has practically no temperature or pressure error. According to the manufacturer, the Navy Oceanographic Office, Instru- mentation Center, issues a Fact-sheet (1FS-001-64) in which they state: "The relationship between the sound velocity (C), and the "Sing-Around'' frequency (F), as stated by the manufacturer is: C = .205916F +.15 m/sec. This equation was observed to result in an error of +.31 to -.05 m/sec. However, if the relationship for this particular instrument is changed to: C = .206451F - 3.96 m/sec, the error is within +.1 m/sec." Made by Ramsay Engineering Company, 707 North Anaheim Boulevard, Anaheim, California, Tel: 714-774-5511. The Navy Oceanographic Office, Instrumentation Center, tested this instrument (Reference 32) and issued a fact-sheet (1FS-006-66) in which they state: "The relationship between sound velocity (C) and the sing-around frequency (F) as stated in the manufacturer's specifications is C = 0.5 F. Using the MK-V Deep Sea Probe, sound velocity measurements were made in distilled water for the temperature range of 0°C to 30°C. Comparison of test results with standard values (Martin Greenspan and Carroll E. Tschiegg, "Speed of Sound in Water by a Direct Method," Journal of Research of the National Bureau of Standards, 59:4 30 Arthur D. Little Inc. Tay i Actus uh ties ) } UNG ty a 1 ee aie tS A nay a i oar aan Daan Dee; tak i (6) (7) (8) (9) (10) ((atit)) (October 1957), Research Paper 2795, p. 253.) showed an error range of +0.5 m/sec at 1405 m/sec to -0.1 m/sec at 1505 m/sec when the manufacturer's equation, C = 0.5 F, was used. However, if the equation for this particular instrument is changed to C = 0.499565 F + 0.025 (t - 20°C) + 1.15 m/sec, the error based on the same test results is within a range of +0.1 m/sec. A calibration equation which includes a term to compensate for effects of temperature change on the. sound- path length of the instrument is used. The equation is of the form C = a+b (t - t,.) + dF where (a), (b) and (d) are constants and (t - t,.) is the difference between in- strument temperature (t) and the temperature (t,.) at which the manufacturer has established zero temperature correction." This temperature coefficient is due to the use of stainless steel as a base for the acoustical path. Why the temperature coefficient of the ACF instrument should be half this value is not clear. The expansion of the stainless steel base should result in aT. C. of velocity of: a = 18 x Ho” << 1500 = Sa ae ee ee ee oy: "aL Acs 43se9 uUssueN SLIJOWUTIOTOA esn AQtutrTes B8urpntpouy siahey voejyang jo a0ueSTSieg ATTeqd pue ~Teuoseas PEN ee eee SBoIyY pe IeTeS UT ydeasouzoyqAyieg eT qepuedxy “d° LA'S qse9 uosueNy Sle IUTIOTEA sn sTeuuey) dseq pue siahey] voejsAng jo aodueqISsTsieg ATTeq pue [euoseas PEON , AeOWTIOTOA , AOJOWTIOTOA eTqepuedxg, eTqepuedxq, io ydei30 i0 yde1z3o0 -wisyzAyjeg —-wisy .AUuIeg eTqepuedxy os[qepuedxy espn osn yond MOTeq yong ur eTFJOAd eTFFOAd AREIOTOA ATIOTOA Peon POON qong aoezans Teuuey) que3i1aau0) daa SINAWHYINOTY AAVN c-AL ATEVL eqeq AjtutTes y3tm ydes3 owzoyzAu7eg aT qepuedxy AO Jepunos oyoq pue , ta VUTIOTOA eT qe puedxg , esp edots ‘yjdeq wo7}0g pue eT FJOAd AQTIOTOA P2ON matt eus SUOTJENITS TeVTIOe] 42 Arthur D. Little, Inc. B. Problems and Recommendations 1. Depth Accuracy Accurate depth location is required for two reasons: to ensure correct profile determination and,in the case of indirect methods, for the application of a correct pressure correction to the sound speed measured. Expendable instruments are necessarily located by a precalculated sink rate and elapsed time. Possible errors in depth location by this method have been considered and compared with experimental data. The analyses have neglected some possible hydrodynamical problems (e.g. Ref- erence 38). Other instruments are located in depth by means of pressure transducers or by inverted echo sounder. When using pressure transducers, such as the ''Vibrotron,'' depth is calculated from hydrostatic pressure. Problems may arise when measuring great depths in regions of gravity anom- alies or when the exact density of the water is in question. The inverted echo sounder, on the other hand, is particularly suited for use in this system. This is a method where depth is calculated from the round trip time of a pulse from the echo sounder which is attached to the velocimeter to the sea surface and back to the echo sounder at the velocimeter. Because the velocimeter measures sound velocity during descent, the output of the echo sounder can be reduced using these true velocity data directly, thus avoiding all errors due to uncertainties in water density and local gravity. 2. Velocity Accuracy The accuracy of the available velocimeters as listed in Table III-1l is excellent for any naval requirement. Temperature-coefficients of the acoustical path and of the gain and delay of transducers and amplifier, as well as pressure coefficients due to buckling of bulkheads, etc. and other 43 Arthur D. Little, Inc. rerasitntns minie~4 aaa cml sagt bees oa pen sonal tay on: ii nin “boon, ; G7) yada ao.ew ah pal bemowaii hs ator wh i i aa stoma a ee ah ' wine: taro, on sat Latins yeah ay staat al hedge. wee lee tibia, atdaiaaye’ 4 ait ce ater sith Phy hoe CREDA Ripe daniel ee ia dae ‘uit hail eeeeet iy systematic errors can be removed by applying suitable corrections. However, certain defects can occur under field conditions which cause errors which would be significant even in a tactical situation. The most serious of these defects are spurious triggering due to precursors (i.e., the second and later echoes of previous pulses ) and triggering on cycles other than the first in the transmitted pulse. Thus with the usual N.B.S. type of velocimeter, a triggering error of one period of the carrier frequency leads to a velocity error of about 3 meters per second. The causes of spurious triggering are listed below: a. Depending on the geometric arrangement of the transducers and reflectors that determine the sound path, multiple re- flections of past pulses could arrive almost simultaneously with the desired directly transmitted pulse. b. Random noise could add or subtract from the signal amplitude, thus shifting the moment of triggering. c. The attenuation in the sound path can change in operation, either by misalignment of the transducers or reflectors, by changes in the gain of the electronics or of the transducers, or by changes in the attenuation of the sea water. The re- sulting change in signal level at the receiver input will then change the moment of triggering. d. A transmitting transducer that is relatively undamped will put out a train of waves rather than one pulse and these extra oscillations can confuse the receiver input. 44 Arthur D. Little Inc. sachinga is We nie aa ‘al ‘gu ss nan ef ae Liessrnan Problem (a) can be solved by proper layout of the sound path, by suppressing spurious oscillations (as in Problem (d) ) and by gating the receiver "on" during a predetermined "window" only. Noise, Problem (b), can be overcome by providing sufficient signal. The influence of attenuation, Problem (c), can be eliminated by an automatic gain control circuit to adjust either the transmitted output power or the receiver amplification so as to deliver a constant peak amplitude signal to the threshold circuit. The T.R.G. under development by NUS Corporation will include A.G.C. In view of these possible problems and the relatively large investment of time in each run, it seems therefore advisable to use two velocimeters together whenever possible for oceanographic survey purposes. 3. Expendable Velocimeters For the specific purpose of sonar range prediction, a velocimeter should be designed to be able to distinguish the fine structure, so that the first derivative of the velocity profile can be reliably computed. The absolute values of depths and velocity, i.e. scale of the velocity profile, are of lesser importance. It is in this respect that the B.T. methods can prove unsatis-— factory, because they cannot take into account any salinity anomalies and therefore may not always reproduce the velocity profile in the upper layers above the main thermocline correctly in shallow water. Any instrument useful for sonar-range prediction must therefore have high sensitivity to small changes in velocity, small hysteresis, i.e. good short term reproducibility, and its speed of response must be appro- 45 Arthur D. Little Inc. rs y e alah (ely tees, L priate to its sinking rate. If it operates on a digital principle or if it has digital readout, its output must be quantized sufficiently finely so as not to obscure any detail that occurred in the velocity profile, even though this detail may be finer than the long term stability or accuracy of the instrument. On the other hand, the comparatively modest absolute accuracies required in this special application may permit the use of principles probably not accurate enough for oceanographic work. Thus, phase measurements on free propagating continuous waves, simple time/ distance methods, or the counting of modes of high order in resonators which operate at fixed frequencies could be employed. It is, however, dif- ficult to think of anything more simple and efficient than the sing around circuit. Taking advantage of these special conditions, it would be reason- able to consider the design of an expendable instrument of bulk and cost comparable to the expendable Bathythermograph. Two such designs, the "Ex- pendable Sound Velocimeter" of Dyna Empire, Inc., and the "Aquasonde" of Tele-Dynamics Division of American Bosch Arma Corporation have already been tested by the Naval Oceanographic Office (see Table III-2). Other systems, such as a freefalling expendable single wire system similar in appearance to the XBT (Reference 31) have been proposed and undoubtedly others have been thought of, designed, and tested. 4. Possibilities of Hybrid Systems for Maximum Cost Effectiveness, and Recommendations Not the least among the considerations that must be kept in mind by instrumentation system designers is that users of successful systems place great and unquestioning faith on the system and its performance. This 46 Arthur D. Little Inc. Spent ae a , My anes cre a ed ECO, : atten Or. hu t oi is especially so in a military instrumentation system where doctrine may establish the manner of utilization of a system or of its components. The user comes to rely on the schemata of the system rather than upon an inter- pretation of the meanings of each of the measured variables in the system. A particular problem in system performance arises when there is no alternative method either for measuring an important variable or for assessing the context in which the measurement is being made. A case in point which particularly applies here is the use of a bathythermograph aboard a ship in an ASW exercise. In most operating areas, it would be quite reasonable to assume a fairly uniform salinity structure and therefore be able to infer local sound speed structure from the temperature trace recorded. However, if the area is a so-called shallow water area, that is, one where internal waves or strong salinity gradients are present, the tem- perature profile measurement at only one point may not provide adequate information useful to the sonar operator. In fact, it may lead him to erroneous conclusions. Historical data will allow him to hazard a guess as to the utility of the measurements he is receiving, but in coastal areas or off the mouths of sizable rivers, the exact meander of the fresh water plume and its consequent eddies may be unknown. Some indication of the existence of such a problem can be gained by monitoring the intake water salinity, but it is unlikely that under stress the sonar crew will be able to quickly recognize such indications, even if they are capable of inter- preting them. A study has been made (Reference 39) which shows that a rather careful measurement of conductivity must be made in order to yield sound velocity structure to the accuracy a sonar officer would like to have in 47 Arthur D. Little, Inc. : iat sa ie 7 fans wit: haflsesine 2 te ~ a * o i! Tk tw wad: pbb, ii sla ia 8 maid vs felt shee ony LL ental Sith THY halt? Wapiti “elt, Oe, acon wine nia ; oT Ailsey & ie: Sabi at jahpow 60 Sp Loken. fe A areas of brackish or "marble-ized" water. The complexity and expense of a suitable device (XSTD), at this time may be too high to be justified as an expendable instrument in normal situations, but such a determination can- not be made until the instrument, currently under development at NAVOCEANO, is field tested. Interim alternatives in the form of hybrid systems are possible, however, and while they would be compromises, they would add only a fraction to the cost of an expendable BT cast, and need by made only as frequently as necessary to determine the oceanographic conditions. The first involves salinity recognition and the second direct sound speed measurement. When the XBT probe is falling through water that is of lower sa- linity than inferential sound speed calculations would permit within accuracy requirements of the tactical situation, an indication can be ob- tained. It is possible to modify the expendable BT so that the output of the XBT is modulated, that is, interrupted when it is in brackish water. This rate of interruption would vary with conductivity below any given value of conductivity, with no interruption above that value. It is possible that such a circuit could give sufficient information that the BT information derived could be used within certain depth intervals if perhaps not in others. It would not be necessary that every BT cast use the modified device, but only one every nth cast when in waters where oceanographic conditions are known to be variable. The record would show where brackish water existed, and appropriate action could then be taken, such as use of an XSV. 48 Arthur D. Little Inc. Pan ity ua: ee aun 4 pe ll ia ‘anion | a ive ua tub vee! saad we sti ie m | bi “ia nn bau a ip bibl wisi 4, a Piya poi rine het, oe ‘etgunts i aa hae : a a riba nil bovis We us Au uy i a . Pe ee an : i. ; “ he paar » iG 7a ie plesorsoi’ ro ab, ihn visinalen anhae TRE e it hos neha af ny 0 soi tanher ae ie 2 ial a i . 4 teaming wu ne : sei asec AS i ee shy mnt Pa oa Ma jee i: 7 F i ee +h io aa a jira, wale a he ah mr tea baiivens a) Ae ig i iv in anion, hus ts " ae Pa re ath «plo ‘eine we ih Anika a if bavi: ait hee yl iin Mi. 4 Upon casual examination, it appears possible that the only modification required to the ordinary XBT would be addition of one semi- conductor and one extra electrode, in addition to the one already used on the XBT, since no accurate measurement is being attempted. Since there would be no requirement for batteries or extra cable conductors, the hydro- dynamics and operational requirements would not be altered. Availability of such an XCT would allow more judicious use of the XSV suggested in the previous section and would considerable raise the permissable cost of the XSV since it would then only be used when its particular capabilities were necessary. Thus it appears possible that a family of expendable devices may be found to be a more practical solution than any one type of expendable device, regardless of how inexpensive it may be possible to make it. The XBT can be modified (Reference 31) to measure the sound velocity profile (SVP) in the ocean directly by installing a hydrophone tuned to the corner frequency of the ships pinger in the freefalling probe unit. The p.r.f. of the pinger would be varied on a sing around basis by signals from the XBT. The direct measurement of SVP circumvents the necessity of approximating salinity effects in using the temperature profile for range prediction and would be useful for tactical ASW purposes. The modified XBT unit could simultaniously provide temperature profile data for synoptic prediction purposes. A preliminary error analysis indicated that a sound velocity profile could be produced with a precision which is roughly com- parable with the +0.15°C temperature capability of the XBT. This XSV would be the only mobile velocimeter system to measure true group velocity a low frequencies, because its long acoustical path 49 Arthur D. Little Inc. “a on 7 +i | i, “ha mo A o 7 Hf } | : | yeas otitny Aion ne bin aghaane ripe \ a i ; . ait ote bw i a (im er tvew, ah pony ‘UK va bi | | oy ads kanye sy, atte at fn ects with ‘eabualiyper 7 a Henetiang ia: voli tr hdl Ah - tRNA iad id indi val ne ae : _ ae : “tows ai: equ: a a ebinnater sthbos i sae e hy —— sae ran Died wiguibel Dae P nati ie aoutinahgh! ois an i - | niahyts yi arid bane ii i he lawl Beha: nae Pua te he aa i 7 7 ow itrasgan ond inl atk ave ty sm tere ae Na: vie. fa | ” alae ¥ pate, ween! a | ae 7. uaiet 0 ahve wiv daiveigtim) sits Apes ne er ‘bat yeben, ont “coming any aon xy heise i an fli) > tm a 2 uns 68 Rene elitonl et) he aig shyly steigtahinl Dr . if bavi # Anis bean bbink! ‘ich ii: iOTUR: ‘ey acai is Ms; ane i pio (dans ar fatty BOSSE NEG 14 Pah ainnaibiane ‘ep sen st it ty Veta dais ik isis = i between probe and ship permits pulse propagation measurement at low carrier frequencies. However, if dispersion can be proven to be a negligible problem, this advantage disappears. On the other hand the XSV has the inherent dis- advantage of its principle of measuring the integral of sound velocity over the full instantaneous depth of the probe and this adds one differentiation to the computation of the velocity profile. In summary, it can be said that further study may show that the most cost-effective shipboard routine for ascertaining oceanographic con- ditions for sonar pupposes may include one or more hybrid or special purpose devices in addition to the regular expendable BT. It is thus our finding in this limited study that the Navy should continue the development and de- ployment of the expendable BT, that the development of the XSV should be investigated, and that a hybrid expendable conductivity—-temperature device (XCT) be explored as a means of determining the variability of onshore and shallow-water sonar oceanographic conditions, pending the availability of an XSTD. 50 Arthur D. Little, Inc. e Baca a nanan ” irVhnna ies hana wma nwa i eRe Bh ty Ot wre ‘wit’ th Hoven Wikia: Se iat act: tl ua fx VBR ws aan ain a ne on bea! a ped fo. at nad dbp ‘any snot eek | ae je by wee tbat nt nae shiny Land a hike sw ers fh! he eet f in V. CONCLUSIONS As a result of this study we have reached the following conclusions: 1. The accuracy requirements for acoustic velocity data may be either relative or absolute. Relative accuracy in the shape of the velocity profiles is usually sufficient for tactical needs. However, for long range surveillance in the deep sound channel, high absolute accuracy may be required. 2. Velocimeters are well suited to surveying the deep sound channel. The use of velocimeters can reduce the error in location of a distant object by a factor of at least 2 to 4 times the possible error which would occur if X.B.T.'s were employed. The principle error in the use of an X.B.T. at great depth is in the pressure correction term because of the error in depth location. The projected range error due to computed velocity at the sound channel axis using an X.B.T. based on a temperature error of + 0.2°F and 2 to 3% projected depth error at about 1500 meters depth (1) is between 0.04 and 0.062. 3. The present velocimeters are adequate for survey purposes. The manufacturers are aware of the problems and attempt solutions. However, there has been in fact little development beyond the original sing-around system, presumably since the returns from limited sales of velocimeters do not warrant extensive development programs. (L) Comprehensive tests performed in May 1968 aboard the NRL research vessel Mizar have shown that the standard deviation of the fall rates of 6,000 ft. deep X.B.T.'s is less than 1%. If any bias on this data is shown to be less than 1% then the absolute values of the fall rate at various depths will be known to an accuracy of 1%. Thus this precision between X.B.T.'s would allow prediction of B.T. depth to 1%. A report detailing these tests is now in preparation for NAVSHIPS, Code OOVIH. 51 Arthur A. Uittle Inc. a4 lan pe wea TRON . Hen ies Rani aa hand it | i sins ia edu ei one ‘eta relaiel ahah aman Dy Piaokeiad oi dae ae nes Ai iat i i eee eo Ac aoe ae vesie e THO CM BN ays RS ht lie Nel eine, at: ci re aed duit bial jt eee ta by peut $n Pryde mont aay, ‘paid ; ey bal ; “ . geal ve io Wy 4 ay, C a Tia: ; ah nik ante 4, The present expendable bathythermograph is adequate for current tactical needs in determining the sound velocity profile in deep water. A possible problem area which, however, does not often occur in the deep ocean is the case of a nearly isothermal surface duct where salinity data may be necessary to determine the velocity gradient. The development of the surface vessel launched X.B.T. is complete. 5. A study should be made of the adequacy of temperature data alone to indicate the presence of a surface duct in shallow water where sharp discontinuities in salinity can occur. The study should indicate what advantage might be gained by the use of expendable velocimeters or conductivity probes if such were available. The study should include possible errors in the pressure correction term due to sharp changes in density which will be common to all expendable instrumentation. 6. Dependent on the results of the study (5.) a development program should be aimed at producing an inexpensive expendable velocimeter and/or conductivity probe, not to compete with the X.B.T., but to supplement it for use in areas where temperature data might not suffice. There have eee attempts at and suggestions of designs. The program need not necessarily first compete with the current price of the X.B.T., but rather should aim at producing an accurate instrument at a cost of perhaps $100 to $200. If such an instrument were successful the decrease in cost for large scale production could be predicted. 52 Arthur A AUittle Inc. | mate ant” aan eae nau vans ne et | eka oho ve pl) hte ee eadigans ? ee Nagoran, ie rr 8 Sere wrote Ye yaa ont 6 atin, 8 Bogota him :) ier thie hod tuts *, ou maui Wye 7" ay pamauiet ah, ania ; i ah wabsabad ‘Aki ehh ce crea thins vaeing at ‘eit | un : ag neosonh malar: ah otinians Be ey act | ee lege ai ith ~ oigudn whut biveln dae att ~absindpRY®. “whom | is ‘io anna 7 my sana caved na? wh tag Hes kv wi re aes Mi omy ‘ 4 | 7 i nonrierremratint wid inven, iy wh sth oA, ‘eal nea cu ; xeepore saowsto tees #: teh) chun ed) siti vai i apne, ee - seule’ shawl We se vvalerre audit ie wont vs, ae ‘ee n | | esintingut ‘er ue. i sei iota Ahi eg vigain hy 98k te Naa rare aval Gust | aaa ae “vakohiy. mat ¥ hes. vie Le pa anwar wk weal ale Four jane eta itt sana BY, seit tant es ‘fit io 8, aa ane ssn Ue PL Rh, oe ine wntes dion Ans ra NV aaah inst ieee Me ; ee ee Ban ie, R09 a his side eniel nivel oe. we anes ‘alte: Co); pie 7 by iar watel THY. Gaod Wh “wrnaaond ails Johyyirin i 3 tegen wind iy dion he et - | iphiathone: sel tid Oe ea ne nisi je oa ‘ 7 ' Drees p H) oa ¥ y } f Y ‘oe : ‘Ay 7 : 4 oe \ . Ley i j x ab 1é 4 Waly os 7 ; s a0 1 q ay » ame 4 7 ‘ i 1 ae y 5 eid x i J q A : 1. We pi af iv Fy. an t ee fend : ! ; ? OI REFERENCES (1) W. Wilson, 1960: Speed of Sound in Sea Water as a Function of Temperature, Pressure and Salinity. J.A.S.A. 32: 641. (2) J. R. Lovett, 1962: The S.V.T.P. Instrument and Some Application to Oceanography. Marine Sciences Instrumentation. Vol l. Plenum Press. (3) Arthur D. Little Report to the Department of the Navy, Bureau of Ships. No. 4150866. August 1966. Expendable Bathythermograph (XBT) System Evaluation (For Tactical Sonar Application). (4) W.H.O.I., 1960: Atlas of the Atlantic Ocean. Ed. Fuglister. (5) M. Ewing & J. L. Worzel, 1948: Geological Society of America. Mem. 27. (6) A. T. Piip, 1964: Fine Structure and Stability of the Sound Channel in the Ocean. J.A.S.A. 36, No. 10: 1948. (7) G. M. Bryan, M. Trucher and J. I. Ewing, 1963: Long Range Sonar Studies in the South Atlantic. J.A.S.A. 35: 273. (8) A. C. Kibblewhite, R. N. Denham and P. H. Barker, 1965: Long Range Sound Propagation Study in the Southern Ocean--Project Neptunes JistAcomAe 64 NOmtc O29 (9) J. A. Pierce, et. al., 1948: Loran. M.I.T. Radiation Laboratory Series. Vol 4. Mc Graw Hill. (10) M. K. Robinson, 1966: Summary of Computer Analyzed Temperature Data for Pacific and Atlantic Oceans. Third U.S. Navy Symposium on Military Oceanography. San Diego. (11) L. Liebermann, 1951: The Effect of Temperature Inhomogeneity in the Ocean on the Propagation of Sound. J.A.S.A. 23: 563. (12) E. J. Skudrzyk, 1962: Thermal Microstructure in the Sea and its Contribution to Sound Level Fluctuations. Underwater Acoustics. V. M. Albers Ed. Plenum Press. 53 Arthur D.ULittle Inc. PMO Pitas (13) Eden, H. F., and M. Mohr, 1966. Acoustic Effects of Internal Waves in the Ocean. Third U.S. Symposium on Military Oceanography, San Diego. (14) E. Bullard, 1954: Proceedings of the Royal Society. A 222: 408 (15) R. Gerard, M. B. Langseth and M. Ewing, 1962: Thermal Gradient Measurement in the Water and Bottom Sediment of the Western Atlantic. Journal of Geophysical Research. 67: 785. (16) R. S. Bennin, C. S. Clay and J. A. Smith, 1966: In Situ Sound Velocity Measurements in Ocean Bottom Sediments. ¥orty-seventh A.G.M. of the American Geophysical Union. Washington D.C. (17) K. V. Mackenzie, 1960: Formulas for the Computation of Sound Speed in Sea Water. J.A.S.A. 32: 100. (18) W. Wilson, 1959: Speed of Sound in Distilled Water as a Function of Temperature and Pressure. J.A.S.A. 31: 1067. (19) S. Kuwahara, 1938: Japanese Journal of Astronomy and Geophysics. I 8 ike (20) E. C. Lafond, 1962: Towed Sea Temperature Structure Profiles. Marine Sciences Instrumentation. Vol 2. Plenum Press. (21) T. M. Dauphinee and H. Preston-Thomas, 1962: The Measurement of Ocean Temperature: Temperature Measurement and Control in Science and Industry. Vol. 3, Part I. Ed. Herzfeld. Reinhold. (22) Marine Sciences Instrumentation. Vol 3. Plenum Press. 1965. (23) Arthur D. Little Report to the Department of the Navy, Bureau of Ships, No. 4010765. July 1965. Cost Effectiveness: Mechanical Bathythermograph versus the Expendable Bathythermograph. (conf.) (24) Arthur D. Little Report to the Department of the Navy, Bureau of Ships, No. 4071165. November 1965. Experimental Evaluation of Expendable Bathythermograph. 54 Arthur D. Little, Inc. per Denes Wee ae nA EE i spite ie pes ne are 7 es, eae hovel ap oleh Fy! Cae Oh a! a Blah, io . os fil, ah “ " ie vase = | a8 (25) Jacobson, A. W., 1948. An Instrument for Recording Continuously the Salinity, Temperature, and Depth of Sea Water. Transactions iL, WUD, (26) M. Greenspan, G. E. Tschiegg, (NBS) 1963: A Sing Around Velocimeter for Measuring the Speed of Sound in the Sea. Underwater Acoustics. Lecture 5: p. 87. Plenum Press. ; @p J. R. Lovett, S. H. Sessions: An Instrument for Continuous Deep Sea Measuring of Velocity of Sound, Temperature, and Pressure. (NOTS) Navweps Report 7650. (28) F. J. Suellentrop, A. E. Brown, Eric Rule, 1961: An Instrument for the Direct Measurement of the Speed of Sound in the Ocean. Marine Sciences Instrumentation. Vol I: 186. Plenum Press. (29) R. A. Stahl, R. L. Miller (Teledynamics), 1965: The Aquasonde. Marine Sciences Instrumentation. Vol III: Plenum Press. (30) Proposal by Dyna Empire, Inc., to U.S. Naval Oceanographic Office, and Report by U.S. Naval Oceanographic Office to the Bureau of Ships of January 7, 1965. (31) Arthur D. Little, Inc., Invention Record. Index No. 2587. 1966. Expendable Sound Velocimeter, and supporting memoranda. (32) Instrumentation Fact-sheets. U.S. Naval Oceanographic Office, Instrumentation Department, Code 6110, Washington. (33) M. Greenspan, C. E. Tschiegg (NBS), 1959: Tables of the Speed of Sound in Water. J.A.S.A. 31:75 (34) Frank K. Brown, 1953: The Development of an Underwater Sound Velocity Meter Using a Cylindrical Resonator. University of Michigan Thesis. (35) John D. Shaffer, 1960: Low Frequency Underwater Sound Velocity Misteeres = INSIL Sle alshlY)o 55 Arthur D. Little Inc. (36) M. Greenspan, C. E. Tschiegg, R. E. Breckenridge (NBS), 1957: A Progressive Wave Velocimeter and the Speed of Sound in Water. VoAoSohtc 28 HOS. (37) W. G. Neubauer, L. R. Dragonette, 1964: Experimental Determinatio of the Freefield Sound Speed in Water. J.A.S.A. 36: 1685. (38) Alex H. Haynes, Walter L. Reid, Gerald F. Appell, 1965: Depth Accuracy of Expendable Oceanographic Instruments. Marine Sciences Instrumentation. Vol III. Plenum Press. (39) Arthur D. Little, Inc., Working Memorandum No. 26-1. October 1967. Schimke, G. R. Salinity Accuracy Requirement. 56 Arthur D. Little, Inc. APPENDIX A Insonification of the ocean is strongly dependent on the structure of the velocity profile. Figures A-1, A-2, A-3 show three velocity profiles (VP-1, VP-2, VP-3) and associated ray plots for a source at 15 feet below the surface. A semilogarithmic depth scale was used on the ray plots to emphasize the shallow depths. Rays were terminated after a maximum of nine surface bounces. VP-1l has an initial positive gradient, the velocity is 5022.0 ft/sec at the surface, and 5024.5 ft/sec at 100 feet. VP-3 has an initial gradient that is the negative reflection of VP-1, i.e., it has a velocity of 5026.0 ft/sec at the surface and 5024.5 ft/sec at 100 feet. VP-2 has an iso-velocity layer down to approximately 400 feet with a velocity of 5024.5 ft/sec. Figure A-4, A-5 shows ray plots of convergence zones. The first and second convergence zone are shown for VP-1; the first convergence zone is shown for VP-3. Rays were allowed a maximum of three surface bounces and two bottom bounces before termination. The structure of the layer depth is relatively insignificant when considering zone operation. A-1 Arthur J. Little Inc. Os) Py nha aaaaed Bee ee os Oe LN if 4 ST Ss ee cor * uli A wht i | : j i a Walter aha y ah 1) leas Se ; K a Sard a aN: BEEP IR UN PEE FIGURE A-1 3 POSITIVE INITIAL GRADIENT (2) moe oo oe 8 8 a 8 a i a Go a a ei a 0 | eo | HESS sa0 i 1 i ed is oo acbeepe CEC I 4Uee ie a a Fe = favEo nee | Ao GE 0 | : si a a : 9 go ao fs a oo ee EN Nl i a 0 CoC LSU oo Doe oe eS | 5 i oe ao ato | i Ss a oo 0 per) lS i a a HESoues io I 0 SS a a ! Be I oo el le Sa 1 oe 8 a SoU noe SRE ERESEHeE mmo Sor 70 i BSG fia] ee 2 a a (CA eC SU a oe a SSE a ee S002 (See an Aenea Bes VOC MEON OC IoC o Gi mmeseace ane SE EEE SS Se eee ee ae ie es : SIS Ba ete eee eno eee na a ar a sera oi JO Ss ooo coo 2 Ooi Se ae a Se Re I a sDeScamuE LC BASS ae (ea ale LEO SSRCoPooES Sons i a B6Gr | SoBe GUS OS UCr Poe a i on 403 3 4.680X10 4920x410" Oy +03 03 4,960x10" 5.000x30 5.oaox10°0> 5.080x10" SOUND WEROCIIY = Bisel 7s FIGURE A-1 (Continued) RAY PLOT FOR VP-1 21,5 to 41.5. in 0.15— STEPS INITIAL ANGLES alsis) NI [alels! W \ a WA > = : t a =n oO onyaitta i ae Ee) a a a ee Geir = ee a Se 3 20 Ges 40 RANGE IN KYDS Saye in HE TH UN SEE FIGURE A-2 VES ISOVELOCITY INITIAL GRADIENT a | i a ia a A SS a a Ooo Soc OS cco sc Ueno oe Oe mooonn ood sendosencs reco) ff 0 A a a SS eee GELS aaes == ee Oe Seen = ame a ea aa i a a ea a ad SS a a dS ea GE LEEEGGnHGRGnnn 5000x1079 5 .040x10" coy SOUND VELOCITY — FEET/S 5 .080xi0 BlelP vial UN REET a.xi077! +04 Sa SS a SS Ge Gee ee eS) o 6 ROVE e teal wes 0 4 A ¥ 7 Ba SS Ee tf ;4 OIE FIGURE A-2 (Continued) RAY PLOT FOR VP-2 INITIAL ANGLES -1.5 TO 41.5 INO.15° stTEps = SS o 30 “RANGE IN KYDS rer Ary, eres v9 LY ees ve ary) : ons if rays ; i} i VR ace Va 2 iy il Uy 4 | ke iy te ge Peniearteniiness BNE tial, Sok aiid oh WHA wa! 7! y Te) cea) Ap 1¢ s : A | ; i a Wie) x ye ee an Hives ae six: on eng febjogie ; om) heaypen 1g oe eee 1 os. - i. cee ay weer Rune 7 Ay: eae dala see ia a, reach ha ha Fr aN DePirl IN Peel FIGURE A-3 3 NEGATIVE INITIAL GRADIENT a i a a a ep ah a a a a a a ie i ea a 68 0G OE 2 a a aaa TL a a eS a i ec OA | Sa am wee 5S 8 eo SHORE LETeeees at a Bae Heo eee Cee cased i | a na a i saooo| | | | | aera Eee Ee FEE EEEECECEPE ESSE I ao] er | 0S ce) oe 4. eeoxs0” 4,920X10 4.960X10° + §.000x10 ? 5040x210" 5.080x10° SOUNE! WelOGiiy = PESI/S aed ey | ; cueaeen _ eae . ae ; ees | ina a eo, | prop Pans Wt ne Jog: (perp Ae eh | 9 j ] A | 4 i ¥. ie 1 ie mote | ae ; Wat Te oy : ‘ee ; frp Mea : ; 1 ; i ¥ ie va Der al IN FeleT FIGURE A-3 (Continued) RAY PLOT FOR VP-3 INITIAL ANGLES -1.5 TO 41.5 IN 0.15° STEPS ~ a= Cee : 20 " RANGE IN KYDS WM BY) el one i sernlinnpete AD fat Aciiaelaet neat 4 Wen : wie ie DEI Ne Reet) FIGURE A-4 RAY PLOT SHOWING CONVERGENCE ZONES FOR VP-1 2000| | 4000 6000 10000 RANGE IN KYDS A-8 Bs a ah i Sey Perry ey es Dithadeuit ” ries oT _ as ey tia ok . ih Hi Paley ¥ DEP WR MIN PEEL FIGURE A-5 RAY PLOT SHOWING CONVERGENCE ZONE FOR VP-3 Fa i a A Fes SE ee eeaemEn A Ty ia ian Ys a ean al Ng a PR ee iS 18000 [FOr eS a a A i | | RANGE IN KYDS (eSeeeees Saas 80 ioo e, i) i, Ae ay kind Boyes ripe } a! hes i? ayer yi : } Security Classification DOCUMENT CONTROL DATA-R&D (Security classification of title, body of abstract and indexing annotation must be entered when the overall report is classified) ORIGINATING ACTIVITY (Corporate author) - 2a. REPORT SECURITY CLASSIFICATION Unclassifiéd Neewre Wo Ibslteeila, Ine, REPORT TITLE The Measurement of Sound Speed in the Sea: Velocimeters and Other Devices 4. DESCRIPTIVE NOTES (Type of report and inclusive dates) 5. AUTHOR(S) (First name, middle initial, last name) H. Francis Eden, Emma M. Duchane, Peter Von Thuna, and Richard H. Van Haagen 6. REPORT DATE 7a. TOTAL NO. OF PAGES 7b. NO. OF REFS | May 1967 63 39 Ba. CONTRACT OR GRANT NO. 9a. ORIGINATOR'S REPORT NUMBER(S) Beguine eh 1930568 SF-101-03-21 9b. OTHER REPORT NO(S) (Any other numbers that may be assigned this report) 10. DISTRIBUTION STATEMENT All distribution of this report is controlled. Qualified D.D.C. users shall request through their cognizant codes to obtain the report. 11. SUPPLEMENTARY NOTES 12. SPONSORING MILITARY ACTIVITY Dept. of the Navy Naval Ship Systems Command 13. ABSTRACT Data on the speed of sound in the sea is a prime requirement in Navy operations. pnly relative accuracy in the measurements of sound speed as a function of depth is eeded in tactical situations because it is the shape of the velocity profile that is pf major importance to hull mounted sonar. Absolute accuracy in the measurement of sound speed is desirable, however, in order to exploit longer range propagation in the deep sound channel. ! The sound speed can be measured either indirectly, by computation from temperature pnd salinity or density data, or directly, by using "velocimeters". In tactical Bituations, temperature data alone has sufficed for the indirect method except under ertain near shore conditions, where skilled guesswork has been required. Present velocimeters based on the "sing-around" principle are adequate for oceano- praphic survey purposes, from the points of view of accuracy and depth capability. owever, their connecting cables render them unsuitable for tactical use; in general, he manufacturers of velocimeters are aware of and have been attempting solutions to the problems that arise from this application. The value of temperature data alone for defining the presence of a surface duct in phallow water where sharp salinity changes can occur should be studied. We believe that} here exists a case for the development of an expendable instrument or of a synoptic echnique for tactical use to supplement the now successful expendable bathythermograph. DD "1473 Security Classification Security Classification KEY WORDS Security Classification ont bey FOR OFFICIAL USE ONLY AOReT Hee Ro DE Tt! tie i Nee CAMBRIDGE e WASHINGTON e EDINBURGH e CHICAGO SANTA MONICA e MEXICO CITY ¢ LONDON e ZURICH BRUSSELS e SAN FRANCISCO e NEW YORK e TORONTO FOR OFFICIAL USE ONLY